diff options
Diffstat (limited to 'drivers/net/ethernet')
539 files changed, 29481 insertions, 9157 deletions
diff --git a/drivers/net/ethernet/3com/3c509.c b/drivers/net/ethernet/3com/3c509.c index 8cafd06ff0c4..b762176a1406 100644 --- a/drivers/net/ethernet/3com/3c509.c +++ b/drivers/net/ethernet/3com/3c509.c @@ -60,8 +60,6 @@ */ #define DRV_NAME "3c509" -#define DRV_VERSION "1.20" -#define DRV_RELDATE "04Feb2008" /* A few values that may be tweaked. */ @@ -87,13 +85,12 @@ #include <linux/device.h> #include <linux/eisa.h> #include <linux/bitops.h> +#include <linux/vermagic.h> #include <linux/uaccess.h> #include <asm/io.h> #include <asm/irq.h> -static char version[] = DRV_NAME ".c:" DRV_VERSION " " DRV_RELDATE " becker@scyld.com\n"; - #ifdef EL3_DEBUG static int el3_debug = EL3_DEBUG; #else @@ -547,8 +544,6 @@ static int el3_common_init(struct net_device *dev) dev->name, dev->base_addr, if_names[(dev->if_port & 0x03)], dev->dev_addr, dev->irq); - if (el3_debug > 0) - pr_info("%s", version); return 0; } @@ -1143,7 +1138,6 @@ el3_netdev_set_ecmd(struct net_device *dev, static void el3_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); } static int el3_get_link_ksettings(struct net_device *dev, diff --git a/drivers/net/ethernet/3com/3c515.c b/drivers/net/ethernet/3com/3c515.c index 1e233e2f0a5a..90312fcd6319 100644 --- a/drivers/net/ethernet/3com/3c515.c +++ b/drivers/net/ethernet/3com/3c515.c @@ -22,12 +22,8 @@ */ +#include <linux/vermagic.h> #define DRV_NAME "3c515" -#define DRV_VERSION "0.99t-ac" -#define DRV_RELDATE "28-Oct-2002" - -static char *version = -DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " becker@scyld.com and others\n"; #define CORKSCREW 1 @@ -84,7 +80,6 @@ static int max_interrupt_work = 20; MODULE_AUTHOR("Donald Becker <becker@scyld.com>"); MODULE_DESCRIPTION("3Com 3c515 Corkscrew driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); /* "Knobs" for adjusting internal parameters. */ /* Put out somewhat more debugging messages. (0 - no msg, 1 minimal msgs). */ @@ -418,8 +413,6 @@ int init_module(void) int found = 0; if (debug >= 0) corkscrew_debug = debug; - if (corkscrew_debug) - pr_debug("%s", version); while (corkscrew_scan(-1)) found++; return found ? 0 : -ENODEV; @@ -429,16 +422,10 @@ int init_module(void) struct net_device *tc515_probe(int unit) { struct net_device *dev = corkscrew_scan(unit); - static int printed; if (!dev) return ERR_PTR(-ENODEV); - if (corkscrew_debug > 0 && !printed) { - printed = 1; - pr_debug("%s", version); - } - return dev; } #endif /* not MODULE */ @@ -1540,7 +1527,6 @@ static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); snprintf(info->bus_info, sizeof(info->bus_info), "ISA 0x%lx", dev->base_addr); } diff --git a/drivers/net/ethernet/3com/3c589_cs.c b/drivers/net/ethernet/3com/3c589_cs.c index d47cde6c5f08..09816e84314d 100644 --- a/drivers/net/ethernet/3com/3c589_cs.c +++ b/drivers/net/ethernet/3com/3c589_cs.c @@ -23,7 +23,6 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define DRV_NAME "3c589_cs" -#define DRV_VERSION "1.162-ac" #include <linux/module.h> #include <linux/kernel.h> @@ -482,7 +481,6 @@ static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); snprintf(info->bus_info, sizeof(info->bus_info), "PCMCIA 0x%lx", dev->base_addr); } diff --git a/drivers/net/ethernet/3com/typhoon.c b/drivers/net/ethernet/3com/typhoon.c index 14fce6658106..5ed33c2c4742 100644 --- a/drivers/net/ethernet/3com/typhoon.c +++ b/drivers/net/ethernet/3com/typhoon.c @@ -127,7 +127,6 @@ static const int multicast_filter_limit = 32; #include "typhoon.h" MODULE_AUTHOR("David Dillow <dave@thedillows.org>"); -MODULE_VERSION("1.0"); MODULE_LICENSE("GPL"); MODULE_FIRMWARE(FIRMWARE_NAME); MODULE_DESCRIPTION("3Com Typhoon Family (3C990, 3CR990, and variants)"); @@ -311,7 +310,7 @@ enum state_values { * cannot pass a read, so this forces current writes to post. */ #define typhoon_post_pci_writes(x) \ - do { if(likely(use_mmio)) ioread32(x+TYPHOON_REG_HEARTBEAT); } while(0) + do { if (likely(use_mmio)) ioread32(x+TYPHOON_REG_HEARTBEAT); } while (0) /* We'll wait up to six seconds for a reset, and half a second normally. */ @@ -381,7 +380,7 @@ typhoon_reset(void __iomem *ioaddr, int wait_type) int i, err = 0; int timeout; - if(wait_type == WaitNoSleep) + if (wait_type == WaitNoSleep) timeout = TYPHOON_RESET_TIMEOUT_NOSLEEP; else timeout = TYPHOON_RESET_TIMEOUT_SLEEP; @@ -394,13 +393,13 @@ typhoon_reset(void __iomem *ioaddr, int wait_type) udelay(1); iowrite32(TYPHOON_RESET_NONE, ioaddr + TYPHOON_REG_SOFT_RESET); - if(wait_type != NoWait) { - for(i = 0; i < timeout; i++) { - if(ioread32(ioaddr + TYPHOON_REG_STATUS) == + if (wait_type != NoWait) { + for (i = 0; i < timeout; i++) { + if (ioread32(ioaddr + TYPHOON_REG_STATUS) == TYPHOON_STATUS_WAITING_FOR_HOST) goto out; - if(wait_type == WaitSleep) + if (wait_type == WaitSleep) schedule_timeout_uninterruptible(1); else udelay(TYPHOON_UDELAY); @@ -423,7 +422,7 @@ out: * which should be enough (I've see it work well at 100us, but still * saw occasional problems.) */ - if(wait_type == WaitSleep) + if (wait_type == WaitSleep) msleep(5); else udelay(500); @@ -435,8 +434,8 @@ typhoon_wait_status(void __iomem *ioaddr, u32 wait_value) { int i, err = 0; - for(i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) { - if(ioread32(ioaddr + TYPHOON_REG_STATUS) == wait_value) + for (i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) { + if (ioread32(ioaddr + TYPHOON_REG_STATUS) == wait_value) goto out; udelay(TYPHOON_UDELAY); } @@ -450,7 +449,7 @@ out: static inline void typhoon_media_status(struct net_device *dev, struct resp_desc *resp) { - if(resp->parm1 & TYPHOON_MEDIA_STAT_NO_LINK) + if (resp->parm1 & TYPHOON_MEDIA_STAT_NO_LINK) netif_carrier_off(dev); else netif_carrier_on(dev); @@ -466,7 +465,7 @@ typhoon_hello(struct typhoon *tp) * card in a long while. If the lock is held, then we're in the * process of issuing a command, so we don't need to respond. */ - if(spin_trylock(&tp->command_lock)) { + if (spin_trylock(&tp->command_lock)) { cmd = (struct cmd_desc *)(ring->ringBase + ring->lastWrite); typhoon_inc_cmd_index(&ring->lastWrite, 1); @@ -490,32 +489,32 @@ typhoon_process_response(struct typhoon *tp, int resp_size, cleared = le32_to_cpu(indexes->respCleared); ready = le32_to_cpu(indexes->respReady); - while(cleared != ready) { + while (cleared != ready) { resp = (struct resp_desc *)(base + cleared); count = resp->numDesc + 1; - if(resp_save && resp->seqNo) { - if(count > resp_size) { + if (resp_save && resp->seqNo) { + if (count > resp_size) { resp_save->flags = TYPHOON_RESP_ERROR; goto cleanup; } wrap_len = 0; len = count * sizeof(*resp); - if(unlikely(cleared + len > RESPONSE_RING_SIZE)) { + if (unlikely(cleared + len > RESPONSE_RING_SIZE)) { wrap_len = cleared + len - RESPONSE_RING_SIZE; len = RESPONSE_RING_SIZE - cleared; } memcpy(resp_save, resp, len); - if(unlikely(wrap_len)) { + if (unlikely(wrap_len)) { resp_save += len / sizeof(*resp); memcpy(resp_save, base, wrap_len); } resp_save = NULL; - } else if(resp->cmd == TYPHOON_CMD_READ_MEDIA_STATUS) { + } else if (resp->cmd == TYPHOON_CMD_READ_MEDIA_STATUS) { typhoon_media_status(tp->dev, resp); - } else if(resp->cmd == TYPHOON_CMD_HELLO_RESP) { + } else if (resp->cmd == TYPHOON_CMD_HELLO_RESP) { typhoon_hello(tp); } else { netdev_err(tp->dev, @@ -589,19 +588,19 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd, freeCmd = typhoon_num_free_cmd(tp); freeResp = typhoon_num_free_resp(tp); - if(freeCmd < num_cmd || freeResp < num_resp) { + if (freeCmd < num_cmd || freeResp < num_resp) { netdev_err(tp->dev, "no descs for cmd, had (needed) %d (%d) cmd, %d (%d) resp\n", freeCmd, num_cmd, freeResp, num_resp); err = -ENOMEM; goto out; } - if(cmd->flags & TYPHOON_CMD_RESPOND) { + if (cmd->flags & TYPHOON_CMD_RESPOND) { /* If we're expecting a response, but the caller hasn't given * us a place to put it, we'll provide one. */ tp->awaiting_resp = 1; - if(resp == NULL) { + if (resp == NULL) { resp = &local_resp; num_resp = 1; } @@ -609,13 +608,13 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd, wrap_len = 0; len = num_cmd * sizeof(*cmd); - if(unlikely(ring->lastWrite + len > COMMAND_RING_SIZE)) { + if (unlikely(ring->lastWrite + len > COMMAND_RING_SIZE)) { wrap_len = ring->lastWrite + len - COMMAND_RING_SIZE; len = COMMAND_RING_SIZE - ring->lastWrite; } memcpy(ring->ringBase + ring->lastWrite, cmd, len); - if(unlikely(wrap_len)) { + if (unlikely(wrap_len)) { struct cmd_desc *wrap_ptr = cmd; wrap_ptr += len / sizeof(*cmd); memcpy(ring->ringBase, wrap_ptr, wrap_len); @@ -629,7 +628,7 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd, iowrite32(ring->lastWrite, tp->ioaddr + TYPHOON_REG_CMD_READY); typhoon_post_pci_writes(tp->ioaddr); - if((cmd->flags & TYPHOON_CMD_RESPOND) == 0) + if ((cmd->flags & TYPHOON_CMD_RESPOND) == 0) goto out; /* Ugh. We'll be here about 8ms, spinning our thumbs, unable to @@ -649,14 +648,14 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd, * wait here. */ got_resp = 0; - for(i = 0; i < TYPHOON_WAIT_TIMEOUT && !got_resp; i++) { - if(indexes->respCleared != indexes->respReady) + for (i = 0; i < TYPHOON_WAIT_TIMEOUT && !got_resp; i++) { + if (indexes->respCleared != indexes->respReady) got_resp = typhoon_process_response(tp, num_resp, resp); udelay(TYPHOON_UDELAY); } - if(!got_resp) { + if (!got_resp) { err = -ETIMEDOUT; goto out; } @@ -664,11 +663,11 @@ typhoon_issue_command(struct typhoon *tp, int num_cmd, struct cmd_desc *cmd, /* Collect the error response even if we don't care about the * rest of the response */ - if(resp->flags & TYPHOON_RESP_ERROR) + if (resp->flags & TYPHOON_RESP_ERROR) err = -EIO; out: - if(tp->awaiting_resp) { + if (tp->awaiting_resp) { tp->awaiting_resp = 0; smp_wmb(); @@ -679,7 +678,7 @@ out: * time. So, check for it, and interrupt ourselves if this * is the case. */ - if(indexes->respCleared != indexes->respReady) + if (indexes->respCleared != indexes->respReady) iowrite32(1, tp->ioaddr + TYPHOON_REG_SELF_INTERRUPT); } @@ -749,7 +748,7 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) * between marking the queue awake and updating the cleared index. * Just loop and it will appear. This comes from the acenic driver. */ - while(unlikely(typhoon_num_free_tx(txRing) < (numDesc + 2))) + while (unlikely(typhoon_num_free_tx(txRing) < (numDesc + 2))) smp_rmb(); first_txd = (struct tx_desc *) (txRing->ringBase + txRing->lastWrite); @@ -761,7 +760,7 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) first_txd->tx_addr = (u64)((unsigned long) skb); first_txd->processFlags = 0; - if(skb->ip_summed == CHECKSUM_PARTIAL) { + if (skb->ip_summed == CHECKSUM_PARTIAL) { /* The 3XP will figure out if this is UDP/TCP */ first_txd->processFlags |= TYPHOON_TX_PF_TCP_CHKSUM; first_txd->processFlags |= TYPHOON_TX_PF_UDP_CHKSUM; @@ -789,7 +788,7 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) /* No need to worry about padding packet -- the firmware pads * it with zeros to ETH_ZLEN for us. */ - if(skb_shinfo(skb)->nr_frags == 0) { + if (skb_shinfo(skb)->nr_frags == 0) { skb_dma = pci_map_single(tp->tx_pdev, skb->data, skb->len, PCI_DMA_TODEVICE); txd->flags = TYPHOON_FRAG_DESC | TYPHOON_DESC_VALID; @@ -841,14 +840,14 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) */ numDesc = MAX_SKB_FRAGS + TSO_NUM_DESCRIPTORS + 1; - if(typhoon_num_free_tx(txRing) < (numDesc + 2)) { + if (typhoon_num_free_tx(txRing) < (numDesc + 2)) { netif_stop_queue(dev); /* A Tx complete IRQ could have gotten between, making * the ring free again. Only need to recheck here, since * Tx is serialized. */ - if(typhoon_num_free_tx(txRing) >= (numDesc + 2)) + if (typhoon_num_free_tx(txRing) >= (numDesc + 2)) netif_wake_queue(dev); } @@ -864,7 +863,7 @@ typhoon_set_rx_mode(struct net_device *dev) __le16 filter; filter = TYPHOON_RX_FILTER_DIRECTED | TYPHOON_RX_FILTER_BROADCAST; - if(dev->flags & IFF_PROMISC) { + if (dev->flags & IFF_PROMISC) { filter |= TYPHOON_RX_FILTER_PROMISCOUS; } else if ((netdev_mc_count(dev) > multicast_filter_limit) || (dev->flags & IFF_ALLMULTI)) { @@ -906,7 +905,7 @@ typhoon_do_get_stats(struct typhoon *tp) INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_READ_STATS); err = typhoon_issue_command(tp, 1, &xp_cmd, 7, xp_resp); - if(err < 0) + if (err < 0) return err; /* 3Com's Linux driver uses txMultipleCollisions as it's @@ -954,10 +953,10 @@ typhoon_get_stats(struct net_device *dev) struct net_device_stats *saved = &tp->stats_saved; smp_rmb(); - if(tp->card_state == Sleeping) + if (tp->card_state == Sleeping) return saved; - if(typhoon_do_get_stats(tp) < 0) { + if (typhoon_do_get_stats(tp) < 0) { netdev_err(dev, "error getting stats\n"); return saved; } @@ -974,12 +973,12 @@ typhoon_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) struct resp_desc xp_resp[3]; smp_rmb(); - if(tp->card_state == Sleeping) { + if (tp->card_state == Sleeping) { strlcpy(info->fw_version, "Sleep image", sizeof(info->fw_version)); } else { INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_READ_VERSIONS); - if(typhoon_issue_command(tp, 1, &xp_cmd, 3, xp_resp) < 0) { + if (typhoon_issue_command(tp, 1, &xp_cmd, 3, xp_resp) < 0) { strlcpy(info->fw_version, "Unknown runtime", sizeof(info->fw_version)); } else { @@ -1026,7 +1025,7 @@ typhoon_get_link_ksettings(struct net_device *dev, break; } - if(tp->capabilities & TYPHOON_FIBER) { + if (tp->capabilities & TYPHOON_FIBER) { supported |= SUPPORTED_FIBRE; advertising |= ADVERTISED_FIBRE; cmd->base.port = PORT_FIBRE; @@ -1043,7 +1042,7 @@ typhoon_get_link_ksettings(struct net_device *dev, cmd->base.speed = tp->speed; cmd->base.duplex = tp->duplex; cmd->base.phy_address = 0; - if(tp->xcvr_select == TYPHOON_XCVR_AUTONEG) + if (tp->xcvr_select == TYPHOON_XCVR_AUTONEG) cmd->base.autoneg = AUTONEG_ENABLE; else cmd->base.autoneg = AUTONEG_DISABLE; @@ -1091,7 +1090,7 @@ typhoon_set_link_ksettings(struct net_device *dev, INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_XCVR_SELECT); xp_cmd.parm1 = xcvr; err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto out; tp->xcvr_select = xcvr; @@ -1114,9 +1113,9 @@ typhoon_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) wol->supported = WAKE_PHY | WAKE_MAGIC; wol->wolopts = 0; - if(tp->wol_events & TYPHOON_WAKE_LINK_EVENT) + if (tp->wol_events & TYPHOON_WAKE_LINK_EVENT) wol->wolopts |= WAKE_PHY; - if(tp->wol_events & TYPHOON_WAKE_MAGIC_PKT) + if (tp->wol_events & TYPHOON_WAKE_MAGIC_PKT) wol->wolopts |= WAKE_MAGIC; memset(&wol->sopass, 0, sizeof(wol->sopass)); } @@ -1126,13 +1125,13 @@ typhoon_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct typhoon *tp = netdev_priv(dev); - if(wol->wolopts & ~(WAKE_PHY | WAKE_MAGIC)) + if (wol->wolopts & ~(WAKE_PHY | WAKE_MAGIC)) return -EINVAL; tp->wol_events = 0; - if(wol->wolopts & WAKE_PHY) + if (wol->wolopts & WAKE_PHY) tp->wol_events |= TYPHOON_WAKE_LINK_EVENT; - if(wol->wolopts & WAKE_MAGIC) + if (wol->wolopts & WAKE_MAGIC) tp->wol_events |= TYPHOON_WAKE_MAGIC_PKT; return 0; @@ -1163,8 +1162,8 @@ typhoon_wait_interrupt(void __iomem *ioaddr) { int i, err = 0; - for(i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) { - if(ioread32(ioaddr + TYPHOON_REG_INTR_STATUS) & + for (i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) { + if (ioread32(ioaddr + TYPHOON_REG_INTR_STATUS) & TYPHOON_INTR_BOOTCMD) goto out; udelay(TYPHOON_UDELAY); @@ -1356,7 +1355,7 @@ typhoon_download_firmware(struct typhoon *tp) */ err = -ENOMEM; dpage = pci_alloc_consistent(pdev, PAGE_SIZE, &dpage_dma); - if(!dpage) { + if (!dpage) { netdev_err(tp->dev, "no DMA mem for firmware\n"); goto err_out; } @@ -1369,7 +1368,7 @@ typhoon_download_firmware(struct typhoon *tp) ioaddr + TYPHOON_REG_INTR_MASK); err = -ETIMEDOUT; - if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) { + if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) { netdev_err(tp->dev, "card ready timeout\n"); goto err_out_irq; } @@ -1398,16 +1397,16 @@ typhoon_download_firmware(struct typhoon *tp) * last write to the command register to post, so * we don't need a typhoon_post_pci_writes() after it. */ - for(i = 0; i < numSections; i++) { + for (i = 0; i < numSections; i++) { sHdr = (struct typhoon_section_header *) image_data; image_data += sizeof(struct typhoon_section_header); load_addr = le32_to_cpu(sHdr->startAddr); section_len = le32_to_cpu(sHdr->len); - while(section_len) { + while (section_len) { len = min_t(u32, section_len, PAGE_SIZE); - if(typhoon_wait_interrupt(ioaddr) < 0 || + if (typhoon_wait_interrupt(ioaddr) < 0 || ioread32(ioaddr + TYPHOON_REG_STATUS) != TYPHOON_STATUS_WAITING_FOR_SEGMENT) { netdev_err(tp->dev, "segment ready timeout\n"); @@ -1440,7 +1439,7 @@ typhoon_download_firmware(struct typhoon *tp) } } - if(typhoon_wait_interrupt(ioaddr) < 0 || + if (typhoon_wait_interrupt(ioaddr) < 0 || ioread32(ioaddr + TYPHOON_REG_STATUS) != TYPHOON_STATUS_WAITING_FOR_SEGMENT) { netdev_err(tp->dev, "final segment ready timeout\n"); @@ -1449,7 +1448,7 @@ typhoon_download_firmware(struct typhoon *tp) iowrite32(TYPHOON_BOOTCMD_DNLD_COMPLETE, ioaddr + TYPHOON_REG_COMMAND); - if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_BOOT) < 0) { + if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_BOOT) < 0) { netdev_err(tp->dev, "boot ready timeout, status 0x%0x\n", ioread32(ioaddr + TYPHOON_REG_STATUS)); goto err_out_irq; @@ -1472,7 +1471,7 @@ typhoon_boot_3XP(struct typhoon *tp, u32 initial_status) { void __iomem *ioaddr = tp->ioaddr; - if(typhoon_wait_status(ioaddr, initial_status) < 0) { + if (typhoon_wait_status(ioaddr, initial_status) < 0) { netdev_err(tp->dev, "boot ready timeout\n"); goto out_timeout; } @@ -1483,7 +1482,7 @@ typhoon_boot_3XP(struct typhoon *tp, u32 initial_status) iowrite32(TYPHOON_BOOTCMD_REG_BOOT_RECORD, ioaddr + TYPHOON_REG_COMMAND); - if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_RUNNING) < 0) { + if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_RUNNING) < 0) { netdev_err(tp->dev, "boot finish timeout (status 0x%x)\n", ioread32(ioaddr + TYPHOON_REG_STATUS)); goto out_timeout; @@ -1513,17 +1512,17 @@ typhoon_clean_tx(struct typhoon *tp, struct transmit_ring *txRing, int dma_len; int type; - while(lastRead != le32_to_cpu(*index)) { + while (lastRead != le32_to_cpu(*index)) { tx = (struct tx_desc *) (txRing->ringBase + lastRead); type = tx->flags & TYPHOON_TYPE_MASK; - if(type == TYPHOON_TX_DESC) { + if (type == TYPHOON_TX_DESC) { /* This tx_desc describes a packet. */ unsigned long ptr = tx->tx_addr; struct sk_buff *skb = (struct sk_buff *) ptr; dev_kfree_skb_irq(skb); - } else if(type == TYPHOON_FRAG_DESC) { + } else if (type == TYPHOON_FRAG_DESC) { /* This tx_desc describes a memory mapping. Free it. */ skb_dma = (dma_addr_t) le32_to_cpu(tx->frag.addr); @@ -1548,7 +1547,7 @@ typhoon_tx_complete(struct typhoon *tp, struct transmit_ring *txRing, /* This will need changing if we start to use the Hi Tx ring. */ lastRead = typhoon_clean_tx(tp, txRing, index); - if(netif_queue_stopped(tp->dev) && typhoon_num_free(txRing->lastWrite, + if (netif_queue_stopped(tp->dev) && typhoon_num_free(txRing->lastWrite, lastRead, TXLO_ENTRIES) > (numDesc + 2)) netif_wake_queue(tp->dev); @@ -1564,7 +1563,7 @@ typhoon_recycle_rx_skb(struct typhoon *tp, u32 idx) struct basic_ring *ring = &tp->rxBuffRing; struct rx_free *r; - if((ring->lastWrite + sizeof(*r)) % (RXFREE_ENTRIES * sizeof(*r)) == + if ((ring->lastWrite + sizeof(*r)) % (RXFREE_ENTRIES * sizeof(*r)) == le32_to_cpu(indexes->rxBuffCleared)) { /* no room in ring, just drop the skb */ @@ -1595,12 +1594,12 @@ typhoon_alloc_rx_skb(struct typhoon *tp, u32 idx) rxb->skb = NULL; - if((ring->lastWrite + sizeof(*r)) % (RXFREE_ENTRIES * sizeof(*r)) == + if ((ring->lastWrite + sizeof(*r)) % (RXFREE_ENTRIES * sizeof(*r)) == le32_to_cpu(indexes->rxBuffCleared)) return -ENOMEM; skb = netdev_alloc_skb(tp->dev, PKT_BUF_SZ); - if(!skb) + if (!skb) return -ENOMEM; #if 0 @@ -1647,7 +1646,7 @@ typhoon_rx(struct typhoon *tp, struct basic_ring *rxRing, volatile __le32 * read received = 0; local_ready = le32_to_cpu(*ready); rxaddr = le32_to_cpu(*cleared); - while(rxaddr != local_ready && budget > 0) { + while (rxaddr != local_ready && budget > 0) { rx = (struct rx_desc *) (rxRing->ringBase + rxaddr); idx = rx->addr; rxb = &tp->rxbuffers[idx]; @@ -1656,14 +1655,14 @@ typhoon_rx(struct typhoon *tp, struct basic_ring *rxRing, volatile __le32 * read typhoon_inc_rx_index(&rxaddr, 1); - if(rx->flags & TYPHOON_RX_ERROR) { + if (rx->flags & TYPHOON_RX_ERROR) { typhoon_recycle_rx_skb(tp, idx); continue; } pkt_len = le16_to_cpu(rx->frameLen); - if(pkt_len < rx_copybreak && + if (pkt_len < rx_copybreak && (new_skb = netdev_alloc_skb(tp->dev, pkt_len + 2)) != NULL) { skb_reserve(new_skb, 2); pci_dma_sync_single_for_cpu(tp->pdev, dma_addr, @@ -1685,7 +1684,7 @@ typhoon_rx(struct typhoon *tp, struct basic_ring *rxRing, volatile __le32 * read new_skb->protocol = eth_type_trans(new_skb, tp->dev); csum_bits = rx->rxStatus & (TYPHOON_RX_IP_CHK_GOOD | TYPHOON_RX_UDP_CHK_GOOD | TYPHOON_RX_TCP_CHK_GOOD); - if(csum_bits == + if (csum_bits == (TYPHOON_RX_IP_CHK_GOOD | TYPHOON_RX_TCP_CHK_GOOD) || csum_bits == (TYPHOON_RX_IP_CHK_GOOD | TYPHOON_RX_UDP_CHK_GOOD)) { @@ -1711,11 +1710,11 @@ typhoon_fill_free_ring(struct typhoon *tp) { u32 i; - for(i = 0; i < RXENT_ENTRIES; i++) { + for (i = 0; i < RXENT_ENTRIES; i++) { struct rxbuff_ent *rxb = &tp->rxbuffers[i]; - if(rxb->skb) + if (rxb->skb) continue; - if(typhoon_alloc_rx_skb(tp, i) < 0) + if (typhoon_alloc_rx_skb(tp, i) < 0) break; } } @@ -1728,25 +1727,25 @@ typhoon_poll(struct napi_struct *napi, int budget) int work_done; rmb(); - if(!tp->awaiting_resp && indexes->respReady != indexes->respCleared) + if (!tp->awaiting_resp && indexes->respReady != indexes->respCleared) typhoon_process_response(tp, 0, NULL); - if(le32_to_cpu(indexes->txLoCleared) != tp->txLoRing.lastRead) + if (le32_to_cpu(indexes->txLoCleared) != tp->txLoRing.lastRead) typhoon_tx_complete(tp, &tp->txLoRing, &indexes->txLoCleared); work_done = 0; - if(indexes->rxHiCleared != indexes->rxHiReady) { + if (indexes->rxHiCleared != indexes->rxHiReady) { work_done += typhoon_rx(tp, &tp->rxHiRing, &indexes->rxHiReady, &indexes->rxHiCleared, budget); } - if(indexes->rxLoCleared != indexes->rxLoReady) { + if (indexes->rxLoCleared != indexes->rxLoReady) { work_done += typhoon_rx(tp, &tp->rxLoRing, &indexes->rxLoReady, &indexes->rxLoCleared, budget - work_done); } - if(le32_to_cpu(indexes->rxBuffCleared) == tp->rxBuffRing.lastWrite) { + if (le32_to_cpu(indexes->rxBuffCleared) == tp->rxBuffRing.lastWrite) { /* rxBuff ring is empty, try to fill it. */ typhoon_fill_free_ring(tp); } @@ -1770,7 +1769,7 @@ typhoon_interrupt(int irq, void *dev_instance) u32 intr_status; intr_status = ioread32(ioaddr + TYPHOON_REG_INTR_STATUS); - if(!(intr_status & TYPHOON_INTR_HOST_INT)) + if (!(intr_status & TYPHOON_INTR_HOST_INT)) return IRQ_NONE; iowrite32(intr_status, ioaddr + TYPHOON_REG_INTR_STATUS); @@ -1790,9 +1789,9 @@ typhoon_free_rx_rings(struct typhoon *tp) { u32 i; - for(i = 0; i < RXENT_ENTRIES; i++) { + for (i = 0; i < RXENT_ENTRIES; i++) { struct rxbuff_ent *rxb = &tp->rxbuffers[i]; - if(rxb->skb) { + if (rxb->skb) { pci_unmap_single(tp->pdev, rxb->dma_addr, PKT_BUF_SZ, PCI_DMA_FROMDEVICE); dev_kfree_skb(rxb->skb); @@ -1812,7 +1811,7 @@ typhoon_sleep(struct typhoon *tp, pci_power_t state, __le16 events) INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_ENABLE_WAKE_EVENTS); xp_cmd.parm1 = events; err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) { + if (err < 0) { netdev_err(tp->dev, "typhoon_sleep(): wake events cmd err %d\n", err); return err; @@ -1820,12 +1819,12 @@ typhoon_sleep(struct typhoon *tp, pci_power_t state, __le16 events) INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_GOTO_SLEEP); err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) { + if (err < 0) { netdev_err(tp->dev, "typhoon_sleep(): sleep cmd err %d\n", err); return err; } - if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_SLEEPING) < 0) + if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_SLEEPING) < 0) return -ETIMEDOUT; /* Since we cannot monitor the status of the link while sleeping, @@ -1852,7 +1851,7 @@ typhoon_wakeup(struct typhoon *tp, int wait_type) * the old firmware pay for the reset. */ iowrite32(TYPHOON_BOOTCMD_WAKEUP, ioaddr + TYPHOON_REG_COMMAND); - if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_HOST) < 0 || + if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_WAITING_FOR_HOST) < 0 || (tp->capabilities & TYPHOON_WAKEUP_NEEDS_RESET)) return typhoon_reset(ioaddr, wait_type); @@ -1871,12 +1870,12 @@ typhoon_start_runtime(struct typhoon *tp) typhoon_fill_free_ring(tp); err = typhoon_download_firmware(tp); - if(err < 0) { + if (err < 0) { netdev_err(tp->dev, "cannot load runtime on 3XP\n"); goto error_out; } - if(typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_BOOT) < 0) { + if (typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_BOOT) < 0) { netdev_err(tp->dev, "cannot boot 3XP\n"); err = -EIO; goto error_out; @@ -1885,14 +1884,14 @@ typhoon_start_runtime(struct typhoon *tp) INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_MAX_PKT_SIZE); xp_cmd.parm1 = cpu_to_le16(PKT_BUF_SZ); err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto error_out; INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_MAC_ADDRESS); xp_cmd.parm1 = cpu_to_le16(ntohs(*(__be16 *)&dev->dev_addr[0])); xp_cmd.parm2 = cpu_to_le32(ntohl(*(__be32 *)&dev->dev_addr[2])); err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto error_out; /* Disable IRQ coalescing -- we can reenable it when 3Com gives @@ -1901,38 +1900,38 @@ typhoon_start_runtime(struct typhoon *tp) INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_IRQ_COALESCE_CTRL); xp_cmd.parm1 = 0; err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto error_out; INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_XCVR_SELECT); xp_cmd.parm1 = tp->xcvr_select; err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto error_out; INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_VLAN_TYPE_WRITE); xp_cmd.parm1 = cpu_to_le16(ETH_P_8021Q); err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto error_out; INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_OFFLOAD_TASKS); xp_cmd.parm2 = tp->offload; xp_cmd.parm3 = tp->offload; err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto error_out; typhoon_set_rx_mode(dev); INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_TX_ENABLE); err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto error_out; INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_RX_ENABLE); err = typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(err < 0) + if (err < 0) goto error_out; tp->card_state = Running; @@ -1972,13 +1971,13 @@ typhoon_stop_runtime(struct typhoon *tp, int wait_type) /* Wait 1/2 sec for any outstanding transmits to occur * We'll cleanup after the reset if this times out. */ - for(i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) { - if(indexes->txLoCleared == cpu_to_le32(txLo->lastWrite)) + for (i = 0; i < TYPHOON_WAIT_TIMEOUT; i++) { + if (indexes->txLoCleared == cpu_to_le32(txLo->lastWrite)) break; udelay(TYPHOON_UDELAY); } - if(i == TYPHOON_WAIT_TIMEOUT) + if (i == TYPHOON_WAIT_TIMEOUT) netdev_err(tp->dev, "halt timed out waiting for Tx to complete\n"); INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_TX_DISABLE); @@ -1995,16 +1994,16 @@ typhoon_stop_runtime(struct typhoon *tp, int wait_type) INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_HALT); typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL); - if(typhoon_wait_status(ioaddr, TYPHOON_STATUS_HALTED) < 0) + if (typhoon_wait_status(ioaddr, TYPHOON_STATUS_HALTED) < 0) netdev_err(tp->dev, "timed out waiting for 3XP to halt\n"); - if(typhoon_reset(ioaddr, wait_type) < 0) { + if (typhoon_reset(ioaddr, wait_type) < 0) { netdev_err(tp->dev, "unable to reset 3XP\n"); return -ETIMEDOUT; } /* cleanup any outstanding Tx packets */ - if(indexes->txLoCleared != cpu_to_le32(txLo->lastWrite)) { + if (indexes->txLoCleared != cpu_to_le32(txLo->lastWrite)) { indexes->txLoCleared = cpu_to_le32(txLo->lastWrite); typhoon_clean_tx(tp, &tp->txLoRing, &indexes->txLoCleared); } @@ -2017,7 +2016,7 @@ typhoon_tx_timeout(struct net_device *dev, unsigned int txqueue) { struct typhoon *tp = netdev_priv(dev); - if(typhoon_reset(tp->ioaddr, WaitNoSleep) < 0) { + if (typhoon_reset(tp->ioaddr, WaitNoSleep) < 0) { netdev_warn(dev, "could not reset in tx timeout\n"); goto truly_dead; } @@ -2026,7 +2025,7 @@ typhoon_tx_timeout(struct net_device *dev, unsigned int txqueue) typhoon_clean_tx(tp, &tp->txLoRing, &tp->indexes->txLoCleared); typhoon_free_rx_rings(tp); - if(typhoon_start_runtime(tp) < 0) { + if (typhoon_start_runtime(tp) < 0) { netdev_err(dev, "could not start runtime in tx timeout\n"); goto truly_dead; } @@ -2051,20 +2050,20 @@ typhoon_open(struct net_device *dev) goto out; err = typhoon_wakeup(tp, WaitSleep); - if(err < 0) { + if (err < 0) { netdev_err(dev, "unable to wakeup device\n"); goto out_sleep; } err = request_irq(dev->irq, typhoon_interrupt, IRQF_SHARED, dev->name, dev); - if(err < 0) + if (err < 0) goto out_sleep; napi_enable(&tp->napi); err = typhoon_start_runtime(tp); - if(err < 0) { + if (err < 0) { napi_disable(&tp->napi); goto out_irq; } @@ -2076,13 +2075,13 @@ out_irq: free_irq(dev->irq, dev); out_sleep: - if(typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) { + if (typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) { netdev_err(dev, "unable to reboot into sleep img\n"); typhoon_reset(tp->ioaddr, NoWait); goto out; } - if(typhoon_sleep(tp, PCI_D3hot, 0) < 0) + if (typhoon_sleep(tp, PCI_D3hot, 0) < 0) netdev_err(dev, "unable to go back to sleep\n"); out: @@ -2097,7 +2096,7 @@ typhoon_close(struct net_device *dev) netif_stop_queue(dev); napi_disable(&tp->napi); - if(typhoon_stop_runtime(tp, WaitSleep) < 0) + if (typhoon_stop_runtime(tp, WaitSleep) < 0) netdev_err(dev, "unable to stop runtime\n"); /* Make sure there is no irq handler running on a different CPU. */ @@ -2106,10 +2105,10 @@ typhoon_close(struct net_device *dev) typhoon_free_rx_rings(tp); typhoon_init_rings(tp); - if(typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) + if (typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) netdev_err(dev, "unable to boot sleep image\n"); - if(typhoon_sleep(tp, PCI_D3hot, 0) < 0) + if (typhoon_sleep(tp, PCI_D3hot, 0) < 0) netdev_err(dev, "unable to put card to sleep\n"); return 0; @@ -2124,15 +2123,15 @@ typhoon_resume(struct pci_dev *pdev) /* If we're down, resume when we are upped. */ - if(!netif_running(dev)) + if (!netif_running(dev)) return 0; - if(typhoon_wakeup(tp, WaitNoSleep) < 0) { + if (typhoon_wakeup(tp, WaitNoSleep) < 0) { netdev_err(dev, "critical: could not wake up in resume\n"); goto reset; } - if(typhoon_start_runtime(tp) < 0) { + if (typhoon_start_runtime(tp) < 0) { netdev_err(dev, "critical: could not start runtime in resume\n"); goto reset; } @@ -2154,16 +2153,16 @@ typhoon_suspend(struct pci_dev *pdev, pm_message_t state) /* If we're down, we're already suspended. */ - if(!netif_running(dev)) + if (!netif_running(dev)) return 0; /* TYPHOON_OFFLOAD_VLAN is always on now, so this doesn't work */ - if(tp->wol_events & TYPHOON_WAKE_MAGIC_PKT) + if (tp->wol_events & TYPHOON_WAKE_MAGIC_PKT) netdev_warn(dev, "cannot do WAKE_MAGIC with VLAN offloading\n"); netif_device_detach(dev); - if(typhoon_stop_runtime(tp, WaitNoSleep) < 0) { + if (typhoon_stop_runtime(tp, WaitNoSleep) < 0) { netdev_err(dev, "unable to stop runtime\n"); goto need_resume; } @@ -2171,7 +2170,7 @@ typhoon_suspend(struct pci_dev *pdev, pm_message_t state) typhoon_free_rx_rings(tp); typhoon_init_rings(tp); - if(typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) { + if (typhoon_boot_3XP(tp, TYPHOON_STATUS_WAITING_FOR_HOST) < 0) { netdev_err(dev, "unable to boot sleep image\n"); goto need_resume; } @@ -2179,19 +2178,19 @@ typhoon_suspend(struct pci_dev *pdev, pm_message_t state) INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_MAC_ADDRESS); xp_cmd.parm1 = cpu_to_le16(ntohs(*(__be16 *)&dev->dev_addr[0])); xp_cmd.parm2 = cpu_to_le32(ntohl(*(__be32 *)&dev->dev_addr[2])); - if(typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL) < 0) { + if (typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL) < 0) { netdev_err(dev, "unable to set mac address in suspend\n"); goto need_resume; } INIT_COMMAND_NO_RESPONSE(&xp_cmd, TYPHOON_CMD_SET_RX_FILTER); xp_cmd.parm1 = TYPHOON_RX_FILTER_DIRECTED | TYPHOON_RX_FILTER_BROADCAST; - if(typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL) < 0) { + if (typhoon_issue_command(tp, 1, &xp_cmd, 0, NULL) < 0) { netdev_err(dev, "unable to set rx filter in suspend\n"); goto need_resume; } - if(typhoon_sleep(tp, pci_choose_state(pdev, state), tp->wol_events) < 0) { + if (typhoon_sleep(tp, pci_choose_state(pdev, state), tp->wol_events) < 0) { netdev_err(dev, "unable to put card to sleep\n"); goto need_resume; } @@ -2211,10 +2210,10 @@ typhoon_test_mmio(struct pci_dev *pdev) int mode = 0; u32 val; - if(!ioaddr) + if (!ioaddr) goto out; - if(ioread32(ioaddr + TYPHOON_REG_STATUS) != + if (ioread32(ioaddr + TYPHOON_REG_STATUS) != TYPHOON_STATUS_WAITING_FOR_HOST) goto out_unmap; @@ -2227,12 +2226,12 @@ typhoon_test_mmio(struct pci_dev *pdev) * The 50usec delay is arbitrary -- it could probably be smaller. */ val = ioread32(ioaddr + TYPHOON_REG_INTR_STATUS); - if((val & TYPHOON_INTR_SELF) == 0) { + if ((val & TYPHOON_INTR_SELF) == 0) { iowrite32(1, ioaddr + TYPHOON_REG_SELF_INTERRUPT); ioread32(ioaddr + TYPHOON_REG_INTR_STATUS); udelay(50); val = ioread32(ioaddr + TYPHOON_REG_INTR_STATUS); - if(val & TYPHOON_INTR_SELF) + if (val & TYPHOON_INTR_SELF) mode = 1; } @@ -2245,7 +2244,7 @@ out_unmap: pci_iounmap(pdev, ioaddr); out: - if(!mode) + if (!mode) pr_info("%s: falling back to port IO\n", pci_name(pdev)); return mode; } @@ -2276,7 +2275,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) const char *err_msg; dev = alloc_etherdev(sizeof(*tp)); - if(dev == NULL) { + if (dev == NULL) { err_msg = "unable to alloc new net device"; err = -ENOMEM; goto error_out; @@ -2284,55 +2283,55 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) SET_NETDEV_DEV(dev, &pdev->dev); err = pci_enable_device(pdev); - if(err < 0) { + if (err < 0) { err_msg = "unable to enable device"; goto error_out_dev; } err = pci_set_mwi(pdev); - if(err < 0) { + if (err < 0) { err_msg = "unable to set MWI"; goto error_out_disable; } err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); - if(err < 0) { + if (err < 0) { err_msg = "No usable DMA configuration"; goto error_out_mwi; } /* sanity checks on IO and MMIO BARs */ - if(!(pci_resource_flags(pdev, 0) & IORESOURCE_IO)) { + if (!(pci_resource_flags(pdev, 0) & IORESOURCE_IO)) { err_msg = "region #1 not a PCI IO resource, aborting"; err = -ENODEV; goto error_out_mwi; } - if(pci_resource_len(pdev, 0) < 128) { + if (pci_resource_len(pdev, 0) < 128) { err_msg = "Invalid PCI IO region size, aborting"; err = -ENODEV; goto error_out_mwi; } - if(!(pci_resource_flags(pdev, 1) & IORESOURCE_MEM)) { + if (!(pci_resource_flags(pdev, 1) & IORESOURCE_MEM)) { err_msg = "region #1 not a PCI MMIO resource, aborting"; err = -ENODEV; goto error_out_mwi; } - if(pci_resource_len(pdev, 1) < 128) { + if (pci_resource_len(pdev, 1) < 128) { err_msg = "Invalid PCI MMIO region size, aborting"; err = -ENODEV; goto error_out_mwi; } err = pci_request_regions(pdev, KBUILD_MODNAME); - if(err < 0) { + if (err < 0) { err_msg = "could not request regions"; goto error_out_mwi; } /* map our registers */ - if(use_mmio != 0 && use_mmio != 1) + if (use_mmio != 0 && use_mmio != 1) use_mmio = typhoon_test_mmio(pdev); ioaddr = pci_iomap(pdev, use_mmio, 128); @@ -2346,7 +2345,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) */ shared = pci_alloc_consistent(pdev, sizeof(struct typhoon_shared), &shared_dma); - if(!shared) { + if (!shared) { err_msg = "could not allocate DMA memory"; err = -ENOMEM; goto error_out_remap; @@ -2426,7 +2425,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) * seem to need a little extra help to get started. Since we don't * know how to nudge it along, just kick it. */ - if(xp_resp[0].numDesc != 0) + if (xp_resp[0].numDesc != 0) tp->capabilities |= TYPHOON_WAKEUP_NEEDS_RESET; err = typhoon_sleep(tp, PCI_D3hot, 0); @@ -2471,14 +2470,14 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) /* xp_resp still contains the response to the READ_VERSIONS command. * For debugging, let the user know what version he has. */ - if(xp_resp[0].numDesc == 0) { + if (xp_resp[0].numDesc == 0) { /* This is the Typhoon 1.0 type Sleep Image, last 16 bits * of version is Month/Day of build. */ u16 monthday = le32_to_cpu(xp_resp[0].parm2) & 0xffff; netdev_info(dev, "Typhoon 1.0 Sleep Image built %02u/%02u/2000\n", monthday >> 8, monthday & 0xff); - } else if(xp_resp[0].numDesc == 2) { + } else if (xp_resp[0].numDesc == 2) { /* This is the Typhoon 1.1+ type Sleep Image */ u32 sleep_ver = le32_to_cpu(xp_resp[0].parm2); diff --git a/drivers/net/ethernet/3com/typhoon.h b/drivers/net/ethernet/3com/typhoon.h index 88187fc84aa3..2f634c64d5d1 100644 --- a/drivers/net/ethernet/3com/typhoon.h +++ b/drivers/net/ethernet/3com/typhoon.h @@ -366,7 +366,7 @@ struct resp_desc { memset(_ptr, 0, sizeof(struct cmd_desc)); \ _ptr->flags = TYPHOON_CMD_DESC | TYPHOON_DESC_VALID; \ _ptr->cmd = command; \ - } while(0) + } while (0) /* We set seqNo to 1 if we're expecting a response from this command */ #define INIT_COMMAND_WITH_RESPONSE(x, command) \ @@ -376,7 +376,7 @@ struct resp_desc { _ptr->flags |= TYPHOON_DESC_VALID; \ _ptr->cmd = command; \ _ptr->seqNo = 1; \ - } while(0) + } while (0) /* TYPHOON_CMD_SET_RX_FILTER filter bits (cmd.parm1) */ diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c index 165d18405b0c..2db42211329f 100644 --- a/drivers/net/ethernet/adaptec/starfire.c +++ b/drivers/net/ethernet/adaptec/starfire.c @@ -27,8 +27,6 @@ */ #define DRV_NAME "starfire" -#define DRV_VERSION "2.1" -#define DRV_RELDATE "July 6, 2008" #include <linux/interrupt.h> #include <linux/module.h> @@ -47,6 +45,7 @@ #include <asm/processor.h> /* Processor type for cache alignment. */ #include <linux/uaccess.h> #include <asm/io.h> +#include <linux/vermagic.h> /* * The current frame processor firmware fails to checksum a fragment @@ -165,15 +164,9 @@ static int rx_copybreak /* = 0 */; #define FIRMWARE_RX "adaptec/starfire_rx.bin" #define FIRMWARE_TX "adaptec/starfire_tx.bin" -/* These identify the driver base version and may not be removed. */ -static const char version[] = -KERN_INFO "starfire.c:v1.03 7/26/2000 Written by Donald Becker <becker@scyld.com>\n" -" (unofficial 2.2/2.4 kernel port, version " DRV_VERSION ", " DRV_RELDATE ")\n"; - MODULE_AUTHOR("Donald Becker <becker@scyld.com>"); MODULE_DESCRIPTION("Adaptec Starfire Ethernet driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); MODULE_FIRMWARE(FIRMWARE_RX); MODULE_FIRMWARE(FIRMWARE_TX); @@ -654,13 +647,6 @@ static int starfire_init_one(struct pci_dev *pdev, int drv_flags, io_size; int boguscnt; -/* when built into the kernel, we only print version if device is found */ -#ifndef MODULE - static int printed_version; - if (!printed_version++) - printk(version); -#endif - if (pci_enable_device (pdev)) return -EIO; @@ -1853,7 +1839,6 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct netdev_private *np = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } @@ -2073,8 +2058,6 @@ static int __init starfire_init (void) { /* when a module, this is printed whether or not devices are found in probe */ #ifdef MODULE - printk(version); - printk(KERN_INFO DRV_NAME ": polling (NAPI) enabled\n"); #endif diff --git a/drivers/net/ethernet/aeroflex/greth.c b/drivers/net/ethernet/aeroflex/greth.c index 2a9f8643629c..bf546118dbc6 100644 --- a/drivers/net/ethernet/aeroflex/greth.c +++ b/drivers/net/ethernet/aeroflex/greth.c @@ -1114,9 +1114,7 @@ static void greth_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *in strlcpy(info->driver, dev_driver_string(greth->dev), sizeof(info->driver)); - strlcpy(info->version, "revision: 1.0", sizeof(info->version)); strlcpy(info->bus_info, greth->dev->bus->name, sizeof(info->bus_info)); - strlcpy(info->fw_version, "N/A", sizeof(info->fw_version)); } static void greth_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *p) diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index cb6a761d5c11..1b19385ad8a9 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -2958,7 +2958,6 @@ static void et131x_get_drvinfo(struct net_device *netdev, struct et131x_adapter *adapter = netdev_priv(netdev); strlcpy(info->driver, DRIVER_NAME, sizeof(info->driver)); - strlcpy(info->version, DRIVER_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(adapter->pdev), sizeof(info->bus_info)); } diff --git a/drivers/net/ethernet/agere/et131x.h b/drivers/net/ethernet/agere/et131x.h index be9a11c02526..d0e922584d8a 100644 --- a/drivers/net/ethernet/agere/et131x.h +++ b/drivers/net/ethernet/agere/et131x.h @@ -46,7 +46,6 @@ */ #define DRIVER_NAME "et131x" -#define DRIVER_VERSION "v2.0" /* EEPROM registers */ diff --git a/drivers/net/ethernet/alacritech/slicoss.c b/drivers/net/ethernet/alacritech/slicoss.c index 9daef4c8feef..6234fcd844ee 100644 --- a/drivers/net/ethernet/alacritech/slicoss.c +++ b/drivers/net/ethernet/alacritech/slicoss.c @@ -26,7 +26,6 @@ #include "slic.h" #define DRV_NAME "slicoss" -#define DRV_VERSION "1.0" static const struct pci_device_id slic_id_tbl[] = { { PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH, @@ -1533,7 +1532,6 @@ static void slic_get_drvinfo(struct net_device *dev, struct slic_device *sdev = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(sdev->pdev), sizeof(info->bus_info)); } @@ -1852,4 +1850,3 @@ module_pci_driver(slic_driver); MODULE_DESCRIPTION("Alacritech non-accelerated SLIC driver"); MODULE_AUTHOR("Lino Sanfilippo <LinoSanfilippo@gmx.de>"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c index 22cadfbeedfb..18d3b4340bd4 100644 --- a/drivers/net/ethernet/allwinner/sun4i-emac.c +++ b/drivers/net/ethernet/allwinner/sun4i-emac.c @@ -33,7 +33,6 @@ #include "sun4i-emac.h" #define DRV_NAME "sun4i-emac" -#define DRV_VERSION "1.02" #define EMAC_MAX_FRAME_LEN 0x0600 @@ -212,7 +211,6 @@ static void emac_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, dev_name(&dev->dev), sizeof(info->bus_info)); } diff --git a/drivers/net/ethernet/alteon/acenic.c b/drivers/net/ethernet/alteon/acenic.c index f366faf88eee..5d192d551623 100644 --- a/drivers/net/ethernet/alteon/acenic.c +++ b/drivers/net/ethernet/alteon/acenic.c @@ -2699,9 +2699,8 @@ static void ace_get_drvinfo(struct net_device *dev, struct ace_private *ap = netdev_priv(dev); strlcpy(info->driver, "acenic", sizeof(info->driver)); - snprintf(info->version, sizeof(info->version), "%i.%i.%i", - ap->firmware_major, ap->firmware_minor, - ap->firmware_fix); + snprintf(info->fw_version, sizeof(info->version), "%i.%i.%i", + ap->firmware_major, ap->firmware_minor, ap->firmware_fix); if (ap->pdev) strlcpy(info->bus_info, pci_name(ap->pdev), diff --git a/drivers/net/ethernet/altera/altera_tse_ethtool.c b/drivers/net/ethernet/altera/altera_tse_ethtool.c index 23823464f2e7..4299f1301149 100644 --- a/drivers/net/ethernet/altera/altera_tse_ethtool.c +++ b/drivers/net/ethernet/altera/altera_tse_ethtool.c @@ -67,7 +67,6 @@ static void tse_get_drvinfo(struct net_device *dev, u32 rev = ioread32(&priv->mac_dev->megacore_revision); strcpy(info->driver, "altera_tse"); - strcpy(info->version, "v8.0"); snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "v%d.%d", rev & 0xFFFF, (rev & 0xFFFF0000) >> 16); sprintf(info->bus_info, "platform"); diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c index 1fb58f9ad80b..a250046b8e18 100644 --- a/drivers/net/ethernet/amazon/ena/ena_com.c +++ b/drivers/net/ethernet/amazon/ena/ena_com.c @@ -1067,18 +1067,14 @@ static void ena_com_hash_key_fill_default_key(struct ena_com_dev *ena_dev) static int ena_com_hash_key_allocate(struct ena_com_dev *ena_dev) { struct ena_rss *rss = &ena_dev->rss; - struct ena_admin_feature_rss_flow_hash_control *hash_key; struct ena_admin_get_feat_resp get_resp; int rc; - hash_key = (ena_dev->rss).hash_key; - rc = ena_com_get_feature_ex(ena_dev, &get_resp, ENA_ADMIN_RSS_HASH_FUNCTION, ena_dev->rss.hash_key_dma_addr, sizeof(ena_dev->rss.hash_key), 0); if (unlikely(rc)) { - hash_key = NULL; return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c index ced1d577b62a..9cc28b4b2627 100644 --- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c +++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c @@ -221,7 +221,7 @@ static void ena_queue_strings(struct ena_adapter *adapter, u8 **data) snprintf(*data, ETH_GSTRING_LEN, "queue_%u_tx_%s", i, ena_stats->name); - (*data) += ETH_GSTRING_LEN; + (*data) += ETH_GSTRING_LEN; } /* Rx stats */ for (j = 0; j < ENA_STATS_ARRAY_RX; j++) { @@ -404,7 +404,6 @@ static void ena_get_drvinfo(struct net_device *dev, struct ena_adapter *adapter = netdev_priv(dev); strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(adapter->pdev), sizeof(info->bus_info)); } @@ -674,7 +673,6 @@ static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, * supports getting/setting the hash function. */ rc = ena_com_get_hash_function(adapter->ena_dev, &ena_func, key); - if (rc) { if (rc == -EOPNOTSUPP) { key = NULL; @@ -685,9 +683,6 @@ static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, return rc; } - if (rc) - return rc; - switch (ena_func) { case ENA_ADMIN_TOEPLITZ: func = ETH_RSS_HASH_TOP; @@ -831,6 +826,8 @@ static int ena_set_tunable(struct net_device *netdev, } static const struct ethtool_ops ena_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_link_ksettings = ena_get_link_ksettings, .get_drvinfo = ena_get_drvinfo, .get_msglevel = ena_get_msglevel, diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index cada6e7e30f4..2cc765df8da3 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -49,12 +49,9 @@ #include <linux/bpf_trace.h> #include "ena_pci_id_tbl.h" -static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n"; - MODULE_AUTHOR("Amazon.com, Inc. or its affiliates"); MODULE_DESCRIPTION(DEVICE_NAME); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_MODULE_VERSION); /* Time in jiffies before concluding the transmitter is hung. */ #define TX_TIMEOUT (5 * HZ) @@ -460,10 +457,9 @@ static void ena_xdp_unregister_rxq_info(struct ena_ring *rx_ring) xdp_rxq_info_unreg(&rx_ring->xdp_rxq); } -void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter, - struct bpf_prog *prog, - int first, - int count) +static void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter, + struct bpf_prog *prog, + int first, int count) { struct ena_ring *rx_ring; int i = 0; @@ -481,8 +477,8 @@ void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter, } } -void ena_xdp_exchange_program(struct ena_adapter *adapter, - struct bpf_prog *prog) +static void ena_xdp_exchange_program(struct ena_adapter *adapter, + struct bpf_prog *prog) { struct bpf_prog *old_bpf_prog = xchg(&adapter->xdp_bpf_prog, prog); @@ -1555,7 +1551,7 @@ static void ena_set_rx_hash(struct ena_ring *rx_ring, } } -int ena_xdp_handle_buff(struct ena_ring *rx_ring, struct xdp_buff *xdp) +static int ena_xdp_handle_buff(struct ena_ring *rx_ring, struct xdp_buff *xdp) { struct ena_rx_buffer *rx_info; int ret; @@ -3104,9 +3100,9 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev, strncpy(host_info->os_dist_str, utsname()->release, sizeof(host_info->os_dist_str) - 1); host_info->driver_version = - (DRV_MODULE_VER_MAJOR) | - (DRV_MODULE_VER_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) | - (DRV_MODULE_VER_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) | + (DRV_MODULE_GEN_MAJOR) | + (DRV_MODULE_GEN_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) | + (DRV_MODULE_GEN_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) | ("K"[0] << ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT); host_info->num_cpus = num_online_cpus(); @@ -3486,10 +3482,8 @@ static int ena_restore_device(struct ena_adapter *adapter) netif_carrier_on(adapter->netdev); mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ)); + dev_err(&pdev->dev, "Device reset completed successfully\n"); adapter->last_keep_alive_jiffies = jiffies; - dev_err(&pdev->dev, - "Device reset completed successfully, Driver info: %s\n", - version); return rc; err_disable_msix: @@ -4127,8 +4121,6 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) dev_dbg(&pdev->dev, "%s\n", __func__); - dev_info_once(&pdev->dev, "%s", version); - rc = pci_enable_device_mem(pdev); if (rc) { dev_err(&pdev->dev, "pci_enable_device_mem() failed!\n"); @@ -4471,8 +4463,6 @@ static struct pci_driver ena_pci_driver = { static int __init ena_init(void) { - pr_info("%s", version); - ena_wq = create_singlethread_workqueue(DRV_MODULE_NAME); if (!ena_wq) { pr_err("Failed to create workqueue\n"); diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h index 8795e0b1dc3c..97dfd0c67e84 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.h +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h @@ -45,16 +45,16 @@ #include "ena_com.h" #include "ena_eth_com.h" -#define DRV_MODULE_VER_MAJOR 2 -#define DRV_MODULE_VER_MINOR 1 -#define DRV_MODULE_VER_SUBMINOR 0 +#define DRV_MODULE_GEN_MAJOR 2 +#define DRV_MODULE_GEN_MINOR 1 +#define DRV_MODULE_GEN_SUBMINOR 0 #define DRV_MODULE_NAME "ena" -#ifndef DRV_MODULE_VERSION -#define DRV_MODULE_VERSION \ - __stringify(DRV_MODULE_VER_MAJOR) "." \ - __stringify(DRV_MODULE_VER_MINOR) "." \ - __stringify(DRV_MODULE_VER_SUBMINOR) "K" +#ifndef DRV_MODULE_GENERATION +#define DRV_MODULE_GENERATION \ + __stringify(DRV_MODULE_GEN_MAJOR) "." \ + __stringify(DRV_MODULE_GEN_MINOR) "." \ + __stringify(DRV_MODULE_GEN_SUBMINOR) "K" #endif #define DEVICE_NAME "Elastic Network Adapter (ENA)" diff --git a/drivers/net/ethernet/amd/amd8111e.c b/drivers/net/ethernet/amd/amd8111e.c index 0f3b743425e8..7a1286f8e983 100644 --- a/drivers/net/ethernet/amd/amd8111e.c +++ b/drivers/net/ethernet/amd/amd8111e.c @@ -84,9 +84,8 @@ Revision History: #include "amd8111e.h" #define MODULE_NAME "amd8111e" -#define MODULE_VERS "3.0.7" MODULE_AUTHOR("Advanced Micro Devices, Inc."); -MODULE_DESCRIPTION ("AMD8111 based 10/100 Ethernet Controller. Driver Version "MODULE_VERS); +MODULE_DESCRIPTION("AMD8111 based 10/100 Ethernet Controller."); MODULE_LICENSE("GPL"); module_param_array(speed_duplex, int, NULL, 0); MODULE_PARM_DESC(speed_duplex, "Set device speed and duplex modes, 0: Auto Negotiate, 1: 10Mbps Half Duplex, 2: 10Mbps Full Duplex, 3: 100Mbps Half Duplex, 4: 100Mbps Full Duplex"); @@ -1366,7 +1365,6 @@ static void amd8111e_get_drvinfo(struct net_device *dev, struct amd8111e_priv *lp = netdev_priv(dev); struct pci_dev *pci_dev = lp->pci_dev; strlcpy(info->driver, MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, MODULE_VERS, sizeof(info->version)); snprintf(info->fw_version, sizeof(info->fw_version), "%u", chip_version); strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info)); @@ -1875,7 +1873,6 @@ static int amd8111e_probe_one(struct pci_dev *pdev, /* display driver and device information */ chip_version = (readl(lp->mmio + CHIPID) & 0xf0000000)>>28; - dev_info(&pdev->dev, "AMD-8111e Driver Version: %s\n", MODULE_VERS); dev_info(&pdev->dev, "[ Rev %x ] PCI 10/100BaseT Ethernet %pM\n", chip_version, dev->dev_addr); if (lp->ext_phy_id) diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c index 089a4fbc61a0..9f6e3cc2ce80 100644 --- a/drivers/net/ethernet/amd/au1000_eth.c +++ b/drivers/net/ethernet/amd/au1000_eth.c @@ -63,14 +63,12 @@ static int au1000_debug = 3; NETIF_MSG_LINK) #define DRV_NAME "au1000_eth" -#define DRV_VERSION "1.7" #define DRV_AUTHOR "Pete Popov <ppopov@embeddedalley.com>" #define DRV_DESC "Au1xxx on-chip Ethernet driver" MODULE_AUTHOR(DRV_AUTHOR); MODULE_DESCRIPTION(DRV_DESC); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); /* AU1000 MAC registers and bits */ #define MAC_CONTROL 0x0 @@ -656,7 +654,6 @@ au1000_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) struct au1000_private *aup = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); snprintf(info->bus_info, sizeof(info->bus_info), "%s %d", DRV_NAME, aup->mac_id); } @@ -1290,8 +1287,6 @@ static int au1000_probe(struct platform_device *pdev) netdev_info(dev, "Au1xx0 Ethernet found at 0x%lx, irq %d\n", (unsigned long)base->start, irq); - pr_info_once("%s version %s %s\n", DRV_NAME, DRV_VERSION, DRV_AUTHOR); - return 0; err_out: diff --git a/drivers/net/ethernet/amd/nmclan_cs.c b/drivers/net/ethernet/amd/nmclan_cs.c index 023aecf6ab30..11c0b13edd30 100644 --- a/drivers/net/ethernet/amd/nmclan_cs.c +++ b/drivers/net/ethernet/amd/nmclan_cs.c @@ -114,8 +114,6 @@ Log: nmclan_cs.c,v #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define DRV_NAME "nmclan_cs" -#define DRV_VERSION "0.16" - /* ---------------------------------------------------------------------------- Conditional Compilation Options @@ -367,7 +365,7 @@ typedef struct _mace_private { char tx_free_frames; /* Number of free transmit frame buffers */ char tx_irq_disabled; /* MACE TX interrupt disabled */ - + spinlock_t bank_lock; /* Must be held if you step off bank 0 */ } mace_private; @@ -444,7 +442,7 @@ static int nmclan_probe(struct pcmcia_device *link) lp = netdev_priv(dev); lp->p_dev = link; link->priv = dev; - + spin_lock_init(&lp->bank_lock); link->resource[0]->end = 32; link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO; @@ -817,7 +815,6 @@ static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); snprintf(info->bus_info, sizeof(info->bus_info), "PCMCIA 0x%lx", dev->base_addr); } @@ -1110,7 +1107,7 @@ static int mace_rx(struct net_device *dev, unsigned char RxCnt) if (pkt_len & 1) *(skb_tail_pointer(skb) - 1) = inb(ioaddr + AM2150_RCV); skb->protocol = eth_type_trans(skb, dev); - + netif_rx(skb); /* Send the packet to the upper (protocol) layers. */ dev->stats.rx_packets++; diff --git a/drivers/net/ethernet/amd/pcnet32.c b/drivers/net/ethernet/amd/pcnet32.c index dc7d88227e76..07e8211eea51 100644 --- a/drivers/net/ethernet/amd/pcnet32.c +++ b/drivers/net/ethernet/amd/pcnet32.c @@ -24,13 +24,9 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define DRV_NAME "pcnet32" -#define DRV_VERSION "1.35" #define DRV_RELDATE "21.Apr.2008" #define PFX DRV_NAME ": " -static const char *const version = - DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " tsbogend@alpha.franken.de\n"; - #include <linux/module.h> #include <linux/kernel.h> #include <linux/sched.h> @@ -809,7 +805,6 @@ static void pcnet32_get_drvinfo(struct net_device *dev, struct pcnet32_private *lp = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); if (lp->pci_dev) strlcpy(info->bus_info, pci_name(lp->pci_dev), sizeof(info->bus_info)); @@ -3006,8 +3001,6 @@ MODULE_LICENSE("GPL"); static int __init pcnet32_init_module(void) { - pr_info("%s", version); - pcnet32_debug = netif_msg_init(debug, PCNET32_MSG_DEFAULT); if ((tx_start_pt >= 0) && (tx_start_pt <= 3)) diff --git a/drivers/net/ethernet/amd/sunlance.c b/drivers/net/ethernet/amd/sunlance.c index b00e00881253..a21b2e60157e 100644 --- a/drivers/net/ethernet/amd/sunlance.c +++ b/drivers/net/ethernet/amd/sunlance.c @@ -105,14 +105,9 @@ static char lancestr[] = "LANCE"; #include <asm/irq.h> #define DRV_NAME "sunlance" -#define DRV_VERSION "2.02" #define DRV_RELDATE "8/24/03" #define DRV_AUTHOR "Miguel de Icaza (miguel@nuclecu.unam.mx)" -static char version[] = - DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n"; - -MODULE_VERSION(DRV_VERSION); MODULE_AUTHOR(DRV_AUTHOR); MODULE_DESCRIPTION("Sun Lance ethernet driver"); MODULE_LICENSE("GPL"); @@ -1282,7 +1277,6 @@ static void lance_free_hwresources(struct lance_private *lp) static void sparc_lance_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, "sunlance", sizeof(info->driver)); - strlcpy(info->version, "2.02", sizeof(info->version)); } static const struct ethtool_ops sparc_lance_ethtool_ops = { @@ -1305,7 +1299,6 @@ static int sparc_lance_probe_one(struct platform_device *op, struct platform_device *lebuffer) { struct device_node *dp = op->dev.of_node; - static unsigned version_printed; struct lance_private *lp; struct net_device *dev; int i; @@ -1316,9 +1309,6 @@ static int sparc_lance_probe_one(struct platform_device *op, lp = netdev_priv(dev); - if (sparc_lance_debug && version_printed++ == 0) - printk (KERN_INFO "%s", version); - spin_lock_init(&lp->lock); /* Copy the IDPROM ethernet address to the device structure, later we diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c index 8083173f1a8f..61f39a0e04f9 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c @@ -405,7 +405,6 @@ static void xgbe_get_drvinfo(struct net_device *netdev, struct xgbe_hw_features *hw_feat = &pdata->hw_feat; strlcpy(drvinfo->driver, XGBE_DRV_NAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, XGBE_DRV_VERSION, sizeof(drvinfo->version)); strlcpy(drvinfo->bus_info, dev_name(pdata->dev), sizeof(drvinfo->bus_info)); snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%d.%d.%d", @@ -451,30 +450,6 @@ static int xgbe_set_coalesce(struct net_device *netdev, unsigned int rx_frames, rx_riwt, rx_usecs; unsigned int tx_frames; - /* Check for not supported parameters */ - if ((ec->rx_coalesce_usecs_irq) || - (ec->rx_max_coalesced_frames_irq) || - (ec->tx_coalesce_usecs) || - (ec->tx_coalesce_usecs_irq) || - (ec->tx_max_coalesced_frames_irq) || - (ec->stats_block_coalesce_usecs) || - (ec->use_adaptive_rx_coalesce) || - (ec->use_adaptive_tx_coalesce) || - (ec->pkt_rate_low) || - (ec->rx_coalesce_usecs_low) || - (ec->rx_max_coalesced_frames_low) || - (ec->tx_coalesce_usecs_low) || - (ec->tx_max_coalesced_frames_low) || - (ec->pkt_rate_high) || - (ec->rx_coalesce_usecs_high) || - (ec->rx_max_coalesced_frames_high) || - (ec->tx_coalesce_usecs_high) || - (ec->tx_max_coalesced_frames_high) || - (ec->rate_sample_interval)) { - netdev_err(netdev, "unsupported coalescing parameter\n"); - return -EOPNOTSUPP; - } - rx_riwt = hw_if->usec_to_riwt(pdata, ec->rx_coalesce_usecs); rx_usecs = ec->rx_coalesce_usecs; rx_frames = ec->rx_max_coalesced_frames; @@ -838,6 +813,8 @@ out: } static const struct ethtool_ops xgbe_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = xgbe_get_drvinfo, .get_msglevel = xgbe_get_msglevel, .set_msglevel = xgbe_set_msglevel, diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-main.c b/drivers/net/ethernet/amd/xgbe/xgbe-main.c index 7ce9c69e9c44..2a70714a791d 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-main.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-main.c @@ -127,7 +127,6 @@ MODULE_AUTHOR("Tom Lendacky <thomas.lendacky@amd.com>"); MODULE_LICENSE("Dual BSD/GPL"); -MODULE_VERSION(XGBE_DRV_VERSION); MODULE_DESCRIPTION(XGBE_DRV_DESC); static int debug = -1; diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h index 47bcbcf58048..5897e46faca5 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe.h @@ -135,7 +135,6 @@ #include <linux/list.h> #define XGBE_DRV_NAME "amd-xgbe" -#define XGBE_DRV_VERSION "1.0.3" #define XGBE_DRV_DESC "AMD 10 Gigabit Ethernet Driver" /* Descriptor related defines */ diff --git a/drivers/net/ethernet/apm/xgene-v2/ethtool.c b/drivers/net/ethernet/apm/xgene-v2/ethtool.c index a58250c1b57a..b78d1a99fe81 100644 --- a/drivers/net/ethernet/apm/xgene-v2/ethtool.c +++ b/drivers/net/ethernet/apm/xgene-v2/ethtool.c @@ -89,8 +89,6 @@ static void xge_get_drvinfo(struct net_device *ndev, struct platform_device *pdev = pdata->pdev; strcpy(info->driver, "xgene-enet-v2"); - strcpy(info->version, XGENE_ENET_V2_VERSION); - snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "N/A"); sprintf(info->bus_info, "%s", pdev->name); } diff --git a/drivers/net/ethernet/apm/xgene-v2/main.c b/drivers/net/ethernet/apm/xgene-v2/main.c index c48f60996761..860c18fb7aae 100644 --- a/drivers/net/ethernet/apm/xgene-v2/main.c +++ b/drivers/net/ethernet/apm/xgene-v2/main.c @@ -741,5 +741,4 @@ module_platform_driver(xge_driver); MODULE_DESCRIPTION("APM X-Gene SoC Ethernet v2 driver"); MODULE_AUTHOR("Iyappan Subramanian <isubramanian@apm.com>"); -MODULE_VERSION(XGENE_ENET_V2_VERSION); MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/apm/xgene-v2/main.h b/drivers/net/ethernet/apm/xgene-v2/main.h index d41439d2709d..b3985a7be59d 100644 --- a/drivers/net/ethernet/apm/xgene-v2/main.h +++ b/drivers/net/ethernet/apm/xgene-v2/main.h @@ -28,7 +28,6 @@ #include "ring.h" #include "ethtool.h" -#define XGENE_ENET_V2_VERSION "v1.0" #define XGENE_ENET_STD_MTU 1536 #define XGENE_ENET_MIN_FRAME 60 #define IRQ_ID_SIZE 16 diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c b/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c index 246dec27140d..ada70425b48c 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c @@ -103,8 +103,6 @@ static void xgene_get_drvinfo(struct net_device *ndev, struct platform_device *pdev = pdata->pdev; strcpy(info->driver, "xgene_enet"); - strcpy(info->version, XGENE_DRV_VERSION); - snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "N/A"); sprintf(info->bus_info, "%s", pdev->name); } diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index 6aee2f0fc0db..5f1fc6582d74 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -2179,7 +2179,6 @@ static struct platform_driver xgene_enet_driver = { module_platform_driver(xgene_enet_driver); MODULE_DESCRIPTION("APM X-Gene SoC Ethernet driver"); -MODULE_VERSION(XGENE_DRV_VERSION); MODULE_AUTHOR("Iyappan Subramanian <isubramanian@apm.com>"); MODULE_AUTHOR("Keyur Chudgar <kchudgar@apm.com>"); MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h index 18f4923b1723..d35a338120cf 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h @@ -28,7 +28,6 @@ #include "xgene_enet_ring2.h" #include "../../../phy/mdio-xgene.h" -#define XGENE_DRV_VERSION "v1.0" #define ETHER_MIN_PACKET 64 #define ETHER_STD_PACKET 1518 #define XGENE_ENET_STD_MTU 1536 diff --git a/drivers/net/ethernet/aquantia/Kconfig b/drivers/net/ethernet/aquantia/Kconfig index 350a48e4f124..76a44b2546ff 100644 --- a/drivers/net/ethernet/aquantia/Kconfig +++ b/drivers/net/ethernet/aquantia/Kconfig @@ -20,6 +20,7 @@ config AQTION tristate "aQuantia AQtion(tm) Support" depends on PCI depends on X86_64 || ARM64 || COMPILE_TEST + depends on MACSEC || MACSEC=n ---help--- This enables the support for the aQuantia AQtion(tm) Ethernet card. diff --git a/drivers/net/ethernet/aquantia/atlantic/Makefile b/drivers/net/ethernet/aquantia/atlantic/Makefile index 6e0a6e234483..8b555665a33a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/Makefile +++ b/drivers/net/ethernet/aquantia/atlantic/Makefile @@ -8,6 +8,8 @@ obj-$(CONFIG_AQTION) += atlantic.o +ccflags-y += -I$(src) + atlantic-objs := aq_main.o \ aq_nic.o \ aq_pci_func.o \ @@ -22,6 +24,9 @@ atlantic-objs := aq_main.o \ hw_atl/hw_atl_b0.o \ hw_atl/hw_atl_utils.o \ hw_atl/hw_atl_utils_fw2x.o \ - hw_atl/hw_atl_llh.o + hw_atl/hw_atl_llh.o \ + macsec/macsec_api.o + +atlantic-$(CONFIG_MACSEC) += aq_macsec.o atlantic-$(CONFIG_PTP_1588_CLOCK) += aq_ptp.o
\ No newline at end of file diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h b/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h index f0c41f7408e5..7560f5506e55 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h @@ -9,8 +9,6 @@ #ifndef AQ_CFG_H #define AQ_CFG_H -#include <generated/utsrelease.h> - #define AQ_CFG_VECS_DEF 8U #define AQ_CFG_TCS_DEF 1U @@ -85,7 +83,5 @@ #define AQ_CFG_DRV_AUTHOR "aQuantia" #define AQ_CFG_DRV_DESC "aQuantia Corporation(R) Network Driver" #define AQ_CFG_DRV_NAME "atlantic" -#define AQ_CFG_DRV_VERSION UTS_RELEASE \ - AQ_CFG_DRV_VERSION_SUFFIX #endif /* AQ_CFG_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_common.h b/drivers/net/ethernet/aquantia/atlantic/aq_common.h index 42ea8d8daa46..c8c402b013bb 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_common.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_common.h @@ -12,7 +12,6 @@ #include <linux/etherdevice.h> #include <linux/pci.h> #include <linux/if_vlan.h> -#include "ver.h" #include "aq_cfg.h" #include "aq_utils.h" diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c index 7b55633d2cb9..7241cf92b43a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c @@ -11,6 +11,7 @@ #include "aq_vec.h" #include "aq_ptp.h" #include "aq_filters.h" +#include "aq_macsec.h" #include <linux/ptp_clock_kernel.h> @@ -96,6 +97,62 @@ static const char aq_ethtool_queue_stat_names[][ETH_GSTRING_LEN] = { "Queue[%d] InErrors", }; +#if IS_ENABLED(CONFIG_MACSEC) +static const char aq_macsec_stat_names[][ETH_GSTRING_LEN] = { + "MACSec InCtlPackets", + "MACSec InTaggedMissPackets", + "MACSec InUntaggedMissPackets", + "MACSec InNotagPackets", + "MACSec InUntaggedPackets", + "MACSec InBadTagPackets", + "MACSec InNoSciPackets", + "MACSec InUnknownSciPackets", + "MACSec InCtrlPortPassPackets", + "MACSec InUnctrlPortPassPackets", + "MACSec InCtrlPortFailPackets", + "MACSec InUnctrlPortFailPackets", + "MACSec InTooLongPackets", + "MACSec InIgpocCtlPackets", + "MACSec InEccErrorPackets", + "MACSec InUnctrlHitDropRedir", + "MACSec OutCtlPackets", + "MACSec OutUnknownSaPackets", + "MACSec OutUntaggedPackets", + "MACSec OutTooLong", + "MACSec OutEccErrorPackets", + "MACSec OutUnctrlHitDropRedir", +}; + +static const char *aq_macsec_txsc_stat_names[] = { + "MACSecTXSC%d ProtectedPkts", + "MACSecTXSC%d EncryptedPkts", + "MACSecTXSC%d ProtectedOctets", + "MACSecTXSC%d EncryptedOctets", +}; + +static const char *aq_macsec_txsa_stat_names[] = { + "MACSecTXSC%dSA%d HitDropRedirect", + "MACSecTXSC%dSA%d Protected2Pkts", + "MACSecTXSC%dSA%d ProtectedPkts", + "MACSecTXSC%dSA%d EncryptedPkts", +}; + +static const char *aq_macsec_rxsa_stat_names[] = { + "MACSecRXSC%dSA%d UntaggedHitPkts", + "MACSecRXSC%dSA%d CtrlHitDrpRedir", + "MACSecRXSC%dSA%d NotUsingSa", + "MACSecRXSC%dSA%d UnusedSa", + "MACSecRXSC%dSA%d NotValidPkts", + "MACSecRXSC%dSA%d InvalidPkts", + "MACSecRXSC%dSA%d OkPkts", + "MACSecRXSC%dSA%d LatePkts", + "MACSecRXSC%dSA%d DelayedPkts", + "MACSecRXSC%dSA%d UncheckedPkts", + "MACSecRXSC%dSA%d ValidatedOctets", + "MACSecRXSC%dSA%d DecryptedOctets", +}; +#endif + static const char aq_ethtool_priv_flag_names[][ETH_GSTRING_LEN] = { "DMASystemLoopback", "PKTSystemLoopback", @@ -104,18 +161,38 @@ static const char aq_ethtool_priv_flag_names[][ETH_GSTRING_LEN] = { "PHYExternalLoopback", }; +static u32 aq_ethtool_n_stats(struct net_device *ndev) +{ + struct aq_nic_s *nic = netdev_priv(ndev); + struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(nic); + u32 n_stats = ARRAY_SIZE(aq_ethtool_stat_names) + + ARRAY_SIZE(aq_ethtool_queue_stat_names) * cfg->vecs; + +#if IS_ENABLED(CONFIG_MACSEC) + if (nic->macsec_cfg) { + n_stats += ARRAY_SIZE(aq_macsec_stat_names) + + ARRAY_SIZE(aq_macsec_txsc_stat_names) * + aq_macsec_tx_sc_cnt(nic) + + ARRAY_SIZE(aq_macsec_txsa_stat_names) * + aq_macsec_tx_sa_cnt(nic) + + ARRAY_SIZE(aq_macsec_rxsa_stat_names) * + aq_macsec_rx_sa_cnt(nic); + } +#endif + + return n_stats; +} + static void aq_ethtool_stats(struct net_device *ndev, struct ethtool_stats *stats, u64 *data) { struct aq_nic_s *aq_nic = netdev_priv(ndev); - struct aq_nic_cfg_s *cfg; - - cfg = aq_nic_get_cfg(aq_nic); - memset(data, 0, (ARRAY_SIZE(aq_ethtool_stat_names) + - ARRAY_SIZE(aq_ethtool_queue_stat_names) * - cfg->vecs) * sizeof(u64)); - aq_nic_get_stats(aq_nic, data); + memset(data, 0, aq_ethtool_n_stats(ndev) * sizeof(u64)); + data = aq_nic_get_stats(aq_nic, data); +#if IS_ENABLED(CONFIG_MACSEC) + data = aq_macsec_get_stats(aq_nic, data); +#endif } static void aq_ethtool_get_drvinfo(struct net_device *ndev, @@ -123,16 +200,13 @@ static void aq_ethtool_get_drvinfo(struct net_device *ndev, { struct pci_dev *pdev = to_pci_dev(ndev->dev.parent); struct aq_nic_s *aq_nic = netdev_priv(ndev); - struct aq_nic_cfg_s *cfg; u32 firmware_version; u32 regs_count; - cfg = aq_nic_get_cfg(aq_nic); firmware_version = aq_nic_get_fw_version(aq_nic); regs_count = aq_nic_get_regs_count(aq_nic); strlcat(drvinfo->driver, AQ_CFG_DRV_NAME, sizeof(drvinfo->driver)); - strlcat(drvinfo->version, AQ_CFG_DRV_VERSION, sizeof(drvinfo->version)); snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%u.%u.%u", firmware_version >> 24, @@ -140,8 +214,7 @@ static void aq_ethtool_get_drvinfo(struct net_device *ndev, strlcpy(drvinfo->bus_info, pdev ? pci_name(pdev) : "", sizeof(drvinfo->bus_info)); - drvinfo->n_stats = ARRAY_SIZE(aq_ethtool_stat_names) + - cfg->vecs * ARRAY_SIZE(aq_ethtool_queue_stat_names); + drvinfo->n_stats = aq_ethtool_n_stats(ndev); drvinfo->testinfo_len = 0; drvinfo->regdump_len = regs_count; drvinfo->eedump_len = 0; @@ -154,6 +227,9 @@ static void aq_ethtool_get_strings(struct net_device *ndev, struct aq_nic_cfg_s *cfg; u8 *p = data; int i, si; +#if IS_ENABLED(CONFIG_MACSEC) + int sa; +#endif cfg = aq_nic_get_cfg(aq_nic); @@ -171,6 +247,60 @@ static void aq_ethtool_get_strings(struct net_device *ndev, p += ETH_GSTRING_LEN; } } +#if IS_ENABLED(CONFIG_MACSEC) + if (!aq_nic->macsec_cfg) + break; + + memcpy(p, aq_macsec_stat_names, sizeof(aq_macsec_stat_names)); + p = p + sizeof(aq_macsec_stat_names); + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + struct aq_macsec_txsc *aq_txsc; + + if (!(test_bit(i, &aq_nic->macsec_cfg->txsc_idx_busy))) + continue; + + for (si = 0; + si < ARRAY_SIZE(aq_macsec_txsc_stat_names); + si++) { + snprintf(p, ETH_GSTRING_LEN, + aq_macsec_txsc_stat_names[si], i); + p += ETH_GSTRING_LEN; + } + aq_txsc = &aq_nic->macsec_cfg->aq_txsc[i]; + for (sa = 0; sa < MACSEC_NUM_AN; sa++) { + if (!(test_bit(sa, &aq_txsc->tx_sa_idx_busy))) + continue; + for (si = 0; + si < ARRAY_SIZE(aq_macsec_txsa_stat_names); + si++) { + snprintf(p, ETH_GSTRING_LEN, + aq_macsec_txsa_stat_names[si], + i, sa); + p += ETH_GSTRING_LEN; + } + } + } + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + struct aq_macsec_rxsc *aq_rxsc; + + if (!(test_bit(i, &aq_nic->macsec_cfg->rxsc_idx_busy))) + continue; + + aq_rxsc = &aq_nic->macsec_cfg->aq_rxsc[i]; + for (sa = 0; sa < MACSEC_NUM_AN; sa++) { + if (!(test_bit(sa, &aq_rxsc->rx_sa_idx_busy))) + continue; + for (si = 0; + si < ARRAY_SIZE(aq_macsec_rxsa_stat_names); + si++) { + snprintf(p, ETH_GSTRING_LEN, + aq_macsec_rxsa_stat_names[si], + i, sa); + p += ETH_GSTRING_LEN; + } + } + } +#endif break; case ETH_SS_PRIV_FLAGS: memcpy(p, aq_ethtool_priv_flag_names, @@ -210,16 +340,11 @@ static int aq_ethtool_set_phys_id(struct net_device *ndev, static int aq_ethtool_get_sset_count(struct net_device *ndev, int stringset) { - struct aq_nic_s *aq_nic = netdev_priv(ndev); - struct aq_nic_cfg_s *cfg; int ret = 0; - cfg = aq_nic_get_cfg(aq_nic); - switch (stringset) { case ETH_SS_STATS: - ret = ARRAY_SIZE(aq_ethtool_stat_names) + - cfg->vecs * ARRAY_SIZE(aq_ethtool_queue_stat_names); + ret = aq_ethtool_n_stats(ndev); break; case ETH_SS_PRIV_FLAGS: ret = ARRAY_SIZE(aq_ethtool_priv_flag_names); @@ -387,21 +512,10 @@ static int aq_ethtool_set_coalesce(struct net_device *ndev, cfg = aq_nic_get_cfg(aq_nic); - /* This is not yet supported - */ - if (coal->use_adaptive_rx_coalesce || coal->use_adaptive_tx_coalesce) - return -EOPNOTSUPP; - /* Atlantic only supports timing based coalescing */ if (coal->rx_max_coalesced_frames > 1 || - coal->rx_coalesce_usecs_irq || - coal->rx_max_coalesced_frames_irq) - return -EOPNOTSUPP; - - if (coal->tx_max_coalesced_frames > 1 || - coal->tx_coalesce_usecs_irq || - coal->tx_max_coalesced_frames_irq) + coal->tx_max_coalesced_frames > 1) return -EOPNOTSUPP; /* We do not support frame counting. Check this @@ -743,6 +857,8 @@ static int aq_ethtool_set_priv_flags(struct net_device *ndev, u32 flags) } const struct ethtool_ops aq_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_link = aq_ethtool_get_link, .get_regs_len = aq_ethtool_get_regs_len, .get_regs = aq_ethtool_get_regs, diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h index 251767c31f7e..7d71bc7dc500 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h @@ -343,6 +343,12 @@ struct aq_fw_ops { int (*get_eee_rate)(struct aq_hw_s *self, u32 *rate, u32 *supported_rates); + + u32 (*get_link_capabilities)(struct aq_hw_s *self); + + int (*send_macsec_req)(struct aq_hw_s *self, + struct macsec_msg_fw_request *msg, + struct macsec_msg_fw_response *resp); }; #endif /* AQ_HW_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c new file mode 100644 index 000000000000..0b3e234a54aa --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c @@ -0,0 +1,1777 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include "aq_macsec.h" +#include "aq_nic.h" +#include <linux/rtnetlink.h> + +#include "macsec/macsec_api.h" +#define AQ_MACSEC_KEY_LEN_128_BIT 16 +#define AQ_MACSEC_KEY_LEN_192_BIT 24 +#define AQ_MACSEC_KEY_LEN_256_BIT 32 + +enum aq_clear_type { + /* update HW configuration */ + AQ_CLEAR_HW = BIT(0), + /* update SW configuration (busy bits, pointers) */ + AQ_CLEAR_SW = BIT(1), + /* update both HW and SW configuration */ + AQ_CLEAR_ALL = AQ_CLEAR_HW | AQ_CLEAR_SW, +}; + +static int aq_clear_txsc(struct aq_nic_s *nic, const int txsc_idx, + enum aq_clear_type clear_type); +static int aq_clear_txsa(struct aq_nic_s *nic, struct aq_macsec_txsc *aq_txsc, + const int sa_num, enum aq_clear_type clear_type); +static int aq_clear_rxsc(struct aq_nic_s *nic, const int rxsc_idx, + enum aq_clear_type clear_type); +static int aq_clear_rxsa(struct aq_nic_s *nic, struct aq_macsec_rxsc *aq_rxsc, + const int sa_num, enum aq_clear_type clear_type); +static int aq_clear_secy(struct aq_nic_s *nic, const struct macsec_secy *secy, + enum aq_clear_type clear_type); +static int aq_apply_macsec_cfg(struct aq_nic_s *nic); +static int aq_apply_secy_cfg(struct aq_nic_s *nic, + const struct macsec_secy *secy); + +static void aq_ether_addr_to_mac(u32 mac[2], unsigned char *emac) +{ + u32 tmp[2] = { 0 }; + + memcpy(((u8 *)tmp) + 2, emac, ETH_ALEN); + + mac[0] = swab32(tmp[1]); + mac[1] = swab32(tmp[0]); +} + +/* There's a 1:1 mapping between SecY and TX SC */ +static int aq_get_txsc_idx_from_secy(struct aq_macsec_cfg *macsec_cfg, + const struct macsec_secy *secy) +{ + int i; + + if (unlikely(!secy)) + return -1; + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (macsec_cfg->aq_txsc[i].sw_secy == secy) + return i; + } + return -1; +} + +static int aq_get_rxsc_idx_from_rxsc(struct aq_macsec_cfg *macsec_cfg, + const struct macsec_rx_sc *rxsc) +{ + int i; + + if (unlikely(!rxsc)) + return -1; + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (macsec_cfg->aq_rxsc[i].sw_rxsc == rxsc) + return i; + } + + return -1; +} + +static int aq_get_txsc_idx_from_sc_idx(const enum aq_macsec_sc_sa sc_sa, + const int sc_idx) +{ + switch (sc_sa) { + case aq_macsec_sa_sc_4sa_8sc: + return sc_idx >> 2; + case aq_macsec_sa_sc_2sa_16sc: + return sc_idx >> 1; + case aq_macsec_sa_sc_1sa_32sc: + return sc_idx; + default: + WARN_ONCE(true, "Invalid sc_sa"); + } + return -1; +} + +/* Rotate keys u32[8] */ +static void aq_rotate_keys(u32 (*key)[8], const int key_len) +{ + u32 tmp[8] = { 0 }; + + memcpy(&tmp, key, sizeof(tmp)); + memset(*key, 0, sizeof(*key)); + + if (key_len == AQ_MACSEC_KEY_LEN_128_BIT) { + (*key)[0] = swab32(tmp[3]); + (*key)[1] = swab32(tmp[2]); + (*key)[2] = swab32(tmp[1]); + (*key)[3] = swab32(tmp[0]); + } else if (key_len == AQ_MACSEC_KEY_LEN_192_BIT) { + (*key)[0] = swab32(tmp[5]); + (*key)[1] = swab32(tmp[4]); + (*key)[2] = swab32(tmp[3]); + (*key)[3] = swab32(tmp[2]); + (*key)[4] = swab32(tmp[1]); + (*key)[5] = swab32(tmp[0]); + } else if (key_len == AQ_MACSEC_KEY_LEN_256_BIT) { + (*key)[0] = swab32(tmp[7]); + (*key)[1] = swab32(tmp[6]); + (*key)[2] = swab32(tmp[5]); + (*key)[3] = swab32(tmp[4]); + (*key)[4] = swab32(tmp[3]); + (*key)[5] = swab32(tmp[2]); + (*key)[6] = swab32(tmp[1]); + (*key)[7] = swab32(tmp[0]); + } else { + pr_warn("Rotate_keys: invalid key_len\n"); + } +} + +#define STATS_2x32_TO_64(stat_field) \ + (((u64)stat_field[1] << 32) | stat_field[0]) + +static int aq_get_macsec_common_stats(struct aq_hw_s *hw, + struct aq_macsec_common_stats *stats) +{ + struct aq_mss_ingress_common_counters ingress_counters; + struct aq_mss_egress_common_counters egress_counters; + int ret; + + /* MACSEC counters */ + ret = aq_mss_get_ingress_common_counters(hw, &ingress_counters); + if (unlikely(ret)) + return ret; + + stats->in.ctl_pkts = STATS_2x32_TO_64(ingress_counters.ctl_pkts); + stats->in.tagged_miss_pkts = + STATS_2x32_TO_64(ingress_counters.tagged_miss_pkts); + stats->in.untagged_miss_pkts = + STATS_2x32_TO_64(ingress_counters.untagged_miss_pkts); + stats->in.notag_pkts = STATS_2x32_TO_64(ingress_counters.notag_pkts); + stats->in.untagged_pkts = + STATS_2x32_TO_64(ingress_counters.untagged_pkts); + stats->in.bad_tag_pkts = + STATS_2x32_TO_64(ingress_counters.bad_tag_pkts); + stats->in.no_sci_pkts = STATS_2x32_TO_64(ingress_counters.no_sci_pkts); + stats->in.unknown_sci_pkts = + STATS_2x32_TO_64(ingress_counters.unknown_sci_pkts); + stats->in.ctrl_prt_pass_pkts = + STATS_2x32_TO_64(ingress_counters.ctrl_prt_pass_pkts); + stats->in.unctrl_prt_pass_pkts = + STATS_2x32_TO_64(ingress_counters.unctrl_prt_pass_pkts); + stats->in.ctrl_prt_fail_pkts = + STATS_2x32_TO_64(ingress_counters.ctrl_prt_fail_pkts); + stats->in.unctrl_prt_fail_pkts = + STATS_2x32_TO_64(ingress_counters.unctrl_prt_fail_pkts); + stats->in.too_long_pkts = + STATS_2x32_TO_64(ingress_counters.too_long_pkts); + stats->in.igpoc_ctl_pkts = + STATS_2x32_TO_64(ingress_counters.igpoc_ctl_pkts); + stats->in.ecc_error_pkts = + STATS_2x32_TO_64(ingress_counters.ecc_error_pkts); + stats->in.unctrl_hit_drop_redir = + STATS_2x32_TO_64(ingress_counters.unctrl_hit_drop_redir); + + ret = aq_mss_get_egress_common_counters(hw, &egress_counters); + if (unlikely(ret)) + return ret; + stats->out.ctl_pkts = STATS_2x32_TO_64(egress_counters.ctl_pkt); + stats->out.unknown_sa_pkts = + STATS_2x32_TO_64(egress_counters.unknown_sa_pkts); + stats->out.untagged_pkts = + STATS_2x32_TO_64(egress_counters.untagged_pkts); + stats->out.too_long = STATS_2x32_TO_64(egress_counters.too_long); + stats->out.ecc_error_pkts = + STATS_2x32_TO_64(egress_counters.ecc_error_pkts); + stats->out.unctrl_hit_drop_redir = + STATS_2x32_TO_64(egress_counters.unctrl_hit_drop_redir); + + return 0; +} + +static int aq_get_rxsa_stats(struct aq_hw_s *hw, const int sa_idx, + struct aq_macsec_rx_sa_stats *stats) +{ + struct aq_mss_ingress_sa_counters i_sa_counters; + int ret; + + ret = aq_mss_get_ingress_sa_counters(hw, &i_sa_counters, sa_idx); + if (unlikely(ret)) + return ret; + + stats->untagged_hit_pkts = + STATS_2x32_TO_64(i_sa_counters.untagged_hit_pkts); + stats->ctrl_hit_drop_redir_pkts = + STATS_2x32_TO_64(i_sa_counters.ctrl_hit_drop_redir_pkts); + stats->not_using_sa = STATS_2x32_TO_64(i_sa_counters.not_using_sa); + stats->unused_sa = STATS_2x32_TO_64(i_sa_counters.unused_sa); + stats->not_valid_pkts = STATS_2x32_TO_64(i_sa_counters.not_valid_pkts); + stats->invalid_pkts = STATS_2x32_TO_64(i_sa_counters.invalid_pkts); + stats->ok_pkts = STATS_2x32_TO_64(i_sa_counters.ok_pkts); + stats->late_pkts = STATS_2x32_TO_64(i_sa_counters.late_pkts); + stats->delayed_pkts = STATS_2x32_TO_64(i_sa_counters.delayed_pkts); + stats->unchecked_pkts = STATS_2x32_TO_64(i_sa_counters.unchecked_pkts); + stats->validated_octets = + STATS_2x32_TO_64(i_sa_counters.validated_octets); + stats->decrypted_octets = + STATS_2x32_TO_64(i_sa_counters.decrypted_octets); + + return 0; +} + +static int aq_get_txsa_stats(struct aq_hw_s *hw, const int sa_idx, + struct aq_macsec_tx_sa_stats *stats) +{ + struct aq_mss_egress_sa_counters e_sa_counters; + int ret; + + ret = aq_mss_get_egress_sa_counters(hw, &e_sa_counters, sa_idx); + if (unlikely(ret)) + return ret; + + stats->sa_hit_drop_redirect = + STATS_2x32_TO_64(e_sa_counters.sa_hit_drop_redirect); + stats->sa_protected2_pkts = + STATS_2x32_TO_64(e_sa_counters.sa_protected2_pkts); + stats->sa_protected_pkts = + STATS_2x32_TO_64(e_sa_counters.sa_protected_pkts); + stats->sa_encrypted_pkts = + STATS_2x32_TO_64(e_sa_counters.sa_encrypted_pkts); + + return 0; +} + +static int aq_get_txsa_next_pn(struct aq_hw_s *hw, const int sa_idx, u32 *pn) +{ + struct aq_mss_egress_sa_record sa_rec; + int ret; + + ret = aq_mss_get_egress_sa_record(hw, &sa_rec, sa_idx); + if (likely(!ret)) + *pn = sa_rec.next_pn; + + return ret; +} + +static int aq_get_rxsa_next_pn(struct aq_hw_s *hw, const int sa_idx, u32 *pn) +{ + struct aq_mss_ingress_sa_record sa_rec; + int ret; + + ret = aq_mss_get_ingress_sa_record(hw, &sa_rec, sa_idx); + if (likely(!ret)) + *pn = (!sa_rec.sat_nextpn) ? sa_rec.next_pn : 0; + + return ret; +} + +static int aq_get_txsc_stats(struct aq_hw_s *hw, const int sc_idx, + struct aq_macsec_tx_sc_stats *stats) +{ + struct aq_mss_egress_sc_counters e_sc_counters; + int ret; + + ret = aq_mss_get_egress_sc_counters(hw, &e_sc_counters, sc_idx); + if (unlikely(ret)) + return ret; + + stats->sc_protected_pkts = + STATS_2x32_TO_64(e_sc_counters.sc_protected_pkts); + stats->sc_encrypted_pkts = + STATS_2x32_TO_64(e_sc_counters.sc_encrypted_pkts); + stats->sc_protected_octets = + STATS_2x32_TO_64(e_sc_counters.sc_protected_octets); + stats->sc_encrypted_octets = + STATS_2x32_TO_64(e_sc_counters.sc_encrypted_octets); + + return 0; +} + +static int aq_mdo_dev_open(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + int ret = 0; + + if (ctx->prepare) + return 0; + + if (netif_carrier_ok(nic->ndev)) + ret = aq_apply_secy_cfg(nic, ctx->secy); + + return ret; +} + +static int aq_mdo_dev_stop(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + int i; + + if (ctx->prepare) + return 0; + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (nic->macsec_cfg->txsc_idx_busy & BIT(i)) + aq_clear_secy(nic, nic->macsec_cfg->aq_txsc[i].sw_secy, + AQ_CLEAR_HW); + } + + return 0; +} + +static int aq_set_txsc(struct aq_nic_s *nic, const int txsc_idx) +{ + struct aq_macsec_txsc *aq_txsc = &nic->macsec_cfg->aq_txsc[txsc_idx]; + struct aq_mss_egress_class_record tx_class_rec = { 0 }; + const struct macsec_secy *secy = aq_txsc->sw_secy; + struct aq_mss_egress_sc_record sc_rec = { 0 }; + unsigned int sc_idx = aq_txsc->hw_sc_idx; + struct aq_hw_s *hw = nic->aq_hw; + int ret = 0; + + aq_ether_addr_to_mac(tx_class_rec.mac_sa, secy->netdev->dev_addr); + + put_unaligned_be64((__force u64)secy->sci, tx_class_rec.sci); + tx_class_rec.sci_mask = 0; + + tx_class_rec.sa_mask = 0x3f; + + tx_class_rec.action = 0; /* forward to SA/SC table */ + tx_class_rec.valid = 1; + + tx_class_rec.sc_idx = sc_idx; + + tx_class_rec.sc_sa = nic->macsec_cfg->sc_sa; + + ret = aq_mss_set_egress_class_record(hw, &tx_class_rec, txsc_idx); + if (ret) + return ret; + + sc_rec.protect = secy->protect_frames; + if (secy->tx_sc.encrypt) + sc_rec.tci |= BIT(1); + if (secy->tx_sc.scb) + sc_rec.tci |= BIT(2); + if (secy->tx_sc.send_sci) + sc_rec.tci |= BIT(3); + if (secy->tx_sc.end_station) + sc_rec.tci |= BIT(4); + /* The C bit is clear if and only if the Secure Data is + * exactly the same as the User Data and the ICV is 16 octets long. + */ + if (!(secy->icv_len == 16 && !secy->tx_sc.encrypt)) + sc_rec.tci |= BIT(0); + + sc_rec.an_roll = 0; + + switch (secy->key_len) { + case AQ_MACSEC_KEY_LEN_128_BIT: + sc_rec.sak_len = 0; + break; + case AQ_MACSEC_KEY_LEN_192_BIT: + sc_rec.sak_len = 1; + break; + case AQ_MACSEC_KEY_LEN_256_BIT: + sc_rec.sak_len = 2; + break; + default: + WARN_ONCE(true, "Invalid sc_sa"); + return -EINVAL; + } + + sc_rec.curr_an = secy->tx_sc.encoding_sa; + sc_rec.valid = 1; + sc_rec.fresh = 1; + + return aq_mss_set_egress_sc_record(hw, &sc_rec, sc_idx); +} + +static u32 aq_sc_idx_max(const enum aq_macsec_sc_sa sc_sa) +{ + u32 result = 0; + + switch (sc_sa) { + case aq_macsec_sa_sc_4sa_8sc: + result = 8; + break; + case aq_macsec_sa_sc_2sa_16sc: + result = 16; + break; + case aq_macsec_sa_sc_1sa_32sc: + result = 32; + break; + default: + break; + }; + + return result; +} + +static u32 aq_to_hw_sc_idx(const u32 sc_idx, const enum aq_macsec_sc_sa sc_sa) +{ + switch (sc_sa) { + case aq_macsec_sa_sc_4sa_8sc: + return sc_idx << 2; + case aq_macsec_sa_sc_2sa_16sc: + return sc_idx << 1; + case aq_macsec_sa_sc_1sa_32sc: + return sc_idx; + default: + WARN_ONCE(true, "Invalid sc_sa"); + }; + + return sc_idx; +} + +static enum aq_macsec_sc_sa sc_sa_from_num_an(const int num_an) +{ + enum aq_macsec_sc_sa sc_sa = aq_macsec_sa_sc_not_used; + + switch (num_an) { + case 4: + sc_sa = aq_macsec_sa_sc_4sa_8sc; + break; + case 2: + sc_sa = aq_macsec_sa_sc_2sa_16sc; + break; + case 1: + sc_sa = aq_macsec_sa_sc_1sa_32sc; + break; + default: + break; + } + + return sc_sa; +} + +static int aq_mdo_add_secy(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + const struct macsec_secy *secy = ctx->secy; + enum aq_macsec_sc_sa sc_sa; + u32 txsc_idx; + int ret = 0; + + if (secy->xpn) + return -EOPNOTSUPP; + + sc_sa = sc_sa_from_num_an(MACSEC_NUM_AN); + if (sc_sa == aq_macsec_sa_sc_not_used) + return -EINVAL; + + if (hweight32(cfg->txsc_idx_busy) >= aq_sc_idx_max(sc_sa)) + return -ENOSPC; + + txsc_idx = ffz(cfg->txsc_idx_busy); + if (txsc_idx == AQ_MACSEC_MAX_SC) + return -ENOSPC; + + if (ctx->prepare) + return 0; + + cfg->sc_sa = sc_sa; + cfg->aq_txsc[txsc_idx].hw_sc_idx = aq_to_hw_sc_idx(txsc_idx, sc_sa); + cfg->aq_txsc[txsc_idx].sw_secy = secy; + + if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev)) + ret = aq_set_txsc(nic, txsc_idx); + + set_bit(txsc_idx, &cfg->txsc_idx_busy); + + return 0; +} + +static int aq_mdo_upd_secy(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + const struct macsec_secy *secy = ctx->secy; + int txsc_idx; + int ret = 0; + + txsc_idx = aq_get_txsc_idx_from_secy(nic->macsec_cfg, secy); + if (txsc_idx < 0) + return -ENOENT; + + if (ctx->prepare) + return 0; + + if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev)) + ret = aq_set_txsc(nic, txsc_idx); + + return ret; +} + +static int aq_clear_txsc(struct aq_nic_s *nic, const int txsc_idx, + enum aq_clear_type clear_type) +{ + struct aq_macsec_txsc *tx_sc = &nic->macsec_cfg->aq_txsc[txsc_idx]; + struct aq_mss_egress_class_record tx_class_rec = { 0 }; + struct aq_mss_egress_sc_record sc_rec = { 0 }; + struct aq_hw_s *hw = nic->aq_hw; + int ret = 0; + int sa_num; + + for_each_set_bit (sa_num, &tx_sc->tx_sa_idx_busy, AQ_MACSEC_MAX_SA) { + ret = aq_clear_txsa(nic, tx_sc, sa_num, clear_type); + if (ret) + return ret; + } + + if (clear_type & AQ_CLEAR_HW) { + ret = aq_mss_set_egress_class_record(hw, &tx_class_rec, + txsc_idx); + if (ret) + return ret; + + sc_rec.fresh = 1; + ret = aq_mss_set_egress_sc_record(hw, &sc_rec, + tx_sc->hw_sc_idx); + if (ret) + return ret; + } + + if (clear_type & AQ_CLEAR_SW) { + clear_bit(txsc_idx, &nic->macsec_cfg->txsc_idx_busy); + nic->macsec_cfg->aq_txsc[txsc_idx].sw_secy = NULL; + } + + return ret; +} + +static int aq_mdo_del_secy(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + int ret = 0; + + if (ctx->prepare) + return 0; + + if (!nic->macsec_cfg) + return 0; + + ret = aq_clear_secy(nic, ctx->secy, AQ_CLEAR_ALL); + + return ret; +} + +static int aq_update_txsa(struct aq_nic_s *nic, const unsigned int sc_idx, + const struct macsec_secy *secy, + const struct macsec_tx_sa *tx_sa, + const unsigned char *key, const unsigned char an) +{ + const u32 next_pn = tx_sa->next_pn_halves.lower; + struct aq_mss_egress_sakey_record key_rec; + const unsigned int sa_idx = sc_idx | an; + struct aq_mss_egress_sa_record sa_rec; + struct aq_hw_s *hw = nic->aq_hw; + int ret = 0; + + memset(&sa_rec, 0, sizeof(sa_rec)); + sa_rec.valid = tx_sa->active; + sa_rec.fresh = 1; + sa_rec.next_pn = next_pn; + + ret = aq_mss_set_egress_sa_record(hw, &sa_rec, sa_idx); + if (ret) + return ret; + + if (!key) + return ret; + + memset(&key_rec, 0, sizeof(key_rec)); + memcpy(&key_rec.key, key, secy->key_len); + + aq_rotate_keys(&key_rec.key, secy->key_len); + + ret = aq_mss_set_egress_sakey_record(hw, &key_rec, sa_idx); + + return ret; +} + +static int aq_mdo_add_txsa(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + const struct macsec_secy *secy = ctx->secy; + struct aq_macsec_txsc *aq_txsc; + int txsc_idx; + int ret = 0; + + txsc_idx = aq_get_txsc_idx_from_secy(cfg, secy); + if (txsc_idx < 0) + return -EINVAL; + + if (ctx->prepare) + return 0; + + aq_txsc = &cfg->aq_txsc[txsc_idx]; + set_bit(ctx->sa.assoc_num, &aq_txsc->tx_sa_idx_busy); + + memcpy(aq_txsc->tx_sa_key[ctx->sa.assoc_num], ctx->sa.key, + secy->key_len); + + if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev)) + ret = aq_update_txsa(nic, aq_txsc->hw_sc_idx, secy, + ctx->sa.tx_sa, ctx->sa.key, + ctx->sa.assoc_num); + + return ret; +} + +static int aq_mdo_upd_txsa(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + const struct macsec_secy *secy = ctx->secy; + struct aq_macsec_txsc *aq_txsc; + int txsc_idx; + int ret = 0; + + txsc_idx = aq_get_txsc_idx_from_secy(cfg, secy); + if (txsc_idx < 0) + return -EINVAL; + + if (ctx->prepare) + return 0; + + aq_txsc = &cfg->aq_txsc[txsc_idx]; + if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev)) + ret = aq_update_txsa(nic, aq_txsc->hw_sc_idx, secy, + ctx->sa.tx_sa, NULL, ctx->sa.assoc_num); + + return ret; +} + +static int aq_clear_txsa(struct aq_nic_s *nic, struct aq_macsec_txsc *aq_txsc, + const int sa_num, enum aq_clear_type clear_type) +{ + const int sa_idx = aq_txsc->hw_sc_idx | sa_num; + struct aq_hw_s *hw = nic->aq_hw; + int ret = 0; + + if (clear_type & AQ_CLEAR_SW) + clear_bit(sa_num, &aq_txsc->tx_sa_idx_busy); + + if ((clear_type & AQ_CLEAR_HW) && netif_carrier_ok(nic->ndev)) { + struct aq_mss_egress_sakey_record key_rec; + struct aq_mss_egress_sa_record sa_rec; + + memset(&sa_rec, 0, sizeof(sa_rec)); + sa_rec.fresh = 1; + + ret = aq_mss_set_egress_sa_record(hw, &sa_rec, sa_idx); + if (ret) + return ret; + + memset(&key_rec, 0, sizeof(key_rec)); + return aq_mss_set_egress_sakey_record(hw, &key_rec, sa_idx); + } + + return 0; +} + +static int aq_mdo_del_txsa(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + int txsc_idx; + int ret = 0; + + txsc_idx = aq_get_txsc_idx_from_secy(cfg, ctx->secy); + if (txsc_idx < 0) + return -EINVAL; + + if (ctx->prepare) + return 0; + + ret = aq_clear_txsa(nic, &cfg->aq_txsc[txsc_idx], ctx->sa.assoc_num, + AQ_CLEAR_ALL); + + return ret; +} + +static int aq_rxsc_validate_frames(const enum macsec_validation_type validate) +{ + switch (validate) { + case MACSEC_VALIDATE_DISABLED: + return 2; + case MACSEC_VALIDATE_CHECK: + return 1; + case MACSEC_VALIDATE_STRICT: + return 0; + default: + WARN_ONCE(true, "Invalid validation type"); + } + + return 0; +} + +static int aq_set_rxsc(struct aq_nic_s *nic, const u32 rxsc_idx) +{ + const struct aq_macsec_rxsc *aq_rxsc = + &nic->macsec_cfg->aq_rxsc[rxsc_idx]; + struct aq_mss_ingress_preclass_record pre_class_record; + const struct macsec_rx_sc *rx_sc = aq_rxsc->sw_rxsc; + const struct macsec_secy *secy = aq_rxsc->sw_secy; + const u32 hw_sc_idx = aq_rxsc->hw_sc_idx; + struct aq_mss_ingress_sc_record sc_record; + struct aq_hw_s *hw = nic->aq_hw; + int ret = 0; + + memset(&pre_class_record, 0, sizeof(pre_class_record)); + put_unaligned_be64((__force u64)rx_sc->sci, pre_class_record.sci); + pre_class_record.sci_mask = 0xff; + /* match all MACSEC ethertype packets */ + pre_class_record.eth_type = ETH_P_MACSEC; + pre_class_record.eth_type_mask = 0x3; + + aq_ether_addr_to_mac(pre_class_record.mac_sa, (char *)&rx_sc->sci); + pre_class_record.sa_mask = 0x3f; + + pre_class_record.an_mask = nic->macsec_cfg->sc_sa; + pre_class_record.sc_idx = hw_sc_idx; + /* strip SecTAG & forward for decryption */ + pre_class_record.action = 0x0; + pre_class_record.valid = 1; + + ret = aq_mss_set_ingress_preclass_record(hw, &pre_class_record, + 2 * rxsc_idx + 1); + if (ret) + return ret; + + /* If SCI is absent, then match by SA alone */ + pre_class_record.sci_mask = 0; + pre_class_record.sci_from_table = 1; + + ret = aq_mss_set_ingress_preclass_record(hw, &pre_class_record, + 2 * rxsc_idx); + if (ret) + return ret; + + memset(&sc_record, 0, sizeof(sc_record)); + sc_record.validate_frames = + aq_rxsc_validate_frames(secy->validate_frames); + if (secy->replay_protect) { + sc_record.replay_protect = 1; + sc_record.anti_replay_window = secy->replay_window; + } + sc_record.valid = 1; + sc_record.fresh = 1; + + ret = aq_mss_set_ingress_sc_record(hw, &sc_record, hw_sc_idx); + if (ret) + return ret; + + return ret; +} + +static int aq_mdo_add_rxsc(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + const u32 rxsc_idx_max = aq_sc_idx_max(cfg->sc_sa); + u32 rxsc_idx; + int ret = 0; + + if (hweight32(cfg->rxsc_idx_busy) >= rxsc_idx_max) + return -ENOSPC; + + rxsc_idx = ffz(cfg->rxsc_idx_busy); + if (rxsc_idx >= rxsc_idx_max) + return -ENOSPC; + + if (ctx->prepare) + return 0; + + cfg->aq_rxsc[rxsc_idx].hw_sc_idx = aq_to_hw_sc_idx(rxsc_idx, + cfg->sc_sa); + cfg->aq_rxsc[rxsc_idx].sw_secy = ctx->secy; + cfg->aq_rxsc[rxsc_idx].sw_rxsc = ctx->rx_sc; + + if (netif_carrier_ok(nic->ndev) && netif_running(ctx->secy->netdev)) + ret = aq_set_rxsc(nic, rxsc_idx); + + if (ret < 0) + return ret; + + set_bit(rxsc_idx, &cfg->rxsc_idx_busy); + + return 0; +} + +static int aq_mdo_upd_rxsc(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + int rxsc_idx; + int ret = 0; + + rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, ctx->rx_sc); + if (rxsc_idx < 0) + return -ENOENT; + + if (ctx->prepare) + return 0; + + if (netif_carrier_ok(nic->ndev) && netif_running(ctx->secy->netdev)) + ret = aq_set_rxsc(nic, rxsc_idx); + + return ret; +} + +static int aq_clear_rxsc(struct aq_nic_s *nic, const int rxsc_idx, + enum aq_clear_type clear_type) +{ + struct aq_macsec_rxsc *rx_sc = &nic->macsec_cfg->aq_rxsc[rxsc_idx]; + struct aq_hw_s *hw = nic->aq_hw; + int ret = 0; + int sa_num; + + for_each_set_bit (sa_num, &rx_sc->rx_sa_idx_busy, AQ_MACSEC_MAX_SA) { + ret = aq_clear_rxsa(nic, rx_sc, sa_num, clear_type); + if (ret) + return ret; + } + + if (clear_type & AQ_CLEAR_HW) { + struct aq_mss_ingress_preclass_record pre_class_record; + struct aq_mss_ingress_sc_record sc_record; + + memset(&pre_class_record, 0, sizeof(pre_class_record)); + memset(&sc_record, 0, sizeof(sc_record)); + + ret = aq_mss_set_ingress_preclass_record(hw, &pre_class_record, + 2 * rxsc_idx); + if (ret) + return ret; + + ret = aq_mss_set_ingress_preclass_record(hw, &pre_class_record, + 2 * rxsc_idx + 1); + if (ret) + return ret; + + sc_record.fresh = 1; + ret = aq_mss_set_ingress_sc_record(hw, &sc_record, + rx_sc->hw_sc_idx); + if (ret) + return ret; + } + + if (clear_type & AQ_CLEAR_SW) { + clear_bit(rxsc_idx, &nic->macsec_cfg->rxsc_idx_busy); + rx_sc->sw_secy = NULL; + rx_sc->sw_rxsc = NULL; + } + + return ret; +} + +static int aq_mdo_del_rxsc(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + enum aq_clear_type clear_type = AQ_CLEAR_SW; + int rxsc_idx; + int ret = 0; + + rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, ctx->rx_sc); + if (rxsc_idx < 0) + return -ENOENT; + + if (ctx->prepare) + return 0; + + if (netif_carrier_ok(nic->ndev)) + clear_type = AQ_CLEAR_ALL; + + ret = aq_clear_rxsc(nic, rxsc_idx, clear_type); + + return ret; +} + +static int aq_update_rxsa(struct aq_nic_s *nic, const unsigned int sc_idx, + const struct macsec_secy *secy, + const struct macsec_rx_sa *rx_sa, + const unsigned char *key, const unsigned char an) +{ + struct aq_mss_ingress_sakey_record sa_key_record; + const u32 next_pn = rx_sa->next_pn_halves.lower; + struct aq_mss_ingress_sa_record sa_record; + struct aq_hw_s *hw = nic->aq_hw; + const int sa_idx = sc_idx | an; + int ret = 0; + + memset(&sa_record, 0, sizeof(sa_record)); + sa_record.valid = rx_sa->active; + sa_record.fresh = 1; + sa_record.next_pn = next_pn; + + ret = aq_mss_set_ingress_sa_record(hw, &sa_record, sa_idx); + if (ret) + return ret; + + if (!key) + return ret; + + memset(&sa_key_record, 0, sizeof(sa_key_record)); + memcpy(&sa_key_record.key, key, secy->key_len); + + switch (secy->key_len) { + case AQ_MACSEC_KEY_LEN_128_BIT: + sa_key_record.key_len = 0; + break; + case AQ_MACSEC_KEY_LEN_192_BIT: + sa_key_record.key_len = 1; + break; + case AQ_MACSEC_KEY_LEN_256_BIT: + sa_key_record.key_len = 2; + break; + default: + return -1; + } + + aq_rotate_keys(&sa_key_record.key, secy->key_len); + + ret = aq_mss_set_ingress_sakey_record(hw, &sa_key_record, sa_idx); + + return ret; +} + +static int aq_mdo_add_rxsa(struct macsec_context *ctx) +{ + const struct macsec_rx_sc *rx_sc = ctx->sa.rx_sa->sc; + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + const struct macsec_secy *secy = ctx->secy; + struct aq_macsec_rxsc *aq_rxsc; + int rxsc_idx; + int ret = 0; + + rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, rx_sc); + if (rxsc_idx < 0) + return -EINVAL; + + if (ctx->prepare) + return 0; + + aq_rxsc = &nic->macsec_cfg->aq_rxsc[rxsc_idx]; + set_bit(ctx->sa.assoc_num, &aq_rxsc->rx_sa_idx_busy); + + memcpy(aq_rxsc->rx_sa_key[ctx->sa.assoc_num], ctx->sa.key, + secy->key_len); + + if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev)) + ret = aq_update_rxsa(nic, aq_rxsc->hw_sc_idx, secy, + ctx->sa.rx_sa, ctx->sa.key, + ctx->sa.assoc_num); + + return ret; +} + +static int aq_mdo_upd_rxsa(struct macsec_context *ctx) +{ + const struct macsec_rx_sc *rx_sc = ctx->sa.rx_sa->sc; + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + const struct macsec_secy *secy = ctx->secy; + int rxsc_idx; + int ret = 0; + + rxsc_idx = aq_get_rxsc_idx_from_rxsc(cfg, rx_sc); + if (rxsc_idx < 0) + return -EINVAL; + + if (ctx->prepare) + return 0; + + if (netif_carrier_ok(nic->ndev) && netif_running(secy->netdev)) + ret = aq_update_rxsa(nic, cfg->aq_rxsc[rxsc_idx].hw_sc_idx, + secy, ctx->sa.rx_sa, NULL, + ctx->sa.assoc_num); + + return ret; +} + +static int aq_clear_rxsa(struct aq_nic_s *nic, struct aq_macsec_rxsc *aq_rxsc, + const int sa_num, enum aq_clear_type clear_type) +{ + int sa_idx = aq_rxsc->hw_sc_idx | sa_num; + struct aq_hw_s *hw = nic->aq_hw; + int ret = 0; + + if (clear_type & AQ_CLEAR_SW) + clear_bit(sa_num, &aq_rxsc->rx_sa_idx_busy); + + if ((clear_type & AQ_CLEAR_HW) && netif_carrier_ok(nic->ndev)) { + struct aq_mss_ingress_sakey_record sa_key_record; + struct aq_mss_ingress_sa_record sa_record; + + memset(&sa_key_record, 0, sizeof(sa_key_record)); + memset(&sa_record, 0, sizeof(sa_record)); + sa_record.fresh = 1; + ret = aq_mss_set_ingress_sa_record(hw, &sa_record, sa_idx); + if (ret) + return ret; + + return aq_mss_set_ingress_sakey_record(hw, &sa_key_record, + sa_idx); + } + + return ret; +} + +static int aq_mdo_del_rxsa(struct macsec_context *ctx) +{ + const struct macsec_rx_sc *rx_sc = ctx->sa.rx_sa->sc; + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + int rxsc_idx; + int ret = 0; + + rxsc_idx = aq_get_rxsc_idx_from_rxsc(cfg, rx_sc); + if (rxsc_idx < 0) + return -EINVAL; + + if (ctx->prepare) + return 0; + + ret = aq_clear_rxsa(nic, &cfg->aq_rxsc[rxsc_idx], ctx->sa.assoc_num, + AQ_CLEAR_ALL); + + return ret; +} + +static int aq_mdo_get_dev_stats(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_common_stats *stats = &nic->macsec_cfg->stats; + struct aq_hw_s *hw = nic->aq_hw; + + if (ctx->prepare) + return 0; + + aq_get_macsec_common_stats(hw, stats); + + ctx->stats.dev_stats->OutPktsUntagged = stats->out.untagged_pkts; + ctx->stats.dev_stats->InPktsUntagged = stats->in.untagged_pkts; + ctx->stats.dev_stats->OutPktsTooLong = stats->out.too_long; + ctx->stats.dev_stats->InPktsNoTag = stats->in.notag_pkts; + ctx->stats.dev_stats->InPktsBadTag = stats->in.bad_tag_pkts; + ctx->stats.dev_stats->InPktsUnknownSCI = stats->in.unknown_sci_pkts; + ctx->stats.dev_stats->InPktsNoSCI = stats->in.no_sci_pkts; + ctx->stats.dev_stats->InPktsOverrun = 0; + + return 0; +} + +static int aq_mdo_get_tx_sc_stats(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_tx_sc_stats *stats; + struct aq_hw_s *hw = nic->aq_hw; + struct aq_macsec_txsc *aq_txsc; + int txsc_idx; + + txsc_idx = aq_get_txsc_idx_from_secy(nic->macsec_cfg, ctx->secy); + if (txsc_idx < 0) + return -ENOENT; + + if (ctx->prepare) + return 0; + + aq_txsc = &nic->macsec_cfg->aq_txsc[txsc_idx]; + stats = &aq_txsc->stats; + aq_get_txsc_stats(hw, aq_txsc->hw_sc_idx, stats); + + ctx->stats.tx_sc_stats->OutPktsProtected = stats->sc_protected_pkts; + ctx->stats.tx_sc_stats->OutPktsEncrypted = stats->sc_encrypted_pkts; + ctx->stats.tx_sc_stats->OutOctetsProtected = stats->sc_protected_octets; + ctx->stats.tx_sc_stats->OutOctetsEncrypted = stats->sc_encrypted_octets; + + return 0; +} + +static int aq_mdo_get_tx_sa_stats(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + struct aq_macsec_tx_sa_stats *stats; + struct aq_hw_s *hw = nic->aq_hw; + const struct macsec_secy *secy; + struct aq_macsec_txsc *aq_txsc; + struct macsec_tx_sa *tx_sa; + unsigned int sa_idx; + int txsc_idx; + u32 next_pn; + int ret; + + txsc_idx = aq_get_txsc_idx_from_secy(cfg, ctx->secy); + if (txsc_idx < 0) + return -EINVAL; + + if (ctx->prepare) + return 0; + + aq_txsc = &cfg->aq_txsc[txsc_idx]; + sa_idx = aq_txsc->hw_sc_idx | ctx->sa.assoc_num; + stats = &aq_txsc->tx_sa_stats[ctx->sa.assoc_num]; + ret = aq_get_txsa_stats(hw, sa_idx, stats); + if (ret) + return ret; + + ctx->stats.tx_sa_stats->OutPktsProtected = stats->sa_protected_pkts; + ctx->stats.tx_sa_stats->OutPktsEncrypted = stats->sa_encrypted_pkts; + + secy = aq_txsc->sw_secy; + tx_sa = rcu_dereference_bh(secy->tx_sc.sa[ctx->sa.assoc_num]); + ret = aq_get_txsa_next_pn(hw, sa_idx, &next_pn); + if (ret == 0) { + spin_lock_bh(&tx_sa->lock); + tx_sa->next_pn = next_pn; + spin_unlock_bh(&tx_sa->lock); + } + + return ret; +} + +static int aq_mdo_get_rx_sc_stats(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + struct aq_macsec_rx_sa_stats *stats; + struct aq_hw_s *hw = nic->aq_hw; + struct aq_macsec_rxsc *aq_rxsc; + unsigned int sa_idx; + int rxsc_idx; + int ret = 0; + int i; + + rxsc_idx = aq_get_rxsc_idx_from_rxsc(cfg, ctx->rx_sc); + if (rxsc_idx < 0) + return -ENOENT; + + if (ctx->prepare) + return 0; + + aq_rxsc = &cfg->aq_rxsc[rxsc_idx]; + for (i = 0; i < MACSEC_NUM_AN; i++) { + if (!test_bit(i, &aq_rxsc->rx_sa_idx_busy)) + continue; + + stats = &aq_rxsc->rx_sa_stats[i]; + sa_idx = aq_rxsc->hw_sc_idx | i; + ret = aq_get_rxsa_stats(hw, sa_idx, stats); + if (ret) + break; + + ctx->stats.rx_sc_stats->InOctetsValidated += + stats->validated_octets; + ctx->stats.rx_sc_stats->InOctetsDecrypted += + stats->decrypted_octets; + ctx->stats.rx_sc_stats->InPktsUnchecked += + stats->unchecked_pkts; + ctx->stats.rx_sc_stats->InPktsDelayed += stats->delayed_pkts; + ctx->stats.rx_sc_stats->InPktsOK += stats->ok_pkts; + ctx->stats.rx_sc_stats->InPktsInvalid += stats->invalid_pkts; + ctx->stats.rx_sc_stats->InPktsLate += stats->late_pkts; + ctx->stats.rx_sc_stats->InPktsNotValid += stats->not_valid_pkts; + ctx->stats.rx_sc_stats->InPktsNotUsingSA += stats->not_using_sa; + ctx->stats.rx_sc_stats->InPktsUnusedSA += stats->unused_sa; + } + + return ret; +} + +static int aq_mdo_get_rx_sa_stats(struct macsec_context *ctx) +{ + struct aq_nic_s *nic = netdev_priv(ctx->netdev); + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + struct aq_macsec_rx_sa_stats *stats; + struct aq_hw_s *hw = nic->aq_hw; + struct aq_macsec_rxsc *aq_rxsc; + struct macsec_rx_sa *rx_sa; + unsigned int sa_idx; + int rxsc_idx; + u32 next_pn; + int ret; + + rxsc_idx = aq_get_rxsc_idx_from_rxsc(cfg, ctx->rx_sc); + if (rxsc_idx < 0) + return -EINVAL; + + if (ctx->prepare) + return 0; + + aq_rxsc = &cfg->aq_rxsc[rxsc_idx]; + stats = &aq_rxsc->rx_sa_stats[ctx->sa.assoc_num]; + sa_idx = aq_rxsc->hw_sc_idx | ctx->sa.assoc_num; + ret = aq_get_rxsa_stats(hw, sa_idx, stats); + if (ret) + return ret; + + ctx->stats.rx_sa_stats->InPktsOK = stats->ok_pkts; + ctx->stats.rx_sa_stats->InPktsInvalid = stats->invalid_pkts; + ctx->stats.rx_sa_stats->InPktsNotValid = stats->not_valid_pkts; + ctx->stats.rx_sa_stats->InPktsNotUsingSA = stats->not_using_sa; + ctx->stats.rx_sa_stats->InPktsUnusedSA = stats->unused_sa; + + rx_sa = rcu_dereference_bh(aq_rxsc->sw_rxsc->sa[ctx->sa.assoc_num]); + ret = aq_get_rxsa_next_pn(hw, sa_idx, &next_pn); + if (ret == 0) { + spin_lock_bh(&rx_sa->lock); + rx_sa->next_pn = next_pn; + spin_unlock_bh(&rx_sa->lock); + } + + return ret; +} + +static int apply_txsc_cfg(struct aq_nic_s *nic, const int txsc_idx) +{ + struct aq_macsec_txsc *aq_txsc = &nic->macsec_cfg->aq_txsc[txsc_idx]; + const struct macsec_secy *secy = aq_txsc->sw_secy; + struct macsec_tx_sa *tx_sa; + int ret = 0; + int i; + + if (!netif_running(secy->netdev)) + return ret; + + ret = aq_set_txsc(nic, txsc_idx); + if (ret) + return ret; + + for (i = 0; i < MACSEC_NUM_AN; i++) { + tx_sa = rcu_dereference_bh(secy->tx_sc.sa[i]); + if (tx_sa) { + ret = aq_update_txsa(nic, aq_txsc->hw_sc_idx, secy, + tx_sa, aq_txsc->tx_sa_key[i], i); + if (ret) + return ret; + } + } + + return ret; +} + +static int apply_rxsc_cfg(struct aq_nic_s *nic, const int rxsc_idx) +{ + struct aq_macsec_rxsc *aq_rxsc = &nic->macsec_cfg->aq_rxsc[rxsc_idx]; + const struct macsec_secy *secy = aq_rxsc->sw_secy; + struct macsec_rx_sa *rx_sa; + int ret = 0; + int i; + + if (!netif_running(secy->netdev)) + return ret; + + ret = aq_set_rxsc(nic, rxsc_idx); + if (ret) + return ret; + + for (i = 0; i < MACSEC_NUM_AN; i++) { + rx_sa = rcu_dereference_bh(aq_rxsc->sw_rxsc->sa[i]); + if (rx_sa) { + ret = aq_update_rxsa(nic, aq_rxsc->hw_sc_idx, secy, + rx_sa, aq_rxsc->rx_sa_key[i], i); + if (ret) + return ret; + } + } + + return ret; +} + +static int aq_clear_secy(struct aq_nic_s *nic, const struct macsec_secy *secy, + enum aq_clear_type clear_type) +{ + struct macsec_rx_sc *rx_sc; + int txsc_idx; + int rxsc_idx; + int ret = 0; + + txsc_idx = aq_get_txsc_idx_from_secy(nic->macsec_cfg, secy); + if (txsc_idx >= 0) { + ret = aq_clear_txsc(nic, txsc_idx, clear_type); + if (ret) + return ret; + } + + for (rx_sc = rcu_dereference_bh(secy->rx_sc); rx_sc; + rx_sc = rcu_dereference_bh(rx_sc->next)) { + rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, rx_sc); + if (rxsc_idx < 0) + continue; + + ret = aq_clear_rxsc(nic, rxsc_idx, clear_type); + if (ret) + return ret; + } + + return ret; +} + +static int aq_apply_secy_cfg(struct aq_nic_s *nic, + const struct macsec_secy *secy) +{ + struct macsec_rx_sc *rx_sc; + int txsc_idx; + int rxsc_idx; + int ret = 0; + + txsc_idx = aq_get_txsc_idx_from_secy(nic->macsec_cfg, secy); + if (txsc_idx >= 0) + apply_txsc_cfg(nic, txsc_idx); + + for (rx_sc = rcu_dereference_bh(secy->rx_sc); rx_sc && rx_sc->active; + rx_sc = rcu_dereference_bh(rx_sc->next)) { + rxsc_idx = aq_get_rxsc_idx_from_rxsc(nic->macsec_cfg, rx_sc); + if (unlikely(rxsc_idx < 0)) + continue; + + ret = apply_rxsc_cfg(nic, rxsc_idx); + if (ret) + return ret; + } + + return ret; +} + +static int aq_apply_macsec_cfg(struct aq_nic_s *nic) +{ + int ret = 0; + int i; + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (nic->macsec_cfg->txsc_idx_busy & BIT(i)) { + ret = apply_txsc_cfg(nic, i); + if (ret) + return ret; + } + } + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (nic->macsec_cfg->rxsc_idx_busy & BIT(i)) { + ret = apply_rxsc_cfg(nic, i); + if (ret) + return ret; + } + } + + return ret; +} + +static int aq_sa_from_sa_idx(const enum aq_macsec_sc_sa sc_sa, const int sa_idx) +{ + switch (sc_sa) { + case aq_macsec_sa_sc_4sa_8sc: + return sa_idx & 3; + case aq_macsec_sa_sc_2sa_16sc: + return sa_idx & 1; + case aq_macsec_sa_sc_1sa_32sc: + return 0; + default: + WARN_ONCE(true, "Invalid sc_sa"); + } + return -EINVAL; +} + +static int aq_sc_idx_from_sa_idx(const enum aq_macsec_sc_sa sc_sa, + const int sa_idx) +{ + switch (sc_sa) { + case aq_macsec_sa_sc_4sa_8sc: + return sa_idx & ~3; + case aq_macsec_sa_sc_2sa_16sc: + return sa_idx & ~1; + case aq_macsec_sa_sc_1sa_32sc: + return sa_idx; + default: + WARN_ONCE(true, "Invalid sc_sa"); + } + return -EINVAL; +} + +static void aq_check_txsa_expiration(struct aq_nic_s *nic) +{ + u32 egress_sa_expired, egress_sa_threshold_expired; + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + struct aq_hw_s *hw = nic->aq_hw; + struct aq_macsec_txsc *aq_txsc; + const struct macsec_secy *secy; + int sc_idx = 0, txsc_idx = 0; + enum aq_macsec_sc_sa sc_sa; + struct macsec_tx_sa *tx_sa; + unsigned char an = 0; + int ret; + int i; + + sc_sa = cfg->sc_sa; + + ret = aq_mss_get_egress_sa_expired(hw, &egress_sa_expired); + if (unlikely(ret)) + return; + + ret = aq_mss_get_egress_sa_threshold_expired(hw, + &egress_sa_threshold_expired); + + for (i = 0; i < AQ_MACSEC_MAX_SA; i++) { + if (egress_sa_expired & BIT(i)) { + an = aq_sa_from_sa_idx(sc_sa, i); + sc_idx = aq_sc_idx_from_sa_idx(sc_sa, i); + txsc_idx = aq_get_txsc_idx_from_sc_idx(sc_sa, sc_idx); + if (txsc_idx < 0) + continue; + + aq_txsc = &cfg->aq_txsc[txsc_idx]; + if (!(cfg->txsc_idx_busy & BIT(txsc_idx))) { + netdev_warn(nic->ndev, + "PN threshold expired on invalid TX SC"); + continue; + } + + secy = aq_txsc->sw_secy; + if (!netif_running(secy->netdev)) { + netdev_warn(nic->ndev, + "PN threshold expired on down TX SC"); + continue; + } + + if (unlikely(!(aq_txsc->tx_sa_idx_busy & BIT(an)))) { + netdev_warn(nic->ndev, + "PN threshold expired on invalid TX SA"); + continue; + } + + tx_sa = rcu_dereference_bh(secy->tx_sc.sa[an]); + macsec_pn_wrapped((struct macsec_secy *)secy, tx_sa); + } + } + + aq_mss_set_egress_sa_expired(hw, egress_sa_expired); + if (likely(!ret)) + aq_mss_set_egress_sa_threshold_expired(hw, + egress_sa_threshold_expired); +} + +const struct macsec_ops aq_macsec_ops = { + .mdo_dev_open = aq_mdo_dev_open, + .mdo_dev_stop = aq_mdo_dev_stop, + .mdo_add_secy = aq_mdo_add_secy, + .mdo_upd_secy = aq_mdo_upd_secy, + .mdo_del_secy = aq_mdo_del_secy, + .mdo_add_rxsc = aq_mdo_add_rxsc, + .mdo_upd_rxsc = aq_mdo_upd_rxsc, + .mdo_del_rxsc = aq_mdo_del_rxsc, + .mdo_add_rxsa = aq_mdo_add_rxsa, + .mdo_upd_rxsa = aq_mdo_upd_rxsa, + .mdo_del_rxsa = aq_mdo_del_rxsa, + .mdo_add_txsa = aq_mdo_add_txsa, + .mdo_upd_txsa = aq_mdo_upd_txsa, + .mdo_del_txsa = aq_mdo_del_txsa, + .mdo_get_dev_stats = aq_mdo_get_dev_stats, + .mdo_get_tx_sc_stats = aq_mdo_get_tx_sc_stats, + .mdo_get_tx_sa_stats = aq_mdo_get_tx_sa_stats, + .mdo_get_rx_sc_stats = aq_mdo_get_rx_sc_stats, + .mdo_get_rx_sa_stats = aq_mdo_get_rx_sa_stats, +}; + +int aq_macsec_init(struct aq_nic_s *nic) +{ + struct aq_macsec_cfg *cfg; + u32 caps_lo; + + if (!nic->aq_fw_ops->get_link_capabilities) + return 0; + + caps_lo = nic->aq_fw_ops->get_link_capabilities(nic->aq_hw); + + if (!(caps_lo & BIT(CAPS_LO_MACSEC))) + return 0; + + nic->macsec_cfg = kzalloc(sizeof(*cfg), GFP_KERNEL); + if (!nic->macsec_cfg) + return -ENOMEM; + + nic->ndev->features |= NETIF_F_HW_MACSEC; + nic->ndev->macsec_ops = &aq_macsec_ops; + + return 0; +} + +void aq_macsec_free(struct aq_nic_s *nic) +{ + kfree(nic->macsec_cfg); + nic->macsec_cfg = NULL; +} + +int aq_macsec_enable(struct aq_nic_s *nic) +{ + u32 ctl_ether_types[1] = { ETH_P_PAE }; + struct macsec_msg_fw_response resp = { 0 }; + struct macsec_msg_fw_request msg = { 0 }; + struct aq_hw_s *hw = nic->aq_hw; + int num_ctl_ether_types = 0; + int index = 0, tbl_idx; + int ret; + + if (!nic->macsec_cfg) + return 0; + + rtnl_lock(); + + if (nic->aq_fw_ops->send_macsec_req) { + struct macsec_cfg_request cfg = { 0 }; + + cfg.enabled = 1; + cfg.egress_threshold = 0xffffffff; + cfg.ingress_threshold = 0xffffffff; + cfg.interrupts_enabled = 1; + + msg.msg_type = macsec_cfg_msg; + msg.cfg = cfg; + + ret = nic->aq_fw_ops->send_macsec_req(hw, &msg, &resp); + if (ret) + goto unlock; + } + + /* Init Ethertype bypass filters */ + for (index = 0; index < ARRAY_SIZE(ctl_ether_types); index++) { + struct aq_mss_ingress_prectlf_record rx_prectlf_rec; + struct aq_mss_egress_ctlf_record tx_ctlf_rec; + + if (ctl_ether_types[index] == 0) + continue; + + memset(&tx_ctlf_rec, 0, sizeof(tx_ctlf_rec)); + tx_ctlf_rec.eth_type = ctl_ether_types[index]; + tx_ctlf_rec.match_type = 4; /* Match eth_type only */ + tx_ctlf_rec.match_mask = 0xf; /* match for eth_type */ + tx_ctlf_rec.action = 0; /* Bypass MACSEC modules */ + tbl_idx = NUMROWS_EGRESSCTLFRECORD - num_ctl_ether_types - 1; + aq_mss_set_egress_ctlf_record(hw, &tx_ctlf_rec, tbl_idx); + + memset(&rx_prectlf_rec, 0, sizeof(rx_prectlf_rec)); + rx_prectlf_rec.eth_type = ctl_ether_types[index]; + rx_prectlf_rec.match_type = 4; /* Match eth_type only */ + rx_prectlf_rec.match_mask = 0xf; /* match for eth_type */ + rx_prectlf_rec.action = 0; /* Bypass MACSEC modules */ + tbl_idx = + NUMROWS_INGRESSPRECTLFRECORD - num_ctl_ether_types - 1; + aq_mss_set_ingress_prectlf_record(hw, &rx_prectlf_rec, tbl_idx); + + num_ctl_ether_types++; + } + + ret = aq_apply_macsec_cfg(nic); + +unlock: + rtnl_unlock(); + return ret; +} + +void aq_macsec_work(struct aq_nic_s *nic) +{ + if (!nic->macsec_cfg) + return; + + if (!netif_carrier_ok(nic->ndev)) + return; + + rtnl_lock(); + aq_check_txsa_expiration(nic); + rtnl_unlock(); +} + +int aq_macsec_rx_sa_cnt(struct aq_nic_s *nic) +{ + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + int i, cnt = 0; + + if (!cfg) + return 0; + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (!test_bit(i, &cfg->rxsc_idx_busy)) + continue; + cnt += hweight_long(cfg->aq_rxsc[i].rx_sa_idx_busy); + } + + return cnt; +} + +int aq_macsec_tx_sc_cnt(struct aq_nic_s *nic) +{ + if (!nic->macsec_cfg) + return 0; + + return hweight_long(nic->macsec_cfg->txsc_idx_busy); +} + +int aq_macsec_tx_sa_cnt(struct aq_nic_s *nic) +{ + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + int i, cnt = 0; + + if (!cfg) + return 0; + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (!test_bit(i, &cfg->txsc_idx_busy)) + continue; + cnt += hweight_long(cfg->aq_txsc[i].tx_sa_idx_busy); + } + + return cnt; +} + +static int aq_macsec_update_stats(struct aq_nic_s *nic) +{ + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + struct aq_hw_s *hw = nic->aq_hw; + struct aq_macsec_txsc *aq_txsc; + struct aq_macsec_rxsc *aq_rxsc; + int i, sa_idx, assoc_num; + int ret = 0; + + aq_get_macsec_common_stats(hw, &cfg->stats); + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (!(cfg->txsc_idx_busy & BIT(i))) + continue; + aq_txsc = &cfg->aq_txsc[i]; + + ret = aq_get_txsc_stats(hw, aq_txsc->hw_sc_idx, + &aq_txsc->stats); + if (ret) + return ret; + + for (assoc_num = 0; assoc_num < MACSEC_NUM_AN; assoc_num++) { + if (!test_bit(assoc_num, &aq_txsc->tx_sa_idx_busy)) + continue; + sa_idx = aq_txsc->hw_sc_idx | assoc_num; + ret = aq_get_txsa_stats(hw, sa_idx, + &aq_txsc->tx_sa_stats[assoc_num]); + if (ret) + return ret; + } + } + + for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { + if (!(test_bit(i, &cfg->rxsc_idx_busy))) + continue; + aq_rxsc = &cfg->aq_rxsc[i]; + + for (assoc_num = 0; assoc_num < MACSEC_NUM_AN; assoc_num++) { + if (!test_bit(assoc_num, &aq_rxsc->rx_sa_idx_busy)) + continue; + sa_idx = aq_rxsc->hw_sc_idx | assoc_num; + + ret = aq_get_rxsa_stats(hw, sa_idx, + &aq_rxsc->rx_sa_stats[assoc_num]); + if (ret) + return ret; + } + } + + return ret; +} + +u64 *aq_macsec_get_stats(struct aq_nic_s *nic, u64 *data) +{ + struct aq_macsec_cfg *cfg = nic->macsec_cfg; + struct aq_macsec_common_stats *common_stats; + struct aq_macsec_tx_sc_stats *txsc_stats; + struct aq_macsec_tx_sa_stats *txsa_stats; + struct aq_macsec_rx_sa_stats *rxsa_stats; + struct aq_macsec_txsc *aq_txsc; + struct aq_macsec_rxsc *aq_rxsc; + unsigned int assoc_num; + unsigned int sc_num; + unsigned int i = 0U; + + if (!cfg) + return data; + + aq_macsec_update_stats(nic); + + common_stats = &cfg->stats; + data[i] = common_stats->in.ctl_pkts; + data[++i] = common_stats->in.tagged_miss_pkts; + data[++i] = common_stats->in.untagged_miss_pkts; + data[++i] = common_stats->in.notag_pkts; + data[++i] = common_stats->in.untagged_pkts; + data[++i] = common_stats->in.bad_tag_pkts; + data[++i] = common_stats->in.no_sci_pkts; + data[++i] = common_stats->in.unknown_sci_pkts; + data[++i] = common_stats->in.ctrl_prt_pass_pkts; + data[++i] = common_stats->in.unctrl_prt_pass_pkts; + data[++i] = common_stats->in.ctrl_prt_fail_pkts; + data[++i] = common_stats->in.unctrl_prt_fail_pkts; + data[++i] = common_stats->in.too_long_pkts; + data[++i] = common_stats->in.igpoc_ctl_pkts; + data[++i] = common_stats->in.ecc_error_pkts; + data[++i] = common_stats->in.unctrl_hit_drop_redir; + data[++i] = common_stats->out.ctl_pkts; + data[++i] = common_stats->out.unknown_sa_pkts; + data[++i] = common_stats->out.untagged_pkts; + data[++i] = common_stats->out.too_long; + data[++i] = common_stats->out.ecc_error_pkts; + data[++i] = common_stats->out.unctrl_hit_drop_redir; + + for (sc_num = 0; sc_num < AQ_MACSEC_MAX_SC; sc_num++) { + if (!(test_bit(sc_num, &cfg->txsc_idx_busy))) + continue; + + aq_txsc = &cfg->aq_txsc[sc_num]; + txsc_stats = &aq_txsc->stats; + + data[++i] = txsc_stats->sc_protected_pkts; + data[++i] = txsc_stats->sc_encrypted_pkts; + data[++i] = txsc_stats->sc_protected_octets; + data[++i] = txsc_stats->sc_encrypted_octets; + + for (assoc_num = 0; assoc_num < MACSEC_NUM_AN; assoc_num++) { + if (!test_bit(assoc_num, &aq_txsc->tx_sa_idx_busy)) + continue; + + txsa_stats = &aq_txsc->tx_sa_stats[assoc_num]; + + data[++i] = txsa_stats->sa_hit_drop_redirect; + data[++i] = txsa_stats->sa_protected2_pkts; + data[++i] = txsa_stats->sa_protected_pkts; + data[++i] = txsa_stats->sa_encrypted_pkts; + } + } + + for (sc_num = 0; sc_num < AQ_MACSEC_MAX_SC; sc_num++) { + if (!(test_bit(sc_num, &cfg->rxsc_idx_busy))) + continue; + + aq_rxsc = &cfg->aq_rxsc[sc_num]; + + for (assoc_num = 0; assoc_num < MACSEC_NUM_AN; assoc_num++) { + if (!test_bit(assoc_num, &aq_rxsc->rx_sa_idx_busy)) + continue; + + rxsa_stats = &aq_rxsc->rx_sa_stats[assoc_num]; + + data[++i] = rxsa_stats->untagged_hit_pkts; + data[++i] = rxsa_stats->ctrl_hit_drop_redir_pkts; + data[++i] = rxsa_stats->not_using_sa; + data[++i] = rxsa_stats->unused_sa; + data[++i] = rxsa_stats->not_valid_pkts; + data[++i] = rxsa_stats->invalid_pkts; + data[++i] = rxsa_stats->ok_pkts; + data[++i] = rxsa_stats->late_pkts; + data[++i] = rxsa_stats->delayed_pkts; + data[++i] = rxsa_stats->unchecked_pkts; + data[++i] = rxsa_stats->validated_octets; + data[++i] = rxsa_stats->decrypted_octets; + } + } + + i++; + + data += i; + + return data; +} diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_macsec.h b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.h new file mode 100644 index 000000000000..f5fba8b8cdea --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.h @@ -0,0 +1,133 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef AQ_MACSEC_H +#define AQ_MACSEC_H + +#include <linux/netdevice.h> +#if IS_ENABLED(CONFIG_MACSEC) + +#include "net/macsec.h" + +struct aq_nic_s; + +#define AQ_MACSEC_MAX_SC 32 +#define AQ_MACSEC_MAX_SA 32 + +enum aq_macsec_sc_sa { + aq_macsec_sa_sc_4sa_8sc, + aq_macsec_sa_sc_not_used, + aq_macsec_sa_sc_2sa_16sc, + aq_macsec_sa_sc_1sa_32sc, +}; + +struct aq_macsec_common_stats { + /* Ingress Common Counters */ + struct { + u64 ctl_pkts; + u64 tagged_miss_pkts; + u64 untagged_miss_pkts; + u64 notag_pkts; + u64 untagged_pkts; + u64 bad_tag_pkts; + u64 no_sci_pkts; + u64 unknown_sci_pkts; + u64 ctrl_prt_pass_pkts; + u64 unctrl_prt_pass_pkts; + u64 ctrl_prt_fail_pkts; + u64 unctrl_prt_fail_pkts; + u64 too_long_pkts; + u64 igpoc_ctl_pkts; + u64 ecc_error_pkts; + u64 unctrl_hit_drop_redir; + } in; + + /* Egress Common Counters */ + struct { + u64 ctl_pkts; + u64 unknown_sa_pkts; + u64 untagged_pkts; + u64 too_long; + u64 ecc_error_pkts; + u64 unctrl_hit_drop_redir; + } out; +}; + +/* Ingress SA Counters */ +struct aq_macsec_rx_sa_stats { + u64 untagged_hit_pkts; + u64 ctrl_hit_drop_redir_pkts; + u64 not_using_sa; + u64 unused_sa; + u64 not_valid_pkts; + u64 invalid_pkts; + u64 ok_pkts; + u64 late_pkts; + u64 delayed_pkts; + u64 unchecked_pkts; + u64 validated_octets; + u64 decrypted_octets; +}; + +/* Egress SA Counters */ +struct aq_macsec_tx_sa_stats { + u64 sa_hit_drop_redirect; + u64 sa_protected2_pkts; + u64 sa_protected_pkts; + u64 sa_encrypted_pkts; +}; + +/* Egress SC Counters */ +struct aq_macsec_tx_sc_stats { + u64 sc_protected_pkts; + u64 sc_encrypted_pkts; + u64 sc_protected_octets; + u64 sc_encrypted_octets; +}; + +struct aq_macsec_txsc { + u32 hw_sc_idx; + unsigned long tx_sa_idx_busy; + const struct macsec_secy *sw_secy; + u8 tx_sa_key[MACSEC_NUM_AN][MACSEC_KEYID_LEN]; + struct aq_macsec_tx_sc_stats stats; + struct aq_macsec_tx_sa_stats tx_sa_stats[MACSEC_NUM_AN]; +}; + +struct aq_macsec_rxsc { + u32 hw_sc_idx; + unsigned long rx_sa_idx_busy; + const struct macsec_secy *sw_secy; + const struct macsec_rx_sc *sw_rxsc; + u8 rx_sa_key[MACSEC_NUM_AN][MACSEC_KEYID_LEN]; + struct aq_macsec_rx_sa_stats rx_sa_stats[MACSEC_NUM_AN]; +}; + +struct aq_macsec_cfg { + enum aq_macsec_sc_sa sc_sa; + /* Egress channel configuration */ + unsigned long txsc_idx_busy; + struct aq_macsec_txsc aq_txsc[AQ_MACSEC_MAX_SC]; + /* Ingress channel configuration */ + unsigned long rxsc_idx_busy; + struct aq_macsec_rxsc aq_rxsc[AQ_MACSEC_MAX_SC]; + /* Statistics / counters */ + struct aq_macsec_common_stats stats; +}; + +extern const struct macsec_ops aq_macsec_ops; + +int aq_macsec_init(struct aq_nic_s *nic); +void aq_macsec_free(struct aq_nic_s *nic); +int aq_macsec_enable(struct aq_nic_s *nic); +void aq_macsec_work(struct aq_nic_s *nic); +u64 *aq_macsec_get_stats(struct aq_nic_s *nic, u64 *data); +int aq_macsec_rx_sa_cnt(struct aq_nic_s *nic); +int aq_macsec_tx_sc_cnt(struct aq_nic_s *nic); +int aq_macsec_tx_sa_cnt(struct aq_nic_s *nic); + +#endif + +#endif /* AQ_MACSEC_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.c b/drivers/net/ethernet/aquantia/atlantic/aq_main.c index 538f460a3da7..9fcab646cbd5 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_main.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.c @@ -19,7 +19,6 @@ #include <linux/udp.h> MODULE_LICENSE("GPL v2"); -MODULE_VERSION(AQ_CFG_DRV_VERSION); MODULE_AUTHOR(AQ_CFG_DRV_AUTHOR); MODULE_DESCRIPTION(AQ_CFG_DRV_DESC); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index e95f6a6bef73..a369705a786a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -11,6 +11,7 @@ #include "aq_vec.h" #include "aq_hw.h" #include "aq_pci_func.h" +#include "aq_macsec.h" #include "aq_main.h" #include "aq_phy.h" #include "aq_ptp.h" @@ -176,6 +177,9 @@ static int aq_nic_update_link_status(struct aq_nic_s *self) aq_utils_obj_clear(&self->flags, AQ_NIC_LINK_DOWN); netif_carrier_on(self->ndev); +#if IS_ENABLED(CONFIG_MACSEC) + aq_macsec_enable(self); +#endif netif_tx_wake_all_queues(self->ndev); } if (netif_carrier_ok(self->ndev) && !self->link_status.mbps) { @@ -217,6 +221,10 @@ static void aq_nic_service_task(struct work_struct *work) if (err) return; +#if IS_ENABLED(CONFIG_MACSEC) + aq_macsec_work(self); +#endif + mutex_lock(&self->fwreq_mutex); if (self->aq_fw_ops->update_stats) self->aq_fw_ops->update_stats(self->aq_hw); @@ -262,6 +270,10 @@ int aq_nic_ndev_register(struct aq_nic_s *self) if (err) goto err_exit; +#if IS_ENABLED(CONFIG_MACSEC) + aq_macsec_init(self); +#endif + mutex_lock(&self->fwreq_mutex); err = self->aq_fw_ops->get_mac_permanent(self->aq_hw, self->ndev->dev_addr); @@ -296,6 +308,10 @@ int aq_nic_ndev_register(struct aq_nic_s *self) goto err_exit; err_exit: +#if IS_ENABLED(CONFIG_MACSEC) + if (err) + aq_macsec_free(self); +#endif return err; } @@ -765,7 +781,7 @@ int aq_nic_get_regs_count(struct aq_nic_s *self) return self->aq_nic_cfg.aq_hw_caps->mac_regs_count; } -void aq_nic_get_stats(struct aq_nic_s *self, u64 *data) +u64 *aq_nic_get_stats(struct aq_nic_s *self, u64 *data) { struct aq_vec_s *aq_vec = NULL; struct aq_stats_s *stats; @@ -815,7 +831,10 @@ void aq_nic_get_stats(struct aq_nic_s *self, u64 *data) aq_vec_get_sw_stats(aq_vec, data, &count); } + data += count; + err_exit:; + return data; } static void aq_nic_update_ndev_stats(struct aq_nic_s *self) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h index a752f8bb4b08..0663b8d0220d 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h @@ -17,6 +17,7 @@ struct aq_ring_s; struct aq_hw_ops; struct aq_fw_s; struct aq_vec_s; +struct aq_macsec_cfg; struct aq_ptp_s; enum aq_rx_filter_type; @@ -129,6 +130,9 @@ struct aq_nic_s { u32 irqvecs; /* mutex to serialize FW interface access operations */ struct mutex fwreq_mutex; +#if IS_ENABLED(CONFIG_MACSEC) + struct aq_macsec_cfg *macsec_cfg; +#endif /* PTP support */ struct aq_ptp_s *aq_ptp; struct aq_hw_rx_fltrs_s aq_hw_rx_fltrs; @@ -154,7 +158,7 @@ unsigned int aq_nic_map_skb(struct aq_nic_s *self, struct sk_buff *skb, int aq_nic_xmit(struct aq_nic_s *self, struct sk_buff *skb); int aq_nic_get_regs(struct aq_nic_s *self, struct ethtool_regs *regs, void *p); int aq_nic_get_regs_count(struct aq_nic_s *self); -void aq_nic_get_stats(struct aq_nic_s *self, u64 *data); +u64 *aq_nic_get_stats(struct aq_nic_s *self, u64 *data); int aq_nic_stop(struct aq_nic_s *self); void aq_nic_deinit(struct aq_nic_s *self, bool link_down); void aq_nic_set_power(struct aq_nic_s *self); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c index 78b6f3248756..2edf137a7030 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c @@ -18,6 +18,7 @@ #include "hw_atl/hw_atl_b0.h" #include "aq_filters.h" #include "aq_drvinfo.h" +#include "aq_macsec.h" static const struct pci_device_id aq_pci_tbl[] = { { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_0001), }, @@ -324,6 +325,10 @@ static void aq_pci_remove(struct pci_dev *pdev) aq_clear_rxnfc_all_rules(self); if (self->ndev->reg_state == NETREG_REGISTERED) unregister_netdev(self->ndev); + +#if IS_ENABLED(CONFIG_MACSEC) + aq_macsec_free(self); +#endif aq_nic_free_vectors(self); aq_pci_free_irq_vectors(self); iounmap(self->aq_hw->mmio); diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h index 42f0c5c6ec2d..b15513914636 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h @@ -225,7 +225,7 @@ struct __packed offload_info { struct offload_port_info ports; struct offload_ka_info kas; struct offload_rr_info rrs; - u8 buf[0]; + u8 buf[]; }; struct __packed hw_atl_utils_fw_rpc { @@ -319,6 +319,32 @@ struct __packed hw_atl_utils_settings { u32 media_detect; }; +enum macsec_msg_type { + macsec_cfg_msg = 0, + macsec_add_rx_sc_msg, + macsec_add_tx_sc_msg, + macsec_add_rx_sa_msg, + macsec_add_tx_sa_msg, + macsec_get_stats_msg, +}; + +struct __packed macsec_cfg_request { + u32 enabled; + u32 egress_threshold; + u32 ingress_threshold; + u32 interrupts_enabled; +}; + +struct __packed macsec_msg_fw_request { + u32 msg_id; /* not used */ + u32 msg_type; + struct macsec_cfg_request cfg; +}; + +struct __packed macsec_msg_fw_response { + u32 result; +}; + enum hw_atl_rx_action_with_traffic { HW_ATL_RX_DISCARD, HW_ATL_RX_HOST, @@ -437,34 +463,43 @@ enum hw_atl_fw2x_caps_lo { CAPS_LO_2P5GBASET_FD, CAPS_LO_5GBASET_FD = 10, CAPS_LO_10GBASET_FD, + CAPS_LO_AUTONEG, + CAPS_LO_SMBUS_READ, + CAPS_LO_SMBUS_WRITE, + CAPS_LO_MACSEC = 15, + CAPS_LO_RESERVED1, + CAPS_LO_WAKE_ON_LINK_FORCED, + CAPS_LO_HIGH_TEMP_WARNING = 29, + CAPS_LO_DRIVER_SCRATCHPAD = 30, + CAPS_LO_GLOBAL_FAULT = 31 }; /* 0x374 * Status register */ enum hw_atl_fw2x_caps_hi { - CAPS_HI_RESERVED1 = 0, + CAPS_HI_TPO2EN = 0, CAPS_HI_10BASET_EEE, CAPS_HI_RESERVED2, CAPS_HI_PAUSE, CAPS_HI_ASYMMETRIC_PAUSE, CAPS_HI_100BASETX_EEE = 5, - CAPS_HI_RESERVED3, - CAPS_HI_RESERVED4, + CAPS_HI_PHY_BUF_SEND, + CAPS_HI_PHY_BUF_RECV, CAPS_HI_1000BASET_FD_EEE, CAPS_HI_2P5GBASET_FD_EEE, CAPS_HI_5GBASET_FD_EEE = 10, CAPS_HI_10GBASET_FD_EEE, CAPS_HI_FW_REQUEST, - CAPS_HI_RESERVED6, - CAPS_HI_RESERVED7, - CAPS_HI_RESERVED8 = 15, - CAPS_HI_RESERVED9, + CAPS_HI_PHY_LOG, + CAPS_HI_EEE_AUTO_DISABLE_SETTINGS, + CAPS_HI_PFC = 15, + CAPS_HI_WAKE_ON_LINK, CAPS_HI_CABLE_DIAG, CAPS_HI_TEMPERATURE, CAPS_HI_DOWNSHIFT, CAPS_HI_PTP_AVB_EN_FW2X = 20, - CAPS_HI_MEDIA_DETECT, + CAPS_HI_THERMAL_SHUTDOWN, CAPS_HI_LINK_DROP, CAPS_HI_SLEEP_PROXY, CAPS_HI_WOL, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c index 77a4ed64830f..1ad10cc14918 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c @@ -55,6 +55,8 @@ #define HW_ATL_FW2X_CAP_EEE_5G_MASK BIT(CAPS_HI_5GBASET_FD_EEE) #define HW_ATL_FW2X_CAP_EEE_10G_MASK BIT(CAPS_HI_10GBASET_FD_EEE) +#define HW_ATL_FW2X_CAP_MACSEC BIT(CAPS_LO_MACSEC) + #define HAL_ATLANTIC_WOL_FILTERS_COUNT 8 #define HAL_ATLANTIC_UTILS_FW2X_MSG_WOL 0x0E @@ -86,6 +88,7 @@ static int aq_fw2x_set_state(struct aq_hw_s *self, static u32 aq_fw2x_mbox_get(struct aq_hw_s *self); static u32 aq_fw2x_rpc_get(struct aq_hw_s *self); static int aq_fw2x_settings_get(struct aq_hw_s *self, u32 *addr); +static u32 aq_fw2x_state_get(struct aq_hw_s *self); static u32 aq_fw2x_state2_get(struct aq_hw_s *self); static int aq_fw2x_init(struct aq_hw_s *self) @@ -619,11 +622,75 @@ static int aq_fw2x_settings_get(struct aq_hw_s *self, u32 *addr) return err; } +static u32 aq_fw2x_state_get(struct aq_hw_s *self) +{ + return aq_hw_read_reg(self, HW_ATL_FW2X_MPI_STATE_ADDR); +} + static u32 aq_fw2x_state2_get(struct aq_hw_s *self) { return aq_hw_read_reg(self, HW_ATL_FW2X_MPI_STATE2_ADDR); } +static u32 aq_fw2x_get_link_capabilities(struct aq_hw_s *self) +{ + int err = 0; + u32 offset; + u32 val; + + offset = self->mbox_addr + + offsetof(struct hw_atl_utils_mbox, info.caps_lo); + + err = hw_atl_utils_fw_downld_dwords(self, offset, &val, 1); + + if (err) + return 0; + + return val; +} + +static int aq_fw2x_send_macsec_req(struct aq_hw_s *hw, + struct macsec_msg_fw_request *req, + struct macsec_msg_fw_response *response) +{ + u32 low_status, low_req = 0; + u32 dword_cnt; + u32 caps_lo; + u32 offset; + int err; + + if (!req || !response) + return -EINVAL; + + caps_lo = aq_fw2x_get_link_capabilities(hw); + if (!(caps_lo & BIT(CAPS_LO_MACSEC))) + return -EOPNOTSUPP; + + /* Write macsec request to cfg memory */ + dword_cnt = (sizeof(*req) + sizeof(u32) - 1) / sizeof(u32); + err = hw_atl_write_fwcfg_dwords(hw, (void *)req, dword_cnt); + if (err < 0) + return err; + + /* Toggle 0x368.CAPS_LO_MACSEC bit */ + low_req = aq_hw_read_reg(hw, HW_ATL_FW2X_MPI_CONTROL_ADDR); + low_req ^= HW_ATL_FW2X_CAP_MACSEC; + aq_hw_write_reg(hw, HW_ATL_FW2X_MPI_CONTROL_ADDR, low_req); + + /* Wait FW to report back */ + err = readx_poll_timeout_atomic(aq_fw2x_state_get, hw, low_status, + low_req != (low_status & BIT(CAPS_LO_MACSEC)), 1U, 10000U); + if (err) + return -EIO; + + /* Read status of write operation */ + offset = hw->rpc_addr + sizeof(u32); + err = hw_atl_utils_fw_downld_dwords(hw, offset, (u32 *)(void *)response, + sizeof(*response) / sizeof(u32)); + + return err; +} + const struct aq_fw_ops aq_fw_2x_ops = { .init = aq_fw2x_init, .deinit = aq_fw2x_deinit, @@ -645,4 +712,6 @@ const struct aq_fw_ops aq_fw_2x_ops = { .led_control = aq_fw2x_led_control, .set_phyloopback = aq_fw2x_set_phyloopback, .adjust_ptp = aq_fw3x_adjust_ptp, + .get_link_capabilities = aq_fw2x_get_link_capabilities, + .send_macsec_req = aq_fw2x_send_macsec_req, }; diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Egress_registers.h b/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Egress_registers.h new file mode 100644 index 000000000000..71d08ea80b54 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Egress_registers.h @@ -0,0 +1,73 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef MSS_EGRESS_REGS_HEADER +#define MSS_EGRESS_REGS_HEADER + +#define MSS_EGRESS_CTL_REGISTER_ADDR 0x00005002 +#define MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR 0x00005060 +#define MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR 0x00005062 +#define MSS_EGRESS_LUT_ADDR_CTL_REGISTER_ADDR 0x00005080 +#define MSS_EGRESS_LUT_CTL_REGISTER_ADDR 0x00005081 +#define MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR 0x000050A0 + +struct mss_egress_ctl_register { + union { + struct { + unsigned int soft_reset : 1; + unsigned int drop_kay_packet : 1; + unsigned int drop_egprc_lut_miss : 1; + unsigned int gcm_start : 1; + unsigned int gcm_test_mode : 1; + unsigned int unmatched_use_sc_0 : 1; + unsigned int drop_invalid_sa_sc_packets : 1; + unsigned int reserved0 : 1; + /* Should always be set to 0. */ + unsigned int external_classification_enable : 1; + unsigned int icv_lsb_8bytes_enable : 1; + unsigned int high_prio : 1; + unsigned int clear_counter : 1; + unsigned int clear_global_time : 1; + unsigned int ethertype_explicit_sectag_lsb : 3; + } bits_0; + unsigned short word_0; + }; + union { + struct { + unsigned int ethertype_explicit_sectag_msb : 13; + unsigned int reserved0 : 3; + } bits_1; + unsigned short word_1; + }; +}; + +struct mss_egress_lut_addr_ctl_register { + union { + struct { + unsigned int lut_addr : 9; + unsigned int reserved0 : 3; + /* 0x0 : Egress MAC Control FIlter (CTLF) LUT + * 0x1 : Egress Classification LUT + * 0x2 : Egress SC/SA LUT + * 0x3 : Egress SMIB + */ + unsigned int lut_select : 4; + } bits_0; + unsigned short word_0; + }; +}; + +struct mss_egress_lut_ctl_register { + union { + struct { + unsigned int reserved0 : 14; + unsigned int lut_read : 1; + unsigned int lut_write : 1; + } bits_0; + unsigned short word_0; + }; +}; + +#endif /* MSS_EGRESS_REGS_HEADER */ diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Ingress_registers.h b/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Ingress_registers.h new file mode 100644 index 000000000000..d4c00d9a0fc6 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/macsec/MSS_Ingress_registers.h @@ -0,0 +1,77 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef MSS_INGRESS_REGS_HEADER +#define MSS_INGRESS_REGS_HEADER + +#define MSS_INGRESS_CTL_REGISTER_ADDR 0x0000800E +#define MSS_INGRESS_LUT_ADDR_CTL_REGISTER_ADDR 0x00008080 +#define MSS_INGRESS_LUT_CTL_REGISTER_ADDR 0x00008081 +#define MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR 0x000080A0 + +struct mss_ingress_ctl_register { + union { + struct { + unsigned int soft_reset : 1; + unsigned int operation_point_to_point : 1; + unsigned int create_sci : 1; + /* Unused */ + unsigned int mask_short_length_error : 1; + unsigned int drop_kay_packet : 1; + unsigned int drop_igprc_miss : 1; + /* Unused */ + unsigned int check_icv : 1; + unsigned int clear_global_time : 1; + unsigned int clear_count : 1; + unsigned int high_prio : 1; + unsigned int remove_sectag : 1; + unsigned int global_validate_frames : 2; + unsigned int icv_lsb_8bytes_enabled : 1; + unsigned int reserved0 : 2; + } bits_0; + unsigned short word_0; + }; + union { + struct { + unsigned int reserved0 : 16; + } bits_1; + unsigned short word_1; + }; +}; + +struct mss_ingress_lut_addr_ctl_register { + union { + struct { + unsigned int lut_addr : 9; + unsigned int reserved0 : 3; + /* 0x0 : Ingress Pre-Security MAC Control FIlter + * (IGPRCTLF) LUT + * 0x1 : Ingress Pre-Security Classification LUT (IGPRC) + * 0x2 : Ingress Packet Format (IGPFMT) SAKey LUT + * 0x3 : Ingress Packet Format (IGPFMT) SC/SA LUT + * 0x4 : Ingress Post-Security Classification LUT + * (IGPOC) + * 0x5 : Ingress Post-Security MAC Control Filter + * (IGPOCTLF) LUT + * 0x6 : Ingress MIB (IGMIB) + */ + unsigned int lut_select : 4; + } bits_0; + unsigned short word_0; + }; +}; + +struct mss_ingress_lut_ctl_register { + union { + struct { + unsigned int reserved0 : 14; + unsigned int lut_read : 1; + unsigned int lut_write : 1; + } bits_0; + unsigned short word_0; + }; +}; + +#endif /* MSS_INGRESS_REGS_HEADER */ diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c new file mode 100644 index 000000000000..97901c114bfa --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c @@ -0,0 +1,2473 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include "macsec_api.h" +#include <linux/mdio.h> +#include "MSS_Ingress_registers.h" +#include "MSS_Egress_registers.h" +#include "aq_phy.h" + +#define AQ_API_CALL_SAFE(func, ...) \ +({ \ + int ret; \ + do { \ + ret = aq_mss_mdio_sem_get(hw); \ + if (unlikely(ret)) \ + break; \ + \ + ret = func(__VA_ARGS__); \ + \ + aq_mss_mdio_sem_put(hw); \ + } while (0); \ + ret; \ +}) + +/******************************************************************************* + * MDIO wrappers + ******************************************************************************/ +static int aq_mss_mdio_sem_get(struct aq_hw_s *hw) +{ + u32 val; + + return readx_poll_timeout_atomic(hw_atl_sem_mdio_get, hw, val, + val == 1U, 10U, 100000U); +} + +static void aq_mss_mdio_sem_put(struct aq_hw_s *hw) +{ + hw_atl_reg_glb_cpu_sem_set(hw, 1U, HW_ATL_FW_SM_MDIO); +} + +static int aq_mss_mdio_read(struct aq_hw_s *hw, u16 mmd, u16 addr, u16 *data) +{ + *data = aq_mdio_read_word(hw, mmd, addr); + return (*data != 0xffff) ? 0 : -ETIME; +} + +static int aq_mss_mdio_write(struct aq_hw_s *hw, u16 mmd, u16 addr, u16 data) +{ + aq_mdio_write_word(hw, mmd, addr, data); + return 0; +} + +/******************************************************************************* + * MACSEC config and status + ******************************************************************************/ + +static int set_raw_ingress_record(struct aq_hw_s *hw, u16 *packed_record, + u8 num_words, u8 table_id, + u16 table_index) +{ + struct mss_ingress_lut_addr_ctl_register lut_sel_reg; + struct mss_ingress_lut_ctl_register lut_op_reg; + + unsigned int i; + + /* NOTE: MSS registers must always be read/written as adjacent pairs. + * For instance, to write either or both 1E.80A0 and 80A1, we have to: + * 1. Write 1E.80A0 first + * 2. Then write 1E.80A1 + * + * For HHD devices: These writes need to be performed consecutively, and + * to ensure this we use the PIF mailbox to delegate the reads/writes to + * the FW. + * + * For EUR devices: Not need to use the PIF mailbox; it is safe to + * write to the registers directly. + */ + + /* Write the packed record words to the data buffer registers. */ + for (i = 0; i < num_words; i += 2) { + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + i, + packed_record[i]); + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + i + + 1, + packed_record[i + 1]); + } + + /* Clear out the unused data buffer registers. */ + for (i = num_words; i < 24; i += 2) { + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + i, + 0); + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + i + 1, 0); + } + + /* Select the table and row index to write to */ + lut_sel_reg.bits_0.lut_select = table_id; + lut_sel_reg.bits_0.lut_addr = table_index; + + lut_op_reg.bits_0.lut_read = 0; + lut_op_reg.bits_0.lut_write = 1; + + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_ADDR_CTL_REGISTER_ADDR, + lut_sel_reg.word_0); + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, MSS_INGRESS_LUT_CTL_REGISTER_ADDR, + lut_op_reg.word_0); + + return 0; +} + +/*! Read the specified Ingress LUT table row. + * packed_record - [OUT] The table row data (raw). + */ +static int get_raw_ingress_record(struct aq_hw_s *hw, u16 *packed_record, + u8 num_words, u8 table_id, + u16 table_index) +{ + struct mss_ingress_lut_addr_ctl_register lut_sel_reg; + struct mss_ingress_lut_ctl_register lut_op_reg; + int ret; + + unsigned int i; + + /* Select the table and row index to read */ + lut_sel_reg.bits_0.lut_select = table_id; + lut_sel_reg.bits_0.lut_addr = table_index; + + lut_op_reg.bits_0.lut_read = 1; + lut_op_reg.bits_0.lut_write = 0; + + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_ADDR_CTL_REGISTER_ADDR, + lut_sel_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_CTL_REGISTER_ADDR, + lut_op_reg.word_0); + if (unlikely(ret)) + return ret; + + memset(packed_record, 0, sizeof(u16) * num_words); + + for (i = 0; i < num_words; i += 2) { + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + + i, + &packed_record[i]); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_INGRESS_LUT_DATA_CTL_REGISTER_ADDR + + i + 1, + &packed_record[i + 1]); + if (unlikely(ret)) + return ret; + } + + return 0; +} + +/*! Write packed_record to the specified Egress LUT table row. */ +static int set_raw_egress_record(struct aq_hw_s *hw, u16 *packed_record, + u8 num_words, u8 table_id, + u16 table_index) +{ + struct mss_egress_lut_addr_ctl_register lut_sel_reg; + struct mss_egress_lut_ctl_register lut_op_reg; + + unsigned int i; + + /* Write the packed record words to the data buffer registers. */ + for (i = 0; i < num_words; i += 2) { + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + i, + packed_record[i]); + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + i + 1, + packed_record[i + 1]); + } + + /* Clear out the unused data buffer registers. */ + for (i = num_words; i < 28; i += 2) { + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + i, 0); + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + i + 1, + 0); + } + + /* Select the table and row index to write to */ + lut_sel_reg.bits_0.lut_select = table_id; + lut_sel_reg.bits_0.lut_addr = table_index; + + lut_op_reg.bits_0.lut_read = 0; + lut_op_reg.bits_0.lut_write = 1; + + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_ADDR_CTL_REGISTER_ADDR, + lut_sel_reg.word_0); + aq_mss_mdio_write(hw, MDIO_MMD_VEND1, MSS_EGRESS_LUT_CTL_REGISTER_ADDR, + lut_op_reg.word_0); + + return 0; +} + +static int get_raw_egress_record(struct aq_hw_s *hw, u16 *packed_record, + u8 num_words, u8 table_id, + u16 table_index) +{ + struct mss_egress_lut_addr_ctl_register lut_sel_reg; + struct mss_egress_lut_ctl_register lut_op_reg; + int ret; + + unsigned int i; + + /* Select the table and row index to read */ + lut_sel_reg.bits_0.lut_select = table_id; + lut_sel_reg.bits_0.lut_addr = table_index; + + lut_op_reg.bits_0.lut_read = 1; + lut_op_reg.bits_0.lut_write = 0; + + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_ADDR_CTL_REGISTER_ADDR, + lut_sel_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_CTL_REGISTER_ADDR, + lut_op_reg.word_0); + if (unlikely(ret)) + return ret; + + memset(packed_record, 0, sizeof(u16) * num_words); + + for (i = 0; i < num_words; i += 2) { + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + + i, + &packed_record[i]); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_EGRESS_LUT_DATA_CTL_REGISTER_ADDR + + i + 1, + &packed_record[i + 1]); + if (unlikely(ret)) + return ret; + } + + return 0; +} + +static int +set_ingress_prectlf_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_prectlf_record *rec, + u16 table_index) +{ + u16 packed_record[6]; + + if (table_index >= NUMROWS_INGRESSPRECTLFRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 6); + + packed_record[0] = rec->sa_da[0] & 0xFFFF; + packed_record[1] = (rec->sa_da[0] >> 16) & 0xFFFF; + packed_record[2] = rec->sa_da[1] & 0xFFFF; + packed_record[3] = rec->eth_type & 0xFFFF; + packed_record[4] = rec->match_mask & 0xFFFF; + packed_record[5] = rec->match_type & 0xF; + packed_record[5] |= (rec->action & 0x1) << 4; + + return set_raw_ingress_record(hw, packed_record, 6, 0, + ROWOFFSET_INGRESSPRECTLFRECORD + + table_index); +} + +int aq_mss_set_ingress_prectlf_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_prectlf_record *rec, + u16 table_index) +{ + return AQ_API_CALL_SAFE(set_ingress_prectlf_record, hw, rec, + table_index); +} + +static int get_ingress_prectlf_record(struct aq_hw_s *hw, + struct aq_mss_ingress_prectlf_record *rec, + u16 table_index) +{ + u16 packed_record[6]; + int ret; + + if (table_index >= NUMROWS_INGRESSPRECTLFRECORD) + return -EINVAL; + + /* If the row that we want to read is odd, first read the previous even + * row, throw that value away, and finally read the desired row. + * This is a workaround for EUR devices that allows us to read + * odd-numbered rows. For HHD devices: this workaround will not work, + * so don't bother; odd-numbered rows are not readable. + */ + if ((table_index % 2) > 0) { + ret = get_raw_ingress_record(hw, packed_record, 6, 0, + ROWOFFSET_INGRESSPRECTLFRECORD + + table_index - 1); + if (unlikely(ret)) + return ret; + } + + ret = get_raw_ingress_record(hw, packed_record, 6, 0, + ROWOFFSET_INGRESSPRECTLFRECORD + + table_index); + if (unlikely(ret)) + return ret; + + rec->sa_da[0] = packed_record[0]; + rec->sa_da[0] |= packed_record[1] << 16; + + rec->sa_da[1] = packed_record[2]; + + rec->eth_type = packed_record[3]; + + rec->match_mask = packed_record[4]; + + rec->match_type = packed_record[5] & 0xF; + + rec->action = (packed_record[5] >> 4) & 0x1; + + return 0; +} + +int aq_mss_get_ingress_prectlf_record(struct aq_hw_s *hw, + struct aq_mss_ingress_prectlf_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_ingress_prectlf_record, hw, rec, + table_index); +} + +static int +set_ingress_preclass_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_preclass_record *rec, + u16 table_index) +{ + u16 packed_record[20]; + + if (table_index >= NUMROWS_INGRESSPRECLASSRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 20); + + packed_record[0] = rec->sci[0] & 0xFFFF; + packed_record[1] = (rec->sci[0] >> 16) & 0xFFFF; + + packed_record[2] = rec->sci[1] & 0xFFFF; + packed_record[3] = (rec->sci[1] >> 16) & 0xFFFF; + + packed_record[4] = rec->tci & 0xFF; + + packed_record[4] |= (rec->encr_offset & 0xFF) << 8; + + packed_record[5] = rec->eth_type & 0xFFFF; + + packed_record[6] = rec->snap[0] & 0xFFFF; + packed_record[7] = (rec->snap[0] >> 16) & 0xFFFF; + + packed_record[8] = rec->snap[1] & 0xFF; + + packed_record[8] |= (rec->llc & 0xFF) << 8; + packed_record[9] = (rec->llc >> 8) & 0xFFFF; + + packed_record[10] = rec->mac_sa[0] & 0xFFFF; + packed_record[11] = (rec->mac_sa[0] >> 16) & 0xFFFF; + + packed_record[12] = rec->mac_sa[1] & 0xFFFF; + + packed_record[13] = rec->mac_da[0] & 0xFFFF; + packed_record[14] = (rec->mac_da[0] >> 16) & 0xFFFF; + + packed_record[15] = rec->mac_da[1] & 0xFFFF; + + packed_record[16] = rec->lpbk_packet & 0x1; + + packed_record[16] |= (rec->an_mask & 0x3) << 1; + + packed_record[16] |= (rec->tci_mask & 0x3F) << 3; + + packed_record[16] |= (rec->sci_mask & 0x7F) << 9; + packed_record[17] = (rec->sci_mask >> 7) & 0x1; + + packed_record[17] |= (rec->eth_type_mask & 0x3) << 1; + + packed_record[17] |= (rec->snap_mask & 0x1F) << 3; + + packed_record[17] |= (rec->llc_mask & 0x7) << 8; + + packed_record[17] |= (rec->_802_2_encapsulate & 0x1) << 11; + + packed_record[17] |= (rec->sa_mask & 0xF) << 12; + packed_record[18] = (rec->sa_mask >> 4) & 0x3; + + packed_record[18] |= (rec->da_mask & 0x3F) << 2; + + packed_record[18] |= (rec->lpbk_mask & 0x1) << 8; + + packed_record[18] |= (rec->sc_idx & 0x1F) << 9; + + packed_record[18] |= (rec->proc_dest & 0x1) << 14; + + packed_record[18] |= (rec->action & 0x1) << 15; + packed_record[19] = (rec->action >> 1) & 0x1; + + packed_record[19] |= (rec->ctrl_unctrl & 0x1) << 1; + + packed_record[19] |= (rec->sci_from_table & 0x1) << 2; + + packed_record[19] |= (rec->reserved & 0xF) << 3; + + packed_record[19] |= (rec->valid & 0x1) << 7; + + return set_raw_ingress_record(hw, packed_record, 20, 1, + ROWOFFSET_INGRESSPRECLASSRECORD + + table_index); +} + +int aq_mss_set_ingress_preclass_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_preclass_record *rec, + u16 table_index) +{ + int err = AQ_API_CALL_SAFE(set_ingress_preclass_record, hw, rec, + table_index); + + WARN_ONCE(err, "%s failed with %d\n", __func__, err); + + return err; +} + +static int +get_ingress_preclass_record(struct aq_hw_s *hw, + struct aq_mss_ingress_preclass_record *rec, + u16 table_index) +{ + u16 packed_record[20]; + int ret; + + if (table_index >= NUMROWS_INGRESSPRECLASSRECORD) + return -EINVAL; + + /* If the row that we want to read is odd, first read the previous even + * row, throw that value away, and finally read the desired row. + */ + if ((table_index % 2) > 0) { + ret = get_raw_ingress_record(hw, packed_record, 20, 1, + ROWOFFSET_INGRESSPRECLASSRECORD + + table_index - 1); + if (unlikely(ret)) + return ret; + } + + ret = get_raw_ingress_record(hw, packed_record, 20, 1, + ROWOFFSET_INGRESSPRECLASSRECORD + + table_index); + if (unlikely(ret)) + return ret; + + rec->sci[0] = packed_record[0]; + rec->sci[0] |= packed_record[1] << 16; + + rec->sci[1] = packed_record[2]; + rec->sci[1] |= packed_record[3] << 16; + + rec->tci = packed_record[4] & 0xFF; + + rec->encr_offset = (packed_record[4] >> 8) & 0xFF; + + rec->eth_type = packed_record[5]; + + rec->snap[0] = packed_record[6]; + rec->snap[0] |= packed_record[7] << 16; + + rec->snap[1] = packed_record[8] & 0xFF; + + rec->llc = (packed_record[8] >> 8) & 0xFF; + rec->llc = packed_record[9] << 8; + + rec->mac_sa[0] = packed_record[10]; + rec->mac_sa[0] |= packed_record[11] << 16; + + rec->mac_sa[1] = packed_record[12]; + + rec->mac_da[0] = packed_record[13]; + rec->mac_da[0] |= packed_record[14] << 16; + + rec->mac_da[1] = packed_record[15]; + + rec->lpbk_packet = packed_record[16] & 0x1; + + rec->an_mask = (packed_record[16] >> 1) & 0x3; + + rec->tci_mask = (packed_record[16] >> 3) & 0x3F; + + rec->sci_mask = (packed_record[16] >> 9) & 0x7F; + rec->sci_mask |= (packed_record[17] & 0x1) << 7; + + rec->eth_type_mask = (packed_record[17] >> 1) & 0x3; + + rec->snap_mask = (packed_record[17] >> 3) & 0x1F; + + rec->llc_mask = (packed_record[17] >> 8) & 0x7; + + rec->_802_2_encapsulate = (packed_record[17] >> 11) & 0x1; + + rec->sa_mask = (packed_record[17] >> 12) & 0xF; + rec->sa_mask |= (packed_record[18] & 0x3) << 4; + + rec->da_mask = (packed_record[18] >> 2) & 0x3F; + + rec->lpbk_mask = (packed_record[18] >> 8) & 0x1; + + rec->sc_idx = (packed_record[18] >> 9) & 0x1F; + + rec->proc_dest = (packed_record[18] >> 14) & 0x1; + + rec->action = (packed_record[18] >> 15) & 0x1; + rec->action |= (packed_record[19] & 0x1) << 1; + + rec->ctrl_unctrl = (packed_record[19] >> 1) & 0x1; + + rec->sci_from_table = (packed_record[19] >> 2) & 0x1; + + rec->reserved = (packed_record[19] >> 3) & 0xF; + + rec->valid = (packed_record[19] >> 7) & 0x1; + + return 0; +} + +int aq_mss_get_ingress_preclass_record(struct aq_hw_s *hw, + struct aq_mss_ingress_preclass_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_ingress_preclass_record, hw, rec, + table_index); +} + +static int set_ingress_sc_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sc_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + + if (table_index >= NUMROWS_INGRESSSCRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 8); + + packed_record[0] = rec->stop_time & 0xFFFF; + packed_record[1] = (rec->stop_time >> 16) & 0xFFFF; + + packed_record[2] = rec->start_time & 0xFFFF; + packed_record[3] = (rec->start_time >> 16) & 0xFFFF; + + packed_record[4] = rec->validate_frames & 0x3; + + packed_record[4] |= (rec->replay_protect & 0x1) << 2; + + packed_record[4] |= (rec->anti_replay_window & 0x1FFF) << 3; + packed_record[5] = (rec->anti_replay_window >> 13) & 0xFFFF; + packed_record[6] = (rec->anti_replay_window >> 29) & 0x7; + + packed_record[6] |= (rec->receiving & 0x1) << 3; + + packed_record[6] |= (rec->fresh & 0x1) << 4; + + packed_record[6] |= (rec->an_rol & 0x1) << 5; + + packed_record[6] |= (rec->reserved & 0x3FF) << 6; + packed_record[7] = (rec->reserved >> 10) & 0x7FFF; + + packed_record[7] |= (rec->valid & 0x1) << 15; + + return set_raw_ingress_record(hw, packed_record, 8, 3, + ROWOFFSET_INGRESSSCRECORD + table_index); +} + +int aq_mss_set_ingress_sc_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sc_record *rec, + u16 table_index) +{ + int err = AQ_API_CALL_SAFE(set_ingress_sc_record, hw, rec, table_index); + + WARN_ONCE(err, "%s failed with %d\n", __func__, err); + + return err; +} + +static int get_ingress_sc_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sc_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + int ret; + + if (table_index >= NUMROWS_INGRESSSCRECORD) + return -EINVAL; + + ret = get_raw_ingress_record(hw, packed_record, 8, 3, + ROWOFFSET_INGRESSSCRECORD + table_index); + if (unlikely(ret)) + return ret; + + rec->stop_time = packed_record[0]; + rec->stop_time |= packed_record[1] << 16; + + rec->start_time = packed_record[2]; + rec->start_time |= packed_record[3] << 16; + + rec->validate_frames = packed_record[4] & 0x3; + + rec->replay_protect = (packed_record[4] >> 2) & 0x1; + + rec->anti_replay_window = (packed_record[4] >> 3) & 0x1FFF; + rec->anti_replay_window |= packed_record[5] << 13; + rec->anti_replay_window |= (packed_record[6] & 0x7) << 29; + + rec->receiving = (packed_record[6] >> 3) & 0x1; + + rec->fresh = (packed_record[6] >> 4) & 0x1; + + rec->an_rol = (packed_record[6] >> 5) & 0x1; + + rec->reserved = (packed_record[6] >> 6) & 0x3FF; + rec->reserved |= (packed_record[7] & 0x7FFF) << 10; + + rec->valid = (packed_record[7] >> 15) & 0x1; + + return 0; +} + +int aq_mss_get_ingress_sc_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sc_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_ingress_sc_record, hw, rec, table_index); +} + +static int set_ingress_sa_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sa_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + + if (table_index >= NUMROWS_INGRESSSARECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 8); + + packed_record[0] = rec->stop_time & 0xFFFF; + packed_record[1] = (rec->stop_time >> 16) & 0xFFFF; + + packed_record[2] = rec->start_time & 0xFFFF; + packed_record[3] = (rec->start_time >> 16) & 0xFFFF; + + packed_record[4] = rec->next_pn & 0xFFFF; + packed_record[5] = (rec->next_pn >> 16) & 0xFFFF; + + packed_record[6] = rec->sat_nextpn & 0x1; + + packed_record[6] |= (rec->in_use & 0x1) << 1; + + packed_record[6] |= (rec->fresh & 0x1) << 2; + + packed_record[6] |= (rec->reserved & 0x1FFF) << 3; + packed_record[7] = (rec->reserved >> 13) & 0x7FFF; + + packed_record[7] |= (rec->valid & 0x1) << 15; + + return set_raw_ingress_record(hw, packed_record, 8, 3, + ROWOFFSET_INGRESSSARECORD + table_index); +} + +int aq_mss_set_ingress_sa_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sa_record *rec, + u16 table_index) +{ + int err = AQ_API_CALL_SAFE(set_ingress_sa_record, hw, rec, table_index); + + WARN_ONCE(err, "%s failed with %d\n", __func__, err); + + return err; +} + +static int get_ingress_sa_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sa_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + int ret; + + if (table_index >= NUMROWS_INGRESSSARECORD) + return -EINVAL; + + ret = get_raw_ingress_record(hw, packed_record, 8, 3, + ROWOFFSET_INGRESSSARECORD + table_index); + if (unlikely(ret)) + return ret; + + rec->stop_time = packed_record[0]; + rec->stop_time |= packed_record[1] << 16; + + rec->start_time = packed_record[2]; + rec->start_time |= packed_record[3] << 16; + + rec->next_pn = packed_record[4]; + rec->next_pn |= packed_record[5] << 16; + + rec->sat_nextpn = packed_record[6] & 0x1; + + rec->in_use = (packed_record[6] >> 1) & 0x1; + + rec->fresh = (packed_record[6] >> 2) & 0x1; + + rec->reserved = (packed_record[6] >> 3) & 0x1FFF; + rec->reserved |= (packed_record[7] & 0x7FFF) << 13; + + rec->valid = (packed_record[7] >> 15) & 0x1; + + return 0; +} + +int aq_mss_get_ingress_sa_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sa_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_ingress_sa_record, hw, rec, table_index); +} + +static int +set_ingress_sakey_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sakey_record *rec, + u16 table_index) +{ + u16 packed_record[18]; + + if (table_index >= NUMROWS_INGRESSSAKEYRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 18); + + packed_record[0] = rec->key[0] & 0xFFFF; + packed_record[1] = (rec->key[0] >> 16) & 0xFFFF; + + packed_record[2] = rec->key[1] & 0xFFFF; + packed_record[3] = (rec->key[1] >> 16) & 0xFFFF; + + packed_record[4] = rec->key[2] & 0xFFFF; + packed_record[5] = (rec->key[2] >> 16) & 0xFFFF; + + packed_record[6] = rec->key[3] & 0xFFFF; + packed_record[7] = (rec->key[3] >> 16) & 0xFFFF; + + packed_record[8] = rec->key[4] & 0xFFFF; + packed_record[9] = (rec->key[4] >> 16) & 0xFFFF; + + packed_record[10] = rec->key[5] & 0xFFFF; + packed_record[11] = (rec->key[5] >> 16) & 0xFFFF; + + packed_record[12] = rec->key[6] & 0xFFFF; + packed_record[13] = (rec->key[6] >> 16) & 0xFFFF; + + packed_record[14] = rec->key[7] & 0xFFFF; + packed_record[15] = (rec->key[7] >> 16) & 0xFFFF; + + packed_record[16] = rec->key_len & 0x3; + + return set_raw_ingress_record(hw, packed_record, 18, 2, + ROWOFFSET_INGRESSSAKEYRECORD + + table_index); +} + +int aq_mss_set_ingress_sakey_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sakey_record *rec, + u16 table_index) +{ + int err = AQ_API_CALL_SAFE(set_ingress_sakey_record, hw, rec, + table_index); + + WARN_ONCE(err, "%s failed with %d\n", __func__, err); + + return err; +} + +static int get_ingress_sakey_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sakey_record *rec, + u16 table_index) +{ + u16 packed_record[18]; + int ret; + + if (table_index >= NUMROWS_INGRESSSAKEYRECORD) + return -EINVAL; + + ret = get_raw_ingress_record(hw, packed_record, 18, 2, + ROWOFFSET_INGRESSSAKEYRECORD + + table_index); + if (unlikely(ret)) + return ret; + + rec->key[0] = packed_record[0]; + rec->key[0] |= packed_record[1] << 16; + + rec->key[1] = packed_record[2]; + rec->key[1] |= packed_record[3] << 16; + + rec->key[2] = packed_record[4]; + rec->key[2] |= packed_record[5] << 16; + + rec->key[3] = packed_record[6]; + rec->key[3] |= packed_record[7] << 16; + + rec->key[4] = packed_record[8]; + rec->key[4] |= packed_record[9] << 16; + + rec->key[5] = packed_record[10]; + rec->key[5] |= packed_record[11] << 16; + + rec->key[6] = packed_record[12]; + rec->key[6] |= packed_record[13] << 16; + + rec->key[7] = packed_record[14]; + rec->key[7] |= packed_record[15] << 16; + + rec->key_len = (rec->key_len & 0xFFFFFFFC) | + (packed_record[16] & 0x3); + + return 0; +} + +int aq_mss_get_ingress_sakey_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sakey_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_ingress_sakey_record, hw, rec, table_index); +} + +static int +set_ingress_postclass_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_postclass_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + + if (table_index >= NUMROWS_INGRESSPOSTCLASSRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 8); + + packed_record[0] = rec->byte0 & 0xFF; + + packed_record[0] |= (rec->byte1 & 0xFF) << 8; + + packed_record[1] = rec->byte2 & 0xFF; + + packed_record[1] |= (rec->byte3 & 0xFF) << 8; + + packed_record[2] = rec->eth_type & 0xFFFF; + + packed_record[3] = rec->eth_type_valid & 0x1; + + packed_record[3] |= (rec->vlan_id & 0xFFF) << 1; + + packed_record[3] |= (rec->vlan_up & 0x7) << 13; + + packed_record[4] = rec->vlan_valid & 0x1; + + packed_record[4] |= (rec->sai & 0x1F) << 1; + + packed_record[4] |= (rec->sai_hit & 0x1) << 6; + + packed_record[4] |= (rec->eth_type_mask & 0xF) << 7; + + packed_record[4] |= (rec->byte3_location & 0x1F) << 11; + packed_record[5] = (rec->byte3_location >> 5) & 0x1; + + packed_record[5] |= (rec->byte3_mask & 0x3) << 1; + + packed_record[5] |= (rec->byte2_location & 0x3F) << 3; + + packed_record[5] |= (rec->byte2_mask & 0x3) << 9; + + packed_record[5] |= (rec->byte1_location & 0x1F) << 11; + packed_record[6] = (rec->byte1_location >> 5) & 0x1; + + packed_record[6] |= (rec->byte1_mask & 0x3) << 1; + + packed_record[6] |= (rec->byte0_location & 0x3F) << 3; + + packed_record[6] |= (rec->byte0_mask & 0x3) << 9; + + packed_record[6] |= (rec->eth_type_valid_mask & 0x3) << 11; + + packed_record[6] |= (rec->vlan_id_mask & 0x7) << 13; + packed_record[7] = (rec->vlan_id_mask >> 3) & 0x1; + + packed_record[7] |= (rec->vlan_up_mask & 0x3) << 1; + + packed_record[7] |= (rec->vlan_valid_mask & 0x3) << 3; + + packed_record[7] |= (rec->sai_mask & 0x3) << 5; + + packed_record[7] |= (rec->sai_hit_mask & 0x3) << 7; + + packed_record[7] |= (rec->firstlevel_actions & 0x1) << 9; + + packed_record[7] |= (rec->secondlevel_actions & 0x1) << 10; + + packed_record[7] |= (rec->reserved & 0xF) << 11; + + packed_record[7] |= (rec->valid & 0x1) << 15; + + return set_raw_ingress_record(hw, packed_record, 8, 4, + ROWOFFSET_INGRESSPOSTCLASSRECORD + + table_index); +} + +int aq_mss_set_ingress_postclass_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_postclass_record *rec, + u16 table_index) +{ + return AQ_API_CALL_SAFE(set_ingress_postclass_record, hw, rec, + table_index); +} + +static int +get_ingress_postclass_record(struct aq_hw_s *hw, + struct aq_mss_ingress_postclass_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + int ret; + + if (table_index >= NUMROWS_INGRESSPOSTCLASSRECORD) + return -EINVAL; + + /* If the row that we want to read is odd, first read the previous even + * row, throw that value away, and finally read the desired row. + */ + if ((table_index % 2) > 0) { + ret = get_raw_ingress_record(hw, packed_record, 8, 4, + ROWOFFSET_INGRESSPOSTCLASSRECORD + + table_index - 1); + if (unlikely(ret)) + return ret; + } + + ret = get_raw_ingress_record(hw, packed_record, 8, 4, + ROWOFFSET_INGRESSPOSTCLASSRECORD + + table_index); + if (unlikely(ret)) + return ret; + + rec->byte0 = packed_record[0] & 0xFF; + + rec->byte1 = (packed_record[0] >> 8) & 0xFF; + + rec->byte2 = packed_record[1] & 0xFF; + + rec->byte3 = (packed_record[1] >> 8) & 0xFF; + + rec->eth_type = packed_record[2]; + + rec->eth_type_valid = packed_record[3] & 0x1; + + rec->vlan_id = (packed_record[3] >> 1) & 0xFFF; + + rec->vlan_up = (packed_record[3] >> 13) & 0x7; + + rec->vlan_valid = packed_record[4] & 0x1; + + rec->sai = (packed_record[4] >> 1) & 0x1F; + + rec->sai_hit = (packed_record[4] >> 6) & 0x1; + + rec->eth_type_mask = (packed_record[4] >> 7) & 0xF; + + rec->byte3_location = (packed_record[4] >> 11) & 0x1F; + rec->byte3_location |= (packed_record[5] & 0x1) << 5; + + rec->byte3_mask = (packed_record[5] >> 1) & 0x3; + + rec->byte2_location = (packed_record[5] >> 3) & 0x3F; + + rec->byte2_mask = (packed_record[5] >> 9) & 0x3; + + rec->byte1_location = (packed_record[5] >> 11) & 0x1F; + rec->byte1_location |= (packed_record[6] & 0x1) << 5; + + rec->byte1_mask = (packed_record[6] >> 1) & 0x3; + + rec->byte0_location = (packed_record[6] >> 3) & 0x3F; + + rec->byte0_mask = (packed_record[6] >> 9) & 0x3; + + rec->eth_type_valid_mask = (packed_record[6] >> 11) & 0x3; + + rec->vlan_id_mask = (packed_record[6] >> 13) & 0x7; + rec->vlan_id_mask |= (packed_record[7] & 0x1) << 3; + + rec->vlan_up_mask = (packed_record[7] >> 1) & 0x3; + + rec->vlan_valid_mask = (packed_record[7] >> 3) & 0x3; + + rec->sai_mask = (packed_record[7] >> 5) & 0x3; + + rec->sai_hit_mask = (packed_record[7] >> 7) & 0x3; + + rec->firstlevel_actions = (packed_record[7] >> 9) & 0x1; + + rec->secondlevel_actions = (packed_record[7] >> 10) & 0x1; + + rec->reserved = (packed_record[7] >> 11) & 0xF; + + rec->valid = (packed_record[7] >> 15) & 0x1; + + return 0; +} + +int aq_mss_get_ingress_postclass_record(struct aq_hw_s *hw, + struct aq_mss_ingress_postclass_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_ingress_postclass_record, hw, rec, + table_index); +} + +static int +set_ingress_postctlf_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_postctlf_record *rec, + u16 table_index) +{ + u16 packed_record[6]; + + if (table_index >= NUMROWS_INGRESSPOSTCTLFRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 6); + + packed_record[0] = rec->sa_da[0] & 0xFFFF; + packed_record[1] = (rec->sa_da[0] >> 16) & 0xFFFF; + + packed_record[2] = rec->sa_da[1] & 0xFFFF; + + packed_record[3] = rec->eth_type & 0xFFFF; + + packed_record[4] = rec->match_mask & 0xFFFF; + + packed_record[5] = rec->match_type & 0xF; + + packed_record[5] |= (rec->action & 0x1) << 4; + + return set_raw_ingress_record(hw, packed_record, 6, 5, + ROWOFFSET_INGRESSPOSTCTLFRECORD + + table_index); +} + +int aq_mss_set_ingress_postctlf_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_postctlf_record *rec, + u16 table_index) +{ + return AQ_API_CALL_SAFE(set_ingress_postctlf_record, hw, rec, + table_index); +} + +static int +get_ingress_postctlf_record(struct aq_hw_s *hw, + struct aq_mss_ingress_postctlf_record *rec, + u16 table_index) +{ + u16 packed_record[6]; + int ret; + + if (table_index >= NUMROWS_INGRESSPOSTCTLFRECORD) + return -EINVAL; + + /* If the row that we want to read is odd, first read the previous even + * row, throw that value away, and finally read the desired row. + */ + if ((table_index % 2) > 0) { + ret = get_raw_ingress_record(hw, packed_record, 6, 5, + ROWOFFSET_INGRESSPOSTCTLFRECORD + + table_index - 1); + if (unlikely(ret)) + return ret; + } + + ret = get_raw_ingress_record(hw, packed_record, 6, 5, + ROWOFFSET_INGRESSPOSTCTLFRECORD + + table_index); + if (unlikely(ret)) + return ret; + + rec->sa_da[0] = packed_record[0]; + rec->sa_da[0] |= packed_record[1] << 16; + + rec->sa_da[1] = packed_record[2]; + + rec->eth_type = packed_record[3]; + + rec->match_mask = packed_record[4]; + + rec->match_type = packed_record[5] & 0xF; + + rec->action = (packed_record[5] >> 4) & 0x1; + + return 0; +} + +int aq_mss_get_ingress_postctlf_record(struct aq_hw_s *hw, + struct aq_mss_ingress_postctlf_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_ingress_postctlf_record, hw, rec, + table_index); +} + +static int set_egress_ctlf_record(struct aq_hw_s *hw, + const struct aq_mss_egress_ctlf_record *rec, + u16 table_index) +{ + u16 packed_record[6]; + + if (table_index >= NUMROWS_EGRESSCTLFRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 6); + + packed_record[0] = rec->sa_da[0] & 0xFFFF; + packed_record[1] = (rec->sa_da[0] >> 16) & 0xFFFF; + packed_record[2] = rec->sa_da[1] & 0xFFFF; + + packed_record[3] = rec->eth_type & 0xFFFF; + + packed_record[4] = rec->match_mask & 0xFFFF; + + packed_record[5] = rec->match_type & 0xF; + + packed_record[5] |= (rec->action & 0x1) << 4; + + return set_raw_egress_record(hw, packed_record, 6, 0, + ROWOFFSET_EGRESSCTLFRECORD + table_index); +} + +int aq_mss_set_egress_ctlf_record(struct aq_hw_s *hw, + const struct aq_mss_egress_ctlf_record *rec, + u16 table_index) +{ + return AQ_API_CALL_SAFE(set_egress_ctlf_record, hw, rec, table_index); +} + +static int get_egress_ctlf_record(struct aq_hw_s *hw, + struct aq_mss_egress_ctlf_record *rec, + u16 table_index) +{ + u16 packed_record[6]; + int ret; + + if (table_index >= NUMROWS_EGRESSCTLFRECORD) + return -EINVAL; + + /* If the row that we want to read is odd, first read the previous even + * row, throw that value away, and finally read the desired row. + */ + if ((table_index % 2) > 0) { + ret = get_raw_egress_record(hw, packed_record, 6, 0, + ROWOFFSET_EGRESSCTLFRECORD + + table_index - 1); + if (unlikely(ret)) + return ret; + } + + ret = get_raw_egress_record(hw, packed_record, 6, 0, + ROWOFFSET_EGRESSCTLFRECORD + table_index); + if (unlikely(ret)) + return ret; + + rec->sa_da[0] = packed_record[0]; + rec->sa_da[0] |= packed_record[1] << 16; + + rec->sa_da[1] = packed_record[2]; + + rec->eth_type = packed_record[3]; + + rec->match_mask = packed_record[4]; + + rec->match_type = packed_record[5] & 0xF; + + rec->action = (packed_record[5] >> 4) & 0x1; + + return 0; +} + +int aq_mss_get_egress_ctlf_record(struct aq_hw_s *hw, + struct aq_mss_egress_ctlf_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_egress_ctlf_record, hw, rec, table_index); +} + +static int set_egress_class_record(struct aq_hw_s *hw, + const struct aq_mss_egress_class_record *rec, + u16 table_index) +{ + u16 packed_record[28]; + + if (table_index >= NUMROWS_EGRESSCLASSRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 28); + + packed_record[0] = rec->vlan_id & 0xFFF; + + packed_record[0] |= (rec->vlan_up & 0x7) << 12; + + packed_record[0] |= (rec->vlan_valid & 0x1) << 15; + + packed_record[1] = rec->byte3 & 0xFF; + + packed_record[1] |= (rec->byte2 & 0xFF) << 8; + + packed_record[2] = rec->byte1 & 0xFF; + + packed_record[2] |= (rec->byte0 & 0xFF) << 8; + + packed_record[3] = rec->tci & 0xFF; + + packed_record[3] |= (rec->sci[0] & 0xFF) << 8; + packed_record[4] = (rec->sci[0] >> 8) & 0xFFFF; + packed_record[5] = (rec->sci[0] >> 24) & 0xFF; + + packed_record[5] |= (rec->sci[1] & 0xFF) << 8; + packed_record[6] = (rec->sci[1] >> 8) & 0xFFFF; + packed_record[7] = (rec->sci[1] >> 24) & 0xFF; + + packed_record[7] |= (rec->eth_type & 0xFF) << 8; + packed_record[8] = (rec->eth_type >> 8) & 0xFF; + + packed_record[8] |= (rec->snap[0] & 0xFF) << 8; + packed_record[9] = (rec->snap[0] >> 8) & 0xFFFF; + packed_record[10] = (rec->snap[0] >> 24) & 0xFF; + + packed_record[10] |= (rec->snap[1] & 0xFF) << 8; + + packed_record[11] = rec->llc & 0xFFFF; + packed_record[12] = (rec->llc >> 16) & 0xFF; + + packed_record[12] |= (rec->mac_sa[0] & 0xFF) << 8; + packed_record[13] = (rec->mac_sa[0] >> 8) & 0xFFFF; + packed_record[14] = (rec->mac_sa[0] >> 24) & 0xFF; + + packed_record[14] |= (rec->mac_sa[1] & 0xFF) << 8; + packed_record[15] = (rec->mac_sa[1] >> 8) & 0xFF; + + packed_record[15] |= (rec->mac_da[0] & 0xFF) << 8; + packed_record[16] = (rec->mac_da[0] >> 8) & 0xFFFF; + packed_record[17] = (rec->mac_da[0] >> 24) & 0xFF; + + packed_record[17] |= (rec->mac_da[1] & 0xFF) << 8; + packed_record[18] = (rec->mac_da[1] >> 8) & 0xFF; + + packed_record[18] |= (rec->pn & 0xFF) << 8; + packed_record[19] = (rec->pn >> 8) & 0xFFFF; + packed_record[20] = (rec->pn >> 24) & 0xFF; + + packed_record[20] |= (rec->byte3_location & 0x3F) << 8; + + packed_record[20] |= (rec->byte3_mask & 0x1) << 14; + + packed_record[20] |= (rec->byte2_location & 0x1) << 15; + packed_record[21] = (rec->byte2_location >> 1) & 0x1F; + + packed_record[21] |= (rec->byte2_mask & 0x1) << 5; + + packed_record[21] |= (rec->byte1_location & 0x3F) << 6; + + packed_record[21] |= (rec->byte1_mask & 0x1) << 12; + + packed_record[21] |= (rec->byte0_location & 0x7) << 13; + packed_record[22] = (rec->byte0_location >> 3) & 0x7; + + packed_record[22] |= (rec->byte0_mask & 0x1) << 3; + + packed_record[22] |= (rec->vlan_id_mask & 0x3) << 4; + + packed_record[22] |= (rec->vlan_up_mask & 0x1) << 6; + + packed_record[22] |= (rec->vlan_valid_mask & 0x1) << 7; + + packed_record[22] |= (rec->tci_mask & 0xFF) << 8; + + packed_record[23] = rec->sci_mask & 0xFF; + + packed_record[23] |= (rec->eth_type_mask & 0x3) << 8; + + packed_record[23] |= (rec->snap_mask & 0x1F) << 10; + + packed_record[23] |= (rec->llc_mask & 0x1) << 15; + packed_record[24] = (rec->llc_mask >> 1) & 0x3; + + packed_record[24] |= (rec->sa_mask & 0x3F) << 2; + + packed_record[24] |= (rec->da_mask & 0x3F) << 8; + + packed_record[24] |= (rec->pn_mask & 0x3) << 14; + packed_record[25] = (rec->pn_mask >> 2) & 0x3; + + packed_record[25] |= (rec->eight02dot2 & 0x1) << 2; + + packed_record[25] |= (rec->tci_sc & 0x1) << 3; + + packed_record[25] |= (rec->tci_87543 & 0x1) << 4; + + packed_record[25] |= (rec->exp_sectag_en & 0x1) << 5; + + packed_record[25] |= (rec->sc_idx & 0x1F) << 6; + + packed_record[25] |= (rec->sc_sa & 0x3) << 11; + + packed_record[25] |= (rec->debug & 0x1) << 13; + + packed_record[25] |= (rec->action & 0x3) << 14; + + packed_record[26] = (rec->valid & 0x1) << 3; + + return set_raw_egress_record(hw, packed_record, 28, 1, + ROWOFFSET_EGRESSCLASSRECORD + table_index); +} + +int aq_mss_set_egress_class_record(struct aq_hw_s *hw, + const struct aq_mss_egress_class_record *rec, + u16 table_index) +{ + return AQ_API_CALL_SAFE(set_egress_class_record, hw, rec, table_index); +} + +static int get_egress_class_record(struct aq_hw_s *hw, + struct aq_mss_egress_class_record *rec, + u16 table_index) +{ + u16 packed_record[28]; + int ret; + + if (table_index >= NUMROWS_EGRESSCLASSRECORD) + return -EINVAL; + + /* If the row that we want to read is odd, first read the previous even + * row, throw that value away, and finally read the desired row. + */ + if ((table_index % 2) > 0) { + ret = get_raw_egress_record(hw, packed_record, 28, 1, + ROWOFFSET_EGRESSCLASSRECORD + + table_index - 1); + if (unlikely(ret)) + return ret; + } + + ret = get_raw_egress_record(hw, packed_record, 28, 1, + ROWOFFSET_EGRESSCLASSRECORD + table_index); + if (unlikely(ret)) + return ret; + + rec->vlan_id = packed_record[0] & 0xFFF; + + rec->vlan_up = (packed_record[0] >> 12) & 0x7; + + rec->vlan_valid = (packed_record[0] >> 15) & 0x1; + + rec->byte3 = packed_record[1] & 0xFF; + + rec->byte2 = (packed_record[1] >> 8) & 0xFF; + + rec->byte1 = packed_record[2] & 0xFF; + + rec->byte0 = (packed_record[2] >> 8) & 0xFF; + + rec->tci = packed_record[3] & 0xFF; + + rec->sci[0] = (packed_record[3] >> 8) & 0xFF; + rec->sci[0] |= packed_record[4] << 8; + rec->sci[0] |= (packed_record[5] & 0xFF) << 24; + + rec->sci[1] = (packed_record[5] >> 8) & 0xFF; + rec->sci[1] |= packed_record[6] << 8; + rec->sci[1] |= (packed_record[7] & 0xFF) << 24; + + rec->eth_type = (packed_record[7] >> 8) & 0xFF; + rec->eth_type |= (packed_record[8] & 0xFF) << 8; + + rec->snap[0] = (packed_record[8] >> 8) & 0xFF; + rec->snap[0] |= packed_record[9] << 8; + rec->snap[0] |= (packed_record[10] & 0xFF) << 24; + + rec->snap[1] = (packed_record[10] >> 8) & 0xFF; + + rec->llc = packed_record[11]; + rec->llc |= (packed_record[12] & 0xFF) << 16; + + rec->mac_sa[0] = (packed_record[12] >> 8) & 0xFF; + rec->mac_sa[0] |= packed_record[13] << 8; + rec->mac_sa[0] |= (packed_record[14] & 0xFF) << 24; + + rec->mac_sa[1] = (packed_record[14] >> 8) & 0xFF; + rec->mac_sa[1] |= (packed_record[15] & 0xFF) << 8; + + rec->mac_da[0] = (packed_record[15] >> 8) & 0xFF; + rec->mac_da[0] |= packed_record[16] << 8; + rec->mac_da[0] |= (packed_record[17] & 0xFF) << 24; + + rec->mac_da[1] = (packed_record[17] >> 8) & 0xFF; + rec->mac_da[1] |= (packed_record[18] & 0xFF) << 8; + + rec->pn = (packed_record[18] >> 8) & 0xFF; + rec->pn |= packed_record[19] << 8; + rec->pn |= (packed_record[20] & 0xFF) << 24; + + rec->byte3_location = (packed_record[20] >> 8) & 0x3F; + + rec->byte3_mask = (packed_record[20] >> 14) & 0x1; + + rec->byte2_location = (packed_record[20] >> 15) & 0x1; + rec->byte2_location |= (packed_record[21] & 0x1F) << 1; + + rec->byte2_mask = (packed_record[21] >> 5) & 0x1; + + rec->byte1_location = (packed_record[21] >> 6) & 0x3F; + + rec->byte1_mask = (packed_record[21] >> 12) & 0x1; + + rec->byte0_location = (packed_record[21] >> 13) & 0x7; + rec->byte0_location |= (packed_record[22] & 0x7) << 3; + + rec->byte0_mask = (packed_record[22] >> 3) & 0x1; + + rec->vlan_id_mask = (packed_record[22] >> 4) & 0x3; + + rec->vlan_up_mask = (packed_record[22] >> 6) & 0x1; + + rec->vlan_valid_mask = (packed_record[22] >> 7) & 0x1; + + rec->tci_mask = (packed_record[22] >> 8) & 0xFF; + + rec->sci_mask = packed_record[23] & 0xFF; + + rec->eth_type_mask = (packed_record[23] >> 8) & 0x3; + + rec->snap_mask = (packed_record[23] >> 10) & 0x1F; + + rec->llc_mask = (packed_record[23] >> 15) & 0x1; + rec->llc_mask |= (packed_record[24] & 0x3) << 1; + + rec->sa_mask = (packed_record[24] >> 2) & 0x3F; + + rec->da_mask = (packed_record[24] >> 8) & 0x3F; + + rec->pn_mask = (packed_record[24] >> 14) & 0x3; + rec->pn_mask |= (packed_record[25] & 0x3) << 2; + + rec->eight02dot2 = (packed_record[25] >> 2) & 0x1; + + rec->tci_sc = (packed_record[25] >> 3) & 0x1; + + rec->tci_87543 = (packed_record[25] >> 4) & 0x1; + + rec->exp_sectag_en = (packed_record[25] >> 5) & 0x1; + + rec->sc_idx = (packed_record[25] >> 6) & 0x1F; + + rec->sc_sa = (packed_record[25] >> 11) & 0x3; + + rec->debug = (packed_record[25] >> 13) & 0x1; + + rec->action = (packed_record[25] >> 14) & 0x3; + + rec->valid = (packed_record[26] >> 3) & 0x1; + + return 0; +} + +int aq_mss_get_egress_class_record(struct aq_hw_s *hw, + struct aq_mss_egress_class_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_egress_class_record, hw, rec, table_index); +} + +static int set_egress_sc_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sc_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + + if (table_index >= NUMROWS_EGRESSSCRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 8); + + packed_record[0] = rec->start_time & 0xFFFF; + packed_record[1] = (rec->start_time >> 16) & 0xFFFF; + + packed_record[2] = rec->stop_time & 0xFFFF; + packed_record[3] = (rec->stop_time >> 16) & 0xFFFF; + + packed_record[4] = rec->curr_an & 0x3; + + packed_record[4] |= (rec->an_roll & 0x1) << 2; + + packed_record[4] |= (rec->tci & 0x3F) << 3; + + packed_record[4] |= (rec->enc_off & 0x7F) << 9; + packed_record[5] = (rec->enc_off >> 7) & 0x1; + + packed_record[5] |= (rec->protect & 0x1) << 1; + + packed_record[5] |= (rec->recv & 0x1) << 2; + + packed_record[5] |= (rec->fresh & 0x1) << 3; + + packed_record[5] |= (rec->sak_len & 0x3) << 4; + + packed_record[7] |= (rec->valid & 0x1) << 15; + + return set_raw_egress_record(hw, packed_record, 8, 2, + ROWOFFSET_EGRESSSCRECORD + table_index); +} + +int aq_mss_set_egress_sc_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sc_record *rec, + u16 table_index) +{ + return AQ_API_CALL_SAFE(set_egress_sc_record, hw, rec, table_index); +} + +static int get_egress_sc_record(struct aq_hw_s *hw, + struct aq_mss_egress_sc_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + int ret; + + if (table_index >= NUMROWS_EGRESSSCRECORD) + return -EINVAL; + + ret = get_raw_egress_record(hw, packed_record, 8, 2, + ROWOFFSET_EGRESSSCRECORD + table_index); + if (unlikely(ret)) + return ret; + + rec->start_time = packed_record[0]; + rec->start_time |= packed_record[1] << 16; + + rec->stop_time = packed_record[2]; + rec->stop_time |= packed_record[3] << 16; + + rec->curr_an = packed_record[4] & 0x3; + + rec->an_roll = (packed_record[4] >> 2) & 0x1; + + rec->tci = (packed_record[4] >> 3) & 0x3F; + + rec->enc_off = (packed_record[4] >> 9) & 0x7F; + rec->enc_off |= (packed_record[5] & 0x1) << 7; + + rec->protect = (packed_record[5] >> 1) & 0x1; + + rec->recv = (packed_record[5] >> 2) & 0x1; + + rec->fresh = (packed_record[5] >> 3) & 0x1; + + rec->sak_len = (packed_record[5] >> 4) & 0x3; + + rec->valid = (packed_record[7] >> 15) & 0x1; + + return 0; +} + +int aq_mss_get_egress_sc_record(struct aq_hw_s *hw, + struct aq_mss_egress_sc_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_egress_sc_record, hw, rec, table_index); +} + +static int set_egress_sa_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sa_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + + if (table_index >= NUMROWS_EGRESSSARECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 8); + + packed_record[0] = rec->start_time & 0xFFFF; + packed_record[1] = (rec->start_time >> 16) & 0xFFFF; + + packed_record[2] = rec->stop_time & 0xFFFF; + packed_record[3] = (rec->stop_time >> 16) & 0xFFFF; + + packed_record[4] = rec->next_pn & 0xFFFF; + packed_record[5] = (rec->next_pn >> 16) & 0xFFFF; + + packed_record[6] = rec->sat_pn & 0x1; + + packed_record[6] |= (rec->fresh & 0x1) << 1; + + packed_record[7] = (rec->valid & 0x1) << 15; + + return set_raw_egress_record(hw, packed_record, 8, 2, + ROWOFFSET_EGRESSSARECORD + table_index); +} + +int aq_mss_set_egress_sa_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sa_record *rec, + u16 table_index) +{ + int err = AQ_API_CALL_SAFE(set_egress_sa_record, hw, rec, table_index); + + WARN_ONCE(err, "%s failed with %d\n", __func__, err); + + return err; +} + +static int get_egress_sa_record(struct aq_hw_s *hw, + struct aq_mss_egress_sa_record *rec, + u16 table_index) +{ + u16 packed_record[8]; + int ret; + + if (table_index >= NUMROWS_EGRESSSARECORD) + return -EINVAL; + + ret = get_raw_egress_record(hw, packed_record, 8, 2, + ROWOFFSET_EGRESSSARECORD + table_index); + if (unlikely(ret)) + return ret; + + rec->start_time = packed_record[0]; + rec->start_time |= packed_record[1] << 16; + + rec->stop_time = packed_record[2]; + rec->stop_time |= packed_record[3] << 16; + + rec->next_pn = packed_record[4]; + rec->next_pn |= packed_record[5] << 16; + + rec->sat_pn = packed_record[6] & 0x1; + + rec->fresh = (packed_record[6] >> 1) & 0x1; + + rec->valid = (packed_record[7] >> 15) & 0x1; + + return 0; +} + +int aq_mss_get_egress_sa_record(struct aq_hw_s *hw, + struct aq_mss_egress_sa_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_egress_sa_record, hw, rec, table_index); +} + +static int set_egress_sakey_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sakey_record *rec, + u16 table_index) +{ + u16 packed_record[16]; + int ret; + + if (table_index >= NUMROWS_EGRESSSAKEYRECORD) + return -EINVAL; + + memset(packed_record, 0, sizeof(u16) * 16); + + packed_record[0] = rec->key[0] & 0xFFFF; + packed_record[1] = (rec->key[0] >> 16) & 0xFFFF; + + packed_record[2] = rec->key[1] & 0xFFFF; + packed_record[3] = (rec->key[1] >> 16) & 0xFFFF; + + packed_record[4] = rec->key[2] & 0xFFFF; + packed_record[5] = (rec->key[2] >> 16) & 0xFFFF; + + packed_record[6] = rec->key[3] & 0xFFFF; + packed_record[7] = (rec->key[3] >> 16) & 0xFFFF; + + packed_record[8] = rec->key[4] & 0xFFFF; + packed_record[9] = (rec->key[4] >> 16) & 0xFFFF; + + packed_record[10] = rec->key[5] & 0xFFFF; + packed_record[11] = (rec->key[5] >> 16) & 0xFFFF; + + packed_record[12] = rec->key[6] & 0xFFFF; + packed_record[13] = (rec->key[6] >> 16) & 0xFFFF; + + packed_record[14] = rec->key[7] & 0xFFFF; + packed_record[15] = (rec->key[7] >> 16) & 0xFFFF; + + ret = set_raw_egress_record(hw, packed_record, 8, 2, + ROWOFFSET_EGRESSSAKEYRECORD + table_index); + if (unlikely(ret)) + return ret; + ret = set_raw_egress_record(hw, packed_record + 8, 8, 2, + ROWOFFSET_EGRESSSAKEYRECORD + table_index - + 32); + if (unlikely(ret)) + return ret; + + return 0; +} + +int aq_mss_set_egress_sakey_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sakey_record *rec, + u16 table_index) +{ + int err = AQ_API_CALL_SAFE(set_egress_sakey_record, hw, rec, + table_index); + + WARN_ONCE(err, "%s failed with %d\n", __func__, err); + + return err; +} + +static int get_egress_sakey_record(struct aq_hw_s *hw, + struct aq_mss_egress_sakey_record *rec, + u16 table_index) +{ + u16 packed_record[16]; + int ret; + + if (table_index >= NUMROWS_EGRESSSAKEYRECORD) + return -EINVAL; + + ret = get_raw_egress_record(hw, packed_record, 8, 2, + ROWOFFSET_EGRESSSAKEYRECORD + table_index); + if (unlikely(ret)) + return ret; + ret = get_raw_egress_record(hw, packed_record + 8, 8, 2, + ROWOFFSET_EGRESSSAKEYRECORD + table_index - + 32); + if (unlikely(ret)) + return ret; + + rec->key[0] = packed_record[0]; + rec->key[0] |= packed_record[1] << 16; + + rec->key[1] = packed_record[2]; + rec->key[1] |= packed_record[3] << 16; + + rec->key[2] = packed_record[4]; + rec->key[2] |= packed_record[5] << 16; + + rec->key[3] = packed_record[6]; + rec->key[3] |= packed_record[7] << 16; + + rec->key[4] = packed_record[8]; + rec->key[4] |= packed_record[9] << 16; + + rec->key[5] = packed_record[10]; + rec->key[5] |= packed_record[11] << 16; + + rec->key[6] = packed_record[12]; + rec->key[6] |= packed_record[13] << 16; + + rec->key[7] = packed_record[14]; + rec->key[7] |= packed_record[15] << 16; + + return 0; +} + +int aq_mss_get_egress_sakey_record(struct aq_hw_s *hw, + struct aq_mss_egress_sakey_record *rec, + u16 table_index) +{ + memset(rec, 0, sizeof(*rec)); + + return AQ_API_CALL_SAFE(get_egress_sakey_record, hw, rec, table_index); +} + +static int get_egress_sc_counters(struct aq_hw_s *hw, + struct aq_mss_egress_sc_counters *counters, + u16 sc_index) +{ + u16 packed_record[4]; + int ret; + + if (sc_index >= NUMROWS_EGRESSSCRECORD) + return -EINVAL; + + ret = get_raw_egress_record(hw, packed_record, 4, 3, sc_index * 8 + 4); + if (unlikely(ret)) + return ret; + counters->sc_protected_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->sc_protected_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, sc_index * 8 + 5); + if (unlikely(ret)) + return ret; + counters->sc_encrypted_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->sc_encrypted_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, sc_index * 8 + 6); + if (unlikely(ret)) + return ret; + counters->sc_protected_octets[0] = + packed_record[0] | (packed_record[1] << 16); + counters->sc_protected_octets[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, sc_index * 8 + 7); + if (unlikely(ret)) + return ret; + counters->sc_encrypted_octets[0] = + packed_record[0] | (packed_record[1] << 16); + counters->sc_encrypted_octets[1] = + packed_record[2] | (packed_record[3] << 16); + + return 0; +} + +int aq_mss_get_egress_sc_counters(struct aq_hw_s *hw, + struct aq_mss_egress_sc_counters *counters, + u16 sc_index) +{ + memset(counters, 0, sizeof(*counters)); + + return AQ_API_CALL_SAFE(get_egress_sc_counters, hw, counters, sc_index); +} + +static int get_egress_sa_counters(struct aq_hw_s *hw, + struct aq_mss_egress_sa_counters *counters, + u16 sa_index) +{ + u16 packed_record[4]; + int ret; + + if (sa_index >= NUMROWS_EGRESSSARECORD) + return -EINVAL; + + ret = get_raw_egress_record(hw, packed_record, 4, 3, sa_index * 8 + 0); + if (unlikely(ret)) + return ret; + counters->sa_hit_drop_redirect[0] = + packed_record[0] | (packed_record[1] << 16); + counters->sa_hit_drop_redirect[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, sa_index * 8 + 1); + if (unlikely(ret)) + return ret; + counters->sa_protected2_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->sa_protected2_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, sa_index * 8 + 2); + if (unlikely(ret)) + return ret; + counters->sa_protected_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->sa_protected_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, sa_index * 8 + 3); + if (unlikely(ret)) + return ret; + counters->sa_encrypted_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->sa_encrypted_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + return 0; +} + +int aq_mss_get_egress_sa_counters(struct aq_hw_s *hw, + struct aq_mss_egress_sa_counters *counters, + u16 sa_index) +{ + memset(counters, 0, sizeof(*counters)); + + return AQ_API_CALL_SAFE(get_egress_sa_counters, hw, counters, sa_index); +} + +static int +get_egress_common_counters(struct aq_hw_s *hw, + struct aq_mss_egress_common_counters *counters) +{ + u16 packed_record[4]; + int ret; + + ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 0); + if (unlikely(ret)) + return ret; + counters->ctl_pkt[0] = packed_record[0] | (packed_record[1] << 16); + counters->ctl_pkt[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 1); + if (unlikely(ret)) + return ret; + counters->unknown_sa_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->unknown_sa_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 2); + if (unlikely(ret)) + return ret; + counters->untagged_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->untagged_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 3); + if (unlikely(ret)) + return ret; + counters->too_long[0] = packed_record[0] | (packed_record[1] << 16); + counters->too_long[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 4); + if (unlikely(ret)) + return ret; + counters->ecc_error_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->ecc_error_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_egress_record(hw, packed_record, 4, 3, 256 + 5); + if (unlikely(ret)) + return ret; + counters->unctrl_hit_drop_redir[0] = + packed_record[0] | (packed_record[1] << 16); + counters->unctrl_hit_drop_redir[1] = + packed_record[2] | (packed_record[3] << 16); + + return 0; +} + +int aq_mss_get_egress_common_counters(struct aq_hw_s *hw, + struct aq_mss_egress_common_counters *counters) +{ + memset(counters, 0, sizeof(*counters)); + + return AQ_API_CALL_SAFE(get_egress_common_counters, hw, counters); +} + +static int clear_egress_counters(struct aq_hw_s *hw) +{ + struct mss_egress_ctl_register ctl_reg; + int ret; + + memset(&ctl_reg, 0, sizeof(ctl_reg)); + + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, MSS_EGRESS_CTL_REGISTER_ADDR, + &ctl_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_EGRESS_CTL_REGISTER_ADDR + 4, + &ctl_reg.word_1); + if (unlikely(ret)) + return ret; + + /* Toggle the Egress MIB clear bit 0->1->0 */ + ctl_reg.bits_0.clear_counter = 0; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_CTL_REGISTER_ADDR + 4, + ctl_reg.word_1); + if (unlikely(ret)) + return ret; + + ctl_reg.bits_0.clear_counter = 1; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_CTL_REGISTER_ADDR + 4, + ctl_reg.word_1); + if (unlikely(ret)) + return ret; + + ctl_reg.bits_0.clear_counter = 0; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_CTL_REGISTER_ADDR + 4, + ctl_reg.word_1); + if (unlikely(ret)) + return ret; + + return 0; +} + +int aq_mss_clear_egress_counters(struct aq_hw_s *hw) +{ + return AQ_API_CALL_SAFE(clear_egress_counters, hw); +} + +static int get_ingress_sa_counters(struct aq_hw_s *hw, + struct aq_mss_ingress_sa_counters *counters, + u16 sa_index) +{ + u16 packed_record[4]; + int ret; + + if (sa_index >= NUMROWS_INGRESSSARECORD) + return -EINVAL; + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 0); + if (unlikely(ret)) + return ret; + counters->untagged_hit_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->untagged_hit_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 1); + if (unlikely(ret)) + return ret; + counters->ctrl_hit_drop_redir_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->ctrl_hit_drop_redir_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 2); + if (unlikely(ret)) + return ret; + counters->not_using_sa[0] = packed_record[0] | (packed_record[1] << 16); + counters->not_using_sa[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 3); + if (unlikely(ret)) + return ret; + counters->unused_sa[0] = packed_record[0] | (packed_record[1] << 16); + counters->unused_sa[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 4); + if (unlikely(ret)) + return ret; + counters->not_valid_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->not_valid_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 5); + if (unlikely(ret)) + return ret; + counters->invalid_pkts[0] = packed_record[0] | (packed_record[1] << 16); + counters->invalid_pkts[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 6); + if (unlikely(ret)) + return ret; + counters->ok_pkts[0] = packed_record[0] | (packed_record[1] << 16); + counters->ok_pkts[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 7); + if (unlikely(ret)) + return ret; + counters->late_pkts[0] = packed_record[0] | (packed_record[1] << 16); + counters->late_pkts[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 8); + if (unlikely(ret)) + return ret; + counters->delayed_pkts[0] = packed_record[0] | (packed_record[1] << 16); + counters->delayed_pkts[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 9); + if (unlikely(ret)) + return ret; + counters->unchecked_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->unchecked_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 10); + if (unlikely(ret)) + return ret; + counters->validated_octets[0] = + packed_record[0] | (packed_record[1] << 16); + counters->validated_octets[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, + sa_index * 12 + 11); + if (unlikely(ret)) + return ret; + counters->decrypted_octets[0] = + packed_record[0] | (packed_record[1] << 16); + counters->decrypted_octets[1] = + packed_record[2] | (packed_record[3] << 16); + + return 0; +} + +int aq_mss_get_ingress_sa_counters(struct aq_hw_s *hw, + struct aq_mss_ingress_sa_counters *counters, + u16 sa_index) +{ + memset(counters, 0, sizeof(*counters)); + + return AQ_API_CALL_SAFE(get_ingress_sa_counters, hw, counters, + sa_index); +} + +static int +get_ingress_common_counters(struct aq_hw_s *hw, + struct aq_mss_ingress_common_counters *counters) +{ + u16 packed_record[4]; + int ret; + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 0); + if (unlikely(ret)) + return ret; + counters->ctl_pkts[0] = packed_record[0] | (packed_record[1] << 16); + counters->ctl_pkts[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 1); + if (unlikely(ret)) + return ret; + counters->tagged_miss_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->tagged_miss_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 2); + if (unlikely(ret)) + return ret; + counters->untagged_miss_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->untagged_miss_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 3); + if (unlikely(ret)) + return ret; + counters->notag_pkts[0] = packed_record[0] | (packed_record[1] << 16); + counters->notag_pkts[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 4); + if (unlikely(ret)) + return ret; + counters->untagged_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->untagged_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 5); + if (unlikely(ret)) + return ret; + counters->bad_tag_pkts[0] = packed_record[0] | (packed_record[1] << 16); + counters->bad_tag_pkts[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 6); + if (unlikely(ret)) + return ret; + counters->no_sci_pkts[0] = packed_record[0] | (packed_record[1] << 16); + counters->no_sci_pkts[1] = packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 7); + if (unlikely(ret)) + return ret; + counters->unknown_sci_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->unknown_sci_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 8); + if (unlikely(ret)) + return ret; + counters->ctrl_prt_pass_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->ctrl_prt_pass_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 9); + if (unlikely(ret)) + return ret; + counters->unctrl_prt_pass_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->unctrl_prt_pass_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 10); + if (unlikely(ret)) + return ret; + counters->ctrl_prt_fail_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->ctrl_prt_fail_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 11); + if (unlikely(ret)) + return ret; + counters->unctrl_prt_fail_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->unctrl_prt_fail_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 12); + if (unlikely(ret)) + return ret; + counters->too_long_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->too_long_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 13); + if (unlikely(ret)) + return ret; + counters->igpoc_ctl_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->igpoc_ctl_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 14); + if (unlikely(ret)) + return ret; + counters->ecc_error_pkts[0] = + packed_record[0] | (packed_record[1] << 16); + counters->ecc_error_pkts[1] = + packed_record[2] | (packed_record[3] << 16); + + ret = get_raw_ingress_record(hw, packed_record, 4, 6, 385 + 15); + if (unlikely(ret)) + return ret; + counters->unctrl_hit_drop_redir[0] = + packed_record[0] | (packed_record[1] << 16); + counters->unctrl_hit_drop_redir[1] = + packed_record[2] | (packed_record[3] << 16); + + return 0; +} + +int aq_mss_get_ingress_common_counters(struct aq_hw_s *hw, + struct aq_mss_ingress_common_counters *counters) +{ + memset(counters, 0, sizeof(*counters)); + + return AQ_API_CALL_SAFE(get_ingress_common_counters, hw, counters); +} + +static int clear_ingress_counters(struct aq_hw_s *hw) +{ + struct mss_ingress_ctl_register ctl_reg; + int ret; + + memset(&ctl_reg, 0, sizeof(ctl_reg)); + + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_INGRESS_CTL_REGISTER_ADDR, &ctl_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_INGRESS_CTL_REGISTER_ADDR + 4, + &ctl_reg.word_1); + if (unlikely(ret)) + return ret; + + /* Toggle the Ingress MIB clear bit 0->1->0 */ + ctl_reg.bits_0.clear_count = 0; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_CTL_REGISTER_ADDR + 4, + ctl_reg.word_1); + if (unlikely(ret)) + return ret; + + ctl_reg.bits_0.clear_count = 1; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_CTL_REGISTER_ADDR + 4, + ctl_reg.word_1); + if (unlikely(ret)) + return ret; + + ctl_reg.bits_0.clear_count = 0; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_CTL_REGISTER_ADDR, ctl_reg.word_0); + if (unlikely(ret)) + return ret; + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_INGRESS_CTL_REGISTER_ADDR + 4, + ctl_reg.word_1); + if (unlikely(ret)) + return ret; + + return 0; +} + +int aq_mss_clear_ingress_counters(struct aq_hw_s *hw) +{ + return AQ_API_CALL_SAFE(clear_ingress_counters, hw); +} + +static int get_egress_sa_expired(struct aq_hw_s *hw, u32 *expired) +{ + u16 val; + int ret; + + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR, + &val); + if (unlikely(ret)) + return ret; + + *expired = val; + + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR + 1, + &val); + if (unlikely(ret)) + return ret; + + *expired |= val << 16; + + return 0; +} + +int aq_mss_get_egress_sa_expired(struct aq_hw_s *hw, u32 *expired) +{ + *expired = 0; + + return AQ_API_CALL_SAFE(get_egress_sa_expired, hw, expired); +} + +static int get_egress_sa_threshold_expired(struct aq_hw_s *hw, + u32 *expired) +{ + u16 val; + int ret; + + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR, &val); + if (unlikely(ret)) + return ret; + + *expired = val; + + ret = aq_mss_mdio_read(hw, MDIO_MMD_VEND1, + MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR + 1, &val); + if (unlikely(ret)) + return ret; + + *expired |= val << 16; + + return 0; +} + +int aq_mss_get_egress_sa_threshold_expired(struct aq_hw_s *hw, + u32 *expired) +{ + *expired = 0; + + return AQ_API_CALL_SAFE(get_egress_sa_threshold_expired, hw, expired); +} + +static int set_egress_sa_expired(struct aq_hw_s *hw, u32 expired) +{ + int ret; + + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR, + expired & 0xFFFF); + if (unlikely(ret)) + return ret; + + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_SA_EXPIRED_STATUS_REGISTER_ADDR + 1, + expired >> 16); + if (unlikely(ret)) + return ret; + + return 0; +} + +int aq_mss_set_egress_sa_expired(struct aq_hw_s *hw, u32 expired) +{ + return AQ_API_CALL_SAFE(set_egress_sa_expired, hw, expired); +} + +static int set_egress_sa_threshold_expired(struct aq_hw_s *hw, u32 expired) +{ + int ret; + + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR, + expired & 0xFFFF); + if (unlikely(ret)) + return ret; + + ret = aq_mss_mdio_write(hw, MDIO_MMD_VEND1, + MSS_EGRESS_SA_THRESHOLD_EXPIRED_STATUS_REGISTER_ADDR + 1, + expired >> 16); + if (unlikely(ret)) + return ret; + + return 0; +} + +int aq_mss_set_egress_sa_threshold_expired(struct aq_hw_s *hw, u32 expired) +{ + return AQ_API_CALL_SAFE(set_egress_sa_threshold_expired, hw, expired); +} diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.h b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.h new file mode 100644 index 000000000000..ff03cc462a37 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.h @@ -0,0 +1,323 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef __MACSEC_API_H__ +#define __MACSEC_API_H__ + +#include "aq_hw.h" +#include "macsec_struct.h" + +#define NUMROWS_INGRESSPRECTLFRECORD 24 +#define ROWOFFSET_INGRESSPRECTLFRECORD 0 + +#define NUMROWS_INGRESSPRECLASSRECORD 48 +#define ROWOFFSET_INGRESSPRECLASSRECORD 0 + +#define NUMROWS_INGRESSPOSTCLASSRECORD 48 +#define ROWOFFSET_INGRESSPOSTCLASSRECORD 0 + +#define NUMROWS_INGRESSSCRECORD 32 +#define ROWOFFSET_INGRESSSCRECORD 0 + +#define NUMROWS_INGRESSSARECORD 32 +#define ROWOFFSET_INGRESSSARECORD 32 + +#define NUMROWS_INGRESSSAKEYRECORD 32 +#define ROWOFFSET_INGRESSSAKEYRECORD 0 + +#define NUMROWS_INGRESSPOSTCTLFRECORD 24 +#define ROWOFFSET_INGRESSPOSTCTLFRECORD 0 + +#define NUMROWS_EGRESSCTLFRECORD 24 +#define ROWOFFSET_EGRESSCTLFRECORD 0 + +#define NUMROWS_EGRESSCLASSRECORD 48 +#define ROWOFFSET_EGRESSCLASSRECORD 0 + +#define NUMROWS_EGRESSSCRECORD 32 +#define ROWOFFSET_EGRESSSCRECORD 0 + +#define NUMROWS_EGRESSSARECORD 32 +#define ROWOFFSET_EGRESSSARECORD 32 + +#define NUMROWS_EGRESSSAKEYRECORD 32 +#define ROWOFFSET_EGRESSSAKEYRECORD 96 + +/*! Read the raw table data from the specified row of the Egress CTL + * Filter table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 23). + */ +int aq_mss_get_egress_ctlf_record(struct aq_hw_s *hw, + struct aq_mss_egress_ctlf_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Egress CTL Filter table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write(max 23). + */ +int aq_mss_set_egress_ctlf_record(struct aq_hw_s *hw, + const struct aq_mss_egress_ctlf_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Egress + * Packet Classifier table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 47). + */ +int aq_mss_get_egress_class_record(struct aq_hw_s *hw, + struct aq_mss_egress_class_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Egress Packet Classifier table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write (max 47). + */ +int aq_mss_set_egress_class_record(struct aq_hw_s *hw, + const struct aq_mss_egress_class_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Egress SC + * Lookup table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 31). + */ +int aq_mss_get_egress_sc_record(struct aq_hw_s *hw, + struct aq_mss_egress_sc_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Egress SC Lookup table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write (max 31). + */ +int aq_mss_set_egress_sc_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sc_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Egress SA + * Lookup table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 31). + */ +int aq_mss_get_egress_sa_record(struct aq_hw_s *hw, + struct aq_mss_egress_sa_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Egress SA Lookup table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write (max 31). + */ +int aq_mss_set_egress_sa_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sa_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Egress SA + * Key Lookup table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 31). + */ +int aq_mss_get_egress_sakey_record(struct aq_hw_s *hw, + struct aq_mss_egress_sakey_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Egress SA Key Lookup table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write (max 31). + */ +int aq_mss_set_egress_sakey_record(struct aq_hw_s *hw, + const struct aq_mss_egress_sakey_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Ingress + * Pre-MACSec CTL Filter table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 23). + */ +int aq_mss_get_ingress_prectlf_record(struct aq_hw_s *hw, + struct aq_mss_ingress_prectlf_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Ingress Pre-MACSec CTL Filter table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write(max 23). + */ +int aq_mss_set_ingress_prectlf_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_prectlf_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Ingress + * Pre-MACSec Packet Classifier table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 47). + */ +int aq_mss_get_ingress_preclass_record(struct aq_hw_s *hw, + struct aq_mss_ingress_preclass_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Ingress Pre-MACSec Packet Classifier table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write(max 47). + */ +int aq_mss_set_ingress_preclass_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_preclass_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Ingress SC + * Lookup table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 31). + */ +int aq_mss_get_ingress_sc_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sc_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Ingress SC Lookup table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write(max 31). + */ +int aq_mss_set_ingress_sc_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sc_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Ingress SA + * Lookup table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 31). + */ +int aq_mss_get_ingress_sa_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sa_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Ingress SA Lookup table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write(max 31). + */ +int aq_mss_set_ingress_sa_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sa_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Ingress SA + * Key Lookup table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 31). + */ +int aq_mss_get_ingress_sakey_record(struct aq_hw_s *hw, + struct aq_mss_ingress_sakey_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Ingress SA Key Lookup table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write(max 31). + */ +int aq_mss_set_ingress_sakey_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_sakey_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Ingress + * Post-MACSec Packet Classifier table, and unpack it into the + * fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 48). + */ +int aq_mss_get_ingress_postclass_record(struct aq_hw_s *hw, + struct aq_mss_ingress_postclass_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Ingress Post-MACSec Packet Classifier table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write(max 48). + */ +int aq_mss_set_ingress_postclass_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_postclass_record *rec, + u16 table_index); + +/*! Read the raw table data from the specified row of the Ingress + * Post-MACSec CTL Filter table, and unpack it into the fields of rec. + * rec - [OUT] The raw table row data will be unpacked into the fields of rec. + * table_index - The table row to read (max 23). + */ +int aq_mss_get_ingress_postctlf_record(struct aq_hw_s *hw, + struct aq_mss_ingress_postctlf_record *rec, + u16 table_index); + +/*! Pack the fields of rec, and write the packed data into the + * specified row of the Ingress Post-MACSec CTL Filter table. + * rec - [IN] The bitfield values to write to the table row. + * table_index - The table row to write(max 23). + */ +int aq_mss_set_ingress_postctlf_record(struct aq_hw_s *hw, + const struct aq_mss_ingress_postctlf_record *rec, + u16 table_index); + +/*! Read the counters for the specified SC, and unpack them into the + * fields of counters. + * counters - [OUT] The raw table row data will be unpacked here. + * sc_index - The table row to read (max 31). + */ +int aq_mss_get_egress_sc_counters(struct aq_hw_s *hw, + struct aq_mss_egress_sc_counters *counters, + u16 sc_index); + +/*! Read the counters for the specified SA, and unpack them into the + * fields of counters. + * counters - [OUT] The raw table row data will be unpacked here. + * sa_index - The table row to read (max 31). + */ +int aq_mss_get_egress_sa_counters(struct aq_hw_s *hw, + struct aq_mss_egress_sa_counters *counters, + u16 sa_index); + +/*! Read the counters for the common egress counters, and unpack them + * into the fields of counters. + * counters - [OUT] The raw table row data will be unpacked here. + */ +int aq_mss_get_egress_common_counters(struct aq_hw_s *hw, + struct aq_mss_egress_common_counters *counters); + +/*! Clear all Egress counters to 0.*/ +int aq_mss_clear_egress_counters(struct aq_hw_s *hw); + +/*! Read the counters for the specified SA, and unpack them into the + * fields of counters. + * counters - [OUT] The raw table row data will be unpacked here. + * sa_index - The table row to read (max 31). + */ +int aq_mss_get_ingress_sa_counters(struct aq_hw_s *hw, + struct aq_mss_ingress_sa_counters *counters, + u16 sa_index); + +/*! Read the counters for the common ingress counters, and unpack them + * into the fields of counters. + * counters - [OUT] The raw table row data will be unpacked here. + */ +int aq_mss_get_ingress_common_counters(struct aq_hw_s *hw, + struct aq_mss_ingress_common_counters *counters); + +/*! Clear all Ingress counters to 0. */ +int aq_mss_clear_ingress_counters(struct aq_hw_s *hw); + +/*! Get Egress SA expired. */ +int aq_mss_get_egress_sa_expired(struct aq_hw_s *hw, u32 *expired); +/*! Get Egress SA threshold expired. */ +int aq_mss_get_egress_sa_threshold_expired(struct aq_hw_s *hw, + u32 *expired); +/*! Set Egress SA expired. */ +int aq_mss_set_egress_sa_expired(struct aq_hw_s *hw, u32 expired); +/*! Set Egress SA threshold expired. */ +int aq_mss_set_egress_sa_threshold_expired(struct aq_hw_s *hw, + u32 expired); + +#endif /* __MACSEC_API_H__ */ diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_struct.h b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_struct.h new file mode 100644 index 000000000000..b6119dcc3bb9 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_struct.h @@ -0,0 +1,914 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef _MACSEC_STRUCT_H_ +#define _MACSEC_STRUCT_H_ + +/*! Represents the bitfields of a single row in the Egress CTL Filter + * table. + */ +struct aq_mss_egress_ctlf_record { + /*! This is used to store the 48 bit value used to compare SA, DA or + * halfDA+half SA value. + */ + u32 sa_da[2]; + /*! This is used to store the 16 bit ethertype value used for + * comparison. + */ + u32 eth_type; + /*! The match mask is per-nibble. 0 means don't care, i.e. every value + * will match successfully. The total data is 64 bit, i.e. 16 nibbles + * masks. + */ + u32 match_mask; + /*! 0: No compare, i.e. This entry is not used + * 1: compare DA only + * 2: compare SA only + * 3: compare half DA + half SA + * 4: compare ether type only + * 5: compare DA + ethertype + * 6: compare SA + ethertype + * 7: compare DA+ range. + */ + u32 match_type; + /*! 0: Bypass the remaining modules if matched. + * 1: Forward to next module for more classifications. + */ + u32 action; +}; + +/*! Represents the bitfields of a single row in the Egress Packet + * Classifier table. + */ +struct aq_mss_egress_class_record { + /*! VLAN ID field. */ + u32 vlan_id; + /*! VLAN UP field. */ + u32 vlan_up; + /*! VLAN Present in the Packet. */ + u32 vlan_valid; + /*! The 8 bit value used to compare with extracted value for byte 3. */ + u32 byte3; + /*! The 8 bit value used to compare with extracted value for byte 2. */ + u32 byte2; + /*! The 8 bit value used to compare with extracted value for byte 1. */ + u32 byte1; + /*! The 8 bit value used to compare with extracted value for byte 0. */ + u32 byte0; + /*! The 8 bit TCI field used to compare with extracted value. */ + u32 tci; + /*! The 64 bit SCI field in the SecTAG. */ + u32 sci[2]; + /*! The 16 bit Ethertype (in the clear) field used to compare with + * extracted value. + */ + u32 eth_type; + /*! This is to specify the 40bit SNAP header if the SNAP header's mask + * is enabled. + */ + u32 snap[2]; + /*! This is to specify the 24bit LLC header if the LLC header's mask is + * enabled. + */ + u32 llc; + /*! The 48 bit MAC_SA field used to compare with extracted value. */ + u32 mac_sa[2]; + /*! The 48 bit MAC_DA field used to compare with extracted value. */ + u32 mac_da[2]; + /*! The 32 bit Packet number used to compare with extracted value. */ + u32 pn; + /*! 0~63: byte location used extracted by packets comparator, which + * can be anything from the first 64 bytes of the MAC packets. + * This byte location counted from MAC' DA address. i.e. set to 0 + * will point to byte 0 of DA address. + */ + u32 byte3_location; + /*! 0: don't care + * 1: enable comparison of extracted byte pointed by byte 3 location. + */ + u32 byte3_mask; + /*! 0~63: byte location used extracted by packets comparator, which + * can be anything from the first 64 bytes of the MAC packets. + * This byte location counted from MAC' DA address. i.e. set to 0 + * will point to byte 0 of DA address. + */ + u32 byte2_location; + /*! 0: don't care + * 1: enable comparison of extracted byte pointed by byte 2 location. + */ + u32 byte2_mask; + /*! 0~63: byte location used extracted by packets comparator, which + * can be anything from the first 64 bytes of the MAC packets. + * This byte location counted from MAC' DA address. i.e. set to 0 + * will point to byte 0 of DA address. + */ + u32 byte1_location; + /*! 0: don't care + * 1: enable comparison of extracted byte pointed by byte 1 location. + */ + u32 byte1_mask; + /*! 0~63: byte location used extracted by packets comparator, which + * can be anything from the first 64 bytes of the MAC packets. + * This byte location counted from MAC' DA address. i.e. set to 0 + * will point to byte 0 of DA address. + */ + u32 byte0_location; + /*! 0: don't care + * 1: enable comparison of extracted byte pointed by byte 0 location. + */ + u32 byte0_mask; + /*! Mask is per-byte. + * 0: don't care + * 1: enable comparison of extracted VLAN ID field. + */ + u32 vlan_id_mask; + /*! 0: don't care + * 1: enable comparison of extracted VLAN UP field. + */ + u32 vlan_up_mask; + /*! 0: don't care + * 1: enable comparison of extracted VLAN Valid field. + */ + u32 vlan_valid_mask; + /*! This is bit mask to enable comparison the 8 bit TCI field, + * including the AN field. + * For explicit SECTAG, AN is hardware controlled. For sending + * packet w/ explicit SECTAG, rest of the TCI fields are directly + * from the SECTAG. + */ + u32 tci_mask; + /*! Mask is per-byte. + * 0: don't care + * 1: enable comparison of SCI + * Note: If this field is not 0, this means the input packet's + * SECTAG is explicitly tagged and MACSEC module will only update + * the MSDU. + * PN number is hardware controlled. + */ + u32 sci_mask; + /*! Mask is per-byte. + * 0: don't care + * 1: enable comparison of Ethertype. + */ + u32 eth_type_mask; + /*! Mask is per-byte. + * 0: don't care and no SNAP header exist. + * 1: compare the SNAP header. + * If this bit is set to 1, the extracted filed will assume the + * SNAP header exist as encapsulated in 802.3 (RFC 1042). I.E. the + * next 5 bytes after the the LLC header is SNAP header. + */ + u32 snap_mask; + /*! 0: don't care and no LLC header exist. + * 1: compare the LLC header. + * If this bit is set to 1, the extracted filed will assume the + * LLC header exist as encapsulated in 802.3 (RFC 1042). I.E. the + * next three bytes after the 802.3MAC header is LLC header. + */ + u32 llc_mask; + /*! Mask is per-byte. + * 0: don't care + * 1: enable comparison of MAC_SA. + */ + u32 sa_mask; + /*! Mask is per-byte. + * 0: don't care + * 1: enable comparison of MAC_DA. + */ + u32 da_mask; + /*! Mask is per-byte. */ + u32 pn_mask; + /*! Reserved. This bit should be always 0. */ + u32 eight02dot2; + /*! 1: For explicit sectag case use TCI_SC from table + * 0: use TCI_SC from explicit sectag. + */ + u32 tci_sc; + /*! 1: For explicit sectag case,use TCI_V,ES,SCB,E,C from table + * 0: use TCI_V,ES,SCB,E,C from explicit sectag. + */ + u32 tci_87543; + /*! 1: indicates that incoming packet has explicit sectag. */ + u32 exp_sectag_en; + /*! If packet matches and tagged as controlled-packet, this SC/SA + * index is used for later SC and SA table lookup. + */ + u32 sc_idx; + /*! This field is used to specify how many SA entries are + * associated with 1 SC entry. + * 2'b00: 1 SC has 4 SA. + * SC index is equivalent to {SC_Index[4:2], 1'b0}. + * SA index is equivalent to {SC_Index[4:2], SC entry's current AN[1:0] + * 2'b10: 1 SC has 2 SA. + * SC index is equivalent to SC_Index[4:1] + * SA index is equivalent to {SC_Index[4:1], SC entry's current AN[0]} + * 2'b11: 1 SC has 1 SA. No SC entry exists for the specific SA. + * SA index is equivalent to SC_Index[4:0] + * Note: if specified as 2'b11, hardware AN roll over is not + * supported. + */ + u32 sc_sa; + /*! 0: the packets will be sent to MAC FIFO + * 1: The packets will be sent to Debug/Loopback FIFO. + * If the above's action is drop, this bit has no meaning. + */ + u32 debug; + /*! 0: forward to remaining modules + * 1: bypass the next encryption modules. This packet is considered + * un-control packet. + * 2: drop + * 3: Reserved. + */ + u32 action; + /*! 0: Not valid entry. This entry is not used + * 1: valid entry. + */ + u32 valid; +}; + +/*! Represents the bitfields of a single row in the Egress SC Lookup table. */ +struct aq_mss_egress_sc_record { + /*! This is to specify when the SC was first used. Set by HW. */ + u32 start_time; + /*! This is to specify when the SC was last used. Set by HW. */ + u32 stop_time; + /*! This is to specify which of the SA entries are used by current HW. + * Note: This value need to be set by SW after reset. It will be + * automatically updated by HW, if AN roll over is enabled. + */ + u32 curr_an; + /*! 0: Clear the SA Valid Bit after PN expiry. + * 1: Do not Clear the SA Valid bit after PN expiry of the current SA. + * When the Enable AN roll over is set, S/W does not need to + * program the new SA's and the H/W will automatically roll over + * between the SA's without session expiry. + * For normal operation, Enable AN Roll over will be set to '0' + * and in which case, the SW needs to program the new SA values + * after the current PN expires. + */ + u32 an_roll; + /*! This is the TCI field used if packet is not explicitly tagged. */ + u32 tci; + /*! This value indicates the offset where the decryption will start. + * [[Values of 0, 4, 8-50]. + */ + u32 enc_off; + /*! 0: Do not protect frames, all the packets will be forwarded + * unchanged. MIB counter (OutPktsUntagged) will be updated. + * 1: Protect. + */ + u32 protect; + /*! 0: when none of the SA related to SC has inUse set. + * 1: when either of the SA related to the SC has inUse set. + * This bit is set by HW. + */ + u32 recv; + /*! 0: H/W Clears this bit on the first use. + * 1: SW updates this entry, when programming the SC Table. + */ + u32 fresh; + /*! AES Key size + * 00 - 128bits + * 01 - 192bits + * 10 - 256bits + * 11 - Reserved. + */ + u32 sak_len; + /*! 0: Invalid SC + * 1: Valid SC. + */ + u32 valid; +}; + +/*! Represents the bitfields of a single row in the Egress SA Lookup table. */ +struct aq_mss_egress_sa_record { + /*! This is to specify when the SC was first used. Set by HW. */ + u32 start_time; + /*! This is to specify when the SC was last used. Set by HW. */ + u32 stop_time; + /*! This is set by SW and updated by HW to store the Next PN number + * used for encryption. + */ + u32 next_pn; + /*! The Next_PN number is going to wrapped around from 0xFFFF_FFFF + * to 0. set by HW. + */ + u32 sat_pn; + /*! 0: This SA is in use. + * 1: This SA is Fresh and set by SW. + */ + u32 fresh; + /*! 0: Invalid SA + * 1: Valid SA. + */ + u32 valid; +}; + +/*! Represents the bitfields of a single row in the Egress SA Key + * Lookup table. + */ +struct aq_mss_egress_sakey_record { + /*! Key for AES-GCM processing. */ + u32 key[8]; +}; + +/*! Represents the bitfields of a single row in the Ingress Pre-MACSec + * CTL Filter table. + */ +struct aq_mss_ingress_prectlf_record { + /*! This is used to store the 48 bit value used to compare SA, DA + * or halfDA+half SA value. + */ + u32 sa_da[2]; + /*! This is used to store the 16 bit ethertype value used for + * comparison. + */ + u32 eth_type; + /*! The match mask is per-nibble. 0 means don't care, i.e. every + * value will match successfully. The total data is 64 bit, i.e. + * 16 nibbles masks. + */ + u32 match_mask; + /*! 0: No compare, i.e. This entry is not used + * 1: compare DA only + * 2: compare SA only + * 3: compare half DA + half SA + * 4: compare ether type only + * 5: compare DA + ethertype + * 6: compare SA + ethertype + * 7: compare DA+ range. + */ + u32 match_type; + /*! 0: Bypass the remaining modules if matched. + * 1: Forward to next module for more classifications. + */ + u32 action; +}; + +/*! Represents the bitfields of a single row in the Ingress Pre-MACSec + * Packet Classifier table. + */ +struct aq_mss_ingress_preclass_record { + /*! The 64 bit SCI field used to compare with extracted value. + * Should have SCI value in case TCI[SCI_SEND] == 0. This will be + * used for ICV calculation. + */ + u32 sci[2]; + /*! The 8 bit TCI field used to compare with extracted value. */ + u32 tci; + /*! 8 bit encryption offset. */ + u32 encr_offset; + /*! The 16 bit Ethertype (in the clear) field used to compare with + * extracted value. + */ + u32 eth_type; + /*! This is to specify the 40bit SNAP header if the SNAP header's + * mask is enabled. + */ + u32 snap[2]; + /*! This is to specify the 24bit LLC header if the LLC header's + * mask is enabled. + */ + u32 llc; + /*! The 48 bit MAC_SA field used to compare with extracted value. */ + u32 mac_sa[2]; + /*! The 48 bit MAC_DA field used to compare with extracted value. */ + u32 mac_da[2]; + /*! 0: this is to compare with non-LPBK packet + * 1: this is to compare with LPBK packet. + * This value is used to compare with a controlled-tag which goes + * with the packet when looped back from Egress port. + */ + u32 lpbk_packet; + /*! The value of this bit mask will affects how the SC index and SA + * index created. + * 2'b00: 1 SC has 4 SA. + * SC index is equivalent to {SC_Index[4:2], 1'b0}. + * SA index is equivalent to {SC_Index[4:2], SECTAG's AN[1:0]} + * Here AN bits are not compared. + * 2'b10: 1 SC has 2 SA. + * SC index is equivalent to SC_Index[4:1] + * SA index is equivalent to {SC_Index[4:1], SECTAG's AN[0]} + * Compare AN[1] field only + * 2'b11: 1 SC has 1 SA. No SC entry exists for the specific SA. + * SA index is equivalent to SC_Index[4:0] + * AN[1:0] bits are compared. + * NOTE: This design is to supports different usage of AN. User + * can either ping-pong buffer 2 SA by using only the AN[0] bit. + * Or use 4 SA per SC by use AN[1:0] bits. Or even treat each SA + * as independent. i.e. AN[1:0] is just another matching pointer + * to select SA. + */ + u32 an_mask; + /*! This is bit mask to enable comparison the upper 6 bits TCI + * field, which does not include the AN field. + * 0: don't compare + * 1: enable comparison of the bits. + */ + u32 tci_mask; + /*! 0: don't care + * 1: enable comparison of SCI. + */ + u32 sci_mask; + /*! Mask is per-byte. + * 0: don't care + * 1: enable comparison of Ethertype. + */ + u32 eth_type_mask; + /*! Mask is per-byte. + * 0: don't care and no SNAP header exist. + * 1: compare the SNAP header. + * If this bit is set to 1, the extracted filed will assume the + * SNAP header exist as encapsulated in 802.3 (RFC 1042). I.E. the + * next 5 bytes after the the LLC header is SNAP header. + */ + u32 snap_mask; + /*! Mask is per-byte. + * 0: don't care and no LLC header exist. + * 1: compare the LLC header. + * If this bit is set to 1, the extracted filed will assume the + * LLC header exist as encapsulated in 802.3 (RFC 1042). I.E. the + * next three bytes after the 802.3MAC header is LLC header. + */ + u32 llc_mask; + /*! Reserved. This bit should be always 0. */ + u32 _802_2_encapsulate; + /*! Mask is per-byte. + * 0: don't care + * 1: enable comparison of MAC_SA. + */ + u32 sa_mask; + /*! Mask is per-byte. + * 0: don't care + * 1: enable comparison of MAC_DA. + */ + u32 da_mask; + /*! 0: don't care + * 1: enable checking if this is loopback packet or not. + */ + u32 lpbk_mask; + /*! If packet matches and tagged as controlled-packet. This SC/SA + * index is used for later SC and SA table lookup. + */ + u32 sc_idx; + /*! 0: the packets will be sent to MAC FIFO + * 1: The packets will be sent to Debug/Loopback FIFO. + * If the above's action is drop. This bit has no meaning. + */ + u32 proc_dest; + /*! 0: Process: Forward to next two modules for 802.1AE decryption. + * 1: Process but keep SECTAG: Forward to next two modules for + * 802.1AE decryption but keep the MACSEC header with added error + * code information. ICV will be stripped for all control packets. + * 2: Bypass: Bypass the next two decryption modules but processed + * by post-classification. + * 3: Drop: drop this packet and update counts accordingly. + */ + u32 action; + /*! 0: This is a controlled-port packet if matched. + * 1: This is an uncontrolled-port packet if matched. + */ + u32 ctrl_unctrl; + /*! Use the SCI value from the Table if 'SC' bit of the input + * packet is not present. + */ + u32 sci_from_table; + /*! Reserved. */ + u32 reserved; + /*! 0: Not valid entry. This entry is not used + * 1: valid entry. + */ + u32 valid; +}; + +/*! Represents the bitfields of a single row in the Ingress SC Lookup table. */ +struct aq_mss_ingress_sc_record { + /*! This is to specify when the SC was first used. Set by HW. */ + u32 stop_time; + /*! This is to specify when the SC was first used. Set by HW. */ + u32 start_time; + /*! 0: Strict + * 1: Check + * 2: Disabled. + */ + u32 validate_frames; + /*! 1: Replay control enabled. + * 0: replay control disabled. + */ + u32 replay_protect; + /*! This is to specify the window range for anti-replay. Default is 0. + * 0: is strict order enforcement. + */ + u32 anti_replay_window; + /*! 0: when none of the SA related to SC has inUse set. + * 1: when either of the SA related to the SC has inUse set. + * This bit is set by HW. + */ + u32 receiving; + /*! 0: when hardware processed the SC for the first time, it clears + * this bit + * 1: This bit is set by SW, when it sets up the SC. + */ + u32 fresh; + /*! 0: The AN number will not automatically roll over if Next_PN is + * saturated. + * 1: The AN number will automatically roll over if Next_PN is + * saturated. + * Rollover is valid only after expiry. Normal roll over between + * SA's should be normal process. + */ + u32 an_rol; + /*! Reserved. */ + u32 reserved; + /*! 0: Invalid SC + * 1: Valid SC. + */ + u32 valid; +}; + +/*! Represents the bitfields of a single row in the Ingress SA Lookup table. */ +struct aq_mss_ingress_sa_record { + /*! This is to specify when the SC was first used. Set by HW. */ + u32 stop_time; + /*! This is to specify when the SC was first used. Set by HW. */ + u32 start_time; + /*! This is updated by HW to store the expected NextPN number for + * anti-replay. + */ + u32 next_pn; + /*! The Next_PN number is going to wrapped around from 0XFFFF_FFFF + * to 0. set by HW. + */ + u32 sat_nextpn; + /*! 0: This SA is not yet used. + * 1: This SA is inUse. + */ + u32 in_use; + /*! 0: when hardware processed the SC for the first time, it clears + * this timer + * 1: This bit is set by SW, when it sets up the SC. + */ + u32 fresh; + /*! Reserved. */ + u32 reserved; + /*! 0: Invalid SA. + * 1: Valid SA. + */ + u32 valid; +}; + +/*! Represents the bitfields of a single row in the Ingress SA Key + * Lookup table. + */ +struct aq_mss_ingress_sakey_record { + /*! Key for AES-GCM processing. */ + u32 key[8]; + /*! AES key size + * 00 - 128bits + * 01 - 192bits + * 10 - 256bits + * 11 - reserved. + */ + u32 key_len; +}; + +/*! Represents the bitfields of a single row in the Ingress Post- + * MACSec Packet Classifier table. + */ +struct aq_mss_ingress_postclass_record { + /*! The 8 bit value used to compare with extracted value for byte 0. */ + u32 byte0; + /*! The 8 bit value used to compare with extracted value for byte 1. */ + u32 byte1; + /*! The 8 bit value used to compare with extracted value for byte 2. */ + u32 byte2; + /*! The 8 bit value used to compare with extracted value for byte 3. */ + u32 byte3; + /*! Ethertype in the packet. */ + u32 eth_type; + /*! Ether Type value > 1500 (0x5dc). */ + u32 eth_type_valid; + /*! VLAN ID after parsing. */ + u32 vlan_id; + /*! VLAN priority after parsing. */ + u32 vlan_up; + /*! Valid VLAN coding. */ + u32 vlan_valid; + /*! SA index. */ + u32 sai; + /*! SAI hit, i.e. controlled packet. */ + u32 sai_hit; + /*! Mask for payload ethertype field. */ + u32 eth_type_mask; + /*! 0~63: byte location used extracted by packets comparator, which + * can be anything from the first 64 bytes of the MAC packets. + * This byte location counted from MAC' DA address. i.e. set to 0 + * will point to byte 0 of DA address. + */ + u32 byte3_location; + /*! Mask for Byte Offset 3. */ + u32 byte3_mask; + /*! 0~63: byte location used extracted by packets comparator, which + * can be anything from the first 64 bytes of the MAC packets. + * This byte location counted from MAC' DA address. i.e. set to 0 + * will point to byte 0 of DA address. + */ + u32 byte2_location; + /*! Mask for Byte Offset 2. */ + u32 byte2_mask; + /*! 0~63: byte location used extracted by packets comparator, which + * can be anything from the first 64 bytes of the MAC packets. + * This byte location counted from MAC' DA address. i.e. set to 0 + * will point to byte 0 of DA address. + */ + u32 byte1_location; + /*! Mask for Byte Offset 1. */ + u32 byte1_mask; + /*! 0~63: byte location used extracted by packets comparator, which + * can be anything from the first 64 bytes of the MAC packets. + * This byte location counted from MAC' DA address. i.e. set to 0 + * will point to byte 0 of DA address. + */ + u32 byte0_location; + /*! Mask for Byte Offset 0. */ + u32 byte0_mask; + /*! Mask for Ethertype valid field. Indicates 802.3 vs. Other. */ + u32 eth_type_valid_mask; + /*! Mask for VLAN ID field. */ + u32 vlan_id_mask; + /*! Mask for VLAN UP field. */ + u32 vlan_up_mask; + /*! Mask for VLAN valid field. */ + u32 vlan_valid_mask; + /*! Mask for SAI. */ + u32 sai_mask; + /*! Mask for SAI_HIT. */ + u32 sai_hit_mask; + /*! Action if only first level matches and second level does not. + * 0: pass + * 1: drop (fail). + */ + u32 firstlevel_actions; + /*! Action if both first and second level matched. + * 0: pass + * 1: drop (fail). + */ + u32 secondlevel_actions; + /*! Reserved. */ + u32 reserved; + /*! 0: Not valid entry. This entry is not used + * 1: valid entry. + */ + u32 valid; +}; + +/*! Represents the bitfields of a single row in the Ingress Post- + * MACSec CTL Filter table. + */ +struct aq_mss_ingress_postctlf_record { + /*! This is used to store the 48 bit value used to compare SA, DA + * or halfDA+half SA value. + */ + u32 sa_da[2]; + /*! This is used to store the 16 bit ethertype value used for + * comparison. + */ + u32 eth_type; + /*! The match mask is per-nibble. 0 means don't care, i.e. every + * value will match successfully. The total data is 64 bit, i.e. + * 16 nibbles masks. + */ + u32 match_mask; + /*! 0: No compare, i.e. This entry is not used + * 1: compare DA only + * 2: compare SA only + * 3: compare half DA + half SA + * 4: compare ether type only + * 5: compare DA + ethertype + * 6: compare SA + ethertype + * 7: compare DA+ range. + */ + u32 match_type; + /*! 0: Bypass the remaining modules if matched. + * 1: Forward to next module for more classifications. + */ + u32 action; +}; + +/*! Represents the Egress MIB counters for a single SC. Counters are + * 64 bits, lower 32 bits in field[0]. + */ +struct aq_mss_egress_sc_counters { + /*! The number of integrity protected but not encrypted packets + * for this transmitting SC. + */ + u32 sc_protected_pkts[2]; + /*! The number of integrity protected and encrypted packets for + * this transmitting SC. + */ + u32 sc_encrypted_pkts[2]; + /*! The number of plain text octets that are integrity protected + * but not encrypted on the transmitting SC. + */ + u32 sc_protected_octets[2]; + /*! The number of plain text octets that are integrity protected + * and encrypted on the transmitting SC. + */ + u32 sc_encrypted_octets[2]; +}; + +/*! Represents the Egress MIB counters for a single SA. Counters are + * 64 bits, lower 32 bits in field[0]. + */ +struct aq_mss_egress_sa_counters { + /*! The number of dropped packets for this transmitting SA. */ + u32 sa_hit_drop_redirect[2]; + /*! TODO */ + u32 sa_protected2_pkts[2]; + /*! The number of integrity protected but not encrypted packets + * for this transmitting SA. + */ + u32 sa_protected_pkts[2]; + /*! The number of integrity protected and encrypted packets for + * this transmitting SA. + */ + u32 sa_encrypted_pkts[2]; +}; + +/*! Represents the common Egress MIB counters; the counter not + * associated with a particular SC/SA. Counters are 64 bits, lower 32 + * bits in field[0]. + */ +struct aq_mss_egress_common_counters { + /*! The number of transmitted packets classified as MAC_CTL packets. */ + u32 ctl_pkt[2]; + /*! The number of transmitted packets that did not match any rows + * in the Egress Packet Classifier table. + */ + u32 unknown_sa_pkts[2]; + /*! The number of transmitted packets where the SC table entry has + * protect=0 (so packets are forwarded unchanged). + */ + u32 untagged_pkts[2]; + /*! The number of transmitted packets discarded because the packet + * length is greater than the ifMtu of the Common Port interface. + */ + u32 too_long[2]; + /*! The number of transmitted packets for which table memory was + * affected by an ECC error during processing. + */ + u32 ecc_error_pkts[2]; + /*! The number of transmitted packets for where the matched row in + * the Egress Packet Classifier table has action=drop. + */ + u32 unctrl_hit_drop_redir[2]; +}; + +/*! Represents the Ingress MIB counters for a single SA. Counters are + * 64 bits, lower 32 bits in field[0]. + */ +struct aq_mss_ingress_sa_counters { + /*! For this SA, the number of received packets without a SecTAG. */ + u32 untagged_hit_pkts[2]; + /*! For this SA, the number of received packets that were dropped. */ + u32 ctrl_hit_drop_redir_pkts[2]; + /*! For this SA which is not currently in use, the number of + * received packets that have been discarded, and have either the + * packets encrypted or the matched row in the Ingress SC Lookup + * table has validate_frames=Strict. + */ + u32 not_using_sa[2]; + /*! For this SA which is not currently in use, the number of + * received, unencrypted, packets with the matched row in the + * Ingress SC Lookup table has validate_frames!=Strict. + */ + u32 unused_sa[2]; + /*! For this SA, the number discarded packets with the condition + * that the packets are not valid and one of the following + * conditions are true: either the matched row in the Ingress SC + * Lookup table has validate_frames=Strict or the packets + * encrypted. + */ + u32 not_valid_pkts[2]; + /*! For this SA, the number of packets with the condition that the + * packets are not valid and the matched row in the Ingress SC + * Lookup table has validate_frames=Check. + */ + u32 invalid_pkts[2]; + /*! For this SA, the number of validated packets. */ + u32 ok_pkts[2]; + /*! For this SC, the number of received packets that have been + * discarded with the condition: the matched row in the Ingress + * SC Lookup table has replay_protect=1 and the PN of the packet + * is lower than the lower bound replay check PN. + */ + u32 late_pkts[2]; + /*! For this SA, the number of packets with the condition that the + * PN of the packets is lower than the lower bound replay + * protection PN. + */ + u32 delayed_pkts[2]; + /*! For this SC, the number of packets with the following condition: + * - the matched row in the Ingress SC Lookup table has + * replay_protect=0 or + * - the matched row in the Ingress SC Lookup table has + * replay_protect=1 and the packet is not encrypted and the + * integrity check has failed or + * - the matched row in the Ingress SC Lookup table has + * replay_protect=1 and the packet is encrypted and integrity + * check has failed. + */ + u32 unchecked_pkts[2]; + /*! The number of octets of plaintext recovered from received + * packets that were integrity protected but not encrypted. + */ + u32 validated_octets[2]; + /*! The number of octets of plaintext recovered from received + * packets that were integrity protected and encrypted. + */ + u32 decrypted_octets[2]; +}; + +/*! Represents the common Ingress MIB counters; the counter not + * associated with a particular SA. Counters are 64 bits, lower 32 + * bits in field[0]. + */ +struct aq_mss_ingress_common_counters { + /*! The number of received packets classified as MAC_CTL packets. */ + u32 ctl_pkts[2]; + /*! The number of received packets with the MAC security tag + * (SecTAG), not matching any rows in the Ingress Pre-MACSec + * Packet Classifier table. + */ + u32 tagged_miss_pkts[2]; + /*! The number of received packets without the MAC security tag + * (SecTAG), not matching any rows in the Ingress Pre-MACSec + * Packet Classifier table. + */ + u32 untagged_miss_pkts[2]; + /*! The number of received packets discarded without the MAC + * security tag (SecTAG) and with the matched row in the Ingress + * SC Lookup table having validate_frames=Strict. + */ + u32 notag_pkts[2]; + /*! The number of received packets without the MAC security tag + * (SecTAG) and with the matched row in the Ingress SC Lookup + * table having validate_frames!=Strict. + */ + u32 untagged_pkts[2]; + /*! The number of received packets discarded with an invalid + * SecTAG or a zero value PN or an invalid ICV. + */ + u32 bad_tag_pkts[2]; + /*! The number of received packets discarded with unknown SCI + * information with the condition: + * the matched row in the Ingress SC Lookup table has + * validate_frames=Strict or the C bit in the SecTAG is set. + */ + u32 no_sci_pkts[2]; + /*! The number of received packets with unknown SCI with the condition: + * The matched row in the Ingress SC Lookup table has + * validate_frames!=Strict and the C bit in the SecTAG is not set. + */ + u32 unknown_sci_pkts[2]; + /*! The number of received packets by the controlled port service + * that passed the Ingress Post-MACSec Packet Classifier table + * check. + */ + u32 ctrl_prt_pass_pkts[2]; + /*! The number of received packets by the uncontrolled port + * service that passed the Ingress Post-MACSec Packet Classifier + * table check. + */ + u32 unctrl_prt_pass_pkts[2]; + /*! The number of received packets by the controlled port service + * that failed the Ingress Post-MACSec Packet Classifier table + * check. + */ + u32 ctrl_prt_fail_pkts[2]; + /*! The number of received packets by the uncontrolled port + * service that failed the Ingress Post-MACSec Packet Classifier + * table check. + */ + u32 unctrl_prt_fail_pkts[2]; + /*! The number of received packets discarded because the packet + * length is greater than the ifMtu of the Common Port interface. + */ + u32 too_long_pkts[2]; + /*! The number of received packets classified as MAC_CTL by the + * Ingress Post-MACSec CTL Filter table. + */ + u32 igpoc_ctl_pkts[2]; + /*! The number of received packets for which table memory was + * affected by an ECC error during processing. + */ + u32 ecc_error_pkts[2]; + /*! The number of received packets by the uncontrolled port + * service that were dropped. + */ + u32 unctrl_hit_drop_redir[2]; +}; + +#endif diff --git a/drivers/net/ethernet/aquantia/atlantic/ver.h b/drivers/net/ethernet/aquantia/atlantic/ver.h deleted file mode 100644 index 597654b51e01..000000000000 --- a/drivers/net/ethernet/aquantia/atlantic/ver.h +++ /dev/null @@ -1,12 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved - */ - -#ifndef VER_H -#define VER_H - -#define AQ_CFG_DRV_VERSION_SUFFIX "-kern" - -#endif /* VER_H */ diff --git a/drivers/net/ethernet/arc/emac.h b/drivers/net/ethernet/arc/emac.h index d9efbc8d783b..d820ae03a966 100644 --- a/drivers/net/ethernet/arc/emac.h +++ b/drivers/net/ethernet/arc/emac.h @@ -130,7 +130,6 @@ struct arc_emac_mdio_bus_data { */ struct arc_emac_priv { const char *drv_name; - const char *drv_version; void (*set_mac_speed)(void *priv, unsigned int speed); /* Devices */ diff --git a/drivers/net/ethernet/arc/emac_arc.c b/drivers/net/ethernet/arc/emac_arc.c index 539166112993..1c7736b7eaf7 100644 --- a/drivers/net/ethernet/arc/emac_arc.c +++ b/drivers/net/ethernet/arc/emac_arc.c @@ -15,7 +15,6 @@ #include "emac.h" #define DRV_NAME "emac_arc" -#define DRV_VERSION "1.0" static int emac_arc_probe(struct platform_device *pdev) { @@ -36,7 +35,6 @@ static int emac_arc_probe(struct platform_device *pdev) priv = netdev_priv(ndev); priv->drv_name = DRV_NAME; - priv->drv_version = DRV_VERSION; err = of_get_phy_mode(dev->of_node, &interface); if (err) { diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c index 17bda4e8cc45..38cd968b6a3b 100644 --- a/drivers/net/ethernet/arc/emac_main.c +++ b/drivers/net/ethernet/arc/emac_main.c @@ -92,7 +92,6 @@ static void arc_emac_get_drvinfo(struct net_device *ndev, struct arc_emac_priv *priv = netdev_priv(ndev); strlcpy(info->driver, priv->drv_name, sizeof(info->driver)); - strlcpy(info->version, priv->drv_version, sizeof(info->version)); } static const struct ethtool_ops arc_emac_ethtool_ops = { diff --git a/drivers/net/ethernet/arc/emac_rockchip.c b/drivers/net/ethernet/arc/emac_rockchip.c index aae231c5224f..48ecdf15eddc 100644 --- a/drivers/net/ethernet/arc/emac_rockchip.c +++ b/drivers/net/ethernet/arc/emac_rockchip.c @@ -16,7 +16,6 @@ #include "emac.h" #define DRV_NAME "rockchip_emac" -#define DRV_VERSION "1.1" struct emac_rockchip_soc_data { unsigned int grf_offset; @@ -112,7 +111,6 @@ static int emac_rockchip_probe(struct platform_device *pdev) priv = netdev_priv(ndev); priv->emac.drv_name = DRV_NAME; - priv->emac.drv_version = DRV_VERSION; priv->emac.set_mac_speed = emac_rockchip_set_mac_speed; err = of_get_phy_mode(dev->of_node, &interface); diff --git a/drivers/net/ethernet/atheros/Kconfig b/drivers/net/ethernet/atheros/Kconfig index 0058051ba925..2720bde5034e 100644 --- a/drivers/net/ethernet/atheros/Kconfig +++ b/drivers/net/ethernet/atheros/Kconfig @@ -20,7 +20,7 @@ if NET_VENDOR_ATHEROS config AG71XX tristate "Atheros AR7XXX/AR9XXX built-in ethernet mac support" depends on ATH79 - select PHYLIB + select PHYLINK help If you wish to compile a kernel for AR7XXX/91XXX and enable ethernet support, then you should always answer Y to this. diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c index e95687a780fb..02b7705393ca 100644 --- a/drivers/net/ethernet/atheros/ag71xx.c +++ b/drivers/net/ethernet/atheros/ag71xx.c @@ -32,6 +32,7 @@ #include <linux/of_mdio.h> #include <linux/of_net.h> #include <linux/of_platform.h> +#include <linux/phylink.h> #include <linux/regmap.h> #include <linux/reset.h> #include <linux/clk.h> @@ -314,6 +315,8 @@ struct ag71xx { dma_addr_t stop_desc_dma; phy_interface_t phy_if_mode; + struct phylink *phylink; + struct phylink_config phylink_config; struct delayed_work restart_work; struct timer_list oom_timer; @@ -845,24 +848,91 @@ static void ag71xx_hw_start(struct ag71xx *ag) netif_wake_queue(ag->ndev); } -static void ag71xx_link_adjust(struct ag71xx *ag, bool update) +static void ag71xx_mac_config(struct phylink_config *config, unsigned int mode, + const struct phylink_link_state *state) { - struct phy_device *phydev = ag->ndev->phydev; - u32 cfg2; - u32 ifctl; - u32 fifo5; + struct ag71xx *ag = netdev_priv(to_net_dev(config->dev)); - if (!phydev->link && update) { - ag71xx_hw_stop(ag); + if (phylink_autoneg_inband(mode)) return; - } if (!ag71xx_is(ag, AR7100) && !ag71xx_is(ag, AR9130)) ag71xx_fast_reset(ag); + if (ag->tx_ring.desc_split) { + ag->fifodata[2] &= 0xffff; + ag->fifodata[2] |= ((2048 - ag->tx_ring.desc_split) / 4) << 16; + } + + ag71xx_wr(ag, AG71XX_REG_FIFO_CFG3, ag->fifodata[2]); +} + +static void ag71xx_mac_validate(struct phylink_config *config, + unsigned long *supported, + struct phylink_link_state *state) +{ + __ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, }; + + if (state->interface != PHY_INTERFACE_MODE_NA && + state->interface != PHY_INTERFACE_MODE_GMII && + state->interface != PHY_INTERFACE_MODE_MII) { + bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS); + return; + } + + phylink_set(mask, MII); + + phylink_set(mask, Autoneg); + phylink_set(mask, 10baseT_Half); + phylink_set(mask, 10baseT_Full); + phylink_set(mask, 100baseT_Half); + phylink_set(mask, 100baseT_Full); + + if (state->interface == PHY_INTERFACE_MODE_NA || + state->interface == PHY_INTERFACE_MODE_GMII) { + phylink_set(mask, 1000baseT_Full); + phylink_set(mask, 1000baseX_Full); + } + + bitmap_and(supported, supported, mask, + __ETHTOOL_LINK_MODE_MASK_NBITS); + bitmap_and(state->advertising, state->advertising, mask, + __ETHTOOL_LINK_MODE_MASK_NBITS); +} + +static void ag71xx_mac_pcs_get_state(struct phylink_config *config, + struct phylink_link_state *state) +{ + state->link = 0; +} + +static void ag71xx_mac_an_restart(struct phylink_config *config) +{ + /* Not Supported */ +} + +static void ag71xx_mac_link_down(struct phylink_config *config, + unsigned int mode, phy_interface_t interface) +{ + struct ag71xx *ag = netdev_priv(to_net_dev(config->dev)); + + ag71xx_hw_stop(ag); +} + +static void ag71xx_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause) +{ + struct ag71xx *ag = netdev_priv(to_net_dev(config->dev)); + u32 cfg2; + u32 ifctl; + u32 fifo5; + cfg2 = ag71xx_rr(ag, AG71XX_REG_MAC_CFG2); cfg2 &= ~(MAC_CFG2_IF_1000 | MAC_CFG2_IF_10_100 | MAC_CFG2_FDX); - cfg2 |= (phydev->duplex) ? MAC_CFG2_FDX : 0; + cfg2 |= duplex ? MAC_CFG2_FDX : 0; ifctl = ag71xx_rr(ag, AG71XX_REG_MAC_IFCTL); ifctl &= ~(MAC_IFCTL_SPEED); @@ -870,7 +940,7 @@ static void ag71xx_link_adjust(struct ag71xx *ag, bool update) fifo5 = ag71xx_rr(ag, AG71XX_REG_FIFO_CFG5); fifo5 &= ~FIFO_CFG5_BM; - switch (phydev->speed) { + switch (speed) { case SPEED_1000: cfg2 |= MAC_CFG2_IF_1000; fifo5 |= FIFO_CFG5_BM; @@ -883,72 +953,38 @@ static void ag71xx_link_adjust(struct ag71xx *ag, bool update) cfg2 |= MAC_CFG2_IF_10_100; break; default: - WARN(1, "not supported speed %i\n", phydev->speed); return; } - if (ag->tx_ring.desc_split) { - ag->fifodata[2] &= 0xffff; - ag->fifodata[2] |= ((2048 - ag->tx_ring.desc_split) / 4) << 16; - } - - ag71xx_wr(ag, AG71XX_REG_FIFO_CFG3, ag->fifodata[2]); - ag71xx_wr(ag, AG71XX_REG_MAC_CFG2, cfg2); ag71xx_wr(ag, AG71XX_REG_FIFO_CFG5, fifo5); ag71xx_wr(ag, AG71XX_REG_MAC_IFCTL, ifctl); ag71xx_hw_start(ag); - - if (update) - phy_print_status(phydev); } -static void ag71xx_phy_link_adjust(struct net_device *ndev) -{ - struct ag71xx *ag = netdev_priv(ndev); - - ag71xx_link_adjust(ag, true); -} +static const struct phylink_mac_ops ag71xx_phylink_mac_ops = { + .validate = ag71xx_mac_validate, + .mac_pcs_get_state = ag71xx_mac_pcs_get_state, + .mac_an_restart = ag71xx_mac_an_restart, + .mac_config = ag71xx_mac_config, + .mac_link_down = ag71xx_mac_link_down, + .mac_link_up = ag71xx_mac_link_up, +}; -static int ag71xx_phy_connect(struct ag71xx *ag) +static int ag71xx_phylink_setup(struct ag71xx *ag) { - struct device_node *np = ag->pdev->dev.of_node; - struct net_device *ndev = ag->ndev; - struct device_node *phy_node; - struct phy_device *phydev; - int ret; - - if (of_phy_is_fixed_link(np)) { - ret = of_phy_register_fixed_link(np); - if (ret < 0) { - netif_err(ag, probe, ndev, "Failed to register fixed PHY link: %d\n", - ret); - return ret; - } + struct phylink *phylink; - phy_node = of_node_get(np); - } else { - phy_node = of_parse_phandle(np, "phy-handle", 0); - } + ag->phylink_config.dev = &ag->ndev->dev; + ag->phylink_config.type = PHYLINK_NETDEV; - if (!phy_node) { - netif_err(ag, probe, ndev, "Could not find valid phy node\n"); - return -ENODEV; - } - - phydev = of_phy_connect(ag->ndev, phy_node, ag71xx_phy_link_adjust, - 0, ag->phy_if_mode); - - of_node_put(phy_node); - - if (!phydev) { - netif_err(ag, probe, ndev, "Could not connect to PHY device\n"); - return -ENODEV; - } - - phy_attached_info(phydev); + phylink = phylink_create(&ag->phylink_config, ag->pdev->dev.fwnode, + ag->phy_if_mode, &ag71xx_phylink_mac_ops); + if (IS_ERR(phylink)) + return PTR_ERR(phylink); + ag->phylink = phylink; return 0; } @@ -1239,6 +1275,13 @@ static int ag71xx_open(struct net_device *ndev) unsigned int max_frame_len; int ret; + ret = phylink_of_phy_connect(ag->phylink, ag->pdev->dev.of_node, 0); + if (ret) { + netif_err(ag, link, ndev, "phylink_of_phy_connect filed with err: %i\n", + ret); + goto err; + } + max_frame_len = ag71xx_max_frame_len(ndev->mtu); ag->rx_buf_size = SKB_DATA_ALIGN(max_frame_len + NET_SKB_PAD + NET_IP_ALIGN); @@ -1251,11 +1294,7 @@ static int ag71xx_open(struct net_device *ndev) if (ret) goto err; - ret = ag71xx_phy_connect(ag); - if (ret) - goto err; - - phy_start(ndev->phydev); + phylink_start(ag->phylink); return 0; @@ -1268,8 +1307,8 @@ static int ag71xx_stop(struct net_device *ndev) { struct ag71xx *ag = netdev_priv(ndev); - phy_stop(ndev->phydev); - phy_disconnect(ndev->phydev); + phylink_stop(ag->phylink); + phylink_disconnect_phy(ag->phylink); ag71xx_hw_disable(ag); return 0; @@ -1414,13 +1453,14 @@ static void ag71xx_restart_work_func(struct work_struct *work) { struct ag71xx *ag = container_of(work, struct ag71xx, restart_work.work); - struct net_device *ndev = ag->ndev; rtnl_lock(); ag71xx_hw_disable(ag); ag71xx_hw_enable(ag); - if (ndev->phydev->link) - ag71xx_link_adjust(ag, false); + + phylink_stop(ag->phylink); + phylink_start(ag->phylink); + rtnl_unlock(); } @@ -1759,6 +1799,12 @@ static int ag71xx_probe(struct platform_device *pdev) platform_set_drvdata(pdev, ndev); + err = ag71xx_phylink_setup(ag); + if (err) { + netif_err(ag, probe, ndev, "failed to setup phylink (%d)\n", err); + goto err_mdio_remove; + } + err = register_netdev(ndev); if (err) { netif_err(ag, probe, ndev, "unable to register net device\n"); diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c index 1dcbc486eca9..b9b4edb913c1 100644 --- a/drivers/net/ethernet/atheros/alx/main.c +++ b/drivers/net/ethernet/atheros/alx/main.c @@ -1416,10 +1416,7 @@ static int alx_tso(struct sk_buff *skb, struct alx_txd *first) 0, IPPROTO_TCP, 0); first->word1 |= 1 << TPD_IPV4_SHIFT; } else if (skb_is_gso_v6(skb)) { - ipv6_hdr(skb)->payload_len = 0; - tcp_hdr(skb)->check = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - 0, IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); /* LSOv2: the first TPD only provides the packet length */ first->adrl.l.pkt_len = skb->len; first->word1 |= 1 << TPD_LSO_V2_SHIFT; diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c.h b/drivers/net/ethernet/atheros/atl1c/atl1c.h index 60b2febd7315..a0562a90fb6d 100644 --- a/drivers/net/ethernet/atheros/atl1c/atl1c.h +++ b/drivers/net/ethernet/atheros/atl1c/atl1c.h @@ -583,7 +583,6 @@ struct atl1c_adapter { readl(((a)->hw_addr + reg) + ((offset) << 2))) extern char atl1c_driver_name[]; -extern char atl1c_driver_version[]; void atl1c_reinit_locked(struct atl1c_adapter *adapter); s32 atl1c_reset_hw(struct atl1c_hw *hw); diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c b/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c index b5a70a36fa04..e2eb7b8c63a0 100644 --- a/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c +++ b/drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c @@ -221,8 +221,6 @@ static void atl1c_get_drvinfo(struct net_device *netdev, struct atl1c_adapter *adapter = netdev_priv(netdev); strlcpy(drvinfo->driver, atl1c_driver_name, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, atl1c_driver_version, - sizeof(drvinfo->version)); strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), sizeof(drvinfo->bus_info)); } diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c index 4c0b1f8551dd..00bd7bd55794 100644 --- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c +++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c @@ -8,9 +8,7 @@ #include "atl1c.h" -#define ATL1C_DRV_VERSION "1.0.1.1-NAPI" char atl1c_driver_name[] = "atl1c"; -char atl1c_driver_version[] = ATL1C_DRV_VERSION; /* * atl1c_pci_tbl - PCI Device ID Table @@ -37,7 +35,6 @@ MODULE_AUTHOR("Jie Yang"); MODULE_AUTHOR("Qualcomm Atheros Inc., <nic-devel@qualcomm.com>"); MODULE_DESCRIPTION("Qualcomm Atheros 100/1000M Ethernet Network Driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(ATL1C_DRV_VERSION); static int atl1c_stop_mac(struct atl1c_hw *hw); static void atl1c_disable_l0s_l1(struct atl1c_hw *hw); @@ -2025,10 +2022,8 @@ static int atl1c_tso_csum(struct atl1c_adapter *adapter, "IPV6 tso with zero data??\n"); goto check_sum; } else - tcp_hdr(skb)->check = ~csum_ipv6_magic( - &ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - 0, IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); + etpd->word1 |= 1 << TPD_LSO_EN_SHIFT; etpd->word1 |= 1 << TPD_LSO_VER_SHIFT; etpd->pkt_len = cpu_to_le32(skb->len); @@ -2644,8 +2639,6 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_register; } - if (netif_msg_probe(adapter)) - dev_info(&pdev->dev, "version %s\n", ATL1C_DRV_VERSION); cards_found++; return 0; diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e.h b/drivers/net/ethernet/atheros/atl1e/atl1e.h index e9893da50995..9fcad783c939 100644 --- a/drivers/net/ethernet/atheros/atl1e/atl1e.h +++ b/drivers/net/ethernet/atheros/atl1e/atl1e.h @@ -482,7 +482,6 @@ struct atl1e_adapter { readl(((a)->hw_addr + reg) + ((offset) << 2))) extern char atl1e_driver_name[]; -extern char atl1e_driver_version[]; void atl1e_check_options(struct atl1e_adapter *adapter); int atl1e_up(struct atl1e_adapter *adapter); diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c b/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c index c6b9e7ea8e38..0cbde352d1ba 100644 --- a/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c +++ b/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c @@ -307,8 +307,6 @@ static void atl1e_get_drvinfo(struct net_device *netdev, struct atl1e_adapter *adapter = netdev_priv(netdev); strlcpy(drvinfo->driver, atl1e_driver_name, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, atl1e_driver_version, - sizeof(drvinfo->version)); strlcpy(drvinfo->fw_version, "L1e", sizeof(drvinfo->fw_version)); strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), sizeof(drvinfo->bus_info)); diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c index e0d89942d537..223ef846123e 100644 --- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c +++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c @@ -8,10 +8,7 @@ #include "atl1e.h" -#define DRV_VERSION "1.0.0.7-NAPI" - char atl1e_driver_name[] = "ATL1E"; -char atl1e_driver_version[] = DRV_VERSION; #define PCI_DEVICE_ID_ATTANSIC_L1E 0x1026 /* * atl1e_pci_tbl - PCI Device ID Table @@ -33,7 +30,6 @@ MODULE_DEVICE_TABLE(pci, atl1e_pci_tbl); MODULE_AUTHOR("Atheros Corporation, <xiong.huang@atheros.com>, Jie Yang <jie.yang@atheros.com>"); MODULE_DESCRIPTION("Atheros 1000M Ethernet Network Driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); static void atl1e_setup_mac_ctrl(struct atl1e_adapter *adapter); diff --git a/drivers/net/ethernet/atheros/atlx/atl1.c b/drivers/net/ethernet/atheros/atlx/atl1.c index b498fd6a47d0..271e7034fa70 100644 --- a/drivers/net/ethernet/atheros/atlx/atl1.c +++ b/drivers/net/ethernet/atheros/atlx/atl1.c @@ -65,12 +65,10 @@ #include "atl1.h" -#define ATLX_DRIVER_VERSION "2.1.3" MODULE_AUTHOR("Xiong Huang <xiong.huang@atheros.com>, " "Chris Snook <csnook@redhat.com>, " "Jay Cliburn <jcliburn@gmail.com>"); MODULE_LICENSE("GPL"); -MODULE_VERSION(ATLX_DRIVER_VERSION); /* Temporary hack for merging atl1 and atl2 */ #include "atlx.c" @@ -2965,8 +2963,6 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* get device revision number */ adapter->hw.dev_rev = ioread16(adapter->hw.hw_addr + (REG_MASTER_CTRL + 2)); - if (netif_msg_probe(adapter)) - dev_info(&pdev->dev, "version %s\n", ATLX_DRIVER_VERSION); /* set default ring resource counts */ adapter->rfd_ring.count = adapter->rrd_ring.count = ATL1_DEFAULT_RFD; @@ -3344,8 +3340,6 @@ static void atl1_get_drvinfo(struct net_device *netdev, struct atl1_adapter *adapter = netdev_priv(netdev); strlcpy(drvinfo->driver, ATLX_DRIVER_NAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, ATLX_DRIVER_VERSION, - sizeof(drvinfo->version)); strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), sizeof(drvinfo->bus_info)); } diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c index b81a4e0c5b57..c915852b8892 100644 --- a/drivers/net/ethernet/atheros/atlx/atl2.c +++ b/drivers/net/ethernet/atheros/atlx/atl2.c @@ -36,18 +36,12 @@ #include "atl2.h" -#define ATL2_DRV_VERSION "2.2.3" - static const char atl2_driver_name[] = "atl2"; -static const char atl2_driver_string[] = "Atheros(R) L2 Ethernet Driver"; -static const char atl2_copyright[] = "Copyright (c) 2007 Atheros Corporation."; -static const char atl2_driver_version[] = ATL2_DRV_VERSION; static const struct ethtool_ops atl2_ethtool_ops; MODULE_AUTHOR("Atheros Corporation <xiong.huang@atheros.com>, Chris Snook <csnook@redhat.com>"); MODULE_DESCRIPTION("Atheros Fast Ethernet Network Driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(ATL2_DRV_VERSION); /* * atl2_pci_tbl - PCI Device ID Table @@ -1688,9 +1682,6 @@ static struct pci_driver atl2_driver = { */ static int __init atl2_init_module(void) { - printk(KERN_INFO "%s - version %s\n", atl2_driver_string, - atl2_driver_version); - printk(KERN_INFO "%s\n", atl2_copyright); return pci_register_driver(&atl2_driver); } module_init(atl2_init_module); @@ -2011,8 +2002,6 @@ static void atl2_get_drvinfo(struct net_device *netdev, struct atl2_adapter *adapter = netdev_priv(netdev); strlcpy(drvinfo->driver, atl2_driver_name, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, atl2_driver_version, - sizeof(drvinfo->version)); strlcpy(drvinfo->fw_version, "L2", sizeof(drvinfo->fw_version)); strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), sizeof(drvinfo->bus_info)); diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index a780b7215021..6fb620e25208 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -40,7 +40,6 @@ #include "b44.h" #define DRV_MODULE_NAME "b44" -#define DRV_MODULE_VERSION "2.0" #define DRV_DESCRIPTION "Broadcom 44xx/47xx 10/100 PCI ethernet driver" #define B44_DEF_MSG_ENABLE \ @@ -97,7 +96,6 @@ MODULE_AUTHOR("Felix Fietkau, Florian Schirmer, Pekka Pietikainen, David S. Miller"); MODULE_DESCRIPTION(DRV_DESCRIPTION); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_MODULE_VERSION); static int b44_debug = -1; /* -1 == use B44_DEF_MSG_ENABLE as value */ module_param(b44_debug, int, 0); @@ -1791,7 +1789,6 @@ static void b44_get_drvinfo (struct net_device *dev, struct ethtool_drvinfo *inf struct ssb_bus *bus = bp->sdev->bus; strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); switch (bus->bustype) { case SSB_BUSTYPE_PCI: strlcpy(info->bus_info, pci_name(bus->host_pci), sizeof(info->bus_info)); @@ -2347,8 +2344,6 @@ static int b44_init_one(struct ssb_device *sdev, instance++; - pr_info_once("%s version %s\n", DRV_DESCRIPTION, DRV_MODULE_VERSION); - dev = alloc_etherdev(sizeof(*bp)); if (!dev) { err = -ENOMEM; diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c index 620cd3fc1fbc..916824cca3fd 100644 --- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c +++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c @@ -22,7 +22,6 @@ #include "bcm63xx_enet.h" static char bcm_enet_driver_name[] = "bcm63xx_enet"; -static char bcm_enet_driver_version[] = "1.0"; static int copybreak __read_mostly = 128; module_param(copybreak, int, 0); @@ -1304,9 +1303,6 @@ static void bcm_enet_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { strlcpy(drvinfo->driver, bcm_enet_driver_name, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, bcm_enet_driver_version, - sizeof(drvinfo->version)); - strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version)); strlcpy(drvinfo->bus_info, "bcm63xx", sizeof(drvinfo->bus_info)); } @@ -1706,7 +1702,6 @@ static int bcm_enet_probe(struct platform_device *pdev) if (!res_irq || !res_irq_rx || !res_irq_tx) return -ENODEV; - ret = 0; dev = alloc_etherdev(sizeof(*priv)); if (!dev) return -ENOMEM; @@ -2529,10 +2524,8 @@ static int bcm_enetsw_get_sset_count(struct net_device *netdev, static void bcm_enetsw_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { - strncpy(drvinfo->driver, bcm_enet_driver_name, 32); - strncpy(drvinfo->version, bcm_enet_driver_version, 32); - strncpy(drvinfo->fw_version, "N/A", 32); - strncpy(drvinfo->bus_info, "bcm63xx", 32); + strncpy(drvinfo->driver, bcm_enet_driver_name, sizeof(drvinfo->driver)); + strncpy(drvinfo->bus_info, "bcm63xx", sizeof(drvinfo->bus_info)); } static void bcm_enetsw_get_ethtool_stats(struct net_device *netdev, diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 15b31cddc054..af7ce5c5488c 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -287,7 +287,6 @@ static void bcm_sysport_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver)); - strlcpy(info->version, "0.1", sizeof(info->version)); strlcpy(info->bus_info, "platform", sizeof(info->bus_info)); } @@ -624,8 +623,7 @@ static int bcm_sysport_set_coalesce(struct net_device *dev, return -EINVAL; if ((ec->tx_coalesce_usecs == 0 && ec->tx_max_coalesced_frames == 0) || - (ec->rx_coalesce_usecs == 0 && ec->rx_max_coalesced_frames == 0) || - ec->use_adaptive_tx_coalesce) + (ec->rx_coalesce_usecs == 0 && ec->rx_max_coalesced_frames == 0)) return -EINVAL; for (i = 0; i < dev->num_tx_queues; i++) @@ -2210,6 +2208,9 @@ static int bcm_sysport_set_rxnfc(struct net_device *dev, } static const struct ethtool_ops bcm_sysport_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_drvinfo = bcm_sysport_get_drvinfo, .get_msglevel = bcm_sysport_get_msglvl, .set_msglevel = bcm_sysport_set_msglvl, diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 1bb07a5d82c9..98ec1b8a7d8e 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -1248,6 +1248,14 @@ static int bgmac_set_mac_address(struct net_device *net_dev, void *addr) return 0; } +static int bgmac_change_mtu(struct net_device *net_dev, int mtu) +{ + struct bgmac *bgmac = netdev_priv(net_dev); + + bgmac_write(bgmac, BGMAC_RXMAX_LENGTH, 32 + mtu); + return 0; +} + static const struct net_device_ops bgmac_netdev_ops = { .ndo_open = bgmac_open, .ndo_stop = bgmac_stop, @@ -1256,6 +1264,7 @@ static const struct net_device_ops bgmac_netdev_ops = { .ndo_set_mac_address = bgmac_set_mac_address, .ndo_validate_addr = eth_validate_addr, .ndo_do_ioctl = phy_do_ioctl_running, + .ndo_change_mtu = bgmac_change_mtu, }; /************************************************** @@ -1530,6 +1539,9 @@ int bgmac_enet_probe(struct bgmac *bgmac) net_dev->hw_features = net_dev->features; net_dev->vlan_features = net_dev->features; + /* Omit FCS from max MTU size */ + net_dev->max_mtu = BGMAC_RX_MAX_FRAME_SIZE - ETH_FCS_LEN; + err = register_netdev(bgmac->net_dev); if (err) { dev_err(bgmac->dev, "Cannot register net device\n"); diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index 40d02fec2747..351c598a3ec6 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -351,7 +351,7 @@ #define BGMAC_DESC_CTL0_IOC 0x20000000 /* IRQ on complete */ #define BGMAC_DESC_CTL0_EOF 0x40000000 /* End of frame */ #define BGMAC_DESC_CTL0_SOF 0x80000000 /* Start of frame */ -#define BGMAC_DESC_CTL1_LEN 0x00001FFF +#define BGMAC_DESC_CTL1_LEN 0x00003FFF #define BGMAC_PHY_NOREGS BRCM_PSEUDO_PHY_ADDR #define BGMAC_PHY_MASK 0x1F @@ -366,7 +366,8 @@ #define BGMAC_RX_FRAME_OFFSET 30 /* There are 2 unused bytes between header and real data */ #define BGMAC_RX_BUF_OFFSET (NET_SKB_PAD + NET_IP_ALIGN - \ BGMAC_RX_FRAME_OFFSET) -#define BGMAC_RX_MAX_FRAME_SIZE 1536 /* Copied from b44/tg3 */ +/* Jumbo frame size with FCS */ +#define BGMAC_RX_MAX_FRAME_SIZE 9724 #define BGMAC_RX_BUF_SIZE (BGMAC_RX_FRAME_OFFSET + BGMAC_RX_MAX_FRAME_SIZE) #define BGMAC_RX_ALLOC_SIZE (SKB_DATA_ALIGN(BGMAC_RX_BUF_SIZE + BGMAC_RX_BUF_OFFSET) + \ SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c index dbb7874607ca..e1c236cab2a7 100644 --- a/drivers/net/ethernet/broadcom/bnx2.c +++ b/drivers/net/ethernet/broadcom/bnx2.c @@ -59,8 +59,6 @@ #include "bnx2_fw.h" #define DRV_MODULE_NAME "bnx2" -#define DRV_MODULE_VERSION "2.2.6" -#define DRV_MODULE_RELDATE "January 29, 2014" #define FW_MIPS_FILE_06 "bnx2/bnx2-mips-06-6.2.3.fw" #define FW_RV2P_FILE_06 "bnx2/bnx2-rv2p-06-6.0.15.fw" #define FW_MIPS_FILE_09 "bnx2/bnx2-mips-09-6.2.1b.fw" @@ -72,13 +70,9 @@ /* Time in jiffies before concluding the transmitter is hung. */ #define TX_TIMEOUT (5*HZ) -static char version[] = - "QLogic " DRV_MODULE_NAME " Gigabit Ethernet Driver v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; - MODULE_AUTHOR("Michael Chan <mchan@broadcom.com>"); MODULE_DESCRIPTION("QLogic BCM5706/5708/5709/5716 Driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_MODULE_VERSION); MODULE_FIRMWARE(FW_MIPS_FILE_06); MODULE_FIRMWARE(FW_RV2P_FILE_06); MODULE_FIRMWARE(FW_MIPS_FILE_09); @@ -7048,7 +7042,6 @@ bnx2_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) struct bnx2 *bp = netdev_priv(dev); strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info)); strlcpy(info->fw_version, bp->fw_version, sizeof(info->fw_version)); } @@ -7819,6 +7812,11 @@ static int bnx2_set_channels(struct net_device *dev, } static const struct ethtool_ops bnx2_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USECS_IRQ | + ETHTOOL_COALESCE_MAX_FRAMES_IRQ | + ETHTOOL_COALESCE_STATS_BLOCK_USECS, .get_drvinfo = bnx2_get_drvinfo, .get_regs_len = bnx2_get_regs_len, .get_regs = bnx2_get_regs, @@ -8562,15 +8560,11 @@ static const struct net_device_ops bnx2_netdev_ops = { static int bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { - static int version_printed = 0; struct net_device *dev; struct bnx2 *bp; int rc; char str[40]; - if (version_printed++ == 0) - pr_info("%s", version); - /* dev zeroed in init_etherdev */ dev = alloc_etherdev_mq(sizeof(*bp), TX_MAX_RINGS); if (!dev) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 6026b53137aa..4f5b2b81be3d 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -32,8 +32,14 @@ * (you will need to reboot afterwards) */ /* #define BNX2X_STOP_ON_ERROR */ +/* FIXME: Delete the DRV_MODULE_VERSION below, but please be warned + * that it is not an easy task because such change has all chances + * to break this driver due to amount of abuse of in-kernel interfaces + * between modules and FW. + * + * DO NOT UPDATE DRV_MODULE_VERSION below. + */ #define DRV_MODULE_VERSION "1.713.36-0" -#define DRV_MODULE_RELDATE "2014/02/10" #define BNX2X_BC_VER 0x040200 #if defined(CONFIG_DCB) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c index 4a0ba6801c9e..7cea33803f7f 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c @@ -1112,13 +1112,6 @@ static void bnx2x_get_drvinfo(struct net_device *dev, u32 mbi; strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); - - memset(version, 0, sizeof(version)); - snprintf(version, ETHTOOL_FWVERS_LEN, " storm %d.%d.%d.%d", - BCM_5710_FW_MAJOR_VERSION, BCM_5710_FW_MINOR_VERSION, - BCM_5710_FW_REVISION_VERSION, BCM_5710_FW_ENGINEERING_VERSION); - strlcat(info->version, version, sizeof(info->version)); if (SHMEM2_HAS(bp, extended_dev_info_shared_addr)) { ext_dev_info_offset = SHMEM2_RD(bp, @@ -3663,6 +3656,7 @@ static int bnx2x_get_ts_info(struct net_device *dev, } static const struct ethtool_ops bnx2x_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_drvinfo = bnx2x_get_drvinfo, .get_regs_len = bnx2x_get_regs_len, .get_regs = bnx2x_get_regs, diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 1c26fa962233..db5107e7937c 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -81,17 +81,12 @@ /* Time in jiffies before concluding the transmitter is hung */ #define TX_TIMEOUT (5*HZ) -static char version[] = - "QLogic 5771x/578xx 10/20-Gigabit Ethernet Driver " - DRV_MODULE_NAME " " DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; - MODULE_AUTHOR("Eliezer Tamir"); MODULE_DESCRIPTION("QLogic " "BCM57710/57711/57711E/" "57712/57712_MF/57800/57800_MF/57810/57810_MF/" "57840/57840_MF Driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_MODULE_VERSION); MODULE_FIRMWARE(FW_FILE_NAME_E1); MODULE_FIRMWARE(FW_FILE_NAME_E1H); MODULE_FIRMWARE(FW_FILE_NAME_E2); @@ -14480,8 +14475,6 @@ static int __init bnx2x_init(void) { int ret; - pr_info("%s", version); - bnx2x_wq = create_singlethread_workqueue("bnx2x"); if (bnx2x_wq == NULL) { pr_err("Cannot create workqueue\n"); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index d28b406a26b1..fead64f1ad90 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -70,12 +70,8 @@ #define BNXT_TX_TIMEOUT (5 * HZ) -static const char version[] = - "Broadcom NetXtreme-C/E driver " DRV_MODULE_NAME " v" DRV_MODULE_VERSION "\n"; - MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("Broadcom BCM573xx network driver"); -MODULE_VERSION(DRV_MODULE_VERSION); #define BNXT_RX_OFFSET (NET_SKB_PAD + NET_IP_ALIGN) #define BNXT_RX_DMA_OFFSET NET_SKB_PAD @@ -2166,6 +2162,7 @@ static int __bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, struct tx_cmp *txcmp; cpr->has_more_work = 0; + cpr->had_work_done = 1; while (1) { int rc; @@ -2179,7 +2176,6 @@ static int __bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, * reading any further. */ dma_rmb(); - cpr->had_work_done = 1; if (TX_CMP_TYPE(txcmp) == CMP_TYPE_TX_L2_CMP) { tx_pkts++; /* return full budget so NAPI will complete. */ @@ -2396,7 +2392,7 @@ static int __bnxt_poll_cqs(struct bnxt *bp, struct bnxt_napi *bnapi, int budget) } static void __bnxt_poll_cqs_done(struct bnxt *bp, struct bnxt_napi *bnapi, - u64 dbr_type, bool all) + u64 dbr_type) { struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring; int i; @@ -2405,7 +2401,7 @@ static void __bnxt_poll_cqs_done(struct bnxt *bp, struct bnxt_napi *bnapi, struct bnxt_cp_ring_info *cpr2 = cpr->cp_ring_arr[i]; struct bnxt_db_info *db; - if (cpr2 && (all || cpr2->had_work_done)) { + if (cpr2 && cpr2->had_work_done) { db = &cpr2->cp_db; writeq(db->db_key64 | dbr_type | RING_CMP(cpr2->cp_raw_cons), db->doorbell); @@ -2428,22 +2424,16 @@ static int bnxt_poll_p5(struct napi_struct *napi, int budget) if (cpr->has_more_work) { cpr->has_more_work = 0; work_done = __bnxt_poll_cqs(bp, bnapi, budget); - if (cpr->has_more_work) { - __bnxt_poll_cqs_done(bp, bnapi, DBR_TYPE_CQ, false); - return work_done; - } - __bnxt_poll_cqs_done(bp, bnapi, DBR_TYPE_CQ_ARMALL, true); - if (napi_complete_done(napi, work_done)) - BNXT_DB_NQ_ARM_P5(&cpr->cp_db, cpr->cp_raw_cons); - return work_done; } while (1) { cons = RING_CMP(raw_cons); nqcmp = &cpr->nq_desc_ring[CP_RING(cons)][CP_IDX(cons)]; if (!NQ_CMP_VALID(nqcmp, raw_cons)) { - __bnxt_poll_cqs_done(bp, bnapi, DBR_TYPE_CQ_ARMALL, - false); + if (cpr->has_more_work) + break; + + __bnxt_poll_cqs_done(bp, bnapi, DBR_TYPE_CQ_ARMALL); cpr->cp_raw_cons = raw_cons; if (napi_complete_done(napi, work_done)) BNXT_DB_NQ_ARM_P5(&cpr->cp_db, @@ -2463,16 +2453,17 @@ static int bnxt_poll_p5(struct napi_struct *napi, int budget) cpr2 = cpr->cp_ring_arr[idx]; work_done += __bnxt_poll_work(bp, cpr2, budget - work_done); - cpr->has_more_work = cpr2->has_more_work; + cpr->has_more_work |= cpr2->has_more_work; } else { bnxt_hwrm_handler(bp, (struct tx_cmp *)nqcmp); } raw_cons = NEXT_RAW_CMP(raw_cons); - if (cpr->has_more_work) - break; } - __bnxt_poll_cqs_done(bp, bnapi, DBR_TYPE_CQ, true); - cpr->cp_raw_cons = raw_cons; + __bnxt_poll_cqs_done(bp, bnapi, DBR_TYPE_CQ); + if (raw_cons != cpr->cp_raw_cons) { + cpr->cp_raw_cons = raw_cons; + BNXT_DB_NQ_P5(&cpr->cp_db, raw_cons); + } return work_done; } @@ -4170,6 +4161,7 @@ static int bnxt_hwrm_to_stderr(u32 hwrm_err) case HWRM_ERR_CODE_NO_BUFFER: return -ENOMEM; case HWRM_ERR_CODE_HOT_RESET_PROGRESS: + case HWRM_ERR_CODE_BUSY: return -EAGAIN; case HWRM_ERR_CODE_CMD_NOT_SUPPORTED: return -EOPNOTSUPP; @@ -5069,10 +5061,8 @@ vnic_mru: return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } -static int bnxt_hwrm_vnic_free_one(struct bnxt *bp, u16 vnic_id) +static void bnxt_hwrm_vnic_free_one(struct bnxt *bp, u16 vnic_id) { - u32 rc = 0; - if (bp->vnic_info[vnic_id].fw_vnic_id != INVALID_HW_RING_ID) { struct hwrm_vnic_free_input req = {0}; @@ -5080,10 +5070,9 @@ static int bnxt_hwrm_vnic_free_one(struct bnxt *bp, u16 vnic_id) req.vnic_id = cpu_to_le32(bp->vnic_info[vnic_id].fw_vnic_id); - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); + hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); bp->vnic_info[vnic_id].fw_vnic_id = INVALID_HW_RING_ID; } - return rc; } static void bnxt_hwrm_vnic_free(struct bnxt *bp) @@ -5200,14 +5189,13 @@ static int bnxt_hwrm_ring_grp_alloc(struct bnxt *bp) return rc; } -static int bnxt_hwrm_ring_grp_free(struct bnxt *bp) +static void bnxt_hwrm_ring_grp_free(struct bnxt *bp) { u16 i; - u32 rc = 0; struct hwrm_ring_grp_free_input req = {0}; if (!bp->grp_info || (bp->flags & BNXT_FLAG_CHIP_P5)) - return 0; + return; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_RING_GRP_FREE, -1, -1); @@ -5218,12 +5206,10 @@ static int bnxt_hwrm_ring_grp_free(struct bnxt *bp) req.ring_group_id = cpu_to_le32(bp->grp_info[i].fw_grp_id); - rc = _hwrm_send_message(bp, &req, sizeof(req), - HWRM_CMD_TIMEOUT); + _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); bp->grp_info[i].fw_grp_id = INVALID_HW_RING_ID; } mutex_unlock(&bp->hwrm_cmd_lock); - return rc; } static int hwrm_ring_alloc_send_msg(struct bnxt *bp, @@ -5847,8 +5833,7 @@ bnxt_hwrm_reserve_pf_rings(struct bnxt *bp, int tx_rings, int rx_rings, if (bp->hwrm_spec_code < 0x10601) bp->hw_resc.resv_tx_rings = tx_rings; - rc = bnxt_hwrm_get_rings(bp); - return rc; + return bnxt_hwrm_get_rings(bp); } static int @@ -5869,8 +5854,7 @@ bnxt_hwrm_reserve_vf_rings(struct bnxt *bp, int tx_rings, int rx_rings, if (rc) return rc; - rc = bnxt_hwrm_get_rings(bp); - return rc; + return bnxt_hwrm_get_rings(bp); } static int bnxt_hwrm_reserve_rings(struct bnxt *bp, int tx, int rx, int grp, @@ -6030,7 +6014,6 @@ static int bnxt_hwrm_check_vf_rings(struct bnxt *bp, int tx_rings, int rx_rings, { struct hwrm_func_vf_cfg_input req = {0}; u32 flags; - int rc; if (!BNXT_NEW_RM(bp)) return 0; @@ -6047,8 +6030,8 @@ static int bnxt_hwrm_check_vf_rings(struct bnxt *bp, int tx_rings, int rx_rings, flags |= FUNC_VF_CFG_REQ_FLAGS_RING_GRP_ASSETS_TEST; req.flags = cpu_to_le32(flags); - rc = hwrm_send_message_silent(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message_silent(bp, &req, sizeof(req), + HWRM_CMD_TIMEOUT); } static int bnxt_hwrm_check_pf_rings(struct bnxt *bp, int tx_rings, int rx_rings, @@ -6057,7 +6040,6 @@ static int bnxt_hwrm_check_pf_rings(struct bnxt *bp, int tx_rings, int rx_rings, { struct hwrm_func_cfg_input req = {0}; u32 flags; - int rc; __bnxt_hwrm_reserve_pf_rings(bp, &req, tx_rings, rx_rings, ring_grps, cp_rings, stats, vnics); @@ -6075,8 +6057,8 @@ static int bnxt_hwrm_check_pf_rings(struct bnxt *bp, int tx_rings, int rx_rings, } req.flags = cpu_to_le32(flags); - rc = hwrm_send_message_silent(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message_silent(bp, &req, sizeof(req), + HWRM_CMD_TIMEOUT); } static int bnxt_hwrm_check_rings(struct bnxt *bp, int tx_rings, int rx_rings, @@ -6315,16 +6297,16 @@ int bnxt_hwrm_set_coal(struct bnxt *bp) return rc; } -static int bnxt_hwrm_stat_ctx_free(struct bnxt *bp) +static void bnxt_hwrm_stat_ctx_free(struct bnxt *bp) { - int rc = 0, i; struct hwrm_stat_ctx_free_input req = {0}; + int i; if (!bp->bnapi) - return 0; + return; if (BNXT_CHIP_TYPE_NITRO_A0(bp)) - return 0; + return; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_STAT_CTX_FREE, -1, -1); @@ -6336,14 +6318,13 @@ static int bnxt_hwrm_stat_ctx_free(struct bnxt *bp) if (cpr->hw_stats_ctx_id != INVALID_STATS_CTX_ID) { req.stat_ctx_id = cpu_to_le32(cpr->hw_stats_ctx_id); - rc = _hwrm_send_message(bp, &req, sizeof(req), - HWRM_CMD_TIMEOUT); + _hwrm_send_message(bp, &req, sizeof(req), + HWRM_CMD_TIMEOUT); cpr->hw_stats_ctx_id = INVALID_STATS_CTX_ID; } } mutex_unlock(&bp->hwrm_cmd_lock); - return rc; } static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp) @@ -6548,8 +6529,8 @@ static int bnxt_hwrm_func_backing_store_cfg(struct bnxt *bp, u32 enables) __le64 *pg_dir; u32 flags = 0; u8 *pg_attr; - int i, rc; u32 ena; + int i; if (!ctx) return 0; @@ -6636,8 +6617,7 @@ static int bnxt_hwrm_func_backing_store_cfg(struct bnxt *bp, u32 enables) bnxt_hwrm_set_pg_attr(&ctx_pg->ring_mem, pg_attr, pg_dir); } req.flags = cpu_to_le32(flags); - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static int bnxt_alloc_ctx_mem_blk(struct bnxt *bp, @@ -7243,7 +7223,7 @@ static int __bnxt_hwrm_ver_get(struct bnxt *bp, bool silent) static int bnxt_hwrm_ver_get(struct bnxt *bp) { struct hwrm_ver_get_output *resp = bp->hwrm_cmd_resp_addr; - u32 dev_caps_cfg; + u32 dev_caps_cfg, hwrm_ver; int rc; bp->hwrm_max_req_len = HWRM_MAX_REQ_LEN; @@ -7263,6 +7243,19 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp) resp->hwrm_intf_upd_8b); netdev_warn(bp->dev, "Please update firmware with HWRM interface 1.0.0 or newer.\n"); } + + hwrm_ver = HWRM_VERSION_MAJOR << 16 | HWRM_VERSION_MINOR << 8 | + HWRM_VERSION_UPDATE; + + if (bp->hwrm_spec_code > hwrm_ver) + snprintf(bp->hwrm_ver_supp, FW_VER_STR_LEN, "%d.%d.%d", + HWRM_VERSION_MAJOR, HWRM_VERSION_MINOR, + HWRM_VERSION_UPDATE); + else + snprintf(bp->hwrm_ver_supp, FW_VER_STR_LEN, "%d.%d.%d", + resp->hwrm_intf_maj_8b, resp->hwrm_intf_min_8b, + resp->hwrm_intf_upd_8b); + snprintf(bp->fw_ver_str, BC_HWRM_STR_LEN, "%d.%d.%d.%d", resp->hwrm_fw_maj_8b, resp->hwrm_fw_min_8b, resp->hwrm_fw_bld_8b, resp->hwrm_fw_rsvd_8b); @@ -7341,7 +7334,6 @@ int bnxt_hwrm_fw_set_time(struct bnxt *bp) static int bnxt_hwrm_port_qstats(struct bnxt *bp) { - int rc; struct bnxt_pf_info *pf = &bp->pf; struct hwrm_port_qstats_input req = {0}; @@ -7352,8 +7344,7 @@ static int bnxt_hwrm_port_qstats(struct bnxt *bp) req.port_id = cpu_to_le16(pf->port_id); req.tx_stat_host_addr = cpu_to_le64(bp->hw_tx_port_stats_map); req.rx_stat_host_addr = cpu_to_le64(bp->hw_rx_port_stats_map); - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static int bnxt_hwrm_port_qstats_ext(struct bnxt *bp) @@ -7515,7 +7506,6 @@ static void bnxt_hwrm_resource_free(struct bnxt *bp, bool close_path, static int bnxt_hwrm_set_br_mode(struct bnxt *bp, u16 br_mode) { struct hwrm_func_cfg_input req = {0}; - int rc; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_CFG, -1, -1); req.fid = cpu_to_le16(0xffff); @@ -7526,14 +7516,12 @@ static int bnxt_hwrm_set_br_mode(struct bnxt *bp, u16 br_mode) req.evb_mode = FUNC_CFG_REQ_EVB_MODE_VEPA; else return -EINVAL; - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static int bnxt_hwrm_set_cache_line_size(struct bnxt *bp, int size) { struct hwrm_func_cfg_input req = {0}; - int rc; if (BNXT_VF(bp) || bp->hwrm_spec_code < 0x10803) return 0; @@ -7545,8 +7533,7 @@ static int bnxt_hwrm_set_cache_line_size(struct bnxt *bp, int size) if (size == 128) req.options = FUNC_CFG_REQ_OPTIONS_CACHE_LINESIZE_SIZE_128; - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static int __bnxt_setup_vnic(struct bnxt *bp, u16 vnic_id) @@ -8804,6 +8791,7 @@ static int bnxt_hwrm_if_change(struct bnxt *bp, bool up) bnxt_free_ctx_mem(bp); kfree(bp->ctx); bp->ctx = NULL; + bnxt_dcb_free(bp); rc = bnxt_fw_init_one(bp); if (rc) { set_bit(BNXT_STATE_ABORT_ERR, &bp->state); @@ -8899,14 +8887,12 @@ int bnxt_hwrm_alloc_wol_fltr(struct bnxt *bp) int bnxt_hwrm_free_wol_fltr(struct bnxt *bp) { struct hwrm_wol_filter_free_input req = {0}; - int rc; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_WOL_FILTER_FREE, -1, -1); req.port_id = cpu_to_le16(bp->pf.port_id); req.enables = cpu_to_le32(WOL_FILTER_FREE_REQ_ENABLES_WOL_FILTER_ID); req.wol_filter_id = bp->wol_filter_id; - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static u16 bnxt_hwrm_get_wol_fltrs(struct bnxt *bp, u16 handle) @@ -11464,6 +11450,8 @@ static void bnxt_remove_one(struct pci_dev *pdev) bnxt_sriov_disable(bp); bnxt_dl_fw_reporters_destroy(bp, true); + if (BNXT_PF(bp)) + devlink_port_type_clear(&bp->dl_port); pci_disable_pcie_error_reporting(pdev); unregister_netdev(dev); bnxt_dl_unregister(bp); @@ -11764,30 +11752,82 @@ static int bnxt_init_mac_addr(struct bnxt *bp) return rc; } +#define BNXT_VPD_LEN 512 +static void bnxt_vpd_read_info(struct bnxt *bp) +{ + struct pci_dev *pdev = bp->pdev; + int i, len, pos, ro_size; + ssize_t vpd_size; + u8 *vpd_data; + + vpd_data = kmalloc(BNXT_VPD_LEN, GFP_KERNEL); + if (!vpd_data) + return; + + vpd_size = pci_read_vpd(pdev, 0, BNXT_VPD_LEN, vpd_data); + if (vpd_size <= 0) { + netdev_err(bp->dev, "Unable to read VPD\n"); + goto exit; + } + + i = pci_vpd_find_tag(vpd_data, 0, vpd_size, PCI_VPD_LRDT_RO_DATA); + if (i < 0) { + netdev_err(bp->dev, "VPD READ-Only not found\n"); + goto exit; + } + + ro_size = pci_vpd_lrdt_size(&vpd_data[i]); + i += PCI_VPD_LRDT_TAG_SIZE; + if (i + ro_size > vpd_size) + goto exit; + + pos = pci_vpd_find_info_keyword(vpd_data, i, ro_size, + PCI_VPD_RO_KEYWORD_PARTNO); + if (pos < 0) + goto read_sn; + + len = pci_vpd_info_field_size(&vpd_data[pos]); + pos += PCI_VPD_INFO_FLD_HDR_SIZE; + if (len + pos > vpd_size) + goto read_sn; + + strlcpy(bp->board_partno, &vpd_data[pos], min(len, BNXT_VPD_FLD_LEN)); + +read_sn: + pos = pci_vpd_find_info_keyword(vpd_data, i, ro_size, + PCI_VPD_RO_KEYWORD_SERIALNO); + if (pos < 0) + goto exit; + + len = pci_vpd_info_field_size(&vpd_data[pos]); + pos += PCI_VPD_INFO_FLD_HDR_SIZE; + if (len + pos > vpd_size) + goto exit; + + strlcpy(bp->board_serialno, &vpd_data[pos], min(len, BNXT_VPD_FLD_LEN)); +exit: + kfree(vpd_data); +} + static int bnxt_pcie_dsn_get(struct bnxt *bp, u8 dsn[]) { struct pci_dev *pdev = bp->pdev; - int pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_DSN); - u32 dw; + u64 qword; - if (!pos) { - netdev_info(bp->dev, "Unable do read adapter's DSN\n"); + qword = pci_get_dsn(pdev); + if (!qword) { + netdev_info(bp->dev, "Unable to read adapter's DSN\n"); return -EOPNOTSUPP; } - /* DSN (two dw) is at an offset of 4 from the cap pos */ - pos += 4; - pci_read_config_dword(pdev, pos, &dw); - put_unaligned_le32(dw, &dsn[0]); - pci_read_config_dword(pdev, pos + 4, &dw); - put_unaligned_le32(dw, &dsn[4]); + put_unaligned_le64(qword, dsn); + bp->flags |= BNXT_FLAG_DSN_VALID; return 0; } static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { - static int version_printed; struct net_device *dev; struct bnxt *bp; int rc, max_irqs; @@ -11795,9 +11835,6 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (pci_is_bridge(pdev)) return -ENODEV; - if (version_printed++ == 0) - pr_info("%s", version); - /* Clear any pending DMA transactions from crash kernel * while loading driver in capture kernel. */ @@ -11829,6 +11866,8 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) dev->ethtool_ops = &bnxt_ethtool_ops; pci_set_drvdata(pdev, dev); + bnxt_vpd_read_info(bp); + rc = bnxt_alloc_hwrm_resources(bp); if (rc) goto init_err_pci_clean; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 63b170658532..f2caa2756f5b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -12,8 +12,10 @@ #define BNXT_H #define DRV_MODULE_NAME "bnxt_en" -#define DRV_MODULE_VERSION "1.10.1" +/* DO NOT CHANGE DRV_VER_* defines + * FIXME: Delete them + */ #define DRV_VER_MAJ 1 #define DRV_VER_MIN 10 #define DRV_VER_UPD 1 @@ -1498,6 +1500,10 @@ struct bnxt { (chip_num) == CHIP_NUM_58804 || \ (chip_num) == CHIP_NUM_58808) +#define BNXT_VPD_FLD_LEN 32 + char board_partno[BNXT_VPD_FLD_LEN]; + char board_serialno[BNXT_VPD_FLD_LEN]; + struct net_device *dev; struct pci_dev *pdev; @@ -1728,6 +1734,7 @@ struct bnxt { #define BC_HWRM_STR_LEN 21 #define PHY_VER_STR_LEN (FW_VER_STR_LEN - BC_HWRM_STR_LEN) char fw_ver_str[FW_VER_STR_LEN]; + char hwrm_ver_supp[FW_VER_STR_LEN]; __be16 vxlan_port; u8 vxlan_port_cnt; __le16 vxlan_fw_dst_port_id; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c index b1511bcffb1b..02b27551d34d 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c @@ -39,8 +39,8 @@ static int bnxt_queue_to_tc(struct bnxt *bp, u8 queue_id) static int bnxt_hwrm_queue_pri2cos_cfg(struct bnxt *bp, struct ieee_ets *ets) { struct hwrm_queue_pri2cos_cfg_input req = {0}; - int rc = 0, i; u8 *pri2cos; + int i; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_PRI2COS_CFG, -1, -1); req.flags = cpu_to_le32(QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_BIDIR | @@ -56,8 +56,7 @@ static int bnxt_hwrm_queue_pri2cos_cfg(struct bnxt *bp, struct ieee_ets *ets) qidx = bp->tc_to_qidx[ets->prio_tc[i]]; pri2cos[i] = bp->q_info[qidx].queue_id; } - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static int bnxt_hwrm_queue_pri2cos_qcfg(struct bnxt *bp, struct ieee_ets *ets) @@ -93,8 +92,8 @@ static int bnxt_hwrm_queue_cos2bw_cfg(struct bnxt *bp, struct ieee_ets *ets, { struct hwrm_queue_cos2bw_cfg_input req = {0}; struct bnxt_cos2bw_cfg cos2bw; - int rc = 0, i; void *data; + int i; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_COS2BW_CFG, -1, -1); for (i = 0; i < max_tc; i++) { @@ -128,8 +127,7 @@ static int bnxt_hwrm_queue_cos2bw_cfg(struct bnxt *bp, struct ieee_ets *ets, req.unused_0 = 0; } } - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static int bnxt_hwrm_queue_cos2bw_qcfg(struct bnxt *bp, struct ieee_ets *ets) @@ -236,7 +234,6 @@ static int bnxt_hwrm_queue_pfc_cfg(struct bnxt *bp, struct ieee_pfc *pfc) unsigned int tc_mask = 0, pri_mask = 0; u8 i, pri, lltc_count = 0; bool need_q_remap = false; - int rc; if (!my_ets) return -EINVAL; @@ -267,15 +264,11 @@ static int bnxt_hwrm_queue_pfc_cfg(struct bnxt *bp, struct ieee_pfc *pfc) } if (need_q_remap) - rc = bnxt_queue_remap(bp, tc_mask); + bnxt_queue_remap(bp, tc_mask); bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_PFCENABLE_CFG, -1, -1); req.flags = cpu_to_le32(pri_mask); - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - if (rc) - return rc; - - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static int bnxt_hwrm_queue_pfc_qcfg(struct bnxt *bp, struct ieee_pfc *pfc) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c index d3c93ccee86a..a812beb46325 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c @@ -150,7 +150,7 @@ void bnxt_dl_fw_reporters_create(struct bnxt *bp) health->fw_reset_reporter = devlink_health_reporter_create(bp->dl, &bnxt_dl_fw_reset_reporter_ops, - 0, true, bp); + 0, bp); if (IS_ERR(health->fw_reset_reporter)) { netdev_warn(bp->dev, "Failed to create FW fatal health reporter, rc = %ld\n", PTR_ERR(health->fw_reset_reporter)); @@ -166,7 +166,7 @@ err_recovery: health->fw_reporter = devlink_health_reporter_create(bp->dl, &bnxt_dl_fw_reporter_ops, - 0, false, bp); + 0, bp); if (IS_ERR(health->fw_reporter)) { netdev_warn(bp->dev, "Failed to create FW health reporter, rc = %ld\n", PTR_ERR(health->fw_reporter)); @@ -182,7 +182,7 @@ err_recovery: health->fw_fatal_reporter = devlink_health_reporter_create(bp->dl, &bnxt_dl_fw_fatal_reporter_ops, - 0, true, bp); + 0, bp); if (IS_ERR(health->fw_fatal_reporter)) { netdev_warn(bp->dev, "Failed to create FW fatal health reporter, rc = %ld\n", PTR_ERR(health->fw_fatal_reporter)); @@ -403,6 +403,14 @@ static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req, if (rc) return rc; + if (strlen(bp->board_partno)) { + rc = devlink_info_version_fixed_put(req, + DEVLINK_INFO_VERSION_GENERIC_BOARD_ID, + bp->board_partno); + if (rc) + return rc; + } + sprintf(buf, "%X", bp->chip_num); rc = devlink_info_version_fixed_put(req, DEVLINK_INFO_VERSION_GENERIC_ASIC_ID, buf); @@ -471,13 +479,19 @@ static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req, ver_resp->roce_fw_bld_8b, ver_resp->roce_fw_rsvd_8b); } rc = devlink_info_version_running_put(req, - DEVLINK_INFO_VERSION_GENERIC_FW_APP, fw_ver); + DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, fw_ver); + if (rc) + return rc; + + rc = devlink_info_version_running_put(req, + DEVLINK_INFO_VERSION_GENERIC_FW_MGMT_API, + bp->hwrm_ver_supp); if (rc) return rc; if (!(bp->flags & BNXT_FLAG_CHIP_P5)) { rc = devlink_info_version_running_put(req, - DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, mgmt_ver); + DEVLINK_INFO_VERSION_GENERIC_FW_NCSI, mgmt_ver); if (rc) return rc; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 3f8a1ded662a..34046a6286e8 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -1236,7 +1236,6 @@ static void bnxt_get_drvinfo(struct net_device *dev, struct bnxt *bp = netdev_priv(dev); strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); strlcpy(info->fw_version, bp->fw_ver_str, sizeof(info->fw_version)); strlcpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info)); info->n_stats = bnxt_get_num_stats(bp); @@ -2605,7 +2604,7 @@ static int bnxt_set_phys_id(struct net_device *dev, struct bnxt_led_cfg *led_cfg; u8 led_state; __le16 duration; - int i, rc; + int i; if (!bp->num_leds || BNXT_VF(bp)) return -EOPNOTSUPP; @@ -2631,8 +2630,7 @@ static int bnxt_set_phys_id(struct net_device *dev, led_cfg->led_blink_off = duration; led_cfg->led_group_id = bp->leds[i].led_group_id; } - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } static int bnxt_hwrm_selftest_irq(struct bnxt *bp, u16 cmpl_ring) @@ -3471,6 +3469,12 @@ void bnxt_ethtool_free(struct bnxt *bp) } const struct ethtool_ops bnxt_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USECS_IRQ | + ETHTOOL_COALESCE_MAX_FRAMES_IRQ | + ETHTOOL_COALESCE_STATS_BLOCK_USECS | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_link_ksettings = bnxt_get_link_ksettings, .set_link_ksettings = bnxt_set_link_ksettings, .get_pauseparam = bnxt_get_pauseparam, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c index 2aba1e02a8f4..6ea3df6da18c 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c @@ -138,7 +138,6 @@ static bool bnxt_is_trusted_vf(struct bnxt *bp, struct bnxt_vf_info *vf) static int bnxt_hwrm_set_trusted_vf(struct bnxt *bp, struct bnxt_vf_info *vf) { struct hwrm_func_cfg_input req = {0}; - int rc; if (!(bp->fw_cap & BNXT_FW_CAP_TRUSTED_VF)) return 0; @@ -149,8 +148,7 @@ static int bnxt_hwrm_set_trusted_vf(struct bnxt *bp, struct bnxt_vf_info *vf) req.flags = cpu_to_le32(FUNC_CFG_REQ_FLAGS_TRUSTED_VF_ENABLE); else req.flags = cpu_to_le32(FUNC_CFG_REQ_FLAGS_TRUSTED_VF_DISABLE); - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - return rc; + return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } int bnxt_set_vf_trust(struct net_device *dev, int vf_id, bool trusted) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c index 9bec256b0934..782ea0771221 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c @@ -279,7 +279,8 @@ bnxt_tc_parse_pedit(struct bnxt *bp, struct bnxt_tc_actions *actions, static int bnxt_tc_parse_actions(struct bnxt *bp, struct bnxt_tc_actions *actions, - struct flow_action *flow_action) + struct flow_action *flow_action, + struct netlink_ext_ack *extack) { /* Used to store the L2 rewrite mask for dmac (6 bytes) followed by * smac (6 bytes) if rewrite of both is specified, otherwise either @@ -299,6 +300,9 @@ static int bnxt_tc_parse_actions(struct bnxt *bp, return -EINVAL; } + if (!flow_action_basic_hw_stats_check(flow_action, extack)) + return -EOPNOTSUPP; + flow_action_for_each(i, act, flow_action) { switch (act->id) { case FLOW_ACTION_DROP: @@ -491,7 +495,8 @@ static int bnxt_tc_parse_flow(struct bnxt *bp, flow->tun_mask.tp_src = match.mask->src; } - return bnxt_tc_parse_actions(bp, &flow->actions, &rule->action); + return bnxt_tc_parse_actions(bp, &flow->actions, &rule->action, + tc_flow_cmd->common.extack); } static int bnxt_hwrm_cfa_flow_free(struct bnxt *bp, @@ -1634,7 +1639,7 @@ static int bnxt_tc_get_flow_stats(struct bnxt *bp, spin_unlock(&flow->stats_lock); flow_stats_update(&tc_flow_cmd->stats, stats.bytes, stats.packets, - lastused); + lastused, FLOW_ACTION_HW_STATS_DELAYED); return 0; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c index 6f2faf81c1ae..4b5c8fd76a51 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c @@ -219,7 +219,6 @@ static void bnxt_vf_rep_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); } static int bnxt_vf_rep_get_port_parent_id(struct net_device *dev, diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 1d678bee2cc9..d975338bf78d 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -7,6 +7,7 @@ #define pr_fmt(fmt) "bcmgenet: " fmt +#include <linux/acpi.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/sched.h> @@ -621,10 +622,6 @@ static int bcmgenet_set_coalesce(struct net_device *dev, * always generate an interrupt either after MBDONE packets have been * transmitted, or when the ring is empty. */ - if (ec->tx_coalesce_usecs || ec->tx_coalesce_usecs_high || - ec->tx_coalesce_usecs_irq || ec->tx_coalesce_usecs_low || - ec->use_adaptive_tx_coalesce) - return -EOPNOTSUPP; /* Program all TX queues with the same values, as there is no * ethtool knob to do coalescing on a per-queue basis @@ -814,7 +811,6 @@ static void bcmgenet_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, "bcmgenet", sizeof(info->driver)); - strlcpy(info->version, "v2.0", sizeof(info->version)); } static int bcmgenet_get_sset_count(struct net_device *dev, int string_set) @@ -1049,6 +1045,9 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) /* standard ethtool support functions. */ static const struct ethtool_ops bcmgenet_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .begin = bcmgenet_begin, .complete = bcmgenet_complete, .get_strings = bcmgenet_get_strings, @@ -2713,6 +2712,21 @@ static void bcmgenet_set_hw_addr(struct bcmgenet_priv *priv, bcmgenet_umac_writel(priv, (addr[4] << 8) | addr[5], UMAC_MAC1); } +static void bcmgenet_get_hw_addr(struct bcmgenet_priv *priv, + unsigned char *addr) +{ + u32 addr_tmp; + + addr_tmp = bcmgenet_umac_readl(priv, UMAC_MAC0); + addr[0] = addr_tmp >> 24; + addr[1] = (addr_tmp >> 16) & 0xff; + addr[2] = (addr_tmp >> 8) & 0xff; + addr[3] = addr_tmp & 0xff; + addr_tmp = bcmgenet_umac_readl(priv, UMAC_MAC1); + addr[4] = (addr_tmp >> 8) & 0xff; + addr[5] = addr_tmp & 0xff; +} + /* Returns a reusable dma control register value */ static u32 bcmgenet_dma_disable(struct bcmgenet_priv *priv) { @@ -3408,10 +3422,8 @@ static int bcmgenet_probe(struct platform_device *pdev) const struct bcmgenet_plat_data *pdata; struct bcmgenet_priv *priv; struct net_device *dev; - const void *macaddr; unsigned int i; int err = -EIO; - const char *phy_mode_str; /* Up to GENET_MAX_MQ_CNT + 1 TX queues and RX queues */ dev = alloc_etherdev_mqs(sizeof(*priv), GENET_MAX_MQ_CNT + 1, @@ -3440,11 +3452,6 @@ static int bcmgenet_probe(struct platform_device *pdev) } priv->wol_irq = platform_get_irq_optional(pdev, 2); - if (dn) - macaddr = of_get_mac_address(dn); - else - macaddr = pd->mac_address; - priv->base = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(priv->base)) { err = PTR_ERR(priv->base); @@ -3455,12 +3462,6 @@ static int bcmgenet_probe(struct platform_device *pdev) SET_NETDEV_DEV(dev, &pdev->dev); dev_set_drvdata(&pdev->dev, dev); - if (IS_ERR_OR_NULL(macaddr) || !is_valid_ether_addr(macaddr)) { - dev_warn(&pdev->dev, "using random Ethernet MAC\n"); - eth_hw_addr_random(dev); - } else { - ether_addr_copy(dev->dev_addr, macaddr); - } dev->watchdog_timeo = 2 * HZ; dev->ethtool_ops = &bcmgenet_ethtool_ops; dev->netdev_ops = &bcmgenet_netdev_ops; @@ -3489,8 +3490,9 @@ static int bcmgenet_probe(struct platform_device *pdev) priv->dev = dev; priv->pdev = pdev; - if (of_id) { - pdata = of_id->data; + + pdata = device_get_match_data(&pdev->dev); + if (pdata) { priv->version = pdata->version; priv->dma_max_burst_length = pdata->dma_max_burst_length; } else { @@ -3500,7 +3502,7 @@ static int bcmgenet_probe(struct platform_device *pdev) priv->clk = devm_clk_get(&priv->pdev->dev, "enet"); if (IS_ERR(priv->clk)) { - dev_warn(&priv->pdev->dev, "failed to get enet clock\n"); + dev_dbg(&priv->pdev->dev, "failed to get enet clock\n"); priv->clk = NULL; } @@ -3524,23 +3526,34 @@ static int bcmgenet_probe(struct platform_device *pdev) priv->clk_wol = devm_clk_get(&priv->pdev->dev, "enet-wol"); if (IS_ERR(priv->clk_wol)) { - dev_warn(&priv->pdev->dev, "failed to get enet-wol clock\n"); + dev_dbg(&priv->pdev->dev, "failed to get enet-wol clock\n"); priv->clk_wol = NULL; } priv->clk_eee = devm_clk_get(&priv->pdev->dev, "enet-eee"); if (IS_ERR(priv->clk_eee)) { - dev_warn(&priv->pdev->dev, "failed to get enet-eee clock\n"); + dev_dbg(&priv->pdev->dev, "failed to get enet-eee clock\n"); priv->clk_eee = NULL; } /* If this is an internal GPHY, power it on now, before UniMAC is * brought out of reset as absolutely no UniMAC activity is allowed */ - if (dn && !of_property_read_string(dn, "phy-mode", &phy_mode_str) && - !strcasecmp(phy_mode_str, "internal")) + if (device_get_phy_mode(&pdev->dev) == PHY_INTERFACE_MODE_INTERNAL) bcmgenet_power_up(priv, GENET_POWER_PASSIVE); + if ((pd) && (!IS_ERR_OR_NULL(pd->mac_address))) + ether_addr_copy(dev->dev_addr, pd->mac_address); + else + if (!device_get_mac_address(&pdev->dev, dev->dev_addr, ETH_ALEN)) + if (has_acpi_companion(&pdev->dev)) + bcmgenet_get_hw_addr(priv, dev->dev_addr); + + if (!is_valid_ether_addr(dev->dev_addr)) { + dev_warn(&pdev->dev, "using random Ethernet MAC\n"); + eth_hw_addr_random(dev); + } + reset_umac(priv); err = bcmgenet_mii_init(dev); @@ -3713,6 +3726,12 @@ static int bcmgenet_suspend(struct device *d) static SIMPLE_DEV_PM_OPS(bcmgenet_pm_ops, bcmgenet_suspend, bcmgenet_resume); +static const struct acpi_device_id genet_acpi_match[] = { + { "BCM6E4E", (kernel_ulong_t)&bcm2711_plat_data }, + { }, +}; +MODULE_DEVICE_TABLE(acpi, genet_acpi_match); + static struct platform_driver bcmgenet_driver = { .probe = bcmgenet_probe, .remove = bcmgenet_remove, @@ -3721,6 +3740,7 @@ static struct platform_driver bcmgenet_driver = { .name = "bcmgenet", .of_match_table = bcmgenet_match, .pm = &bcmgenet_pm_ops, + .acpi_match_table = ACPI_PTR(genet_acpi_match), }, }; module_platform_driver(bcmgenet_driver); diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index b5930f80039d..511d553a4d11 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -5,7 +5,7 @@ * Copyright (c) 2014-2017 Broadcom */ - +#include <linux/acpi.h> #include <linux/types.h> #include <linux/delay.h> #include <linux/wait.h> @@ -284,7 +284,8 @@ int bcmgenet_mii_config(struct net_device *dev, bool init) int bcmgenet_mii_probe(struct net_device *dev) { struct bcmgenet_priv *priv = netdev_priv(dev); - struct device_node *dn = priv->pdev->dev.of_node; + struct device *kdev = &priv->pdev->dev; + struct device_node *dn = kdev->of_node; struct phy_device *phydev; u32 phy_flags = 0; int ret; @@ -307,7 +308,27 @@ int bcmgenet_mii_probe(struct net_device *dev) return -ENODEV; } } else { - phydev = dev->phydev; + if (has_acpi_companion(kdev)) { + char mdio_bus_id[MII_BUS_ID_SIZE]; + struct mii_bus *unimacbus; + + snprintf(mdio_bus_id, MII_BUS_ID_SIZE, "%s-%d", + UNIMAC_MDIO_DRV_NAME, priv->pdev->id); + + unimacbus = mdio_find_bus(mdio_bus_id); + if (!unimacbus) { + pr_err("Unable to find mii\n"); + return -ENODEV; + } + phydev = phy_find_first(unimacbus); + put_device(&unimacbus->dev); + if (!phydev) { + pr_err("Unable to find PHY\n"); + return -ENODEV; + } + } else { + phydev = dev->phydev; + } phydev->dev_flags = phy_flags; ret = phy_connect_direct(dev, phydev, bcmgenet_mii_setup, @@ -428,9 +449,12 @@ static int bcmgenet_mii_register(struct bcmgenet_priv *priv) /* Retain this platform_device pointer for later cleanup */ priv->mii_pdev = ppdev; ppdev->dev.parent = &pdev->dev; - ppdev->dev.of_node = bcmgenet_mii_of_find_mdio(priv); - if (pdata) + if (dn) + ppdev->dev.of_node = bcmgenet_mii_of_find_mdio(priv); + else if (pdata) bcmgenet_mii_pdata_init(priv, &ppd); + else + ppd.phy_mask = ~0; ret = platform_device_add_resources(ppdev, &res, 1); if (ret) @@ -450,12 +474,33 @@ out: return ret; } +static int bcmgenet_phy_interface_init(struct bcmgenet_priv *priv) +{ + struct device *kdev = &priv->pdev->dev; + int phy_mode = device_get_phy_mode(kdev); + + if (phy_mode < 0) { + dev_err(kdev, "invalid PHY mode property\n"); + return phy_mode; + } + + priv->phy_interface = phy_mode; + + /* We need to specifically look up whether this PHY interface is + * internal or not *before* we even try to probe the PHY driver + * over MDIO as we may have shut down the internal PHY for power + * saving purposes. + */ + if (priv->phy_interface == PHY_INTERFACE_MODE_INTERNAL) + priv->internal_phy = true; + + return 0; +} + static int bcmgenet_mii_of_init(struct bcmgenet_priv *priv) { struct device_node *dn = priv->pdev->dev.of_node; - struct device *kdev = &priv->pdev->dev; struct phy_device *phydev; - phy_interface_t phy_mode; int ret; /* Fetch the PHY phandle */ @@ -473,23 +518,12 @@ static int bcmgenet_mii_of_init(struct bcmgenet_priv *priv) } /* Get the link mode */ - ret = of_get_phy_mode(dn, &phy_mode); - if (ret) { - dev_err(kdev, "invalid PHY mode property\n"); + ret = bcmgenet_phy_interface_init(priv); + if (ret) return ret; - } - - priv->phy_interface = phy_mode; - - /* We need to specifically look up whether this PHY interface is internal - * or not *before* we even try to probe the PHY driver over MDIO as we - * may have shut down the internal PHY for power saving purposes. - */ - if (priv->phy_interface == PHY_INTERFACE_MODE_INTERNAL) - priv->internal_phy = true; /* Make sure we initialize MoCA PHYs with a link down */ - if (phy_mode == PHY_INTERFACE_MODE_MOCA) { + if (priv->phy_interface == PHY_INTERFACE_MODE_MOCA) { phydev = of_phy_find_device(dn); if (phydev) { phydev->link = 0; @@ -554,10 +588,13 @@ static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv) static int bcmgenet_mii_bus_init(struct bcmgenet_priv *priv) { - struct device_node *dn = priv->pdev->dev.of_node; + struct device *kdev = &priv->pdev->dev; + struct device_node *dn = kdev->of_node; if (dn) return bcmgenet_mii_of_init(priv); + else if (has_acpi_companion(kdev)) + return bcmgenet_phy_interface_init(priv); else return bcmgenet_mii_pd_init(priv); } diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 88466255bf66..ff98a82b7bc4 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -96,11 +96,9 @@ static inline void _tg3_flag_clear(enum TG3_FLAGS flag, unsigned long *bits) _tg3_flag_clear(TG3_FLAG_##flag, (tp)->tg3_flags) #define DRV_MODULE_NAME "tg3" +/* DO NOT UPDATE TG3_*_NUM defines */ #define TG3_MAJ_NUM 3 #define TG3_MIN_NUM 137 -#define DRV_MODULE_VERSION \ - __stringify(TG3_MAJ_NUM) "." __stringify(TG3_MIN_NUM) -#define DRV_MODULE_RELDATE "May 11, 2014" #define RESET_KIND_SHUTDOWN 0 #define RESET_KIND_INIT 1 @@ -222,13 +220,9 @@ static inline void _tg3_flag_clear(enum TG3_FLAGS flag, unsigned long *bits) #define FIRMWARE_TG3TSO "tigon/tg3_tso.bin" #define FIRMWARE_TG3TSO5 "tigon/tg3_tso5.bin" -static char version[] = - DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")"; - MODULE_AUTHOR("David S. Miller (davem@redhat.com) and Jeff Garzik (jgarzik@pobox.com)"); MODULE_DESCRIPTION("Broadcom Tigon3 ethernet driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_MODULE_VERSION); MODULE_FIRMWARE(FIRMWARE_TG3); MODULE_FIRMWARE(FIRMWARE_TG3TSO); MODULE_FIRMWARE(FIRMWARE_TG3TSO5); @@ -12317,7 +12311,6 @@ static void tg3_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info struct tg3 *tp = netdev_priv(dev); strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); strlcpy(info->fw_version, tp->fw_ver, sizeof(info->fw_version)); strlcpy(info->bus_info, pci_name(tp->pdev), sizeof(info->bus_info)); } @@ -14160,6 +14153,11 @@ static int tg3_get_eee(struct net_device *dev, struct ethtool_eee *edata) } static const struct ethtool_ops tg3_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USECS_IRQ | + ETHTOOL_COALESCE_MAX_FRAMES_IRQ | + ETHTOOL_COALESCE_STATS_BLOCK_USECS, .get_drvinfo = tg3_get_drvinfo, .get_regs_len = tg3_get_regs_len, .get_regs = tg3_get_regs, @@ -17625,8 +17623,6 @@ static int tg3_init_one(struct pci_dev *pdev, u64 dma_mask, persist_dma_mask; netdev_features_t features = 0; - printk_once(KERN_INFO "%s\n", version); - err = pci_enable_device(pdev); if (err) { dev_err(&pdev->dev, "Cannot enable PCI device, aborting\n"); diff --git a/drivers/net/ethernet/brocade/bna/bnad.c b/drivers/net/ethernet/brocade/bna/bnad.c index 01a50a4b2113..cc80bbbefe87 100644 --- a/drivers/net/ethernet/brocade/bna/bnad.c +++ b/drivers/net/ethernet/brocade/bna/bnad.c @@ -2504,12 +2504,7 @@ bnad_tso_prepare(struct bnad *bnad, struct sk_buff *skb) IPPROTO_TCP, 0); BNAD_UPDATE_CTR(bnad, tso4); } else { - struct ipv6hdr *ipv6h = ipv6_hdr(skb); - - ipv6h->payload_len = 0; - tcp_hdr(skb)->check = - ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr, 0, - IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); BNAD_UPDATE_CTR(bnad, tso6); } @@ -3847,9 +3842,6 @@ bnad_module_init(void) { int err; - pr_info("bna: QLogic BR-series 10G Ethernet driver - version: %s\n", - BNAD_VERSION); - bfa_nw_ioc_auto_recover(bnad_ioc_auto_recover); err = pci_register_driver(&bnad_pci_driver); @@ -3874,6 +3866,5 @@ module_exit(bnad_module_exit); MODULE_AUTHOR("Brocade"); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("QLogic BR-series 10G PCIe Ethernet driver"); -MODULE_VERSION(BNAD_VERSION); MODULE_FIRMWARE(CNA_FW_FILE_CT); MODULE_FIRMWARE(CNA_FW_FILE_CT2); diff --git a/drivers/net/ethernet/brocade/bna/bnad.h b/drivers/net/ethernet/brocade/bna/bnad.h index 492a02d54f14..627a93ce38ab 100644 --- a/drivers/net/ethernet/brocade/bna/bnad.h +++ b/drivers/net/ethernet/brocade/bna/bnad.h @@ -64,8 +64,6 @@ struct bnad_rx_ctrl { #define BNAD_NAME "bna" #define BNAD_NAME_LEN 64 -#define BNAD_VERSION "3.2.25.1" - #define BNAD_MAILBOX_MSIX_INDEX 0 #define BNAD_MAILBOX_MSIX_VECTORS 1 #define BNAD_INTX_TX_IB_BITMASK 0x1 @@ -253,7 +251,7 @@ struct bnad_rx_unmap_q { int alloc_order; u32 map_size; enum bnad_rxbuf_type type; - struct bnad_rx_unmap unmap[0] ____cacheline_aligned; + struct bnad_rx_unmap unmap[] ____cacheline_aligned; }; #define BNAD_PCI_DEV_IS_CAT2(_bnad) \ diff --git a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c index b764c9ff9ad1..588c4804d10a 100644 --- a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c +++ b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c @@ -284,7 +284,6 @@ bnad_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) unsigned long flags; strlcpy(drvinfo->driver, BNAD_NAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, BNAD_VERSION, sizeof(drvinfo->version)); ioc_attr = kzalloc(sizeof(*ioc_attr), GFP_KERNEL); if (ioc_attr) { @@ -1116,6 +1115,9 @@ out: } static const struct ethtool_ops bnad_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_TX_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_drvinfo = bnad_get_drvinfo, .get_wol = bnad_get_wol, .get_link = ethtool_op_get_link, diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index a3f0f27fc79a..ab827fb4b6b9 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -1200,7 +1200,6 @@ struct macb { unsigned int dma_burst_length; phy_interface_t phy_interface; - int speed; /* AT91RM9200 transmit */ struct sk_buff *skb; /* holds skb until xmit interrupt completes */ diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 2c28da1737fe..a0e8c5bbabc0 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -571,37 +571,20 @@ static void macb_mac_config(struct phylink_config *config, unsigned int mode, old_ctrl = ctrl = macb_or_gem_readl(bp, NCFGR); - /* Clear all the bits we might set later */ - ctrl &= ~(MACB_BIT(SPD) | MACB_BIT(FD) | MACB_BIT(PAE)); - if (bp->caps & MACB_CAPS_MACB_IS_EMAC) { if (state->interface == PHY_INTERFACE_MODE_RMII) ctrl |= MACB_BIT(RM9200_RMII); } else { - ctrl &= ~(GEM_BIT(GBE) | GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL)); - - /* We do not support MLO_PAUSE_RX yet */ - if (state->pause & MLO_PAUSE_TX) - ctrl |= MACB_BIT(PAE); + ctrl &= ~(GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL)); if (state->interface == PHY_INTERFACE_MODE_SGMII) ctrl |= GEM_BIT(SGMIIEN) | GEM_BIT(PCSSEL); } - if (state->speed == SPEED_1000) - ctrl |= GEM_BIT(GBE); - else if (state->speed == SPEED_100) - ctrl |= MACB_BIT(SPD); - - if (state->duplex) - ctrl |= MACB_BIT(FD); - /* Apply the new configuration, if any */ if (old_ctrl ^ ctrl) macb_or_gem_writel(bp, NCFGR, ctrl); - bp->speed = state->speed; - spin_unlock_irqrestore(&bp->lock, flags); } @@ -626,16 +609,42 @@ static void macb_mac_link_down(struct phylink_config *config, unsigned int mode, netif_tx_stop_all_queues(ndev); } -static void macb_mac_link_up(struct phylink_config *config, unsigned int mode, - phy_interface_t interface, struct phy_device *phy) +static void macb_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause) { struct net_device *ndev = to_net_dev(config->dev); struct macb *bp = netdev_priv(ndev); struct macb_queue *queue; + unsigned long flags; unsigned int q; + u32 ctrl; + + spin_lock_irqsave(&bp->lock, flags); + + ctrl = macb_or_gem_readl(bp, NCFGR); + + ctrl &= ~(MACB_BIT(SPD) | MACB_BIT(FD)); + + if (speed == SPEED_100) + ctrl |= MACB_BIT(SPD); + + if (duplex) + ctrl |= MACB_BIT(FD); if (!(bp->caps & MACB_CAPS_MACB_IS_EMAC)) { - macb_set_tx_clk(bp->tx_clk, bp->speed, ndev); + ctrl &= ~(GEM_BIT(GBE) | MACB_BIT(PAE)); + + if (speed == SPEED_1000) + ctrl |= GEM_BIT(GBE); + + /* We do not support MLO_PAUSE_RX yet */ + if (tx_pause) + ctrl |= MACB_BIT(PAE); + + macb_set_tx_clk(bp->tx_clk, speed, ndev); /* Initialize rings & buffers as clearing MACB_BIT(TE) in link down * cleared the pipeline and control registers. @@ -648,6 +657,10 @@ static void macb_mac_link_up(struct phylink_config *config, unsigned int mode, bp->rx_intr_mask | MACB_TX_INT_FLAGS | MACB_BIT(HRESP)); } + macb_or_gem_writel(bp, NCFGR, ctrl); + + spin_unlock_irqrestore(&bp->lock, flags); + /* Enable Rx and Tx */ macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(RE) | MACB_BIT(TE)); @@ -724,6 +737,9 @@ static int macb_mdiobus_register(struct macb *bp) { struct device_node *child, *np = bp->pdev->dev.of_node; + if (of_phy_is_fixed_link(np)) + return mdiobus_register(bp->mii_bus); + /* Only create the PHY from the device tree if at least one PHY is * described. Otherwise scan the entire MDIO bus. We do this to support * old device tree that did not follow the best practices and did not @@ -4429,8 +4445,6 @@ static int macb_probe(struct platform_device *pdev) else bp->phy_interface = interface; - bp->speed = SPEED_UNKNOWN; - /* IP specific init */ err = init(pdev); if (err) diff --git a/drivers/net/ethernet/cavium/common/cavium_ptp.c b/drivers/net/ethernet/cavium/common/cavium_ptp.c index b821c9e1604c..81ff9ac73f9a 100644 --- a/drivers/net/ethernet/cavium/common/cavium_ptp.c +++ b/drivers/net/ethernet/cavium/common/cavium_ptp.c @@ -13,6 +13,9 @@ #define DRV_NAME "cavium_ptp" #define PCI_DEVICE_ID_CAVIUM_PTP 0xA00C +#define PCI_SUBSYS_DEVID_88XX_PTP 0xA10C +#define PCI_SUBSYS_DEVID_81XX_PTP 0XA20C +#define PCI_SUBSYS_DEVID_83XX_PTP 0xA30C #define PCI_DEVICE_ID_CAVIUM_RST 0xA00E #define PCI_PTP_BAR_NO 0 @@ -321,7 +324,12 @@ static void cavium_ptp_remove(struct pci_dev *pdev) } static const struct pci_device_id cavium_ptp_id_table[] = { - { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_CAVIUM_PTP) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_CAVIUM_PTP, + PCI_VENDOR_ID_CAVIUM, PCI_SUBSYS_DEVID_88XX_PTP) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_CAVIUM_PTP, + PCI_VENDOR_ID_CAVIUM, PCI_SUBSYS_DEVID_81XX_PTP) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_CAVIUM_PTP, + PCI_VENDOR_ID_CAVIUM, PCI_SUBSYS_DEVID_83XX_PTP) }, { 0, } }; diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index abe5d0dac851..16eebfc52109 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -442,7 +442,6 @@ lio_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) memset(drvinfo, 0, sizeof(struct ethtool_drvinfo)); strcpy(drvinfo->driver, "liquidio"); - strcpy(drvinfo->version, LIQUIDIO_VERSION); strncpy(drvinfo->fw_version, oct->fw_info.liquidio_firmware_version, ETHTOOL_FWVERS_LEN); strncpy(drvinfo->bus_info, pci_name(oct->pci_dev), 32); @@ -459,7 +458,6 @@ lio_get_vf_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) memset(drvinfo, 0, sizeof(struct ethtool_drvinfo)); strcpy(drvinfo->driver, "liquidio_vf"); - strcpy(drvinfo->version, LIQUIDIO_VERSION); strncpy(drvinfo->fw_version, oct->fw_info.liquidio_firmware_version, ETHTOOL_FWVERS_LEN); strncpy(drvinfo->bus_info, pci_name(oct->pci_dev), 32); @@ -3099,7 +3097,17 @@ static int lio_set_fecparam(struct net_device *netdev, return 0; } +#define LIO_ETHTOOL_COALESCE (ETHTOOL_COALESCE_RX_USECS | \ + ETHTOOL_COALESCE_MAX_FRAMES | \ + ETHTOOL_COALESCE_USE_ADAPTIVE | \ + ETHTOOL_COALESCE_RX_MAX_FRAMES_LOW | \ + ETHTOOL_COALESCE_TX_MAX_FRAMES_LOW | \ + ETHTOOL_COALESCE_RX_MAX_FRAMES_HIGH | \ + ETHTOOL_COALESCE_TX_MAX_FRAMES_HIGH | \ + ETHTOOL_COALESCE_PKT_RATE_RX_USECS) + static const struct ethtool_ops lio_ethtool_ops = { + .supported_coalesce_params = LIO_ETHTOOL_COALESCE, .get_link_ksettings = lio_get_link_ksettings, .set_link_ksettings = lio_set_link_ksettings, .get_fecparam = lio_get_fecparam, @@ -3130,6 +3138,7 @@ static const struct ethtool_ops lio_ethtool_ops = { }; static const struct ethtool_ops lio_vf_ethtool_ops = { + .supported_coalesce_params = LIO_ETHTOOL_COALESCE, .get_link_ksettings = lio_get_link_ksettings, .get_link = ethtool_op_get_link, .get_drvinfo = lio_get_vf_drvinfo, diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index eab05b5534ea..66d31c018c7e 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -39,7 +39,6 @@ MODULE_AUTHOR("Cavium Networks, <support@cavium.com>"); MODULE_DESCRIPTION("Cavium LiquidIO Intelligent Server Adapter Driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(LIQUIDIO_VERSION); MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_210SV_NAME "_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX); MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_210NV_NAME @@ -1376,7 +1375,6 @@ static int octeon_chip_specific_setup(struct octeon_device *oct) { u32 dev_id, rev_id; int ret = 1; - char *s; pci_read_config_dword(oct->pci_dev, 0, &dev_id); pci_read_config_dword(oct->pci_dev, 8, &rev_id); @@ -1386,13 +1384,11 @@ static int octeon_chip_specific_setup(struct octeon_device *oct) case OCTEON_CN68XX_PCIID: oct->chip_id = OCTEON_CN68XX; ret = lio_setup_cn68xx_octeon_device(oct); - s = "CN68XX"; break; case OCTEON_CN66XX_PCIID: oct->chip_id = OCTEON_CN66XX; ret = lio_setup_cn66xx_octeon_device(oct); - s = "CN66XX"; break; case OCTEON_CN23XX_PCIID_PF: @@ -1405,22 +1401,13 @@ static int octeon_chip_specific_setup(struct octeon_device *oct) pci_sriov_set_totalvfs(oct->pci_dev, oct->sriov_info.max_vfs); #endif - s = "CN23XX"; break; default: - s = "?"; dev_err(&oct->pci_dev->dev, "Unknown device found (dev_id: %x)\n", dev_id); } - if (!ret) - dev_info(&oct->pci_dev->dev, "%s PASS%d.%d %s Version: %s\n", s, - OCTEON_MAJOR_REV(oct), - OCTEON_MINOR_REV(oct), - octeon_get_conf(oct)->card_name, - LIQUIDIO_VERSION); - return ret; } diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c index 7a77544a54f5..bbd9bfa4a989 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c @@ -32,7 +32,6 @@ MODULE_AUTHOR("Cavium Networks, <support@cavium.com>"); MODULE_DESCRIPTION("Cavium LiquidIO Intelligent Server Adapter Virtual Function Driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(LIQUIDIO_VERSION); static int debug = -1; module_param(debug, int, 0644); @@ -2352,8 +2351,8 @@ static int octeon_device_init(struct octeon_device *oct) } atomic_set(&oct->status, OCT_DEV_MSIX_ALLOC_VECTOR_DONE); - dev_info(&oct->pci_dev->dev, "OCTEON_CN23XX VF Version: %s, %d ioqs\n", - LIQUIDIO_VERSION, oct->sriov_info.rings_per_vf); + dev_info(&oct->pci_dev->dev, "OCTEON_CN23XX VF: %d ioqs\n", + oct->sriov_info.rings_per_vf); /* Setup the interrupt handler and record the INT SUM register address*/ if (octeon_setup_interrupt(oct, oct->sriov_info.rings_per_vf)) diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index a5e0e9f17959..4da90757cd3f 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -25,17 +25,11 @@ #include "octeon_config.h" -#define LIQUIDIO_PACKAGE "" #define LIQUIDIO_BASE_MAJOR_VERSION 1 #define LIQUIDIO_BASE_MINOR_VERSION 7 #define LIQUIDIO_BASE_MICRO_VERSION 2 #define LIQUIDIO_BASE_VERSION __stringify(LIQUIDIO_BASE_MAJOR_VERSION) "." \ __stringify(LIQUIDIO_BASE_MINOR_VERSION) -#define LIQUIDIO_MICRO_VERSION "." __stringify(LIQUIDIO_BASE_MICRO_VERSION) -#define LIQUIDIO_VERSION LIQUIDIO_PACKAGE \ - __stringify(LIQUIDIO_BASE_MAJOR_VERSION) "." \ - __stringify(LIQUIDIO_BASE_MINOR_VERSION) \ - "." __stringify(LIQUIDIO_BASE_MICRO_VERSION) struct lio_version { u16 major; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c index dfc77507b159..0d2831d10f65 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c @@ -127,7 +127,7 @@ struct octeon_pci_console_desc { u32 pad; /* must be 64 bit aligned here... */ /* Array of addresses of octeon_pci_console structures */ - u64 console_addr_array[0]; + u64 console_addr_array[]; /* Implicit storage for console_addr_array */ }; @@ -840,17 +840,11 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data, return -EINVAL; } - if (strncmp(LIQUIDIO_PACKAGE, h->version, strlen(LIQUIDIO_PACKAGE))) { - dev_err(&oct->pci_dev->dev, "Unmatched firmware package type. Expected %s, got %s.\n", - LIQUIDIO_PACKAGE, h->version); - return -EINVAL; - } - - if (memcmp(LIQUIDIO_BASE_VERSION, h->version + strlen(LIQUIDIO_PACKAGE), + if (memcmp(LIQUIDIO_BASE_VERSION, h->version, strlen(LIQUIDIO_BASE_VERSION))) { dev_err(&oct->pci_dev->dev, "Unmatched firmware version. Expected %s.x, got %s.\n", LIQUIDIO_BASE_VERSION, - h->version + strlen(LIQUIDIO_PACKAGE)); + h->version); return -EINVAL; } diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c index e9575887a4f8..9d868403d86c 100644 --- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c +++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c @@ -28,7 +28,6 @@ #include <asm/octeon/cvmx-agl-defs.h> #define DRV_NAME "octeon_mgmt" -#define DRV_VERSION "2.0" #define DRV_DESCRIPTION \ "Cavium Networks Octeon MII (management) port Network Driver" @@ -1340,9 +1339,6 @@ static void octeon_mgmt_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); - strlcpy(info->fw_version, "N/A", sizeof(info->fw_version)); - strlcpy(info->bus_info, "N/A", sizeof(info->bus_info)); } static int octeon_mgmt_nway_reset(struct net_device *dev) @@ -1517,7 +1513,6 @@ static int octeon_mgmt_probe(struct platform_device *pdev) if (result) goto err; - dev_info(&pdev->dev, "Version " DRV_VERSION "\n"); return 0; err: @@ -1574,4 +1569,3 @@ module_exit(octeon_mgmt_mod_exit); MODULE_DESCRIPTION(DRV_DESCRIPTION); MODULE_AUTHOR("David Daney"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c index 5e0b16bb95a0..83dabcffc789 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c @@ -16,7 +16,6 @@ #include "../common/cavium_ptp.h" #define DRV_NAME "nicvf" -#define DRV_VERSION "1.0" struct nicvf_stat { char name[ETH_GSTRING_LEN]; @@ -192,7 +191,6 @@ static void nicvf_get_drvinfo(struct net_device *netdev, struct nicvf *nic = netdev_priv(netdev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(nic->pdev), sizeof(info->bus_info)); } diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index 016957285f99..b4b33368698f 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -126,8 +126,7 @@ static void nicvf_write_to_mbx(struct nicvf *nic, union nic_mbx *mbx) int nicvf_send_msg_to_pf(struct nicvf *nic, union nic_mbx *mbx) { - int timeout = NIC_MBOX_MSG_TIMEOUT; - int sleep = 10; + unsigned long timeout; int ret = 0; mutex_lock(&nic->rx_mode_mtx); @@ -137,6 +136,7 @@ int nicvf_send_msg_to_pf(struct nicvf *nic, union nic_mbx *mbx) nicvf_write_to_mbx(nic, mbx); + timeout = jiffies + msecs_to_jiffies(NIC_MBOX_MSG_TIMEOUT); /* Wait for previous message to be acked, timeout 2sec */ while (!nic->pf_acked) { if (nic->pf_nacked) { @@ -146,11 +146,10 @@ int nicvf_send_msg_to_pf(struct nicvf *nic, union nic_mbx *mbx) ret = -EINVAL; break; } - msleep(sleep); + usleep_range(8000, 10000); if (nic->pf_acked) break; - timeout -= sleep; - if (!timeout) { + if (time_after(jiffies, timeout)) { netdev_err(nic->netdev, "PF didn't ACK to mbox msg 0x%02x from VF%d\n", (mbx->msg.msg & 0xFF), nic->vf_id); diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index 4ab57d33a87e..069e7413f1ef 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -1179,13 +1179,12 @@ void nicvf_sq_disable(struct nicvf *nic, int qidx) void nicvf_sq_free_used_descs(struct net_device *netdev, struct snd_queue *sq, int qidx) { - u64 head, tail; + u64 head; struct sk_buff *skb; struct nicvf *nic = netdev_priv(netdev); struct sq_hdr_subdesc *hdr; head = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_HEAD, qidx) >> 4; - tail = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_TAIL, qidx) >> 4; while (sq->head != head) { hdr = (struct sq_hdr_subdesc *)GET_SQ_DESC(sq, sq->head); if (hdr->subdesc_type != SQ_DESC_TYPE_HEADER) { diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h index bc2427c49b89..2460451fc48f 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h @@ -100,8 +100,8 @@ * RED accepts pkt if unused CQE < 2304 & >= 2560 * DROPs pkts if unused CQE < 2304 */ -#define RQ_PASS_CQ_LVL 192ULL -#define RQ_DROP_CQ_LVL 184ULL +#define RQ_PASS_CQ_LVL 224ULL +#define RQ_DROP_CQ_LVL 216ULL /* RED and Backpressure levels of RBDR for pkt reception * For RBDR, level is a measure of fullness i.e 0x0 means empty diff --git a/drivers/net/ethernet/chelsio/cxgb/common.h b/drivers/net/ethernet/chelsio/cxgb/common.h index 94b9482f14a5..6475060649e9 100644 --- a/drivers/net/ethernet/chelsio/cxgb/common.h +++ b/drivers/net/ethernet/chelsio/cxgb/common.h @@ -55,7 +55,6 @@ #define DRV_DESCRIPTION "Chelsio 10Gb Ethernet Driver" #define DRV_NAME "cxgb" -#define DRV_VERSION "2.2" #define CH_DEVICE(devid, ssid, idx) \ { PCI_VENDOR_ID_CHELSIO, devid, PCI_ANY_ID, ssid, 0, 0, idx } diff --git a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c index 0ccdde366ae1..99736796e1a0 100644 --- a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c +++ b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c @@ -429,7 +429,6 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) struct adapter *adapter = dev->ml_priv; strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(adapter->pdev), sizeof(info->bus_info)); } @@ -794,6 +793,9 @@ static int get_eeprom(struct net_device *dev, struct ethtool_eeprom *e, } static const struct ethtool_ops t1_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX | + ETHTOOL_COALESCE_RATE_SAMPLE_INTERVAL, .get_drvinfo = get_drvinfo, .get_msglevel = get_msglevel, .set_msglevel = set_msglevel, @@ -984,8 +986,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) struct adapter *adapter = NULL; struct port_info *pi; - pr_info_once("%s - version %s\n", DRV_DESCRIPTION, DRV_VERSION); - err = pci_enable_device(pdev); if (err) return err; diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_ioctl.h b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_ioctl.h index b19e4376ba76..401827b82aa1 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_ioctl.h +++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_ioctl.h @@ -79,7 +79,7 @@ struct ch_mem_range { uint32_t addr; uint32_t len; uint32_t version; - uint8_t buf[0]; + uint8_t buf[]; }; struct ch_qset_params { diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c index 883cfa9c4b6d..42c6e9379882 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c +++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c @@ -105,7 +105,6 @@ static const struct pci_device_id cxgb3_pci_tbl[] = { MODULE_DESCRIPTION(DRV_DESC); MODULE_AUTHOR("Chelsio Communications"); MODULE_LICENSE("Dual BSD/GPL"); -MODULE_VERSION(DRV_VERSION); MODULE_DEVICE_TABLE(pci, cxgb3_pci_tbl); static int dflt_msg_enable = DFLT_MSG_ENABLE; @@ -1629,7 +1628,6 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) spin_unlock(&adapter->stats_lock); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(adapter->pdev), sizeof(info->bus_info)); if (fw_vers) @@ -2106,6 +2104,7 @@ static void get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) } static const struct ethtool_ops cxgb_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = get_drvinfo, .get_msglevel = get_msglevel, .set_msglevel = set_msglevel, @@ -3210,8 +3209,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) struct adapter *adapter = NULL; struct port_info *pi; - pr_info_once("%s - version %s\n", DRV_DESC, DRV_VERSION); - if (!cxgb3_wq) { cxgb3_wq = create_singlethread_workqueue(DRV_NAME); if (!cxgb3_wq) { diff --git a/drivers/net/ethernet/chelsio/cxgb3/t3_cpl.h b/drivers/net/ethernet/chelsio/cxgb3/t3_cpl.h index 852c399a8b0a..68bb5f39f3f1 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/t3_cpl.h +++ b/drivers/net/ethernet/chelsio/cxgb3/t3_cpl.h @@ -1448,7 +1448,7 @@ struct cpl_rdma_terminate { #endif __be32 msn; __be32 mo; - __u8 data[0]; + __u8 data[]; }; /* cpl_rdma_terminate.tid_len fields */ diff --git a/drivers/net/ethernet/chelsio/cxgb3/version.h b/drivers/net/ethernet/chelsio/cxgb3/version.h index 165bfb91487a..b4b2547efc86 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/version.h +++ b/drivers/net/ethernet/chelsio/cxgb3/version.h @@ -34,8 +34,6 @@ #define __CHELSIO_VERSION_H #define DRV_DESC "Chelsio T3 Network Driver" #define DRV_NAME "cxgb3" -/* Driver version */ -#define DRV_VERSION "1.1.5-ko" /* Firmware version */ #define FW_VERSION_MAJOR 7 diff --git a/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.h b/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.h index a0e0ae19649f..290c1058069a 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.h +++ b/drivers/net/ethernet/chelsio/cxgb4/clip_tbl.h @@ -29,7 +29,7 @@ struct clip_tbl { atomic_t nfree; struct list_head ce_free_head; void *cl_list; - struct list_head hash_list[0]; + struct list_head hash_list[]; }; enum { diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h b/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h index f5be3ee1bdb4..dcab94cc2dee 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h @@ -82,7 +82,7 @@ struct cudbg_ulprx_la { struct cudbg_tp_la { u32 size; u32 mode; - u8 data[0]; + u8 data[]; }; static const char * const cudbg_region[] = { @@ -134,7 +134,7 @@ struct cudbg_meminfo { struct cudbg_cim_pif_la { int size; - u8 data[0]; + u8 data[]; }; struct cudbg_clk_info { @@ -339,13 +339,13 @@ struct cudbg_qdesc_entry { u32 qid; u32 desc_size; u32 num_desc; - u8 data[0]; /* Must be last */ + u8 data[]; /* Must be last */ }; struct cudbg_qdesc_info { u32 qdesc_entry_size; u32 num_queues; - u8 data[0]; /* Must be last */ + u8 data[]; /* Must be last */ }; #define IREG_NUM_ELEM 4 diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index 8b7d156f79d3..e46a14f44a6f 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -640,6 +640,7 @@ enum { /* adapter flags */ enum { ULP_CRYPTO_LOOKASIDE = 1 << 0, ULP_CRYPTO_IPSEC_INLINE = 1 << 1, + ULP_CRYPTO_KTLS_INLINE = 1 << 3, }; struct rx_sw_desc; @@ -1485,9 +1486,8 @@ static inline unsigned int qtimer_val(const struct adapter *adap, return idx < SGE_NTIMERS ? adap->sge.timer_val[idx] : 0; } -/* driver version & name used for ethtool_drvinfo */ +/* driver name used for ethtool_drvinfo */ extern char cxgb4_driver_name[]; -extern const char cxgb4_driver_version[]; void t4_os_portmod_changed(struct adapter *adap, int port_id); void t4_os_link_changed(struct adapter *adap, int port_id, int link_stat); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c index de30d61af065..ebed99f3d4cf 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c @@ -984,7 +984,7 @@ static const char * const devlog_facility_strings[] = { struct devlog_info { unsigned int nentries; /* number of entries in log[] */ unsigned int first; /* first [temporal] entry in log[] */ - struct fw_devlog_e log[0]; /* Firmware Device Log */ + struct fw_devlog_e log[]; /* Firmware Device Log */ }; /* Dump a Firmaware Device Log entry. @@ -3409,6 +3409,41 @@ static int chcr_stats_show(struct seq_file *seq, void *v) atomic_read(&adap->chcr_stats.tls_pdu_rx)); seq_printf(seq, "TLS Keys (DDR) Count: %10u\n", atomic_read(&adap->chcr_stats.tls_key)); +#ifdef CONFIG_CHELSIO_TLS_DEVICE + seq_puts(seq, "\nChelsio KTLS Crypto Accelerator Stats\n"); + seq_printf(seq, "Tx HW offload contexts added: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_ctx)); + seq_printf(seq, "Tx connection created: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_connection_open)); + seq_printf(seq, "Tx connection failed: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_connection_fail)); + seq_printf(seq, "Tx connection closed: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_connection_close)); + seq_printf(seq, "Packets passed for encryption : %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_encrypted_packets)); + seq_printf(seq, "Bytes passed for encryption : %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_encrypted_bytes)); + seq_printf(seq, "Tx records send: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_send_records)); + seq_printf(seq, "Tx partial start of records: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_start_pkts)); + seq_printf(seq, "Tx partial middle of records: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_middle_pkts)); + seq_printf(seq, "Tx partial end of record: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_end_pkts)); + seq_printf(seq, "Tx complete records: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_complete_pkts)); + seq_printf(seq, "TX trim pkts : %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_trimmed_pkts)); + seq_printf(seq, "Tx out of order packets: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_ooo)); + seq_printf(seq, "Tx drop pkts before HW offload: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_skip_no_sync_data)); + seq_printf(seq, "Tx drop not synced packets: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_drop_no_sync_data)); + seq_printf(seq, "Tx drop bypass req: %20llu\n", + atomic64_read(&adap->chcr_stats.ktls_tx_drop_bypass_req)); +#endif return 0; } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.h index ba95e13d52da..1471cf0deb58 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.h @@ -49,7 +49,7 @@ struct seq_tab { unsigned int rows; /* # of entries */ unsigned char width; /* size in bytes of each entry */ unsigned char skip_first; /* whether the first line is a header */ - char data[0]; /* the table data */ + char data[]; /* the table data */ }; static inline unsigned int hex2val(char c) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c index c837382ee522..9fd496732b2c 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c @@ -106,6 +106,15 @@ static char adapter_stats_strings[][ETH_GSTRING_LEN] = { "db_empty ", "write_coal_success ", "write_coal_fail ", +#ifdef CONFIG_CHELSIO_TLS_DEVICE + "tx_tls_encrypted_packets", + "tx_tls_encrypted_bytes ", + "tx_tls_ctx ", + "tx_tls_ooo ", + "tx_tls_skip_no_sync_data", + "tx_tls_drop_no_sync_data", + "tx_tls_drop_bypass_req ", +#endif }; static char loopback_stats_strings[][ETH_GSTRING_LEN] = { @@ -170,15 +179,11 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) u32 exprom_vers; strlcpy(info->driver, cxgb4_driver_name, sizeof(info->driver)); - strlcpy(info->version, cxgb4_driver_version, - sizeof(info->version)); strlcpy(info->bus_info, pci_name(adapter->pdev), sizeof(info->bus_info)); info->regdump_len = get_regs_len(dev); - if (!adapter->params.fw_vers) - strcpy(info->fw_version, "N/A"); - else + if (adapter->params.fw_vers) snprintf(info->fw_version, sizeof(info->fw_version), "%u.%u.%u.%u, TP %u.%u.%u.%u", FW_HDR_FW_VER_MAJOR_G(adapter->params.fw_vers), @@ -236,6 +241,15 @@ struct adapter_stats { u64 db_empty; u64 wc_success; u64 wc_fail; +#ifdef CONFIG_CHELSIO_TLS_DEVICE + u64 tx_tls_encrypted_packets; + u64 tx_tls_encrypted_bytes; + u64 tx_tls_ctx; + u64 tx_tls_ooo; + u64 tx_tls_skip_no_sync_data; + u64 tx_tls_drop_no_sync_data; + u64 tx_tls_drop_bypass_req; +#endif }; static void collect_sge_port_stats(const struct adapter *adap, @@ -1580,6 +1594,10 @@ static int cxgb4_set_priv_flags(struct net_device *netdev, u32 flags) } static const struct ethtool_ops cxgb_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_RX_MAX_FRAMES | + ETHTOOL_COALESCE_TX_USECS_IRQ | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_link_ksettings = get_link_ksettings, .set_link_ksettings = set_link_ksettings, .get_fecparam = get_fecparam, diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c index fc05248984fc..796555255207 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c @@ -438,13 +438,118 @@ int cxgb4_get_filter_counters(struct net_device *dev, unsigned int fidx, return get_filter_count(adapter, fidx, hitcnt, bytecnt, hash); } -int cxgb4_get_free_ftid(struct net_device *dev, int family) +static bool cxgb4_filter_prio_in_range(struct tid_info *t, u32 idx, u8 nslots, + u32 prio) +{ + struct filter_entry *prev_tab, *next_tab, *prev_fe, *next_fe; + u32 prev_ftid, next_ftid; + + /* Only insert the rule if both of the following conditions + * are met: + * 1. The immediate previous rule has priority <= @prio. + * 2. The immediate next rule has priority >= @prio. + */ + + /* High Priority (HPFILTER) region always has higher priority + * than normal FILTER region. So, all rules in HPFILTER region + * must have prio value <= rules in normal FILTER region. + */ + if (idx < t->nhpftids) { + /* Don't insert if there's a rule already present at @idx + * in HPFILTER region. + */ + if (test_bit(idx, t->hpftid_bmap)) + return false; + + next_tab = t->hpftid_tab; + next_ftid = find_next_bit(t->hpftid_bmap, t->nhpftids, idx); + if (next_ftid >= t->nhpftids) { + /* No next entry found in HPFILTER region. + * See if there's any next entry in normal + * FILTER region. + */ + next_ftid = find_first_bit(t->ftid_bmap, t->nftids); + if (next_ftid >= t->nftids) + next_ftid = idx; + else + next_tab = t->ftid_tab; + } + + /* Search for the closest previous filter entry in HPFILTER + * region. No need to search in normal FILTER region because + * there can never be any entry in normal FILTER region whose + * prio value is < last entry in HPFILTER region. + */ + prev_ftid = find_last_bit(t->hpftid_bmap, idx); + if (prev_ftid >= idx) + prev_ftid = idx; + + prev_tab = t->hpftid_tab; + } else { + idx -= t->nhpftids; + + /* Don't insert if there's a rule already present at @idx + * in normal FILTER region. + */ + if (test_bit(idx, t->ftid_bmap)) + return false; + + prev_tab = t->ftid_tab; + prev_ftid = find_last_bit(t->ftid_bmap, idx); + if (prev_ftid >= idx) { + /* No previous entry found in normal FILTER + * region. See if there's any previous entry + * in HPFILTER region. + */ + prev_ftid = find_last_bit(t->hpftid_bmap, t->nhpftids); + if (prev_ftid >= t->nhpftids) + prev_ftid = idx; + else + prev_tab = t->hpftid_tab; + } + + /* Search for the closest next filter entry in normal + * FILTER region. No need to search in HPFILTER region + * because there can never be any entry in HPFILTER + * region whose prio value is > first entry in normal + * FILTER region. + */ + next_ftid = find_next_bit(t->ftid_bmap, t->nftids, idx); + if (next_ftid >= t->nftids) + next_ftid = idx; + + next_tab = t->ftid_tab; + } + + next_fe = &next_tab[next_ftid]; + + /* See if the filter entry belongs to an IPv6 rule, which + * occupy 4 slots on T5 and 2 slots on T6. Adjust the + * reference to the previously inserted filter entry + * accordingly. + */ + prev_fe = &prev_tab[prev_ftid & ~(nslots - 1)]; + if (!prev_fe->fs.type) + prev_fe = &prev_tab[prev_ftid]; + + if ((prev_fe->valid && prev_fe->fs.tc_prio > prio) || + (next_fe->valid && next_fe->fs.tc_prio < prio)) + return false; + + return true; +} + +int cxgb4_get_free_ftid(struct net_device *dev, u8 family, bool hash_en, + u32 tc_prio) { struct adapter *adap = netdev2adap(dev); struct tid_info *t = &adap->tids; + u32 bmap_ftid, max_ftid; + struct filter_entry *f; + unsigned long *bmap; bool found = false; - u8 i, n, cnt; - int ftid; + u8 i, cnt, n; + int ftid = 0; /* IPv4 occupy 1 slot. IPv6 occupy 2 slots on T6 and 4 slots * on T5. @@ -456,34 +561,127 @@ int cxgb4_get_free_ftid(struct net_device *dev, int family) n += 2; } - if (n > t->nftids) - return -ENOMEM; - - /* Find free filter slots from the end of TCAM. Appropriate - * checks must be done by caller later to ensure the prio - * passed by TC doesn't conflict with prio saved by existing - * rules in the TCAM. + /* There are 3 filter regions available in hardware in + * following order of priority: + * + * 1. High Priority (HPFILTER) region (Highest Priority). + * 2. HASH region. + * 3. Normal FILTER region (Lowest Priority). + * + * Entries in HPFILTER and normal FILTER region have index + * 0 as the highest priority and the rules will be scanned + * in ascending order until either a rule hits or end of + * the region is reached. + * + * All HASH region entries have same priority. The set of + * fields to match in headers are pre-determined. The same + * set of header match fields must be compulsorily specified + * in all the rules wanting to get inserted in HASH region. + * Hence, HASH region is an exact-match region. A HASH is + * generated for a rule based on the values in the + * pre-determined set of header match fields. The generated + * HASH serves as an index into the HASH region. There can + * never be 2 rules having the same HASH. Hardware will + * compute a HASH for every incoming packet based on the + * values in the pre-determined set of header match fields + * and uses it as an index to check if there's a rule + * inserted in the HASH region at the specified index. If + * there's a rule inserted, then it's considered as a filter + * hit. Otherwise, it's a filter miss and normal FILTER region + * is scanned afterwards. */ + spin_lock_bh(&t->ftid_lock); - ftid = t->nftids - 1; - while (ftid >= n - 1) { + + ftid = (tc_prio <= t->nhpftids) ? 0 : t->nhpftids; + max_ftid = t->nftids + t->nhpftids; + while (ftid < max_ftid) { + if (ftid < t->nhpftids) { + /* If the new rule wants to get inserted into + * HPFILTER region, but its prio is greater + * than the rule with the highest prio in HASH + * region, then reject the rule. + */ + if (t->tc_hash_tids_max_prio && + tc_prio > t->tc_hash_tids_max_prio) + break; + + /* If there's not enough slots available + * in HPFILTER region, then move on to + * normal FILTER region immediately. + */ + if (ftid + n > t->nhpftids) { + ftid = t->nhpftids; + continue; + } + + bmap = t->hpftid_bmap; + bmap_ftid = ftid; + } else if (hash_en) { + /* Ensure priority is >= last rule in HPFILTER + * region. + */ + ftid = find_last_bit(t->hpftid_bmap, t->nhpftids); + if (ftid < t->nhpftids) { + f = &t->hpftid_tab[ftid]; + if (f->valid && tc_prio < f->fs.tc_prio) + break; + } + + /* Ensure priority is <= first rule in normal + * FILTER region. + */ + ftid = find_first_bit(t->ftid_bmap, t->nftids); + if (ftid < t->nftids) { + f = &t->ftid_tab[ftid]; + if (f->valid && tc_prio > f->fs.tc_prio) + break; + } + + found = true; + ftid = t->nhpftids; + goto out_unlock; + } else { + /* If the new rule wants to get inserted into + * normal FILTER region, but its prio is less + * than the rule with the highest prio in HASH + * region, then reject the rule. + */ + if (t->tc_hash_tids_max_prio && + tc_prio < t->tc_hash_tids_max_prio) + break; + + if (ftid + n > max_ftid) + break; + + bmap = t->ftid_bmap; + bmap_ftid = ftid - t->nhpftids; + } + cnt = 0; for (i = 0; i < n; i++) { - if (test_bit(ftid - i, t->ftid_bmap)) + if (test_bit(bmap_ftid + i, bmap)) break; cnt++; } + if (cnt == n) { - ftid &= ~(n - 1); - found = true; - break; + /* Ensure the new rule's prio doesn't conflict + * with existing rules. + */ + if (cxgb4_filter_prio_in_range(t, ftid, n, + tc_prio)) { + ftid &= ~(n - 1); + found = true; + break; + } } - ftid -= n; + ftid += n; } - spin_unlock_bh(&t->ftid_lock); - ftid += t->nhpftids; +out_unlock: + spin_unlock_bh(&t->ftid_lock); return found ? ftid : -ENOMEM; } @@ -555,73 +753,6 @@ static void cxgb4_clear_hpftid(struct tid_info *t, int fidx, int family) spin_unlock_bh(&t->ftid_lock); } -bool cxgb4_filter_prio_in_range(struct net_device *dev, u32 idx, u32 prio) -{ - struct filter_entry *prev_fe, *next_fe, *tab; - struct adapter *adap = netdev2adap(dev); - u32 prev_ftid, next_ftid, max_tid; - struct tid_info *t = &adap->tids; - unsigned long *bmap; - bool valid = true; - - if (idx < t->nhpftids) { - bmap = t->hpftid_bmap; - tab = t->hpftid_tab; - max_tid = t->nhpftids; - } else { - idx -= t->nhpftids; - bmap = t->ftid_bmap; - tab = t->ftid_tab; - max_tid = t->nftids; - } - - /* Only insert the rule if both of the following conditions - * are met: - * 1. The immediate previous rule has priority <= @prio. - * 2. The immediate next rule has priority >= @prio. - */ - spin_lock_bh(&t->ftid_lock); - - /* Don't insert if there's a rule already present at @idx. */ - if (test_bit(idx, bmap)) { - valid = false; - goto out_unlock; - } - - next_ftid = find_next_bit(bmap, max_tid, idx); - if (next_ftid >= max_tid) - next_ftid = idx; - - next_fe = &tab[next_ftid]; - - prev_ftid = find_last_bit(bmap, idx); - if (prev_ftid >= idx) - prev_ftid = idx; - - /* See if the filter entry belongs to an IPv6 rule, which - * occupy 4 slots on T5 and 2 slots on T6. Adjust the - * reference to the previously inserted filter entry - * accordingly. - */ - if (CHELSIO_CHIP_VERSION(adap->params.chip) < CHELSIO_T6) { - prev_fe = &tab[prev_ftid & ~0x3]; - if (!prev_fe->fs.type) - prev_fe = &tab[prev_ftid]; - } else { - prev_fe = &tab[prev_ftid & ~0x1]; - if (!prev_fe->fs.type) - prev_fe = &tab[prev_ftid]; - } - - if ((prev_fe->valid && prio < prev_fe->fs.tc_prio) || - (next_fe->valid && prio > next_fe->fs.tc_prio)) - valid = false; - -out_unlock: - spin_unlock_bh(&t->ftid_lock); - return valid; -} - /* Delete the filter at a specified index. */ static int del_filter_wr(struct adapter *adapter, int fidx) { diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.h index b3e4a645043d..b0751c0611ec 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.h @@ -53,5 +53,4 @@ void clear_all_filters(struct adapter *adapter); void init_hash_filter(struct adapter *adap); bool is_filter_exact_match(struct adapter *adap, struct ch_filter_specification *fs); -bool cxgb4_filter_prio_in_range(struct net_device *dev, u32 idx, u32 prio); #endif /* __CXGB4_FILTER_H */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 97f90edbc068..75fde0d4d493 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -90,11 +90,6 @@ char cxgb4_driver_name[] = KBUILD_MODNAME; -#ifdef DRV_VERSION -#undef DRV_VERSION -#endif -#define DRV_VERSION "2.0.0-ko" -const char cxgb4_driver_version[] = DRV_VERSION; #define DRV_DESC "Chelsio T4/T5/T6 Network Driver" #define DFLT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK | \ @@ -137,7 +132,6 @@ const char cxgb4_driver_version[] = DRV_VERSION; MODULE_DESCRIPTION(DRV_DESC); MODULE_AUTHOR("Chelsio Communications"); MODULE_LICENSE("Dual BSD/GPL"); -MODULE_VERSION(DRV_VERSION); MODULE_DEVICE_TABLE(pci, cxgb4_pci_tbl); MODULE_FIRMWARE(FW4_FNAME); MODULE_FIRMWARE(FW5_FNAME); @@ -3626,8 +3620,6 @@ static void cxgb4_mgmt_get_drvinfo(struct net_device *dev, struct adapter *adapter = netdev2adap(dev); strlcpy(info->driver, cxgb4_driver_name, sizeof(info->driver)); - strlcpy(info->version, cxgb4_driver_version, - sizeof(info->version)); strlcpy(info->bus_info, pci_name(adapter->pdev), sizeof(info->bus_info)); } @@ -6086,8 +6078,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) int i, err; u32 whoami; - printk_once(KERN_INFO "%s - version %s\n", DRV_DESC, DRV_VERSION); - err = pci_request_regions(pdev, KBUILD_MODNAME); if (err) { /* Just info, some other driver may have claimed the device. */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c index bb5513bdd293..4a5fa9eba0b6 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c @@ -544,7 +544,8 @@ static bool valid_pedit_action(struct net_device *dev, } int cxgb4_validate_flow_actions(struct net_device *dev, - struct flow_action *actions) + struct flow_action *actions, + struct netlink_ext_ack *extack) { struct flow_action_entry *act; bool act_redir = false; @@ -552,6 +553,9 @@ int cxgb4_validate_flow_actions(struct net_device *dev, bool act_vlan = false; int i; + if (!flow_action_basic_hw_stats_check(actions, extack)) + return -EOPNOTSUPP; + flow_action_for_each(i, act, actions) { switch (act->id) { case FLOW_ACTION_ACCEPT: @@ -631,6 +635,64 @@ int cxgb4_validate_flow_actions(struct net_device *dev, return 0; } +static void cxgb4_tc_flower_hash_prio_add(struct adapter *adap, u32 tc_prio) +{ + spin_lock_bh(&adap->tids.ftid_lock); + if (adap->tids.tc_hash_tids_max_prio < tc_prio) + adap->tids.tc_hash_tids_max_prio = tc_prio; + spin_unlock_bh(&adap->tids.ftid_lock); +} + +static void cxgb4_tc_flower_hash_prio_del(struct adapter *adap, u32 tc_prio) +{ + struct tid_info *t = &adap->tids; + struct ch_tc_flower_entry *fe; + struct rhashtable_iter iter; + u32 found = 0; + + spin_lock_bh(&t->ftid_lock); + /* Bail if the current rule is not the one with the max + * prio. + */ + if (t->tc_hash_tids_max_prio != tc_prio) + goto out_unlock; + + /* Search for the next rule having the same or next lower + * max prio. + */ + rhashtable_walk_enter(&adap->flower_tbl, &iter); + do { + rhashtable_walk_start(&iter); + + fe = rhashtable_walk_next(&iter); + while (!IS_ERR_OR_NULL(fe)) { + if (fe->fs.hash && + fe->fs.tc_prio <= t->tc_hash_tids_max_prio) { + t->tc_hash_tids_max_prio = fe->fs.tc_prio; + found++; + + /* Bail if we found another rule + * having the same prio as the + * current max one. + */ + if (fe->fs.tc_prio == tc_prio) + break; + } + + fe = rhashtable_walk_next(&iter); + } + + rhashtable_walk_stop(&iter); + } while (fe == ERR_PTR(-EAGAIN)); + rhashtable_walk_exit(&iter); + + if (!found) + t->tc_hash_tids_max_prio = 0; + +out_unlock: + spin_unlock_bh(&t->ftid_lock); +} + int cxgb4_tc_flower_replace(struct net_device *dev, struct flow_cls_offload *cls) { @@ -640,9 +702,10 @@ int cxgb4_tc_flower_replace(struct net_device *dev, struct ch_tc_flower_entry *ch_flower; struct ch_filter_specification *fs; struct filter_ctx ctx; + u8 inet_family; int fidx, ret; - if (cxgb4_validate_flow_actions(dev, &rule->action)) + if (cxgb4_validate_flow_actions(dev, &rule->action, extack)) return -EOPNOTSUPP; if (cxgb4_validate_flow_match(dev, cls)) @@ -660,39 +723,32 @@ int cxgb4_tc_flower_replace(struct net_device *dev, cxgb4_process_flow_actions(dev, &rule->action, fs); fs->hash = is_filter_exact_match(adap, fs); - if (fs->hash) { - fidx = 0; - } else { - u8 inet_family; - - inet_family = fs->type ? PF_INET6 : PF_INET; + inet_family = fs->type ? PF_INET6 : PF_INET; - /* Note that TC uses prio 0 to indicate stack to - * generate automatic prio and hence doesn't pass prio - * 0 to driver. However, the hardware TCAM index - * starts from 0. Hence, the -1 here. - */ - if (cls->common.prio <= (adap->tids.nftids + - adap->tids.nhpftids)) { - fidx = cls->common.prio - 1; - if (fidx < adap->tids.nhpftids) - fs->prio = 1; - } else { - fidx = cxgb4_get_free_ftid(dev, inet_family); - } + /* Get a free filter entry TID, where we can insert this new + * rule. Only insert rule if its prio doesn't conflict with + * existing rules. + */ + fidx = cxgb4_get_free_ftid(dev, inet_family, fs->hash, + cls->common.prio); + if (fidx < 0) { + NL_SET_ERR_MSG_MOD(extack, + "No free LETCAM index available"); + ret = -ENOMEM; + goto free_entry; + } - /* Only insert FLOWER rule if its priority doesn't - * conflict with existing rules in the LETCAM. - */ - if (fidx < 0 || - !cxgb4_filter_prio_in_range(dev, fidx, cls->common.prio)) { - NL_SET_ERR_MSG_MOD(extack, - "No free LETCAM index available"); - ret = -ENOMEM; - goto free_entry; - } + if (fidx < adap->tids.nhpftids) { + fs->prio = 1; + fs->hash = 0; } + /* If the rule can be inserted into HASH region, then ignore + * the index to normal FILTER region. + */ + if (fs->hash) + fidx = 0; + fs->tc_prio = cls->common.prio; fs->tc_cookie = cls->cookie; @@ -723,6 +779,9 @@ int cxgb4_tc_flower_replace(struct net_device *dev, if (ret) goto del_filter; + if (fs->hash) + cxgb4_tc_flower_hash_prio_add(adap, cls->common.prio); + return 0; del_filter: @@ -738,12 +797,17 @@ int cxgb4_tc_flower_destroy(struct net_device *dev, { struct adapter *adap = netdev2adap(dev); struct ch_tc_flower_entry *ch_flower; + u32 tc_prio; + bool hash; int ret; ch_flower = ch_flower_lookup(adap, cls->cookie); if (!ch_flower) return -ENOENT; + hash = ch_flower->fs.hash; + tc_prio = ch_flower->fs.tc_prio; + ret = cxgb4_del_filter(dev, ch_flower->filter_id, &ch_flower->fs); if (ret) goto err; @@ -756,6 +820,9 @@ int cxgb4_tc_flower_destroy(struct net_device *dev, } kfree_rcu(ch_flower, rcu); + if (hash) + cxgb4_tc_flower_hash_prio_del(adap, tc_prio); + err: return ret; } @@ -836,7 +903,8 @@ int cxgb4_tc_flower_stats(struct net_device *dev, ofld_stats->last_used = jiffies; flow_stats_update(&cls->stats, bytes - ofld_stats->byte_count, packets - ofld_stats->packet_count, - ofld_stats->last_used); + ofld_stats->last_used, + FLOW_ACTION_HW_STATS_IMMEDIATE); ofld_stats->packet_count = packets; ofld_stats->byte_count = bytes; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h index e132516e9868..0a30c96b81ff 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h @@ -112,7 +112,8 @@ void cxgb4_process_flow_actions(struct net_device *in, struct flow_action *actions, struct ch_filter_specification *fs); int cxgb4_validate_flow_actions(struct net_device *dev, - struct flow_action *actions); + struct flow_action *actions, + struct netlink_ext_ack *extack); int cxgb4_tc_flower_replace(struct net_device *dev, struct flow_cls_offload *cls); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c index 1b7681a4eb32..c88c47a14fbb 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c @@ -198,22 +198,14 @@ static int cxgb4_matchall_alloc_filter(struct net_device *dev, struct ch_filter_specification *fs; int ret, fidx; - /* Note that TC uses prio 0 to indicate stack to generate - * automatic prio and hence doesn't pass prio 0 to driver. - * However, the hardware TCAM index starts from 0. Hence, the - * -1 here. 1 slot is enough to create a wildcard matchall - * VIID rule. + /* Get a free filter entry TID, where we can insert this new + * rule. Only insert rule if its prio doesn't conflict with + * existing rules. + * + * 1 slot is enough to create a wildcard matchall VIID rule. */ - if (cls->common.prio <= (adap->tids.nftids + adap->tids.nhpftids)) - fidx = cls->common.prio - 1; - else - fidx = cxgb4_get_free_ftid(dev, PF_INET); - - /* Only insert MATCHALL rule if its priority doesn't conflict - * with existing rules in the LETCAM. - */ - if (fidx < 0 || - !cxgb4_filter_prio_in_range(dev, fidx, cls->common.prio)) { + fidx = cxgb4_get_free_ftid(dev, PF_INET, false, cls->common.prio); + if (fidx < 0) { NL_SET_ERR_MSG_MOD(extack, "No free LETCAM index available"); return -ENOMEM; @@ -286,7 +278,8 @@ int cxgb4_tc_matchall_replace(struct net_device *dev, } ret = cxgb4_validate_flow_actions(dev, - &cls_matchall->rule->action); + &cls_matchall->rule->action, + extack); if (ret) return ret; @@ -353,7 +346,8 @@ int cxgb4_tc_matchall_stats(struct net_device *dev, flow_stats_update(&cls_matchall->stats, bytes - tc_port_matchall->ingress.bytes, packets - tc_port_matchall->ingress.packets, - tc_port_matchall->ingress.last_used); + tc_port_matchall->ingress.last_used, + FLOW_ACTION_HW_STATS_IMMEDIATE); tc_port_matchall->ingress.packets = packets; tc_port_matchall->ingress.bytes = bytes; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c index 269b8d9e25e0..3f3c11e54d97 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c @@ -155,9 +155,10 @@ int cxgb4_config_knode(struct net_device *dev, struct tc_cls_u32_offload *cls) struct ch_filter_specification fs; struct cxgb4_tc_u32_table *t; struct cxgb4_link *link; - unsigned int filter_id; u32 uhtid, link_uhtid; bool is_ipv6 = false; + u8 inet_family; + int filter_id; int ret; if (!can_tc_u32_offload(dev)) @@ -166,18 +167,15 @@ int cxgb4_config_knode(struct net_device *dev, struct tc_cls_u32_offload *cls) if (protocol != htons(ETH_P_IP) && protocol != htons(ETH_P_IPV6)) return -EOPNOTSUPP; - /* Note that TC uses prio 0 to indicate stack to generate - * automatic prio and hence doesn't pass prio 0 to driver. - * However, the hardware TCAM index starts from 0. Hence, the - * -1 here. - */ - filter_id = TC_U32_NODE(cls->knode.handle) - 1; + inet_family = (protocol == htons(ETH_P_IPV6)) ? PF_INET6 : PF_INET; - /* Only insert U32 rule if its priority doesn't conflict with - * existing rules in the LETCAM. + /* Get a free filter entry TID, where we can insert this new + * rule. Only insert rule if its prio doesn't conflict with + * existing rules. */ - if (filter_id >= adapter->tids.nftids + adapter->tids.nhpftids || - !cxgb4_filter_prio_in_range(dev, filter_id, cls->common.prio)) { + filter_id = cxgb4_get_free_ftid(dev, inet_family, false, + TC_U32_NODE(cls->knode.handle)); + if (filter_id < 0) { NL_SET_ERR_MSG_MOD(extack, "No free LETCAM index available"); return -ENOMEM; @@ -358,23 +356,65 @@ int cxgb4_delete_knode(struct net_device *dev, struct tc_cls_u32_offload *cls) struct cxgb4_link *link = NULL; struct cxgb4_tc_u32_table *t; struct filter_entry *f; + bool found = false; u32 handle, uhtid; + u8 nslots; int ret; if (!can_tc_u32_offload(dev)) return -EOPNOTSUPP; /* Fetch the location to delete the filter. */ - filter_id = TC_U32_NODE(cls->knode.handle) - 1; - if (filter_id >= adapter->tids.nftids + adapter->tids.nhpftids) - return -ERANGE; + max_tids = adapter->tids.nhpftids + adapter->tids.nftids; + + spin_lock_bh(&adapter->tids.ftid_lock); + filter_id = 0; + while (filter_id < max_tids) { + if (filter_id < adapter->tids.nhpftids) { + i = filter_id; + f = &adapter->tids.hpftid_tab[i]; + if (f->valid && f->fs.tc_cookie == cls->knode.handle) { + found = true; + break; + } - if (filter_id < adapter->tids.nhpftids) - f = &adapter->tids.hpftid_tab[filter_id]; - else - f = &adapter->tids.ftid_tab[filter_id - adapter->tids.nhpftids]; + i = find_next_bit(adapter->tids.hpftid_bmap, + adapter->tids.nhpftids, i + 1); + if (i >= adapter->tids.nhpftids) { + filter_id = adapter->tids.nhpftids; + continue; + } + + filter_id = i; + } else { + i = filter_id - adapter->tids.nhpftids; + f = &adapter->tids.ftid_tab[i]; + if (f->valid && f->fs.tc_cookie == cls->knode.handle) { + found = true; + break; + } + + i = find_next_bit(adapter->tids.ftid_bmap, + adapter->tids.nftids, i + 1); + if (i >= adapter->tids.nftids) + break; + + filter_id = i + adapter->tids.nhpftids; + } + + nslots = 0; + if (f->fs.type) { + nslots++; + if (CHELSIO_CHIP_VERSION(adapter->params.chip) < + CHELSIO_T6) + nslots += 2; + } + + filter_id += nslots; + } + spin_unlock_bh(&adapter->tids.ftid_lock); - if (cls->knode.handle != f->fs.tc_cookie) + if (!found) return -ERANGE; t = adapter->tc_u32; @@ -407,7 +447,6 @@ int cxgb4_delete_knode(struct net_device *dev, struct tc_cls_u32_offload *cls) /* If a link is being deleted, then delete all filters * associated with the link. */ - max_tids = adapter->tids.nftids; for (i = 0; i < t->size; i++) { link = &t->table[i]; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32_parse.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32_parse.h index a4b99edcc339..125868c6770a 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32_parse.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32_parse.h @@ -289,6 +289,6 @@ struct cxgb4_link { struct cxgb4_tc_u32_table { unsigned int size; /* number of entries in table */ - struct cxgb4_link table[0]; /* Jump table */ + struct cxgb4_link table[]; /* Jump table */ }; #endif /* __CXGB4_TC_U32_PARSE_H */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c index cce33d279094..e65b52375dd8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c @@ -662,6 +662,25 @@ static int uld_attach(struct adapter *adap, unsigned int uld) return 0; } +#ifdef CONFIG_CHELSIO_TLS_DEVICE +/* cxgb4_set_ktls_feature: request FW to enable/disable ktls settings. + * @adap: adapter info + * @enable: 1 to enable / 0 to disable ktls settings. + */ +static void cxgb4_set_ktls_feature(struct adapter *adap, bool enable) +{ + u32 params = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | + FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_KTLS_TX_HW) | + FW_PARAMS_PARAM_Y_V(enable)); + int ret = 0; + + ret = t4_set_params(adap, adap->mbox, adap->pf, 0, 1, ¶ms, ¶ms); + /* if fw returns failure, clear the ktls flag */ + if (ret) + adap->params.crypto &= ~ULP_CRYPTO_KTLS_INLINE; +} +#endif + /* cxgb4_register_uld - register an upper-layer driver * @type: the ULD type * @p: the ULD methods @@ -698,6 +717,12 @@ void cxgb4_register_uld(enum cxgb4_uld type, } if (adap->flags & CXGB4_FULL_INIT_DONE) enable_rx_uld(adap, type); +#ifdef CONFIG_CHELSIO_TLS_DEVICE + /* send mbox to enable ktls related settings. */ + if (type == CXGB4_ULD_CRYPTO && + (adap->params.crypto & FW_CAPS_CONFIG_TX_TLS_HW)) + cxgb4_set_ktls_feature(adap, 1); +#endif if (adap->uld[type].add) goto free_irq; ret = setup_sge_txq_uld(adap, type, p); @@ -750,6 +775,13 @@ int cxgb4_unregister_uld(enum cxgb4_uld type) continue; cxgb4_shutdown_uld_adapter(adap, type); + +#ifdef CONFIG_CHELSIO_TLS_DEVICE + /* send mbox to disable ktls related settings. */ + if (type == CXGB4_ULD_CRYPTO && + (adap->params.crypto & FW_CAPS_CONFIG_TX_TLS_HW)) + cxgb4_set_ktls_feature(adap, 0); +#endif } mutex_unlock(&uld_mutex); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h index d9d27bc1ae67..be831317520a 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h @@ -149,6 +149,8 @@ struct tid_info { atomic_t conns_in_use; /* lock for setting/clearing filter bitmap */ spinlock_t ftid_lock; + + unsigned int tc_hash_tids_max_prio; }; static inline void *lookup_tid(const struct tid_info *t, unsigned int tid) @@ -263,7 +265,8 @@ struct filter_ctx { struct ch_filter_specification; -int cxgb4_get_free_ftid(struct net_device *dev, int family); +int cxgb4_get_free_ftid(struct net_device *dev, u8 family, bool hash_en, + u32 tc_prio); int __cxgb4_set_filter(struct net_device *dev, int filter_id, struct ch_filter_specification *fs, struct filter_ctx *ctx); @@ -357,6 +360,26 @@ struct chcr_stats_debug { atomic_t tls_pdu_tx; atomic_t tls_pdu_rx; atomic_t tls_key; +#ifdef CONFIG_CHELSIO_TLS_DEVICE + atomic64_t ktls_tx_connection_open; + atomic64_t ktls_tx_connection_fail; + atomic64_t ktls_tx_connection_close; + atomic64_t ktls_tx_send_records; + atomic64_t ktls_tx_end_pkts; + atomic64_t ktls_tx_start_pkts; + atomic64_t ktls_tx_middle_pkts; + atomic64_t ktls_tx_retransmit_pkts; + atomic64_t ktls_tx_complete_pkts; + atomic64_t ktls_tx_trimmed_pkts; + atomic64_t ktls_tx_encrypted_packets; + atomic64_t ktls_tx_encrypted_bytes; + atomic64_t ktls_tx_ctx; + atomic64_t ktls_tx_ooo; + atomic64_t ktls_tx_skip_no_sync_data; + atomic64_t ktls_tx_drop_no_sync_data; + atomic64_t ktls_tx_drop_bypass_req; + +#endif }; #define OCQ_WIN_OFFSET(pdev, vres) \ diff --git a/drivers/net/ethernet/chelsio/cxgb4/l2t.c b/drivers/net/ethernet/chelsio/cxgb4/l2t.c index 1a16449e9deb..72b37a66c7d8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/l2t.c +++ b/drivers/net/ethernet/chelsio/cxgb4/l2t.c @@ -59,7 +59,7 @@ struct l2t_data { rwlock_t lock; atomic_t nfree; /* number of free entries */ struct l2t_entry *rover; /* starting point for next allocation */ - struct l2t_entry l2tab[0]; /* MUST BE LAST */ + struct l2t_entry l2tab[]; /* MUST BE LAST */ }; static inline unsigned int vlan_prio(const struct l2t_entry *e) @@ -700,6 +700,17 @@ static char l2e_state(const struct l2t_entry *e) } } +bool cxgb4_check_l2t_valid(struct l2t_entry *e) +{ + bool valid; + + spin_lock(&e->lock); + valid = (e->state == L2T_STATE_VALID); + spin_unlock(&e->lock); + return valid; +} +EXPORT_SYMBOL(cxgb4_check_l2t_valid); + static int l2t_seq_show(struct seq_file *seq, void *v) { if (v == SEQ_START_TOKEN) diff --git a/drivers/net/ethernet/chelsio/cxgb4/l2t.h b/drivers/net/ethernet/chelsio/cxgb4/l2t.h index 79665bd8f881..340fecb28a13 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/l2t.h +++ b/drivers/net/ethernet/chelsio/cxgb4/l2t.h @@ -122,6 +122,7 @@ struct l2t_entry *t4_l2t_alloc_switching(struct adapter *adap, u16 vlan, u8 port, u8 *dmac); struct l2t_data *t4_init_l2t(unsigned int l2t_start, unsigned int l2t_end); void do_l2t_write_rpl(struct adapter *p, const struct cpl_l2t_write_rpl *rpl); +bool cxgb4_check_l2t_valid(struct l2t_entry *e); extern const struct file_operations t4_l2t_fops; #endif /* __CXGB4_L2T_H */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/sched.h b/drivers/net/ethernet/chelsio/cxgb4/sched.h index 5cc74a5a1774..5f8b871d79af 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sched.h +++ b/drivers/net/ethernet/chelsio/cxgb4/sched.h @@ -82,7 +82,7 @@ struct sched_class { struct sched_table { /* per port scheduling table */ u8 sched_size; - struct sched_class tab[0]; + struct sched_class tab[]; }; static inline bool can_sched(struct net_device *dev) diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index cab3d17e0e1a..f5dd34db4b54 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c @@ -1418,6 +1418,11 @@ static netdev_tx_t cxgb4_eth_xmit(struct sk_buff *skb, struct net_device *dev) return adap->uld[CXGB4_ULD_CRYPTO].tx_handler(skb, dev); #endif /* CHELSIO_IPSEC_INLINE */ +#ifdef CONFIG_CHELSIO_TLS_DEVICE + if (skb->decrypted) + return adap->uld[CXGB4_ULD_CRYPTO].tx_handler(skb, dev); +#endif /* CHELSIO_TLS_DEVICE */ + qidx = skb_get_queue_mapping(skb); if (ptp_enabled) { spin_lock(&adap->ptp_lock); diff --git a/drivers/net/ethernet/chelsio/cxgb4/smt.h b/drivers/net/ethernet/chelsio/cxgb4/smt.h index 1268d6e93a47..541249d78914 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/smt.h +++ b/drivers/net/ethernet/chelsio/cxgb4/smt.h @@ -66,7 +66,7 @@ struct smt_entry { struct smt_data { unsigned int smt_size; rwlock_t lock; - struct smt_entry smtab[0]; + struct smt_entry smtab[]; }; struct smt_data *t4_init_smt(void); diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index 844fdcf55118..239f678a94ed 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -1379,8 +1379,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x9608, 0x9638, 0x9640, 0x96f4, 0x9800, 0x9808, - 0x9820, 0x983c, - 0x9850, 0x9864, + 0x9810, 0x9864, 0x9c00, 0x9c6c, 0x9c80, 0x9cec, 0x9d00, 0x9d6c, @@ -1389,7 +1388,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x9e80, 0x9eec, 0x9f00, 0x9f6c, 0x9f80, 0xa020, - 0xd004, 0xd004, + 0xd000, 0xd004, 0xd010, 0xd03c, 0xdfc0, 0xdfe0, 0xe000, 0x1106c, @@ -1430,10 +1429,8 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x1a0b0, 0x1a0e4, 0x1a0ec, 0x1a0f8, 0x1a100, 0x1a108, - 0x1a114, 0x1a120, - 0x1a128, 0x1a130, - 0x1a138, 0x1a138, - 0x1a190, 0x1a1c4, + 0x1a114, 0x1a130, + 0x1a138, 0x1a1c4, 0x1a1fc, 0x1a1fc, 0x1e008, 0x1e00c, 0x1e040, 0x1e044, @@ -2162,8 +2159,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x9640, 0x9704, 0x9710, 0x971c, 0x9800, 0x9808, - 0x9820, 0x983c, - 0x9850, 0x9864, + 0x9810, 0x9864, 0x9c00, 0x9c6c, 0x9c80, 0x9cec, 0x9d00, 0x9d6c, @@ -2172,7 +2168,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x9e80, 0x9eec, 0x9f00, 0x9f6c, 0x9f80, 0xa020, - 0xd004, 0xd03c, + 0xd000, 0xd03c, 0xd100, 0xd118, 0xd200, 0xd214, 0xd220, 0xd234, @@ -2240,10 +2236,8 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x1a0b0, 0x1a0e4, 0x1a0ec, 0x1a0f8, 0x1a100, 0x1a108, - 0x1a114, 0x1a120, - 0x1a128, 0x1a130, - 0x1a138, 0x1a138, - 0x1a190, 0x1a1c4, + 0x1a114, 0x1a130, + 0x1a138, 0x1a1c4, 0x1a1fc, 0x1a1fc, 0x1e008, 0x1e00c, 0x1e040, 0x1e044, @@ -4480,7 +4474,7 @@ static void tp_intr_handler(struct adapter *adapter) */ static void sge_intr_handler(struct adapter *adapter) { - u64 v; + u32 v = 0, perr; u32 err; static const struct intr_info sge_intr_info[] = { @@ -4515,13 +4509,29 @@ static void sge_intr_handler(struct adapter *adapter) { 0 } }; - v = (u64)t4_read_reg(adapter, SGE_INT_CAUSE1_A) | - ((u64)t4_read_reg(adapter, SGE_INT_CAUSE2_A) << 32); - if (v) { - dev_alert(adapter->pdev_dev, "SGE parity error (%#llx)\n", - (unsigned long long)v); - t4_write_reg(adapter, SGE_INT_CAUSE1_A, v); - t4_write_reg(adapter, SGE_INT_CAUSE2_A, v >> 32); + perr = t4_read_reg(adapter, SGE_INT_CAUSE1_A); + if (perr) { + v |= perr; + dev_alert(adapter->pdev_dev, "SGE Cause1 Parity Error %#x\n", + perr); + } + + perr = t4_read_reg(adapter, SGE_INT_CAUSE2_A); + if (perr) { + v |= perr; + dev_alert(adapter->pdev_dev, "SGE Cause2 Parity Error %#x\n", + perr); + } + + if (CHELSIO_CHIP_VERSION(adapter->params.chip) >= CHELSIO_T5) { + perr = t4_read_reg(adapter, SGE_INT_CAUSE5_A); + /* Parity error (CRC) for err_T_RxCRC is trivial, ignore it */ + perr &= ~ERR_T_RXCRC_F; + if (perr) { + v |= perr; + dev_alert(adapter->pdev_dev, + "SGE Cause5 Parity Error %#x\n", perr); + } } v |= t4_handle_intr_status(adapter, SGE_INT_CAUSE3_A, sge_intr_info); diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h index 575c6abcdae7..fed5f93bf620 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h @@ -47,6 +47,7 @@ enum { CPL_CLOSE_LISTSRV_REQ = 0x9, CPL_ABORT_REQ = 0xA, CPL_ABORT_RPL = 0xB, + CPL_TX_DATA = 0xC, CPL_RX_DATA_ACK = 0xD, CPL_TX_PKT = 0xE, CPL_L2T_WRITE_REQ = 0x12, @@ -705,6 +706,14 @@ struct cpl_set_tcb_field { __be64 val; }; +struct cpl_set_tcb_field_core { + union opcode_tid ot; + __be16 reply_ctrl; + __be16 word_cookie; + __be64 mask; + __be64 val; +}; + /* cpl_set_tcb_field.word_cookie fields */ #define TCB_WORD_S 0 #define TCB_WORD_V(x) ((x) << TCB_WORD_S) @@ -1462,6 +1471,16 @@ struct cpl_tx_data { #define TX_FORCE_S 13 #define TX_FORCE_V(x) ((x) << TX_FORCE_S) +#define TX_DATA_MSS_S 16 +#define TX_DATA_MSS_M 0xFFFF +#define TX_DATA_MSS_V(x) ((x) << TX_DATA_MSS_S) +#define TX_DATA_MSS_G(x) (((x) >> TX_DATA_MSS_S) & TX_DATA_MSS_M) + +#define TX_LENGTH_S 0 +#define TX_LENGTH_M 0xFFFF +#define TX_LENGTH_V(x) ((x) << TX_LENGTH_S) +#define TX_LENGTH_G(x) (((x) >> TX_LENGTH_S) & TX_LENGTH_M) + #define T6_TX_FORCE_S 20 #define T6_TX_FORCE_V(x) ((x) << T6_TX_FORCE_S) #define T6_TX_FORCE_F T6_TX_FORCE_V(1U) @@ -1471,6 +1490,15 @@ struct cpl_tx_data { #define TX_SHOVE_S 14 #define TX_SHOVE_V(x) ((x) << TX_SHOVE_S) +#define TX_SHOVE_F TX_SHOVE_V(1U) + +#define TX_BYPASS_S 21 +#define TX_BYPASS_V(x) ((x) << TX_BYPASS_S) +#define TX_BYPASS_F TX_BYPASS_V(1U) + +#define TX_PUSH_S 22 +#define TX_PUSH_V(x) ((x) << TX_PUSH_S) +#define TX_PUSH_F TX_PUSH_V(1U) #define TX_ULP_MODE_S 10 #define TX_ULP_MODE_M 0x7 @@ -1511,7 +1539,7 @@ struct ulptx_sgl { __be32 cmd_nsge; __be32 len0; __be64 addr0; - struct ulptx_sge_pair sge[0]; + struct ulptx_sge_pair sge[]; }; struct ulptx_idata { diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h index a957a6e4d4c4..bb20e50ddb84 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h @@ -487,6 +487,12 @@ #define ERROR_QID_M 0x1ffffU #define ERROR_QID_G(x) (((x) >> ERROR_QID_S) & ERROR_QID_M) +#define SGE_INT_CAUSE5_A 0x110c + +#define ERR_T_RXCRC_S 31 +#define ERR_T_RXCRC_V(x) ((x) << ERR_T_RXCRC_S) +#define ERR_T_RXCRC_F ERR_T_RXCRC_V(1U) + #define HP_INT_THRESH_S 28 #define HP_INT_THRESH_M 0xfU #define HP_INT_THRESH_V(x) ((x) << HP_INT_THRESH_S) diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h b/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h index 1b9afb192f7f..50232e063f49 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h @@ -35,6 +35,11 @@ #ifndef __T4_TCB_H #define __T4_TCB_H +#define TCB_L2T_IX_W 0 +#define TCB_L2T_IX_S 12 +#define TCB_L2T_IX_M 0xfffULL +#define TCB_L2T_IX_V(x) ((x) << TCB_L2T_IX_S) + #define TCB_SMAC_SEL_W 0 #define TCB_SMAC_SEL_S 24 #define TCB_SMAC_SEL_M 0xffULL @@ -45,11 +50,6 @@ #define TCB_T_FLAGS_M 0xffffffffffffffffULL #define TCB_T_FLAGS_V(x) ((__u64)(x) << TCB_T_FLAGS_S) -#define TCB_RQ_START_W 30 -#define TCB_RQ_START_S 0 -#define TCB_RQ_START_M 0x3ffffffULL -#define TCB_RQ_START_V(x) ((x) << TCB_RQ_START_S) - #define TF_CCTRL_ECE_S 60 #define TF_CCTRL_CWR_S 61 #define TF_CCTRL_RFR_S 62 @@ -59,6 +59,11 @@ #define TCB_RSS_INFO_M 0x3ffULL #define TCB_RSS_INFO_V(x) ((x) << TCB_RSS_INFO_S) +#define TCB_T_STATE_W 3 +#define TCB_T_STATE_S 16 +#define TCB_T_STATE_M 0xfULL +#define TCB_T_STATE_V(x) ((x) << TCB_T_STATE_S) + #define TCB_TIMESTAMP_W 5 #define TCB_TIMESTAMP_S 0 #define TCB_TIMESTAMP_M 0xffffffffULL @@ -69,13 +74,60 @@ #define TCB_RTT_TS_RECENT_AGE_M 0xffffffffULL #define TCB_RTT_TS_RECENT_AGE_V(x) ((x) << TCB_RTT_TS_RECENT_AGE_S) +#define TCB_T_RTSEQ_RECENT_W 7 +#define TCB_T_RTSEQ_RECENT_S 0 +#define TCB_T_RTSEQ_RECENT_M 0xffffffffULL +#define TCB_T_RTSEQ_RECENT_V(x) ((x) << TCB_T_RTSEQ_RECENT_S) + +#define TCB_TX_MAX_W 9 +#define TCB_TX_MAX_S 0 +#define TCB_TX_MAX_M 0xffffffffULL +#define TCB_TX_MAX_V(x) ((x) << TCB_TX_MAX_S) + #define TCB_SND_UNA_RAW_W 10 +#define TCB_SND_UNA_RAW_S 0 +#define TCB_SND_UNA_RAW_M 0xfffffffULL +#define TCB_SND_UNA_RAW_V(x) ((x) << TCB_SND_UNA_RAW_S) + +#define TCB_SND_NXT_RAW_W 10 +#define TCB_SND_NXT_RAW_S 28 +#define TCB_SND_NXT_RAW_M 0xfffffffULL +#define TCB_SND_NXT_RAW_V(x) ((x) << TCB_SND_NXT_RAW_S) + +#define TCB_SND_MAX_RAW_W 11 +#define TCB_SND_MAX_RAW_S 24 +#define TCB_SND_MAX_RAW_M 0xfffffffULL +#define TCB_SND_MAX_RAW_V(x) ((x) << TCB_SND_MAX_RAW_S) + +#define TCB_RCV_NXT_W 16 +#define TCB_RCV_NXT_S 10 +#define TCB_RCV_NXT_M 0xffffffffULL +#define TCB_RCV_NXT_V(x) ((x) << TCB_RCV_NXT_S) + +#define TCB_RCV_WND_W 17 +#define TCB_RCV_WND_S 10 +#define TCB_RCV_WND_M 0xffffffULL +#define TCB_RCV_WND_V(x) ((x) << TCB_RCV_WND_S) + #define TCB_RX_FRAG2_PTR_RAW_W 27 #define TCB_RX_FRAG3_LEN_RAW_W 29 #define TCB_RX_FRAG3_START_IDX_OFFSET_RAW_W 30 #define TCB_PDU_HDR_LEN_W 31 +#define TCB_RQ_START_W 30 +#define TCB_RQ_START_S 0 +#define TCB_RQ_START_M 0x3ffffffULL +#define TCB_RQ_START_V(x) ((x) << TCB_RQ_START_S) + #define TF_RX_PDU_OUT_S 49 #define TF_RX_PDU_OUT_V(x) ((__u64)(x) << TF_RX_PDU_OUT_S) +#define TF_CORE_BYPASS_S 63 +#define TF_CORE_BYPASS_V(x) ((__u64)(x) << TF_CORE_BYPASS_S) +#define TF_CORE_BYPASS_F TF_CORE_BYPASS_V(1) + +#define TF_NON_OFFLOAD_S 1 +#define TF_NON_OFFLOAD_V(x) ((x) << TF_NON_OFFLOAD_S) +#define TF_NON_OFFLOAD_F TF_NON_OFFLOAD_V(1) + #endif /* __T4_TCB_H */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h index accad1101ad1..68fe734b9b37 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h @@ -737,7 +737,7 @@ struct fw_flowc_mnemval { struct fw_flowc_wr { __be32 op_to_nparams; __be32 flowid_len16; - struct fw_flowc_mnemval mnemval[0]; + struct fw_flowc_mnemval mnemval[]; }; #define FW_FLOWC_WR_NPARAMS_S 0 @@ -1205,6 +1205,7 @@ enum fw_caps_config_crypto { FW_CAPS_CONFIG_CRYPTO_LOOKASIDE = 0x00000001, FW_CAPS_CONFIG_TLS_INLINE = 0x00000002, FW_CAPS_CONFIG_IPSEC_INLINE = 0x00000004, + FW_CAPS_CONFIG_TX_TLS_HW = 0x00000008, }; enum fw_caps_config_fcoe { @@ -1328,6 +1329,7 @@ enum fw_params_param_dev { FW_PARAMS_PARAM_DEV_DBQ_TIMERTICK = 0x2A, FW_PARAMS_PARAM_DEV_NUM_TM_CLASS = 0x2B, FW_PARAMS_PARAM_DEV_FILTER = 0x2E, + FW_PARAMS_PARAM_DEV_KTLS_TX_HW = 0x31, }; /* diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index f4d41f968afa..9cc3541a7e1c 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -55,7 +55,6 @@ /* * Generic information about the driver. */ -#define DRV_VERSION "2.0.0-ko" #define DRV_DESC "Chelsio T4/T5/T6 Virtual Function (VF) Network Driver" /* @@ -1556,7 +1555,6 @@ static void cxgb4vf_get_drvinfo(struct net_device *dev, struct adapter *adapter = netdev2adap(dev); strlcpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version)); strlcpy(drvinfo->bus_info, pci_name(to_pci_dev(dev->dev.parent)), sizeof(drvinfo->bus_info)); snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), @@ -1921,6 +1919,8 @@ static void cxgb4vf_get_wol(struct net_device *dev, NETIF_F_GRO | NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA) static const struct ethtool_ops cxgb4vf_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_RX_MAX_FRAMES, .get_link_ksettings = cxgb4vf_get_link_ksettings, .get_fecparam = cxgb4vf_get_fecparam, .get_drvinfo = cxgb4vf_get_drvinfo, @@ -2934,12 +2934,6 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev, unsigned int pf; /* - * Print our driver banner the first time we're called to initialize a - * device. - */ - pr_info_once("%s - version %s\n", DRV_DESC, DRV_VERSION); - - /* * Initialize generic PCI device state. */ err = pci_enable_device(pdev); @@ -3454,7 +3448,6 @@ static void cxgb4vf_pci_shutdown(struct pci_dev *pdev) MODULE_DESCRIPTION(DRV_DESC); MODULE_AUTHOR("Chelsio Communications"); MODULE_LICENSE("Dual BSD/GPL"); -MODULE_VERSION(DRV_VERSION); MODULE_DEVICE_TABLE(pci, cxgb4vf_pci_tbl); static struct pci_driver cxgb4vf_driver = { diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c index 21034536c9c5..854d87e1125c 100644 --- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c +++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c @@ -35,7 +35,6 @@ */ #define DRV_NAME "libcxgb" -#define DRV_VERSION "1.0.0-ko" #define pr_fmt(fmt) DRV_NAME ": " fmt #include <linux/kernel.h> @@ -530,5 +529,4 @@ EXPORT_SYMBOL(cxgbi_tagmask_set); MODULE_AUTHOR("Chelsio Communications"); MODULE_DESCRIPTION("Chelsio common library"); -MODULE_VERSION(DRV_VERSION); MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h index 7b02c200dd1e..1b4156461ba1 100644 --- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h +++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h @@ -122,7 +122,7 @@ struct cxgbi_ppm_pool { unsigned int base; /* base index */ unsigned int next; /* next possible free index */ spinlock_t lock; /* ppm pool lock */ - unsigned long bmap[0]; + unsigned long bmap[]; } ____cacheline_aligned_in_smp; struct cxgbi_ppm { @@ -145,7 +145,7 @@ struct cxgbi_ppm { unsigned int next; unsigned int max_index_in_edram; unsigned long *ppod_bmap; - struct cxgbi_ppod_data ppod_data[0]; + struct cxgbi_ppod_data ppod_data[]; }; #define DDP_THRESHOLD 512 diff --git a/drivers/net/ethernet/cirrus/ep93xx_eth.c b/drivers/net/ethernet/cirrus/ep93xx_eth.c index f37c9a08c4cf..9f5e5ec69991 100644 --- a/drivers/net/ethernet/cirrus/ep93xx_eth.c +++ b/drivers/net/ethernet/cirrus/ep93xx_eth.c @@ -24,7 +24,6 @@ #include <linux/platform_data/eth-ep93xx.h> #define DRV_MODULE_NAME "ep93xx-eth" -#define DRV_MODULE_VERSION "0.1" #define RX_QUEUE_ENTRIES 64 #define TX_QUEUE_ENTRIES 8 @@ -691,7 +690,6 @@ static int ep93xx_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) static void ep93xx_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); } static int ep93xx_get_link_ksettings(struct net_device *dev, diff --git a/drivers/net/ethernet/cisco/enic/enic.h b/drivers/net/ethernet/cisco/enic/enic.h index 0dd64acd2a3f..18f3aeb88f22 100644 --- a/drivers/net/ethernet/cisco/enic/enic.h +++ b/drivers/net/ethernet/cisco/enic/enic.h @@ -33,8 +33,6 @@ #define DRV_NAME "enic" #define DRV_DESCRIPTION "Cisco VIC Ethernet NIC Driver" -#define DRV_VERSION "2.3.0.53" -#define DRV_COPYRIGHT "Copyright 2008-2013 Cisco Systems, Inc" #define ENIC_BARS_MAX 6 diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c index ebd5c2cf1efe..4d8e0aa447fb 100644 --- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c +++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c @@ -147,7 +147,6 @@ static void enic_get_drvinfo(struct net_device *netdev, return; strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version)); strlcpy(drvinfo->fw_version, fw_info->fw_version, sizeof(drvinfo->fw_version)); strlcpy(drvinfo->bus_info, pci_name(enic->pdev), @@ -324,25 +323,6 @@ static int enic_coalesce_valid(struct enic *enic, u32 rx_coalesce_usecs_low = min_t(u32, coalesce_usecs_max, ec->rx_coalesce_usecs_low); - if (ec->rx_max_coalesced_frames || - ec->rx_coalesce_usecs_irq || - ec->rx_max_coalesced_frames_irq || - ec->tx_max_coalesced_frames || - ec->tx_coalesce_usecs_irq || - ec->tx_max_coalesced_frames_irq || - ec->stats_block_coalesce_usecs || - ec->use_adaptive_tx_coalesce || - ec->pkt_rate_low || - ec->rx_max_coalesced_frames_low || - ec->tx_coalesce_usecs_low || - ec->tx_max_coalesced_frames_low || - ec->pkt_rate_high || - ec->rx_max_coalesced_frames_high || - ec->tx_coalesce_usecs_high || - ec->tx_max_coalesced_frames_high || - ec->rate_sample_interval) - return -EINVAL; - if ((vnic_dev_get_intr_mode(enic->vdev) != VNIC_DEV_INTR_MODE_MSIX) && ec->tx_coalesce_usecs) return -EINVAL; @@ -636,6 +616,10 @@ static int enic_get_ts_info(struct net_device *netdev, } static const struct ethtool_ops enic_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX | + ETHTOOL_COALESCE_RX_USECS_LOW | + ETHTOOL_COALESCE_RX_USECS_HIGH, .get_drvinfo = enic_get_drvinfo, .get_msglevel = enic_get_msglevel, .set_msglevel = enic_set_msglevel, diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c index ddf60dc9ad16..cd5fe4f6b54c 100644 --- a/drivers/net/ethernet/cisco/enic/enic_main.c +++ b/drivers/net/ethernet/cisco/enic/enic_main.c @@ -80,7 +80,6 @@ static const struct pci_device_id enic_id_table[] = { MODULE_DESCRIPTION(DRV_DESCRIPTION); MODULE_AUTHOR("Scott Feldman <scofeldm@cisco.com>"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); MODULE_DEVICE_TABLE(pci, enic_id_table); #define ENIC_LARGE_PKT_THRESHOLD 1000 @@ -696,8 +695,7 @@ static void enic_preload_tcp_csum(struct sk_buff *skb) tcp_hdr(skb)->check = ~csum_tcpudp_magic(ip_hdr(skb)->saddr, ip_hdr(skb)->daddr, 0, IPPROTO_TCP, 0); } else if (skb->protocol == cpu_to_be16(ETH_P_IPV6)) { - tcp_hdr(skb)->check = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, 0, IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); } } @@ -3056,8 +3054,6 @@ static struct pci_driver enic_driver = { static int __init enic_init_module(void) { - pr_info("%s, ver %s\n", DRV_DESCRIPTION, DRV_VERSION); - return pci_register_driver(&enic_driver); } diff --git a/drivers/net/ethernet/cisco/enic/vnic_devcmd.h b/drivers/net/ethernet/cisco/enic/vnic_devcmd.h index fef5a0a0663d..fcc4a3ccdd94 100644 --- a/drivers/net/ethernet/cisco/enic/vnic_devcmd.h +++ b/drivers/net/ethernet/cisco/enic/vnic_devcmd.h @@ -541,7 +541,7 @@ struct vnic_devcmd_notify { struct vnic_devcmd_provinfo { u8 oui[3]; u8 type; - u8 data[0]; + u8 data[]; }; /* These are used in flags field of different filters to denote @@ -648,9 +648,9 @@ enum { #define FILTER_MAX_BUF_SIZE 100 struct filter_tlv { - u_int32_t type; - u_int32_t length; - u_int32_t val[0]; + u32 type; + u32 length; + u32 val[]; }; enum { diff --git a/drivers/net/ethernet/cisco/enic/vnic_vic.h b/drivers/net/ethernet/cisco/enic/vnic_vic.h index 9ef81f148351..057776908828 100644 --- a/drivers/net/ethernet/cisco/enic/vnic_vic.h +++ b/drivers/net/ethernet/cisco/enic/vnic_vic.h @@ -59,7 +59,7 @@ struct vic_provinfo { u16 type; u16 length; u8 value[0]; - } tlv[0]; + } tlv[]; } __packed; #define VIC_PROVINFO_ADD_TLV(vp, tlvtype, tlvlen, data) \ diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c index f30fa8e6ef80..5bff5c2be88b 100644 --- a/drivers/net/ethernet/cortina/gemini.c +++ b/drivers/net/ethernet/cortina/gemini.c @@ -44,7 +44,6 @@ #include "gemini.h" #define DRV_NAME "gmac-gemini" -#define DRV_VERSION "1.0" #define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK) static int debug = -1; @@ -2204,7 +2203,6 @@ static void gmac_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) { strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); strcpy(info->bus_info, netdev->dev_id ? "1" : "0"); } @@ -2224,6 +2222,8 @@ static const struct net_device_ops gmac_351x_ops = { }; static const struct ethtool_ops gmac_351x_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_sset_count = gmac_get_sset_count, .get_strings = gmac_get_strings, .get_ethtool_stats = gmac_get_ethtool_stats, diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c index e94ae9b94dbf..7f7705138262 100644 --- a/drivers/net/ethernet/davicom/dm9000.c +++ b/drivers/net/ethernet/davicom/dm9000.c @@ -42,7 +42,6 @@ #define DM9000_PHY 0x40 /* PHY address 0x01 */ #define CARDNAME "dm9000" -#define DRV_VERSION "1.31" /* * Transmit timeout, default 5 seconds. @@ -543,7 +542,6 @@ static void dm9000_get_drvinfo(struct net_device *dev, struct board_info *dm = to_dm9000_board(dev); strlcpy(info->driver, CARDNAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, to_platform_device(dm->dev)->name, sizeof(info->bus_info)); } diff --git a/drivers/net/ethernet/dec/tulip/de2104x.c b/drivers/net/ethernet/dec/tulip/de2104x.c index 42b798a3fad4..592454f444ce 100644 --- a/drivers/net/ethernet/dec/tulip/de2104x.c +++ b/drivers/net/ethernet/dec/tulip/de2104x.c @@ -30,7 +30,6 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define DRV_NAME "de2104x" -#define DRV_VERSION "0.7" #define DRV_RELDATE "Mar 17, 2004" #include <linux/module.h> @@ -52,14 +51,9 @@ #include <linux/uaccess.h> #include <asm/unaligned.h> -/* These identify the driver base version and may not be removed. */ -static char version[] = -"PCI Ethernet driver v" DRV_VERSION " (" DRV_RELDATE ")"; - MODULE_AUTHOR("Jeff Garzik <jgarzik@pobox.com>"); MODULE_DESCRIPTION("Intel/Digital 21040/1 series PCI Ethernet driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); static int debug = -1; module_param (debug, int, 0); @@ -1603,7 +1597,6 @@ static void de_get_drvinfo (struct net_device *dev,struct ethtool_drvinfo *info) struct de_private *de = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(de->pdev), sizeof(info->bus_info)); } @@ -1980,11 +1973,6 @@ static int de_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) board_idx++; -#ifndef MODULE - if (board_idx == 0) - pr_info("%s\n", version); -#endif - /* allocate a new ethernet device structure, and fill in defaults */ dev = alloc_etherdev(sizeof(struct de_private)); if (!dev) @@ -2196,9 +2184,6 @@ static struct pci_driver de_driver = { static int __init de_init (void) { -#ifdef MODULE - pr_info("%s\n", version); -#endif return pci_register_driver(&de_driver); } diff --git a/drivers/net/ethernet/dec/tulip/dmfe.c b/drivers/net/ethernet/dec/tulip/dmfe.c index 32d470d4122a..c1884fc9ad32 100644 --- a/drivers/net/ethernet/dec/tulip/dmfe.c +++ b/drivers/net/ethernet/dec/tulip/dmfe.c @@ -56,8 +56,6 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define DRV_NAME "dmfe" -#define DRV_VERSION "1.36.4" -#define DRV_RELDATE "2002-01-17" #include <linux/module.h> #include <linux/kernel.h> @@ -280,10 +278,6 @@ enum dmfe_CR6_bits { }; /* Global variable declaration ----------------------------- */ -static int printed_version; -static const char version[] = - "Davicom DM9xxx net driver, version " DRV_VERSION " (" DRV_RELDATE ")"; - static int dmfe_debug; static unsigned char dmfe_media_mode = DMFE_AUTO; static u32 dmfe_cr6_user_set; @@ -364,9 +358,6 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) DMFE_DBUG(0, "dmfe_init_one()", 0); - if (!printed_version++) - pr_info("%s\n", version); - /* * SPARC on-board DM910x chips should be handled by the main * tulip driver, except for early DM9100s. @@ -1081,7 +1072,6 @@ static void dmfe_ethtool_get_drvinfo(struct net_device *dev, struct dmfe_board_info *np = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info)); } @@ -2177,7 +2167,6 @@ static struct pci_driver dmfe_driver = { MODULE_AUTHOR("Sten Wang, sten_wang@davicom.com.tw"); MODULE_DESCRIPTION("Davicom DM910X fast ethernet driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); module_param(debug, int, 0); module_param(mode, byte, 0); @@ -2204,9 +2193,6 @@ static int __init dmfe_init_module(void) { int rc; - pr_info("%s\n", version); - printed_version = 1; - DMFE_DBUG(0, "init_module() ", debug); if (debug) diff --git a/drivers/net/ethernet/dec/tulip/tulip.h b/drivers/net/ethernet/dec/tulip/tulip.h index b458140aeaef..815907259048 100644 --- a/drivers/net/ethernet/dec/tulip/tulip.h +++ b/drivers/net/ethernet/dec/tulip/tulip.h @@ -381,7 +381,7 @@ struct mediatable { unsigned has_reset:6; u32 csr15dir; u32 csr15val; /* 21143 NWay setting. */ - struct medialeaf mleaf[0]; + struct medialeaf mleaf[]; }; diff --git a/drivers/net/ethernet/dec/tulip/tulip_core.c b/drivers/net/ethernet/dec/tulip/tulip_core.c index 9e9d9eee29d9..48ea658aa1a6 100644 --- a/drivers/net/ethernet/dec/tulip/tulip_core.c +++ b/drivers/net/ethernet/dec/tulip/tulip_core.c @@ -12,13 +12,6 @@ #define pr_fmt(fmt) "tulip: " fmt #define DRV_NAME "tulip" -#ifdef CONFIG_TULIP_NAPI -#define DRV_VERSION "1.1.15-NAPI" /* Keep at least for test */ -#else -#define DRV_VERSION "1.1.15" -#endif -#define DRV_RELDATE "Feb 27, 2007" - #include <linux/module.h> #include <linux/pci.h> @@ -37,9 +30,6 @@ #include <asm/prom.h> #endif -static char version[] = - "Linux Tulip driver version " DRV_VERSION " (" DRV_RELDATE ")\n"; - /* A few user-configurable values. */ /* Maximum events (Rx packets, etc.) to handle at each interrupt. */ @@ -109,7 +99,6 @@ static int csr0; MODULE_AUTHOR("The Linux Kernel Team"); MODULE_DESCRIPTION("Digital 21*4* Tulip ethernet driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); module_param(tulip_debug, int, 0); module_param(max_interrupt_work, int, 0); module_param(rx_copybreak, int, 0); @@ -868,7 +857,6 @@ static void tulip_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *in { struct tulip_private *np = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info)); } @@ -1314,11 +1302,6 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) unsigned int eeprom_missing = 0; unsigned int force_csr0 = 0; -#ifndef MODULE - if (tulip_debug > 0) - printk_once(KERN_INFO "%s", version); -#endif - board_idx++; /* @@ -1800,14 +1783,13 @@ static void tulip_set_wolopts (struct pci_dev *pdev, u32 wolopts) void __iomem *ioaddr = tp->base_addr; if (tp->flags & COMET_PM) { - unsigned int tmp; - + tmp = ioread32(ioaddr + CSR18); tmp &= ~(comet_csr18_pmes_sticky | comet_csr18_apm_mode | comet_csr18_d3a); tmp |= comet_csr18_pm_mode; iowrite32(tmp, ioaddr + CSR18); - + /* Set the Wake-up Control/Status Register to the given WOL options*/ tmp = ioread32(ioaddr + CSR13); tmp &= ~(comet_csr13_linkoffe | comet_csr13_linkone | comet_csr13_wfre | comet_csr13_lsce | comet_csr13_mpre); @@ -1969,10 +1951,6 @@ static struct pci_driver tulip_driver = { static int __init tulip_init (void) { -#ifdef MODULE - pr_info("%s", version); -#endif - if (!csr0) { pr_warn("tulip: unknown CPU architecture, using default csr0\n"); /* default to 8 longword cache line alignment */ diff --git a/drivers/net/ethernet/dec/tulip/uli526x.c b/drivers/net/ethernet/dec/tulip/uli526x.c index 117ffe08800d..f726436b1985 100644 --- a/drivers/net/ethernet/dec/tulip/uli526x.c +++ b/drivers/net/ethernet/dec/tulip/uli526x.c @@ -7,8 +7,6 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define DRV_NAME "uli526x" -#define DRV_VERSION "0.9.3" -#define DRV_RELDATE "2005-7-29" #include <linux/module.h> @@ -196,10 +194,6 @@ enum uli526x_CR6_bits { }; /* Global variable declaration ----------------------------- */ -static int printed_version; -static const char version[] = - "ULi M5261/M5263 net driver, version " DRV_VERSION " (" DRV_RELDATE ")"; - static int uli526x_debug; static unsigned char uli526x_media_mode = ULI526X_AUTO; static u32 uli526x_cr6_user_set; @@ -282,9 +276,6 @@ static int uli526x_init_one(struct pci_dev *pdev, ULI526X_DBUG(0, "uli526x_init_one()", 0); - if (!printed_version++) - pr_info("%s\n", version); - /* Init network device */ dev = alloc_etherdev(sizeof(*db)); if (dev == NULL) @@ -972,7 +963,6 @@ static void netdev_get_drvinfo(struct net_device *dev, struct uli526x_board_info *np = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info)); } @@ -1799,9 +1789,6 @@ MODULE_PARM_DESC(mode, "ULi M5261/M5263: Bit 0: 10/100Mbps, bit 2: duplex, bit 8 static int __init uli526x_init_module(void) { - pr_info("%s\n", version); - printed_version = 1; - ULI526X_DBUG(0, "init_module() ", debug); if (debug) diff --git a/drivers/net/ethernet/dec/tulip/winbond-840.c b/drivers/net/ethernet/dec/tulip/winbond-840.c index 7f136488e67c..4d5e4fa53023 100644 --- a/drivers/net/ethernet/dec/tulip/winbond-840.c +++ b/drivers/net/ethernet/dec/tulip/winbond-840.c @@ -47,9 +47,6 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define DRV_NAME "winbond-840" -#define DRV_VERSION "1.01-e" -#define DRV_RELDATE "Sep-11-2006" - /* Automatically extracted configuration info: probe-func: winbond840_probe @@ -139,16 +136,9 @@ static int full_duplex[MAX_UNITS] = {-1, -1, -1, -1, -1, -1, -1, -1}; #undef PKT_BUF_SZ /* tulip.h also defines this */ #define PKT_BUF_SZ 1536 /* Size of each temporary Rx buffer.*/ -/* These identify the driver base version and may not be removed. */ -static const char version[] __initconst = - "v" DRV_VERSION " (2.4 port) " - DRV_RELDATE " Donald Becker <becker@scyld.com>\n" - " http://www.scyld.com/network/drivers.html\n"; - MODULE_AUTHOR("Donald Becker <becker@scyld.com>"); MODULE_DESCRIPTION("Winbond W89c840 Ethernet driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_VERSION); module_param(max_interrupt_work, int, 0); module_param(debug, int, 0); @@ -1385,7 +1375,6 @@ static void netdev_get_drvinfo (struct net_device *dev, struct ethtool_drvinfo * struct netdev_private *np = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } @@ -1650,7 +1639,6 @@ static struct pci_driver w840_driver = { static int __init w840_init(void) { - printk(version); return pci_register_driver(&w840_driver); } diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c index 26c5da032b1e..643090555cc7 100644 --- a/drivers/net/ethernet/dlink/dl2k.c +++ b/drivers/net/ethernet/dlink/dl2k.c @@ -8,8 +8,6 @@ */ #define DRV_NAME "DL2000/TC902x-based linux driver" -#define DRV_VERSION "v1.19" -#define DRV_RELDATE "2007/08/12" #include "dl2k.h" #include <linux/dma-mapping.h> @@ -20,8 +18,6 @@ #define dr16(reg) ioread16(ioaddr + (reg)) #define dr8(reg) ioread8(ioaddr + (reg)) -static char version[] = - KERN_INFO DRV_NAME " " DRV_VERSION " " DRV_RELDATE "\n"; #define MAX_UNITS 8 static int mtu[MAX_UNITS]; static int vlan[MAX_UNITS]; @@ -113,13 +109,9 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent) int chip_idx = ent->driver_data; int err, irq; void __iomem *ioaddr; - static int version_printed; void *ring_space; dma_addr_t ring_dma; - if (!version_printed++) - printk ("%s", version); - err = pci_enable_device (pdev); if (err) return err; @@ -1244,7 +1236,6 @@ static void rio_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info struct netdev_private *np = netdev_priv(dev); strlcpy(info->driver, "dl2k", sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info)); } diff --git a/drivers/net/ethernet/dlink/sundance.c b/drivers/net/ethernet/dlink/sundance.c index b91387c456ba..dc566fcc3ba9 100644 --- a/drivers/net/ethernet/dlink/sundance.c +++ b/drivers/net/ethernet/dlink/sundance.c @@ -23,9 +23,6 @@ */ #define DRV_NAME "sundance" -#define DRV_VERSION "1.2" -#define DRV_RELDATE "11-Sep-2006" - /* The user-configurable values. These may be modified when a driver module is loaded.*/ @@ -101,11 +98,6 @@ static char *media[MAX_UNITS]; #include <linux/ethtool.h> #include <linux/mii.h> -/* These identify the driver base version and may not be removed. */ -static const char version[] = - KERN_INFO DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE - " Written by Donald Becker\n"; - MODULE_AUTHOR("Donald Becker <becker@scyld.com>"); MODULE_DESCRIPTION("Sundance Alta Ethernet driver"); MODULE_LICENSE("GPL"); @@ -516,13 +508,6 @@ static int sundance_probe1(struct pci_dev *pdev, #endif int phy, phy_end, phy_idx = 0; -/* when built into the kernel, we only print version if device is found */ -#ifndef MODULE - static int printed_version; - if (!printed_version++) - printk(version); -#endif - if (pci_enable_device(pdev)) return -EIO; pci_set_master(pdev); @@ -1657,7 +1642,6 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct netdev_private *np = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } @@ -2010,10 +1994,6 @@ static struct pci_driver sundance_driver = { static int __init sundance_init(void) { -/* when a module, this is printed whether or not devices are found in probe */ -#ifdef MODULE - printk(version); -#endif return pci_register_driver(&sundance_driver); } diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index 5f8fa1145db6..057a508dd6e2 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -729,7 +729,6 @@ static void dnet_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, "0", sizeof(info->bus_info)); } diff --git a/drivers/net/ethernet/dnet.h b/drivers/net/ethernet/dnet.h index 8af6c0705ab3..030724484b49 100644 --- a/drivers/net/ethernet/dnet.h +++ b/drivers/net/ethernet/dnet.h @@ -8,7 +8,6 @@ #define _DNET_H #define DRV_NAME "dnet" -#define DRV_VERSION "0.9.1" #define PFX DRV_NAME ": " /* Register access macros */ diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index cf3e6f2892ff..6e9022083004 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -33,7 +33,6 @@ #include "be_hw.h" #include "be_roce.h" -#define DRV_VER "12.0.0.0" #define DRV_NAME "be2net" #define BE_NAME "Emulex BladeEngine2" #define BE3_NAME "Emulex BladeEngine3" diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c index 022a54a1805b..d6ed1d943762 100644 --- a/drivers/net/ethernet/emulex/benet/be_ethtool.c +++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c @@ -221,7 +221,6 @@ static void be_get_drvinfo(struct net_device *netdev, struct be_adapter *adapter = netdev_priv(netdev); strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, DRV_VER, sizeof(drvinfo->version)); if (!memcmp(adapter->fw_ver, adapter->fw_on_flash, FW_VER_LEN)) strlcpy(drvinfo->fw_version, adapter->fw_ver, sizeof(drvinfo->fw_version)); @@ -1409,6 +1408,9 @@ static int be_set_priv_flags(struct net_device *netdev, u32 flags) } const struct ethtool_ops be_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_USE_ADAPTIVE | + ETHTOOL_COALESCE_USECS_LOW_HIGH, .get_drvinfo = be_get_drvinfo, .get_wol = be_get_wol, .set_wol = be_set_wol, diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 56f59db6ebf2..a7ac23a6862b 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -21,8 +21,7 @@ #include <net/busy_poll.h> #include <net/vxlan.h> -MODULE_VERSION(DRV_VER); -MODULE_DESCRIPTION(DRV_DESC " " DRV_VER); +MODULE_DESCRIPTION(DRV_DESC); MODULE_AUTHOR("Emulex Corporation"); MODULE_LICENSE("GPL"); @@ -5949,8 +5948,6 @@ static int be_probe(struct pci_dev *pdev, const struct pci_device_id *pdev_id) struct net_device *netdev; int status = 0; - dev_info(&pdev->dev, "%s version is %s\n", DRV_NAME, DRV_VER); - status = pci_enable_device(pdev); if (status) goto do_none; diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index 4572797f00d7..835b7816e372 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -30,7 +30,6 @@ #include "ftgmac100.h" #define DRV_NAME "ftgmac100" -#define DRV_VERSION "0.7" /* Arbitrary values, I am not sure the HW has limits */ #define MAX_RX_QUEUE_ENTRIES 1024 @@ -1150,7 +1149,6 @@ static void ftgmac100_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, dev_name(&netdev->dev), sizeof(info->bus_info)); } @@ -1757,9 +1755,6 @@ static int ftgmac100_probe(struct platform_device *pdev) struct device_node *np; int err = 0; - if (!pdev) - return -ENODEV; - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!res) return -ENXIO; diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c index 6c247cbbd23e..32cf54f0e35b 100644 --- a/drivers/net/ethernet/faraday/ftmac100.c +++ b/drivers/net/ethernet/faraday/ftmac100.c @@ -23,7 +23,6 @@ #include "ftmac100.h" #define DRV_NAME "ftmac100" -#define DRV_VERSION "0.2" #define RX_QUEUE_ENTRIES 128 /* must be power of 2 */ #define TX_QUEUE_ENTRIES 16 /* must be power of 2 */ @@ -809,7 +808,6 @@ static void ftmac100_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, dev_name(&netdev->dev), sizeof(info->bus_info)); } @@ -1184,7 +1182,6 @@ static struct platform_driver ftmac100_driver = { *****************************************************************************/ static int __init ftmac100_init(void) { - pr_info("Loading version " DRV_VERSION " ...\n"); return platform_driver_register(&ftmac100_driver); } diff --git a/drivers/net/ethernet/fealnx.c b/drivers/net/ethernet/fealnx.c index 84f10970299a..73e896a7d8fd 100644 --- a/drivers/net/ethernet/fealnx.c +++ b/drivers/net/ethernet/fealnx.c @@ -25,8 +25,6 @@ */ #define DRV_NAME "fealnx" -#define DRV_VERSION "2.52" -#define DRV_RELDATE "Sep-11-2006" static int debug; /* 1-> print debug message */ static int max_interrupt_work = 20; @@ -91,11 +89,6 @@ static int full_duplex[MAX_UNITS] = { -1, -1, -1, -1, -1, -1, -1, -1 }; #include <linux/uaccess.h> #include <asm/byteorder.h> -/* These identify the driver base version and may not be removed. */ -static const char version[] = - KERN_INFO DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE "\n"; - - /* This driver was written to use PCI memory space, however some x86 systems work only with I/O space accesses. */ #ifndef __alpha__ @@ -495,13 +488,6 @@ static int fealnx_init_one(struct pci_dev *pdev, int bar = 1; #endif -/* when built into the kernel, we only print version if device is found */ -#ifndef MODULE - static int printed_version; - if (!printed_version++) - printk(version); -#endif - card_idx++; sprintf(boardname, "fealnx%d", card_idx); @@ -1809,7 +1795,6 @@ static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *i struct netdev_private *np = netdev_priv(dev); strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } @@ -1950,11 +1935,6 @@ static struct pci_driver fealnx_driver = { static int __init fealnx_init(void) { -/* when a module, this is printed whether or not devices are found in probe */ -#ifdef MODULE - printk(version); -#endif - return pci_register_driver(&fealnx_driver); } diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index ca74a684a904..2cd1f8efdfa3 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -259,8 +259,20 @@ static int dpaa_netdev_init(struct net_device *net_dev, net_dev->features |= net_dev->hw_features; net_dev->vlan_features = net_dev->features; - memcpy(net_dev->perm_addr, mac_addr, net_dev->addr_len); - memcpy(net_dev->dev_addr, mac_addr, net_dev->addr_len); + if (is_valid_ether_addr(mac_addr)) { + memcpy(net_dev->perm_addr, mac_addr, net_dev->addr_len); + memcpy(net_dev->dev_addr, mac_addr, net_dev->addr_len); + } else { + eth_hw_addr_random(net_dev); + err = priv->mac_dev->change_addr(priv->mac_dev->fman_mac, + (enet_addr_t *)net_dev->dev_addr); + if (err) { + dev_err(dev, "Failed to set random MAC address\n"); + return -EINVAL; + } + dev_info(dev, "Using random MAC address: %pM\n", + net_dev->dev_addr); + } net_dev->ethtool_ops = &dpaa_ethtool_ops; @@ -2050,7 +2062,7 @@ static inline int dpaa_xmit(struct dpaa_priv *priv, } #ifdef CONFIG_DPAA_ERRATUM_A050385 -int dpaa_a050385_wa(struct net_device *net_dev, struct sk_buff **s) +static int dpaa_a050385_wa(struct net_device *net_dev, struct sk_buff **s) { struct dpaa_priv *priv = netdev_priv(net_dev); struct sk_buff *new_skb, *skb = *s; diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c index 66d150872d48..9db2a02fb531 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c @@ -106,19 +106,8 @@ static int dpaa_set_link_ksettings(struct net_device *net_dev, static void dpaa_get_drvinfo(struct net_device *net_dev, struct ethtool_drvinfo *drvinfo) { - int len; - strlcpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver)); - len = snprintf(drvinfo->version, sizeof(drvinfo->version), - "%X", 0); - len = snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), - "%X", 0); - - if (len >= sizeof(drvinfo->fw_version)) { - /* Truncated output */ - netdev_notice(net_dev, "snprintf() = %d\n", len); - } strlcpy(drvinfo->bus_info, dev_name(net_dev->dev.parent->parent), sizeof(drvinfo->bus_info)); } @@ -536,7 +525,6 @@ static int dpaa_get_coalesce(struct net_device *dev, c->rx_coalesce_usecs = period; c->rx_max_coalesced_frames = thresh; - c->use_adaptive_rx_coalesce = false; return 0; } @@ -551,9 +539,6 @@ static int dpaa_set_coalesce(struct net_device *dev, u8 thresh, prev_thresh; int cpu, res; - if (c->use_adaptive_rx_coalesce) - return -EINVAL; - period = c->rx_coalesce_usecs; thresh = c->rx_max_coalesced_frames; @@ -593,6 +578,8 @@ revert_values: } const struct ethtool_ops dpaa_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_RX_MAX_FRAMES, .get_drvinfo = dpaa_get_drvinfo, .get_msglevel = dpaa_get_msglevel, .set_msglevel = dpaa_set_msglevel, diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c index 7ff147e89426..b6c46639aa4c 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c @@ -1704,10 +1704,15 @@ static int dpaa2_eth_ts_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) static int dpaa2_eth_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { + struct dpaa2_eth_priv *priv = netdev_priv(dev); + if (cmd == SIOCSHWTSTAMP) return dpaa2_eth_ts_ioctl(dev, rq, cmd); - return -EINVAL; + if (priv->mac) + return phylink_mii_ioctl(priv->mac->phylink, rq, cmd); + + return -EOPNOTSUPP; } static bool xdp_mtu_valid(struct dpaa2_eth_priv *priv, int mtu) diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c index 96676abcebd5..94347c695233 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c @@ -79,6 +79,16 @@ static void dpaa2_eth_get_drvinfo(struct net_device *net_dev, sizeof(drvinfo->bus_info)); } +static int dpaa2_eth_nway_reset(struct net_device *net_dev) +{ + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + + if (priv->mac) + return phylink_ethtool_nway_reset(priv->mac->phylink); + + return -EOPNOTSUPP; +} + static int dpaa2_eth_get_link_ksettings(struct net_device *net_dev, struct ethtool_link_ksettings *link_settings) @@ -761,6 +771,7 @@ static int dpaa2_eth_get_ts_info(struct net_device *dev, const struct ethtool_ops dpaa2_ethtool_ops = { .get_drvinfo = dpaa2_eth_get_drvinfo, + .nway_reset = dpaa2_eth_nway_reset, .get_link = ethtool_op_get_link, .get_link_ksettings = dpaa2_eth_get_link_ksettings, .set_link_ksettings = dpaa2_eth_set_link_ksettings, diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c index 84233e467ed1..3ee236c5fc37 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c @@ -123,49 +123,60 @@ static void dpaa2_mac_config(struct phylink_config *config, unsigned int mode, struct dpmac_link_state *dpmac_state = &mac->state; int err; - if (state->speed != SPEED_UNKNOWN) - dpmac_state->rate = state->speed; - - if (state->duplex != DUPLEX_UNKNOWN) { - if (!state->duplex) - dpmac_state->options |= DPMAC_LINK_OPT_HALF_DUPLEX; - else - dpmac_state->options &= ~DPMAC_LINK_OPT_HALF_DUPLEX; - } - if (state->an_enabled) dpmac_state->options |= DPMAC_LINK_OPT_AUTONEG; else dpmac_state->options &= ~DPMAC_LINK_OPT_AUTONEG; - if (state->pause & MLO_PAUSE_RX) - dpmac_state->options |= DPMAC_LINK_OPT_PAUSE; - else - dpmac_state->options &= ~DPMAC_LINK_OPT_PAUSE; - - if (!!(state->pause & MLO_PAUSE_RX) ^ !!(state->pause & MLO_PAUSE_TX)) - dpmac_state->options |= DPMAC_LINK_OPT_ASYM_PAUSE; - else - dpmac_state->options &= ~DPMAC_LINK_OPT_ASYM_PAUSE; - err = dpmac_set_link_state(mac->mc_io, 0, mac->mc_dev->mc_handle, dpmac_state); if (err) - netdev_err(mac->net_dev, "dpmac_set_link_state() = %d\n", err); + netdev_err(mac->net_dev, "%s: dpmac_set_link_state() = %d\n", + __func__, err); } -static void dpaa2_mac_link_up(struct phylink_config *config, unsigned int mode, - phy_interface_t interface, struct phy_device *phy) +static void dpaa2_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause) { struct dpaa2_mac *mac = phylink_to_dpaa2_mac(config); struct dpmac_link_state *dpmac_state = &mac->state; int err; dpmac_state->up = 1; + + if (mac->if_link_type == DPMAC_LINK_TYPE_PHY) { + /* If the DPMAC is configured for PHY mode, we need + * to pass the link parameters to the MC firmware. + */ + dpmac_state->rate = speed; + + if (duplex == DUPLEX_HALF) + dpmac_state->options |= DPMAC_LINK_OPT_HALF_DUPLEX; + else if (duplex == DUPLEX_FULL) + dpmac_state->options &= ~DPMAC_LINK_OPT_HALF_DUPLEX; + + /* This is lossy; the firmware really should take the pause + * enablement status rather than pause/asym pause status. + */ + if (rx_pause) + dpmac_state->options |= DPMAC_LINK_OPT_PAUSE; + else + dpmac_state->options &= ~DPMAC_LINK_OPT_PAUSE; + + if (rx_pause ^ tx_pause) + dpmac_state->options |= DPMAC_LINK_OPT_ASYM_PAUSE; + else + dpmac_state->options &= ~DPMAC_LINK_OPT_ASYM_PAUSE; + } + err = dpmac_set_link_state(mac->mc_io, 0, mac->mc_dev->mc_handle, dpmac_state); if (err) - netdev_err(mac->net_dev, "dpmac_set_link_state() = %d\n", err); + netdev_err(mac->net_dev, "%s: dpmac_set_link_state() = %d\n", + __func__, err); } static void dpaa2_mac_link_down(struct phylink_config *config, @@ -238,6 +249,8 @@ int dpaa2_mac_connect(struct dpaa2_mac *mac) goto err_close_dpmac; } + mac->if_link_type = attr.link_type; + dpmac_node = dpaa2_mac_get_node(attr.id); if (!dpmac_node) { netdev_err(net_dev, "No dpmac@%d node found.\n", attr.id); diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h index 4da8079b9155..2130d9c7d40e 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h @@ -20,6 +20,7 @@ struct dpaa2_mac { struct phylink_config phylink_config; struct phylink *phylink; phy_interface_t if_mode; + enum dpmac_link_type if_link_type; }; bool dpaa2_mac_is_type_fixed(struct fsl_mc_device *dpmac_dev, diff --git a/drivers/net/ethernet/freescale/enetc/Kconfig b/drivers/net/ethernet/freescale/enetc/Kconfig index fe942de19597..2b43848e1363 100644 --- a/drivers/net/ethernet/freescale/enetc/Kconfig +++ b/drivers/net/ethernet/freescale/enetc/Kconfig @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0 config FSL_ENETC tristate "ENETC PF driver" - depends on PCI && PCI_MSI && (ARCH_LAYERSCAPE || COMPILE_TEST) + depends on PCI && PCI_MSI select FSL_ENETC_MDIO select PHYLIB help @@ -13,7 +13,7 @@ config FSL_ENETC config FSL_ENETC_VF tristate "ENETC VF driver" - depends on PCI && PCI_MSI && (ARCH_LAYERSCAPE || COMPILE_TEST) + depends on PCI && PCI_MSI select PHYLIB help This driver supports NXP ENETC gigabit ethernet controller PCIe @@ -23,7 +23,7 @@ config FSL_ENETC_VF config FSL_ENETC_MDIO tristate "ENETC MDIO driver" - depends on PCI && (ARCH_LAYERSCAPE || COMPILE_TEST) + depends on PCI help This driver supports NXP ENETC Central MDIO controller as a PCIe physical function (PF) device. @@ -42,16 +42,6 @@ config FSL_ENETC_PTP_CLOCK If compiled as module (M), the module name is fsl-enetc-ptp. -config FSL_ENETC_HW_TIMESTAMPING - bool "ENETC hardware timestamping support" - depends on FSL_ENETC || FSL_ENETC_VF - help - Enable hardware timestamping support on the Ethernet packets - using the SO_TIMESTAMPING API. Because the RX BD ring dynamic - allocation has not been supported and it is too expensive to use - extended RX BDs if timestamping is not used, this option enables - extended RX BDs in order to support hardware timestamping. - config FSL_ENETC_QOS bool "ENETC hardware Time-sensitive Network support" depends on (FSL_ENETC || FSL_ENETC_VF) && (NET_SCH_TAPRIO || NET_SCH_CBS) diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index 1f79e36116a3..ccf2611f4a20 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -451,7 +451,7 @@ static int enetc_refill_rx_ring(struct enetc_bdr *rx_ring, const int buff_cnt) i = rx_ring->next_to_use; rx_swbd = &rx_ring->rx_swbd[i]; - rxbd = ENETC_RXBD(*rx_ring, i); + rxbd = enetc_rxbd(rx_ring, i); for (j = 0; j < buff_cnt; j++) { /* try reuse page */ @@ -468,13 +468,12 @@ static int enetc_refill_rx_ring(struct enetc_bdr *rx_ring, const int buff_cnt) /* clear 'R" as well */ rxbd->r.lstatus = 0; + rxbd = enetc_rxbd_next(rx_ring, rxbd, i); rx_swbd++; - rxbd++; i++; if (unlikely(i == rx_ring->bd_count)) { i = 0; rx_swbd = rx_ring->rx_swbd; - rxbd = ENETC_RXBD(*rx_ring, 0); } } @@ -488,7 +487,7 @@ static int enetc_refill_rx_ring(struct enetc_bdr *rx_ring, const int buff_cnt) return j; } -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING +#ifdef CONFIG_FSL_ENETC_PTP_CLOCK static void enetc_get_rx_tstamp(struct net_device *ndev, union enetc_rx_bd *rxbd, struct sk_buff *skb) @@ -502,7 +501,8 @@ static void enetc_get_rx_tstamp(struct net_device *ndev, if (le16_to_cpu(rxbd->r.flags) & ENETC_RXBD_FLAG_TSTMP) { lo = enetc_rd(hw, ENETC_SICTR0); hi = enetc_rd(hw, ENETC_SICTR1); - tstamp_lo = le32_to_cpu(rxbd->r.tstamp); + rxbd = enetc_rxbd_ext(rxbd); + tstamp_lo = le32_to_cpu(rxbd->ext.tstamp); if (lo <= tstamp_lo) hi -= 1; @@ -516,7 +516,7 @@ static void enetc_get_rx_tstamp(struct net_device *ndev, static void enetc_get_offloads(struct enetc_bdr *rx_ring, union enetc_rx_bd *rxbd, struct sk_buff *skb) { -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING +#ifdef CONFIG_FSL_ENETC_PTP_CLOCK struct enetc_ndev_priv *priv = netdev_priv(rx_ring->ndev); #endif /* TODO: hashing */ @@ -533,7 +533,7 @@ static void enetc_get_offloads(struct enetc_bdr *rx_ring, if (le16_to_cpu(rxbd->r.flags) & ENETC_RXBD_FLAG_VLAN) __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), le16_to_cpu(rxbd->r.vlan_opt)); -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING +#ifdef CONFIG_FSL_ENETC_PTP_CLOCK if (priv->active_offloads & ENETC_F_RX_TSTAMP) enetc_get_rx_tstamp(rx_ring->ndev, rxbd, skb); #endif @@ -655,7 +655,7 @@ static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring, cleaned_cnt -= count; } - rxbd = ENETC_RXBD(*rx_ring, i); + rxbd = enetc_rxbd(rx_ring, i); bd_status = le32_to_cpu(rxbd->r.lstatus); if (!bd_status) break; @@ -670,12 +670,10 @@ static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring, enetc_get_offloads(rx_ring, rxbd, skb); cleaned_cnt++; - rxbd++; - i++; - if (unlikely(i == rx_ring->bd_count)) { + + rxbd = enetc_rxbd_next(rx_ring, rxbd, i); + if (unlikely(++i == rx_ring->bd_count)) i = 0; - rxbd = ENETC_RXBD(*rx_ring, 0); - } if (unlikely(bd_status & ENETC_RXBD_LSTATUS(ENETC_RXBD_ERR_MASK))) { @@ -683,12 +681,10 @@ static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring, while (!(bd_status & ENETC_RXBD_LSTATUS_F)) { dma_rmb(); bd_status = le32_to_cpu(rxbd->r.lstatus); - rxbd++; - i++; - if (unlikely(i == rx_ring->bd_count)) { + + rxbd = enetc_rxbd_next(rx_ring, rxbd, i); + if (unlikely(++i == rx_ring->bd_count)) i = 0; - rxbd = ENETC_RXBD(*rx_ring, 0); - } } rx_ring->ndev->stats.rx_dropped++; @@ -710,12 +706,10 @@ static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring, enetc_add_rx_buff_to_skb(rx_ring, i, size, skb); cleaned_cnt++; - rxbd++; - i++; - if (unlikely(i == rx_ring->bd_count)) { + + rxbd = enetc_rxbd_next(rx_ring, rxbd, i); + if (unlikely(++i == rx_ring->bd_count)) i = 0; - rxbd = ENETC_RXBD(*rx_ring, 0); - } } rx_byte_cnt += skb->len; @@ -845,15 +839,19 @@ static void enetc_free_tx_resources(struct enetc_ndev_priv *priv) enetc_free_txbdr(priv->tx_ring[i]); } -static int enetc_alloc_rxbdr(struct enetc_bdr *rxr) +static int enetc_alloc_rxbdr(struct enetc_bdr *rxr, bool extended) { + size_t size = sizeof(union enetc_rx_bd); int err; rxr->rx_swbd = vzalloc(rxr->bd_count * sizeof(struct enetc_rx_swbd)); if (!rxr->rx_swbd) return -ENOMEM; - err = enetc_dma_alloc_bdr(rxr, sizeof(union enetc_rx_bd)); + if (extended) + size *= 2; + + err = enetc_dma_alloc_bdr(rxr, size); if (err) { vfree(rxr->rx_swbd); return err; @@ -862,6 +860,7 @@ static int enetc_alloc_rxbdr(struct enetc_bdr *rxr) rxr->next_to_clean = 0; rxr->next_to_use = 0; rxr->next_to_alloc = 0; + rxr->ext_en = extended; return 0; } @@ -881,10 +880,11 @@ static void enetc_free_rxbdr(struct enetc_bdr *rxr) static int enetc_alloc_rx_resources(struct enetc_ndev_priv *priv) { + bool extended = !!(priv->active_offloads & ENETC_F_RX_TSTAMP); int i, err; for (i = 0; i < priv->num_rx_rings; i++) { - err = enetc_alloc_rxbdr(priv->rx_ring[i]); + err = enetc_alloc_rxbdr(priv->rx_ring[i], extended); if (err) goto fail; @@ -1174,9 +1174,10 @@ static void enetc_setup_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring) enetc_rxbdr_wr(hw, idx, ENETC_RBICIR0, ENETC_RBICIR0_ICEN | 0x1); rbmr = ENETC_RBMR_EN; -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING - rbmr |= ENETC_RBMR_BDS; -#endif + + if (rx_ring->ext_en) + rbmr |= ENETC_RBMR_BDS; + if (rx_ring->ndev->features & NETIF_F_HW_VLAN_CTAG_RX) rbmr |= ENETC_RBMR_VTE; @@ -1577,11 +1578,12 @@ int enetc_set_features(struct net_device *ndev, return 0; } -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING +#ifdef CONFIG_FSL_ENETC_PTP_CLOCK static int enetc_hwtstamp_set(struct net_device *ndev, struct ifreq *ifr) { struct enetc_ndev_priv *priv = netdev_priv(ndev); struct hwtstamp_config config; + int ao; if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) return -EFAULT; @@ -1597,6 +1599,7 @@ static int enetc_hwtstamp_set(struct net_device *ndev, struct ifreq *ifr) return -ERANGE; } + ao = priv->active_offloads; switch (config.rx_filter) { case HWTSTAMP_FILTER_NONE: priv->active_offloads &= ~ENETC_F_RX_TSTAMP; @@ -1606,6 +1609,11 @@ static int enetc_hwtstamp_set(struct net_device *ndev, struct ifreq *ifr) config.rx_filter = HWTSTAMP_FILTER_ALL; } + if (netif_running(ndev) && ao != priv->active_offloads) { + enetc_close(ndev); + enetc_open(ndev); + } + return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? -EFAULT : 0; } @@ -1632,7 +1640,7 @@ static int enetc_hwtstamp_get(struct net_device *ndev, struct ifreq *ifr) int enetc_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd) { -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING +#ifdef CONFIG_FSL_ENETC_PTP_CLOCK if (cmd == SIOCSHWTSTAMP) return enetc_hwtstamp_set(ndev, rq); if (cmd == SIOCGHWTSTAMP) diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h index dd4a227ffc7a..56c43f35b633 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.h +++ b/drivers/net/ethernet/freescale/enetc/enetc.h @@ -73,6 +73,7 @@ struct enetc_bdr { dma_addr_t bd_dma_base; u8 tsd_enable; /* Time specific departure */ + bool ext_en; /* enable h/w descriptor extensions */ } ____cacheline_aligned_in_smp; static inline void enetc_bdr_idx_inc(struct enetc_bdr *bdr, int *i) @@ -104,7 +105,37 @@ struct enetc_cbdr { }; #define ENETC_TXBD(BDR, i) (&(((union enetc_tx_bd *)((BDR).bd_base))[i])) -#define ENETC_RXBD(BDR, i) (&(((union enetc_rx_bd *)((BDR).bd_base))[i])) + +static inline union enetc_rx_bd *enetc_rxbd(struct enetc_bdr *rx_ring, int i) +{ + int hw_idx = i; + +#ifdef CONFIG_FSL_ENETC_PTP_CLOCK + if (rx_ring->ext_en) + hw_idx = 2 * i; +#endif + return &(((union enetc_rx_bd *)rx_ring->bd_base)[hw_idx]); +} + +static inline union enetc_rx_bd *enetc_rxbd_next(struct enetc_bdr *rx_ring, + union enetc_rx_bd *rxbd, + int i) +{ + rxbd++; +#ifdef CONFIG_FSL_ENETC_PTP_CLOCK + if (rx_ring->ext_en) + rxbd++; +#endif + if (unlikely(++i == rx_ring->bd_count)) + rxbd = rx_ring->bd_base; + + return rxbd; +} + +static inline union enetc_rx_bd *enetc_rxbd_ext(union enetc_rx_bd *rxbd) +{ + return ++rxbd; +} struct enetc_msg_swbd { void *vaddr; @@ -163,7 +194,7 @@ struct enetc_int_vector { char name[ENETC_INT_NAME_MAX]; struct enetc_bdr rx_ring ____cacheline_aligned_in_smp; - struct enetc_bdr tx_ring[0]; + struct enetc_bdr tx_ring[]; }; struct enetc_cls_rule { diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c index 301ee0dde02d..34bd1f3fb415 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c @@ -195,15 +195,21 @@ static const char tx_ring_stats[][ETH_GSTRING_LEN] = { static int enetc_get_sset_count(struct net_device *ndev, int sset) { struct enetc_ndev_priv *priv = netdev_priv(ndev); + int len; + + if (sset != ETH_SS_STATS) + return -EOPNOTSUPP; - if (sset == ETH_SS_STATS) - return ARRAY_SIZE(enetc_si_counters) + - ARRAY_SIZE(tx_ring_stats) * priv->num_tx_rings + - ARRAY_SIZE(rx_ring_stats) * priv->num_rx_rings + - (enetc_si_is_pf(priv->si) ? - ARRAY_SIZE(enetc_port_counters) : 0); + len = ARRAY_SIZE(enetc_si_counters) + + ARRAY_SIZE(tx_ring_stats) * priv->num_tx_rings + + ARRAY_SIZE(rx_ring_stats) * priv->num_rx_rings; - return -EOPNOTSUPP; + if (!enetc_si_is_pf(priv->si)) + return len; + + len += ARRAY_SIZE(enetc_port_counters); + + return len; } static void enetc_get_strings(struct net_device *ndev, u32 stringset, u8 *data) @@ -568,7 +574,7 @@ static int enetc_get_ts_info(struct net_device *ndev, info->phc_index = -1; } -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING +#ifdef CONFIG_FSL_ENETC_PTP_CLOCK info->so_timestamping = SOF_TIMESTAMPING_TX_HARDWARE | SOF_TIMESTAMPING_RX_HARDWARE | SOF_TIMESTAMPING_RAW_HARDWARE; diff --git a/drivers/net/ethernet/freescale/enetc/enetc_hw.h b/drivers/net/ethernet/freescale/enetc/enetc_hw.h index 62554f28ce07..2a6523136947 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_hw.h +++ b/drivers/net/ethernet/freescale/enetc/enetc_hw.h @@ -418,9 +418,6 @@ union enetc_rx_bd { struct { __le64 addr; u8 reserved[8]; -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING - u8 reserved1[16]; -#endif } w; struct { __le16 inet_csum; @@ -435,11 +432,11 @@ union enetc_rx_bd { }; __le32 lstatus; }; -#ifdef CONFIG_FSL_ENETC_HW_TIMESTAMPING + } r; + struct { __le32 tstamp; u8 reserved[12]; -#endif - } r; + } ext; }; #define ENETC_RXBD_LSTATUS_R BIT(30) @@ -588,7 +585,7 @@ struct tgs_gcl_data { __le32 bth; __le32 ct; __le32 cte; - struct gce entry[0]; + struct gce entry[]; }; struct enetc_cbd { diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c index fc0d7d99e9a1..85e2b741df41 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c @@ -7,14 +7,7 @@ #include <linux/of_net.h> #include "enetc_pf.h" -#define ENETC_DRV_VER_MAJ 1 -#define ENETC_DRV_VER_MIN 0 - -#define ENETC_DRV_VER_STR __stringify(ENETC_DRV_VER_MAJ) "." \ - __stringify(ENETC_DRV_VER_MIN) -static const char enetc_drv_ver[] = ENETC_DRV_VER_STR; #define ENETC_DRV_NAME_STR "ENETC PF driver" -static const char enetc_drv_name[] = ENETC_DRV_NAME_STR; static void enetc_pf_get_primary_mac_addr(struct enetc_hw *hw, int si, u8 *addr) { @@ -803,11 +796,6 @@ static int enetc_of_get_phy(struct enetc_ndev_priv *priv) struct device_node *mdio_np; int err; - if (!np) { - dev_err(priv->dev, "missing ENETC port node\n"); - return -ENODEV; - } - priv->phy_node = of_parse_phandle(np, "phy-handle", 0); if (!priv->phy_node) { if (!of_phy_is_fixed_link(np)) { @@ -929,9 +917,6 @@ static int enetc_pf_probe(struct pci_dev *pdev, netif_carrier_off(ndev); - netif_info(priv, probe, ndev, "%s v%s\n", - enetc_drv_name, enetc_drv_ver); - return 0; err_reg_netdev: @@ -959,9 +944,6 @@ static void enetc_pf_remove(struct pci_dev *pdev) enetc_sriov_configure(pdev, 0); priv = netdev_priv(si->ndev); - netif_info(priv, drv, si->ndev, "%s v%s remove\n", - enetc_drv_name, enetc_drv_ver); - unregister_netdev(si->ndev); enetc_mdio_remove(pf); @@ -995,4 +977,3 @@ module_pci_driver(enetc_pf_driver); MODULE_DESCRIPTION(ENETC_DRV_NAME_STR); MODULE_LICENSE("Dual BSD/GPL"); -MODULE_VERSION(ENETC_DRV_VER_STR); diff --git a/drivers/net/ethernet/freescale/enetc/enetc_vf.c b/drivers/net/ethernet/freescale/enetc/enetc_vf.c index ebd21bf4cfa1..f14576212a0e 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_vf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_vf.c @@ -4,14 +4,7 @@ #include <linux/module.h> #include "enetc.h" -#define ENETC_DRV_VER_MAJ 1 -#define ENETC_DRV_VER_MIN 0 - -#define ENETC_DRV_VER_STR __stringify(ENETC_DRV_VER_MAJ) "." \ - __stringify(ENETC_DRV_VER_MIN) -static const char enetc_drv_ver[] = ENETC_DRV_VER_STR; #define ENETC_DRV_NAME_STR "ENETC VF driver" -static const char enetc_drv_name[] = ENETC_DRV_NAME_STR; /* Messaging */ static void enetc_msg_vsi_write_msg(struct enetc_hw *hw, @@ -201,9 +194,6 @@ static int enetc_vf_probe(struct pci_dev *pdev, netif_carrier_off(ndev); - netif_info(priv, probe, ndev, "%s v%s\n", - enetc_drv_name, enetc_drv_ver); - return 0; err_reg_netdev: @@ -225,8 +215,6 @@ static void enetc_vf_remove(struct pci_dev *pdev) struct enetc_ndev_priv *priv; priv = netdev_priv(si->ndev); - netif_info(priv, drv, si->ndev, "%s v%s remove\n", - enetc_drv_name, enetc_drv_ver); unregister_netdev(si->ndev); enetc_free_msix(priv); @@ -254,4 +242,3 @@ module_pci_driver(enetc_vf_driver); MODULE_DESCRIPTION(ENETC_DRV_NAME_STR); MODULE_LICENSE("Dual BSD/GPL"); -MODULE_VERSION(ENETC_DRV_VER_STR); diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index f79e57f735b3..bd898f5b4da5 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -584,7 +584,7 @@ struct fec_enet_private { int pps_enable; unsigned int next_counter; - u64 ethtool_stats[0]; + u64 ethtool_stats[]; }; void fec_ptp_init(struct platform_device *pdev, int irq_idx); diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 23c5fef2f1ad..c1c267b61647 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -2128,7 +2128,6 @@ static void fec_enet_get_drvinfo(struct net_device *ndev, strlcpy(info->driver, fep->pdev->dev.driver->name, sizeof(info->driver)); - strlcpy(info->version, "Revision: 1.0", sizeof(info->version)); strlcpy(info->bus_info, dev_name(&ndev->dev), sizeof(info->bus_info)); } @@ -2642,6 +2641,8 @@ fec_enet_set_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) } static const struct ethtool_ops fec_enet_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = fec_enet_get_drvinfo, .get_regs_len = fec_enet_get_regs_len, .get_regs = fec_enet_get_regs, @@ -3793,6 +3794,7 @@ static struct platform_driver fec_driver = { .name = DRIVER_NAME, .pm = &fec_pm_ops, .of_match_table = fec_dt_ids, + .suppress_bind_attrs = true, }, .id_table = fec_devtype, .probe = fec_probe, diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c b/drivers/net/ethernet/freescale/fman/fman_dtsec.c index 1ca543ac8f2c..004c266802a8 100644 --- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c +++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c @@ -366,13 +366,26 @@ static void set_dflts(struct dtsec_cfg *cfg) cfg->maximum_frame = DEFAULT_MAXIMUM_FRAME; } +static void set_mac_address(struct dtsec_regs __iomem *regs, u8 *adr) +{ + u32 tmp; + + tmp = (u32)((adr[5] << 24) | + (adr[4] << 16) | (adr[3] << 8) | adr[2]); + iowrite32be(tmp, ®s->macstnaddr1); + + tmp = (u32)((adr[1] << 24) | (adr[0] << 16)); + iowrite32be(tmp, ®s->macstnaddr2); +} + static int init(struct dtsec_regs __iomem *regs, struct dtsec_cfg *cfg, - phy_interface_t iface, u16 iface_speed, u8 *macaddr, + phy_interface_t iface, u16 iface_speed, u64 addr, u32 exception_mask, u8 tbi_addr) { bool is_rgmii, is_sgmii, is_qsgmii; - int i; + enet_addr_t eth_addr; u32 tmp; + int i; /* Soft reset */ iowrite32be(MACCFG1_SOFT_RESET, ®s->maccfg1); @@ -501,12 +514,10 @@ static int init(struct dtsec_regs __iomem *regs, struct dtsec_cfg *cfg, iowrite32be(0xffffffff, ®s->ievent); - tmp = (u32)((macaddr[5] << 24) | - (macaddr[4] << 16) | (macaddr[3] << 8) | macaddr[2]); - iowrite32be(tmp, ®s->macstnaddr1); - - tmp = (u32)((macaddr[1] << 24) | (macaddr[0] << 16)); - iowrite32be(tmp, ®s->macstnaddr2); + if (addr) { + MAKE_ENET_ADDR_FROM_UINT64(addr, eth_addr); + set_mac_address(regs, (u8 *)eth_addr); + } /* HASH */ for (i = 0; i < NUM_OF_HASH_REGS; i++) { @@ -519,18 +530,6 @@ static int init(struct dtsec_regs __iomem *regs, struct dtsec_cfg *cfg, return 0; } -static void set_mac_address(struct dtsec_regs __iomem *regs, u8 *adr) -{ - u32 tmp; - - tmp = (u32)((adr[5] << 24) | - (adr[4] << 16) | (adr[3] << 8) | adr[2]); - iowrite32be(tmp, ®s->macstnaddr1); - - tmp = (u32)((adr[1] << 24) | (adr[0] << 16)); - iowrite32be(tmp, ®s->macstnaddr2); -} - static void set_bucket(struct dtsec_regs __iomem *regs, int bucket, bool enable) { @@ -556,10 +555,6 @@ static int check_init_parameters(struct fman_mac *dtsec) pr_err("1G MAC driver supports 1G or lower speeds\n"); return -EINVAL; } - if (dtsec->addr == 0) { - pr_err("Ethernet MAC Must have a valid MAC Address\n"); - return -EINVAL; - } if ((dtsec->dtsec_drv_param)->rx_prepend > MAX_PACKET_ALIGNMENT) { pr_err("packetAlignmentPadding can't be > than %d\n", @@ -1391,9 +1386,8 @@ int dtsec_init(struct fman_mac *dtsec) { struct dtsec_regs __iomem *regs = dtsec->regs; struct dtsec_cfg *dtsec_drv_param; - int err; u16 max_frm_ln; - enet_addr_t eth_addr; + int err; if (is_init_done(dtsec->dtsec_drv_param)) return -EINVAL; @@ -1410,10 +1404,8 @@ int dtsec_init(struct fman_mac *dtsec) dtsec_drv_param = dtsec->dtsec_drv_param; - MAKE_ENET_ADDR_FROM_UINT64(dtsec->addr, eth_addr); - err = init(dtsec->regs, dtsec_drv_param, dtsec->phy_if, - dtsec->max_speed, (u8 *)eth_addr, dtsec->exceptions, + dtsec->max_speed, dtsec->addr, dtsec->exceptions, dtsec->tbiphy->mdio.addr); if (err) { free_init_resources(dtsec); diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c index 0d2b4ab01f24..a5500ede4070 100644 --- a/drivers/net/ethernet/freescale/fman/fman_memac.c +++ b/drivers/net/ethernet/freescale/fman/fman_memac.c @@ -596,10 +596,6 @@ static void setup_sgmii_internal_phy_base_x(struct fman_mac *memac) static int check_init_parameters(struct fman_mac *memac) { - if (memac->addr == 0) { - pr_err("Ethernet MAC must have a valid MAC address\n"); - return -EINVAL; - } if (!memac->exception_cb) { pr_err("Uninitialized exception handler\n"); return -EINVAL; @@ -1057,8 +1053,10 @@ int memac_init(struct fman_mac *memac) } /* MAC Address */ - MAKE_ENET_ADDR_FROM_UINT64(memac->addr, eth_addr); - add_addr_in_paddr(memac->regs, (u8 *)eth_addr, 0); + if (memac->addr != 0) { + MAKE_ENET_ADDR_FROM_UINT64(memac->addr, eth_addr); + add_addr_in_paddr(memac->regs, (u8 *)eth_addr, 0); + } fixed_link = memac_drv_param->fixed_link; diff --git a/drivers/net/ethernet/freescale/fman/fman_tgec.c b/drivers/net/ethernet/freescale/fman/fman_tgec.c index f75b9c11b2d2..8c7eb878d5b4 100644 --- a/drivers/net/ethernet/freescale/fman/fman_tgec.c +++ b/drivers/net/ethernet/freescale/fman/fman_tgec.c @@ -273,10 +273,6 @@ static int check_init_parameters(struct fman_mac *tgec) pr_err("10G MAC driver only support 10G speed\n"); return -EINVAL; } - if (tgec->addr == 0) { - pr_err("Ethernet 10G MAC Must have valid MAC Address\n"); - return -EINVAL; - } if (!tgec->exception_cb) { pr_err("uninitialized exception_cb\n"); return -EINVAL; @@ -706,8 +702,10 @@ int tgec_init(struct fman_mac *tgec) cfg = tgec->cfg; - MAKE_ENET_ADDR_FROM_UINT64(tgec->addr, eth_addr); - set_mac_address(tgec->regs, (u8 *)eth_addr); + if (tgec->addr) { + MAKE_ENET_ADDR_FROM_UINT64(tgec->addr, eth_addr); + set_mac_address(tgec->regs, (u8 *)eth_addr); + } /* interrupts */ /* FM_10G_REM_N_LCL_FLT_EX_10GMAC_ERRATA_SW005 Errata workaround */ diff --git a/drivers/net/ethernet/freescale/fman/mac.c b/drivers/net/ethernet/freescale/fman/mac.c index 55f2122c3217..43427c5b9396 100644 --- a/drivers/net/ethernet/freescale/fman/mac.c +++ b/drivers/net/ethernet/freescale/fman/mac.c @@ -724,12 +724,10 @@ static int mac_probe(struct platform_device *_of_dev) /* Get the MAC address */ mac_addr = of_get_mac_address(mac_node); - if (IS_ERR(mac_addr)) { - dev_err(dev, "of_get_mac_address(%pOF) failed\n", mac_node); - err = -EINVAL; - goto _return_of_get_parent; - } - ether_addr_copy(mac_dev->addr, mac_addr); + if (IS_ERR(mac_addr)) + dev_warn(dev, "of_get_mac_address(%pOF) failed\n", mac_node); + else + ether_addr_copy(mac_dev->addr, mac_addr); /* Get the port handles */ nph = of_count_phandle_with_args(mac_node, "fsl,fman-ports", NULL); @@ -855,7 +853,8 @@ static int mac_probe(struct platform_device *_of_dev) if (err < 0) dev_err(dev, "fman_set_mac_active_pause() = %d\n", err); - dev_info(dev, "FMan MAC address: %pM\n", mac_dev->addr); + if (!IS_ERR(mac_addr)) + dev_info(dev, "FMan MAC address: %pM\n", mac_dev->addr); priv->eth_dev = dpaa_eth_add_device(fman_id, mac_dev); if (IS_ERR(priv->eth_dev)) { diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c index add61fed33ee..ce85feaac357 100644 --- a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c +++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c @@ -53,7 +53,6 @@ MODULE_AUTHOR("Pantelis Antoniou <panto@intracom.gr>"); MODULE_DESCRIPTION("Freescale Ethernet Driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION(DRV_MODULE_VERSION); static int fs_enet_debug = -1; /* -1 == use FS_ENET_DEF_MSG_ENABLE as value */ module_param(fs_enet_debug, int, 0); @@ -790,7 +789,6 @@ static void fs_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); } static int fs_get_regs_len(struct net_device *dev) diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet.h b/drivers/net/ethernet/freescale/fs_enet/fs_enet.h index 195fae6aec4a..5ff2634bee2f 100644 --- a/drivers/net/ethernet/freescale/fs_enet/fs_enet.h +++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet.h @@ -190,8 +190,6 @@ void fs_cleanup_bds(struct net_device *dev); #define DRV_MODULE_NAME "fs_enet" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "1.1" -#define DRV_MODULE_RELDATE "Sep 22, 2014" /***************************************************************************/ diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index f7e5cafe89a9..b3c69e9038ea 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -103,8 +103,6 @@ #define TX_TIMEOUT (5*HZ) -const char gfar_driver_version[] = "2.0"; - MODULE_AUTHOR("Freescale Semiconductor, Inc"); MODULE_DESCRIPTION("Gianfar Ethernet Driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/freescale/gianfar.h b/drivers/net/ethernet/freescale/gianfar.h index 432c6a818ae5..8ced783f5302 100644 --- a/drivers/net/ethernet/freescale/gianfar.h +++ b/drivers/net/ethernet/freescale/gianfar.h @@ -68,7 +68,6 @@ struct ethtool_rx_list { #define RXBUF_ALIGNMENT 64 #define DRV_NAME "gfar-enet" -extern const char gfar_driver_version[]; /* MAXIMUM NUMBER OF QUEUES SUPPORTED */ #define MAX_TX_QS 0x8 diff --git a/drivers/net/ethernet/freescale/gianfar_ethtool.c b/drivers/net/ethernet/freescale/gianfar_ethtool.c index 3c8e4e2efc07..cc7d4f93da54 100644 --- a/drivers/net/ethernet/freescale/gianfar_ethtool.c +++ b/drivers/net/ethernet/freescale/gianfar_ethtool.c @@ -164,10 +164,6 @@ static void gfar_gdrvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) { strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, gfar_driver_version, - sizeof(drvinfo->version)); - strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version)); - strlcpy(drvinfo->bus_info, "N/A", sizeof(drvinfo->bus_info)); } /* Return the length of the register structure */ @@ -276,35 +272,6 @@ static int gfar_gcoalesce(struct net_device *dev, cvals->tx_coalesce_usecs = gfar_ticks2usecs(priv, txtime); cvals->tx_max_coalesced_frames = txcount; - cvals->use_adaptive_rx_coalesce = 0; - cvals->use_adaptive_tx_coalesce = 0; - - cvals->pkt_rate_low = 0; - cvals->rx_coalesce_usecs_low = 0; - cvals->rx_max_coalesced_frames_low = 0; - cvals->tx_coalesce_usecs_low = 0; - cvals->tx_max_coalesced_frames_low = 0; - - /* When the packet rate is below pkt_rate_high but above - * pkt_rate_low (both measured in packets per second) the - * normal {rx,tx}_* coalescing parameters are used. - */ - - /* When the packet rate is (measured in packets per second) - * is above pkt_rate_high, the {rx,tx}_*_high parameters are - * used. - */ - cvals->pkt_rate_high = 0; - cvals->rx_coalesce_usecs_high = 0; - cvals->rx_max_coalesced_frames_high = 0; - cvals->tx_coalesce_usecs_high = 0; - cvals->tx_max_coalesced_frames_high = 0; - - /* How often to do adaptive coalescing packet rate sampling, - * measured in seconds. Must not be zero. - */ - cvals->rate_sample_interval = 0; - return 0; } @@ -1507,6 +1474,8 @@ static int gfar_get_ts_info(struct net_device *dev, } const struct ethtool_ops gfar_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = gfar_gdrvinfo, .get_regs_len = gfar_reglen, .get_regs = gfar_get_regs, diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c index 0d101c00286f..6e5f6dd169b5 100644 --- a/drivers/net/ethernet/freescale/ucc_geth.c +++ b/drivers/net/ethernet/freescale/ucc_geth.c @@ -3990,5 +3990,4 @@ module_exit(ucc_geth_exit); MODULE_AUTHOR("Freescale Semiconductor, Inc"); MODULE_DESCRIPTION(DRV_DESC); -MODULE_VERSION(DRV_VERSION); MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/freescale/ucc_geth.h b/drivers/net/ethernet/freescale/ucc_geth.h index a86a42131fc7..3fe903972195 100644 --- a/drivers/net/ethernet/freescale/ucc_geth.h +++ b/drivers/net/ethernet/freescale/ucc_geth.h @@ -26,7 +26,6 @@ #define DRV_DESC "QE UCC Gigabit Ethernet Controller" #define DRV_NAME "ucc_geth" -#define DRV_VERSION "1.1" #define NUM_TX_QUEUES 8 #define NUM_RX_QUEUES 8 diff --git a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c index dfebacf443fc..14c08a868190 100644 --- a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c +++ b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c @@ -334,8 +334,6 @@ uec_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version)); - strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version)); strlcpy(drvinfo->bus_info, "QUICC ENGINE", sizeof(drvinfo->bus_info)); } diff --git a/drivers/net/ethernet/hisilicon/hip04_eth.c b/drivers/net/ethernet/hisilicon/hip04_eth.c index d9718b87279d..12f6c2442a7a 100644 --- a/drivers/net/ethernet/hisilicon/hip04_eth.c +++ b/drivers/net/ethernet/hisilicon/hip04_eth.c @@ -811,20 +811,6 @@ static int hip04_set_coalesce(struct net_device *netdev, { struct hip04_priv *priv = netdev_priv(netdev); - /* Check not supported parameters */ - if ((ec->rx_max_coalesced_frames) || (ec->rx_coalesce_usecs_irq) || - (ec->rx_max_coalesced_frames_irq) || (ec->tx_coalesce_usecs_irq) || - (ec->use_adaptive_rx_coalesce) || (ec->use_adaptive_tx_coalesce) || - (ec->pkt_rate_low) || (ec->rx_coalesce_usecs_low) || - (ec->rx_max_coalesced_frames_low) || (ec->tx_coalesce_usecs_high) || - (ec->tx_max_coalesced_frames_low) || (ec->pkt_rate_high) || - (ec->tx_coalesce_usecs_low) || (ec->rx_coalesce_usecs_high) || - (ec->rx_max_coalesced_frames_high) || (ec->rx_coalesce_usecs) || - (ec->tx_max_coalesced_frames_irq) || - (ec->stats_block_coalesce_usecs) || - (ec->tx_max_coalesced_frames_high) || (ec->rate_sample_interval)) - return -EOPNOTSUPP; - if ((ec->tx_coalesce_usecs > HIP04_MAX_TX_COALESCE_USECS || ec->tx_coalesce_usecs < HIP04_MIN_TX_COALESCE_USECS) || (ec->tx_max_coalesced_frames > HIP04_MAX_TX_COALESCE_FRAMES || @@ -845,6 +831,8 @@ static void hip04_get_drvinfo(struct net_device *netdev, } static const struct ethtool_ops hip04_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_TX_USECS | + ETHTOOL_COALESCE_TX_MAX_FRAMES, .get_coalesce = hip04_get_coalesce, .set_coalesce = hip04_set_coalesce, .get_drvinfo = hip04_get_drvinfo, diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.h index 2721f1f1ab42..0f0e16f9afc0 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.h @@ -92,7 +92,7 @@ struct ppe_common_cb { u8 comm_index; /*ppe_common index*/ u32 ppe_num; - struct hns_ppe_cb ppe_cb[0]; + struct hns_ppe_cb ppe_cb[]; }; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h index 3741befb914e..a9f805925699 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h @@ -108,7 +108,7 @@ struct rcb_common_cb { u32 ring_num; u32 desc_num; /* desc num per queue*/ - struct ring_pair_cb ring_pair_cb[0]; + struct ring_pair_cb ring_pair_cb[]; }; int hns_rcb_buf_size2type(u32 buf_size); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index 717fccc2efba..49624acf2473 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -1264,6 +1264,11 @@ static int hns_get_rxnfc(struct net_device *netdev, } static const struct ethtool_ops hns_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE | + ETHTOOL_COALESCE_USECS_LOW_HIGH | + ETHTOOL_COALESCE_MAX_FRAMES_LOW_HIGH, .get_drvinfo = hns_nic_get_drvinfo, .get_link = hns_nic_get_link, .get_ringparam = hns_get_ringparam, diff --git a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h index d87158acdf6f..948e67ef30fd 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h +++ b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h @@ -7,8 +7,6 @@ #include <linux/mutex.h> #include <linux/types.h> -#define HCLGE_MBX_VF_MSG_DATA_NUM 16 - enum HCLGE_MBX_OPCODE { HCLGE_MBX_RESET = 0x01, /* (VF -> PF) assert reset */ HCLGE_MBX_ASSERTING_RESET, /* (PF -> VF) PF is asserting reset*/ @@ -72,10 +70,15 @@ enum hclge_mbx_vlan_cfg_subcode { HCLGE_MBX_GET_PORT_BASE_VLAN_STATE, /* get port based vlan state */ }; -#define HCLGE_MBX_MAX_MSG_SIZE 16 +#define HCLGE_MBX_MAX_MSG_SIZE 14 #define HCLGE_MBX_MAX_RESP_DATA_SIZE 8U -#define HCLGE_MBX_RING_MAP_BASIC_MSG_NUM 3 -#define HCLGE_MBX_RING_NODE_VARIABLE_NUM 3 +#define HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM 4 + +struct hclge_ring_chain_param { + u8 ring_type; + u8 tqp_index; + u8 int_gl_index; +}; struct hclgevf_mbx_resp_status { struct mutex mbx_mutex; /* protects against contending sync cmd resp */ @@ -85,6 +88,41 @@ struct hclgevf_mbx_resp_status { u8 additional_info[HCLGE_MBX_MAX_RESP_DATA_SIZE]; }; +struct hclge_respond_to_vf_msg { + int status; + u8 data[HCLGE_MBX_MAX_RESP_DATA_SIZE]; + u16 len; +}; + +struct hclge_vf_to_pf_msg { + u8 code; + union { + struct { + u8 subcode; + u8 data[HCLGE_MBX_MAX_MSG_SIZE]; + }; + struct { + u8 en_bc; + u8 en_uc; + u8 en_mc; + }; + struct { + u8 vector_id; + u8 ring_num; + struct hclge_ring_chain_param + param[HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM]; + }; + }; +}; + +struct hclge_pf_to_vf_msg { + u16 code; + u16 vf_mbx_msg_code; + u16 vf_mbx_msg_subcode; + u16 resp_status; + u8 resp_data[HCLGE_MBX_MAX_RESP_DATA_SIZE]; +}; + struct hclge_mbx_vf_to_pf_cmd { u8 rsv; u8 mbx_src_vfid; /* Auto filled by IMP */ @@ -92,17 +130,17 @@ struct hclge_mbx_vf_to_pf_cmd { u8 rsv1[1]; u8 msg_len; u8 rsv2[3]; - u8 msg[HCLGE_MBX_MAX_MSG_SIZE]; + struct hclge_vf_to_pf_msg msg; }; -#define HCLGE_MBX_NEED_RESP_BIT BIT(0) +#define HCLGE_MBX_NEED_RESP_B 0 struct hclge_mbx_pf_to_vf_cmd { u8 dest_vfid; u8 rsv[3]; u8 msg_len; u8 rsv1[3]; - u16 msg[8]; + struct hclge_pf_to_vf_msg msg; }; struct hclge_vf_rst_cmd { diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index a3e4081b84ba..5587605d6deb 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -78,6 +78,7 @@ enum hns_desc_type { DESC_TYPE_SKB, + DESC_TYPE_FRAGLIST_SKB, DESC_TYPE_PAGE, }; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c index 1d4ffc5f408a..e1d88095a77e 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c @@ -260,6 +260,8 @@ static void hns3_dbg_help(struct hnae3_handle *h) dev_info(&h->pdev->dev, "dump m7 info\n"); dev_info(&h->pdev->dev, "dump ncl_config <offset> <length>(in hex)\n"); dev_info(&h->pdev->dev, "dump mac tnl status\n"); + dev_info(&h->pdev->dev, "dump loopback\n"); + dev_info(&h->pdev->dev, "dump qs shaper [qs id]\n"); memset(printf_buf, 0, HNS3_DBG_BUF_LEN); strncat(printf_buf, "dump reg [[bios common] [ssu <port_id>]", diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index a7f40aa1a0ea..da98fd7c8eca 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -1107,6 +1107,10 @@ static int hns3_fill_desc(struct hns3_enet_ring *ring, void *priv, return ret; dma = dma_map_single(dev, skb->data, size, DMA_TO_DEVICE); + } else if (type == DESC_TYPE_FRAGLIST_SKB) { + struct sk_buff *skb = (struct sk_buff *)priv; + + dma = dma_map_single(dev, skb->data, size, DMA_TO_DEVICE); } else { frag = (skb_frag_t *)priv; dma = skb_frag_dma_map(dev, frag, 0, size, DMA_TO_DEVICE); @@ -1144,8 +1148,9 @@ static int hns3_fill_desc(struct hns3_enet_ring *ring, void *priv, /* The txbd's baseinfo of DESC_TYPE_PAGE & DESC_TYPE_SKB */ desc_cb->priv = priv; desc_cb->dma = dma + HNS3_MAX_BD_SIZE * k; - desc_cb->type = (type == DESC_TYPE_SKB && !k) ? - DESC_TYPE_SKB : DESC_TYPE_PAGE; + desc_cb->type = ((type == DESC_TYPE_FRAGLIST_SKB || + type == DESC_TYPE_SKB) && !k) ? + type : DESC_TYPE_PAGE; /* now, fill the descriptor */ desc->addr = cpu_to_le64(dma + HNS3_MAX_BD_SIZE * k); @@ -1354,7 +1359,9 @@ static void hns3_clear_desc(struct hns3_enet_ring *ring, int next_to_use_orig) ring_ptr_move_bw(ring, next_to_use); /* unmap the descriptor dma address */ - if (ring->desc_cb[ring->next_to_use].type == DESC_TYPE_SKB) + if (ring->desc_cb[ring->next_to_use].type == DESC_TYPE_SKB || + ring->desc_cb[ring->next_to_use].type == + DESC_TYPE_FRAGLIST_SKB) dma_unmap_single(dev, ring->desc_cb[ring->next_to_use].dma, ring->desc_cb[ring->next_to_use].length, @@ -1447,7 +1454,8 @@ netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev) goto out; skb_walk_frags(skb, frag_skb) { - ret = hns3_fill_skb_to_desc(ring, frag_skb, DESC_TYPE_PAGE); + ret = hns3_fill_skb_to_desc(ring, frag_skb, + DESC_TYPE_FRAGLIST_SKB); if (unlikely(ret < 0)) goto fill_err; @@ -2228,7 +2236,7 @@ static void hns3_reset_prepare(struct pci_dev *pdev) { struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev); - dev_info(&pdev->dev, "hns3 flr prepare\n"); + dev_info(&pdev->dev, "FLR prepare\n"); if (ae_dev && ae_dev->ops && ae_dev->ops->flr_prepare) ae_dev->ops->flr_prepare(ae_dev); } @@ -2237,7 +2245,7 @@ static void hns3_reset_done(struct pci_dev *pdev) { struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev); - dev_info(&pdev->dev, "hns3 flr done\n"); + dev_info(&pdev->dev, "FLR done\n"); if (ae_dev && ae_dev->ops && ae_dev->ops->flr_done) ae_dev->ops->flr_done(ae_dev); } @@ -2356,7 +2364,7 @@ static int hns3_map_buffer(struct hns3_enet_ring *ring, struct hns3_desc_cb *cb) static void hns3_unmap_buffer(struct hns3_enet_ring *ring, struct hns3_desc_cb *cb) { - if (cb->type == DESC_TYPE_SKB) + if (cb->type == DESC_TYPE_SKB || cb->type == DESC_TYPE_FRAGLIST_SKB) dma_unmap_single(ring_to_dev(ring), cb->dma, cb->length, ring_to_dma_dir(ring)); else if (cb->length) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c index c03856e63320..28b81f24afa1 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c @@ -736,7 +736,7 @@ static int hns3_check_ksettings_param(const struct net_device *netdev, if (ops->get_media_type) ops->get_media_type(handle, &media_type, &module_type); - if (cmd->base.duplex != DUPLEX_FULL && + if (cmd->base.duplex == DUPLEX_HALF && media_type != HNAE3_MEDIA_TYPE_COPPER) { netdev_err(netdev, "only copper port supports half duplex!"); @@ -1390,7 +1390,13 @@ static int hns3_set_fecparam(struct net_device *netdev, return ops->set_fec(handle, fec_mode); } +#define HNS3_ETHTOOL_COALESCE (ETHTOOL_COALESCE_USECS | \ + ETHTOOL_COALESCE_USE_ADAPTIVE | \ + ETHTOOL_COALESCE_RX_USECS_HIGH | \ + ETHTOOL_COALESCE_TX_USECS_HIGH) + static const struct ethtool_ops hns3vf_ethtool_ops = { + .supported_coalesce_params = HNS3_ETHTOOL_COALESCE, .get_drvinfo = hns3_get_drvinfo, .get_ringparam = hns3_get_ringparam, .set_ringparam = hns3_set_ringparam, @@ -1416,6 +1422,7 @@ static const struct ethtool_ops hns3vf_ethtool_ops = { }; static const struct ethtool_ops hns3_ethtool_ops = { + .supported_coalesce_params = HNS3_ETHTOOL_COALESCE, .self_test = hns3_self_test, .get_drvinfo = hns3_get_drvinfo, .get_link = hns3_get_link, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c index 67fad80035d3..17228288d4df 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c @@ -87,7 +87,7 @@ static int hclge_dbg_get_dfx_bd_num(struct hclge_dev *hdev, int offset) entries_per_desc = ARRAY_SIZE(desc[0].data); index = offset % entries_per_desc; - return (int)desc[offset / entries_per_desc].data[index]; + return le32_to_cpu(desc[offset / entries_per_desc].data[index]); } static int hclge_dbg_cmd_send(struct hclge_dev *hdev, @@ -145,10 +145,8 @@ static void hclge_dbg_dump_reg_common(struct hclge_dev *hdev, buf_len = sizeof(struct hclge_desc) * bd_num; desc_src = kzalloc(buf_len, GFP_KERNEL); - if (!desc_src) { - dev_err(&hdev->pdev->dev, "call kzalloc failed\n"); + if (!desc_src) return; - } desc = desc_src; ret = hclge_dbg_cmd_send(hdev, desc, index, bd_num, reg_msg->cmd); @@ -179,6 +177,7 @@ static void hclge_dbg_dump_dcb(struct hclge_dev *hdev, const char *cmd_buf) { struct device *dev = &hdev->pdev->dev; struct hclge_dbg_bitmap_cmd *bitmap; + enum hclge_opcode_type cmd; int rq_id, pri_id, qset_id; int port_id, nq_id, pg_id; struct hclge_desc desc[2]; @@ -193,10 +192,10 @@ static void hclge_dbg_dump_dcb(struct hclge_dev *hdev, const char *cmd_buf) return; } - ret = hclge_dbg_cmd_send(hdev, desc, qset_id, 1, - HCLGE_OPC_QSET_DFX_STS); + cmd = HCLGE_OPC_QSET_DFX_STS; + ret = hclge_dbg_cmd_send(hdev, desc, qset_id, 1, cmd); if (ret) - return; + goto err_dcb_cmd_send; bitmap = (struct hclge_dbg_bitmap_cmd *)&desc[0].data[1]; dev_info(dev, "roce_qset_mask: 0x%x\n", bitmap->bit0); @@ -204,48 +203,53 @@ static void hclge_dbg_dump_dcb(struct hclge_dev *hdev, const char *cmd_buf) dev_info(dev, "qs_shaping_pass: 0x%x\n", bitmap->bit2); dev_info(dev, "qs_bp_sts: 0x%x\n", bitmap->bit3); - ret = hclge_dbg_cmd_send(hdev, desc, pri_id, 1, HCLGE_OPC_PRI_DFX_STS); + cmd = HCLGE_OPC_PRI_DFX_STS; + ret = hclge_dbg_cmd_send(hdev, desc, pri_id, 1, cmd); if (ret) - return; + goto err_dcb_cmd_send; bitmap = (struct hclge_dbg_bitmap_cmd *)&desc[0].data[1]; dev_info(dev, "pri_mask: 0x%x\n", bitmap->bit0); dev_info(dev, "pri_cshaping_pass: 0x%x\n", bitmap->bit1); dev_info(dev, "pri_pshaping_pass: 0x%x\n", bitmap->bit2); - ret = hclge_dbg_cmd_send(hdev, desc, pg_id, 1, HCLGE_OPC_PG_DFX_STS); + cmd = HCLGE_OPC_PG_DFX_STS; + ret = hclge_dbg_cmd_send(hdev, desc, pg_id, 1, cmd); if (ret) - return; + goto err_dcb_cmd_send; bitmap = (struct hclge_dbg_bitmap_cmd *)&desc[0].data[1]; dev_info(dev, "pg_mask: 0x%x\n", bitmap->bit0); dev_info(dev, "pg_cshaping_pass: 0x%x\n", bitmap->bit1); dev_info(dev, "pg_pshaping_pass: 0x%x\n", bitmap->bit2); - ret = hclge_dbg_cmd_send(hdev, desc, port_id, 1, - HCLGE_OPC_PORT_DFX_STS); + cmd = HCLGE_OPC_PORT_DFX_STS; + ret = hclge_dbg_cmd_send(hdev, desc, port_id, 1, cmd); if (ret) - return; + goto err_dcb_cmd_send; bitmap = (struct hclge_dbg_bitmap_cmd *)&desc[0].data[1]; dev_info(dev, "port_mask: 0x%x\n", bitmap->bit0); dev_info(dev, "port_shaping_pass: 0x%x\n", bitmap->bit1); - ret = hclge_dbg_cmd_send(hdev, desc, nq_id, 1, HCLGE_OPC_SCH_NQ_CNT); + cmd = HCLGE_OPC_SCH_NQ_CNT; + ret = hclge_dbg_cmd_send(hdev, desc, nq_id, 1, cmd); if (ret) - return; + goto err_dcb_cmd_send; dev_info(dev, "sch_nq_cnt: 0x%x\n", le32_to_cpu(desc[0].data[1])); - ret = hclge_dbg_cmd_send(hdev, desc, nq_id, 1, HCLGE_OPC_SCH_RQ_CNT); + cmd = HCLGE_OPC_SCH_RQ_CNT; + ret = hclge_dbg_cmd_send(hdev, desc, nq_id, 1, cmd); if (ret) - return; + goto err_dcb_cmd_send; dev_info(dev, "sch_rq_cnt: 0x%x\n", le32_to_cpu(desc[0].data[1])); - ret = hclge_dbg_cmd_send(hdev, desc, 0, 2, HCLGE_OPC_TM_INTERNAL_STS); + cmd = HCLGE_OPC_TM_INTERNAL_STS; + ret = hclge_dbg_cmd_send(hdev, desc, 0, 2, cmd); if (ret) - return; + goto err_dcb_cmd_send; dev_info(dev, "pri_bp: 0x%x\n", le32_to_cpu(desc[0].data[1])); dev_info(dev, "fifo_dfx_info: 0x%x\n", le32_to_cpu(desc[0].data[2])); @@ -257,18 +261,18 @@ static void hclge_dbg_dump_dcb(struct hclge_dev *hdev, const char *cmd_buf) dev_info(dev, "SSU_TM_BYPASS_EN: 0x%x\n", le32_to_cpu(desc[1].data[0])); dev_info(dev, "SSU_RESERVE_CFG: 0x%x\n", le32_to_cpu(desc[1].data[1])); - ret = hclge_dbg_cmd_send(hdev, desc, port_id, 1, - HCLGE_OPC_TM_INTERNAL_CNT); + cmd = HCLGE_OPC_TM_INTERNAL_CNT; + ret = hclge_dbg_cmd_send(hdev, desc, port_id, 1, cmd); if (ret) - return; + goto err_dcb_cmd_send; dev_info(dev, "SCH_NIC_NUM: 0x%x\n", le32_to_cpu(desc[0].data[1])); dev_info(dev, "SCH_ROCE_NUM: 0x%x\n", le32_to_cpu(desc[0].data[2])); - ret = hclge_dbg_cmd_send(hdev, desc, port_id, 1, - HCLGE_OPC_TM_INTERNAL_STS_1); + cmd = HCLGE_OPC_TM_INTERNAL_STS_1; + ret = hclge_dbg_cmd_send(hdev, desc, port_id, 1, cmd); if (ret) - return; + goto err_dcb_cmd_send; dev_info(dev, "TC_MAP_SEL: 0x%x\n", le32_to_cpu(desc[0].data[1])); dev_info(dev, "IGU_PFC_PRI_EN: 0x%x\n", le32_to_cpu(desc[0].data[2])); @@ -277,6 +281,12 @@ static void hclge_dbg_dump_dcb(struct hclge_dev *hdev, const char *cmd_buf) le32_to_cpu(desc[0].data[4])); dev_info(dev, "IGU_TX_PRI_MAP_TC_CFG: 0x%x\n", le32_to_cpu(desc[0].data[5])); + return; + +err_dcb_cmd_send: + dev_err(&hdev->pdev->dev, + "failed to dump dcb dfx, cmd = %#x, ret = %d\n", + cmd, ret); } static void hclge_dbg_dump_reg_cmd(struct hclge_dev *hdev, const char *cmd_buf) @@ -310,8 +320,9 @@ static void hclge_title_idx_print(struct hclge_dev *hdev, bool flag, int index, char *false_buf) { if (flag) - dev_info(&hdev->pdev->dev, "%s(%d): %s\n", title_buf, index, - true_buf); + dev_info(&hdev->pdev->dev, "%s(%d): %s weight: %u\n", + title_buf, index, true_buf, + hdev->tm_info.pg_info[0].tc_dwrr[index]); else dev_info(&hdev->pdev->dev, "%s(%d): %s\n", title_buf, index, false_buf); @@ -339,7 +350,8 @@ static void hclge_dbg_dump_tc(struct hclge_dev *hdev) ets_weight = (struct hclge_ets_tc_weight_cmd *)desc.data; - dev_info(&hdev->pdev->dev, "dump tc\n"); + dev_info(&hdev->pdev->dev, "dump tc: %u tc enabled\n", + hdev->tm_info.num_tc); dev_info(&hdev->pdev->dev, "weight_offset: %u\n", ets_weight->weight_offset); @@ -581,7 +593,7 @@ static void hclge_dbg_dump_tm_map(struct hclge_dev *hdev, ret = hclge_cmd_send(&hdev->hw, &desc, 1); if (ret) goto err_tm_map_cmd_send; - qset_id = nq_to_qs_map->qset_id & 0x3FF; + qset_id = le16_to_cpu(nq_to_qs_map->qset_id) & 0x3FF; cmd = HCLGE_OPC_TM_QS_TO_PRI_LINK; map = (struct hclge_qs_to_pri_link_cmd *)desc.data; @@ -621,7 +633,8 @@ static void hclge_dbg_dump_tm_map(struct hclge_dev *hdev, if (ret) goto err_tm_map_cmd_send; - qset_maping[group_id] = bp_to_qs_map_cmd->qs_bit_map; + qset_maping[group_id] = + le32_to_cpu(bp_to_qs_map_cmd->qs_bit_map); } dev_info(&hdev->pdev->dev, "index | tm bp qset maping:\n"); @@ -824,6 +837,7 @@ static void hclge_dbg_dump_mng_table(struct hclge_dev *hdev) struct hclge_mac_ethertype_idx_rd_cmd *req0; char printf_buf[HCLGE_DBG_BUF_LEN]; struct hclge_desc desc; + u32 msg_egress_port; int ret, i; dev_info(&hdev->pdev->dev, "mng tab:\n"); @@ -865,20 +879,21 @@ static void hclge_dbg_dump_mng_table(struct hclge_dev *hdev) HCLGE_DBG_BUF_LEN - strlen(printf_buf), "%x |%04x |%x |%04x|%x |%02x |%02x |", !!(req0->flags & HCLGE_DBG_MNG_MAC_MASK_B), - req0->ethter_type, + le16_to_cpu(req0->ethter_type), !!(req0->flags & HCLGE_DBG_MNG_ETHER_MASK_B), - req0->vlan_tag & HCLGE_DBG_MNG_VLAN_TAG, + le16_to_cpu(req0->vlan_tag) & HCLGE_DBG_MNG_VLAN_TAG, !!(req0->flags & HCLGE_DBG_MNG_VLAN_MASK_B), req0->i_port_bitmap, req0->i_port_direction); + msg_egress_port = le16_to_cpu(req0->egress_port); snprintf(printf_buf + strlen(printf_buf), HCLGE_DBG_BUF_LEN - strlen(printf_buf), - "%d |%d |%02d |%04d|%x\n", - !!(req0->egress_port & HCLGE_DBG_MNG_E_TYPE_B), - req0->egress_port & HCLGE_DBG_MNG_PF_ID, - (req0->egress_port >> 3) & HCLGE_DBG_MNG_VF_ID, - req0->egress_queue, - !!(req0->egress_port & HCLGE_DBG_MNG_DROP_B)); + "%x |%x |%02x |%04x|%x\n", + !!(msg_egress_port & HCLGE_DBG_MNG_E_TYPE_B), + msg_egress_port & HCLGE_DBG_MNG_PF_ID, + (msg_egress_port >> 3) & HCLGE_DBG_MNG_VF_ID, + le16_to_cpu(req0->egress_queue), + !!(msg_egress_port & HCLGE_DBG_MNG_DROP_B)); dev_info(&hdev->pdev->dev, "%s", printf_buf); } @@ -1065,11 +1080,8 @@ static void hclge_dbg_get_m7_stats_info(struct hclge_dev *hdev) buf_len = sizeof(struct hclge_desc) * bd_num; desc_src = kzalloc(buf_len, GFP_KERNEL); - if (!desc_src) { - dev_err(&hdev->pdev->dev, - "allocate desc for get_m7_stats failed\n"); + if (!desc_src) return; - } desc_tmp = desc_src; ret = hclge_dbg_cmd_send(hdev, desc_tmp, 0, bd_num, @@ -1132,7 +1144,7 @@ static void hclge_dbg_dump_ncl_config(struct hclge_dev *hdev, const char *cmd_buf) { #define HCLGE_MAX_NCL_CONFIG_OFFSET 4096 -#define HCLGE_MAX_NCL_CONFIG_LENGTH (20 + 24 * 4) +#define HCLGE_NCL_CONFIG_LENGTH_IN_EACH_CMD (20 + 24 * 4) struct hclge_desc desc[HCLGE_CMD_NCL_CONFIG_BD_NUM]; int bd_num = HCLGE_CMD_NCL_CONFIG_BD_NUM; @@ -1156,8 +1168,8 @@ static void hclge_dbg_dump_ncl_config(struct hclge_dev *hdev, while (length > 0) { data0 = offset; - if (length >= HCLGE_MAX_NCL_CONFIG_LENGTH) - data0 |= HCLGE_MAX_NCL_CONFIG_LENGTH << 16; + if (length >= HCLGE_NCL_CONFIG_LENGTH_IN_EACH_CMD) + data0 |= HCLGE_NCL_CONFIG_LENGTH_IN_EACH_CMD << 16; else data0 |= length << 16; ret = hclge_dbg_cmd_send(hdev, desc, data0, bd_num, @@ -1169,6 +1181,57 @@ static void hclge_dbg_dump_ncl_config(struct hclge_dev *hdev, } } +static void hclge_dbg_dump_loopback(struct hclge_dev *hdev, + const char *cmd_buf) +{ + struct phy_device *phydev = hdev->hw.mac.phydev; + struct hclge_config_mac_mode_cmd *req_app; + struct hclge_serdes_lb_cmd *req_serdes; + struct hclge_desc desc; + u8 loopback_en; + int ret; + + req_app = (struct hclge_config_mac_mode_cmd *)desc.data; + req_serdes = (struct hclge_serdes_lb_cmd *)desc.data; + + dev_info(&hdev->pdev->dev, "mac id: %u\n", hdev->hw.mac.mac_id); + + hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAC_MODE, true); + ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to dump app loopback status, ret = %d\n", ret); + return; + } + + loopback_en = hnae3_get_bit(le32_to_cpu(req_app->txrx_pad_fcs_loop_en), + HCLGE_MAC_APP_LP_B); + dev_info(&hdev->pdev->dev, "app loopback: %s\n", + loopback_en ? "on" : "off"); + + hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_SERDES_LOOPBACK, true); + ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to dump serdes loopback status, ret = %d\n", + ret); + return; + } + + loopback_en = req_serdes->enable & HCLGE_CMD_SERDES_SERIAL_INNER_LOOP_B; + dev_info(&hdev->pdev->dev, "serdes serial loopback: %s\n", + loopback_en ? "on" : "off"); + + loopback_en = req_serdes->enable & + HCLGE_CMD_SERDES_PARALLEL_INNER_LOOP_B; + dev_info(&hdev->pdev->dev, "serdes parallel loopback: %s\n", + loopback_en ? "on" : "off"); + + if (phydev) + dev_info(&hdev->pdev->dev, "phy loopback: %s\n", + phydev->loopback_enabled ? "on" : "off"); +} + /* hclge_dbg_dump_mac_tnl_status: print message about mac tnl interrupt * @hdev: pointer to struct hclge_dev */ @@ -1269,6 +1332,7 @@ int hclge_dbg_run_cmd(struct hnae3_handle *handle, const char *cmd_buf) { #define DUMP_REG "dump reg" #define DUMP_TM_MAP "dump tm map" +#define DUMP_LOOPBACK "dump loopback" struct hclge_vport *vport = hclge_get_vport(handle); struct hclge_dev *hdev = vport->back; @@ -1302,6 +1366,9 @@ int hclge_dbg_run_cmd(struct hnae3_handle *handle, const char *cmd_buf) &cmd_buf[sizeof("dump ncl_config")]); } else if (strncmp(cmd_buf, "dump mac tnl status", 19) == 0) { hclge_dbg_dump_mac_tnl_status(hdev); + } else if (strncmp(cmd_buf, DUMP_LOOPBACK, + strlen(DUMP_LOOPBACK)) == 0) { + hclge_dbg_dump_loopback(hdev, &cmd_buf[sizeof(DUMP_LOOPBACK)]); } else if (strncmp(cmd_buf, "dump qs shaper", 14) == 0) { hclge_dbg_dump_qs_shaper(hdev, &cmd_buf[sizeof("dump qs shaper")]); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c index c85b72dc44d2..50d5ef71756b 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c @@ -1667,9 +1667,6 @@ pci_ers_result_t hclge_handle_hw_ras_error(struct hnae3_ae_dev *ae_dev) hclge_handle_rocee_ras_error(ae_dev); } - if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state)) - goto out; - if (ae_dev->hw_err_reset_req) return PCI_ERS_RESULT_NEED_RESET; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index d3b0cd74ecd2..a758f9ae32be 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -824,6 +824,8 @@ static void hclge_get_mac_stat(struct hnae3_handle *handle, static int hclge_parse_func_status(struct hclge_dev *hdev, struct hclge_func_status_cmd *status) { +#define HCLGE_MAC_ID_MASK 0xF + if (!(status->pf_state & HCLGE_PF_STATE_DONE)) return -EINVAL; @@ -833,6 +835,7 @@ static int hclge_parse_func_status(struct hclge_dev *hdev, else hdev->flag &= ~HCLGE_FLAG_MAIN; + hdev->hw.mac.mac_id = status->mac_id & HCLGE_MAC_ID_MASK; return 0; } @@ -3441,7 +3444,7 @@ static void hclge_do_reset(struct hclge_dev *hdev) u32 val; if (hclge_get_hw_reset_stat(handle)) { - dev_info(&pdev->dev, "Hardware reset not finish\n"); + dev_info(&pdev->dev, "hardware reset not finish\n"); dev_info(&pdev->dev, "func_rst_reg:0x%x, global_rst_reg:0x%x\n", hclge_read_dev(&hdev->hw, HCLGE_FUN_RST_ING), hclge_read_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG)); @@ -3450,20 +3453,20 @@ static void hclge_do_reset(struct hclge_dev *hdev) switch (hdev->reset_type) { case HNAE3_GLOBAL_RESET: + dev_info(&pdev->dev, "global reset requested\n"); val = hclge_read_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG); hnae3_set_bit(val, HCLGE_GLOBAL_RESET_BIT, 1); hclge_write_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG, val); - dev_info(&pdev->dev, "Global Reset requested\n"); break; case HNAE3_FUNC_RESET: - dev_info(&pdev->dev, "PF Reset requested\n"); + dev_info(&pdev->dev, "PF reset requested\n"); /* schedule again to check later */ set_bit(HNAE3_FUNC_RESET, &hdev->reset_pending); hclge_reset_task_schedule(hdev); break; default: dev_warn(&pdev->dev, - "Unsupported reset type: %d\n", hdev->reset_type); + "unsupported reset type: %d\n", hdev->reset_type); break; } } @@ -6765,7 +6768,7 @@ static void hclge_set_timer_task(struct hnae3_handle *handle, bool enable) struct hclge_dev *hdev = vport->back; if (enable) { - hclge_task_schedule(hdev, round_jiffies_relative(HZ)); + hclge_task_schedule(hdev, 0); } else { /* Set the DOWN flag here to disable link updating */ set_bit(HCLGE_STATE_DOWN, &hdev->state); @@ -7353,7 +7356,6 @@ int hclge_add_mc_addr_common(struct hclge_vport *vport, return -EINVAL; } memset(&req, 0, sizeof(req)); - hnae3_set_bit(req.entry_type, HCLGE_MAC_VLAN_BIT0_EN_B, 0); hclge_prepare_mac_addr(&req, addr, true); status = hclge_lookup_mac_vlan_tbl(vport, &req, desc, true); if (status) { @@ -7398,7 +7400,6 @@ int hclge_rm_mc_addr_common(struct hclge_vport *vport, } memset(&req, 0, sizeof(req)); - hnae3_set_bit(req.entry_type, HCLGE_MAC_VLAN_BIT0_EN_B, 0); hclge_prepare_mac_addr(&req, addr, true); status = hclge_lookup_mac_vlan_tbl(vport, &req, desc, true); if (!status) { @@ -7618,11 +7619,17 @@ static int hclge_set_vf_mac(struct hnae3_handle *handle, int vf, } ether_addr_copy(vport->vf_info.mac, mac_addr); - dev_info(&hdev->pdev->dev, - "MAC of VF %d has been set to %pM, and it will be reinitialized!\n", - vf, mac_addr); - return hclge_inform_reset_assert_to_vf(vport); + if (test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) { + dev_info(&hdev->pdev->dev, + "MAC of VF %d has been set to %pM, and it will be reinitialized!\n", + vf, mac_addr); + return hclge_inform_reset_assert_to_vf(vport); + } + + dev_info(&hdev->pdev->dev, "MAC of VF %d has been set to %pM\n", + vf, mac_addr); + return 0; } static int hclge_add_mgr_tbl(struct hclge_dev *hdev, @@ -8979,6 +8986,12 @@ static void hclge_get_media_type(struct hnae3_handle *handle, u8 *media_type, struct hclge_vport *vport = hclge_get_vport(handle); struct hclge_dev *hdev = vport->back; + /* When nic is down, the service task is not running, doesn't update + * the port information per second. Query the port information before + * return the media type, ensure getting the correct media information. + */ + hclge_update_port_info(hdev); + if (media_type) *media_type = hdev->hw.mac.media_type; @@ -9109,8 +9122,8 @@ init_nic_err: static int hclge_init_roce_client_instance(struct hnae3_ae_dev *ae_dev, struct hclge_vport *vport) { - struct hnae3_client *client = vport->roce.client; struct hclge_dev *hdev = ae_dev->priv; + struct hnae3_client *client; int rst_cnt; int ret; @@ -10286,8 +10299,9 @@ static int hclge_dfx_reg_fetch_data(struct hclge_desc *desc_src, int bd_num, static int hclge_get_dfx_reg_len(struct hclge_dev *hdev, int *len) { u32 dfx_reg_type_num = ARRAY_SIZE(hclge_dfx_bd_offset_list); - int data_len_per_desc, data_len, bd_num, i; + int data_len_per_desc, bd_num, i; int bd_num_list[BD_LIST_MAX_NUM]; + u32 data_len; int ret; ret = hclge_get_dfx_reg_bd_num(hdev, bd_num_list, dfx_reg_type_num); @@ -10666,7 +10680,7 @@ static int hclge_init(void) { pr_info("%s is initializing\n", HCLGE_NAME); - hclge_wq = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0, HCLGE_NAME); + hclge_wq = alloc_workqueue("%s", 0, 0, HCLGE_NAME); if (!hclge_wq) { pr_err("%s: failed to create workqueue\n", HCLGE_NAME); return -ENOMEM; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index f78cbb4cc85e..71df23d5f1b4 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -249,6 +249,7 @@ enum HCLGE_MAC_DUPLEX { #define QUERY_ACTIVE_SPEED 1 struct hclge_mac { + u8 mac_id; u8 phy_addr; u8 flag; u8 media_type; /* port media type, e.g. fibre/copper/backplane */ diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c index 3d850f6b1e37..7f24fcb4f96a 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c @@ -5,6 +5,11 @@ #include "hclge_mbx.h" #include "hnae3.h" +static u16 hclge_errno_to_resp(int errno) +{ + return abs(errno); +} + /* hclge_gen_resp_to_vf: used to generate a synchronous response to VF when PF * receives a mailbox message from VF. * @vport: pointer to struct hclge_vport @@ -14,25 +19,25 @@ */ static int hclge_gen_resp_to_vf(struct hclge_vport *vport, struct hclge_mbx_vf_to_pf_cmd *vf_to_pf_req, - int resp_status, - u8 *resp_data, u16 resp_data_len) + struct hclge_respond_to_vf_msg *resp_msg) { struct hclge_mbx_pf_to_vf_cmd *resp_pf_to_vf; struct hclge_dev *hdev = vport->back; enum hclge_cmd_status status; struct hclge_desc desc; + u16 resp; resp_pf_to_vf = (struct hclge_mbx_pf_to_vf_cmd *)desc.data; - if (resp_data_len > HCLGE_MBX_MAX_RESP_DATA_SIZE) { + if (resp_msg->len > HCLGE_MBX_MAX_RESP_DATA_SIZE) { dev_err(&hdev->pdev->dev, "PF fail to gen resp to VF len %u exceeds max len %u\n", - resp_data_len, + resp_msg->len, HCLGE_MBX_MAX_RESP_DATA_SIZE); - /* If resp_data_len is too long, set the value to max length + /* If resp_msg->len is too long, set the value to max length * and return the msg to VF */ - resp_data_len = HCLGE_MBX_MAX_RESP_DATA_SIZE; + resp_msg->len = HCLGE_MBX_MAX_RESP_DATA_SIZE; } hclge_cmd_setup_basic_desc(&desc, HCLGEVF_OPC_MBX_PF_TO_VF, false); @@ -40,18 +45,29 @@ static int hclge_gen_resp_to_vf(struct hclge_vport *vport, resp_pf_to_vf->dest_vfid = vf_to_pf_req->mbx_src_vfid; resp_pf_to_vf->msg_len = vf_to_pf_req->msg_len; - resp_pf_to_vf->msg[0] = HCLGE_MBX_PF_VF_RESP; - resp_pf_to_vf->msg[1] = vf_to_pf_req->msg[0]; - resp_pf_to_vf->msg[2] = vf_to_pf_req->msg[1]; - resp_pf_to_vf->msg[3] = (resp_status == 0) ? 0 : 1; + resp_pf_to_vf->msg.code = HCLGE_MBX_PF_VF_RESP; + resp_pf_to_vf->msg.vf_mbx_msg_code = vf_to_pf_req->msg.code; + resp_pf_to_vf->msg.vf_mbx_msg_subcode = vf_to_pf_req->msg.subcode; + resp = hclge_errno_to_resp(resp_msg->status); + if (resp < SHRT_MAX) { + resp_pf_to_vf->msg.resp_status = resp; + } else { + dev_warn(&hdev->pdev->dev, + "failed to send response to VF, response status %d is out-of-bound\n", + resp); + resp_pf_to_vf->msg.resp_status = EIO; + } - if (resp_data && resp_data_len > 0) - memcpy(&resp_pf_to_vf->msg[4], resp_data, resp_data_len); + if (resp_msg->len > 0) + memcpy(resp_pf_to_vf->msg.resp_data, resp_msg->data, + resp_msg->len); status = hclge_cmd_send(&hdev->hw, &desc, 1); if (status) dev_err(&hdev->pdev->dev, - "PF failed(=%d) to send response to VF\n", status); + "failed to send response to VF, status: %d, vfid: %u, code: %u, subcode: %u.\n", + status, vf_to_pf_req->mbx_src_vfid, + vf_to_pf_req->msg.code, vf_to_pf_req->msg.subcode); return status; } @@ -70,15 +86,15 @@ static int hclge_send_mbx_msg(struct hclge_vport *vport, u8 *msg, u16 msg_len, resp_pf_to_vf->dest_vfid = dest_vfid; resp_pf_to_vf->msg_len = msg_len; - resp_pf_to_vf->msg[0] = mbx_opcode; + resp_pf_to_vf->msg.code = mbx_opcode; - memcpy(&resp_pf_to_vf->msg[1], msg, msg_len); + memcpy(&resp_pf_to_vf->msg.vf_mbx_msg_code, msg, msg_len); status = hclge_cmd_send(&hdev->hw, &desc, 1); if (status) dev_err(&hdev->pdev->dev, - "PF failed(=%d) to send mailbox message to VF\n", - status); + "failed to send mailbox to VF, status: %d, vfid: %u, opcode: %u\n", + status, dest_vfid, mbx_opcode); return status; } @@ -138,21 +154,20 @@ static int hclge_get_ring_chain_from_mbx( { struct hnae3_ring_chain_node *cur_chain, *new_chain; int ring_num; - int i; + int i = 0; - ring_num = req->msg[2]; + ring_num = req->msg.ring_num; - if (ring_num > ((HCLGE_MBX_VF_MSG_DATA_NUM - - HCLGE_MBX_RING_MAP_BASIC_MSG_NUM) / - HCLGE_MBX_RING_NODE_VARIABLE_NUM)) + if (ring_num > HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM) return -ENOMEM; - hnae3_set_bit(ring_chain->flag, HNAE3_RING_TYPE_B, req->msg[3]); + hnae3_set_bit(ring_chain->flag, HNAE3_RING_TYPE_B, + req->msg.param[i].ring_type); ring_chain->tqp_index = - hclge_get_queue_id(vport->nic.kinfo.tqp[req->msg[4]]); + hclge_get_queue_id(vport->nic.kinfo.tqp + [req->msg.param[i].tqp_index]); hnae3_set_field(ring_chain->int_gl_idx, HNAE3_RING_GL_IDX_M, - HNAE3_RING_GL_IDX_S, - req->msg[5]); + HNAE3_RING_GL_IDX_S, req->msg.param[i].int_gl_index); cur_chain = ring_chain; @@ -162,18 +177,15 @@ static int hclge_get_ring_chain_from_mbx( goto err; hnae3_set_bit(new_chain->flag, HNAE3_RING_TYPE_B, - req->msg[HCLGE_MBX_RING_NODE_VARIABLE_NUM * i + - HCLGE_MBX_RING_MAP_BASIC_MSG_NUM]); + req->msg.param[i].ring_type); new_chain->tqp_index = hclge_get_queue_id(vport->nic.kinfo.tqp - [req->msg[HCLGE_MBX_RING_NODE_VARIABLE_NUM * i + - HCLGE_MBX_RING_MAP_BASIC_MSG_NUM + 1]]); + [req->msg.param[i].tqp_index]); hnae3_set_field(new_chain->int_gl_idx, HNAE3_RING_GL_IDX_M, HNAE3_RING_GL_IDX_S, - req->msg[HCLGE_MBX_RING_NODE_VARIABLE_NUM * i + - HCLGE_MBX_RING_MAP_BASIC_MSG_NUM + 2]); + req->msg.param[i].int_gl_index); cur_chain->next = new_chain; cur_chain = new_chain; @@ -189,7 +201,7 @@ static int hclge_map_unmap_ring_to_vf_vector(struct hclge_vport *vport, bool en, struct hclge_mbx_vf_to_pf_cmd *req) { struct hnae3_ring_chain_node ring_chain; - int vector_id = req->msg[1]; + int vector_id = req->msg.vector_id; int ret; memset(&ring_chain, 0, sizeof(ring_chain)); @@ -207,13 +219,9 @@ static int hclge_map_unmap_ring_to_vf_vector(struct hclge_vport *vport, bool en, static int hclge_set_vf_promisc_mode(struct hclge_vport *vport, struct hclge_mbx_vf_to_pf_cmd *req) { -#define HCLGE_MBX_BC_INDEX 1 -#define HCLGE_MBX_UC_INDEX 2 -#define HCLGE_MBX_MC_INDEX 3 - - bool en_bc = req->msg[HCLGE_MBX_BC_INDEX] ? true : false; - bool en_uc = req->msg[HCLGE_MBX_UC_INDEX] ? true : false; - bool en_mc = req->msg[HCLGE_MBX_MC_INDEX] ? true : false; + bool en_bc = req->msg.en_bc ? true : false; + bool en_uc = req->msg.en_uc ? true : false; + bool en_mc = req->msg.en_mc ? true : false; int ret; if (!vport->vf_info.trusted) { @@ -222,8 +230,6 @@ static int hclge_set_vf_promisc_mode(struct hclge_vport *vport, } ret = hclge_set_vport_promisc_mode(vport, en_uc, en_mc, en_bc); - if (req->mbx_need_resp) - hclge_gen_resp_to_vf(vport, req, ret, NULL, 0); vport->vf_info.promisc_enable = (en_uc || en_mc) ? 1 : 0; @@ -244,26 +250,25 @@ void hclge_inform_vf_promisc_info(struct hclge_vport *vport) static int hclge_set_vf_uc_mac_addr(struct hclge_vport *vport, struct hclge_mbx_vf_to_pf_cmd *mbx_req) { - const u8 *mac_addr = (const u8 *)(&mbx_req->msg[2]); +#define HCLGE_MBX_VF_OLD_MAC_ADDR_OFFSET 6 + + const u8 *mac_addr = (const u8 *)(mbx_req->msg.data); struct hclge_dev *hdev = vport->back; int status; - if (mbx_req->msg[1] == HCLGE_MBX_MAC_VLAN_UC_MODIFY) { - const u8 *old_addr = (const u8 *)(&mbx_req->msg[8]); + if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_UC_MODIFY) { + const u8 *old_addr = (const u8 *) + (&mbx_req->msg.data[HCLGE_MBX_VF_OLD_MAC_ADDR_OFFSET]); /* If VF MAC has been configured by the host then it * cannot be overridden by the MAC specified by the VM. */ if (!is_zero_ether_addr(vport->vf_info.mac) && - !ether_addr_equal(mac_addr, vport->vf_info.mac)) { - status = -EPERM; - goto out; - } + !ether_addr_equal(mac_addr, vport->vf_info.mac)) + return -EPERM; - if (!is_valid_ether_addr(mac_addr)) { - status = -EINVAL; - goto out; - } + if (!is_valid_ether_addr(mac_addr)) + return -EINVAL; hclge_rm_uc_addr_common(vport, old_addr); status = hclge_add_uc_addr_common(vport, mac_addr); @@ -275,12 +280,12 @@ static int hclge_set_vf_uc_mac_addr(struct hclge_vport *vport, hclge_add_vport_mac_table(vport, mac_addr, HCLGE_MAC_ADDR_UC); } - } else if (mbx_req->msg[1] == HCLGE_MBX_MAC_VLAN_UC_ADD) { + } else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_UC_ADD) { status = hclge_add_uc_addr_common(vport, mac_addr); if (!status) hclge_add_vport_mac_table(vport, mac_addr, HCLGE_MAC_ADDR_UC); - } else if (mbx_req->msg[1] == HCLGE_MBX_MAC_VLAN_UC_REMOVE) { + } else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_UC_REMOVE) { status = hclge_rm_uc_addr_common(vport, mac_addr); if (!status) hclge_rm_vport_mac_table(vport, mac_addr, @@ -288,33 +293,26 @@ static int hclge_set_vf_uc_mac_addr(struct hclge_vport *vport, } else { dev_err(&hdev->pdev->dev, "failed to set unicast mac addr, unknown subcode %u\n", - mbx_req->msg[1]); + mbx_req->msg.subcode); return -EIO; } -out: - if (mbx_req->mbx_need_resp & HCLGE_MBX_NEED_RESP_BIT) - hclge_gen_resp_to_vf(vport, mbx_req, status, NULL, 0); - - return 0; + return status; } static int hclge_set_vf_mc_mac_addr(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req, - bool gen_resp) + struct hclge_mbx_vf_to_pf_cmd *mbx_req) { - const u8 *mac_addr = (const u8 *)(&mbx_req->msg[2]); + const u8 *mac_addr = (const u8 *)(mbx_req->msg.data); struct hclge_dev *hdev = vport->back; - u8 resp_len = 0; - u8 resp_data; int status; - if (mbx_req->msg[1] == HCLGE_MBX_MAC_VLAN_MC_ADD) { + if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_MC_ADD) { status = hclge_add_mc_addr_common(vport, mac_addr); if (!status) hclge_add_vport_mac_table(vport, mac_addr, HCLGE_MAC_ADDR_MC); - } else if (mbx_req->msg[1] == HCLGE_MBX_MAC_VLAN_MC_REMOVE) { + } else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_MC_REMOVE) { status = hclge_rm_mc_addr_common(vport, mac_addr); if (!status) hclge_rm_vport_mac_table(vport, mac_addr, @@ -322,15 +320,11 @@ static int hclge_set_vf_mc_mac_addr(struct hclge_vport *vport, } else { dev_err(&hdev->pdev->dev, "failed to set mcast mac addr, unknown subcode %u\n", - mbx_req->msg[1]); + mbx_req->msg.subcode); return -EIO; } - if (gen_resp) - hclge_gen_resp_to_vf(vport, mbx_req, status, - &resp_data, resp_len); - - return 0; + return status; } int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid, @@ -351,12 +345,16 @@ int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid, } static int hclge_set_vf_vlan_cfg(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req) + struct hclge_mbx_vf_to_pf_cmd *mbx_req, + struct hclge_respond_to_vf_msg *resp_msg) { +#define HCLGE_MBX_VLAN_STATE_OFFSET 0 +#define HCLGE_MBX_VLAN_INFO_OFFSET 2 + struct hclge_vf_vlan_cfg *msg_cmd; int status = 0; - msg_cmd = (struct hclge_vf_vlan_cfg *)mbx_req->msg; + msg_cmd = (struct hclge_vf_vlan_cfg *)&mbx_req->msg; if (msg_cmd->subcode == HCLGE_MBX_VLAN_FILTER) { struct hnae3_handle *handle = &vport->nic; u16 vlan, proto; @@ -367,38 +365,32 @@ static int hclge_set_vf_vlan_cfg(struct hclge_vport *vport, proto = msg_cmd->proto; status = hclge_set_vlan_filter(handle, cpu_to_be16(proto), vlan, is_kill); - if (mbx_req->mbx_need_resp) - return hclge_gen_resp_to_vf(vport, mbx_req, status, - NULL, 0); } else if (msg_cmd->subcode == HCLGE_MBX_VLAN_RX_OFF_CFG) { struct hnae3_handle *handle = &vport->nic; bool en = msg_cmd->is_kill ? true : false; status = hclge_en_hw_strip_rxvtag(handle, en); - } else if (mbx_req->msg[1] == HCLGE_MBX_PORT_BASE_VLAN_CFG) { + } else if (msg_cmd->subcode == HCLGE_MBX_PORT_BASE_VLAN_CFG) { struct hclge_vlan_info *vlan_info; u16 *state; - state = (u16 *)&mbx_req->msg[2]; - vlan_info = (struct hclge_vlan_info *)&mbx_req->msg[4]; + state = (u16 *)&mbx_req->msg.data[HCLGE_MBX_VLAN_STATE_OFFSET]; + vlan_info = (struct hclge_vlan_info *) + &mbx_req->msg.data[HCLGE_MBX_VLAN_INFO_OFFSET]; status = hclge_update_port_base_vlan_cfg(vport, *state, vlan_info); - } else if (mbx_req->msg[1] == HCLGE_MBX_GET_PORT_BASE_VLAN_STATE) { - u8 state; - - state = vport->port_base_vlan_cfg.state; - status = hclge_gen_resp_to_vf(vport, mbx_req, 0, &state, - sizeof(u8)); + } else if (msg_cmd->subcode == HCLGE_MBX_GET_PORT_BASE_VLAN_STATE) { + resp_msg->data[0] = vport->port_base_vlan_cfg.state; + resp_msg->len = sizeof(u8); } return status; } static int hclge_set_vf_alive(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req, - bool gen_resp) + struct hclge_mbx_vf_to_pf_cmd *mbx_req) { - bool alive = !!mbx_req->msg[2]; + bool alive = !!mbx_req->msg.data[0]; int ret = 0; if (alive) @@ -409,73 +401,76 @@ static int hclge_set_vf_alive(struct hclge_vport *vport, return ret; } -static int hclge_get_vf_tcinfo(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req, - bool gen_resp) +static void hclge_get_vf_tcinfo(struct hclge_vport *vport, + struct hclge_respond_to_vf_msg *resp_msg) { struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo; - u8 vf_tc_map = 0; unsigned int i; - int ret; for (i = 0; i < kinfo->num_tc; i++) - vf_tc_map |= BIT(i); - - ret = hclge_gen_resp_to_vf(vport, mbx_req, 0, &vf_tc_map, - sizeof(vf_tc_map)); + resp_msg->data[0] |= BIT(i); - return ret; + resp_msg->len = sizeof(u8); } -static int hclge_get_vf_queue_info(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req, - bool gen_resp) +static void hclge_get_vf_queue_info(struct hclge_vport *vport, + struct hclge_respond_to_vf_msg *resp_msg) { #define HCLGE_TQPS_RSS_INFO_LEN 6 - u8 resp_data[HCLGE_TQPS_RSS_INFO_LEN]; +#define HCLGE_TQPS_ALLOC_OFFSET 0 +#define HCLGE_TQPS_RSS_SIZE_OFFSET 2 +#define HCLGE_TQPS_RX_BUFFER_LEN_OFFSET 4 + struct hclge_dev *hdev = vport->back; /* get the queue related info */ - memcpy(&resp_data[0], &vport->alloc_tqps, sizeof(u16)); - memcpy(&resp_data[2], &vport->nic.kinfo.rss_size, sizeof(u16)); - memcpy(&resp_data[4], &hdev->rx_buf_len, sizeof(u16)); - - return hclge_gen_resp_to_vf(vport, mbx_req, 0, resp_data, - HCLGE_TQPS_RSS_INFO_LEN); + memcpy(&resp_msg->data[HCLGE_TQPS_ALLOC_OFFSET], + &vport->alloc_tqps, sizeof(u16)); + memcpy(&resp_msg->data[HCLGE_TQPS_RSS_SIZE_OFFSET], + &vport->nic.kinfo.rss_size, sizeof(u16)); + memcpy(&resp_msg->data[HCLGE_TQPS_RX_BUFFER_LEN_OFFSET], + &hdev->rx_buf_len, sizeof(u16)); + resp_msg->len = HCLGE_TQPS_RSS_INFO_LEN; } -static int hclge_get_vf_mac_addr(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req) +static void hclge_get_vf_mac_addr(struct hclge_vport *vport, + struct hclge_respond_to_vf_msg *resp_msg) { - return hclge_gen_resp_to_vf(vport, mbx_req, 0, vport->vf_info.mac, - ETH_ALEN); + ether_addr_copy(resp_msg->data, vport->vf_info.mac); + resp_msg->len = ETH_ALEN; } -static int hclge_get_vf_queue_depth(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req, - bool gen_resp) +static void hclge_get_vf_queue_depth(struct hclge_vport *vport, + struct hclge_respond_to_vf_msg *resp_msg) { #define HCLGE_TQPS_DEPTH_INFO_LEN 4 - u8 resp_data[HCLGE_TQPS_DEPTH_INFO_LEN]; +#define HCLGE_TQPS_NUM_TX_DESC_OFFSET 0 +#define HCLGE_TQPS_NUM_RX_DESC_OFFSET 2 + struct hclge_dev *hdev = vport->back; /* get the queue depth info */ - memcpy(&resp_data[0], &hdev->num_tx_desc, sizeof(u16)); - memcpy(&resp_data[2], &hdev->num_rx_desc, sizeof(u16)); - return hclge_gen_resp_to_vf(vport, mbx_req, 0, resp_data, - HCLGE_TQPS_DEPTH_INFO_LEN); + memcpy(&resp_msg->data[HCLGE_TQPS_NUM_TX_DESC_OFFSET], + &hdev->num_tx_desc, sizeof(u16)); + memcpy(&resp_msg->data[HCLGE_TQPS_NUM_RX_DESC_OFFSET], + &hdev->num_rx_desc, sizeof(u16)); + resp_msg->len = HCLGE_TQPS_DEPTH_INFO_LEN; } -static int hclge_get_vf_media_type(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req) +static void hclge_get_vf_media_type(struct hclge_vport *vport, + struct hclge_respond_to_vf_msg *resp_msg) { +#define HCLGE_VF_MEDIA_TYPE_OFFSET 0 +#define HCLGE_VF_MODULE_TYPE_OFFSET 1 +#define HCLGE_VF_MEDIA_TYPE_LENGTH 2 + struct hclge_dev *hdev = vport->back; - u8 resp_data[2]; - resp_data[0] = hdev->hw.mac.media_type; - resp_data[1] = hdev->hw.mac.module_type; - return hclge_gen_resp_to_vf(vport, mbx_req, 0, resp_data, - sizeof(resp_data)); + resp_msg->data[HCLGE_VF_MEDIA_TYPE_OFFSET] = + hdev->hw.mac.media_type; + resp_msg->data[HCLGE_VF_MODULE_TYPE_OFFSET] = + hdev->hw.mac.module_type; + resp_msg->len = HCLGE_VF_MEDIA_TYPE_LENGTH; } static int hclge_get_link_info(struct hclge_vport *vport, @@ -529,7 +524,7 @@ static void hclge_get_link_mode(struct hclge_vport *vport, advertising = hdev->hw.mac.advertising[0]; supported = hdev->hw.mac.supported[0]; dest_vfid = mbx_req->mbx_src_vfid; - msg_data[0] = mbx_req->msg[2]; + msg_data[0] = mbx_req->msg.data[0]; send_data = msg_data[0] == HCLGE_SUPPORTED ? supported : advertising; @@ -543,29 +538,22 @@ static void hclge_mbx_reset_vf_queue(struct hclge_vport *vport, { u16 queue_id; - memcpy(&queue_id, &mbx_req->msg[2], sizeof(queue_id)); + memcpy(&queue_id, mbx_req->msg.data, sizeof(queue_id)); hclge_reset_vf_queue(vport, queue_id); - - /* send response msg to VF after queue reset complete */ - hclge_gen_resp_to_vf(vport, mbx_req, 0, NULL, 0); } -static void hclge_reset_vf(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req) +static int hclge_reset_vf(struct hclge_vport *vport) { struct hclge_dev *hdev = vport->back; - int ret; dev_warn(&hdev->pdev->dev, "PF received VF reset request from VF %u!", vport->vport_id); - ret = hclge_func_reset_cmd(hdev, vport->vport_id); - hclge_gen_resp_to_vf(vport, mbx_req, ret, NULL, 0); + return hclge_func_reset_cmd(hdev, vport->vport_id); } -static void hclge_vf_keep_alive(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req) +static void hclge_vf_keep_alive(struct hclge_vport *vport) { vport->last_active_jiffies = jiffies; } @@ -573,45 +561,39 @@ static void hclge_vf_keep_alive(struct hclge_vport *vport, static int hclge_set_vf_mtu(struct hclge_vport *vport, struct hclge_mbx_vf_to_pf_cmd *mbx_req) { - int ret; u32 mtu; - memcpy(&mtu, &mbx_req->msg[2], sizeof(mtu)); - ret = hclge_set_vport_mtu(vport, mtu); + memcpy(&mtu, mbx_req->msg.data, sizeof(mtu)); - return hclge_gen_resp_to_vf(vport, mbx_req, ret, NULL, 0); + return hclge_set_vport_mtu(vport, mtu); } -static int hclge_get_queue_id_in_pf(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req) +static void hclge_get_queue_id_in_pf(struct hclge_vport *vport, + struct hclge_mbx_vf_to_pf_cmd *mbx_req, + struct hclge_respond_to_vf_msg *resp_msg) { u16 queue_id, qid_in_pf; - u8 resp_data[2]; - memcpy(&queue_id, &mbx_req->msg[2], sizeof(queue_id)); + memcpy(&queue_id, mbx_req->msg.data, sizeof(queue_id)); qid_in_pf = hclge_covert_handle_qid_global(&vport->nic, queue_id); - memcpy(resp_data, &qid_in_pf, sizeof(qid_in_pf)); - - return hclge_gen_resp_to_vf(vport, mbx_req, 0, resp_data, - sizeof(resp_data)); + memcpy(resp_msg->data, &qid_in_pf, sizeof(qid_in_pf)); + resp_msg->len = sizeof(qid_in_pf); } -static int hclge_get_rss_key(struct hclge_vport *vport, - struct hclge_mbx_vf_to_pf_cmd *mbx_req) +static void hclge_get_rss_key(struct hclge_vport *vport, + struct hclge_mbx_vf_to_pf_cmd *mbx_req, + struct hclge_respond_to_vf_msg *resp_msg) { #define HCLGE_RSS_MBX_RESP_LEN 8 - u8 resp_data[HCLGE_RSS_MBX_RESP_LEN]; struct hclge_dev *hdev = vport->back; u8 index; - index = mbx_req->msg[2]; + index = mbx_req->msg.data[0]; - memcpy(&resp_data[0], + memcpy(resp_msg->data, &hdev->vport[0].rss_hash_key[index * HCLGE_RSS_MBX_RESP_LEN], HCLGE_RSS_MBX_RESP_LEN); - - return hclge_gen_resp_to_vf(vport, mbx_req, 0, resp_data, - HCLGE_RSS_MBX_RESP_LEN); + resp_msg->len = HCLGE_RSS_MBX_RESP_LEN; } static void hclge_link_fail_parse(struct hclge_dev *hdev, u8 link_fail_code) @@ -634,13 +616,10 @@ static void hclge_link_fail_parse(struct hclge_dev *hdev, u8 link_fail_code) static void hclge_handle_link_change_event(struct hclge_dev *hdev, struct hclge_mbx_vf_to_pf_cmd *req) { -#define LINK_STATUS_OFFSET 1 -#define LINK_FAIL_CODE_OFFSET 2 - hclge_task_schedule(hdev, 0); - if (!req->msg[LINK_STATUS_OFFSET]) - hclge_link_fail_parse(hdev, req->msg[LINK_FAIL_CODE_OFFSET]); + if (!req->msg.subcode) + hclge_link_fail_parse(hdev, req->msg.data[0]); } static bool hclge_cmd_crq_empty(struct hclge_hw *hw) @@ -662,12 +641,14 @@ static void hclge_handle_ncsi_error(struct hclge_dev *hdev) void hclge_mbx_handler(struct hclge_dev *hdev) { struct hclge_cmq_ring *crq = &hdev->hw.cmq.crq; + struct hclge_respond_to_vf_msg resp_msg; struct hclge_mbx_vf_to_pf_cmd *req; struct hclge_vport *vport; struct hclge_desc *desc; unsigned int flag; - int ret; + int ret = 0; + memset(&resp_msg, 0, sizeof(resp_msg)); /* handle all the mailbox requests in the queue */ while (!hclge_cmd_crq_empty(&hdev->hw)) { if (test_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state)) { @@ -683,7 +664,7 @@ void hclge_mbx_handler(struct hclge_dev *hdev) if (unlikely(!hnae3_get_bit(flag, HCLGE_CMDQ_RX_OUTVLD_B))) { dev_warn(&hdev->pdev->dev, "dropped invalid mailbox message, code = %u\n", - req->msg[0]); + req->msg.code); /* dropping/not processing this invalid message */ crq->desc[crq->next_to_use].flag = 0; @@ -693,7 +674,7 @@ void hclge_mbx_handler(struct hclge_dev *hdev) vport = &hdev->vport[req->mbx_src_vfid]; - switch (req->msg[0]) { + switch (req->msg.code) { case HCLGE_MBX_MAP_RING_TO_VECTOR: ret = hclge_map_unmap_ring_to_vf_vector(vport, true, req); @@ -717,47 +698,34 @@ void hclge_mbx_handler(struct hclge_dev *hdev) ret); break; case HCLGE_MBX_SET_MULTICAST: - ret = hclge_set_vf_mc_mac_addr(vport, req, false); + ret = hclge_set_vf_mc_mac_addr(vport, req); if (ret) dev_err(&hdev->pdev->dev, "PF fail(%d) to set VF MC MAC Addr\n", ret); break; case HCLGE_MBX_SET_VLAN: - ret = hclge_set_vf_vlan_cfg(vport, req); + ret = hclge_set_vf_vlan_cfg(vport, req, &resp_msg); if (ret) dev_err(&hdev->pdev->dev, "PF failed(%d) to config VF's VLAN\n", ret); break; case HCLGE_MBX_SET_ALIVE: - ret = hclge_set_vf_alive(vport, req, false); + ret = hclge_set_vf_alive(vport, req); if (ret) dev_err(&hdev->pdev->dev, "PF failed(%d) to set VF's ALIVE\n", ret); break; case HCLGE_MBX_GET_QINFO: - ret = hclge_get_vf_queue_info(vport, req, true); - if (ret) - dev_err(&hdev->pdev->dev, - "PF failed(%d) to get Q info for VF\n", - ret); + hclge_get_vf_queue_info(vport, &resp_msg); break; case HCLGE_MBX_GET_QDEPTH: - ret = hclge_get_vf_queue_depth(vport, req, true); - if (ret) - dev_err(&hdev->pdev->dev, - "PF failed(%d) to get Q depth for VF\n", - ret); + hclge_get_vf_queue_depth(vport, &resp_msg); break; - case HCLGE_MBX_GET_TCINFO: - ret = hclge_get_vf_tcinfo(vport, req, true); - if (ret) - dev_err(&hdev->pdev->dev, - "PF failed(%d) to get TC info for VF\n", - ret); + hclge_get_vf_tcinfo(vport, &resp_msg); break; case HCLGE_MBX_GET_LINK_STATUS: ret = hclge_get_link_info(vport, req); @@ -770,10 +738,10 @@ void hclge_mbx_handler(struct hclge_dev *hdev) hclge_mbx_reset_vf_queue(vport, req); break; case HCLGE_MBX_RESET: - hclge_reset_vf(vport, req); + ret = hclge_reset_vf(vport); break; case HCLGE_MBX_KEEP_ALIVE: - hclge_vf_keep_alive(vport, req); + hclge_vf_keep_alive(vport); break; case HCLGE_MBX_SET_MTU: ret = hclge_set_vf_mtu(vport, req); @@ -782,18 +750,10 @@ void hclge_mbx_handler(struct hclge_dev *hdev) "VF fail(%d) to set mtu\n", ret); break; case HCLGE_MBX_GET_QID_IN_PF: - ret = hclge_get_queue_id_in_pf(vport, req); - if (ret) - dev_err(&hdev->pdev->dev, - "PF failed(%d) to get qid for VF\n", - ret); + hclge_get_queue_id_in_pf(vport, req, &resp_msg); break; case HCLGE_MBX_GET_RSS_KEY: - ret = hclge_get_rss_key(vport, req); - if (ret) - dev_err(&hdev->pdev->dev, - "PF fail(%d) to get rss key for VF\n", - ret); + hclge_get_rss_key(vport, req, &resp_msg); break; case HCLGE_MBX_GET_LINK_MODE: hclge_get_link_mode(vport, req); @@ -807,21 +767,13 @@ void hclge_mbx_handler(struct hclge_dev *hdev) hclge_rm_vport_all_vlan_table(vport, true); break; case HCLGE_MBX_GET_MEDIA_TYPE: - ret = hclge_get_vf_media_type(vport, req); - if (ret) - dev_err(&hdev->pdev->dev, - "PF fail(%d) to media type for VF\n", - ret); + hclge_get_vf_media_type(vport, &resp_msg); break; case HCLGE_MBX_PUSH_LINK_STATUS: hclge_handle_link_change_event(hdev, req); break; case HCLGE_MBX_GET_MAC_ADDR: - ret = hclge_get_vf_mac_addr(vport, req); - if (ret) - dev_err(&hdev->pdev->dev, - "PF failed(%d) to get MAC for VF\n", - ret); + hclge_get_vf_mac_addr(vport, &resp_msg); break; case HCLGE_MBX_NCSI_ERROR: hclge_handle_ncsi_error(hdev); @@ -829,11 +781,22 @@ void hclge_mbx_handler(struct hclge_dev *hdev) default: dev_err(&hdev->pdev->dev, "un-supported mailbox message, code = %u\n", - req->msg[0]); + req->msg.code); break; } + + /* PF driver should not reply IMP */ + if (hnae3_get_bit(req->mbx_need_resp, HCLGE_MBX_NEED_RESP_B) && + req->msg.code < HCLGE_MBX_GET_VF_FLR_STATUS) { + resp_msg.status = ret; + hclge_gen_resp_to_vf(vport, req, &resp_msg); + } + crq->desc[crq->next_to_use].flag = 0; hclge_mbx_ring_ptr_move_crq(crq); + + /* reinitialize ret after complete the mbx message processing */ + ret = 0; } /* Write back CMDQ_RQ header pointer, M7 need this pointer */ diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index 0510d85a7f6a..e02d427131ee 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -229,13 +229,25 @@ static void hclgevf_get_stats(struct hnae3_handle *handle, u64 *data) hclgevf_tqps_get_stats(handle, data); } +static void hclgevf_build_send_msg(struct hclge_vf_to_pf_msg *msg, u8 code, + u8 subcode) +{ + if (msg) { + memset(msg, 0, sizeof(struct hclge_vf_to_pf_msg)); + msg->code = code; + msg->subcode = subcode; + } +} + static int hclgevf_get_tc_info(struct hclgevf_dev *hdev) { + struct hclge_vf_to_pf_msg send_msg; u8 resp_msg; int status; - status = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_TCINFO, 0, NULL, 0, - true, &resp_msg, sizeof(resp_msg)); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_TCINFO, 0); + status = hclgevf_send_mbx_msg(hdev, &send_msg, true, &resp_msg, + sizeof(resp_msg)); if (status) { dev_err(&hdev->pdev->dev, "VF request to get TC info from PF failed %d", @@ -251,12 +263,14 @@ static int hclgevf_get_tc_info(struct hclgevf_dev *hdev) static int hclgevf_get_port_base_vlan_filter_state(struct hclgevf_dev *hdev) { struct hnae3_handle *nic = &hdev->nic; + struct hclge_vf_to_pf_msg send_msg; u8 resp_msg; int ret; - ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_VLAN, - HCLGE_MBX_GET_PORT_BASE_VLAN_STATE, - NULL, 0, true, &resp_msg, sizeof(u8)); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN, + HCLGE_MBX_GET_PORT_BASE_VLAN_STATE); + ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, &resp_msg, + sizeof(u8)); if (ret) { dev_err(&hdev->pdev->dev, "VF request to get port based vlan state failed %d", @@ -272,11 +286,16 @@ static int hclgevf_get_port_base_vlan_filter_state(struct hclgevf_dev *hdev) static int hclgevf_get_queue_info(struct hclgevf_dev *hdev) { #define HCLGEVF_TQPS_RSS_INFO_LEN 6 +#define HCLGEVF_TQPS_ALLOC_OFFSET 0 +#define HCLGEVF_TQPS_RSS_SIZE_OFFSET 2 +#define HCLGEVF_TQPS_RX_BUFFER_LEN_OFFSET 4 + u8 resp_msg[HCLGEVF_TQPS_RSS_INFO_LEN]; + struct hclge_vf_to_pf_msg send_msg; int status; - status = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_QINFO, 0, NULL, 0, - true, resp_msg, + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_QINFO, 0); + status = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg, HCLGEVF_TQPS_RSS_INFO_LEN); if (status) { dev_err(&hdev->pdev->dev, @@ -285,9 +304,12 @@ static int hclgevf_get_queue_info(struct hclgevf_dev *hdev) return status; } - memcpy(&hdev->num_tqps, &resp_msg[0], sizeof(u16)); - memcpy(&hdev->rss_size_max, &resp_msg[2], sizeof(u16)); - memcpy(&hdev->rx_buf_len, &resp_msg[4], sizeof(u16)); + memcpy(&hdev->num_tqps, &resp_msg[HCLGEVF_TQPS_ALLOC_OFFSET], + sizeof(u16)); + memcpy(&hdev->rss_size_max, &resp_msg[HCLGEVF_TQPS_RSS_SIZE_OFFSET], + sizeof(u16)); + memcpy(&hdev->rx_buf_len, &resp_msg[HCLGEVF_TQPS_RX_BUFFER_LEN_OFFSET], + sizeof(u16)); return 0; } @@ -295,11 +317,15 @@ static int hclgevf_get_queue_info(struct hclgevf_dev *hdev) static int hclgevf_get_queue_depth(struct hclgevf_dev *hdev) { #define HCLGEVF_TQPS_DEPTH_INFO_LEN 4 +#define HCLGEVF_TQPS_NUM_TX_DESC_OFFSET 0 +#define HCLGEVF_TQPS_NUM_RX_DESC_OFFSET 2 + u8 resp_msg[HCLGEVF_TQPS_DEPTH_INFO_LEN]; + struct hclge_vf_to_pf_msg send_msg; int ret; - ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_QDEPTH, 0, NULL, 0, - true, resp_msg, + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_QDEPTH, 0); + ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg, HCLGEVF_TQPS_DEPTH_INFO_LEN); if (ret) { dev_err(&hdev->pdev->dev, @@ -308,8 +334,10 @@ static int hclgevf_get_queue_depth(struct hclgevf_dev *hdev) return ret; } - memcpy(&hdev->num_tx_desc, &resp_msg[0], sizeof(u16)); - memcpy(&hdev->num_rx_desc, &resp_msg[2], sizeof(u16)); + memcpy(&hdev->num_tx_desc, &resp_msg[HCLGEVF_TQPS_NUM_TX_DESC_OFFSET], + sizeof(u16)); + memcpy(&hdev->num_rx_desc, &resp_msg[HCLGEVF_TQPS_NUM_RX_DESC_OFFSET], + sizeof(u16)); return 0; } @@ -317,14 +345,14 @@ static int hclgevf_get_queue_depth(struct hclgevf_dev *hdev) static u16 hclgevf_get_qid_global(struct hnae3_handle *handle, u16 queue_id) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - u8 msg_data[2], resp_data[2]; + struct hclge_vf_to_pf_msg send_msg; u16 qid_in_pf = 0; + u8 resp_data[2]; int ret; - memcpy(&msg_data[0], &queue_id, sizeof(queue_id)); - - ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_QID_IN_PF, 0, msg_data, - sizeof(msg_data), true, resp_data, + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_QID_IN_PF, 0); + memcpy(send_msg.data, &queue_id, sizeof(queue_id)); + ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_data, sizeof(resp_data)); if (!ret) qid_in_pf = *(u16 *)resp_data; @@ -334,11 +362,13 @@ static u16 hclgevf_get_qid_global(struct hnae3_handle *handle, u16 queue_id) static int hclgevf_get_pf_media_type(struct hclgevf_dev *hdev) { + struct hclge_vf_to_pf_msg send_msg; u8 resp_msg[2]; int ret; - ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_MEDIA_TYPE, 0, NULL, 0, - true, resp_msg, sizeof(resp_msg)); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_MEDIA_TYPE, 0); + ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg, + sizeof(resp_msg)); if (ret) { dev_err(&hdev->pdev->dev, "VF request to get the pf port media type failed %d", @@ -425,11 +455,11 @@ static int hclgevf_knic_setup(struct hclgevf_dev *hdev) static void hclgevf_request_link_info(struct hclgevf_dev *hdev) { + struct hclge_vf_to_pf_msg send_msg; int status; - u8 resp_msg; - status = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_LINK_STATUS, 0, NULL, - 0, false, &resp_msg, sizeof(resp_msg)); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_LINK_STATUS, 0); + status = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); if (status) dev_err(&hdev->pdev->dev, "VF failed to fetch link status(%d) from PF", status); @@ -463,19 +493,16 @@ void hclgevf_update_link_status(struct hclgevf_dev *hdev, int link_state) static void hclgevf_update_link_mode(struct hclgevf_dev *hdev) { -#define HCLGEVF_ADVERTISING 0 -#define HCLGEVF_SUPPORTED 1 - u8 send_msg; - u8 resp_msg; +#define HCLGEVF_ADVERTISING 0 +#define HCLGEVF_SUPPORTED 1 - send_msg = HCLGEVF_ADVERTISING; - hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_LINK_MODE, 0, - &send_msg, sizeof(send_msg), false, - &resp_msg, sizeof(resp_msg)); - send_msg = HCLGEVF_SUPPORTED; - hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_LINK_MODE, 0, - &send_msg, sizeof(send_msg), false, - &resp_msg, sizeof(resp_msg)); + struct hclge_vf_to_pf_msg send_msg; + + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_LINK_MODE, 0); + send_msg.data[0] = HCLGEVF_ADVERTISING; + hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); + send_msg.data[0] = HCLGEVF_SUPPORTED; + hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); } static int hclgevf_set_handle_info(struct hclgevf_dev *hdev) @@ -677,19 +704,19 @@ static int hclgevf_set_rss_tc_mode(struct hclgevf_dev *hdev, u16 rss_size) static int hclgevf_get_rss_hash_key(struct hclgevf_dev *hdev) { #define HCLGEVF_RSS_MBX_RESP_LEN 8 - struct hclgevf_rss_cfg *rss_cfg = &hdev->rss_cfg; u8 resp_msg[HCLGEVF_RSS_MBX_RESP_LEN]; + struct hclge_vf_to_pf_msg send_msg; u16 msg_num, hash_key_index; u8 index; int ret; + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_RSS_KEY, 0); msg_num = (HCLGEVF_RSS_KEY_SIZE + HCLGEVF_RSS_MBX_RESP_LEN - 1) / HCLGEVF_RSS_MBX_RESP_LEN; for (index = 0; index < msg_num; index++) { - ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_RSS_KEY, 0, - &index, sizeof(index), - true, resp_msg, + send_msg.data[0] = index; + ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, resp_msg, HCLGEVF_RSS_MBX_RESP_LEN); if (ret) { dev_err(&hdev->pdev->dev, @@ -1001,44 +1028,32 @@ static int hclgevf_bind_ring_to_vector(struct hnae3_handle *handle, bool en, struct hnae3_ring_chain_node *ring_chain) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + struct hclge_vf_to_pf_msg send_msg; struct hnae3_ring_chain_node *node; - struct hclge_mbx_vf_to_pf_cmd *req; - struct hclgevf_desc desc; - int i = 0; int status; - u8 type; + int i = 0; - req = (struct hclge_mbx_vf_to_pf_cmd *)desc.data; - type = en ? HCLGE_MBX_MAP_RING_TO_VECTOR : + memset(&send_msg, 0, sizeof(send_msg)); + send_msg.code = en ? HCLGE_MBX_MAP_RING_TO_VECTOR : HCLGE_MBX_UNMAP_RING_TO_VECTOR; + send_msg.vector_id = vector_id; for (node = ring_chain; node; node = node->next) { - int idx_offset = HCLGE_MBX_RING_MAP_BASIC_MSG_NUM + - HCLGE_MBX_RING_NODE_VARIABLE_NUM * i; - - if (i == 0) { - hclgevf_cmd_setup_basic_desc(&desc, - HCLGEVF_OPC_MBX_VF_TO_PF, - false); - req->msg[0] = type; - req->msg[1] = vector_id; - } - - req->msg[idx_offset] = + send_msg.param[i].ring_type = hnae3_get_bit(node->flag, HNAE3_RING_TYPE_B); - req->msg[idx_offset + 1] = node->tqp_index; - req->msg[idx_offset + 2] = hnae3_get_field(node->int_gl_idx, - HNAE3_RING_GL_IDX_M, - HNAE3_RING_GL_IDX_S); + + send_msg.param[i].tqp_index = node->tqp_index; + send_msg.param[i].int_gl_index = + hnae3_get_field(node->int_gl_idx, + HNAE3_RING_GL_IDX_M, + HNAE3_RING_GL_IDX_S); i++; - if ((i == (HCLGE_MBX_VF_MSG_DATA_NUM - - HCLGE_MBX_RING_MAP_BASIC_MSG_NUM) / - HCLGE_MBX_RING_NODE_VARIABLE_NUM) || - !node->next) { - req->msg[2] = i; + if (i == HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM || !node->next) { + send_msg.ring_num = i; - status = hclgevf_cmd_send(&hdev->hw, &desc, 1); + status = hclgevf_send_mbx_msg(hdev, &send_msg, false, + NULL, 0); if (status) { dev_err(&hdev->pdev->dev, "Map TQP fail, status is %d.\n", @@ -1046,11 +1061,6 @@ static int hclgevf_bind_ring_to_vector(struct hnae3_handle *handle, bool en, return status; } i = 0; - hclgevf_cmd_setup_basic_desc(&desc, - HCLGEVF_OPC_MBX_VF_TO_PF, - false); - req->msg[0] = type; - req->msg[1] = vector_id; } } @@ -1123,18 +1133,17 @@ static int hclgevf_cmd_set_promisc_mode(struct hclgevf_dev *hdev, bool en_uc_pmc, bool en_mc_pmc, bool en_bc_pmc) { - struct hclge_mbx_vf_to_pf_cmd *req; - struct hclgevf_desc desc; + struct hclge_vf_to_pf_msg send_msg; int ret; - req = (struct hclge_mbx_vf_to_pf_cmd *)desc.data; - hclgevf_cmd_setup_basic_desc(&desc, HCLGEVF_OPC_MBX_VF_TO_PF, false); - req->msg[0] = HCLGE_MBX_SET_PROMISC_MODE; - req->msg[1] = en_bc_pmc ? 1 : 0; - req->msg[2] = en_uc_pmc ? 1 : 0; - req->msg[3] = en_mc_pmc ? 1 : 0; + memset(&send_msg, 0, sizeof(send_msg)); + send_msg.code = HCLGE_MBX_SET_PROMISC_MODE; + send_msg.en_bc = en_bc_pmc ? 1 : 0; + send_msg.en_uc = en_uc_pmc ? 1 : 0; + send_msg.en_mc = en_mc_pmc ? 1 : 0; + + ret = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); - ret = hclgevf_cmd_send(&hdev->hw, &desc, 1); if (ret) dev_err(&hdev->pdev->dev, "Set promisc mode fail, status is %d.\n", ret); @@ -1193,11 +1202,13 @@ static void hclgevf_reset_tqp_stats(struct hnae3_handle *handle) static int hclgevf_get_host_mac_addr(struct hclgevf_dev *hdev, u8 *p) { + struct hclge_vf_to_pf_msg send_msg; u8 host_mac[ETH_ALEN]; int status; - status = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_GET_MAC_ADDR, 0, NULL, 0, - true, host_mac, ETH_ALEN); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_GET_MAC_ADDR, 0); + status = hclgevf_send_mbx_msg(hdev, &send_msg, true, host_mac, + ETH_ALEN); if (status) { dev_err(&hdev->pdev->dev, "fail to get VF MAC from host %d", status); @@ -1229,20 +1240,16 @@ static int hclgevf_set_mac_addr(struct hnae3_handle *handle, void *p, { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); u8 *old_mac_addr = (u8 *)hdev->hw.mac.mac_addr; + struct hclge_vf_to_pf_msg send_msg; u8 *new_mac_addr = (u8 *)p; - u8 msg_data[ETH_ALEN * 2]; - u16 subcode; int status; - ether_addr_copy(msg_data, new_mac_addr); - ether_addr_copy(&msg_data[ETH_ALEN], old_mac_addr); - - subcode = is_first ? HCLGE_MBX_MAC_VLAN_UC_ADD : + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_UNICAST, 0); + send_msg.subcode = is_first ? HCLGE_MBX_MAC_VLAN_UC_ADD : HCLGE_MBX_MAC_VLAN_UC_MODIFY; - - status = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_UNICAST, - subcode, msg_data, sizeof(msg_data), - true, NULL, 0); + ether_addr_copy(send_msg.data, new_mac_addr); + ether_addr_copy(&send_msg.data[ETH_ALEN], old_mac_addr); + status = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0); if (!status) ether_addr_copy(hdev->hw.mac.mac_addr, new_mac_addr); @@ -1253,49 +1260,60 @@ static int hclgevf_add_uc_addr(struct hnae3_handle *handle, const unsigned char *addr) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + struct hclge_vf_to_pf_msg send_msg; - return hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_UNICAST, - HCLGE_MBX_MAC_VLAN_UC_ADD, - addr, ETH_ALEN, false, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_UNICAST, + HCLGE_MBX_MAC_VLAN_UC_ADD); + ether_addr_copy(send_msg.data, addr); + return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); } static int hclgevf_rm_uc_addr(struct hnae3_handle *handle, const unsigned char *addr) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + struct hclge_vf_to_pf_msg send_msg; - return hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_UNICAST, - HCLGE_MBX_MAC_VLAN_UC_REMOVE, - addr, ETH_ALEN, false, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_UNICAST, + HCLGE_MBX_MAC_VLAN_UC_REMOVE); + ether_addr_copy(send_msg.data, addr); + return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); } static int hclgevf_add_mc_addr(struct hnae3_handle *handle, const unsigned char *addr) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + struct hclge_vf_to_pf_msg send_msg; - return hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_MULTICAST, - HCLGE_MBX_MAC_VLAN_MC_ADD, - addr, ETH_ALEN, false, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_MULTICAST, + HCLGE_MBX_MAC_VLAN_MC_ADD); + ether_addr_copy(send_msg.data, addr); + return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); } static int hclgevf_rm_mc_addr(struct hnae3_handle *handle, const unsigned char *addr) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + struct hclge_vf_to_pf_msg send_msg; - return hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_MULTICAST, - HCLGE_MBX_MAC_VLAN_MC_REMOVE, - addr, ETH_ALEN, false, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_MULTICAST, + HCLGE_MBX_MAC_VLAN_MC_REMOVE); + ether_addr_copy(send_msg.data, addr); + return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); } static int hclgevf_set_vlan_filter(struct hnae3_handle *handle, __be16 proto, u16 vlan_id, bool is_kill) { -#define HCLGEVF_VLAN_MBX_MSG_LEN 5 +#define HCLGEVF_VLAN_MBX_IS_KILL_OFFSET 0 +#define HCLGEVF_VLAN_MBX_VLAN_ID_OFFSET 1 +#define HCLGEVF_VLAN_MBX_PROTO_OFFSET 3 + struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - u8 msg_data[HCLGEVF_VLAN_MBX_MSG_LEN]; + struct hclge_vf_to_pf_msg send_msg; int ret; if (vlan_id > HCLGEVF_MAX_VLAN_ID) @@ -1313,16 +1331,18 @@ static int hclgevf_set_vlan_filter(struct hnae3_handle *handle, return -EBUSY; } - msg_data[0] = is_kill; - memcpy(&msg_data[1], &vlan_id, sizeof(vlan_id)); - memcpy(&msg_data[3], &proto, sizeof(proto)); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN, + HCLGE_MBX_VLAN_FILTER); + send_msg.data[HCLGEVF_VLAN_MBX_IS_KILL_OFFSET] = is_kill; + memcpy(&send_msg.data[HCLGEVF_VLAN_MBX_VLAN_ID_OFFSET], &vlan_id, + sizeof(vlan_id)); + memcpy(&send_msg.data[HCLGEVF_VLAN_MBX_PROTO_OFFSET], &proto, + sizeof(proto)); /* when remove hw vlan filter failed, record the vlan id, * and try to remove it from hw later, to be consistence * with stack. */ - ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_VLAN, - HCLGE_MBX_VLAN_FILTER, msg_data, - HCLGEVF_VLAN_MBX_MSG_LEN, true, NULL, 0); + ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0); if (is_kill && ret) set_bit(vlan_id, hdev->vlan_del_fail_bmap); @@ -1355,37 +1375,38 @@ static void hclgevf_sync_vlan_filter(struct hclgevf_dev *hdev) static int hclgevf_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - u8 msg_data; + struct hclge_vf_to_pf_msg send_msg; - msg_data = enable ? 1 : 0; - return hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_VLAN, - HCLGE_MBX_VLAN_RX_OFF_CFG, &msg_data, - 1, false, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN, + HCLGE_MBX_VLAN_RX_OFF_CFG); + send_msg.data[0] = enable ? 1 : 0; + return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); } static int hclgevf_reset_tqp(struct hnae3_handle *handle, u16 queue_id) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - u8 msg_data[2]; + struct hclge_vf_to_pf_msg send_msg; int ret; - memcpy(msg_data, &queue_id, sizeof(queue_id)); - /* disable vf queue before send queue reset msg to PF */ ret = hclgevf_tqp_enable(hdev, queue_id, 0, false); if (ret) return ret; - return hclgevf_send_mbx_msg(hdev, HCLGE_MBX_QUEUE_RESET, 0, msg_data, - sizeof(msg_data), true, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_QUEUE_RESET, 0); + memcpy(send_msg.data, &queue_id, sizeof(queue_id)); + return hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0); } static int hclgevf_set_mtu(struct hnae3_handle *handle, int new_mtu) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + struct hclge_vf_to_pf_msg send_msg; - return hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_MTU, 0, (u8 *)&new_mtu, - sizeof(new_mtu), true, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_MTU, 0); + memcpy(send_msg.data, &new_mtu, sizeof(new_mtu)); + return hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0); } static int hclgevf_notify_client(struct hclgevf_dev *hdev, @@ -1500,11 +1521,12 @@ static int hclgevf_reset_prepare_wait(struct hclgevf_dev *hdev) { #define HCLGEVF_RESET_SYNC_TIME 100 + struct hclge_vf_to_pf_msg send_msg; int ret = 0; if (hdev->reset_type == HNAE3_VF_FUNC_RESET) { - ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_RESET, 0, NULL, - 0, true, NULL, sizeof(u8)); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_RESET, 0); + ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0); hdev->rst_stats.vf_func_rst_cnt++; } @@ -1881,14 +1903,14 @@ static void hclgevf_mailbox_service_task(struct hclgevf_dev *hdev) static void hclgevf_keep_alive(struct hclgevf_dev *hdev) { - u8 respmsg; + struct hclge_vf_to_pf_msg send_msg; int ret; if (test_bit(HCLGEVF_STATE_CMD_DISABLE, &hdev->state)) return; - ret = hclgevf_send_mbx_msg(hdev, HCLGE_MBX_KEEP_ALIVE, 0, NULL, - 0, false, &respmsg, sizeof(respmsg)); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_KEEP_ALIVE, 0); + ret = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); if (ret) dev_err(&hdev->pdev->dev, "VF sends keep alive cmd failed(=%d)\n", ret); @@ -2002,7 +2024,10 @@ static enum hclgevf_evt_cause hclgevf_check_evt_cause(struct hclgevf_dev *hdev, return HCLGEVF_VECTOR0_EVENT_MBX; } - dev_dbg(&hdev->pdev->dev, "vector 0 interrupt from unknown source\n"); + /* print other vector0 event source */ + dev_info(&hdev->pdev->dev, + "vector 0 interrupt from unknown source, cmdq_src = %#x\n", + cmdq_stat_reg); return HCLGEVF_VECTOR0_EVENT_OTHER; } @@ -2124,50 +2149,51 @@ static int hclgevf_config_gro(struct hclgevf_dev *hdev, bool en) return ret; } -static int hclgevf_rss_init_hw(struct hclgevf_dev *hdev) +static void hclgevf_rss_init_cfg(struct hclgevf_dev *hdev) { struct hclgevf_rss_cfg *rss_cfg = &hdev->rss_cfg; - int ret; + struct hclgevf_rss_tuple_cfg *tuple_sets; u32 i; + rss_cfg->hash_algo = HCLGEVF_RSS_HASH_ALGO_TOEPLITZ; rss_cfg->rss_size = hdev->nic.kinfo.rss_size; - + tuple_sets = &rss_cfg->rss_tuple_sets; if (hdev->pdev->revision >= 0x21) { rss_cfg->hash_algo = HCLGEVF_RSS_HASH_ALGO_SIMPLE; memcpy(rss_cfg->rss_hash_key, hclgevf_hash_key, HCLGEVF_RSS_KEY_SIZE); + tuple_sets->ipv4_tcp_en = HCLGEVF_RSS_INPUT_TUPLE_OTHER; + tuple_sets->ipv4_udp_en = HCLGEVF_RSS_INPUT_TUPLE_OTHER; + tuple_sets->ipv4_sctp_en = HCLGEVF_RSS_INPUT_TUPLE_SCTP; + tuple_sets->ipv4_fragment_en = HCLGEVF_RSS_INPUT_TUPLE_OTHER; + tuple_sets->ipv6_tcp_en = HCLGEVF_RSS_INPUT_TUPLE_OTHER; + tuple_sets->ipv6_udp_en = HCLGEVF_RSS_INPUT_TUPLE_OTHER; + tuple_sets->ipv6_sctp_en = HCLGEVF_RSS_INPUT_TUPLE_SCTP; + tuple_sets->ipv6_fragment_en = HCLGEVF_RSS_INPUT_TUPLE_OTHER; + } + + /* Initialize RSS indirect table */ + for (i = 0; i < HCLGEVF_RSS_IND_TBL_SIZE; i++) + rss_cfg->rss_indirection_tbl[i] = i % rss_cfg->rss_size; +} + +static int hclgevf_rss_init_hw(struct hclgevf_dev *hdev) +{ + struct hclgevf_rss_cfg *rss_cfg = &hdev->rss_cfg; + int ret; + + if (hdev->pdev->revision >= 0x21) { ret = hclgevf_set_rss_algo_key(hdev, rss_cfg->hash_algo, rss_cfg->rss_hash_key); if (ret) return ret; - rss_cfg->rss_tuple_sets.ipv4_tcp_en = - HCLGEVF_RSS_INPUT_TUPLE_OTHER; - rss_cfg->rss_tuple_sets.ipv4_udp_en = - HCLGEVF_RSS_INPUT_TUPLE_OTHER; - rss_cfg->rss_tuple_sets.ipv4_sctp_en = - HCLGEVF_RSS_INPUT_TUPLE_SCTP; - rss_cfg->rss_tuple_sets.ipv4_fragment_en = - HCLGEVF_RSS_INPUT_TUPLE_OTHER; - rss_cfg->rss_tuple_sets.ipv6_tcp_en = - HCLGEVF_RSS_INPUT_TUPLE_OTHER; - rss_cfg->rss_tuple_sets.ipv6_udp_en = - HCLGEVF_RSS_INPUT_TUPLE_OTHER; - rss_cfg->rss_tuple_sets.ipv6_sctp_en = - HCLGEVF_RSS_INPUT_TUPLE_SCTP; - rss_cfg->rss_tuple_sets.ipv6_fragment_en = - HCLGEVF_RSS_INPUT_TUPLE_OTHER; - ret = hclgevf_set_rss_input_tuple(hdev, rss_cfg); if (ret) return ret; } - /* Initialize RSS indirect table */ - for (i = 0; i < HCLGEVF_RSS_IND_TBL_SIZE; i++) - rss_cfg->rss_indirection_tbl[i] = i % rss_cfg->rss_size; - ret = hclgevf_set_rss_indir_table(hdev); if (ret) return ret; @@ -2242,12 +2268,16 @@ static void hclgevf_ae_stop(struct hnae3_handle *handle) static int hclgevf_set_alive(struct hnae3_handle *handle, bool alive) { +#define HCLGEVF_STATE_ALIVE 1 +#define HCLGEVF_STATE_NOT_ALIVE 0 + struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - u8 msg_data; + struct hclge_vf_to_pf_msg send_msg; - msg_data = alive ? 1 : 0; - return hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_ALIVE, - 0, &msg_data, 1, false, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_ALIVE, 0); + send_msg.data[0] = alive ? HCLGEVF_STATE_ALIVE : + HCLGEVF_STATE_NOT_ALIVE; + return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); } static int hclgevf_client_start(struct hnae3_handle *handle) @@ -2764,6 +2794,7 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev) goto err_config; /* Initialize RSS for this VF */ + hclgevf_rss_init_cfg(hdev); ret = hclgevf_rss_init_hw(hdev); if (ret) { dev_err(&hdev->pdev->dev, @@ -2801,10 +2832,12 @@ err_cmd_queue_init: static void hclgevf_uninit_hdev(struct hclgevf_dev *hdev) { + struct hclge_vf_to_pf_msg send_msg; + hclgevf_state_uninit(hdev); - hclgevf_send_mbx_msg(hdev, HCLGE_MBX_VF_UNINIT, 0, NULL, 0, - false, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_VF_UNINIT, 0); + hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); if (test_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state)) { hclgevf_misc_irq_uninit(hdev); @@ -2936,6 +2969,8 @@ static int hclgevf_set_channels(struct hnae3_handle *handle, u32 new_tqps_num, for (i = 0; i < HCLGEVF_RSS_IND_TBL_SIZE; i++) rss_indir[i] = i % kinfo->rss_size; + hdev->rss_cfg.rss_size = kinfo->rss_size; + ret = hclgevf_set_rss(handle, rss_indir, NULL, 0); if (ret) dev_err(&hdev->pdev->dev, "set rss indir table fail, ret=%d\n", @@ -3101,16 +3136,17 @@ void hclgevf_update_port_base_vlan_info(struct hclgevf_dev *hdev, u16 state, u8 *port_base_vlan_info, u8 data_size) { struct hnae3_handle *nic = &hdev->nic; + struct hclge_vf_to_pf_msg send_msg; rtnl_lock(); hclgevf_notify_client(hdev, HNAE3_DOWN_CLIENT); rtnl_unlock(); /* send msg to PF and wait update port based vlan info */ - hclgevf_send_mbx_msg(hdev, HCLGE_MBX_SET_VLAN, - HCLGE_MBX_PORT_BASE_VLAN_CFG, - port_base_vlan_info, data_size, - false, NULL, 0); + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN, + HCLGE_MBX_PORT_BASE_VLAN_CFG); + memcpy(send_msg.data, port_base_vlan_info, data_size); + hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); if (state == HNAE3_PORT_BASE_VLAN_DISABLE) nic->port_base_vlan_state = HNAE3_PORT_BASE_VLAN_DISABLE; @@ -3188,7 +3224,7 @@ static int hclgevf_init(void) { pr_info("%s is initializing\n", HCLGEVF_NAME); - hclgevf_wq = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0, HCLGEVF_NAME); + hclgevf_wq = alloc_workqueue("%s", 0, 0, HCLGEVF_NAME); if (!hclgevf_wq) { pr_err("%s: failed to create workqueue\n", HCLGEVF_NAME); return -ENOMEM; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h index fee8d97f323c..3b88d866facc 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h @@ -305,8 +305,8 @@ static inline bool hclgevf_is_reset_pending(struct hclgevf_dev *hdev) return !!hdev->reset_pending; } -int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev, u16 code, u16 subcode, - const u8 *msg_data, u8 msg_len, bool need_resp, +int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev, + struct hclge_vf_to_pf_msg *send_msg, bool need_resp, u8 *resp_data, u16 resp_len); void hclgevf_mbx_handler(struct hclgevf_dev *hdev); void hclgevf_mbx_async_handler(struct hclgevf_dev *hdev); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c index 7cbd715d5e7a..9b8154955f91 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c @@ -5,6 +5,11 @@ #include "hclgevf_main.h" #include "hnae3.h" +static int hclgevf_resp_to_errno(u16 resp_code) +{ + return resp_code ? -resp_code : 0; +} + static void hclgevf_reset_mbx_resp_status(struct hclgevf_dev *hdev) { /* this function should be called with mbx_resp.mbx_mutex held @@ -79,8 +84,8 @@ static int hclgevf_get_mbx_resp(struct hclgevf_dev *hdev, u16 code0, u16 code1, return 0; } -int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev, u16 code, u16 subcode, - const u8 *msg_data, u8 msg_len, bool need_resp, +int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev, + struct hclge_vf_to_pf_msg *send_msg, bool need_resp, u8 *resp_data, u16 resp_len) { struct hclge_mbx_vf_to_pf_cmd *req; @@ -89,21 +94,17 @@ int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev, u16 code, u16 subcode, req = (struct hclge_mbx_vf_to_pf_cmd *)desc.data; - /* first two bytes are reserved for code & subcode */ - if (msg_len > (HCLGE_MBX_MAX_MSG_SIZE - 2)) { + if (!send_msg) { dev_err(&hdev->pdev->dev, - "VF send mbx msg fail, msg len %d exceeds max len %d\n", - msg_len, HCLGE_MBX_MAX_MSG_SIZE); + "failed to send mbx, msg is NULL\n"); return -EINVAL; } hclgevf_cmd_setup_basic_desc(&desc, HCLGEVF_OPC_MBX_VF_TO_PF, false); - req->mbx_need_resp |= need_resp ? HCLGE_MBX_NEED_RESP_BIT : - ~HCLGE_MBX_NEED_RESP_BIT; - req->msg[0] = code; - req->msg[1] = subcode; - if (msg_data) - memcpy(&req->msg[2], msg_data, msg_len); + if (need_resp) + hnae3_set_bit(req->mbx_need_resp, HCLGE_MBX_NEED_RESP_B, 1); + + memcpy(&req->msg, send_msg, sizeof(struct hclge_vf_to_pf_msg)); /* synchronous send */ if (need_resp) { @@ -118,7 +119,8 @@ int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev, u16 code, u16 subcode, return status; } - status = hclgevf_get_mbx_resp(hdev, code, subcode, resp_data, + status = hclgevf_get_mbx_resp(hdev, send_msg->code, + send_msg->subcode, resp_data, resp_len); mutex_unlock(&hdev->mbx_resp.mbx_mutex); } else { @@ -169,7 +171,7 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev) if (unlikely(!hnae3_get_bit(flag, HCLGEVF_CMDQ_RX_OUTVLD_B))) { dev_warn(&hdev->pdev->dev, "dropped invalid mailbox message, code = %u\n", - req->msg[0]); + req->msg.code); /* dropping/not processing this invalid message */ crq->desc[crq->next_to_use].flag = 0; @@ -183,19 +185,21 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev) * timeout and simultaneously queue the async messages for later * prcessing in context of mailbox task i.e. the slow path. */ - switch (req->msg[0]) { + switch (req->msg.code) { case HCLGE_MBX_PF_VF_RESP: if (resp->received_resp) dev_warn(&hdev->pdev->dev, "VF mbx resp flag not clear(%u)\n", - req->msg[1]); + req->msg.vf_mbx_msg_code); resp->received_resp = true; - resp->origin_mbx_msg = (req->msg[1] << 16); - resp->origin_mbx_msg |= req->msg[2]; - resp->resp_status = req->msg[3]; + resp->origin_mbx_msg = + (req->msg.vf_mbx_msg_code << 16); + resp->origin_mbx_msg |= req->msg.vf_mbx_msg_subcode; + resp->resp_status = + hclgevf_resp_to_errno(req->msg.resp_status); - temp = (u8 *)&req->msg[4]; + temp = (u8 *)req->msg.resp_data; for (i = 0; i < HCLGE_MBX_MAX_RESP_DATA_SIZE; i++) { resp->additional_info[i] = *temp; temp++; @@ -220,13 +224,13 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev) HCLGE_MBX_MAX_ARQ_MSG_NUM) { dev_warn(&hdev->pdev->dev, "Async Q full, dropping msg(%u)\n", - req->msg[1]); + req->msg.code); break; } /* tail the async message in arq */ msg_q = hdev->arq.msg_q[hdev->arq.tail]; - memcpy(&msg_q[0], req->msg, + memcpy(&msg_q[0], &req->msg, HCLGE_MBX_MAX_ARQ_MSG_SIZE * sizeof(u16)); hclge_mbx_tail_ptr_move_arq(hdev->arq); atomic_inc(&hdev->arq.count); @@ -237,7 +241,7 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev) default: dev_err(&hdev->pdev->dev, "VF received unsupported(%u) mbx msg from PF\n", - req->msg[0]); + req->msg.code); break; } crq->desc[crq->next_to_use].flag = 0; diff --git a/drivers/net/ethernet/ibm/emac/core.c b/drivers/net/ethernet/ibm/emac/core.c index b7fc17756c51..06248a7db7f2 100644 --- a/drivers/net/ethernet/ibm/emac/core.c +++ b/drivers/net/ethernet/ibm/emac/core.c @@ -872,7 +872,7 @@ static void __emac_mdio_write(struct emac_instance *dev, u8 id, u8 reg, { struct emac_regs __iomem *p = dev->emacp; u32 r = 0; - int n, err = -ETIMEDOUT; + int n; mutex_lock(&dev->mdio_lock); @@ -919,7 +919,6 @@ static void __emac_mdio_write(struct emac_instance *dev, u8 id, u8 reg, goto bail; } } - err = 0; bail: if (emac_has_feature(dev, EMAC_FTR_HAS_RGMII)) rgmii_put_mdio(dev->rgmii_dev, dev->rgmii_port); diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index 84121aab7ff1..96d36ae5049e 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -712,29 +712,36 @@ static int ibmveth_close(struct net_device *netdev) return 0; } -static int netdev_get_link_ksettings(struct net_device *dev, - struct ethtool_link_ksettings *cmd) +static int ibmveth_set_link_ksettings(struct net_device *dev, + const struct ethtool_link_ksettings *cmd) { - u32 supported, advertising; - - supported = (SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg | - SUPPORTED_FIBRE); - advertising = (ADVERTISED_1000baseT_Full | ADVERTISED_Autoneg | - ADVERTISED_FIBRE); - cmd->base.speed = SPEED_1000; - cmd->base.duplex = DUPLEX_FULL; - cmd->base.port = PORT_FIBRE; - cmd->base.phy_address = 0; - cmd->base.autoneg = AUTONEG_ENABLE; - - ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported, - supported); - ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising, - advertising); + struct ibmveth_adapter *adapter = netdev_priv(dev); + + return ethtool_virtdev_set_link_ksettings(dev, cmd, + &adapter->speed, + &adapter->duplex); +} + +static int ibmveth_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *cmd) +{ + struct ibmveth_adapter *adapter = netdev_priv(dev); + + cmd->base.speed = adapter->speed; + cmd->base.duplex = adapter->duplex; + cmd->base.port = PORT_OTHER; return 0; } +static void ibmveth_init_link_settings(struct net_device *dev) +{ + struct ibmveth_adapter *adapter = netdev_priv(dev); + + adapter->speed = SPEED_1000; + adapter->duplex = DUPLEX_FULL; +} + static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { @@ -965,12 +972,13 @@ static void ibmveth_get_ethtool_stats(struct net_device *dev, } static const struct ethtool_ops netdev_ethtool_ops = { - .get_drvinfo = netdev_get_drvinfo, - .get_link = ethtool_op_get_link, - .get_strings = ibmveth_get_strings, - .get_sset_count = ibmveth_get_sset_count, - .get_ethtool_stats = ibmveth_get_ethtool_stats, - .get_link_ksettings = netdev_get_link_ksettings, + .get_drvinfo = netdev_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_strings = ibmveth_get_strings, + .get_sset_count = ibmveth_get_sset_count, + .get_ethtool_stats = ibmveth_get_ethtool_stats, + .get_link_ksettings = ibmveth_get_link_ksettings, + .set_link_ksettings = ibmveth_set_link_ksettings, }; static int ibmveth_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) @@ -978,8 +986,6 @@ static int ibmveth_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return -EOPNOTSUPP; } -#define page_offset(v) ((unsigned long)(v) & ((1 << 12) - 1)) - static int ibmveth_send(struct ibmveth_adapter *adapter, union ibmveth_buf_desc *descs, unsigned long mss) { @@ -1674,6 +1680,7 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id) adapter->netdev = netdev; adapter->mcastFilterSize = be32_to_cpu(*mcastFilterSize_p); adapter->pool_config = 0; + ibmveth_init_link_settings(netdev); netif_napi_add(netdev, &adapter->napi, ibmveth_poll, 16); diff --git a/drivers/net/ethernet/ibm/ibmveth.h b/drivers/net/ethernet/ibm/ibmveth.h index 4e9bf3421f4f..27dfff200166 100644 --- a/drivers/net/ethernet/ibm/ibmveth.h +++ b/drivers/net/ethernet/ibm/ibmveth.h @@ -162,6 +162,9 @@ struct ibmveth_adapter { u64 tx_send_failed; u64 tx_large_packets; u64 rx_large_packets; + /* Ethtool settings */ + u8 duplex; + u32 speed; }; /* diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index 154e2e818ec6..ad34e4335df2 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig @@ -294,6 +294,7 @@ config ICE tristate "Intel(R) Ethernet Connection E800 Series Support" default n depends on PCI_MSI + select NET_DEVLINK ---help--- This driver supports Intel(R) Ethernet Connection E800 Series of devices. For more information on how to identify your adapter, go diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c index be56e631d693..6f45df5690d4 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c +++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c @@ -1852,6 +1852,7 @@ static void e1000_get_strings(struct net_device *netdev, u32 stringset, } static const struct ethtool_ops e1000_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = e1000_get_drvinfo, .get_regs_len = e1000_get_regs_len, .get_regs = e1000_get_regs, diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c index 2bced34c19ba..f7103356ef56 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_main.c +++ b/drivers/net/ethernet/intel/e1000/e1000_main.c @@ -2715,11 +2715,7 @@ static int e1000_tso(struct e1000_adapter *adapter, cmd_length = E1000_TXD_CMD_IP; ipcse = skb_transport_offset(skb) - 1; } else if (skb_is_gso_v6(skb)) { - ipv6_hdr(skb)->payload_len = 0; - tcp_hdr(skb)->check = - ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - 0, IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); ipcse = 0; } ipcss = skb_network_offset(skb); diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c index adce7e319b9e..1d47e2503072 100644 --- a/drivers/net/ethernet/intel/e1000e/ethtool.c +++ b/drivers/net/ethernet/intel/e1000e/ethtool.c @@ -897,6 +897,7 @@ static int e1000_reg_test(struct e1000_adapter *adapter, u64 *data) case e1000_pch_cnp: /* fall through */ case e1000_pch_tgp: + case e1000_pch_adp: mask |= BIT(18); break; default: @@ -1561,6 +1562,7 @@ static void e1000_loopback_cleanup(struct e1000_adapter *adapter) case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: fext_nvm11 = er32(FEXTNVM11); fext_nvm11 &= ~E1000_FEXTNVM11_DISABLE_MULR_FIX; ew32(FEXTNVM11, fext_nvm11); @@ -2305,6 +2307,7 @@ static int e1000e_get_ts_info(struct net_device *netdev, } static const struct ethtool_ops e1000_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = e1000_get_drvinfo, .get_regs_len = e1000_get_regs_len, .get_regs = e1000_get_regs, diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h index f556163481cb..b1447221669e 100644 --- a/drivers/net/ethernet/intel/e1000e/hw.h +++ b/drivers/net/ethernet/intel/e1000e/hw.h @@ -97,6 +97,11 @@ struct e1000_hw; #define E1000_DEV_ID_PCH_TGP_I219_LM14 0x15F9 #define E1000_DEV_ID_PCH_TGP_I219_V14 0x15FA #define E1000_DEV_ID_PCH_TGP_I219_LM15 0x15F4 +#define E1000_DEV_ID_PCH_TGP_I219_V15 0x15F5 +#define E1000_DEV_ID_PCH_ADP_I219_LM16 0x1A1E +#define E1000_DEV_ID_PCH_ADP_I219_V16 0x1A1F +#define E1000_DEV_ID_PCH_ADP_I219_LM17 0x1A1C +#define E1000_DEV_ID_PCH_ADP_I219_V17 0x1A1D #define E1000_REVISION_4 4 @@ -121,6 +126,7 @@ enum e1000_mac_type { e1000_pch_spt, e1000_pch_cnp, e1000_pch_tgp, + e1000_pch_adp, }; enum e1000_media_type { diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c index b4135c50e905..735bf25952fc 100644 --- a/drivers/net/ethernet/intel/e1000e/ich8lan.c +++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c @@ -317,6 +317,7 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw) case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: if (e1000_phy_is_accessible_pchlan(hw)) break; @@ -460,6 +461,7 @@ static s32 e1000_init_phy_params_pchlan(struct e1000_hw *hw) case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: /* In case the PHY needs to be in mdio slow mode, * set slow mode and try to get the PHY id again. */ @@ -703,6 +705,7 @@ static s32 e1000_init_mac_params_ich8lan(struct e1000_hw *hw) case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: case e1000_pchlan: /* check management mode */ mac->ops.check_mng_mode = e1000_check_mng_mode_pchlan; @@ -1642,6 +1645,7 @@ static s32 e1000_get_variants_ich8lan(struct e1000_adapter *adapter) case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: rc = e1000_init_phy_params_pchlan(hw); break; default: @@ -2095,6 +2099,7 @@ static s32 e1000_sw_lcd_config_ich8lan(struct e1000_hw *hw) case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: sw_cfg_mask = E1000_FEXTNVM_SW_CONFIG_ICH8M; break; default: @@ -3133,6 +3138,7 @@ static s32 e1000_valid_nvm_bank_detect_ich8lan(struct e1000_hw *hw, u32 *bank) case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: bank1_offset = nvm->flash_bank_size; act_offset = E1000_ICH_NVM_SIG_WORD; @@ -4077,6 +4083,7 @@ static s32 e1000_validate_nvm_checksum_ich8lan(struct e1000_hw *hw) case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: word = NVM_COMPAT; valid_csum_mask = NVM_COMPAT_VALID_CSUM; break; diff --git a/drivers/net/ethernet/intel/e1000e/mac.c b/drivers/net/ethernet/intel/e1000e/mac.c index e531976f8a67..51512a73fdd0 100644 --- a/drivers/net/ethernet/intel/e1000e/mac.c +++ b/drivers/net/ethernet/intel/e1000e/mac.c @@ -1363,7 +1363,7 @@ s32 e1000e_get_hw_semaphore(struct e1000_hw *hw) if (!(swsm & E1000_SWSM_SMBI)) break; - usleep_range(50, 100); + udelay(100); i++; } @@ -1381,7 +1381,7 @@ s32 e1000e_get_hw_semaphore(struct e1000_hw *hw) if (er32(SWSM) & E1000_SWSM_SWESMBI) break; - usleep_range(50, 100); + udelay(100); } if (i == timeout) { diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index 0f02c7a5ee9b..177c6da80c57 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -3536,6 +3536,7 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca) break; case e1000_pch_cnp: case e1000_pch_tgp: + case e1000_pch_adp: if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI) { /* Stable 24MHz frequency */ incperiod = INCPERIOD_24MHZ; @@ -3807,7 +3808,7 @@ static void e1000_flush_tx_ring(struct e1000_adapter *adapter) tdt = er32(TDT(0)); BUG_ON(tdt != tx_ring->next_to_use); tx_desc = E1000_TX_DESC(*tx_ring, tx_ring->next_to_use); - tx_desc->buffer_addr = tx_ring->dma; + tx_desc->buffer_addr = cpu_to_le64(tx_ring->dma); tx_desc->lower.data = cpu_to_le32(txd_lower | size); tx_desc->upper.data = 0; @@ -4049,6 +4050,7 @@ void e1000e_reset(struct e1000_adapter *adapter) case e1000_pch_cnp: /* fall-through */ case e1000_pch_tgp: + case e1000_pch_adp: fc->refresh_time = 0xFFFF; fc->pause_time = 0xFFFF; @@ -5461,10 +5463,7 @@ static int e1000_tso(struct e1000_ring *tx_ring, struct sk_buff *skb, cmd_length = E1000_TXD_CMD_IP; ipcse = skb_transport_offset(skb) - 1; } else if (skb_is_gso_v6(skb)) { - ipv6_hdr(skb)->payload_len = 0; - tcp_hdr(skb)->check = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - 0, IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); ipcse = 0; } ipcss = skb_network_offset(skb); @@ -7759,6 +7758,11 @@ static const struct pci_device_id e1000_pci_tbl[] = { { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_LM14), board_pch_cnp }, { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_V14), board_pch_cnp }, { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_LM15), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_V15), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM16), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V16), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM17), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V17), board_pch_cnp }, { 0, 0, 0, 0, 0, 0, 0 } /* terminate list */ }; diff --git a/drivers/net/ethernet/intel/e1000e/ptp.c b/drivers/net/ethernet/intel/e1000e/ptp.c index eaa5a0fb99f0..439fda2f5368 100644 --- a/drivers/net/ethernet/intel/e1000e/ptp.c +++ b/drivers/net/ethernet/intel/e1000e/ptp.c @@ -297,6 +297,7 @@ void e1000e_ptp_init(struct e1000_adapter *adapter) case e1000_pch_cnp: /* fall-through */ case e1000_pch_tgp: + case e1000_pch_adp: if ((hw->mac.type < e1000_pch_lpt) || (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI)) { adapter->ptp_clock_info.max_adj = 24000000 - 1; diff --git a/drivers/net/ethernet/intel/fm10k/fm10k.h b/drivers/net/ethernet/intel/fm10k/fm10k.h index f306084ca12c..5b78362b82ac 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k.h +++ b/drivers/net/ethernet/intel/fm10k/fm10k.h @@ -41,7 +41,7 @@ struct fm10k_l2_accel { u16 count; u16 dglort; struct rcu_head rcu; - struct net_device *macvlan[0]; + struct net_device *macvlan[]; }; enum fm10k_ring_state_t { @@ -198,7 +198,7 @@ struct fm10k_q_vector { struct rcu_head rcu; /* to avoid race with update stats on free */ /* for dynamic allocation of rings associated with this q_vector */ - struct fm10k_ring ring[0] ____cacheline_internodealigned_in_smp; + struct fm10k_ring ring[] ____cacheline_internodealigned_in_smp; }; enum fm10k_ring_f_enum { @@ -218,7 +218,7 @@ struct fm10k_iov_data { unsigned int num_vfs; unsigned int next_vf_mbx; struct rcu_head rcu; - struct fm10k_vf_info vf_info[0]; + struct fm10k_vf_info vf_info[]; }; struct fm10k_udp_port { diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c index 68edf55ac906..37fbc646deb9 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c @@ -1151,6 +1151,8 @@ static int fm10k_set_channels(struct net_device *dev, } static const struct ethtool_ops fm10k_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_USE_ADAPTIVE, .get_strings = fm10k_get_strings, .get_sset_count = fm10k_get_sset_count, .get_ethtool_stats = fm10k_get_ethtool_stats, diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index 4833187bd259..e95b8da45e07 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -334,13 +334,13 @@ int i40e_ddp_flash(struct net_device *netdev, struct ethtool_flash *flash); struct i40e_ddp_profile_list { u32 p_count; - struct i40e_profile_info p_info[0]; + struct i40e_profile_info p_info[]; }; struct i40e_ddp_old_profile_list { struct list_head list; size_t old_ddp_size; - u8 old_ddp_buf[0]; + u8 old_ddp_buf[]; }; /* macros related to FLX_PIT */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 317f3f1458db..aa8026b1eb81 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -5249,6 +5249,11 @@ static const struct ethtool_ops i40e_ethtool_recovery_mode_ops = { }; static const struct ethtool_ops i40e_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES_IRQ | + ETHTOOL_COALESCE_USE_ADAPTIVE | + ETHTOOL_COALESCE_RX_USECS_HIGH | + ETHTOOL_COALESCE_TX_USECS_HIGH, .get_drvinfo = i40e_get_drvinfo, .get_regs_len = i40e_get_regs_len, .get_regs = i40e_get_regs, diff --git a/drivers/net/ethernet/intel/iavf/iavf.h b/drivers/net/ethernet/intel/iavf/iavf.h index bd1b1ed323f4..bcd11b4b29df 100644 --- a/drivers/net/ethernet/intel/iavf/iavf.h +++ b/drivers/net/ethernet/intel/iavf/iavf.h @@ -81,7 +81,7 @@ struct iavf_vsi { #define IAVF_TX_DESC(R, i) (&(((struct iavf_tx_desc *)((R)->desc))[i])) #define IAVF_TX_CTXTDESC(R, i) \ (&(((struct iavf_tx_context_desc *)((R)->desc))[i])) -#define IAVF_MAX_REQ_QUEUES 4 +#define IAVF_MAX_REQ_QUEUES 16 #define IAVF_HKEY_ARRAY_SIZE ((IAVF_VFQF_HKEY_MAX_INDEX + 1) * 4) #define IAVF_HLUT_ARRAY_SIZE ((IAVF_VFQF_HLUT_MAX_INDEX + 1) * 4) diff --git a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c index 84c3d8d97ef6..2c39d46b6138 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c +++ b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c @@ -860,7 +860,7 @@ static void iavf_get_channels(struct net_device *netdev, struct iavf_adapter *adapter = netdev_priv(netdev); /* Report maximum channels */ - ch->max_combined = IAVF_MAX_REQ_QUEUES; + ch->max_combined = adapter->vsi_res->num_queue_pairs; ch->max_other = NONQ_VECS; ch->other_count = NONQ_VECS; @@ -881,14 +881,7 @@ static int iavf_set_channels(struct net_device *netdev, struct ethtool_channels *ch) { struct iavf_adapter *adapter = netdev_priv(netdev); - int num_req = ch->combined_count; - - if (num_req != adapter->num_active_queues && - !(adapter->vf_res->vf_cap_flags & - VIRTCHNL_VF_OFFLOAD_REQ_QUEUES)) { - dev_info(&adapter->pdev->dev, "PF is not capable of queue negotiation.\n"); - return -EINVAL; - } + u32 num_req = ch->combined_count; if ((adapter->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_ADQ) && adapter->num_tc) { @@ -899,14 +892,19 @@ static int iavf_set_channels(struct net_device *netdev, /* All of these should have already been checked by ethtool before this * even gets to us, but just to be sure. */ - if (num_req <= 0 || num_req > IAVF_MAX_REQ_QUEUES) + if (num_req > adapter->vsi_res->num_queue_pairs) return -EINVAL; + if (num_req == adapter->num_active_queues) + return 0; + if (ch->rx_count || ch->tx_count || ch->other_count != NONQ_VECS) return -EINVAL; adapter->num_req_queues = num_req; - return iavf_request_queues(adapter, num_req); + adapter->flags |= IAVF_FLAG_REINIT_ITR_NEEDED; + iavf_schedule_reset(adapter); + return 0; } /** @@ -998,6 +996,10 @@ static int iavf_set_rxfh(struct net_device *netdev, const u32 *indir, } static const struct ethtool_ops iavf_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_MAX_FRAMES_IRQ | + ETHTOOL_COALESCE_USE_ADAPTIVE, .get_drvinfo = iavf_get_drvinfo, .get_link = ethtool_op_get_link, .get_ringparam = iavf_get_ringparam, diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c index 62fe56ddcb6e..2050649848ba 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_main.c +++ b/drivers/net/ethernet/intel/iavf/iavf_main.c @@ -3061,9 +3061,6 @@ static int iavf_delete_clsflower(struct iavf_adapter *adapter, static int iavf_setup_tc_cls_flower(struct iavf_adapter *adapter, struct flow_cls_offload *cls_flower) { - if (cls_flower->common.chain_index) - return -EOPNOTSUPP; - switch (cls_flower->command) { case FLOW_CLS_REPLACE: return iavf_configure_clsflower(adapter, cls_flower); @@ -3087,6 +3084,11 @@ static int iavf_setup_tc_cls_flower(struct iavf_adapter *adapter, static int iavf_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv) { + struct iavf_adapter *adapter = cb_priv; + + if (!tc_cls_can_offload_and_chain0(adapter->netdev, type_data)) + return -EOPNOTSUPP; + switch (type) { case TC_SETUP_CLSFLOWER: return iavf_setup_tc_cls_flower(cb_priv, type_data); @@ -3448,7 +3450,7 @@ int iavf_process_config(struct iavf_adapter *adapter) } if (num_req_queues && - num_req_queues != adapter->vsi_res->num_queue_pairs) { + num_req_queues > adapter->vsi_res->num_queue_pairs) { /* Problem. The PF gave us fewer queues than what we had * negotiated in our request. Need a reset to see if we can't * get back to a working state. diff --git a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c index 1ab9cb339acb..d58374c2c33d 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c +++ b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c @@ -397,33 +397,6 @@ void iavf_map_queues(struct iavf_adapter *adapter) } /** - * iavf_request_queues - * @adapter: adapter structure - * @num: number of requested queues - * - * We get a default number of queues from the PF. This enables us to request a - * different number. Returns 0 on success, negative on failure - **/ -int iavf_request_queues(struct iavf_adapter *adapter, int num) -{ - struct virtchnl_vf_res_request vfres; - - if (adapter->current_op != VIRTCHNL_OP_UNKNOWN) { - /* bail because we already have a command pending */ - dev_err(&adapter->pdev->dev, "Cannot request queues, command %d pending\n", - adapter->current_op); - return -EBUSY; - } - - vfres.num_queue_pairs = min_t(int, num, num_online_cpus()); - - adapter->current_op = VIRTCHNL_OP_REQUEST_QUEUES; - adapter->flags |= IAVF_FLAG_REINIT_ITR_NEEDED; - return iavf_send_pf_msg(adapter, VIRTCHNL_OP_REQUEST_QUEUES, - (u8 *)&vfres, sizeof(vfres)); -} - -/** * iavf_add_ether_addrs * @adapter: adapter structure * diff --git a/drivers/net/ethernet/intel/ice/Makefile b/drivers/net/ethernet/intel/ice/Makefile index 59544b0fc086..29c6c6743450 100644 --- a/drivers/net/ethernet/intel/ice/Makefile +++ b/drivers/net/ethernet/intel/ice/Makefile @@ -19,6 +19,7 @@ ice-y := ice_main.o \ ice_txrx.o \ ice_flex_pipe.o \ ice_flow.o \ + ice_devlink.o \ ice_ethtool.o ice-$(CONFIG_PCI_IOV) += ice_virtchnl_pf.o ice_sriov.o ice-$(CONFIG_DCB) += ice_dcb.o ice_dcb_nl.o ice_dcb_lib.o diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index cb10abb14e11..5c11448bfbb3 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -34,6 +34,7 @@ #include <linux/ctype.h> #include <linux/bpf.h> #include <linux/avf/virtchnl.h> +#include <net/devlink.h> #include <net/ipv6.h> #include <net/xdp_sock.h> #include "ice_devids.h" @@ -60,7 +61,6 @@ extern const char ice_drv_ver[]; #define ICE_INT_NAME_STR_LEN (IFNAMSIZ + 16) #define ICE_AQ_LEN 64 #define ICE_MBXSQ_LEN 64 -#define ICE_MBXRQ_LEN 512 #define ICE_MIN_MSIX 2 #define ICE_NO_VSI 0xffff #define ICE_VSI_MAP_CONTIG 0 @@ -70,7 +70,6 @@ extern const char ice_drv_ver[]; #define ICE_Q_WAIT_RETRY_LIMIT 10 #define ICE_Q_WAIT_MAX_RETRY (5 * ICE_Q_WAIT_RETRY_LIMIT) #define ICE_MAX_LG_RSS_QS 256 -#define ICE_MAX_SMALL_RSS_QS 8 #define ICE_RES_VALID_BIT 0x8000 #define ICE_RES_MISC_VEC_ID (ICE_RES_VALID_BIT - 1) #define ICE_INVAL_Q_INDEX 0xffff @@ -212,6 +211,8 @@ enum ice_state { __ICE_SERVICE_SCHED, __ICE_SERVICE_DIS, __ICE_OICR_INTR_DIS, /* Global OICR interrupt disabled */ + __ICE_MDD_VF_PRINT_PENDING, /* set when MDD event handle */ + __ICE_VF_RESETS_DISABLED, /* disable resets during ice_remove */ __ICE_STATE_NBITS /* must be last */ }; @@ -340,12 +341,18 @@ enum ice_pf_flags { ICE_FLAG_FW_LLDP_AGENT, ICE_FLAG_ETHTOOL_CTXT, /* set when ethtool holds RTNL lock */ ICE_FLAG_LEGACY_RX, + ICE_FLAG_MDD_AUTO_RESET_VF, ICE_PF_FLAGS_NBITS /* must be last */ }; struct ice_pf { struct pci_dev *pdev; + /* devlink port data */ + struct devlink_port devlink_port; + + struct devlink_region *nvm_region; + /* OS reserved IRQ details */ struct msix_entry *msix_entries; struct ice_res_tracker *irq_tracker; @@ -361,8 +368,10 @@ struct ice_pf { struct ice_vf *vf; int num_alloc_vfs; /* actual number of VFs allocated */ u16 num_vfs_supported; /* num VFs supported for this PF */ - u16 num_vf_qps; /* num queue pairs per VF */ - u16 num_vf_msix; /* num vectors per VF */ + u16 num_qps_per_vf; + u16 num_msix_per_vf; + /* used to ratelimit the MDD event logging */ + unsigned long last_printed_mdd_jiffies; DECLARE_BITMAP(state, __ICE_STATE_NBITS); DECLARE_BITMAP(flags, ICE_PF_FLAGS_NBITS); unsigned long *avail_txqs; /* bitmap to track PF Tx queue usage */ diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h index 6873998cf145..2381b4014ed6 100644 --- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h +++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h @@ -1232,6 +1232,7 @@ struct ice_aqc_sff_eeprom { * NVM Update commands (indirect 0x0703) */ struct ice_aqc_nvm { +#define ICE_AQC_NVM_MAX_OFFSET 0xFFFFFF __le16 offset_low; u8 offset_high; u8 cmd_flags; @@ -1250,6 +1251,8 @@ struct ice_aqc_nvm { __le32 addr_low; }; +#define ICE_AQC_NVM_START_POINT 0 + /* NVM Checksum Command (direct, 0x0706) */ struct ice_aqc_nvm_checksum { u8 flags; @@ -1661,6 +1664,13 @@ struct ice_aqc_get_pkg_info_resp { struct ice_aqc_get_pkg_info pkg_info[1]; }; +/* Lan Queue Overflow Event (direct, 0x1001) */ +struct ice_aqc_event_lan_overflow { + __le32 prtdcb_ruptq; + __le32 qtx_ctl; + u8 reserved[8]; +}; + /** * struct ice_aq_desc - Admin Queue (AQ) descriptor * @flags: ICE_AQ_FLAG_* flags @@ -1730,6 +1740,7 @@ struct ice_aq_desc { struct ice_aqc_alloc_free_res_cmd sw_res_ctrl; struct ice_aqc_set_event_mask set_event_mask; struct ice_aqc_get_link_status get_link_status; + struct ice_aqc_event_lan_overflow lan_overflow; } params; }; @@ -1756,6 +1767,7 @@ enum ice_aq_err { ICE_AQ_RC_ENOMEM = 9, /* Out of memory */ ICE_AQ_RC_EBUSY = 12, /* Device or resource busy */ ICE_AQ_RC_EEXIST = 13, /* Object already exists */ + ICE_AQ_RC_EINVAL = 14, /* Invalid argument */ ICE_AQ_RC_ENOSPC = 16, /* No space left or allocation failure */ ICE_AQ_RC_ENOSYS = 17, /* Function not implemented */ ICE_AQ_RC_ENOSEC = 24, /* Missing security manifest */ @@ -1860,6 +1872,9 @@ enum ice_adminq_opc { ice_aqc_opc_update_pkg = 0x0C42, ice_aqc_opc_get_pkg_info_list = 0x0C43, + /* Standalone Commands/Events */ + ice_aqc_opc_event_lan_overflow = 0x1001, + /* debug commands */ ice_aqc_opc_fw_logging = 0xFF09, ice_aqc_opc_fw_logging_info = 0xFF10, diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index 81885efadc7a..a19cd6f5436b 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -203,8 +203,7 @@ static void ice_cfg_itr_gran(struct ice_hw *hw) */ static u16 ice_calc_q_handle(struct ice_vsi *vsi, struct ice_ring *ring, u8 tc) { - WARN_ONCE(ice_ring_is_xdp(ring) && tc, - "XDP ring can't belong to TC other than 0"); + WARN_ONCE(ice_ring_is_xdp(ring) && tc, "XDP ring can't belong to TC other than 0\n"); /* Idea here for calculation is that we subtract the number of queue * count from TC that ring belongs to from it's absolute queue index @@ -247,7 +246,6 @@ ice_setup_tx_ctx(struct ice_ring *ring, struct ice_tlan_ctx *tlan_ctx, u16 pf_q) */ switch (vsi->type) { case ICE_VSI_LB: - /* fall through */ case ICE_VSI_PF: tlan_ctx->vmvf_type = ICE_TLAN_CTX_VMVF_TYPE_PF; break; @@ -387,8 +385,8 @@ int ice_setup_rx_ctx(struct ice_ring *ring) /* Enable Flexible Descriptors in the queue context which * allows this driver to select a specific receive descriptor format */ + regval = rd32(hw, QRXFLXP_CNTXT(pf_q)); if (vsi->type != ICE_VSI_VF) { - regval = rd32(hw, QRXFLXP_CNTXT(pf_q)); regval |= (rxdid << QRXFLXP_CNTXT_RXDID_IDX_S) & QRXFLXP_CNTXT_RXDID_IDX_M; @@ -399,8 +397,12 @@ int ice_setup_rx_ctx(struct ice_ring *ring) regval |= (0x03 << QRXFLXP_CNTXT_RXDID_PRIO_S) & QRXFLXP_CNTXT_RXDID_PRIO_M; - wr32(hw, QRXFLXP_CNTXT(pf_q), regval); + } else { + regval &= ~(QRXFLXP_CNTXT_RXDID_IDX_M | + QRXFLXP_CNTXT_RXDID_PRIO_M | + QRXFLXP_CNTXT_TS_M); } + wr32(hw, QRXFLXP_CNTXT(pf_q), regval); /* Absolute queue number out of 2K needs to be passed */ err = ice_write_rxq_ctx(hw, &rlan_ctx, pf_q); @@ -459,17 +461,20 @@ int __ice_vsi_get_qs(struct ice_qs_cfg *qs_cfg) } /** - * ice_vsi_ctrl_rx_ring - Start or stop a VSI's Rx ring + * ice_vsi_ctrl_one_rx_ring - start/stop VSI's Rx ring with no busy wait * @vsi: the VSI being configured - * @ena: start or stop the Rx rings - * @rxq_idx: Rx queue index + * @ena: start or stop the Rx ring + * @rxq_idx: 0-based Rx queue index for the VSI passed in + * @wait: wait or don't wait for configuration to finish in hardware + * + * Return 0 on success and negative on error. */ -int ice_vsi_ctrl_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx) +int +ice_vsi_ctrl_one_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx, bool wait) { int pf_q = vsi->rxq_map[rxq_idx]; struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; - int ret = 0; u32 rx_reg; rx_reg = rd32(hw, QRX_CTRL(pf_q)); @@ -485,13 +490,30 @@ int ice_vsi_ctrl_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx) rx_reg &= ~QRX_CTRL_QENA_REQ_M; wr32(hw, QRX_CTRL(pf_q), rx_reg); - /* wait for the change to finish */ - ret = ice_pf_rxq_wait(pf, pf_q, ena); - if (ret) - dev_err(ice_pf_to_dev(pf), "VSI idx %d Rx ring %d %sable timeout\n", - vsi->idx, pf_q, (ena ? "en" : "dis")); + if (!wait) + return 0; + + ice_flush(hw); + return ice_pf_rxq_wait(pf, pf_q, ena); +} - return ret; +/** + * ice_vsi_wait_one_rx_ring - wait for a VSI's Rx ring to be stopped/started + * @vsi: the VSI being configured + * @ena: true/false to verify Rx ring has been enabled/disabled respectively + * @rxq_idx: 0-based Rx queue index for the VSI passed in + * + * This routine will wait for the given Rx queue of the VSI to reach the + * enabled or disabled state. Returns -ETIMEDOUT in case of failing to reach + * the requested state after multiple retries; else will return 0 in case of + * success. + */ +int ice_vsi_wait_one_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx) +{ + int pf_q = vsi->rxq_map[rxq_idx]; + struct ice_pf *pf = vsi->back; + + return ice_pf_rxq_wait(pf, pf_q, ena); } /** diff --git a/drivers/net/ethernet/intel/ice/ice_base.h b/drivers/net/ethernet/intel/ice/ice_base.h index 407995e8e944..44efdb627043 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.h +++ b/drivers/net/ethernet/intel/ice/ice_base.h @@ -8,7 +8,9 @@ int ice_setup_rx_ctx(struct ice_ring *ring); int __ice_vsi_get_qs(struct ice_qs_cfg *qs_cfg); -int ice_vsi_ctrl_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx); +int +ice_vsi_ctrl_one_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx, bool wait); +int ice_vsi_wait_one_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx); int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi); void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi); void ice_vsi_free_q_vectors(struct ice_vsi *vsi); diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 04d5db0a25bf..2c0d8fd3d5cd 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -6,7 +6,7 @@ #include "ice_adminq_cmd.h" #include "ice_flow.h" -#define ICE_PF_RESET_WAIT_COUNT 200 +#define ICE_PF_RESET_WAIT_COUNT 300 /** * ice_set_mac_type - Sets MAC type @@ -615,29 +615,6 @@ static void ice_get_itr_intrl_gran(struct ice_hw *hw) } /** - * ice_get_nvm_version - get cached NVM version data - * @hw: pointer to the hardware structure - * @oem_ver: 8 bit NVM version - * @oem_build: 16 bit NVM build number - * @oem_patch: 8 NVM patch number - * @ver_hi: high 16 bits of the NVM version - * @ver_lo: low 16 bits of the NVM version - */ -void -ice_get_nvm_version(struct ice_hw *hw, u8 *oem_ver, u16 *oem_build, - u8 *oem_patch, u8 *ver_hi, u8 *ver_lo) -{ - struct ice_nvm_info *nvm = &hw->nvm; - - *oem_ver = (u8)((nvm->oem_ver & ICE_OEM_VER_MASK) >> ICE_OEM_VER_SHIFT); - *oem_patch = (u8)(nvm->oem_ver & ICE_OEM_VER_PATCH_MASK); - *oem_build = (u16)((nvm->oem_ver & ICE_OEM_VER_BUILD_MASK) >> - ICE_OEM_VER_BUILD_SHIFT); - *ver_hi = (nvm->ver & ICE_NVM_VER_HI_MASK) >> ICE_NVM_VER_HI_SHIFT; - *ver_lo = (nvm->ver & ICE_NVM_VER_LO_MASK) >> ICE_NVM_VER_LO_SHIFT; -} - -/** * ice_init_hw - main hardware initialization routine * @hw: pointer to the hardware structure */ @@ -958,72 +935,6 @@ enum ice_status ice_reset(struct ice_hw *hw, enum ice_reset_req req) } /** - * ice_get_pfa_module_tlv - Reads sub module TLV from NVM PFA - * @hw: pointer to hardware structure - * @module_tlv: pointer to module TLV to return - * @module_tlv_len: pointer to module TLV length to return - * @module_type: module type requested - * - * Finds the requested sub module TLV type from the Preserved Field - * Area (PFA) and returns the TLV pointer and length. The caller can - * use these to read the variable length TLV value. - */ -enum ice_status -ice_get_pfa_module_tlv(struct ice_hw *hw, u16 *module_tlv, u16 *module_tlv_len, - u16 module_type) -{ - enum ice_status status; - u16 pfa_len, pfa_ptr; - u16 next_tlv; - - status = ice_read_sr_word(hw, ICE_SR_PFA_PTR, &pfa_ptr); - if (status) { - ice_debug(hw, ICE_DBG_INIT, "Preserved Field Array pointer.\n"); - return status; - } - status = ice_read_sr_word(hw, pfa_ptr, &pfa_len); - if (status) { - ice_debug(hw, ICE_DBG_INIT, "Failed to read PFA length.\n"); - return status; - } - /* Starting with first TLV after PFA length, iterate through the list - * of TLVs to find the requested one. - */ - next_tlv = pfa_ptr + 1; - while (next_tlv < pfa_ptr + pfa_len) { - u16 tlv_sub_module_type; - u16 tlv_len; - - /* Read TLV type */ - status = ice_read_sr_word(hw, next_tlv, &tlv_sub_module_type); - if (status) { - ice_debug(hw, ICE_DBG_INIT, "Failed to read TLV type.\n"); - break; - } - /* Read TLV length */ - status = ice_read_sr_word(hw, next_tlv + 1, &tlv_len); - if (status) { - ice_debug(hw, ICE_DBG_INIT, "Failed to read TLV length.\n"); - break; - } - if (tlv_sub_module_type == module_type) { - if (tlv_len) { - *module_tlv = next_tlv; - *module_tlv_len = tlv_len; - return 0; - } - return ICE_ERR_INVAL_SIZE; - } - /* Check next TLV, i.e. current TLV pointer + length + 2 words - * (for current TLV's type and length) - */ - next_tlv = next_tlv + tlv_len + 2; - } - /* Module does not exist */ - return ICE_ERR_DOES_NOT_EXIST; -} - -/** * ice_copy_rxq_ctx_to_hw * @hw: pointer to the hardware structure * @ice_rxq_ctx: pointer to the rxq context @@ -1181,7 +1092,7 @@ ice_aq_send_cmd(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf, case ice_aqc_opc_release_res: if (le16_to_cpu(cmd->res_id) == ICE_AQC_RES_ID_GLBL_LOCK) break; - /* fall-through */ + fallthrough; default: mutex_lock(&ice_global_cfg_lock_sw); lock_acquired = true; @@ -2703,7 +2614,7 @@ __ice_aq_get_set_rss_lut(struct ice_hw *hw, u16 vsi_id, u8 lut_type, u8 *lut, ICE_AQC_GSET_RSS_LUT_TABLE_SIZE_M; break; } - /* fall-through */ + fallthrough; default: status = ICE_ERR_PARAM; goto ice_aq_get_set_rss_lut_exit; diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h index f9fc005d35a7..8104f3d64d96 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.h +++ b/drivers/net/ethernet/intel/ice/ice_common.h @@ -15,9 +15,6 @@ enum ice_status ice_nvm_validate_checksum(struct ice_hw *hw); enum ice_status ice_init_hw(struct ice_hw *hw); void ice_deinit_hw(struct ice_hw *hw); -enum ice_status -ice_get_pfa_module_tlv(struct ice_hw *hw, u16 *module_tlv, u16 *module_tlv_len, - u16 module_type); enum ice_status ice_check_reset(struct ice_hw *hw); enum ice_status ice_reset(struct ice_hw *hw, enum ice_reset_req req); enum ice_status ice_create_all_ctrlq(struct ice_hw *hw); @@ -38,9 +35,6 @@ enum ice_status ice_alloc_hw_res(struct ice_hw *hw, u16 type, u16 num, bool btm, u16 *res); enum ice_status ice_free_hw_res(struct ice_hw *hw, u16 type, u16 num, u16 *res); -enum ice_status ice_init_nvm(struct ice_hw *hw); -enum ice_status -ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words, u16 *data); enum ice_status ice_aq_alloc_free_res(struct ice_hw *hw, u16 num_entries, struct ice_aqc_alloc_free_res_elem *buf, u16 buf_size, @@ -153,9 +147,6 @@ ice_stat_update40(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, void ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat); -void -ice_get_nvm_version(struct ice_hw *hw, u8 *oem_ver, u16 *oem_build, - u8 *oem_patch, u8 *ver_hi, u8 *ver_lo); enum ice_status ice_sched_query_elem(struct ice_hw *hw, u32 node_teid, struct ice_aqc_get_elem *buf); diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c index 7108fb41b604..7bea09363b42 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c @@ -63,6 +63,26 @@ u8 ice_dcb_get_ena_tc(struct ice_dcbx_cfg *dcbcfg) } /** + * ice_dcb_get_mode - gets the DCB mode + * @port_info: pointer to port info structure + * @host: if set it's HOST if not it's MANAGED + */ +static u8 ice_dcb_get_mode(struct ice_port_info *port_info, bool host) +{ + u8 mode; + + if (host) + mode = DCB_CAP_DCBX_HOST; + else + mode = DCB_CAP_DCBX_LLD_MANAGED; + + if (port_info->local_dcbx_cfg.dcbx_mode & ICE_DCBX_MODE_CEE) + return mode | DCB_CAP_DCBX_VER_CEE; + else + return mode | DCB_CAP_DCBX_VER_IEEE; +} + +/** * ice_dcb_get_num_tc - Get the number of TCs from DCBX config * @dcbcfg: config to retrieve number of TCs from */ @@ -149,6 +169,43 @@ void ice_vsi_cfg_dcb_rings(struct ice_vsi *vsi) } /** + * ice_dcb_bwchk - check if ETS bandwidth input parameters are correct + * @pf: pointer to the PF struct + * @dcbcfg: pointer to DCB config structure + */ +int ice_dcb_bwchk(struct ice_pf *pf, struct ice_dcbx_cfg *dcbcfg) +{ + struct ice_dcb_ets_cfg *etscfg = &dcbcfg->etscfg; + u8 num_tc, total_bw = 0; + int i; + + /* returns number of contigous TCs and 1 TC for non-contigous TCs, + * since at least 1 TC has to be configured + */ + num_tc = ice_dcb_get_num_tc(dcbcfg); + + /* no bandwidth checks required if there's only one TC, so assign + * all bandwidth to TC0 and return + */ + if (num_tc == 1) { + etscfg->tcbwtable[0] = ICE_TC_MAX_BW; + return 0; + } + + for (i = 0; i < num_tc; i++) + total_bw += etscfg->tcbwtable[i]; + + if (!total_bw) { + etscfg->tcbwtable[0] = ICE_TC_MAX_BW; + } else if (total_bw != ICE_TC_MAX_BW) { + dev_err(ice_pf_to_dev(pf), "Invalid config, total bandwidth must equal 100\n"); + return -EINVAL; + } + + return 0; +} + +/** * ice_pf_dcb_cfg - Apply new DCB configuration * @pf: pointer to the PF struct * @new_cfg: DCBX config to apply @@ -182,6 +239,9 @@ int ice_pf_dcb_cfg(struct ice_pf *pf, struct ice_dcbx_cfg *new_cfg, bool locked) return ret; } + if (ice_dcb_bwchk(pf, new_cfg)) + return -EINVAL; + /* Store old config in case FW config fails */ old_cfg = kmemdup(curr_cfg, sizeof(*old_cfg), GFP_KERNEL); if (!old_cfg) @@ -605,14 +665,14 @@ int ice_init_pf_dcb(struct ice_pf *pf, bool locked) ice_cfg_sw_lldp(pf_vsi, false, true); - pf->dcbx_cap = DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE; + pf->dcbx_cap = ice_dcb_get_mode(port_info, true); return 0; } set_bit(ICE_FLAG_FW_LLDP_AGENT, pf->flags); - /* DCBX in FW and LLDP enabled in FW */ - pf->dcbx_cap = DCB_CAP_DCBX_LLD_MANAGED | DCB_CAP_DCBX_VER_IEEE; + /* DCBX/LLDP enabled in FW, set DCBNL mode advertisement */ + pf->dcbx_cap = ice_dcb_get_mode(port_info, false); err = ice_dcb_init_cfg(pf, locked); if (err) @@ -719,7 +779,7 @@ ice_dcb_process_lldp_set_mib_change(struct ice_pf *pf, bool need_reconfig = false; struct ice_port_info *pi; struct ice_vsi *pf_vsi; - u8 type; + u8 mib_type; int ret; /* Not DCB capable or capability disabled */ @@ -734,16 +794,16 @@ ice_dcb_process_lldp_set_mib_change(struct ice_pf *pf, pi = pf->hw.port_info; mib = (struct ice_aqc_lldp_get_mib *)&event->desc.params.raw; /* Ignore if event is not for Nearest Bridge */ - type = ((mib->type >> ICE_AQ_LLDP_BRID_TYPE_S) & - ICE_AQ_LLDP_BRID_TYPE_M); - dev_dbg(dev, "LLDP event MIB bridge type 0x%x\n", type); - if (type != ICE_AQ_LLDP_BRID_TYPE_NEAREST_BRID) + mib_type = ((mib->type >> ICE_AQ_LLDP_BRID_TYPE_S) & + ICE_AQ_LLDP_BRID_TYPE_M); + dev_dbg(dev, "LLDP event MIB bridge type 0x%x\n", mib_type); + if (mib_type != ICE_AQ_LLDP_BRID_TYPE_NEAREST_BRID) return; /* Check MIB Type and return if event for Remote MIB update */ - type = mib->type & ICE_AQ_LLDP_MIB_TYPE_M; - dev_dbg(dev, "LLDP event mib type %s\n", type ? "remote" : "local"); - if (type == ICE_AQ_LLDP_MIB_REMOTE) { + mib_type = mib->type & ICE_AQ_LLDP_MIB_TYPE_M; + dev_dbg(dev, "LLDP event mib type %s\n", mib_type ? "remote" : "local"); + if (mib_type == ICE_AQ_LLDP_MIB_REMOTE) { /* Update the remote cached instance and return */ ret = ice_aq_get_dcb_cfg(pi->hw, ICE_AQ_LLDP_MIB_REMOTE, ICE_AQ_LLDP_BRID_TYPE_NEAREST_BRID, @@ -775,6 +835,8 @@ ice_dcb_process_lldp_set_mib_change(struct ice_pf *pf, goto out; } + pf->dcbx_cap = ice_dcb_get_mode(pi, false); + need_reconfig = ice_dcb_need_recfg(pf, &tmp_dcbx_cfg, &pi->local_dcbx_cfg); ice_dcbnl_flush_apps(pf, &tmp_dcbx_cfg, &pi->local_dcbx_cfg); diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.h b/drivers/net/ethernet/intel/ice/ice_dcb_lib.h index f15e5776f287..37680e815b02 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.h @@ -20,6 +20,7 @@ u8 ice_dcb_get_num_tc(struct ice_dcbx_cfg *dcbcfg); u8 ice_dcb_get_tc(struct ice_vsi *vsi, int queue_index); int ice_pf_dcb_cfg(struct ice_pf *pf, struct ice_dcbx_cfg *new_cfg, bool locked); +int ice_dcb_bwchk(struct ice_pf *pf, struct ice_dcbx_cfg *dcbcfg); void ice_pf_dcb_recfg(struct ice_pf *pf); void ice_vsi_cfg_dcb_rings(struct ice_vsi *vsi); int ice_init_pf_dcb(struct ice_pf *pf, bool locked); diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_nl.c b/drivers/net/ethernet/intel/ice/ice_dcb_nl.c index b61aba428adb..c4c12414083a 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_nl.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb_nl.c @@ -95,14 +95,12 @@ static int ice_dcbnl_setets(struct net_device *netdev, struct ieee_ets *ets) new_cfg->etsrec.prio_table[i] = ets->reco_prio_tc[i]; } - /* max_tc is a 1-8 value count of number of TC's, not a 0-7 value - * for the TC's index number. Add one to value if not zero, and - * for zero set it to the FW's default value - */ - if (max_tc) - max_tc++; - else - max_tc = IEEE_8021QAZ_MAX_TCS; + if (ice_dcb_bwchk(pf, new_cfg)) { + err = -EINVAL; + goto ets_out; + } + + max_tc = pf->hw.func_caps.common_cap.maxtc; new_cfg->etscfg.maxtcs = max_tc; @@ -119,6 +117,7 @@ static int ice_dcbnl_setets(struct net_device *netdev, struct ieee_ets *ets) if (err == ICE_DCB_NO_HW_CHG) err = ICE_DCB_HW_CHG_RST; +ets_out: mutex_unlock(&pf->tc_mutex); return err; } @@ -535,6 +534,30 @@ ice_dcbnl_get_pg_tc_cfg_rx(struct net_device *netdev, int prio, } /** + * ice_dcbnl_set_pg_tc_cfg_rx + * @netdev: relevant netdev struct + * @prio: corresponding user priority + * @prio_type: the traffic priority type + * @pgid: the PG ID + * @bw_pct: BW percentage for corresponding BWG + * @up_map: prio mapped to corresponding TC + * + * lldpad requires this function pointer to be non-NULL to complete CEE config. + */ +static void +ice_dcbnl_set_pg_tc_cfg_rx(struct net_device *netdev, + int __always_unused prio, + u8 __always_unused prio_type, + u8 __always_unused pgid, + u8 __always_unused bw_pct, + u8 __always_unused up_map) +{ + struct ice_pf *pf = ice_netdev_to_pf(netdev); + + dev_dbg(ice_pf_to_dev(pf), "Rx TC PG Config Not Supported.\n"); +} + +/** * ice_dcbnl_get_pg_bwg_cfg_rx - Get CEE PG BW Rx config * @netdev: pointer to netdev struct * @pgid: the corresponding traffic class @@ -554,6 +577,23 @@ ice_dcbnl_get_pg_bwg_cfg_rx(struct net_device *netdev, int __always_unused pgid, } /** + * ice_dcbnl_set_pg_bwg_cfg_rx + * @netdev: the corresponding netdev + * @pgid: corresponding TC + * @bw_pct: BW percentage for given TC + * + * lldpad requires this function pointer to be non-NULL to complete CEE config. + */ +static void +ice_dcbnl_set_pg_bwg_cfg_rx(struct net_device *netdev, int __always_unused pgid, + u8 __always_unused bw_pct) +{ + struct ice_pf *pf = ice_netdev_to_pf(netdev); + + dev_dbg(ice_pf_to_dev(pf), "Rx BWG PG Config Not Supported.\n"); +} + +/** * ice_dcbnl_get_cap - Get DCBX capabilities of adapter * @netdev: pointer to netdev struct * @capid: the capability type @@ -799,6 +839,8 @@ static const struct dcbnl_rtnl_ops dcbnl_ops = { .getpermhwaddr = ice_dcbnl_get_perm_hw_addr, .setpgtccfgtx = ice_dcbnl_set_pg_tc_cfg_tx, .setpgbwgcfgtx = ice_dcbnl_set_pg_bwg_cfg_tx, + .setpgtccfgrx = ice_dcbnl_set_pg_tc_cfg_rx, + .setpgbwgcfgrx = ice_dcbnl_set_pg_bwg_cfg_rx, .getpgtccfgtx = ice_dcbnl_get_pg_tc_cfg_tx, .getpgbwgcfgtx = ice_dcbnl_get_pg_bwg_cfg_tx, .getpgtccfgrx = ice_dcbnl_get_pg_tc_cfg_rx, diff --git a/drivers/net/ethernet/intel/ice/ice_devids.h b/drivers/net/ethernet/intel/ice/ice_devids.h index ce63017c56c7..9d8194671f6a 100644 --- a/drivers/net/ethernet/intel/ice/ice_devids.h +++ b/drivers/net/ethernet/intel/ice/ice_devids.h @@ -5,12 +5,34 @@ #define _ICE_DEVIDS_H_ /* Device IDs */ +/* Intel(R) Ethernet Connection E823-L for backplane */ +#define ICE_DEV_ID_E823L_BACKPLANE 0x124C +/* Intel(R) Ethernet Connection E823-L for SFP */ +#define ICE_DEV_ID_E823L_SFP 0x124D +/* Intel(R) Ethernet Connection E823-L/X557-AT 10GBASE-T */ +#define ICE_DEV_ID_E823L_10G_BASE_T 0x124E +/* Intel(R) Ethernet Connection E823-L 1GbE */ +#define ICE_DEV_ID_E823L_1GBE 0x124F +/* Intel(R) Ethernet Connection E823-L for QSFP */ +#define ICE_DEV_ID_E823L_QSFP 0x151D /* Intel(R) Ethernet Controller E810-C for backplane */ #define ICE_DEV_ID_E810C_BACKPLANE 0x1591 /* Intel(R) Ethernet Controller E810-C for QSFP */ #define ICE_DEV_ID_E810C_QSFP 0x1592 /* Intel(R) Ethernet Controller E810-C for SFP */ #define ICE_DEV_ID_E810C_SFP 0x1593 +/* Intel(R) Ethernet Controller E810-XXV for SFP */ +#define ICE_DEV_ID_E810_XXV_SFP 0x159B +/* Intel(R) Ethernet Connection E823-C for backplane */ +#define ICE_DEV_ID_E823C_BACKPLANE 0x188A +/* Intel(R) Ethernet Connection E823-C for QSFP */ +#define ICE_DEV_ID_E823C_QSFP 0x188B +/* Intel(R) Ethernet Connection E823-C for SFP */ +#define ICE_DEV_ID_E823C_SFP 0x188C +/* Intel(R) Ethernet Connection E823-C/X557-AT 10GBASE-T */ +#define ICE_DEV_ID_E823C_10G_BASE_T 0x188D +/* Intel(R) Ethernet Connection E823-C 1GbE */ +#define ICE_DEV_ID_E823C_SGMII 0x188E /* Intel(R) Ethernet Connection E822-C for backplane */ #define ICE_DEV_ID_E822C_BACKPLANE 0x1890 /* Intel(R) Ethernet Connection E822-C for QSFP */ @@ -21,8 +43,8 @@ #define ICE_DEV_ID_E822C_10G_BASE_T 0x1893 /* Intel(R) Ethernet Connection E822-C 1GbE */ #define ICE_DEV_ID_E822C_SGMII 0x1894 -/* Intel(R) Ethernet Connection E822-X for backplane */ -#define ICE_DEV_ID_E822X_BACKPLANE 0x1897 +/* Intel(R) Ethernet Connection E822-L for backplane */ +#define ICE_DEV_ID_E822L_BACKPLANE 0x1897 /* Intel(R) Ethernet Connection E822-L for SFP */ #define ICE_DEV_ID_E822L_SFP 0x1898 /* Intel(R) Ethernet Connection E822-L/X557-AT 10GBASE-T */ diff --git a/drivers/net/ethernet/intel/ice/ice_devlink.c b/drivers/net/ethernet/intel/ice/ice_devlink.c new file mode 100644 index 000000000000..c6833944b90a --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_devlink.c @@ -0,0 +1,416 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020, Intel Corporation. */ + +#include "ice.h" +#include "ice_lib.h" +#include "ice_devlink.h" + +static int ice_info_get_dsn(struct ice_pf *pf, char *buf, size_t len) +{ + u8 dsn[8]; + + /* Copy the DSN into an array in Big Endian format */ + put_unaligned_be64(pci_get_dsn(pf->pdev), dsn); + + snprintf(buf, len, "%02x-%02x-%02x-%02x-%02x-%02x-%02x-%02x", + dsn[0], dsn[1], dsn[2], dsn[3], + dsn[4], dsn[5], dsn[6], dsn[7]); + + return 0; +} + +static int ice_info_pba(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_hw *hw = &pf->hw; + enum ice_status status; + + status = ice_read_pba_string(hw, (u8 *)buf, len); + if (status) + return -EIO; + + return 0; +} + +static int ice_info_fw_mgmt(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_hw *hw = &pf->hw; + + snprintf(buf, len, "%u.%u.%u", hw->fw_maj_ver, hw->fw_min_ver, + hw->fw_patch); + + return 0; +} + +static int ice_info_fw_api(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_hw *hw = &pf->hw; + + snprintf(buf, len, "%u.%u", hw->api_maj_ver, hw->api_min_ver); + + return 0; +} + +static int ice_info_fw_build(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_hw *hw = &pf->hw; + + snprintf(buf, len, "0x%08x", hw->fw_build); + + return 0; +} + +static int ice_info_orom_ver(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_orom_info *orom = &pf->hw.nvm.orom; + + snprintf(buf, len, "%u.%u.%u", orom->major, orom->build, orom->patch); + + return 0; +} + +static int ice_info_nvm_ver(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_nvm_info *nvm = &pf->hw.nvm; + + snprintf(buf, len, "%x.%02x", nvm->major_ver, nvm->minor_ver); + + return 0; +} + +static int ice_info_eetrack(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_nvm_info *nvm = &pf->hw.nvm; + + snprintf(buf, len, "0x%08x", nvm->eetrack); + + return 0; +} + +static int ice_info_ddp_pkg_name(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_hw *hw = &pf->hw; + + snprintf(buf, len, "%s", hw->active_pkg_name); + + return 0; +} + +static int ice_info_ddp_pkg_version(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_pkg_ver *pkg = &pf->hw.active_pkg_ver; + + snprintf(buf, len, "%u.%u.%u.%u", pkg->major, pkg->minor, pkg->update, + pkg->draft); + + return 0; +} + +#define fixed(key, getter) { ICE_VERSION_FIXED, key, getter } +#define running(key, getter) { ICE_VERSION_RUNNING, key, getter } + +enum ice_version_type { + ICE_VERSION_FIXED, + ICE_VERSION_RUNNING, + ICE_VERSION_STORED, +}; + +static const struct ice_devlink_version { + enum ice_version_type type; + const char *key; + int (*getter)(struct ice_pf *pf, char *buf, size_t len); +} ice_devlink_versions[] = { + fixed(DEVLINK_INFO_VERSION_GENERIC_BOARD_ID, ice_info_pba), + running(DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, ice_info_fw_mgmt), + running("fw.mgmt.api", ice_info_fw_api), + running("fw.mgmt.build", ice_info_fw_build), + running(DEVLINK_INFO_VERSION_GENERIC_FW_UNDI, ice_info_orom_ver), + running("fw.psid.api", ice_info_nvm_ver), + running(DEVLINK_INFO_VERSION_GENERIC_FW_BUNDLE_ID, ice_info_eetrack), + running("fw.app.name", ice_info_ddp_pkg_name), + running(DEVLINK_INFO_VERSION_GENERIC_FW_APP, ice_info_ddp_pkg_version), +}; + +/** + * ice_devlink_info_get - .info_get devlink handler + * @devlink: devlink instance structure + * @req: the devlink info request + * @extack: extended netdev ack structure + * + * Callback for the devlink .info_get operation. Reports information about the + * device. + * + * Return: zero on success or an error code on failure. + */ +static int ice_devlink_info_get(struct devlink *devlink, + struct devlink_info_req *req, + struct netlink_ext_ack *extack) +{ + struct ice_pf *pf = devlink_priv(devlink); + char buf[100]; + size_t i; + int err; + + err = devlink_info_driver_name_put(req, KBUILD_MODNAME); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Unable to set driver name"); + return err; + } + + err = ice_info_get_dsn(pf, buf, sizeof(buf)); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Unable to obtain serial number"); + return err; + } + + err = devlink_info_serial_number_put(req, buf); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Unable to set serial number"); + return err; + } + + for (i = 0; i < ARRAY_SIZE(ice_devlink_versions); i++) { + enum ice_version_type type = ice_devlink_versions[i].type; + const char *key = ice_devlink_versions[i].key; + + err = ice_devlink_versions[i].getter(pf, buf, sizeof(buf)); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Unable to obtain version info"); + return err; + } + + switch (type) { + case ICE_VERSION_FIXED: + err = devlink_info_version_fixed_put(req, key, buf); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Unable to set fixed version"); + return err; + } + break; + case ICE_VERSION_RUNNING: + err = devlink_info_version_running_put(req, key, buf); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Unable to set running version"); + return err; + } + break; + case ICE_VERSION_STORED: + err = devlink_info_version_stored_put(req, key, buf); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Unable to set stored version"); + return err; + } + break; + } + } + + return 0; +} + +static const struct devlink_ops ice_devlink_ops = { + .info_get = ice_devlink_info_get, +}; + +static void ice_devlink_free(void *devlink_ptr) +{ + devlink_free((struct devlink *)devlink_ptr); +} + +/** + * ice_allocate_pf - Allocate devlink and return PF structure pointer + * @dev: the device to allocate for + * + * Allocate a devlink instance for this device and return the private area as + * the PF structure. The devlink memory is kept track of through devres by + * adding an action to remove it when unwinding. + */ +struct ice_pf *ice_allocate_pf(struct device *dev) +{ + struct devlink *devlink; + + devlink = devlink_alloc(&ice_devlink_ops, sizeof(struct ice_pf)); + if (!devlink) + return NULL; + + /* Add an action to teardown the devlink when unwinding the driver */ + if (devm_add_action(dev, ice_devlink_free, devlink)) { + devlink_free(devlink); + return NULL; + } + + return devlink_priv(devlink); +} + +/** + * ice_devlink_register - Register devlink interface for this PF + * @pf: the PF to register the devlink for. + * + * Register the devlink instance associated with this physical function. + * + * Return: zero on success or an error code on failure. + */ +int ice_devlink_register(struct ice_pf *pf) +{ + struct devlink *devlink = priv_to_devlink(pf); + struct device *dev = ice_pf_to_dev(pf); + int err; + + err = devlink_register(devlink, dev); + if (err) { + dev_err(dev, "devlink registration failed: %d\n", err); + return err; + } + + return 0; +} + +/** + * ice_devlink_unregister - Unregister devlink resources for this PF. + * @pf: the PF structure to cleanup + * + * Releases resources used by devlink and cleans up associated memory. + */ +void ice_devlink_unregister(struct ice_pf *pf) +{ + devlink_unregister(priv_to_devlink(pf)); +} + +/** + * ice_devlink_create_port - Create a devlink port for this PF + * @pf: the PF to create a port for + * + * Create and register a devlink_port for this PF. Note that although each + * physical function is connected to a separate devlink instance, the port + * will still be numbered according to the physical function id. + * + * Return: zero on success or an error code on failure. + */ +int ice_devlink_create_port(struct ice_pf *pf) +{ + struct devlink *devlink = priv_to_devlink(pf); + struct ice_vsi *vsi = ice_get_main_vsi(pf); + struct device *dev = ice_pf_to_dev(pf); + int err; + + if (!vsi) { + dev_err(dev, "%s: unable to find main VSI\n", __func__); + return -EIO; + } + + devlink_port_attrs_set(&pf->devlink_port, DEVLINK_PORT_FLAVOUR_PHYSICAL, + pf->hw.pf_id, false, 0, NULL, 0); + err = devlink_port_register(devlink, &pf->devlink_port, pf->hw.pf_id); + if (err) { + dev_err(dev, "devlink_port_register failed: %d\n", err); + return err; + } + + return 0; +} + +/** + * ice_devlink_destroy_port - Destroy the devlink_port for this PF + * @pf: the PF to cleanup + * + * Unregisters the devlink_port structure associated with this PF. + */ +void ice_devlink_destroy_port(struct ice_pf *pf) +{ + devlink_port_type_clear(&pf->devlink_port); + devlink_port_unregister(&pf->devlink_port); +} + +/** + * ice_devlink_nvm_snapshot - Capture a snapshot of the Shadow RAM contents + * @devlink: the devlink instance + * @extack: extended ACK response structure + * @data: on exit points to snapshot data buffer + * + * This function is called in response to the DEVLINK_CMD_REGION_TRIGGER for + * the shadow-ram devlink region. It captures a snapshot of the shadow ram + * contents. This snapshot can later be viewed via the devlink-region + * interface. + * + * @returns zero on success, and updates the data pointer. Returns a non-zero + * error code on failure. + */ +static int ice_devlink_nvm_snapshot(struct devlink *devlink, + struct netlink_ext_ack *extack, u8 **data) +{ + struct ice_pf *pf = devlink_priv(devlink); + struct device *dev = ice_pf_to_dev(pf); + struct ice_hw *hw = &pf->hw; + enum ice_status status; + void *nvm_data; + u32 nvm_size; + + nvm_size = hw->nvm.flash_size; + nvm_data = vzalloc(nvm_size); + if (!nvm_data) + return -ENOMEM; + + status = ice_acquire_nvm(hw, ICE_RES_READ); + if (status) { + dev_dbg(dev, "ice_acquire_nvm failed, err %d aq_err %d\n", + status, hw->adminq.sq_last_status); + NL_SET_ERR_MSG_MOD(extack, "Failed to acquire NVM semaphore"); + vfree(nvm_data); + return -EIO; + } + + status = ice_read_flat_nvm(hw, 0, &nvm_size, nvm_data, false); + if (status) { + dev_dbg(dev, "ice_read_flat_nvm failed after reading %u bytes, err %d aq_err %d\n", + nvm_size, status, hw->adminq.sq_last_status); + NL_SET_ERR_MSG_MOD(extack, "Failed to read NVM contents"); + ice_release_nvm(hw); + vfree(nvm_data); + return -EIO; + } + + ice_release_nvm(hw); + + *data = nvm_data; + + return 0; +} + +static const struct devlink_region_ops ice_nvm_region_ops = { + .name = "nvm-flash", + .destructor = vfree, + .snapshot = ice_devlink_nvm_snapshot, +}; + +/** + * ice_devlink_init_regions - Initialize devlink regions + * @pf: the PF device structure + * + * Create devlink regions used to enable access to dump the contents of the + * flash memory on the device. + */ +void ice_devlink_init_regions(struct ice_pf *pf) +{ + struct devlink *devlink = priv_to_devlink(pf); + struct device *dev = ice_pf_to_dev(pf); + u64 nvm_size; + + nvm_size = pf->hw.nvm.flash_size; + pf->nvm_region = devlink_region_create(devlink, &ice_nvm_region_ops, 1, + nvm_size); + if (IS_ERR(pf->nvm_region)) { + dev_err(dev, "failed to create NVM devlink region, err %ld\n", + PTR_ERR(pf->nvm_region)); + pf->nvm_region = NULL; + } +} + +/** + * ice_devlink_destroy_regions - Destroy devlink regions + * @pf: the PF device structure + * + * Remove previously created regions for this PF. + */ +void ice_devlink_destroy_regions(struct ice_pf *pf) +{ + if (pf->nvm_region) + devlink_region_destroy(pf->nvm_region); +} diff --git a/drivers/net/ethernet/intel/ice/ice_devlink.h b/drivers/net/ethernet/intel/ice/ice_devlink.h new file mode 100644 index 000000000000..6e806a08dc23 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_devlink.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019, Intel Corporation. */ + +#ifndef _ICE_DEVLINK_H_ +#define _ICE_DEVLINK_H_ + +struct ice_pf *ice_allocate_pf(struct device *dev); + +int ice_devlink_register(struct ice_pf *pf); +void ice_devlink_unregister(struct ice_pf *pf); +int ice_devlink_create_port(struct ice_pf *pf); +void ice_devlink_destroy_port(struct ice_pf *pf); + +void ice_devlink_init_regions(struct ice_pf *pf); +void ice_devlink_destroy_regions(struct ice_pf *pf); + +#endif /* _ICE_DEVLINK_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index 77c412a7e7a4..593fb37bd59e 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -157,6 +157,7 @@ struct ice_priv_flag { static const struct ice_priv_flag ice_gstrings_priv_flags[] = { ICE_PRIV_FLAG("link-down-on-close", ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA), ICE_PRIV_FLAG("fw-lldp-agent", ICE_FLAG_FW_LLDP_AGENT), + ICE_PRIV_FLAG("mdd-auto-reset-vf", ICE_FLAG_MDD_AUTO_RESET_VF), ICE_PRIV_FLAG("legacy-rx", ICE_FLAG_LEGACY_RX), }; @@ -166,25 +167,26 @@ static void ice_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { struct ice_netdev_priv *np = netdev_priv(netdev); - u8 oem_ver, oem_patch, nvm_ver_hi, nvm_ver_lo; struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; - u16 oem_build; + struct ice_orom_info *orom; + struct ice_nvm_info *nvm; - strlcpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, ice_drv_ver, sizeof(drvinfo->version)); + nvm = &hw->nvm; + orom = &nvm->orom; + + strscpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver)); + strscpy(drvinfo->version, ice_drv_ver, sizeof(drvinfo->version)); /* Display NVM version (from which the firmware version can be * determined) which contains more pertinent information. */ - ice_get_nvm_version(hw, &oem_ver, &oem_build, &oem_patch, - &nvm_ver_hi, &nvm_ver_lo); snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), - "%x.%02x 0x%x %d.%d.%d", nvm_ver_hi, nvm_ver_lo, - hw->nvm.eetrack, oem_ver, oem_build, oem_patch); + "%x.%02x 0x%x %d.%d.%d", nvm->major_ver, nvm->minor_ver, + nvm->eetrack, orom->major, orom->build, orom->patch); - strlcpy(drvinfo->bus_info, pci_name(pf->pdev), + strscpy(drvinfo->bus_info, pci_name(pf->pdev), sizeof(drvinfo->bus_info)); drvinfo->n_priv_flags = ICE_PRIV_FLAG_ARRAY_SIZE; } @@ -243,7 +245,7 @@ static int ice_get_eeprom_len(struct net_device *netdev) struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_pf *pf = np->vsi->back; - return (int)(pf->hw.nvm.sr_words * sizeof(u16)); + return (int)pf->hw.nvm.flash_size; } static int @@ -251,39 +253,46 @@ ice_get_eeprom(struct net_device *netdev, struct ethtool_eeprom *eeprom, u8 *bytes) { struct ice_netdev_priv *np = netdev_priv(netdev); - u16 first_word, last_word, nwords; struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; enum ice_status status; struct device *dev; int ret = 0; - u16 *buf; + u8 *buf; dev = ice_pf_to_dev(pf); eeprom->magic = hw->vendor_id | (hw->device_id << 16); + netdev_dbg(netdev, "GEEPROM cmd 0x%08x, offset 0x%08x, len 0x%08x\n", + eeprom->cmd, eeprom->offset, eeprom->len); - first_word = eeprom->offset >> 1; - last_word = (eeprom->offset + eeprom->len - 1) >> 1; - nwords = last_word - first_word + 1; - - buf = devm_kcalloc(dev, nwords, sizeof(u16), GFP_KERNEL); + buf = kzalloc(eeprom->len, GFP_KERNEL); if (!buf) return -ENOMEM; - status = ice_read_sr_buf(hw, first_word, &nwords, buf); + status = ice_acquire_nvm(hw, ICE_RES_READ); if (status) { - dev_err(dev, "ice_read_sr_buf failed, err %d aq_err %d\n", + dev_err(dev, "ice_acquire_nvm failed, err %d aq_err %d\n", status, hw->adminq.sq_last_status); - eeprom->len = sizeof(u16) * nwords; ret = -EIO; goto out; } - memcpy(bytes, (u8 *)buf + (eeprom->offset & 1), eeprom->len); + status = ice_read_flat_nvm(hw, eeprom->offset, &eeprom->len, buf, + false); + if (status) { + dev_err(dev, "ice_read_flat_nvm failed, err %d aq_err %d\n", + status, hw->adminq.sq_last_status); + ret = -EIO; + goto release; + } + + memcpy(bytes, buf, eeprom->len); +release: + ice_release_nvm(hw); out: - devm_kfree(dev, buf); + kfree(buf); return ret; } @@ -462,7 +471,7 @@ static int ice_lbtest_prepare_rings(struct ice_vsi *vsi) if (status) goto err_setup_rx_ring; - status = ice_vsi_start_rx_rings(vsi); + status = ice_vsi_start_all_rx_rings(vsi); if (status) goto err_start_rx_ring; @@ -494,7 +503,7 @@ static int ice_lbtest_disable_rings(struct ice_vsi *vsi) netdev_err(vsi->netdev, "Failed to stop Tx rings, VSI %d error %d\n", vsi->vsi_num, status); - status = ice_vsi_stop_rx_rings(vsi); + status = ice_vsi_stop_all_rx_rings(vsi); if (status) netdev_err(vsi->netdev, "Failed to stop Rx rings, VSI %d error %d\n", vsi->vsi_num, status); @@ -672,7 +681,7 @@ static u64 ice_loopback_test(struct net_device *netdev) test_vsi = ice_lb_vsi_setup(pf, pf->hw.port_info); if (!test_vsi) { - netdev_err(netdev, "Failed to create a VSI for the loopback test"); + netdev_err(netdev, "Failed to create a VSI for the loopback test\n"); return 1; } @@ -731,7 +740,7 @@ lbtest_free_frame: devm_kfree(dev, tx_frame); remove_mac_filters: if (ice_remove_mac(&pf->hw, &tmp_list)) - netdev_err(netdev, "Could not remove MAC filter for the test VSI"); + netdev_err(netdev, "Could not remove MAC filter for the test VSI\n"); free_mac_list: ice_free_fltr_list(dev, &tmp_list); lbtest_mac_dis: @@ -744,7 +753,7 @@ lbtest_rings_dis: lbtest_vsi_close: test_vsi->netdev = NULL; if (ice_vsi_release(test_vsi)) - netdev_err(netdev, "Failed to remove the test VSI"); + netdev_err(netdev, "Failed to remove the test VSI\n"); return ret; } @@ -834,7 +843,7 @@ ice_self_test(struct net_device *netdev, struct ethtool_test *eth_test, int status = ice_open(netdev); if (status) { - dev_err(dev, "Could not open device %s, err %d", + dev_err(dev, "Could not open device %s, err %d\n", pf->int_name, status); } } @@ -1091,7 +1100,6 @@ ice_get_fecparam(struct net_device *netdev, struct ethtool_fecparam *fecparam) fecparam->active_fec = ETHTOOL_FEC_BASER; break; case ICE_AQ_LINK_25G_RS_528_FEC_EN: - /* fall through */ case ICE_AQ_LINK_25G_RS_544_FEC_EN: fecparam->active_fec = ETHTOOL_FEC_RS; break; @@ -1132,6 +1140,33 @@ done: } /** + * ice_nway_reset - restart autonegotiation + * @netdev: network interface device structure + */ +static int ice_nway_reset(struct net_device *netdev) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + struct ice_port_info *pi; + enum ice_status status; + + pi = vsi->port_info; + /* If VSI state is up, then restart autoneg with link up */ + if (!test_bit(__ICE_DOWN, vsi->back->state)) + status = ice_aq_set_link_restart_an(pi, true, NULL); + else + status = ice_aq_set_link_restart_an(pi, false, NULL); + + if (status) { + netdev_info(netdev, "link restart failed, err %d aq_err %d\n", + status, pi->hw->adminq.sq_last_status); + return -EIO; + } + + return 0; +} + +/** * ice_get_priv_flags - report device private flags * @netdev: network interface device structure * @@ -1264,6 +1299,8 @@ static int ice_set_priv_flags(struct net_device *netdev, u32 flags) status = ice_cfg_lldp_mib_change(&pf->hw, true); if (status) dev_dbg(dev, "Fail to enable MIB change events\n"); + + ice_nway_reset(netdev); } } if (test_bit(ICE_FLAG_LEGACY_RX, change_flags)) { @@ -1781,7 +1818,6 @@ ice_get_settings_link_up(struct ethtool_link_ksettings *ks, Asym_Pause); break; case ICE_FC_PFC: - /* fall through */ default: ethtool_link_ksettings_del_link_mode(ks, lp_advertising, Pause); ethtool_link_ksettings_del_link_mode(ks, lp_advertising, @@ -2776,30 +2812,6 @@ done: return err; } -static int ice_nway_reset(struct net_device *netdev) -{ - /* restart autonegotiation */ - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *vsi = np->vsi; - struct ice_port_info *pi; - enum ice_status status; - - pi = vsi->port_info; - /* If VSI state is up, then restart autoneg with link up */ - if (!test_bit(__ICE_DOWN, vsi->back->state)) - status = ice_aq_set_link_restart_an(pi, true, NULL); - else - status = ice_aq_set_link_restart_an(pi, false, NULL); - - if (status) { - netdev_info(netdev, "link restart failed, err %d aq_err %d\n", - status, pi->hw->adminq.sq_last_status); - return -EIO; - } - - return 0; -} - /** * ice_get_pauseparam - Get Flow Control status * @netdev: network interface device structure @@ -3453,12 +3465,6 @@ ice_set_rc_coalesce(enum ice_container_type c_type, struct ethtool_coalesce *ec, break; case ICE_TX_CONTAINER: - if (ec->tx_coalesce_usecs_high) { - netdev_info(vsi->netdev, "setting %s-usecs-high is not supported\n", - c_type_str); - return -EINVAL; - } - use_adaptive_coalesce = ec->use_adaptive_tx_coalesce; coalesce_usecs = ec->tx_coalesce_usecs; @@ -3535,53 +3541,6 @@ ice_set_q_coalesce(struct ice_vsi *vsi, struct ethtool_coalesce *ec, int q_num) } /** - * ice_is_coalesce_param_invalid - check for unsupported coalesce parameters - * @netdev: pointer to the netdev associated with this query - * @ec: ethtool structure to fill with driver's coalesce settings - * - * Print netdev info if driver doesn't support one of the parameters - * and return error. When any parameters will be implemented, remove only - * this parameter from param array. - */ -static int -ice_is_coalesce_param_invalid(struct net_device *netdev, - struct ethtool_coalesce *ec) -{ - struct ice_ethtool_not_used { - u32 value; - const char *name; - } param[] = { - {ec->stats_block_coalesce_usecs, "stats-block-usecs"}, - {ec->rate_sample_interval, "sample-interval"}, - {ec->pkt_rate_low, "pkt-rate-low"}, - {ec->pkt_rate_high, "pkt-rate-high"}, - {ec->rx_max_coalesced_frames, "rx-frames"}, - {ec->rx_coalesce_usecs_irq, "rx-usecs-irq"}, - {ec->rx_max_coalesced_frames_irq, "rx-frames-irq"}, - {ec->tx_max_coalesced_frames, "tx-frames"}, - {ec->tx_coalesce_usecs_irq, "tx-usecs-irq"}, - {ec->tx_max_coalesced_frames_irq, "tx-frames-irq"}, - {ec->rx_coalesce_usecs_low, "rx-usecs-low"}, - {ec->rx_max_coalesced_frames_low, "rx-frames-low"}, - {ec->tx_coalesce_usecs_low, "tx-usecs-low"}, - {ec->tx_max_coalesced_frames_low, "tx-frames-low"}, - {ec->rx_max_coalesced_frames_high, "rx-frames-high"}, - {ec->tx_max_coalesced_frames_high, "tx-frames-high"} - }; - int i; - - for (i = 0; i < ARRAY_SIZE(param); i++) { - if (param[i].value) { - netdev_info(netdev, "Setting %s not supported\n", - param[i].name); - return -EINVAL; - } - } - - return 0; -} - -/** * ice_print_if_odd_usecs - print message if user tries to set odd [tx|rx]-usecs * @netdev: netdev used for print * @itr_setting: previous user setting @@ -3621,9 +3580,6 @@ __ice_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec, struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; - if (ice_is_coalesce_param_invalid(netdev, ec)) - return -EINVAL; - if (q_num < 0) { struct ice_q_vector *q_vector = vsi->q_vectors[0]; int v_idx; @@ -3818,6 +3774,9 @@ ice_get_module_eeprom(struct net_device *netdev, } static const struct ethtool_ops ice_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_USE_ADAPTIVE | + ETHTOOL_COALESCE_RX_USECS_HIGH, .get_link_ksettings = ice_get_link_ksettings, .set_link_ksettings = ice_set_link_ksettings, .get_drvinfo = ice_get_drvinfo, @@ -3867,6 +3826,7 @@ static const struct ethtool_ops ice_ethtool_safe_mode_ops = { .get_regs = ice_get_regs, .get_msglevel = ice_get_msglevel, .set_msglevel = ice_set_msglevel, + .get_link = ethtool_op_get_link, .get_eeprom_len = ice_get_eeprom_len, .get_eeprom = ice_get_eeprom, .get_strings = ice_get_strings, diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c index 99208946224c..42bac3ec5526 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c @@ -3471,6 +3471,24 @@ ice_move_vsi(struct ice_hw *hw, enum ice_block blk, u16 vsi, u16 vsig, } /** + * ice_rem_chg_tcam_ent - remove a specific TCAM entry from change list + * @hw: pointer to the HW struct + * @idx: the index of the TCAM entry to remove + * @chg: the list of change structures to search + */ +static void +ice_rem_chg_tcam_ent(struct ice_hw *hw, u16 idx, struct list_head *chg) +{ + struct ice_chs_chg *pos, *tmp; + + list_for_each_entry_safe(tmp, pos, chg, list_entry) + if (tmp->type == ICE_TCAM_ADD && tmp->tcam_idx == idx) { + list_del(&tmp->list_entry); + devm_kfree(ice_hw_to_dev(hw), tmp); + } +} + +/** * ice_prof_tcam_ena_dis - add enable or disable TCAM change * @hw: pointer to the HW struct * @blk: hardware block @@ -3489,14 +3507,19 @@ ice_prof_tcam_ena_dis(struct ice_hw *hw, enum ice_block blk, bool enable, enum ice_status status; struct ice_chs_chg *p; - /* Default: enable means change the low flag bit to don't care */ - u8 dc_msk[ICE_TCAM_KEY_VAL_SZ] = { 0x01, 0x00, 0x00, 0x00, 0x00 }; + u8 vl_msk[ICE_TCAM_KEY_VAL_SZ] = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF }; + u8 dc_msk[ICE_TCAM_KEY_VAL_SZ] = { 0xFF, 0xFF, 0x00, 0x00, 0x00 }; u8 nm_msk[ICE_TCAM_KEY_VAL_SZ] = { 0x00, 0x00, 0x00, 0x00, 0x00 }; - u8 vl_msk[ICE_TCAM_KEY_VAL_SZ] = { 0x01, 0x00, 0x00, 0x00, 0x00 }; /* if disabling, free the TCAM */ if (!enable) { - status = ice_free_tcam_ent(hw, blk, tcam->tcam_idx); + status = ice_rel_tcam_idx(hw, blk, tcam->tcam_idx); + + /* if we have already created a change for this TCAM entry, then + * we need to remove that entry, in order to prevent writing to + * a TCAM entry we no longer will have ownership of. + */ + ice_rem_chg_tcam_ent(hw, tcam->tcam_idx, chg); tcam->tcam_idx = 0; tcam->in_use = 0; return status; @@ -3612,11 +3635,12 @@ ice_adj_prof_priorities(struct ice_hw *hw, enum ice_block blk, u16 vsig, * @blk: hardware block * @vsig: the VSIG to which this profile is to be added * @hdl: the profile handle indicating the profile to add + * @rev: true to add entries to the end of the list * @chg: the change list */ static enum ice_status ice_add_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl, - struct list_head *chg) + bool rev, struct list_head *chg) { /* Masks that ignore flags */ u8 vl_msk[ICE_TCAM_KEY_VAL_SZ] = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF }; @@ -3625,7 +3649,7 @@ ice_add_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl, struct ice_prof_map *map; struct ice_vsig_prof *t; struct ice_chs_chg *p; - u16 i; + u16 vsig_idx, i; /* Get the details on the profile specified by the handle ID */ map = ice_search_prof_id(hw, blk, hdl); @@ -3687,8 +3711,13 @@ ice_add_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl, } /* add profile to VSIG */ - list_add(&t->list, - &hw->blk[blk].xlt2.vsig_tbl[(vsig & ICE_VSIG_IDX_M)].prop_lst); + vsig_idx = vsig & ICE_VSIG_IDX_M; + if (rev) + list_add_tail(&t->list, + &hw->blk[blk].xlt2.vsig_tbl[vsig_idx].prop_lst); + else + list_add(&t->list, + &hw->blk[blk].xlt2.vsig_tbl[vsig_idx].prop_lst); return 0; @@ -3728,7 +3757,7 @@ ice_create_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl, if (status) goto err_ice_create_prof_id_vsig; - status = ice_add_prof_id_vsig(hw, blk, new_vsig, hdl, chg); + status = ice_add_prof_id_vsig(hw, blk, new_vsig, hdl, false, chg); if (status) goto err_ice_create_prof_id_vsig; @@ -3753,11 +3782,13 @@ err_ice_create_prof_id_vsig: * @blk: hardware block * @vsi: the initial VSI that will be in VSIG * @lst: the list of profile that will be added to the VSIG + * @new_vsig: return of new VSIG * @chg: the change list */ static enum ice_status ice_create_vsig_from_lst(struct ice_hw *hw, enum ice_block blk, u16 vsi, - struct list_head *lst, struct list_head *chg) + struct list_head *lst, u16 *new_vsig, + struct list_head *chg) { struct ice_vsig_prof *t; enum ice_status status; @@ -3772,12 +3803,15 @@ ice_create_vsig_from_lst(struct ice_hw *hw, enum ice_block blk, u16 vsi, return status; list_for_each_entry(t, lst, list) { + /* Reverse the order here since we are copying the list */ status = ice_add_prof_id_vsig(hw, blk, vsig, t->profile_cookie, - chg); + true, chg); if (status) return status; } + *new_vsig = vsig; + return 0; } @@ -3899,7 +3933,8 @@ ice_add_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl) * not sharing entries and we can simply add the new * profile to the VSIG. */ - status = ice_add_prof_id_vsig(hw, blk, vsig, hdl, &chg); + status = ice_add_prof_id_vsig(hw, blk, vsig, hdl, false, + &chg); if (status) goto err_ice_add_prof_id_flow; @@ -3910,7 +3945,8 @@ ice_add_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl) } else { /* No match, so we need a new VSIG */ status = ice_create_vsig_from_lst(hw, blk, vsi, - &union_lst, &chg); + &union_lst, &vsig, + &chg); if (status) goto err_ice_add_prof_id_flow; @@ -4076,7 +4112,8 @@ ice_rem_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl) * new VSIG and TCAM entries */ status = ice_create_vsig_from_lst(hw, blk, vsi, - ©, &chg); + ©, &vsig, + &chg); if (status) goto err_ice_rem_prof_id_flow; diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c index a05ceb59863b..3de862a3c789 100644 --- a/drivers/net/ethernet/intel/ice/ice_flow.c +++ b/drivers/net/ethernet/intel/ice/ice_flow.c @@ -694,7 +694,7 @@ out: * ice_flow_set_fld_ext - specifies locations of field from entry's input buffer * @seg: packet segment the field being set belongs to * @fld: field to be set - * @type: type of the field + * @field_type: type of the field * @val_loc: if not ICE_FLOW_FLD_OFF_INVAL, location of the value to match from * entry's input buffer * @mask_loc: if not ICE_FLOW_FLD_OFF_INVAL, location of mask value from entry's @@ -715,16 +715,16 @@ out: */ static void ice_flow_set_fld_ext(struct ice_flow_seg_info *seg, enum ice_flow_field fld, - enum ice_flow_fld_match_type type, u16 val_loc, + enum ice_flow_fld_match_type field_type, u16 val_loc, u16 mask_loc, u16 last_loc) { u64 bit = BIT_ULL(fld); seg->match |= bit; - if (type == ICE_FLOW_FLD_TYPE_RANGE) + if (field_type == ICE_FLOW_FLD_TYPE_RANGE) seg->range |= bit; - seg->fields[fld].type = type; + seg->fields[fld].type = field_type; seg->fields[fld].src.val = val_loc; seg->fields[fld].src.mask = mask_loc; seg->fields[fld].src.last = last_loc; diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h index 6db3d0494127..1d37a9f02c1c 100644 --- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h +++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h @@ -85,6 +85,7 @@ #define QRXFLXP_CNTXT_RXDID_IDX_M ICE_M(0x3F, 0) #define QRXFLXP_CNTXT_RXDID_PRIO_S 8 #define QRXFLXP_CNTXT_RXDID_PRIO_M ICE_M(0x7, 8) +#define QRXFLXP_CNTXT_TS_M BIT(11) #define GLGEN_RSTAT 0x000B8188 #define GLGEN_RSTAT_DEVSTATE_M ICE_M(0x3, 0) #define GLGEN_RSTCTL 0x000B8180 @@ -217,6 +218,8 @@ #define VPLAN_TX_QBASE_VFNUMQ_M ICE_M(0xFF, 16) #define VPLAN_TXQ_MAPENA(_VF) (0x00073800 + ((_VF) * 4)) #define VPLAN_TXQ_MAPENA_TX_ENA_M BIT(0) +#define GL_MDCK_TX_TDPU 0x00049348 +#define GL_MDCK_TX_TDPU_RCU_ANTISPOOF_ITR_DIS_M BIT(1) #define GL_MDET_RX 0x00294C00 #define GL_MDET_RX_QNUM_S 0 #define GL_MDET_RX_QNUM_M ICE_M(0x7FFF, 0) @@ -286,6 +289,8 @@ #define GL_PWR_MODE_CTL 0x000B820C #define GL_PWR_MODE_CTL_CAR_MAX_BW_S 30 #define GL_PWR_MODE_CTL_CAR_MAX_BW_M ICE_M(0x3, 30) +#define GLDCB_RTCTQ_RXQNUM_S 0 +#define GLDCB_RTCTQ_RXQNUM_M ICE_M(0x7FF, 0) #define GLPRT_BPRCL(_i) (0x00381380 + ((_i) * 8)) #define GLPRT_BPTCL(_i) (0x00381240 + ((_i) * 8)) #define GLPRT_CRCERRS(_i) (0x00380100 + ((_i) * 8)) diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index d974e2fa3e63..2f256bf45efc 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -9,11 +9,11 @@ /** * ice_vsi_type_str - maps VSI type enum to string equivalents - * @type: VSI type enum + * @vsi_type: VSI type enum */ -const char *ice_vsi_type_str(enum ice_vsi_type type) +const char *ice_vsi_type_str(enum ice_vsi_type vsi_type) { - switch (type) { + switch (vsi_type) { case ICE_VSI_PF: return "ICE_VSI_PF"; case ICE_VSI_VF: @@ -26,16 +26,26 @@ const char *ice_vsi_type_str(enum ice_vsi_type type) } /** - * ice_vsi_ctrl_rx_rings - Start or stop a VSI's Rx rings + * ice_vsi_ctrl_all_rx_rings - Start or stop a VSI's Rx rings * @vsi: the VSI being configured * @ena: start or stop the Rx rings + * + * First enable/disable all of the Rx rings, flush any remaining writes, and + * then verify that they have all been enabled/disabled successfully. This will + * let all of the register writes complete when enabling/disabling the Rx rings + * before waiting for the change in hardware to complete. */ -static int ice_vsi_ctrl_rx_rings(struct ice_vsi *vsi, bool ena) +static int ice_vsi_ctrl_all_rx_rings(struct ice_vsi *vsi, bool ena) { int i, ret = 0; + for (i = 0; i < vsi->num_rxq; i++) + ice_vsi_ctrl_one_rx_ring(vsi, ena, i, false); + + ice_flush(&vsi->back->hw); + for (i = 0; i < vsi->num_rxq; i++) { - ret = ice_vsi_ctrl_rx_ring(vsi, ena, i); + ret = ice_vsi_wait_one_rx_ring(vsi, ena, i); if (ret) break; } @@ -111,7 +121,6 @@ static void ice_vsi_set_num_desc(struct ice_vsi *vsi) { switch (vsi->type) { case ICE_VSI_PF: - /* fall through */ case ICE_VSI_LB: vsi->num_rx_desc = ICE_DFLT_NUM_RX_DESC; vsi->num_tx_desc = ICE_DFLT_NUM_TX_DESC; @@ -169,12 +178,12 @@ static void ice_vsi_set_num_qs(struct ice_vsi *vsi, u16 vf_id) vf = &pf->vf[vsi->vf_id]; vsi->alloc_txq = vf->num_vf_qs; vsi->alloc_rxq = vf->num_vf_qs; - /* pf->num_vf_msix includes (VF miscellaneous vector + + /* pf->num_msix_per_vf includes (VF miscellaneous vector + * data queue interrupts). Since vsi->num_q_vectors is number * of queues vectors, subtract 1 (ICE_NONQ_VECS_VF) from the * original vector count */ - vsi->num_q_vectors = pf->num_vf_msix - ICE_NONQ_VECS_VF; + vsi->num_q_vectors = pf->num_msix_per_vf - ICE_NONQ_VECS_VF; break; case ICE_VSI_LB: vsi->alloc_txq = 1; @@ -341,13 +350,13 @@ static irqreturn_t ice_msix_clean_rings(int __always_unused irq, void *data) /** * ice_vsi_alloc - Allocates the next available struct VSI in the PF * @pf: board private structure - * @type: type of VSI + * @vsi_type: type of VSI * @vf_id: ID of the VF being configured * * returns a pointer to a VSI on success, NULL on failure. */ static struct ice_vsi * -ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type, u16 vf_id) +ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type vsi_type, u16 vf_id) { struct device *dev = ice_pf_to_dev(pf); struct ice_vsi *vsi = NULL; @@ -368,13 +377,13 @@ ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type, u16 vf_id) if (!vsi) goto unlock_pf; - vsi->type = type; + vsi->type = vsi_type; vsi->back = pf; set_bit(__ICE_DOWN, vsi->state); vsi->idx = pf->next_vsi; - if (type == ICE_VSI_VF) + if (vsi_type == ICE_VSI_VF) ice_vsi_set_num_qs(vsi, vf_id); else ice_vsi_set_num_qs(vsi, ICE_INVAL_VFID); @@ -433,7 +442,7 @@ static int ice_vsi_get_qs(struct ice_vsi *vsi) .scatter_count = ICE_MAX_SCATTER_TXQS, .vsi_map = vsi->txq_map, .vsi_map_offset = 0, - .mapping_mode = vsi->tx_mapping_mode + .mapping_mode = ICE_VSI_MAP_CONTIG }; struct ice_qs_cfg rx_qs_cfg = { .qs_mutex = &pf->avail_q_mutex, @@ -443,18 +452,21 @@ static int ice_vsi_get_qs(struct ice_vsi *vsi) .scatter_count = ICE_MAX_SCATTER_RXQS, .vsi_map = vsi->rxq_map, .vsi_map_offset = 0, - .mapping_mode = vsi->rx_mapping_mode + .mapping_mode = ICE_VSI_MAP_CONTIG }; - int ret = 0; - - vsi->tx_mapping_mode = ICE_VSI_MAP_CONTIG; - vsi->rx_mapping_mode = ICE_VSI_MAP_CONTIG; + int ret; ret = __ice_vsi_get_qs(&tx_qs_cfg); - if (!ret) - ret = __ice_vsi_get_qs(&rx_qs_cfg); + if (ret) + return ret; + vsi->tx_mapping_mode = tx_qs_cfg.mapping_mode; - return ret; + ret = __ice_vsi_get_qs(&rx_qs_cfg); + if (ret) + return ret; + vsi->rx_mapping_mode = rx_qs_cfg.mapping_mode; + + return 0; } /** @@ -559,12 +571,11 @@ static void ice_vsi_set_rss_params(struct ice_vsi *vsi) vsi->rss_lut_type = ICE_AQC_GSET_RSS_LUT_TABLE_TYPE_PF; break; case ICE_VSI_VF: - /* VF VSI will gets a small RSS table - * For VSI_LUT, LUT size should be set to 64 bytes + /* VF VSI will get a small RSS table. + * For VSI_LUT, LUT size should be set to 64 bytes. */ vsi->rss_table_size = ICE_VSIQF_HLUT_ARRAY_SIZE; - vsi->rss_size = min_t(int, num_online_cpus(), - BIT(cap->rss_table_entry_width)); + vsi->rss_size = ICE_MAX_RSS_QS_PER_VF; vsi->rss_lut_type = ICE_AQC_GSET_RSS_LUT_TABLE_TYPE_VSI; break; case ICE_VSI_LB: @@ -672,7 +683,7 @@ static void ice_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt) if (vsi->type == ICE_VSI_PF) max_rss = ICE_MAX_LG_RSS_QS; else - max_rss = ICE_MAX_SMALL_RSS_QS; + max_rss = ICE_MAX_RSS_QS_PER_VF; qcount_rx = min_t(int, rx_numq_tc, max_rss); if (!vsi->req_rxq) qcount_rx = min_t(int, qcount_rx, @@ -804,7 +815,6 @@ static int ice_vsi_init(struct ice_vsi *vsi, bool init_vsi) ctxt->info = vsi->info; switch (vsi->type) { case ICE_VSI_LB: - /* fall through */ case ICE_VSI_PF: ctxt->flags = ICE_AQ_VSI_TYPE_PF; break; @@ -897,6 +907,109 @@ out: } /** + * ice_free_res - free a block of resources + * @res: pointer to the resource + * @index: starting index previously returned by ice_get_res + * @id: identifier to track owner + * + * Returns number of resources freed + */ +int ice_free_res(struct ice_res_tracker *res, u16 index, u16 id) +{ + int count = 0; + int i; + + if (!res || index >= res->end) + return -EINVAL; + + id |= ICE_RES_VALID_BIT; + for (i = index; i < res->end && res->list[i] == id; i++) { + res->list[i] = 0; + count++; + } + + return count; +} + +/** + * ice_search_res - Search the tracker for a block of resources + * @res: pointer to the resource + * @needed: size of the block needed + * @id: identifier to track owner + * + * Returns the base item index of the block, or -ENOMEM for error + */ +static int ice_search_res(struct ice_res_tracker *res, u16 needed, u16 id) +{ + int start = 0, end = 0; + + if (needed > res->end) + return -ENOMEM; + + id |= ICE_RES_VALID_BIT; + + do { + /* skip already allocated entries */ + if (res->list[end++] & ICE_RES_VALID_BIT) { + start = end; + if ((start + needed) > res->end) + break; + } + + if (end == (start + needed)) { + int i = start; + + /* there was enough, so assign it to the requestor */ + while (i != end) + res->list[i++] = id; + + return start; + } + } while (end < res->end); + + return -ENOMEM; +} + +/** + * ice_get_free_res_count - Get free count from a resource tracker + * @res: Resource tracker instance + */ +static u16 ice_get_free_res_count(struct ice_res_tracker *res) +{ + u16 i, count = 0; + + for (i = 0; i < res->end; i++) + if (!(res->list[i] & ICE_RES_VALID_BIT)) + count++; + + return count; +} + +/** + * ice_get_res - get a block of resources + * @pf: board private structure + * @res: pointer to the resource + * @needed: size of the block needed + * @id: identifier to track owner + * + * Returns the base item index of the block, or negative for error + */ +int +ice_get_res(struct ice_pf *pf, struct ice_res_tracker *res, u16 needed, u16 id) +{ + if (!res || !pf) + return -EINVAL; + + if (!needed || needed > res->num_entries || id >= ICE_RES_VALID_BIT) { + dev_err(ice_pf_to_dev(pf), "param err: needed=%d, num_entries = %d id=0x%04x\n", + needed, res->num_entries, id); + return -EINVAL; + } + + return ice_search_res(res, needed, id); +} + +/** * ice_vsi_setup_vector_base - Set up the base vector for the given VSI * @vsi: ptr to the VSI * @@ -928,8 +1041,9 @@ static int ice_vsi_setup_vector_base(struct ice_vsi *vsi) vsi->base_vector = ice_get_res(pf, pf->irq_tracker, num_q_vectors, vsi->idx); if (vsi->base_vector < 0) { - dev_err(dev, "Failed to get tracking for %d vectors for VSI %d, err=%d\n", - num_q_vectors, vsi->vsi_num, vsi->base_vector); + dev_err(dev, "%d MSI-X interrupts available. %s %d failed to get %d MSI-X vectors\n", + ice_get_free_res_count(pf->irq_tracker), + ice_vsi_type_str(vsi->type), vsi->idx, num_q_vectors); return -ENOENT; } pf->num_avail_sw_msix -= num_q_vectors; @@ -1348,7 +1462,9 @@ int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid) list_add(&tmp->list_entry, &tmp_add_list); status = ice_add_vlan(&pf->hw, &tmp_add_list); - if (status) { + if (!status) { + vsi->num_vlan++; + } else { err = -ENODEV; dev_err(dev, "Failure Adding VLAN %d on VSI %i\n", vid, vsi->vsi_num); @@ -1390,10 +1506,12 @@ int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid) list_add(&list->list_entry, &tmp_add_list); status = ice_remove_vlan(&pf->hw, &tmp_add_list); - if (status == ICE_ERR_DOES_NOT_EXIST) { + if (!status) { + vsi->num_vlan--; + } else if (status == ICE_ERR_DOES_NOT_EXIST) { dev_dbg(dev, "Failed to remove VLAN %d on VSI %i, it does not exist, status: %d\n", vid, vsi->vsi_num, status); - } else if (status) { + } else { dev_err(dev, "Error removing VLAN %d on vsi %i error: %d\n", vid, vsi->vsi_num, status); err = -EIO; @@ -1678,25 +1796,25 @@ out: } /** - * ice_vsi_start_rx_rings - start VSI's Rx rings - * @vsi: the VSI whose rings are to be started + * ice_vsi_start_all_rx_rings - start/enable all of a VSI's Rx rings + * @vsi: the VSI whose rings are to be enabled * * Returns 0 on success and a negative value on error */ -int ice_vsi_start_rx_rings(struct ice_vsi *vsi) +int ice_vsi_start_all_rx_rings(struct ice_vsi *vsi) { - return ice_vsi_ctrl_rx_rings(vsi, true); + return ice_vsi_ctrl_all_rx_rings(vsi, true); } /** - * ice_vsi_stop_rx_rings - stop VSI's Rx rings - * @vsi: the VSI + * ice_vsi_stop_all_rx_rings - stop/disable all of a VSI's Rx rings + * @vsi: the VSI whose rings are to be disabled * * Returns 0 on success and a negative value on error */ -int ice_vsi_stop_rx_rings(struct ice_vsi *vsi) +int ice_vsi_stop_all_rx_rings(struct ice_vsi *vsi) { - return ice_vsi_ctrl_rx_rings(vsi, false); + return ice_vsi_ctrl_all_rx_rings(vsi, false); } /** @@ -1756,6 +1874,20 @@ int ice_vsi_stop_xdp_tx_rings(struct ice_vsi *vsi) } /** + * ice_vsi_is_vlan_pruning_ena - check if VLAN pruning is enabled or not + * @vsi: VSI to check whether or not VLAN pruning is enabled. + * + * returns true if Rx VLAN pruning is enabled and false otherwise. + */ +bool ice_vsi_is_vlan_pruning_ena(struct ice_vsi *vsi) +{ + if (!vsi) + return false; + + return (vsi->info.sw_flags2 & ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA); +} + +/** * ice_cfg_vlan_pruning - enable or disable VLAN pruning on the VSI * @vsi: VSI to enable or disable VLAN pruning on * @ena: set to true to enable VLAN pruning and false to disable it @@ -1952,7 +2084,7 @@ void ice_cfg_sw_lldp(struct ice_vsi *vsi, bool tx, bool create) * ice_vsi_setup - Set up a VSI by a given type * @pf: board private structure * @pi: pointer to the port_info instance - * @type: VSI type + * @vsi_type: VSI type * @vf_id: defines VF ID to which this VSI connects. This field is meant to be * used only for ICE_VSI_VF VSI type. For other VSI types, should * fill-in ICE_INVAL_VFID as input. @@ -1964,7 +2096,7 @@ void ice_cfg_sw_lldp(struct ice_vsi *vsi, bool tx, bool create) */ struct ice_vsi * ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, - enum ice_vsi_type type, u16 vf_id) + enum ice_vsi_type vsi_type, u16 vf_id) { u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 }; struct device *dev = ice_pf_to_dev(pf); @@ -1972,10 +2104,10 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, struct ice_vsi *vsi; int ret, i; - if (type == ICE_VSI_VF) - vsi = ice_vsi_alloc(pf, type, vf_id); + if (vsi_type == ICE_VSI_VF) + vsi = ice_vsi_alloc(pf, vsi_type, vf_id); else - vsi = ice_vsi_alloc(pf, type, ICE_INVAL_VFID); + vsi = ice_vsi_alloc(pf, vsi_type, ICE_INVAL_VFID); if (!vsi) { dev_err(dev, "could not allocate VSI\n"); @@ -2025,6 +2157,17 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, if (ret) goto unroll_vector_base; + /* Always add VLAN ID 0 switch rule by default. This is needed + * in order to allow all untagged and 0 tagged priority traffic + * if Rx VLAN pruning is enabled. Also there are cases where we + * don't get the call to add VLAN 0 via ice_vlan_rx_add_vid() + * so this handles those cases (i.e. adding the PF to a bridge + * without the 8021q module loaded). + */ + ret = ice_vsi_add_vlan(vsi, 0); + if (ret) + goto unroll_clear_rings; + ice_vsi_map_rings_to_vectors(vsi); /* Do not exit if configuring RSS had an issue, at least @@ -2104,6 +2247,8 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, return vsi; +unroll_clear_rings: + ice_vsi_clear_rings(vsi); unroll_vector_base: /* reclaim SW interrupts back to the common pool */ ice_free_res(pf->irq_tracker, vsi->base_vector, vsi->idx); @@ -2299,94 +2444,6 @@ void ice_dis_vsi(struct ice_vsi *vsi, bool locked) } /** - * ice_free_res - free a block of resources - * @res: pointer to the resource - * @index: starting index previously returned by ice_get_res - * @id: identifier to track owner - * - * Returns number of resources freed - */ -int ice_free_res(struct ice_res_tracker *res, u16 index, u16 id) -{ - int count = 0; - int i; - - if (!res || index >= res->end) - return -EINVAL; - - id |= ICE_RES_VALID_BIT; - for (i = index; i < res->end && res->list[i] == id; i++) { - res->list[i] = 0; - count++; - } - - return count; -} - -/** - * ice_search_res - Search the tracker for a block of resources - * @res: pointer to the resource - * @needed: size of the block needed - * @id: identifier to track owner - * - * Returns the base item index of the block, or -ENOMEM for error - */ -static int ice_search_res(struct ice_res_tracker *res, u16 needed, u16 id) -{ - int start = 0, end = 0; - - if (needed > res->end) - return -ENOMEM; - - id |= ICE_RES_VALID_BIT; - - do { - /* skip already allocated entries */ - if (res->list[end++] & ICE_RES_VALID_BIT) { - start = end; - if ((start + needed) > res->end) - break; - } - - if (end == (start + needed)) { - int i = start; - - /* there was enough, so assign it to the requestor */ - while (i != end) - res->list[i++] = id; - - return start; - } - } while (end < res->end); - - return -ENOMEM; -} - -/** - * ice_get_res - get a block of resources - * @pf: board private structure - * @res: pointer to the resource - * @needed: size of the block needed - * @id: identifier to track owner - * - * Returns the base item index of the block, or negative for error - */ -int -ice_get_res(struct ice_pf *pf, struct ice_res_tracker *res, u16 needed, u16 id) -{ - if (!res || !pf) - return -EINVAL; - - if (!needed || needed > res->num_entries || id >= ICE_RES_VALID_BIT) { - dev_err(ice_pf_to_dev(pf), "param err: needed=%d, num_entries = %d id=0x%04x\n", - needed, res->num_entries, id); - return -EINVAL; - } - - return ice_search_res(res, needed, id); -} - -/** * ice_vsi_dis_irq - Mask off queue interrupt generation on the VSI * @vsi: the VSI being un-configured */ diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h index e2c0dadce920..04ca00799364 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_lib.h @@ -6,7 +6,7 @@ #include "ice.h" -const char *ice_vsi_type_str(enum ice_vsi_type type); +const char *ice_vsi_type_str(enum ice_vsi_type vsi_type); int ice_add_mac_to_list(struct ice_vsi *vsi, struct list_head *add_list, @@ -30,9 +30,9 @@ int ice_vsi_manage_vlan_insertion(struct ice_vsi *vsi); int ice_vsi_manage_vlan_stripping(struct ice_vsi *vsi, bool ena); -int ice_vsi_start_rx_rings(struct ice_vsi *vsi); +int ice_vsi_start_all_rx_rings(struct ice_vsi *vsi); -int ice_vsi_stop_rx_rings(struct ice_vsi *vsi); +int ice_vsi_stop_all_rx_rings(struct ice_vsi *vsi); int ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, @@ -42,6 +42,8 @@ int ice_vsi_cfg_xdp_txqs(struct ice_vsi *vsi); int ice_vsi_stop_xdp_tx_rings(struct ice_vsi *vsi); +bool ice_vsi_is_vlan_pruning_ena(struct ice_vsi *vsi); + int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc); void ice_cfg_sw_lldp(struct ice_vsi *vsi, bool tx, bool create); @@ -56,7 +58,7 @@ int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc); struct ice_vsi * ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, - enum ice_vsi_type type, u16 vf_id); + enum ice_vsi_type vsi_type, u16 vf_id); void ice_napi_del(struct ice_vsi *vsi); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 5ef28052c0f8..306a4e5b2320 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -10,6 +10,7 @@ #include "ice_lib.h" #include "ice_dcb_lib.h" #include "ice_dcb_nl.h" +#include "ice_devlink.h" #define DRV_VERSION_MAJOR 0 #define DRV_VERSION_MINOR 8 @@ -706,7 +707,6 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) /* Get FEC mode based on negotiated link info */ switch (vsi->port_info->phy.link_info.fec_info) { case ICE_AQ_LINK_25G_RS_528_FEC_EN: - /* fall through */ case ICE_AQ_LINK_25G_RS_544_FEC_EN: fec = "RS-FEC"; break; @@ -1029,6 +1029,9 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type) if (ice_handle_link_event(pf, &event)) dev_err(dev, "Could not handle link event\n"); break; + case ice_aqc_opc_event_lan_overflow: + ice_vf_lan_overflow_event(pf, &event); + break; case ice_mbx_opc_send_msg_to_pf: ice_vc_process_vf_msg(pf, &event); break; @@ -1185,20 +1188,28 @@ static void ice_service_timer(struct timer_list *t) * ice_handle_mdd_event - handle malicious driver detect event * @pf: pointer to the PF structure * - * Called from service task. OICR interrupt handler indicates MDD event + * Called from service task. OICR interrupt handler indicates MDD event. + * VF MDD logging is guarded by net_ratelimit. Additional PF and VF log + * messages are wrapped by netif_msg_[rx|tx]_err. Since VF Rx MDD events + * disable the queue, the PF can be configured to reset the VF using ethtool + * private flag mdd-auto-reset-vf. */ static void ice_handle_mdd_event(struct ice_pf *pf) { struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; - bool mdd_detected = false; u32 reg; int i; - if (!test_and_clear_bit(__ICE_MDD_EVENT_PENDING, pf->state)) + if (!test_and_clear_bit(__ICE_MDD_EVENT_PENDING, pf->state)) { + /* Since the VF MDD event logging is rate limited, check if + * there are pending MDD events. + */ + ice_print_vfs_mdd_events(pf); return; + } - /* find what triggered the MDD event */ + /* find what triggered an MDD event */ reg = rd32(hw, GL_MDET_TX_PQM); if (reg & GL_MDET_TX_PQM_VALID_M) { u8 pf_num = (reg & GL_MDET_TX_PQM_PF_NUM_M) >> @@ -1214,7 +1225,6 @@ static void ice_handle_mdd_event(struct ice_pf *pf) dev_info(dev, "Malicious Driver Detection event %d on TX queue %d PF# %d VF# %d\n", event, queue, pf_num, vf_num); wr32(hw, GL_MDET_TX_PQM, 0xffffffff); - mdd_detected = true; } reg = rd32(hw, GL_MDET_TX_TCLAN); @@ -1232,7 +1242,6 @@ static void ice_handle_mdd_event(struct ice_pf *pf) dev_info(dev, "Malicious Driver Detection event %d on TX queue %d PF# %d VF# %d\n", event, queue, pf_num, vf_num); wr32(hw, GL_MDET_TX_TCLAN, 0xffffffff); - mdd_detected = true; } reg = rd32(hw, GL_MDET_RX); @@ -1250,85 +1259,85 @@ static void ice_handle_mdd_event(struct ice_pf *pf) dev_info(dev, "Malicious Driver Detection event %d on RX queue %d PF# %d VF# %d\n", event, queue, pf_num, vf_num); wr32(hw, GL_MDET_RX, 0xffffffff); - mdd_detected = true; } - if (mdd_detected) { - bool pf_mdd_detected = false; - - reg = rd32(hw, PF_MDET_TX_PQM); - if (reg & PF_MDET_TX_PQM_VALID_M) { - wr32(hw, PF_MDET_TX_PQM, 0xFFFF); - dev_info(dev, "TX driver issue detected, PF reset issued\n"); - pf_mdd_detected = true; - } + /* check to see if this PF caused an MDD event */ + reg = rd32(hw, PF_MDET_TX_PQM); + if (reg & PF_MDET_TX_PQM_VALID_M) { + wr32(hw, PF_MDET_TX_PQM, 0xFFFF); + if (netif_msg_tx_err(pf)) + dev_info(dev, "Malicious Driver Detection event TX_PQM detected on PF\n"); + } - reg = rd32(hw, PF_MDET_TX_TCLAN); - if (reg & PF_MDET_TX_TCLAN_VALID_M) { - wr32(hw, PF_MDET_TX_TCLAN, 0xFFFF); - dev_info(dev, "TX driver issue detected, PF reset issued\n"); - pf_mdd_detected = true; - } + reg = rd32(hw, PF_MDET_TX_TCLAN); + if (reg & PF_MDET_TX_TCLAN_VALID_M) { + wr32(hw, PF_MDET_TX_TCLAN, 0xFFFF); + if (netif_msg_tx_err(pf)) + dev_info(dev, "Malicious Driver Detection event TX_TCLAN detected on PF\n"); + } - reg = rd32(hw, PF_MDET_RX); - if (reg & PF_MDET_RX_VALID_M) { - wr32(hw, PF_MDET_RX, 0xFFFF); - dev_info(dev, "RX driver issue detected, PF reset issued\n"); - pf_mdd_detected = true; - } - /* Queue belongs to the PF initiate a reset */ - if (pf_mdd_detected) { - set_bit(__ICE_NEEDS_RESTART, pf->state); - ice_service_task_schedule(pf); - } + reg = rd32(hw, PF_MDET_RX); + if (reg & PF_MDET_RX_VALID_M) { + wr32(hw, PF_MDET_RX, 0xFFFF); + if (netif_msg_rx_err(pf)) + dev_info(dev, "Malicious Driver Detection event RX detected on PF\n"); } - /* check to see if one of the VFs caused the MDD */ + /* Check to see if one of the VFs caused an MDD event, and then + * increment counters and set print pending + */ ice_for_each_vf(pf, i) { struct ice_vf *vf = &pf->vf[i]; - bool vf_mdd_detected = false; - reg = rd32(hw, VP_MDET_TX_PQM(i)); if (reg & VP_MDET_TX_PQM_VALID_M) { wr32(hw, VP_MDET_TX_PQM(i), 0xFFFF); - vf_mdd_detected = true; - dev_info(dev, "TX driver issue detected on VF %d\n", - i); + vf->mdd_tx_events.count++; + set_bit(__ICE_MDD_VF_PRINT_PENDING, pf->state); + if (netif_msg_tx_err(pf)) + dev_info(dev, "Malicious Driver Detection event TX_PQM detected on VF %d\n", + i); } reg = rd32(hw, VP_MDET_TX_TCLAN(i)); if (reg & VP_MDET_TX_TCLAN_VALID_M) { wr32(hw, VP_MDET_TX_TCLAN(i), 0xFFFF); - vf_mdd_detected = true; - dev_info(dev, "TX driver issue detected on VF %d\n", - i); + vf->mdd_tx_events.count++; + set_bit(__ICE_MDD_VF_PRINT_PENDING, pf->state); + if (netif_msg_tx_err(pf)) + dev_info(dev, "Malicious Driver Detection event TX_TCLAN detected on VF %d\n", + i); } reg = rd32(hw, VP_MDET_TX_TDPU(i)); if (reg & VP_MDET_TX_TDPU_VALID_M) { wr32(hw, VP_MDET_TX_TDPU(i), 0xFFFF); - vf_mdd_detected = true; - dev_info(dev, "TX driver issue detected on VF %d\n", - i); + vf->mdd_tx_events.count++; + set_bit(__ICE_MDD_VF_PRINT_PENDING, pf->state); + if (netif_msg_tx_err(pf)) + dev_info(dev, "Malicious Driver Detection event TX_TDPU detected on VF %d\n", + i); } reg = rd32(hw, VP_MDET_RX(i)); if (reg & VP_MDET_RX_VALID_M) { wr32(hw, VP_MDET_RX(i), 0xFFFF); - vf_mdd_detected = true; - dev_info(dev, "RX driver issue detected on VF %d\n", - i); - } - - if (vf_mdd_detected) { - vf->num_mdd_events++; - if (vf->num_mdd_events && - vf->num_mdd_events <= ICE_MDD_EVENTS_THRESHOLD) - dev_info(dev, "VF %d has had %llu MDD events since last boot, Admin might need to reload AVF driver with this number of events\n", - i, vf->num_mdd_events); + vf->mdd_rx_events.count++; + set_bit(__ICE_MDD_VF_PRINT_PENDING, pf->state); + if (netif_msg_rx_err(pf)) + dev_info(dev, "Malicious Driver Detection event RX detected on VF %d\n", + i); + + /* Since the queue is disabled on VF Rx MDD events, the + * PF can be configured to reset the VF through ethtool + * private flag mdd-auto-reset-vf. + */ + if (test_bit(ICE_FLAG_MDD_AUTO_RESET_VF, pf->flags)) + ice_reset_vf(&pf->vf[i], false); } } + + ice_print_vfs_mdd_events(pf); } /** @@ -1510,7 +1519,7 @@ static void ice_set_ctrlq_len(struct ice_hw *hw) hw->adminq.num_sq_entries = ICE_AQ_LEN; hw->adminq.rq_buf_size = ICE_AQ_MAX_BUF_LEN; hw->adminq.sq_buf_size = ICE_AQ_MAX_BUF_LEN; - hw->mailboxq.num_rq_entries = ICE_MBXRQ_LEN; + hw->mailboxq.num_rq_entries = PF_MBX_ARQLEN_ARQLEN_M; hw->mailboxq.num_sq_entries = ICE_MBXSQ_LEN; hw->mailboxq.rq_buf_size = ICE_MBXQ_MAX_BUF_LEN; hw->mailboxq.sq_buf_size = ICE_MBXQ_MAX_BUF_LEN; @@ -1916,8 +1925,7 @@ ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, if (if_running && !test_and_set_bit(__ICE_DOWN, vsi->state)) { ret = ice_down(vsi); if (ret) { - NL_SET_ERR_MSG_MOD(extack, - "Preparing device for XDP attach failed"); + NL_SET_ERR_MSG_MOD(extack, "Preparing device for XDP attach failed"); return ret; } } @@ -1926,13 +1934,11 @@ ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, vsi->num_xdp_txq = vsi->alloc_txq; xdp_ring_err = ice_prepare_xdp_rings(vsi, prog); if (xdp_ring_err) - NL_SET_ERR_MSG_MOD(extack, - "Setting up XDP Tx resources failed"); + NL_SET_ERR_MSG_MOD(extack, "Setting up XDP Tx resources failed"); } else if (ice_is_xdp_ena_vsi(vsi) && !prog) { xdp_ring_err = ice_destroy_xdp_rings(vsi); if (xdp_ring_err) - NL_SET_ERR_MSG_MOD(extack, - "Freeing XDP Tx resources failed"); + NL_SET_ERR_MSG_MOD(extack, "Freeing XDP Tx resources failed"); } else { ice_vsi_assign_bpf_prog(vsi, prog); } @@ -1965,8 +1971,7 @@ static int ice_xdp(struct net_device *dev, struct netdev_bpf *xdp) struct ice_vsi *vsi = np->vsi; if (vsi->type != ICE_VSI_PF) { - NL_SET_ERR_MSG_MOD(xdp->extack, - "XDP can be loaded only on PF VSI"); + NL_SET_ERR_MSG_MOD(xdp->extack, "XDP can be loaded only on PF VSI"); return -EINVAL; } @@ -1993,6 +1998,14 @@ static void ice_ena_misc_vector(struct ice_pf *pf) struct ice_hw *hw = &pf->hw; u32 val; + /* Disable anti-spoof detection interrupt to prevent spurious event + * interrupts during a function reset. Anti-spoof functionally is + * still supported. + */ + val = rd32(hw, GL_MDCK_TX_TDPU); + val |= GL_MDCK_TX_TDPU_RCU_ANTISPOOF_ITR_DIS_M; + wr32(hw, GL_MDCK_TX_TDPU, val); + /* clear things first */ wr32(hw, PFINT_OICR_ENA, 0); /* disable all */ rd32(hw, PFINT_OICR); /* read to clear */ @@ -2042,8 +2055,16 @@ static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) set_bit(__ICE_MDD_EVENT_PENDING, pf->state); } if (oicr & PFINT_OICR_VFLR_M) { - ena_mask &= ~PFINT_OICR_VFLR_M; - set_bit(__ICE_VFLR_EVENT_PENDING, pf->state); + /* disable any further VFLR event notifications */ + if (test_bit(__ICE_VF_RESETS_DISABLED, pf->state)) { + u32 reg = rd32(hw, PFINT_OICR_ENA); + + reg &= ~PFINT_OICR_VFLR_M; + wr32(hw, PFINT_OICR_ENA, reg); + } else { + ena_mask &= ~PFINT_OICR_VFLR_M; + set_bit(__ICE_VFLR_EVENT_PENDING, pf->state); + } } if (oicr & PFINT_OICR_GRST_M) { @@ -2351,10 +2372,16 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) u8 mac_addr[ETH_ALEN]; int err; + err = ice_devlink_create_port(pf); + if (err) + return err; + netdev = alloc_etherdev_mqs(sizeof(*np), vsi->alloc_txq, vsi->alloc_rxq); - if (!netdev) - return -ENOMEM; + if (!netdev) { + err = -ENOMEM; + goto err_destroy_devlink_port; + } vsi->netdev = netdev; np = netdev_priv(netdev); @@ -2384,7 +2411,9 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) err = register_netdev(vsi->netdev); if (err) - return err; + goto err_destroy_devlink_port; + + devlink_port_type_eth_set(&pf->devlink_port, vsi->netdev); netif_carrier_off(vsi->netdev); @@ -2392,6 +2421,11 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) netif_tx_stop_all_queues(vsi->netdev); return 0; + +err_destroy_devlink_port: + ice_devlink_destroy_port(pf); + + return err; } /** @@ -2461,16 +2495,19 @@ ice_vlan_rx_add_vid(struct net_device *netdev, __always_unused __be16 proto, if (vsi->info.pvid) return -EINVAL; - /* Enable VLAN pruning when VLAN 0 is added */ - if (unlikely(!vid)) { + /* VLAN 0 is added by default during load/reset */ + if (!vid) + return 0; + + /* Enable VLAN pruning when a VLAN other than 0 is added */ + if (!ice_vsi_is_vlan_pruning_ena(vsi)) { ret = ice_cfg_vlan_pruning(vsi, true, false); if (ret) return ret; } - /* Add all VLAN IDs including 0 to the switch filter. VLAN ID 0 is - * needed to continue allowing all untagged packets since VLAN prune - * list is applied to all packets by the switch + /* Add a switch rule for this VLAN ID so its corresponding VLAN tagged + * packets aren't pruned by the device's internal switch on Rx */ ret = ice_vsi_add_vlan(vsi, vid); if (!ret) { @@ -2500,6 +2537,10 @@ ice_vlan_rx_kill_vid(struct net_device *netdev, __always_unused __be16 proto, if (vsi->info.pvid) return -EINVAL; + /* don't allow removal of VLAN 0 */ + if (!vid) + return 0; + /* Make sure ice_vsi_kill_vlan is successful before updating VLAN * information */ @@ -2507,8 +2548,8 @@ ice_vlan_rx_kill_vid(struct net_device *netdev, __always_unused __be16 proto, if (ret) return ret; - /* Disable VLAN pruning when VLAN 0 is removed */ - if (unlikely(!vid)) + /* Disable pruning when VLAN 0 is the only VLAN rule */ + if (vsi->num_vlan == 1 && ice_vsi_is_vlan_pruning_ena(vsi)) ret = ice_cfg_vlan_pruning(vsi, false, false); vsi->vlan_ena = false; @@ -2945,7 +2986,6 @@ ice_log_pkg_init(struct ice_hw *hw, enum ice_status *status) } break; case ICE_ERR_BUF_TOO_SHORT: - /* fall-through */ case ICE_ERR_CFG: dev_err(dev, "The DDP package file is invalid. Entering Safe Mode.\n"); break; @@ -2977,7 +3017,7 @@ ice_log_pkg_init(struct ice_hw *hw, enum ice_status *status) default: break; } - /* fall-through */ + fallthrough; default: dev_err(dev, "An unknown error (%d) occurred when loading the DDP package. Entering Safe Mode.\n", *status); @@ -3069,30 +3109,22 @@ static char *ice_get_opt_fw_name(struct ice_pf *pf) * followed by a EUI-64 identifier (PCIe Device Serial Number) */ struct pci_dev *pdev = pf->pdev; - char *opt_fw_filename = NULL; - u32 dword; - u8 dsn[8]; - int pos; + char *opt_fw_filename; + u64 dsn; /* Determine the name of the optional file using the DSN (two * dwords following the start of the DSN Capability). */ - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_DSN); - if (pos) { - opt_fw_filename = kzalloc(NAME_MAX, GFP_KERNEL); - if (!opt_fw_filename) - return NULL; - - pci_read_config_dword(pdev, pos + 4, &dword); - put_unaligned_le32(dword, &dsn[0]); - pci_read_config_dword(pdev, pos + 8, &dword); - put_unaligned_le32(dword, &dsn[4]); - snprintf(opt_fw_filename, NAME_MAX, - "%sice-%02x%02x%02x%02x%02x%02x%02x%02x.pkg", - ICE_DDP_PKG_PATH, - dsn[7], dsn[6], dsn[5], dsn[4], - dsn[3], dsn[2], dsn[1], dsn[0]); - } + dsn = pci_get_dsn(pdev); + if (!dsn) + return NULL; + + opt_fw_filename = kzalloc(NAME_MAX, GFP_KERNEL); + if (!opt_fw_filename) + return NULL; + + snprintf(opt_fw_filename, NAME_MAX, "%sice-%016llX.pkg", + ICE_DDP_PKG_PATH, dsn); return opt_fw_filename; } @@ -3166,7 +3198,7 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) return err; } - pf = devm_kzalloc(dev, sizeof(*pf), GFP_KERNEL); + pf = ice_allocate_pf(dev); if (!pf) return -ENOMEM; @@ -3204,6 +3236,12 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) pf->msg_enable = netif_msg_init(debug, ICE_DFLT_NETIF_M); + err = ice_devlink_register(pf); + if (err) { + dev_err(dev, "ice_devlink_register failed: %d\n", err); + goto err_exit_unroll; + } + #ifndef CONFIG_DYNAMIC_DEBUG if (debug < -1) hw->debug_mask = debug; @@ -3238,6 +3276,8 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) goto err_init_pf_unroll; } + ice_devlink_init_regions(pf); + pf->num_alloc_vsi = hw->func_caps.guar_num_vsi; if (!pf->num_alloc_vsi) { err = -EIO; @@ -3336,6 +3376,7 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) return 0; err_alloc_sw_unroll: + ice_devlink_destroy_port(pf); set_bit(__ICE_SERVICE_DIS, pf->state); set_bit(__ICE_DOWN, pf->state); devm_kfree(dev, pf->first_sw); @@ -3346,8 +3387,10 @@ err_init_interrupt_unroll: devm_kfree(dev, pf->vsi); err_init_pf_unroll: ice_deinit_pf(pf); + ice_devlink_destroy_regions(pf); ice_deinit_hw(hw); err_exit_unroll: + ice_devlink_unregister(pf); pci_disable_pcie_error_reporting(pdev); return err; } @@ -3370,11 +3413,15 @@ static void ice_remove(struct pci_dev *pdev) msleep(100); } + if (test_bit(ICE_FLAG_SRIOV_ENA, pf->flags)) { + set_bit(__ICE_VF_RESETS_DISABLED, pf->state); + ice_free_vfs(pf); + } + set_bit(__ICE_DOWN, pf->state); ice_service_task_stop(pf); - if (test_bit(ICE_FLAG_SRIOV_ENA, pf->flags)) - ice_free_vfs(pf); + ice_devlink_destroy_port(pf); ice_vsi_release_all(pf); ice_free_irq_msix_misc(pf); ice_for_each_vsi(pf, i) { @@ -3383,7 +3430,10 @@ static void ice_remove(struct pci_dev *pdev) ice_vsi_free_q_vectors(pf->vsi[i]); } ice_deinit_pf(pf); + ice_devlink_destroy_regions(pf); ice_deinit_hw(&pf->hw); + ice_devlink_unregister(pf); + /* Issue a PFR as part of the prescribed driver unload flow. Do not * do it via ice_schedule_reset() since there is no need to rebuild * and the service task is already stopped. @@ -3534,15 +3584,26 @@ static const struct pci_device_id ice_pci_tbl[] = { { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_BACKPLANE), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_QSFP), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_SFP), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810_XXV_SFP), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_BACKPLANE), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_QSFP), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_SFP), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_10G_BASE_T), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_SGMII), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_BACKPLANE), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_QSFP), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_SFP), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_10G_BASE_T), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_SGMII), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822X_BACKPLANE), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_BACKPLANE), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_SFP), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_10G_BASE_T), 0 }, { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_SGMII), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_BACKPLANE), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_SFP), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_10G_BASE_T), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_1GBE), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_QSFP), 0 }, /* required last entry */ { 0, } }; @@ -3961,7 +4022,7 @@ static int ice_up_complete(struct ice_vsi *vsi) * Tx queue group list was configured and the context bits were * programmed using ice_vsi_cfg_txqs */ - err = ice_vsi_start_rx_rings(vsi); + err = ice_vsi_start_all_rx_rings(vsi); if (err) return err; @@ -4340,7 +4401,7 @@ int ice_down(struct ice_vsi *vsi) vsi->vsi_num, tx_err); } - rx_err = ice_vsi_stop_rx_rings(vsi); + rx_err = ice_vsi_stop_all_rx_rings(vsi); if (rx_err) netdev_err(vsi->netdev, "Failed stop Rx rings, VSI %d error %d\n", vsi->vsi_num, rx_err); @@ -5027,6 +5088,7 @@ ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, /** * ice_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: Tx queue */ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) { @@ -5064,13 +5126,13 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) /* Read interrupt register */ val = rd32(hw, GLINT_DYN_CTL(tx_ring->q_vector->reg_idx)); - netdev_info(netdev, "tx_timeout: VSI_num: %d, Q %d, NTC: 0x%x, HW_HEAD: 0x%x, NTU: 0x%x, INT: 0x%x\n", + netdev_info(netdev, "tx_timeout: VSI_num: %d, Q %u, NTC: 0x%x, HW_HEAD: 0x%x, NTU: 0x%x, INT: 0x%x\n", vsi->vsi_num, txqueue, tx_ring->next_to_clean, head, tx_ring->next_to_use, val); } pf->tx_timeout_last_recovery = jiffies; - netdev_info(netdev, "tx_timeout recovery level %d, txqueue %d\n", + netdev_info(netdev, "tx_timeout recovery level %d, txqueue %u\n", pf->tx_timeout_recovery_level, txqueue); switch (pf->tx_timeout_recovery_level) { diff --git a/drivers/net/ethernet/intel/ice/ice_nvm.c b/drivers/net/ethernet/intel/ice/ice_nvm.c index 7525ac50742e..8beb675d676b 100644 --- a/drivers/net/ethernet/intel/ice/ice_nvm.c +++ b/drivers/net/ethernet/intel/ice/ice_nvm.c @@ -11,25 +11,29 @@ * @length: length of the section to be read (in bytes from the offset) * @data: command buffer (size [bytes] = length) * @last_command: tells if this is the last command in a series + * @read_shadow_ram: tell if this is a shadow RAM read * @cd: pointer to command details structure or NULL * * Read the NVM using the admin queue commands (0x0701) */ static enum ice_status ice_aq_read_nvm(struct ice_hw *hw, u16 module_typeid, u32 offset, u16 length, - void *data, bool last_command, struct ice_sq_cd *cd) + void *data, bool last_command, bool read_shadow_ram, + struct ice_sq_cd *cd) { struct ice_aq_desc desc; struct ice_aqc_nvm *cmd; cmd = &desc.params.nvm; - /* In offset the highest byte must be zeroed. */ - if (offset & 0xFF000000) + if (offset > ICE_AQC_NVM_MAX_OFFSET) return ICE_ERR_PARAM; ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_nvm_read); + if (!read_shadow_ram && module_typeid == ICE_AQC_NVM_START_POINT) + cmd->cmd_flags |= ICE_AQC_NVM_FLASH_ONLY; + /* If this is the last command in a series, set the proper flag. */ if (last_command) cmd->cmd_flags |= ICE_AQC_NVM_LAST_CMD; @@ -42,65 +46,64 @@ ice_aq_read_nvm(struct ice_hw *hw, u16 module_typeid, u32 offset, u16 length, } /** - * ice_check_sr_access_params - verify params for Shadow RAM R/W operations. - * @hw: pointer to the HW structure - * @offset: offset in words from module start - * @words: number of words to access + * ice_read_flat_nvm - Read portion of NVM by flat offset + * @hw: pointer to the HW struct + * @offset: offset from beginning of NVM + * @length: (in) number of bytes to read; (out) number of bytes actually read + * @data: buffer to return data in (sized to fit the specified length) + * @read_shadow_ram: if true, read from shadow RAM instead of NVM + * + * Reads a portion of the NVM, as a flat memory space. This function correctly + * breaks read requests across Shadow RAM sectors and ensures that no single + * read request exceeds the maximum 4Kb read for a single AdminQ command. + * + * Returns a status code on failure. Note that the data pointer may be + * partially updated if some reads succeed before a failure. */ -static enum ice_status -ice_check_sr_access_params(struct ice_hw *hw, u32 offset, u16 words) +enum ice_status +ice_read_flat_nvm(struct ice_hw *hw, u32 offset, u32 *length, u8 *data, + bool read_shadow_ram) { - if ((offset + words) > hw->nvm.sr_words) { - ice_debug(hw, ICE_DBG_NVM, - "NVM error: offset beyond SR lmt.\n"); - return ICE_ERR_PARAM; - } + enum ice_status status; + u32 inlen = *length; + u32 bytes_read = 0; + bool last_cmd; - if (words > ICE_SR_SECTOR_SIZE_IN_WORDS) { - /* We can access only up to 4KB (one sector), in one AQ write */ - ice_debug(hw, ICE_DBG_NVM, - "NVM error: tried to access %d words, limit is %d.\n", - words, ICE_SR_SECTOR_SIZE_IN_WORDS); - return ICE_ERR_PARAM; - } + *length = 0; - if (((offset + (words - 1)) / ICE_SR_SECTOR_SIZE_IN_WORDS) != - (offset / ICE_SR_SECTOR_SIZE_IN_WORDS)) { - /* A single access cannot spread over two sectors */ + /* Verify the length of the read if this is for the Shadow RAM */ + if (read_shadow_ram && ((offset + inlen) > (hw->nvm.sr_words * 2u))) { ice_debug(hw, ICE_DBG_NVM, - "NVM error: cannot spread over two sectors.\n"); + "NVM error: requested offset is beyond Shadow RAM limit\n"); return ICE_ERR_PARAM; } - return 0; -} + do { + u32 read_size, sector_offset; -/** - * ice_read_sr_aq - Read Shadow RAM. - * @hw: pointer to the HW structure - * @offset: offset in words from module start - * @words: number of words to read - * @data: buffer for words reads from Shadow RAM - * @last_command: tells the AdminQ that this is the last command - * - * Reads 16-bit word buffers from the Shadow RAM using the admin command. - */ -static enum ice_status -ice_read_sr_aq(struct ice_hw *hw, u32 offset, u16 words, u16 *data, - bool last_command) -{ - enum ice_status status; + /* ice_aq_read_nvm cannot read more than 4Kb at a time. + * Additionally, a read from the Shadow RAM may not cross over + * a sector boundary. Conveniently, the sector size is also + * 4Kb. + */ + sector_offset = offset % ICE_AQ_MAX_BUF_LEN; + read_size = min_t(u32, ICE_AQ_MAX_BUF_LEN - sector_offset, + inlen - bytes_read); - status = ice_check_sr_access_params(hw, offset, words); + last_cmd = !(bytes_read + read_size < inlen); - /* values in "offset" and "words" parameters are sized as words - * (16 bits) but ice_aq_read_nvm expects these values in bytes. - * So do this conversion while calling ice_aq_read_nvm. - */ - if (!status) - status = ice_aq_read_nvm(hw, 0, 2 * offset, 2 * words, data, - last_command, NULL); + status = ice_aq_read_nvm(hw, ICE_AQC_NVM_START_POINT, + offset, read_size, + data + bytes_read, last_cmd, + read_shadow_ram, NULL); + if (status) + break; + bytes_read += read_size; + offset += read_size; + } while (!last_cmd); + + *length = bytes_read; return status; } @@ -110,75 +113,25 @@ ice_read_sr_aq(struct ice_hw *hw, u32 offset, u16 words, u16 *data, * @offset: offset of the Shadow RAM word to read (0x000000 - 0x001FFF) * @data: word read from the Shadow RAM * - * Reads one 16 bit word from the Shadow RAM using the ice_read_sr_aq method. + * Reads one 16 bit word from the Shadow RAM using ice_read_flat_nvm. */ static enum ice_status ice_read_sr_word_aq(struct ice_hw *hw, u16 offset, u16 *data) { + u32 bytes = sizeof(u16); enum ice_status status; + __le16 data_local; - status = ice_read_sr_aq(hw, offset, 1, data, true); - if (!status) - *data = le16_to_cpu(*(__force __le16 *)data); - - return status; -} - -/** - * ice_read_sr_buf_aq - Reads Shadow RAM buf via AQ - * @hw: pointer to the HW structure - * @offset: offset of the Shadow RAM word to read (0x000000 - 0x001FFF) - * @words: (in) number of words to read; (out) number of words actually read - * @data: words read from the Shadow RAM - * - * Reads 16 bit words (data buf) from the SR using the ice_read_sr_aq - * method. Ownership of the NVM is taken before reading the buffer and later - * released. - */ -static enum ice_status -ice_read_sr_buf_aq(struct ice_hw *hw, u16 offset, u16 *words, u16 *data) -{ - enum ice_status status; - bool last_cmd = false; - u16 words_read = 0; - u16 i = 0; - - do { - u16 read_size, off_w; - - /* Calculate number of bytes we should read in this step. - * It's not allowed to read more than one page at a time or - * to cross page boundaries. - */ - off_w = offset % ICE_SR_SECTOR_SIZE_IN_WORDS; - read_size = off_w ? - min_t(u16, *words, - (ICE_SR_SECTOR_SIZE_IN_WORDS - off_w)) : - min_t(u16, (*words - words_read), - ICE_SR_SECTOR_SIZE_IN_WORDS); - - /* Check if this is last command, if so set proper flag */ - if ((words_read + read_size) >= *words) - last_cmd = true; - - status = ice_read_sr_aq(hw, offset, read_size, - data + words_read, last_cmd); - if (status) - goto read_nvm_buf_aq_exit; - - /* Increment counter for words already read and move offset to - * new read location - */ - words_read += read_size; - offset += read_size; - } while (words_read < *words); - - for (i = 0; i < *words; i++) - data[i] = le16_to_cpu(((__force __le16 *)data)[i]); + /* Note that ice_read_flat_nvm takes into account the 4Kb AdminQ and + * Shadow RAM sector restrictions necessary when reading from the NVM. + */ + status = ice_read_flat_nvm(hw, offset * sizeof(u16), &bytes, + (u8 *)&data_local, true); + if (status) + return status; -read_nvm_buf_aq_exit: - *words = words_read; - return status; + *data = le16_to_cpu(data_local); + return 0; } /** @@ -188,7 +141,7 @@ read_nvm_buf_aq_exit: * * This function will request NVM ownership. */ -static enum ice_status +enum ice_status ice_acquire_nvm(struct ice_hw *hw, enum ice_aq_res_access_type access) { if (hw->nvm.blank_nvm_mode) @@ -203,7 +156,7 @@ ice_acquire_nvm(struct ice_hw *hw, enum ice_aq_res_access_type access) * * This function will release NVM ownership. */ -static void ice_release_nvm(struct ice_hw *hw) +void ice_release_nvm(struct ice_hw *hw) { if (hw->nvm.blank_nvm_mode) return; @@ -233,6 +186,239 @@ enum ice_status ice_read_sr_word(struct ice_hw *hw, u16 offset, u16 *data) } /** + * ice_get_pfa_module_tlv - Reads sub module TLV from NVM PFA + * @hw: pointer to hardware structure + * @module_tlv: pointer to module TLV to return + * @module_tlv_len: pointer to module TLV length to return + * @module_type: module type requested + * + * Finds the requested sub module TLV type from the Preserved Field + * Area (PFA) and returns the TLV pointer and length. The caller can + * use these to read the variable length TLV value. + */ +enum ice_status +ice_get_pfa_module_tlv(struct ice_hw *hw, u16 *module_tlv, u16 *module_tlv_len, + u16 module_type) +{ + enum ice_status status; + u16 pfa_len, pfa_ptr; + u16 next_tlv; + + status = ice_read_sr_word(hw, ICE_SR_PFA_PTR, &pfa_ptr); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Preserved Field Array pointer.\n"); + return status; + } + status = ice_read_sr_word(hw, pfa_ptr, &pfa_len); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Failed to read PFA length.\n"); + return status; + } + /* Starting with first TLV after PFA length, iterate through the list + * of TLVs to find the requested one. + */ + next_tlv = pfa_ptr + 1; + while (next_tlv < pfa_ptr + pfa_len) { + u16 tlv_sub_module_type; + u16 tlv_len; + + /* Read TLV type */ + status = ice_read_sr_word(hw, next_tlv, &tlv_sub_module_type); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Failed to read TLV type.\n"); + break; + } + /* Read TLV length */ + status = ice_read_sr_word(hw, next_tlv + 1, &tlv_len); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Failed to read TLV length.\n"); + break; + } + if (tlv_sub_module_type == module_type) { + if (tlv_len) { + *module_tlv = next_tlv; + *module_tlv_len = tlv_len; + return 0; + } + return ICE_ERR_INVAL_SIZE; + } + /* Check next TLV, i.e. current TLV pointer + length + 2 words + * (for current TLV's type and length) + */ + next_tlv = next_tlv + tlv_len + 2; + } + /* Module does not exist */ + return ICE_ERR_DOES_NOT_EXIST; +} + +/** + * ice_read_pba_string - Reads part number string from NVM + * @hw: pointer to hardware structure + * @pba_num: stores the part number string from the NVM + * @pba_num_size: part number string buffer length + * + * Reads the part number string from the NVM. + */ +enum ice_status +ice_read_pba_string(struct ice_hw *hw, u8 *pba_num, u32 pba_num_size) +{ + u16 pba_tlv, pba_tlv_len; + enum ice_status status; + u16 pba_word, pba_size; + u16 i; + + status = ice_get_pfa_module_tlv(hw, &pba_tlv, &pba_tlv_len, + ICE_SR_PBA_BLOCK_PTR); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Failed to read PBA Block TLV.\n"); + return status; + } + + /* pba_size is the next word */ + status = ice_read_sr_word(hw, (pba_tlv + 2), &pba_size); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Failed to read PBA Section size.\n"); + return status; + } + + if (pba_tlv_len < pba_size) { + ice_debug(hw, ICE_DBG_INIT, "Invalid PBA Block TLV size.\n"); + return ICE_ERR_INVAL_SIZE; + } + + /* Subtract one to get PBA word count (PBA Size word is included in + * total size) + */ + pba_size--; + if (pba_num_size < (((u32)pba_size * 2) + 1)) { + ice_debug(hw, ICE_DBG_INIT, "Buffer too small for PBA data.\n"); + return ICE_ERR_PARAM; + } + + for (i = 0; i < pba_size; i++) { + status = ice_read_sr_word(hw, (pba_tlv + 2 + 1) + i, &pba_word); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Failed to read PBA Block word %d.\n", i); + return status; + } + + pba_num[(i * 2)] = (pba_word >> 8) & 0xFF; + pba_num[(i * 2) + 1] = pba_word & 0xFF; + } + pba_num[(pba_size * 2)] = '\0'; + + return status; +} + +/** + * ice_get_orom_ver_info - Read Option ROM version information + * @hw: pointer to the HW struct + * + * Read the Combo Image version data from the Boot Configuration TLV and fill + * in the option ROM version data. + */ +static enum ice_status ice_get_orom_ver_info(struct ice_hw *hw) +{ + u16 combo_hi, combo_lo, boot_cfg_tlv, boot_cfg_tlv_len; + struct ice_orom_info *orom = &hw->nvm.orom; + enum ice_status status; + u32 combo_ver; + + status = ice_get_pfa_module_tlv(hw, &boot_cfg_tlv, &boot_cfg_tlv_len, + ICE_SR_BOOT_CFG_PTR); + if (status) { + ice_debug(hw, ICE_DBG_INIT, + "Failed to read Boot Configuration Block TLV.\n"); + return status; + } + + /* Boot Configuration Block must have length at least 2 words + * (Combo Image Version High and Combo Image Version Low) + */ + if (boot_cfg_tlv_len < 2) { + ice_debug(hw, ICE_DBG_INIT, + "Invalid Boot Configuration Block TLV size.\n"); + return ICE_ERR_INVAL_SIZE; + } + + status = ice_read_sr_word(hw, (boot_cfg_tlv + ICE_NVM_OROM_VER_OFF), + &combo_hi); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Failed to read OROM_VER hi.\n"); + return status; + } + + status = ice_read_sr_word(hw, (boot_cfg_tlv + ICE_NVM_OROM_VER_OFF + 1), + &combo_lo); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Failed to read OROM_VER lo.\n"); + return status; + } + + combo_ver = ((u32)combo_hi << 16) | combo_lo; + + orom->major = (u8)((combo_ver & ICE_OROM_VER_MASK) >> + ICE_OROM_VER_SHIFT); + orom->patch = (u8)(combo_ver & ICE_OROM_VER_PATCH_MASK); + orom->build = (u16)((combo_ver & ICE_OROM_VER_BUILD_MASK) >> + ICE_OROM_VER_BUILD_SHIFT); + + return 0; +} + +/** + * ice_discover_flash_size - Discover the available flash size. + * @hw: pointer to the HW struct + * + * The device flash could be up to 16MB in size. However, it is possible that + * the actual size is smaller. Use bisection to determine the accessible size + * of flash memory. + */ +static enum ice_status ice_discover_flash_size(struct ice_hw *hw) +{ + u32 min_size = 0, max_size = ICE_AQC_NVM_MAX_OFFSET + 1; + enum ice_status status; + + status = ice_acquire_nvm(hw, ICE_RES_READ); + if (status) + return status; + + while ((max_size - min_size) > 1) { + u32 offset = (max_size + min_size) / 2; + u32 len = 1; + u8 data; + + status = ice_read_flat_nvm(hw, offset, &len, &data, false); + if (status == ICE_ERR_AQ_ERROR && + hw->adminq.sq_last_status == ICE_AQ_RC_EINVAL) { + ice_debug(hw, ICE_DBG_NVM, + "%s: New upper bound of %u bytes\n", + __func__, offset); + status = 0; + max_size = offset; + } else if (!status) { + ice_debug(hw, ICE_DBG_NVM, + "%s: New lower bound of %u bytes\n", + __func__, offset); + min_size = offset; + } else { + /* an unexpected error occurred */ + goto err_read_flat_nvm; + } + } + + ice_debug(hw, ICE_DBG_NVM, + "Predicted flash size is %u bytes\n", max_size); + + hw->nvm.flash_size = max_size; + +err_read_flat_nvm: + ice_release_nvm(hw); + + return status; +} + +/** * ice_init_nvm - initializes NVM setting * @hw: pointer to the HW struct * @@ -241,9 +427,8 @@ enum ice_status ice_read_sr_word(struct ice_hw *hw, u16 offset, u16 *data) */ enum ice_status ice_init_nvm(struct ice_hw *hw) { - u16 oem_hi, oem_lo, boot_cfg_tlv, boot_cfg_tlv_len; struct ice_nvm_info *nvm = &hw->nvm; - u16 eetrack_lo, eetrack_hi; + u16 eetrack_lo, eetrack_hi, ver; enum ice_status status; u32 fla, gens_stat; u8 sr_size; @@ -269,12 +454,14 @@ enum ice_status ice_init_nvm(struct ice_hw *hw) return ICE_ERR_NVM_BLANK_MODE; } - status = ice_read_sr_word(hw, ICE_SR_NVM_DEV_STARTER_VER, &nvm->ver); + status = ice_read_sr_word(hw, ICE_SR_NVM_DEV_STARTER_VER, &ver); if (status) { ice_debug(hw, ICE_DBG_INIT, "Failed to read DEV starter version.\n"); return status; } + nvm->major_ver = (ver & ICE_NVM_VER_HI_MASK) >> ICE_NVM_VER_HI_SHIFT; + nvm->minor_ver = (ver & ICE_NVM_VER_LO_MASK) >> ICE_NVM_VER_LO_SHIFT; status = ice_read_sr_word(hw, ICE_SR_NVM_EETRACK_LO, &eetrack_lo); if (status) { @@ -289,80 +476,49 @@ enum ice_status ice_init_nvm(struct ice_hw *hw) nvm->eetrack = (eetrack_hi << 16) | eetrack_lo; - /* the following devices do not have boot_cfg_tlv yet */ - if (hw->device_id == ICE_DEV_ID_E822C_BACKPLANE || - hw->device_id == ICE_DEV_ID_E822C_QSFP || - hw->device_id == ICE_DEV_ID_E822C_10G_BASE_T || - hw->device_id == ICE_DEV_ID_E822C_SGMII || - hw->device_id == ICE_DEV_ID_E822C_SFP || - hw->device_id == ICE_DEV_ID_E822X_BACKPLANE || - hw->device_id == ICE_DEV_ID_E822L_SFP || - hw->device_id == ICE_DEV_ID_E822L_10G_BASE_T || - hw->device_id == ICE_DEV_ID_E822L_SGMII) - return status; - - status = ice_get_pfa_module_tlv(hw, &boot_cfg_tlv, &boot_cfg_tlv_len, - ICE_SR_BOOT_CFG_PTR); + status = ice_discover_flash_size(hw); if (status) { - ice_debug(hw, ICE_DBG_INIT, - "Failed to read Boot Configuration Block TLV.\n"); + ice_debug(hw, ICE_DBG_NVM, + "NVM init error: failed to discover flash size.\n"); return status; } - /* Boot Configuration Block must have length at least 2 words - * (Combo Image Version High and Combo Image Version Low) - */ - if (boot_cfg_tlv_len < 2) { - ice_debug(hw, ICE_DBG_INIT, - "Invalid Boot Configuration Block TLV size.\n"); - return ICE_ERR_INVAL_SIZE; - } - - status = ice_read_sr_word(hw, (boot_cfg_tlv + ICE_NVM_OEM_VER_OFF), - &oem_hi); - if (status) { - ice_debug(hw, ICE_DBG_INIT, "Failed to read OEM_VER hi.\n"); + switch (hw->device_id) { + /* the following devices do not have boot_cfg_tlv yet */ + case ICE_DEV_ID_E823C_BACKPLANE: + case ICE_DEV_ID_E823C_QSFP: + case ICE_DEV_ID_E823C_SFP: + case ICE_DEV_ID_E823C_10G_BASE_T: + case ICE_DEV_ID_E823C_SGMII: + case ICE_DEV_ID_E822C_BACKPLANE: + case ICE_DEV_ID_E822C_QSFP: + case ICE_DEV_ID_E822C_10G_BASE_T: + case ICE_DEV_ID_E822C_SGMII: + case ICE_DEV_ID_E822C_SFP: + case ICE_DEV_ID_E822L_BACKPLANE: + case ICE_DEV_ID_E822L_SFP: + case ICE_DEV_ID_E822L_10G_BASE_T: + case ICE_DEV_ID_E822L_SGMII: + case ICE_DEV_ID_E823L_BACKPLANE: + case ICE_DEV_ID_E823L_SFP: + case ICE_DEV_ID_E823L_10G_BASE_T: + case ICE_DEV_ID_E823L_1GBE: + case ICE_DEV_ID_E823L_QSFP: return status; + default: + break; } - status = ice_read_sr_word(hw, (boot_cfg_tlv + ICE_NVM_OEM_VER_OFF + 1), - &oem_lo); + status = ice_get_orom_ver_info(hw); if (status) { - ice_debug(hw, ICE_DBG_INIT, "Failed to read OEM_VER lo.\n"); + ice_debug(hw, ICE_DBG_INIT, "Failed to read Option ROM info.\n"); return status; } - nvm->oem_ver = ((u32)oem_hi << 16) | oem_lo; - return 0; } /** - * ice_read_sr_buf - Reads Shadow RAM buf and acquire lock if necessary - * @hw: pointer to the HW structure - * @offset: offset of the Shadow RAM word to read (0x000000 - 0x001FFF) - * @words: (in) number of words to read; (out) number of words actually read - * @data: words read from the Shadow RAM - * - * Reads 16 bit words (data buf) from the SR using the ice_read_nvm_buf_aq - * method. The buf read is preceded by the NVM ownership take - * and followed by the release. - */ -enum ice_status -ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words, u16 *data) -{ - enum ice_status status; - - status = ice_acquire_nvm(hw, ICE_RES_READ); - if (!status) { - status = ice_read_sr_buf_aq(hw, offset, words, data); - ice_release_nvm(hw); - } - - return status; -} - -/** * ice_nvm_validate_checksum * @hw: pointer to the HW struct * diff --git a/drivers/net/ethernet/intel/ice/ice_nvm.h b/drivers/net/ethernet/intel/ice/ice_nvm.h index a9fa011c22c6..999f273ba6ad 100644 --- a/drivers/net/ethernet/intel/ice/ice_nvm.h +++ b/drivers/net/ethernet/intel/ice/ice_nvm.h @@ -4,5 +4,17 @@ #ifndef _ICE_NVM_H_ #define _ICE_NVM_H_ +enum ice_status +ice_acquire_nvm(struct ice_hw *hw, enum ice_aq_res_access_type access); +void ice_release_nvm(struct ice_hw *hw); +enum ice_status +ice_read_flat_nvm(struct ice_hw *hw, u32 offset, u32 *length, u8 *data, + bool read_shadow_ram); +enum ice_status +ice_get_pfa_module_tlv(struct ice_hw *hw, u16 *module_tlv, u16 *module_tlv_len, + u16 module_type); +enum ice_status +ice_read_pba_string(struct ice_hw *hw, u8 *pba_num, u32 pba_num_size); +enum ice_status ice_init_nvm(struct ice_hw *hw); enum ice_status ice_read_sr_word(struct ice_hw *hw, u16 offset, u16 *data); #endif /* _ICE_NVM_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c index d2db0d04e117..554f567476f3 100644 --- a/drivers/net/ethernet/intel/ice/ice_sriov.c +++ b/drivers/net/ethernet/intel/ice/ice_sriov.c @@ -121,9 +121,7 @@ u32 ice_conv_link_speed_to_virtchnl(bool adv_link_support, u16 link_speed) speed = (u32)VIRTCHNL_LINK_SPEED_25GB; break; case ICE_AQ_LINK_SPEED_40GB: - /* fall through */ case ICE_AQ_LINK_SPEED_50GB: - /* fall through */ case ICE_AQ_LINK_SPEED_100GB: speed = (u32)VIRTCHNL_LINK_SPEED_40GB; break; diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c index 431266081a80..51825a203e35 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.c +++ b/drivers/net/ethernet/intel/ice/ice_switch.c @@ -578,7 +578,7 @@ enum ice_status ice_get_initial_sw_cfg(struct ice_hw *hw) struct ice_aqc_get_sw_cfg_resp_elem *ele; u16 pf_vf_num, swid, vsi_port_num; bool is_vf = false; - u8 type; + u8 res_type; ele = rbuf[i].elements; vsi_port_num = le16_to_cpu(ele->vsi_port_num) & @@ -593,16 +593,16 @@ enum ice_status ice_get_initial_sw_cfg(struct ice_hw *hw) ICE_AQC_GET_SW_CONF_RESP_IS_VF) is_vf = true; - type = le16_to_cpu(ele->vsi_port_num) >> + res_type = le16_to_cpu(ele->vsi_port_num) >> ICE_AQC_GET_SW_CONF_RESP_TYPE_S; - if (type == ICE_AQC_GET_SW_CONF_RESP_VSI) { + if (res_type == ICE_AQC_GET_SW_CONF_RESP_VSI) { /* FW VSI is not needed. Just continue. */ continue; } ice_init_port_info(hw->port_info, vsi_port_num, - type, swid, pf_vf_num, is_vf); + res_type, swid, pf_vf_num, is_vf); } } while (req_desc && !status); @@ -760,7 +760,7 @@ ice_fill_sw_rule(struct ice_hw *hw, struct ice_fltr_info *f_info, break; case ICE_SW_LKUP_ETHERTYPE_MAC: daddr = f_info->l_data.ethertype_mac.mac_addr; - /* fall-through */ + fallthrough; case ICE_SW_LKUP_ETHERTYPE: off = (__force __be16 *)(eth_hdr + ICE_ETH_ETHTYPE_OFFSET); *off = cpu_to_be16(f_info->l_data.ethertype_mac.ethertype); @@ -771,7 +771,7 @@ ice_fill_sw_rule(struct ice_hw *hw, struct ice_fltr_info *f_info, break; case ICE_SW_LKUP_PROMISC_VLAN: vlan_id = f_info->l_data.mac_vlan.vlan_id; - /* fall-through */ + fallthrough; case ICE_SW_LKUP_PROMISC: daddr = f_info->l_data.mac_vlan.mac_addr; break; @@ -958,7 +958,7 @@ ice_update_vsi_list_rule(struct ice_hw *hw, u16 *vsi_handle_arr, u16 num_vsi, struct ice_aqc_sw_rules_elem *s_rule; enum ice_status status; u16 s_rule_size; - u16 type; + u16 rule_type; int i; if (!num_vsi) @@ -970,11 +970,11 @@ ice_update_vsi_list_rule(struct ice_hw *hw, u16 *vsi_handle_arr, u16 num_vsi, lkup_type == ICE_SW_LKUP_ETHERTYPE_MAC || lkup_type == ICE_SW_LKUP_PROMISC || lkup_type == ICE_SW_LKUP_PROMISC_VLAN) - type = remove ? ICE_AQC_SW_RULES_T_VSI_LIST_CLEAR : - ICE_AQC_SW_RULES_T_VSI_LIST_SET; + rule_type = remove ? ICE_AQC_SW_RULES_T_VSI_LIST_CLEAR : + ICE_AQC_SW_RULES_T_VSI_LIST_SET; else if (lkup_type == ICE_SW_LKUP_VLAN) - type = remove ? ICE_AQC_SW_RULES_T_PRUNE_LIST_CLEAR : - ICE_AQC_SW_RULES_T_PRUNE_LIST_SET; + rule_type = remove ? ICE_AQC_SW_RULES_T_PRUNE_LIST_CLEAR : + ICE_AQC_SW_RULES_T_PRUNE_LIST_SET; else return ICE_ERR_PARAM; @@ -992,7 +992,7 @@ ice_update_vsi_list_rule(struct ice_hw *hw, u16 *vsi_handle_arr, u16 num_vsi, cpu_to_le16(ice_get_hw_vsi_num(hw, vsi_handle_arr[i])); } - s_rule->type = cpu_to_le16(type); + s_rule->type = cpu_to_le16(rule_type); s_rule->pdata.vsi_list.number_vsi = cpu_to_le16(num_vsi); s_rule->pdata.vsi_list.index = cpu_to_le16(vsi_list_id); diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 4de61dbedd36..f67e8362958c 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -453,10 +453,10 @@ ice_run_xdp(struct ice_ring *rx_ring, struct xdp_buff *xdp, break; default: bpf_warn_invalid_xdp_action(act); - /* fallthrough -- not supported action */ + fallthrough; case XDP_ABORTED: trace_xdp_exception(rx_ring->netdev, xdp_prog, act); - /* fallthrough -- handle aborts by dropping frame */ + fallthrough; case XDP_DROP: result = ICE_XDP_CONSUMED; break; @@ -1188,7 +1188,6 @@ ice_adjust_itr_by_size_and_speed(struct ice_port_info *port_info, avg_pkt_size + 640); break; case ICE_AQ_LINK_SPEED_10GB: - /* fall through */ default: itr += DIV_ROUND_UP(170 * (avg_pkt_size + 24), avg_pkt_size + 640); diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index db0ef6ba907f..4ce5f92fca4a 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -239,12 +239,21 @@ struct ice_fc_info { enum ice_fc_mode req_mode; /* FC mode requested by caller */ }; +/* Option ROM version information */ +struct ice_orom_info { + u8 major; /* Major version of OROM */ + u8 patch; /* Patch version of OROM */ + u16 build; /* Build version of OROM */ +}; + /* NVM Information */ struct ice_nvm_info { - u32 eetrack; /* NVM data version */ - u32 oem_ver; /* OEM version info */ - u16 sr_words; /* Shadow RAM size in words */ - u16 ver; /* NVM package version */ + struct ice_orom_info orom; /* Option ROM version info */ + u32 eetrack; /* NVM data version */ + u16 sr_words; /* Shadow RAM size in words */ + u32 flash_size; /* Size of available flash in bytes */ + u8 major_ver; /* major version of NVM package */ + u8 minor_ver; /* minor version of dev starter */ u8 blank_nvm_mode; /* is NVM empty (no FW present) */ }; @@ -626,7 +635,8 @@ struct ice_hw_port_stats { /* Checksum and Shadow RAM pointers */ #define ICE_SR_BOOT_CFG_PTR 0x132 -#define ICE_NVM_OEM_VER_OFF 0x02 +#define ICE_NVM_OROM_VER_OFF 0x02 +#define ICE_SR_PBA_BLOCK_PTR 0x16 #define ICE_SR_NVM_DEV_STARTER_VER 0x18 #define ICE_SR_NVM_EETRACK_LO 0x2D #define ICE_SR_NVM_EETRACK_HI 0x2E @@ -634,12 +644,12 @@ struct ice_hw_port_stats { #define ICE_NVM_VER_LO_MASK (0xff << ICE_NVM_VER_LO_SHIFT) #define ICE_NVM_VER_HI_SHIFT 12 #define ICE_NVM_VER_HI_MASK (0xf << ICE_NVM_VER_HI_SHIFT) -#define ICE_OEM_VER_PATCH_SHIFT 0 -#define ICE_OEM_VER_PATCH_MASK (0xff << ICE_OEM_VER_PATCH_SHIFT) -#define ICE_OEM_VER_BUILD_SHIFT 8 -#define ICE_OEM_VER_BUILD_MASK (0xffff << ICE_OEM_VER_BUILD_SHIFT) -#define ICE_OEM_VER_SHIFT 24 -#define ICE_OEM_VER_MASK (0xff << ICE_OEM_VER_SHIFT) +#define ICE_OROM_VER_PATCH_SHIFT 0 +#define ICE_OROM_VER_PATCH_MASK (0xff << ICE_OROM_VER_PATCH_SHIFT) +#define ICE_OROM_VER_BUILD_SHIFT 8 +#define ICE_OROM_VER_BUILD_MASK (0xffff << ICE_OROM_VER_BUILD_SHIFT) +#define ICE_OROM_VER_SHIFT 24 +#define ICE_OROM_VER_MASK (0xff << ICE_OROM_VER_SHIFT) #define ICE_SR_PFA_PTR 0x40 #define ICE_SR_SECTOR_SIZE_IN_WORDS 0x800 #define ICE_SR_WORDS_IN_1KB 512 diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 75c70d432c72..15191a325918 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -91,6 +91,39 @@ ice_set_pfe_link(struct ice_vf *vf, struct virtchnl_pf_event *pfe, } /** + * ice_vf_has_no_qs_ena - check if the VF has any Rx or Tx queues enabled + * @vf: the VF to check + * + * Returns true if the VF has no Rx and no Tx queues enabled and returns false + * otherwise + */ +static bool ice_vf_has_no_qs_ena(struct ice_vf *vf) +{ + return (!bitmap_weight(vf->rxq_ena, ICE_MAX_RSS_QS_PER_VF) && + !bitmap_weight(vf->txq_ena, ICE_MAX_RSS_QS_PER_VF)); +} + +/** + * ice_is_vf_link_up - check if the VF's link is up + * @vf: VF to check if link is up + */ +static bool ice_is_vf_link_up(struct ice_vf *vf) +{ + struct ice_pf *pf = vf->pf; + + if (ice_check_vf_init(pf, vf)) + return false; + + if (ice_vf_has_no_qs_ena(vf)) + return false; + else if (vf->link_forced) + return vf->link_up; + else + return pf->hw.port_info->phy.link_info.link_info & + ICE_AQ_LINK_UP; +} + +/** * ice_vc_notify_vf_link_state - Inform a VF of link status * @vf: pointer to the VF structure * @@ -99,28 +132,16 @@ ice_set_pfe_link(struct ice_vf *vf, struct virtchnl_pf_event *pfe, static void ice_vc_notify_vf_link_state(struct ice_vf *vf) { struct virtchnl_pf_event pfe = { 0 }; - struct ice_link_status *ls; - struct ice_pf *pf = vf->pf; - struct ice_hw *hw; - - hw = &pf->hw; - ls = &hw->port_info->phy.link_info; + struct ice_hw *hw = &vf->pf->hw; pfe.event = VIRTCHNL_EVENT_LINK_CHANGE; pfe.severity = PF_EVENT_SEVERITY_INFO; - /* Always report link is down if the VF queues aren't enabled */ - if (!vf->num_qs_ena) { + if (ice_is_vf_link_up(vf)) + ice_set_pfe_link(vf, &pfe, + hw->port_info->phy.link_info.link_speed, true); + else ice_set_pfe_link(vf, &pfe, ICE_AQ_LINK_SPEED_UNKNOWN, false); - } else if (vf->link_forced) { - u16 link_speed = vf->link_up ? - ls->link_speed : ICE_AQ_LINK_SPEED_UNKNOWN; - - ice_set_pfe_link(vf, &pfe, link_speed, vf->link_up); - } else { - ice_set_pfe_link(vf, &pfe, ls->link_speed, - ls->link_info & ICE_AQ_LINK_UP); - } ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT, VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe, @@ -149,7 +170,12 @@ static void ice_free_vf_res(struct ice_vf *vf) vf->num_mac = 0; } - last_vector_idx = vf->first_vector_idx + pf->num_vf_msix - 1; + last_vector_idx = vf->first_vector_idx + pf->num_msix_per_vf - 1; + + /* clear VF MDD event information */ + memset(&vf->mdd_tx_events, 0, sizeof(vf->mdd_tx_events)); + memset(&vf->mdd_rx_events, 0, sizeof(vf->mdd_rx_events)); + /* Disable interrupts so that VF starts in a known state */ for (i = vf->first_vector_idx; i <= last_vector_idx; i++) { wr32(&pf->hw, GLINT_DYN_CTL(i), GLINT_DYN_CTL_CLEARPBA_M); @@ -180,7 +206,7 @@ static void ice_dis_vf_mappings(struct ice_vf *vf) wr32(hw, VPINT_ALLOC_PCI(vf->vf_id), 0); first = vf->first_vector_idx; - last = first + pf->num_vf_msix - 1; + last = first + pf->num_msix_per_vf - 1; for (v = first; v <= last; v++) { u32 reg; @@ -206,11 +232,7 @@ static void ice_dis_vf_mappings(struct ice_vf *vf) * ice_sriov_free_msix_res - Reset/free any used MSIX resources * @pf: pointer to the PF structure * - * If MSIX entries from the pf->irq_tracker were needed then we need to - * reset the irq_tracker->end and give back the entries we needed to - * num_avail_sw_msix. - * - * If no MSIX entries were taken from the pf->irq_tracker then just clear + * Since no MSIX entries are taken from the pf->irq_tracker then just clear * the pf->sriov_base_vector. * * Returns 0 on success, and -EINVAL on error. @@ -227,11 +249,7 @@ static int ice_sriov_free_msix_res(struct ice_pf *pf) return -EINVAL; /* give back irq_tracker resources used */ - if (pf->sriov_base_vector < res->num_entries) { - res->end = res->num_entries; - pf->num_avail_sw_msix += - res->num_entries - pf->sriov_base_vector; - } + WARN_ON(pf->sriov_base_vector < res->num_entries); pf->sriov_base_vector = 0; @@ -245,9 +263,8 @@ static int ice_sriov_free_msix_res(struct ice_pf *pf) void ice_set_vf_state_qs_dis(struct ice_vf *vf) { /* Clear Rx/Tx enabled queues flag */ - bitmap_zero(vf->txq_ena, ICE_MAX_BASE_QS_PER_VF); - bitmap_zero(vf->rxq_ena, ICE_MAX_BASE_QS_PER_VF); - vf->num_qs_ena = 0; + bitmap_zero(vf->txq_ena, ICE_MAX_RSS_QS_PER_VF); + bitmap_zero(vf->rxq_ena, ICE_MAX_RSS_QS_PER_VF); clear_bit(ICE_VF_STATE_QS_ENA, vf->vf_states); } @@ -263,7 +280,7 @@ static void ice_dis_vf_qs(struct ice_vf *vf) vsi = pf->vsi[vf->lan_vsi_idx]; ice_vsi_stop_lan_tx_rings(vsi, ICE_NO_RESET, vf->vf_id); - ice_vsi_stop_rx_rings(vsi); + ice_vsi_stop_all_rx_rings(vsi); ice_set_vf_state_qs_dis(vf); } @@ -283,11 +300,6 @@ void ice_free_vfs(struct ice_pf *pf) while (test_and_set_bit(__ICE_VF_DIS, pf->state)) usleep_range(1000, 2000); - /* Avoid wait time by stopping all VFs at the same time */ - ice_for_each_vf(pf, i) - if (test_bit(ICE_VF_STATE_QS_ENA, pf->vf[i].vf_states)) - ice_dis_vf_qs(&pf->vf[i]); - /* Disable IOV before freeing resources. This lets any VF drivers * running in the host get themselves cleaned up before we yank * the carpet out from underneath their feet. @@ -297,8 +309,13 @@ void ice_free_vfs(struct ice_pf *pf) else dev_warn(dev, "VFs are assigned - not disabling SR-IOV\n"); + /* Avoid wait time by stopping all VFs at the same time */ + ice_for_each_vf(pf, i) + if (test_bit(ICE_VF_STATE_QS_ENA, pf->vf[i].vf_states)) + ice_dis_vf_qs(&pf->vf[i]); + tmp = pf->num_alloc_vfs; - pf->num_vf_qps = 0; + pf->num_qps_per_vf = 0; pf->num_alloc_vfs = 0; for (i = 0; i < tmp; i++) { if (test_bit(ICE_VF_STATE_INIT, pf->vf[i].vf_states)) { @@ -407,43 +424,15 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr, bool is_pfr) } /** - * ice_vsi_set_pvid_fill_ctxt - Set VSI ctxt for add PVID - * @ctxt: the VSI ctxt to fill - * @vid: the VLAN ID to set as a PVID - */ -static void ice_vsi_set_pvid_fill_ctxt(struct ice_vsi_ctx *ctxt, u16 vid) -{ - ctxt->info.vlan_flags = (ICE_AQ_VSI_VLAN_MODE_UNTAGGED | - ICE_AQ_VSI_PVLAN_INSERT_PVID | - ICE_AQ_VSI_VLAN_EMOD_STR); - ctxt->info.pvid = cpu_to_le16(vid); - ctxt->info.sw_flags2 |= ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA; - ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID | - ICE_AQ_VSI_PROP_SW_VALID); -} - -/** - * ice_vsi_kill_pvid_fill_ctxt - Set VSI ctx for remove PVID - * @ctxt: the VSI ctxt to fill - */ -static void ice_vsi_kill_pvid_fill_ctxt(struct ice_vsi_ctx *ctxt) -{ - ctxt->info.vlan_flags = ICE_AQ_VSI_VLAN_EMOD_NOTHING; - ctxt->info.vlan_flags |= ICE_AQ_VSI_VLAN_MODE_ALL; - ctxt->info.sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA; - ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID | - ICE_AQ_VSI_PROP_SW_VALID); -} - -/** * ice_vsi_manage_pvid - Enable or disable port VLAN for VSI * @vsi: the VSI to update - * @vid: the VLAN ID to set as a PVID + * @pvid_info: VLAN ID and QoS used to set the PVID VSI context field * @enable: true for enable PVID false for disable */ -static int ice_vsi_manage_pvid(struct ice_vsi *vsi, u16 vid, bool enable) +static int ice_vsi_manage_pvid(struct ice_vsi *vsi, u16 pvid_info, bool enable) { struct ice_hw *hw = &vsi->back->hw; + struct ice_aqc_vsi_props *info; struct ice_vsi_ctx *ctxt; enum ice_status status; int ret = 0; @@ -453,20 +442,33 @@ static int ice_vsi_manage_pvid(struct ice_vsi *vsi, u16 vid, bool enable) return -ENOMEM; ctxt->info = vsi->info; - if (enable) - ice_vsi_set_pvid_fill_ctxt(ctxt, vid); - else - ice_vsi_kill_pvid_fill_ctxt(ctxt); + info = &ctxt->info; + if (enable) { + info->vlan_flags = ICE_AQ_VSI_VLAN_MODE_UNTAGGED | + ICE_AQ_VSI_PVLAN_INSERT_PVID | + ICE_AQ_VSI_VLAN_EMOD_STR; + info->sw_flags2 |= ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA; + } else { + info->vlan_flags = ICE_AQ_VSI_VLAN_EMOD_NOTHING | + ICE_AQ_VSI_VLAN_MODE_ALL; + info->sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA; + } + + info->pvid = cpu_to_le16(pvid_info); + info->valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID | + ICE_AQ_VSI_PROP_SW_VALID); status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); if (status) { - dev_info(ice_pf_to_dev(vsi->back), "update VSI for port VLAN failed, err %d aq_err %d\n", + dev_info(ice_hw_to_dev(hw), "update VSI for port VLAN failed, err %d aq_err %d\n", status, hw->adminq.sq_last_status); ret = -EIO; goto out; } - vsi->info = ctxt->info; + vsi->info.vlan_flags = info->vlan_flags; + vsi->info.sw_flags2 = info->sw_flags2; + vsi->info.pvid = info->pvid; out: kfree(ctxt); return ret; @@ -501,7 +503,7 @@ ice_vf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, u16 vf_id) */ static int ice_calc_vf_first_vector_idx(struct ice_pf *pf, struct ice_vf *vf) { - return pf->sriov_base_vector + vf->vf_id * pf->num_vf_msix; + return pf->sriov_base_vector + vf->vf_id * pf->num_msix_per_vf; } /** @@ -533,9 +535,20 @@ static int ice_alloc_vsi_res(struct ice_vf *vf) vf->lan_vsi_num = vsi->vsi_num; /* Check if port VLAN exist before, and restore it accordingly */ - if (vf->port_vlan_id) { - ice_vsi_manage_pvid(vsi, vf->port_vlan_id, true); - ice_vsi_add_vlan(vsi, vf->port_vlan_id & ICE_VLAN_M); + if (vf->port_vlan_info) { + ice_vsi_manage_pvid(vsi, vf->port_vlan_info, true); + if (ice_vsi_add_vlan(vsi, vf->port_vlan_info & VLAN_VID_MASK)) + dev_warn(ice_pf_to_dev(pf), "Failed to add Port VLAN %d filter for VF %d\n", + vf->port_vlan_info & VLAN_VID_MASK, vf->vf_id); + } else { + /* set VLAN 0 filter by default when no port VLAN is + * enabled. If a port VLAN is enabled we don't want + * untagged broadcast/multicast traffic seen on the VF + * interface. + */ + if (ice_vsi_add_vlan(vsi, 0)) + dev_warn(ice_pf_to_dev(pf), "Failed to add VLAN 0 filter for VF %d, MDD events will trigger. Reset the VF, disable spoofchk, or enable 8021q module on the guest\n", + vf->vf_id); } eth_broadcast_addr(broadcast); @@ -583,7 +596,7 @@ static int ice_alloc_vf_res(struct ice_vf *vf) */ tx_rx_queue_left = min_t(int, ice_get_avail_txq_count(pf), ice_get_avail_rxq_count(pf)); - tx_rx_queue_left += ICE_DFLT_QS_PER_VF; + tx_rx_queue_left += pf->num_qps_per_vf; if (vf->num_req_qs && vf->num_req_qs <= tx_rx_queue_left && vf->num_req_qs != vf->num_vf_qs) vf->num_vf_qs = vf->num_req_qs; @@ -629,9 +642,9 @@ static void ice_ena_vf_mappings(struct ice_vf *vf) hw = &pf->hw; vsi = pf->vsi[vf->lan_vsi_idx]; first = vf->first_vector_idx; - last = (first + pf->num_vf_msix) - 1; + last = (first + pf->num_msix_per_vf) - 1; abs_first = first + pf->hw.func_caps.common_cap.msix_vector_first_id; - abs_last = (abs_first + pf->num_vf_msix) - 1; + abs_last = (abs_first + pf->num_msix_per_vf) - 1; abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id; /* VF Vector allocation */ @@ -749,7 +762,7 @@ int ice_calc_vf_reg_idx(struct ice_vf *vf, struct ice_q_vector *q_vector) pf = vf->pf; /* always add one to account for the OICR being the first MSIX */ - return pf->sriov_base_vector + pf->num_vf_msix * vf->vf_id + + return pf->sriov_base_vector + pf->num_msix_per_vf * vf->vf_id + q_vector->v_idx + 1; } @@ -782,127 +795,112 @@ static int ice_get_max_valid_res_idx(struct ice_res_tracker *res) * @num_msix_needed: number of MSIX vectors needed for all SR-IOV VFs * * This function allows SR-IOV resources to be taken from the end of the PF's - * allowed HW MSIX vectors so in many cases the irq_tracker will not - * be needed. In these cases we just set the pf->sriov_base_vector and return - * success. + * allowed HW MSIX vectors so that the irq_tracker will not be affected. We + * just set the pf->sriov_base_vector and return success. * - * If SR-IOV needs to use any pf->irq_tracker entries it updates the - * irq_tracker->end based on the first entry needed for SR-IOV. This makes it - * so any calls to ice_get_res() using the irq_tracker will not try to use - * resources at or beyond the newly set value. + * If there are not enough resources available, return an error. This should + * always be caught by ice_set_per_vf_res(). * * Return 0 on success, and -EINVAL when there are not enough MSIX vectors in * in the PF's space available for SR-IOV. */ static int ice_sriov_set_msix_res(struct ice_pf *pf, u16 num_msix_needed) { - int max_valid_res_idx = ice_get_max_valid_res_idx(pf->irq_tracker); - u16 pf_total_msix_vectors = - pf->hw.func_caps.common_cap.num_msix_vectors; - struct ice_res_tracker *res = pf->irq_tracker; + u16 total_vectors = pf->hw.func_caps.common_cap.num_msix_vectors; + int vectors_used = pf->irq_tracker->num_entries; int sriov_base_vector; - if (max_valid_res_idx < 0) - return max_valid_res_idx; - - sriov_base_vector = pf_total_msix_vectors - num_msix_needed; + sriov_base_vector = total_vectors - num_msix_needed; /* make sure we only grab irq_tracker entries from the list end and * that we have enough available MSIX vectors */ - if (sriov_base_vector <= max_valid_res_idx) + if (sriov_base_vector < vectors_used) return -EINVAL; pf->sriov_base_vector = sriov_base_vector; - /* dip into irq_tracker entries and update used resources */ - if (num_msix_needed > (pf_total_msix_vectors - res->num_entries)) { - pf->num_avail_sw_msix -= - res->num_entries - pf->sriov_base_vector; - res->end = pf->sriov_base_vector; - } - return 0; } /** - * ice_check_avail_res - check if vectors and queues are available + * ice_set_per_vf_res - check if vectors and queues are available * @pf: pointer to the PF structure * - * This function is where we calculate actual number of resources for VF VSIs, - * we don't reserve ahead of time during probe. Returns success if vectors and - * queues resources are available, otherwise returns error code + * First, determine HW interrupts from common pool. If we allocate fewer VFs, we + * get more vectors and can enable more queues per VF. Note that this does not + * grab any vectors from the SW pool already allocated. Also note, that all + * vector counts include one for each VF's miscellaneous interrupt vector + * (i.e. OICR). + * + * Minimum VFs - 2 vectors, 1 queue pair + * Small VFs - 5 vectors, 4 queue pairs + * Medium VFs - 17 vectors, 16 queue pairs + * + * Second, determine number of queue pairs per VF by starting with a pre-defined + * maximum each VF supports. If this is not possible, then we adjust based on + * queue pairs available on the device. + * + * Lastly, set queue and MSI-X VF variables tracked by the PF so it can be used + * by each VF during VF initialization and reset. */ -static int ice_check_avail_res(struct ice_pf *pf) +static int ice_set_per_vf_res(struct ice_pf *pf) { int max_valid_res_idx = ice_get_max_valid_res_idx(pf->irq_tracker); - u16 num_msix, num_txq, num_rxq, num_avail_msix; + int msix_avail_per_vf, msix_avail_for_sriov; struct device *dev = ice_pf_to_dev(pf); + u16 num_msix_per_vf, num_txq, num_rxq; if (!pf->num_alloc_vfs || max_valid_res_idx < 0) return -EINVAL; - /* add 1 to max_valid_res_idx to account for it being 0-based */ - num_avail_msix = pf->hw.func_caps.common_cap.num_msix_vectors - - (max_valid_res_idx + 1); - - /* Grab from HW interrupts common pool - * Note: By the time the user decides it needs more vectors in a VF - * its already too late since one must decide this prior to creating the - * VF interface. So the best we can do is take a guess as to what the - * user might want. - * - * We have two policies for vector allocation: - * 1. if num_alloc_vfs is from 1 to 16, then we consider this as small - * number of NFV VFs used for NFV appliances, since this is a special - * case, we try to assign maximum vectors per VF (65) as much as - * possible, based on determine_resources algorithm. - * 2. if num_alloc_vfs is from 17 to 256, then its large number of - * regular VFs which are not used for any special purpose. Hence try to - * grab default interrupt vectors (5 as supported by AVF driver). - */ - if (pf->num_alloc_vfs <= 16) { - num_msix = ice_determine_res(pf, num_avail_msix, - ICE_MAX_INTR_PER_VF, - ICE_MIN_INTR_PER_VF); - } else if (pf->num_alloc_vfs <= ICE_MAX_VF_COUNT) { - num_msix = ice_determine_res(pf, num_avail_msix, - ICE_DFLT_INTR_PER_VF, - ICE_MIN_INTR_PER_VF); + /* determine MSI-X resources per VF */ + msix_avail_for_sriov = pf->hw.func_caps.common_cap.num_msix_vectors - + pf->irq_tracker->num_entries; + msix_avail_per_vf = msix_avail_for_sriov / pf->num_alloc_vfs; + if (msix_avail_per_vf >= ICE_NUM_VF_MSIX_MED) { + num_msix_per_vf = ICE_NUM_VF_MSIX_MED; + } else if (msix_avail_per_vf >= ICE_NUM_VF_MSIX_SMALL) { + num_msix_per_vf = ICE_NUM_VF_MSIX_SMALL; + } else if (msix_avail_per_vf >= ICE_MIN_INTR_PER_VF) { + num_msix_per_vf = ICE_MIN_INTR_PER_VF; } else { - dev_err(dev, "Number of VFs %d exceeds max VF count %d\n", - pf->num_alloc_vfs, ICE_MAX_VF_COUNT); + dev_err(dev, "Only %d MSI-X interrupts available for SR-IOV. Not enough to support minimum of %d MSI-X interrupts per VF for %d VFs\n", + msix_avail_for_sriov, ICE_MIN_INTR_PER_VF, + pf->num_alloc_vfs); return -EIO; } - if (!num_msix) - return -EIO; - - /* Grab from the common pool - * start by requesting Default queues (4 as supported by AVF driver), - * Note that, the main difference between queues and vectors is, latter - * can only be reserved at init time but queues can be requested by VF - * at runtime through Virtchnl, that is the reason we start by reserving - * few queues. - */ + /* determine queue resources per VF */ num_txq = ice_determine_res(pf, ice_get_avail_txq_count(pf), - ICE_DFLT_QS_PER_VF, ICE_MIN_QS_PER_VF); + min_t(u16, + num_msix_per_vf - ICE_NONQ_VECS_VF, + ICE_MAX_RSS_QS_PER_VF), + ICE_MIN_QS_PER_VF); num_rxq = ice_determine_res(pf, ice_get_avail_rxq_count(pf), - ICE_DFLT_QS_PER_VF, ICE_MIN_QS_PER_VF); - - if (!num_txq || !num_rxq) + min_t(u16, + num_msix_per_vf - ICE_NONQ_VECS_VF, + ICE_MAX_RSS_QS_PER_VF), + ICE_MIN_QS_PER_VF); + + if (!num_txq || !num_rxq) { + dev_err(dev, "Not enough queues to support minimum of %d queue pairs per VF for %d VFs\n", + ICE_MIN_QS_PER_VF, pf->num_alloc_vfs); return -EIO; + } - if (ice_sriov_set_msix_res(pf, num_msix * pf->num_alloc_vfs)) + if (ice_sriov_set_msix_res(pf, num_msix_per_vf * pf->num_alloc_vfs)) { + dev_err(dev, "Unable to set MSI-X resources for %d VFs\n", + pf->num_alloc_vfs); return -EINVAL; + } - /* since AVF driver works with only queue pairs which means, it expects - * to have equal number of Rx and Tx queues, so take the minimum of - * available Tx or Rx queues - */ - pf->num_vf_qps = min_t(int, num_txq, num_rxq); - pf->num_vf_msix = num_msix; + /* only allow equal Tx/Rx queue count (i.e. queue pairs) */ + pf->num_qps_per_vf = min_t(int, num_txq, num_rxq); + pf->num_msix_per_vf = num_msix_per_vf; + dev_info(dev, "Enabling %d VFs with %d vectors and %d queues per VF\n", + pf->num_alloc_vfs, pf->num_msix_per_vf, pf->num_qps_per_vf); return 0; } @@ -943,17 +941,9 @@ static void ice_cleanup_and_realloc_vf(struct ice_vf *vf) /* reallocate VF resources to finish resetting the VSI state */ if (!ice_alloc_vf_res(vf)) { - struct ice_vsi *vsi; - ice_ena_vf_mappings(vf); set_bit(ICE_VF_STATE_ACTIVE, vf->vf_states); clear_bit(ICE_VF_STATE_DIS, vf->vf_states); - - vsi = pf->vsi[vf->lan_vsi_idx]; - if (ice_vsi_add_vlan(vsi, 0)) - dev_warn(ice_pf_to_dev(pf), - "Failed to add VLAN 0 filter for VF %d, MDD events will trigger. Reset the VF, disable spoofchk, or enable 8021q module on the guest", - vf->vf_id); } /* Tell the VF driver the reset is done. This needs to be done only @@ -985,13 +975,13 @@ ice_vf_set_vsi_promisc(struct ice_vf *vf, struct ice_vsi *vsi, u8 promisc_m, if (vsi->num_vlan) { status = ice_set_vlan_vsi_promisc(hw, vsi->idx, promisc_m, rm_promisc); - } else if (vf->port_vlan_id) { + } else if (vf->port_vlan_info) { if (rm_promisc) status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m, - vf->port_vlan_id); + vf->port_vlan_info); else status = ice_set_vsi_promisc(hw, vsi->idx, promisc_m, - vf->port_vlan_id); + vf->port_vlan_info); } else { if (rm_promisc) status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m, @@ -1019,7 +1009,7 @@ static bool ice_config_res_vfs(struct ice_pf *pf) struct ice_hw *hw = &pf->hw; int v; - if (ice_check_avail_res(pf)) { + if (ice_set_per_vf_res(pf)) { dev_err(dev, "Cannot allocate VF resources, try with fewer number of VFs\n"); return false; } @@ -1032,7 +1022,7 @@ static bool ice_config_res_vfs(struct ice_pf *pf) ice_for_each_vf(pf, v) { struct ice_vf *vf = &pf->vf[v]; - vf->num_vf_qs = pf->num_vf_qps; + vf->num_vf_qs = pf->num_qps_per_vf; dev_dbg(dev, "VF-id %d has %d queues configured\n", vf->vf_id, vf->num_vf_qs); ice_cleanup_and_realloc_vf(vf); @@ -1165,9 +1155,10 @@ static bool ice_is_vf_disabled(struct ice_vf *vf) * @vf: pointer to the VF structure * @is_vflr: true if VFLR was issued, false if not * - * Returns true if the VF is reset, false otherwise. + * Returns true if the VF is currently in reset, resets successfully, or resets + * are disabled and false otherwise. */ -static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) +bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) { struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; @@ -1180,6 +1171,12 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) dev = ice_pf_to_dev(pf); + if (test_bit(__ICE_VF_RESETS_DISABLED, pf->state)) { + dev_dbg(dev, "Trying to reset VF %d, but all VF resets are disabled\n", + vf->vf_id); + return true; + } + if (ice_is_vf_disabled(vf)) { dev_dbg(dev, "VF is already disabled, there is no need for resetting it, telling VM, all is fine %d\n", vf->vf_id); @@ -1231,7 +1228,7 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) */ if (test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) || test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states)) { - if (vf->port_vlan_id || vsi->num_vlan) + if (vf->port_vlan_info || vsi->num_vlan) promisc_m = ICE_UCAST_VLAN_PROMISC_BITS; else promisc_m = ICE_UCAST_PROMISC_BITS; @@ -1432,7 +1429,7 @@ static int ice_pci_sriov_ena(struct ice_pf *pf, int num_vfs) if (num_vfs > pf->num_vfs_supported) { dev_err(dev, "Can't enable %d VFs, max VFs supported is %d\n", num_vfs, pf->num_vfs_supported); - return -ENOTSUPP; + return -EOPNOTSUPP; } dev_info(dev, "Allocating %d VFs\n", num_vfs); @@ -1518,6 +1515,72 @@ static void ice_vc_reset_vf(struct ice_vf *vf) } /** + * ice_get_vf_from_pfq - get the VF who owns the PF space queue passed in + * @pf: PF used to index all VFs + * @pfq: queue index relative to the PF's function space + * + * If no VF is found who owns the pfq then return NULL, otherwise return a + * pointer to the VF who owns the pfq + */ +static struct ice_vf *ice_get_vf_from_pfq(struct ice_pf *pf, u16 pfq) +{ + int vf_id; + + ice_for_each_vf(pf, vf_id) { + struct ice_vf *vf = &pf->vf[vf_id]; + struct ice_vsi *vsi; + u16 rxq_idx; + + vsi = pf->vsi[vf->lan_vsi_idx]; + + ice_for_each_rxq(vsi, rxq_idx) + if (vsi->rxq_map[rxq_idx] == pfq) + return vf; + } + + return NULL; +} + +/** + * ice_globalq_to_pfq - convert from global queue index to PF space queue index + * @pf: PF used for conversion + * @globalq: global queue index used to convert to PF space queue index + */ +static u32 ice_globalq_to_pfq(struct ice_pf *pf, u32 globalq) +{ + return globalq - pf->hw.func_caps.common_cap.rxq_first_id; +} + +/** + * ice_vf_lan_overflow_event - handle LAN overflow event for a VF + * @pf: PF that the LAN overflow event happened on + * @event: structure holding the event information for the LAN overflow event + * + * Determine if the LAN overflow event was caused by a VF queue. If it was not + * caused by a VF, do nothing. If a VF caused this LAN overflow event trigger a + * reset on the offending VF. + */ +void +ice_vf_lan_overflow_event(struct ice_pf *pf, struct ice_rq_event_info *event) +{ + u32 gldcb_rtctq, queue; + struct ice_vf *vf; + + gldcb_rtctq = le32_to_cpu(event->desc.params.lan_overflow.prtdcb_ruptq); + dev_dbg(ice_pf_to_dev(pf), "GLDCB_RTCTQ: 0x%08x\n", gldcb_rtctq); + + /* event returns device global Rx queue number */ + queue = (gldcb_rtctq & GLDCB_RTCTQ_RXQNUM_M) >> + GLDCB_RTCTQ_RXQNUM_S; + + vf = ice_get_vf_from_pfq(pf, ice_globalq_to_pfq(pf, queue)); + if (!vf) + return; + + ice_vc_reset_vf(vf); +} + +/** * ice_vc_send_msg_to_vf - Send message to VF * @vf: pointer to the VF info * @v_opcode: virtual channel opcode @@ -1675,7 +1738,7 @@ static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg) vfres->num_vsis = 1; /* Tx and Rx queue are equal for VF */ vfres->num_queue_pairs = vsi->num_txq; - vfres->max_vectors = pf->num_vf_msix; + vfres->max_vectors = pf->num_msix_per_vf; vfres->rss_key_size = ICE_VSIQF_HKEY_ARRAY_SIZE; vfres->rss_lut_size = ICE_VSIQF_HLUT_ARRAY_SIZE; @@ -1981,7 +2044,7 @@ int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena) status = ice_update_vsi(&pf->hw, vf_vsi->idx, ctx, NULL); if (status) { - dev_err(dev, "Failed to %sable spoofchk on VF %d VSI %d\n error %d", + dev_err(dev, "Failed to %sable spoofchk on VF %d VSI %d\n error %d\n", ena ? "en" : "dis", vf->vf_id, vf_vsi->vsi_num, status); ret = -EIO; goto out; @@ -2039,6 +2102,22 @@ error_param: } /** + * ice_vc_validate_vqs_bitmaps - validate Rx/Tx queue bitmaps from VIRTCHNL + * @vqs: virtchnl_queue_select structure containing bitmaps to validate + * + * Return true on successful validation, else false + */ +static bool ice_vc_validate_vqs_bitmaps(struct virtchnl_queue_select *vqs) +{ + if ((!vqs->rx_queues && !vqs->tx_queues) || + vqs->rx_queues >= BIT(ICE_MAX_RSS_QS_PER_VF) || + vqs->tx_queues >= BIT(ICE_MAX_RSS_QS_PER_VF)) + return false; + + return true; +} + +/** * ice_vc_ena_qs_msg * @vf: pointer to the VF info * @msg: pointer to the msg buffer @@ -2065,13 +2144,7 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) goto error_param; } - if (!vqs->rx_queues && !vqs->tx_queues) { - v_ret = VIRTCHNL_STATUS_ERR_PARAM; - goto error_param; - } - - if (vqs->rx_queues > ICE_MAX_BASE_QS_PER_VF || - vqs->tx_queues > ICE_MAX_BASE_QS_PER_VF) { + if (!ice_vc_validate_vqs_bitmaps(vqs)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -2087,7 +2160,7 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) * programmed using ice_vsi_cfg_txqs */ q_map = vqs->rx_queues; - for_each_set_bit(vf_q_id, &q_map, ICE_MAX_BASE_QS_PER_VF) { + for_each_set_bit(vf_q_id, &q_map, ICE_MAX_RSS_QS_PER_VF) { if (!ice_vc_isvalid_q_id(vf, vqs->vsi_id, vf_q_id)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; @@ -2097,7 +2170,7 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) if (test_bit(vf_q_id, vf->rxq_ena)) continue; - if (ice_vsi_ctrl_rx_ring(vsi, true, vf_q_id)) { + if (ice_vsi_ctrl_one_rx_ring(vsi, true, vf_q_id, true)) { dev_err(ice_pf_to_dev(vsi->back), "Failed to enable Rx ring %d on VSI %d\n", vf_q_id, vsi->vsi_num); v_ret = VIRTCHNL_STATUS_ERR_PARAM; @@ -2105,12 +2178,11 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) } set_bit(vf_q_id, vf->rxq_ena); - vf->num_qs_ena++; } vsi = pf->vsi[vf->lan_vsi_idx]; q_map = vqs->tx_queues; - for_each_set_bit(vf_q_id, &q_map, ICE_MAX_BASE_QS_PER_VF) { + for_each_set_bit(vf_q_id, &q_map, ICE_MAX_RSS_QS_PER_VF) { if (!ice_vc_isvalid_q_id(vf, vqs->vsi_id, vf_q_id)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; @@ -2121,7 +2193,6 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) continue; set_bit(vf_q_id, vf->txq_ena); - vf->num_qs_ena++; } /* Set flag to indicate that queues are enabled */ @@ -2163,13 +2234,7 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) goto error_param; } - if (!vqs->rx_queues && !vqs->tx_queues) { - v_ret = VIRTCHNL_STATUS_ERR_PARAM; - goto error_param; - } - - if (vqs->rx_queues > ICE_MAX_BASE_QS_PER_VF || - vqs->tx_queues > ICE_MAX_BASE_QS_PER_VF) { + if (!ice_vc_validate_vqs_bitmaps(vqs)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -2183,7 +2248,7 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) if (vqs->tx_queues) { q_map = vqs->tx_queues; - for_each_set_bit(vf_q_id, &q_map, ICE_MAX_BASE_QS_PER_VF) { + for_each_set_bit(vf_q_id, &q_map, ICE_MAX_RSS_QS_PER_VF) { struct ice_ring *ring = vsi->tx_rings[vf_q_id]; struct ice_txq_meta txq_meta = { 0 }; @@ -2208,14 +2273,23 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) /* Clear enabled queues flag */ clear_bit(vf_q_id, vf->txq_ena); - vf->num_qs_ena--; } } - if (vqs->rx_queues) { - q_map = vqs->rx_queues; + q_map = vqs->rx_queues; + /* speed up Rx queue disable by batching them if possible */ + if (q_map && + bitmap_equal(&q_map, vf->rxq_ena, ICE_MAX_RSS_QS_PER_VF)) { + if (ice_vsi_stop_all_rx_rings(vsi)) { + dev_err(ice_pf_to_dev(vsi->back), "Failed to stop all Rx rings on VSI %d\n", + vsi->vsi_num); + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } - for_each_set_bit(vf_q_id, &q_map, ICE_MAX_BASE_QS_PER_VF) { + bitmap_zero(vf->rxq_ena, ICE_MAX_RSS_QS_PER_VF); + } else if (q_map) { + for_each_set_bit(vf_q_id, &q_map, ICE_MAX_RSS_QS_PER_VF) { if (!ice_vc_isvalid_q_id(vf, vqs->vsi_id, vf_q_id)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; @@ -2225,7 +2299,8 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) if (!test_bit(vf_q_id, vf->rxq_ena)) continue; - if (ice_vsi_ctrl_rx_ring(vsi, false, vf_q_id)) { + if (ice_vsi_ctrl_one_rx_ring(vsi, false, vf_q_id, + true)) { dev_err(ice_pf_to_dev(vsi->back), "Failed to stop Rx ring %d on VSI %d\n", vf_q_id, vsi->vsi_num); v_ret = VIRTCHNL_STATUS_ERR_PARAM; @@ -2234,12 +2309,11 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) /* Clear enabled queues flag */ clear_bit(vf_q_id, vf->rxq_ena); - vf->num_qs_ena--; } } /* Clear enabled queues flag */ - if (v_ret == VIRTCHNL_STATUS_SUCCESS && !vf->num_qs_ena) + if (v_ret == VIRTCHNL_STATUS_SUCCESS && ice_vf_has_no_qs_ena(vf)) clear_bit(ICE_VF_STATE_QS_ENA, vf->vf_states); error_param: @@ -2249,6 +2323,57 @@ error_param: } /** + * ice_cfg_interrupt + * @vf: pointer to the VF info + * @vsi: the VSI being configured + * @vector_id: vector ID + * @map: vector map for mapping vectors to queues + * @q_vector: structure for interrupt vector + * configure the IRQ to queue map + */ +static int +ice_cfg_interrupt(struct ice_vf *vf, struct ice_vsi *vsi, u16 vector_id, + struct virtchnl_vector_map *map, + struct ice_q_vector *q_vector) +{ + u16 vsi_q_id, vsi_q_id_idx; + unsigned long qmap; + + q_vector->num_ring_rx = 0; + q_vector->num_ring_tx = 0; + + qmap = map->rxq_map; + for_each_set_bit(vsi_q_id_idx, &qmap, ICE_MAX_RSS_QS_PER_VF) { + vsi_q_id = vsi_q_id_idx; + + if (!ice_vc_isvalid_q_id(vf, vsi->vsi_num, vsi_q_id)) + return VIRTCHNL_STATUS_ERR_PARAM; + + q_vector->num_ring_rx++; + q_vector->rx.itr_idx = map->rxitr_idx; + vsi->rx_rings[vsi_q_id]->q_vector = q_vector; + ice_cfg_rxq_interrupt(vsi, vsi_q_id, vector_id, + q_vector->rx.itr_idx); + } + + qmap = map->txq_map; + for_each_set_bit(vsi_q_id_idx, &qmap, ICE_MAX_RSS_QS_PER_VF) { + vsi_q_id = vsi_q_id_idx; + + if (!ice_vc_isvalid_q_id(vf, vsi->vsi_num, vsi_q_id)) + return VIRTCHNL_STATUS_ERR_PARAM; + + q_vector->num_ring_tx++; + q_vector->tx.itr_idx = map->txitr_idx; + vsi->tx_rings[vsi_q_id]->q_vector = q_vector; + ice_cfg_txq_interrupt(vsi, vsi_q_id, vector_id, + q_vector->tx.itr_idx); + } + + return VIRTCHNL_STATUS_SUCCESS; +} + +/** * ice_vc_cfg_irq_map_msg * @vf: pointer to the VF info * @msg: pointer to the msg buffer @@ -2258,13 +2383,11 @@ error_param: static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg) { enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; + u16 num_q_vectors_mapped, vsi_id, vector_id; struct virtchnl_irq_map_info *irqmap_info; - u16 vsi_id, vsi_q_id, vector_id; struct virtchnl_vector_map *map; struct ice_pf *pf = vf->pf; - u16 num_q_vectors_mapped; struct ice_vsi *vsi; - unsigned long qmap; int i; irqmap_info = (struct virtchnl_irq_map_info *)msg; @@ -2275,8 +2398,8 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg) * there is actually at least a single VF queue vector mapped */ if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states) || - pf->num_vf_msix < num_q_vectors_mapped || - !irqmap_info->num_vectors) { + pf->num_msix_per_vf < num_q_vectors_mapped || + !num_q_vectors_mapped) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -2297,7 +2420,7 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg) /* vector_id is always 0-based for each VF, and can never be * larger than or equal to the max allowed interrupts per VF */ - if (!(vector_id < ICE_MAX_INTR_PER_VF) || + if (!(vector_id < pf->num_msix_per_vf) || !ice_vc_isvalid_vsi_id(vf, vsi_id) || (!vector_id && (map->rxq_map || map->txq_map))) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; @@ -2318,33 +2441,10 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg) } /* lookout for the invalid queue index */ - qmap = map->rxq_map; - q_vector->num_ring_rx = 0; - for_each_set_bit(vsi_q_id, &qmap, ICE_MAX_BASE_QS_PER_VF) { - if (!ice_vc_isvalid_q_id(vf, vsi_id, vsi_q_id)) { - v_ret = VIRTCHNL_STATUS_ERR_PARAM; - goto error_param; - } - q_vector->num_ring_rx++; - q_vector->rx.itr_idx = map->rxitr_idx; - vsi->rx_rings[vsi_q_id]->q_vector = q_vector; - ice_cfg_rxq_interrupt(vsi, vsi_q_id, vector_id, - q_vector->rx.itr_idx); - } - - qmap = map->txq_map; - q_vector->num_ring_tx = 0; - for_each_set_bit(vsi_q_id, &qmap, ICE_MAX_BASE_QS_PER_VF) { - if (!ice_vc_isvalid_q_id(vf, vsi_id, vsi_q_id)) { - v_ret = VIRTCHNL_STATUS_ERR_PARAM; - goto error_param; - } - q_vector->num_ring_tx++; - q_vector->tx.itr_idx = map->txitr_idx; - vsi->tx_rings[vsi_q_id]->q_vector = q_vector; - ice_cfg_txq_interrupt(vsi, vsi_q_id, vector_id, - q_vector->tx.itr_idx); - } + v_ret = (enum virtchnl_status_code) + ice_cfg_interrupt(vf, vsi, vector_id, map, q_vector); + if (v_ret) + goto error_param; } error_param: @@ -2387,7 +2487,7 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) goto error_param; } - if (qci->num_queue_pairs > ICE_MAX_BASE_QS_PER_VF || + if (qci->num_queue_pairs > ICE_MAX_RSS_QS_PER_VF || qci->num_queue_pairs > min_t(u16, vsi->alloc_txq, vsi->alloc_rxq)) { dev_err(ice_pf_to_dev(pf), "VF-%d requesting more than supported number of queues: %d\n", vf->vf_id, min_t(u16, vsi->alloc_txq, vsi->alloc_rxq)); @@ -2694,16 +2794,16 @@ static int ice_vc_request_qs_msg(struct ice_vf *vf, u8 *msg) if (!req_queues) { dev_err(dev, "VF %d tried to request 0 queues. Ignoring.\n", vf->vf_id); - } else if (req_queues > ICE_MAX_BASE_QS_PER_VF) { + } else if (req_queues > ICE_MAX_RSS_QS_PER_VF) { dev_err(dev, "VF %d tried to request more than %d queues.\n", - vf->vf_id, ICE_MAX_BASE_QS_PER_VF); - vfres->num_queue_pairs = ICE_MAX_BASE_QS_PER_VF; + vf->vf_id, ICE_MAX_RSS_QS_PER_VF); + vfres->num_queue_pairs = ICE_MAX_RSS_QS_PER_VF; } else if (req_queues > cur_queues && req_queues - cur_queues > tx_rx_queue_left) { dev_warn(dev, "VF %d requested %u more queues, but only %u left.\n", vf->vf_id, req_queues - cur_queues, tx_rx_queue_left); vfres->num_queue_pairs = min_t(u16, max_allowed_vf_queues, - ICE_MAX_BASE_QS_PER_VF); + ICE_MAX_RSS_QS_PER_VF); } else { /* request is successful, then reset VF */ vf->num_req_qs = req_queues; @@ -2733,19 +2833,20 @@ int ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, __be16 vlan_proto) { - u16 vlanprio = vlan_id | (qos << ICE_VLAN_PRIORITY_S); struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_vsi *vsi; struct device *dev; struct ice_vf *vf; + u16 vlanprio; int ret; dev = ice_pf_to_dev(pf); if (ice_validate_vf_id(pf, vf_id)) return -EINVAL; - if (vlan_id > ICE_MAX_VLANID || qos > 7) { - dev_err(dev, "Invalid VF Parameters\n"); + if (vlan_id >= VLAN_N_VID || qos > 7) { + dev_err(dev, "Invalid Port VLAN parameters for VF %d, ID %d, QoS %d\n", + vf_id, vlan_id, qos); return -EINVAL; } @@ -2761,40 +2862,52 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, if (ret) return ret; - if (le16_to_cpu(vsi->info.pvid) == vlanprio) { + vlanprio = vlan_id | (qos << VLAN_PRIO_SHIFT); + + if (vf->port_vlan_info == vlanprio) { /* duplicate request, so just return success */ dev_dbg(dev, "Duplicate pvid %d request\n", vlanprio); return 0; } - /* If PVID, then remove all filters on the old VLAN */ - if (vsi->info.pvid) - ice_vsi_kill_vlan(vsi, (le16_to_cpu(vsi->info.pvid) & - VLAN_VID_MASK)); - if (vlan_id || qos) { + /* remove VLAN 0 filter set by default when transitioning from + * no port VLAN to a port VLAN. No change to old port VLAN on + * failure. + */ + ret = ice_vsi_kill_vlan(vsi, 0); + if (ret) + return ret; ret = ice_vsi_manage_pvid(vsi, vlanprio, true); if (ret) return ret; } else { - ice_vsi_manage_pvid(vsi, 0, false); - vsi->info.pvid = 0; + /* add VLAN 0 filter back when transitioning from port VLAN to + * no port VLAN. No change to old port VLAN on failure. + */ + ret = ice_vsi_add_vlan(vsi, 0); + if (ret) + return ret; + ret = ice_vsi_manage_pvid(vsi, 0, false); + if (ret) + return ret; } if (vlan_id) { dev_info(dev, "Setting VLAN %d, QoS 0x%x on VF %d\n", vlan_id, qos, vf_id); - /* add new VLAN filter for each MAC */ + /* add VLAN filter for the port VLAN */ ret = ice_vsi_add_vlan(vsi, vlan_id); if (ret) return ret; } + /* remove old port VLAN filter with valid VLAN ID or QoS fields */ + if (vf->port_vlan_info) + ice_vsi_kill_vlan(vsi, vf->port_vlan_info & VLAN_VID_MASK); - /* The Port VLAN needs to be saved across resets the same as the - * default LAN MAC address. - */ - vf->port_vlan_id = le16_to_cpu(vsi->info.pvid); + /* keep port VLAN information persistent on resets */ + vf->port_vlan_info = le16_to_cpu(vsi->info.pvid); return 0; } @@ -2849,7 +2962,7 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) } for (i = 0; i < vfl->num_elements; i++) { - if (vfl->vlan_id[i] > ICE_MAX_VLANID) { + if (vfl->vlan_id[i] >= VLAN_N_VID) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; dev_err(dev, "invalid VF VLAN id %d\n", vfl->vlan_id[i]); @@ -2911,9 +3024,9 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) goto error_param; } - vsi->num_vlan++; - /* Enable VLAN pruning when VLAN is added */ - if (!vlan_promisc) { + /* Enable VLAN pruning when non-zero VLAN is added */ + if (!vlan_promisc && vid && + !ice_vsi_is_vlan_pruning_ena(vsi)) { status = ice_cfg_vlan_pruning(vsi, true, false); if (status) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; @@ -2921,7 +3034,7 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) vid, status); goto error_param; } - } else { + } else if (vlan_promisc) { /* Enable Ucast/Mcast VLAN promiscuous mode */ promisc_m = ICE_PROMISC_VLAN_TX | ICE_PROMISC_VLAN_RX; @@ -2965,9 +3078,9 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) goto error_param; } - vsi->num_vlan--; - /* Disable VLAN pruning when the last VLAN is removed */ - if (!vsi->num_vlan) + /* Disable VLAN pruning when only VLAN 0 is left */ + if (vsi->num_vlan == 1 && + ice_vsi_is_vlan_pruning_ena(vsi)) ice_cfg_vlan_pruning(vsi, false, false); /* Disable Unicast/Multicast VLAN promiscuous mode */ @@ -3246,14 +3359,12 @@ int ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi) { struct ice_pf *pf = ice_netdev_to_pf(netdev); - struct ice_vsi *vsi; struct ice_vf *vf; if (ice_validate_vf_id(pf, vf_id)) return -EINVAL; vf = &pf->vf[vf_id]; - vsi = pf->vsi[vf->lan_vsi_idx]; if (ice_check_vf_init(pf, vf)) return -EBUSY; @@ -3262,9 +3373,8 @@ ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi) ether_addr_copy(ivi->mac, vf->dflt_lan_addr.addr); /* VF configuration for VLAN and applicable QoS */ - ivi->vlan = le16_to_cpu(vsi->info.pvid) & ICE_VLAN_M; - ivi->qos = (le16_to_cpu(vsi->info.pvid) & ICE_PRIORITY_M) >> - ICE_VLAN_PRIORITY_S; + ivi->vlan = vf->port_vlan_info & VLAN_VID_MASK; + ivi->qos = (vf->port_vlan_info & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT; ivi->trusted = vf->trusted; ivi->spoofchk = vf->spoofchk; @@ -3442,3 +3552,52 @@ int ice_get_vf_stats(struct net_device *netdev, int vf_id, return 0; } + +/** + * ice_print_vfs_mdd_event - print VFs malicious driver detect event + * @pf: pointer to the PF structure + * + * Called from ice_handle_mdd_event to rate limit and print VFs MDD events. + */ +void ice_print_vfs_mdd_events(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + struct ice_hw *hw = &pf->hw; + int i; + + /* check that there are pending MDD events to print */ + if (!test_and_clear_bit(__ICE_MDD_VF_PRINT_PENDING, pf->state)) + return; + + /* VF MDD event logs are rate limited to one second intervals */ + if (time_is_after_jiffies(pf->last_printed_mdd_jiffies + HZ * 1)) + return; + + pf->last_printed_mdd_jiffies = jiffies; + + ice_for_each_vf(pf, i) { + struct ice_vf *vf = &pf->vf[i]; + + /* only print Rx MDD event message if there are new events */ + if (vf->mdd_rx_events.count != vf->mdd_rx_events.last_printed) { + vf->mdd_rx_events.last_printed = + vf->mdd_rx_events.count; + + dev_info(dev, "%d Rx Malicious Driver Detection events detected on PF %d VF %d MAC %pM. mdd-auto-reset-vfs=%s\n", + vf->mdd_rx_events.count, hw->pf_id, i, + vf->dflt_lan_addr.addr, + test_bit(ICE_FLAG_MDD_AUTO_RESET_VF, pf->flags) + ? "on" : "off"); + } + + /* only print Tx MDD event message if there are new events */ + if (vf->mdd_tx_events.count != vf->mdd_tx_events.last_printed) { + vf->mdd_tx_events.last_printed = + vf->mdd_tx_events.count; + + dev_info(dev, "%d Tx Malicious Driver Detection events detected on PF %d VF %d MAC %pM.\n", + vf->mdd_tx_events.count, hw->pf_id, i, + vf->dflt_lan_addr.addr); + } + } +} diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h index ac67982751df..3f9464269bd2 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h @@ -5,11 +5,6 @@ #define _ICE_VIRTCHNL_PF_H_ #include "ice.h" -#define ICE_MAX_VLANID 4095 -#define ICE_VLAN_PRIORITY_S 12 -#define ICE_VLAN_M 0xFFF -#define ICE_PRIORITY_M 0x7000 - /* Restrict number of MAC Addr and VLAN that non-trusted VF can programmed */ #define ICE_MAX_VLAN_PER_VF 8 #define ICE_MAX_MACADDR_PER_VF 12 @@ -26,18 +21,15 @@ #define ICE_PCI_CIAD_WAIT_COUNT 100 #define ICE_PCI_CIAD_WAIT_DELAY_US 1 -/* VF resources default values and limitation */ +/* VF resource constraints */ #define ICE_MAX_VF_COUNT 256 -#define ICE_MAX_QS_PER_VF 256 #define ICE_MIN_QS_PER_VF 1 -#define ICE_DFLT_QS_PER_VF 4 #define ICE_NONQ_VECS_VF 1 #define ICE_MAX_SCATTER_QS_PER_VF 16 -#define ICE_MAX_BASE_QS_PER_VF 16 -#define ICE_MAX_INTR_PER_VF 65 -#define ICE_MAX_POLICY_INTR_PER_VF 33 +#define ICE_MAX_RSS_QS_PER_VF 16 +#define ICE_NUM_VF_MSIX_MED 17 +#define ICE_NUM_VF_MSIX_SMALL 5 #define ICE_MIN_INTR_PER_VF (ICE_MIN_QS_PER_VF + 1) -#define ICE_DFLT_INTR_PER_VF (ICE_DFLT_QS_PER_VF + 1) #define ICE_MAX_VF_RESET_TRIES 40 #define ICE_MAX_VF_RESET_SLEEP_MS 20 @@ -61,6 +53,13 @@ enum ice_virtchnl_cap { ICE_VIRTCHNL_VF_CAP_PRIVILEGE, }; +/* VF MDD events print structure */ +struct ice_mdd_vf_events { + u16 count; /* total count of Rx|Tx events */ + /* count number of the last printed event */ + u16 last_printed; +}; + /* VF information structure */ struct ice_vf { struct ice_pf *pf; @@ -73,9 +72,9 @@ struct ice_vf { struct virtchnl_version_info vf_ver; u32 driver_caps; /* reported by VF driver */ struct virtchnl_ether_addr dflt_lan_addr; - DECLARE_BITMAP(txq_ena, ICE_MAX_BASE_QS_PER_VF); - DECLARE_BITMAP(rxq_ena, ICE_MAX_BASE_QS_PER_VF); - u16 port_vlan_id; + DECLARE_BITMAP(txq_ena, ICE_MAX_RSS_QS_PER_VF); + DECLARE_BITMAP(rxq_ena, ICE_MAX_RSS_QS_PER_VF); + u16 port_vlan_info; /* Port VLAN ID and QoS */ u8 pf_set_mac:1; /* VF MAC address set by VMM admin */ u8 trusted:1; u8 spoofchk:1; @@ -89,14 +88,14 @@ struct ice_vf { unsigned int tx_rate; /* Tx bandwidth limit in Mbps */ DECLARE_BITMAP(vf_states, ICE_VF_STATES_NBITS); /* VF runtime states */ - u64 num_mdd_events; /* number of MDD events detected */ u64 num_inval_msgs; /* number of continuous invalid msgs */ u64 num_valid_msgs; /* number of valid msgs detected */ unsigned long vf_caps; /* VF's adv. capabilities */ u8 num_req_qs; /* num of queue pairs requested by VF */ u16 num_mac; u16 num_vf_qs; /* num of queue configured per VF */ - u16 num_qs_ena; /* total num of Tx/Rx queue enabled */ + struct ice_mdd_vf_events mdd_rx_events; + struct ice_mdd_vf_events mdd_tx_events; }; #ifdef CONFIG_PCI_IOV @@ -111,6 +110,7 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event); void ice_vc_notify_link_state(struct ice_pf *pf); void ice_vc_notify_reset(struct ice_pf *pf); bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr); +bool ice_reset_vf(struct ice_vf *vf, bool is_vflr); int ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, @@ -128,6 +128,9 @@ void ice_set_vf_state_qs_dis(struct ice_vf *vf); int ice_get_vf_stats(struct net_device *netdev, int vf_id, struct ifla_vf_stats *vf_stats); +void +ice_vf_lan_overflow_event(struct ice_pf *pf, struct ice_rq_event_info *event); +void ice_print_vfs_mdd_events(struct ice_pf *pf); #else /* CONFIG_PCI_IOV */ #define ice_process_vflr_event(pf) do {} while (0) #define ice_free_vfs(pf) do {} while (0) @@ -135,6 +138,8 @@ ice_get_vf_stats(struct net_device *netdev, int vf_id, #define ice_vc_notify_link_state(pf) do {} while (0) #define ice_vc_notify_reset(pf) do {} while (0) #define ice_set_vf_state_qs_dis(vf) do {} while (0) +#define ice_vf_lan_overflow_event(pf, event) do {} while (0) +#define ice_print_vfs_mdd_events(pf) do {} while (0) static inline bool ice_reset_all_vfs(struct ice_pf __always_unused *pf, @@ -143,6 +148,12 @@ ice_reset_all_vfs(struct ice_pf __always_unused *pf, return true; } +static inline bool +ice_reset_vf(struct ice_vf __always_unused *vf, bool __always_unused is_vflr) +{ + return true; +} + static inline int ice_sriov_configure(struct pci_dev __always_unused *pdev, int __always_unused num_vfs) diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 4d3407bbd4c4..8279db15e870 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -183,7 +183,7 @@ static int ice_qp_dis(struct ice_vsi *vsi, u16 q_idx) if (err) return err; } - err = ice_vsi_ctrl_rx_ring(vsi, false, q_idx); + err = ice_vsi_ctrl_one_rx_ring(vsi, false, q_idx, true); if (err) return err; @@ -243,7 +243,7 @@ static int ice_qp_ena(struct ice_vsi *vsi, u16 q_idx) ice_qvec_cfg_msix(vsi, q_vector); - err = ice_vsi_ctrl_rx_ring(vsi, true, q_idx); + err = ice_vsi_ctrl_one_rx_ring(vsi, true, q_idx, true); if (err) goto free_buf; @@ -457,7 +457,7 @@ int ice_xsk_umem_setup(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid) if (if_running) { ret = ice_qp_dis(vsi, qid); if (ret) { - netdev_err(vsi->netdev, "ice_qp_dis error = %d", ret); + netdev_err(vsi->netdev, "ice_qp_dis error = %d\n", ret); goto xsk_umem_if_up; } } @@ -471,11 +471,11 @@ xsk_umem_if_up: if (!ret && umem_present) napi_schedule(&vsi->xdp_rings[qid]->q_vector->napi); else if (ret) - netdev_err(vsi->netdev, "ice_qp_ena error = %d", ret); + netdev_err(vsi->netdev, "ice_qp_ena error = %d\n", ret); } if (umem_failure) { - netdev_err(vsi->netdev, "Could not %sable UMEM, error = %d", + netdev_err(vsi->netdev, "Could not %sable UMEM, error = %d\n", umem_present ? "en" : "dis", umem_failure); return umem_failure; } @@ -609,7 +609,7 @@ ice_alloc_buf_slow_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf) */ static bool ice_alloc_rx_bufs_zc(struct ice_ring *rx_ring, int count, - bool alloc(struct ice_ring *, struct ice_rx_buf *)) + bool (*alloc)(struct ice_ring *, struct ice_rx_buf *)) { union ice_32b_rx_flex_desc *rx_desc; u16 ntu = rx_ring->next_to_use; @@ -816,10 +816,10 @@ ice_run_xdp_zc(struct ice_ring *rx_ring, struct xdp_buff *xdp) break; default: bpf_warn_invalid_xdp_action(act); - /* fallthrough -- not supported action */ + fallthrough; case XDP_ABORTED: trace_xdp_exception(rx_ring->netdev, xdp_prog, act); - /* fallthrough -- handle aborts by dropping frame */ + fallthrough; case XDP_DROP: result = ICE_XDP_CONSUMED; break; @@ -841,8 +841,8 @@ int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget) unsigned int total_rx_bytes = 0, total_rx_packets = 0; u16 cleaned_count = ICE_DESC_UNUSED(rx_ring); unsigned int xdp_xmit = 0; + bool failure = false; struct xdp_buff xdp; - bool failure = 0; xdp.rxq = &rx_ring->xdp_rxq; @@ -937,6 +937,15 @@ int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget) ice_finalize_xdp_rx(rx_ring, xdp_xmit); ice_update_rx_ring_stats(rx_ring, total_rx_packets, total_rx_bytes); + if (xsk_umem_uses_need_wakeup(rx_ring->xsk_umem)) { + if (failure || rx_ring->next_to_clean == rx_ring->next_to_use) + xsk_set_rx_need_wakeup(rx_ring->xsk_umem); + else + xsk_clear_rx_need_wakeup(rx_ring->xsk_umem); + + return (int)total_rx_packets; + } + return failure ? budget : (int)total_rx_packets; } @@ -988,6 +997,8 @@ static bool ice_xmit_zc(struct ice_ring *xdp_ring, int budget) if (tx_desc) { ice_xdp_ring_update_tail(xdp_ring); xsk_umem_consume_tx_done(xdp_ring->xsk_umem); + if (xsk_umem_uses_need_wakeup(xdp_ring->xsk_umem)) + xsk_clear_tx_need_wakeup(xdp_ring->xsk_umem); } return budget > 0 && work_done; @@ -1063,6 +1074,13 @@ bool ice_clean_tx_irq_zc(struct ice_ring *xdp_ring, int budget) if (xsk_frames) xsk_umem_complete_tx(xdp_ring->xsk_umem, xsk_frames); + if (xsk_umem_uses_need_wakeup(xdp_ring->xsk_umem)) { + if (xdp_ring->next_to_clean == xdp_ring->next_to_use) + xsk_set_tx_need_wakeup(xdp_ring->xsk_umem); + else + xsk_clear_tx_need_wakeup(xdp_ring->xsk_umem); + } + ice_update_tx_ring_stats(xdp_ring, total_packets, total_bytes); xmit_done = ice_xmit_zc(xdp_ring, ICE_DFLT_IRQ_WORK); diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.h b/drivers/net/ethernet/intel/ice/ice_xsk.h index 3479e1de98fe..8a4ba7c6d549 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.h +++ b/drivers/net/ethernet/intel/ice/ice_xsk.h @@ -24,7 +24,7 @@ ice_xsk_umem_setup(struct ice_vsi __always_unused *vsi, struct xdp_umem __always_unused *umem, u16 __always_unused qid) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline void @@ -63,7 +63,7 @@ static inline int ice_xsk_wakeup(struct net_device __always_unused *netdev, u32 __always_unused queue_id, u32 __always_unused flags) { - return -ENOTSUPP; + return -EOPNOTSUPP; } #define ice_xsk_clean_rx_ring(rx_ring) do {} while (0) diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h index 49b5fa9d4783..0c9282e2aaec 100644 --- a/drivers/net/ethernet/intel/igb/igb.h +++ b/drivers/net/ethernet/intel/igb/igb.h @@ -306,7 +306,7 @@ struct igb_q_vector { char name[IFNAMSIZ + 9]; /* for dynamic allocation of rings associated with this q_vector */ - struct igb_ring ring[0] ____cacheline_internodealigned_in_smp; + struct igb_ring ring[] ____cacheline_internodealigned_in_smp; }; enum e1000_ring_flags_t { diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c index f96ffa83efbe..39d3b76a6f5d 100644 --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c @@ -2183,27 +2183,6 @@ static int igb_set_coalesce(struct net_device *netdev, struct igb_adapter *adapter = netdev_priv(netdev); int i; - if (ec->rx_max_coalesced_frames || - ec->rx_coalesce_usecs_irq || - ec->rx_max_coalesced_frames_irq || - ec->tx_max_coalesced_frames || - ec->tx_coalesce_usecs_irq || - ec->stats_block_coalesce_usecs || - ec->use_adaptive_rx_coalesce || - ec->use_adaptive_tx_coalesce || - ec->pkt_rate_low || - ec->rx_coalesce_usecs_low || - ec->rx_max_coalesced_frames_low || - ec->tx_coalesce_usecs_low || - ec->tx_max_coalesced_frames_low || - ec->pkt_rate_high || - ec->rx_coalesce_usecs_high || - ec->rx_max_coalesced_frames_high || - ec->tx_coalesce_usecs_high || - ec->tx_max_coalesced_frames_high || - ec->rate_sample_interval) - return -ENOTSUPP; - if ((ec->rx_coalesce_usecs > IGB_MAX_ITR_USECS) || ((ec->rx_coalesce_usecs > 3) && (ec->rx_coalesce_usecs < IGB_MIN_ITR_USECS)) || @@ -3477,6 +3456,7 @@ static int igb_set_priv_flags(struct net_device *netdev, u32 priv_flags) } static const struct ethtool_ops igb_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_drvinfo = igb_get_drvinfo, .get_regs_len = igb_get_regs_len, .get_regs = igb_get_regs, diff --git a/drivers/net/ethernet/intel/igb/igb_hwmon.c b/drivers/net/ethernet/intel/igb/igb_hwmon.c index 3b83747b2700..21a29a0ca7f4 100644 --- a/drivers/net/ethernet/intel/igb/igb_hwmon.c +++ b/drivers/net/ethernet/intel/igb/igb_hwmon.c @@ -198,11 +198,11 @@ int igb_sysfs_init(struct igb_adapter *adapter) } /* init i2c_client */ - client = i2c_new_device(&adapter->i2c_adap, &i350_sensor_info); - if (client == NULL) { + client = i2c_new_client_device(&adapter->i2c_adap, &i350_sensor_info); + if (IS_ERR(client)) { dev_info(&adapter->pdev->dev, "Failed to create new i2c device.\n"); - rc = -ENODEV; + rc = PTR_ERR(client); goto exit; } adapter->i2c_client = client; diff --git a/drivers/net/ethernet/intel/igbvf/ethtool.c b/drivers/net/ethernet/intel/igbvf/ethtool.c index 3ae358b35227..9217d150e286 100644 --- a/drivers/net/ethernet/intel/igbvf/ethtool.c +++ b/drivers/net/ethernet/intel/igbvf/ethtool.c @@ -424,6 +424,7 @@ static void igbvf_get_strings(struct net_device *netdev, u32 stringset, } static const struct ethtool_ops igbvf_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = igbvf_get_drvinfo, .get_regs_len = igbvf_get_regs_len, .get_regs = igbvf_get_regs, diff --git a/drivers/net/ethernet/intel/igc/Makefile b/drivers/net/ethernet/intel/igc/Makefile index 49fb1e1965cd..e3c164c12e10 100644 --- a/drivers/net/ethernet/intel/igc/Makefile +++ b/drivers/net/ethernet/intel/igc/Makefile @@ -8,4 +8,4 @@ obj-$(CONFIG_IGC) += igc.o igc-objs := igc_main.o igc_mac.o igc_i225.o igc_base.o igc_nvm.o igc_phy.o \ -igc_ethtool.o igc_ptp.o +igc_ethtool.o igc_ptp.o igc_dump.o diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h index 52066bdbbad0..a1f845a2aa80 100644 --- a/drivers/net/ethernet/intel/igc/igc.h +++ b/drivers/net/ethernet/intel/igc/igc.h @@ -42,6 +42,10 @@ int igc_del_mac_steering_filter(struct igc_adapter *adapter, const u8 *addr, u8 queue, u8 flags); void igc_update_stats(struct igc_adapter *adapter); +/* igc_dump declarations */ +void igc_rings_dump(struct igc_adapter *adapter); +void igc_regs_dump(struct igc_adapter *adapter); + extern char igc_driver_name[]; extern char igc_driver_version[]; @@ -53,10 +57,13 @@ extern char igc_driver_version[]; /* Interrupt defines */ #define IGC_START_ITR 648 /* ~6000 ints/sec */ + +/* Flags definitions */ #define IGC_FLAG_HAS_MSI BIT(0) #define IGC_FLAG_QUEUE_PAIRS BIT(3) #define IGC_FLAG_DMAC BIT(4) #define IGC_FLAG_PTP BIT(8) +#define IGC_FLAG_WOL_SUPPORTED BIT(8) #define IGC_FLAG_NEED_LINK_UPDATE BIT(9) #define IGC_FLAG_MEDIA_RESET BIT(10) #define IGC_FLAG_MAS_ENABLE BIT(12) @@ -108,7 +115,7 @@ extern char igc_driver_version[]; #define IGC_RX_HDR_LEN IGC_RXBUFFER_256 /* Transmit and receive latency (for PTP timestamps) */ -/* FIXME: These values were estimated using the ones that i210 has as +/* FIXME: These values were estimated using the ones that i225 has as * basis, they seem to provide good numbers with ptp4l/phc2sys, but we * need to confirm them. */ @@ -319,7 +326,7 @@ struct igc_q_vector { struct net_device poll_dev; /* for dynamic allocation of rings associated with this q_vector */ - struct igc_ring ring[0] ____cacheline_internodealigned_in_smp; + struct igc_ring ring[] ____cacheline_internodealigned_in_smp; }; #define MAX_ETYPE_FILTER (4 - 1) @@ -552,6 +559,7 @@ int igc_erase_filter(struct igc_adapter *adapter, void igc_ptp_init(struct igc_adapter *adapter); void igc_ptp_reset(struct igc_adapter *adapter); +void igc_ptp_suspend(struct igc_adapter *adapter); void igc_ptp_stop(struct igc_adapter *adapter); void igc_ptp_rx_rgtstamp(struct igc_q_vector *q_vector, struct sk_buff *skb); void igc_ptp_rx_pktstamp(struct igc_q_vector *q_vector, void *va, diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h index 58efa7a02c68..4ddccccf42cc 100644 --- a/drivers/net/ethernet/intel/igc/igc_defines.h +++ b/drivers/net/ethernet/intel/igc/igc_defines.h @@ -16,7 +16,10 @@ /* Wake Up Filter Control */ #define IGC_WUFC_LNKC 0x00000001 /* Link Status Change Wakeup Enable */ +#define IGC_WUFC_MAG 0x00000002 /* Magic Packet Wakeup Enable */ +#define IGC_WUFC_EX 0x00000004 /* Directed Exact Wakeup Enable */ #define IGC_WUFC_MC 0x00000008 /* Directed Multicast Wakeup Enable */ +#define IGC_WUFC_BC 0x00000010 /* Broadcast Wakeup Enable */ #define IGC_CTRL_ADVD3WUC 0x00100000 /* D3 WUC */ @@ -259,6 +262,9 @@ #define IGC_GPIE_EIAME 0x40000000 #define IGC_GPIE_PBA 0x80000000 +/* Receive Descriptor bit definitions */ +#define IGC_RXD_STAT_DD 0x01 /* Descriptor Done */ + /* Transmit Descriptor bit definitions */ #define IGC_TXD_DTYP_D 0x00100000 /* Data Descriptor */ #define IGC_TXD_DTYP_C 0x00000000 /* Context Descriptor */ diff --git a/drivers/net/ethernet/intel/igc/igc_dump.c b/drivers/net/ethernet/intel/igc/igc_dump.c new file mode 100644 index 000000000000..657ab50ae296 --- /dev/null +++ b/drivers/net/ethernet/intel/igc/igc_dump.c @@ -0,0 +1,323 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 Intel Corporation */ + +#include "igc.h" + +struct igc_reg_info { + u32 ofs; + char *name; +}; + +static const struct igc_reg_info igc_reg_info_tbl[] = { + /* General Registers */ + {IGC_CTRL, "CTRL"}, + {IGC_STATUS, "STATUS"}, + {IGC_CTRL_EXT, "CTRL_EXT"}, + {IGC_MDIC, "MDIC"}, + + /* Interrupt Registers */ + {IGC_ICR, "ICR"}, + + /* RX Registers */ + {IGC_RCTL, "RCTL"}, + {IGC_RDLEN(0), "RDLEN"}, + {IGC_RDH(0), "RDH"}, + {IGC_RDT(0), "RDT"}, + {IGC_RXDCTL(0), "RXDCTL"}, + {IGC_RDBAL(0), "RDBAL"}, + {IGC_RDBAH(0), "RDBAH"}, + + /* TX Registers */ + {IGC_TCTL, "TCTL"}, + {IGC_TDBAL(0), "TDBAL"}, + {IGC_TDBAH(0), "TDBAH"}, + {IGC_TDLEN(0), "TDLEN"}, + {IGC_TDH(0), "TDH"}, + {IGC_TDT(0), "TDT"}, + {IGC_TXDCTL(0), "TXDCTL"}, + {IGC_TDFH, "TDFH"}, + {IGC_TDFT, "TDFT"}, + {IGC_TDFHS, "TDFHS"}, + {IGC_TDFPC, "TDFPC"}, + + /* List Terminator */ + {} +}; + +/* igc_regdump - register printout routine */ +static void igc_regdump(struct igc_hw *hw, struct igc_reg_info *reginfo) +{ + int n = 0; + char rname[16]; + u32 regs[8]; + + switch (reginfo->ofs) { + case IGC_RDLEN(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_RDLEN(n)); + break; + case IGC_RDH(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_RDH(n)); + break; + case IGC_RDT(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_RDT(n)); + break; + case IGC_RXDCTL(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_RXDCTL(n)); + break; + case IGC_RDBAL(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_RDBAL(n)); + break; + case IGC_RDBAH(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_RDBAH(n)); + break; + case IGC_TDBAL(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_RDBAL(n)); + break; + case IGC_TDBAH(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_TDBAH(n)); + break; + case IGC_TDLEN(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_TDLEN(n)); + break; + case IGC_TDH(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_TDH(n)); + break; + case IGC_TDT(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_TDT(n)); + break; + case IGC_TXDCTL(0): + for (n = 0; n < 4; n++) + regs[n] = rd32(IGC_TXDCTL(n)); + break; + default: + pr_info("%-15s %08x\n", reginfo->name, rd32(reginfo->ofs)); + return; + } + + snprintf(rname, 16, "%s%s", reginfo->name, "[0-3]"); + pr_info("%-15s %08x %08x %08x %08x\n", rname, regs[0], regs[1], + regs[2], regs[3]); +} + +/* igc_rings_dump - Tx-rings and Rx-rings */ +void igc_rings_dump(struct igc_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + struct my_u0 { u64 a; u64 b; } *u0; + union igc_adv_tx_desc *tx_desc; + union igc_adv_rx_desc *rx_desc; + struct igc_ring *tx_ring; + struct igc_ring *rx_ring; + u32 staterr; + u16 i, n; + + if (!netif_msg_hw(adapter)) + return; + + /* Print netdevice Info */ + if (netdev) { + dev_info(&adapter->pdev->dev, "Net device Info\n"); + pr_info("Device Name state trans_start\n"); + pr_info("%-15s %016lX %016lX\n", netdev->name, + netdev->state, dev_trans_start(netdev)); + } + + /* Print TX Ring Summary */ + if (!netdev || !netif_running(netdev)) + goto exit; + + dev_info(&adapter->pdev->dev, "TX Rings Summary\n"); + pr_info("Queue [NTU] [NTC] [bi(ntc)->dma ] leng ntw timestamp\n"); + for (n = 0; n < adapter->num_tx_queues; n++) { + struct igc_tx_buffer *buffer_info; + + tx_ring = adapter->tx_ring[n]; + buffer_info = &tx_ring->tx_buffer_info[tx_ring->next_to_clean]; + + pr_info(" %5d %5X %5X %016llX %04X %p %016llX\n", + n, tx_ring->next_to_use, tx_ring->next_to_clean, + (u64)dma_unmap_addr(buffer_info, dma), + dma_unmap_len(buffer_info, len), + buffer_info->next_to_watch, + (u64)buffer_info->time_stamp); + } + + /* Print TX Rings */ + if (!netif_msg_tx_done(adapter)) + goto rx_ring_summary; + + dev_info(&adapter->pdev->dev, "TX Rings Dump\n"); + + /* Transmit Descriptor Formats + * + * Advanced Transmit Descriptor + * +--------------------------------------------------------------+ + * 0 | Buffer Address [63:0] | + * +--------------------------------------------------------------+ + * 8 | PAYLEN | PORTS |CC|IDX | STA | DCMD |DTYP|MAC|RSV| DTALEN | + * +--------------------------------------------------------------+ + * 63 46 45 40 39 38 36 35 32 31 24 15 0 + */ + + for (n = 0; n < adapter->num_tx_queues; n++) { + tx_ring = adapter->tx_ring[n]; + pr_info("------------------------------------\n"); + pr_info("TX QUEUE INDEX = %d\n", tx_ring->queue_index); + pr_info("------------------------------------\n"); + pr_info("T [desc] [address 63:0 ] [PlPOCIStDDM Ln] [bi->dma ] leng ntw timestamp bi->skb\n"); + + for (i = 0; tx_ring->desc && (i < tx_ring->count); i++) { + const char *next_desc; + struct igc_tx_buffer *buffer_info; + + tx_desc = IGC_TX_DESC(tx_ring, i); + buffer_info = &tx_ring->tx_buffer_info[i]; + u0 = (struct my_u0 *)tx_desc; + if (i == tx_ring->next_to_use && + i == tx_ring->next_to_clean) + next_desc = " NTC/U"; + else if (i == tx_ring->next_to_use) + next_desc = " NTU"; + else if (i == tx_ring->next_to_clean) + next_desc = " NTC"; + else + next_desc = ""; + + pr_info("T [0x%03X] %016llX %016llX %016llX %04X %p %016llX %p%s\n", + i, le64_to_cpu(u0->a), + le64_to_cpu(u0->b), + (u64)dma_unmap_addr(buffer_info, dma), + dma_unmap_len(buffer_info, len), + buffer_info->next_to_watch, + (u64)buffer_info->time_stamp, + buffer_info->skb, next_desc); + + if (netif_msg_pktdata(adapter) && buffer_info->skb) + print_hex_dump(KERN_INFO, "", + DUMP_PREFIX_ADDRESS, + 16, 1, buffer_info->skb->data, + dma_unmap_len(buffer_info, len), + true); + } + } + + /* Print RX Rings Summary */ +rx_ring_summary: + dev_info(&adapter->pdev->dev, "RX Rings Summary\n"); + pr_info("Queue [NTU] [NTC]\n"); + for (n = 0; n < adapter->num_rx_queues; n++) { + rx_ring = adapter->rx_ring[n]; + pr_info(" %5d %5X %5X\n", + n, rx_ring->next_to_use, rx_ring->next_to_clean); + } + + /* Print RX Rings */ + if (!netif_msg_rx_status(adapter)) + goto exit; + + dev_info(&adapter->pdev->dev, "RX Rings Dump\n"); + + /* Advanced Receive Descriptor (Read) Format + * 63 1 0 + * +-----------------------------------------------------+ + * 0 | Packet Buffer Address [63:1] |A0/NSE| + * +----------------------------------------------+------+ + * 8 | Header Buffer Address [63:1] | DD | + * +-----------------------------------------------------+ + * + * + * Advanced Receive Descriptor (Write-Back) Format + * + * 63 48 47 32 31 30 21 20 17 16 4 3 0 + * +------------------------------------------------------+ + * 0 | Packet IP |SPH| HDR_LEN | RSV|Packet| RSS | + * | Checksum Ident | | | | Type | Type | + * +------------------------------------------------------+ + * 8 | VLAN Tag | Length | Extended Error | Extended Status | + * +------------------------------------------------------+ + * 63 48 47 32 31 20 19 0 + */ + + for (n = 0; n < adapter->num_rx_queues; n++) { + rx_ring = adapter->rx_ring[n]; + pr_info("------------------------------------\n"); + pr_info("RX QUEUE INDEX = %d\n", rx_ring->queue_index); + pr_info("------------------------------------\n"); + pr_info("R [desc] [ PktBuf A0] [ HeadBuf DD] [bi->dma ] [bi->skb] <-- Adv Rx Read format\n"); + pr_info("RWB[desc] [PcsmIpSHl PtRs] [vl er S cks ln] ---------------- [bi->skb] <-- Adv Rx Write-Back format\n"); + + for (i = 0; i < rx_ring->count; i++) { + const char *next_desc; + struct igc_rx_buffer *buffer_info; + + buffer_info = &rx_ring->rx_buffer_info[i]; + rx_desc = IGC_RX_DESC(rx_ring, i); + u0 = (struct my_u0 *)rx_desc; + staterr = le32_to_cpu(rx_desc->wb.upper.status_error); + + if (i == rx_ring->next_to_use) + next_desc = " NTU"; + else if (i == rx_ring->next_to_clean) + next_desc = " NTC"; + else + next_desc = ""; + + if (staterr & IGC_RXD_STAT_DD) { + /* Descriptor Done */ + pr_info("%s[0x%03X] %016llX %016llX ---------------- %s\n", + "RWB", i, + le64_to_cpu(u0->a), + le64_to_cpu(u0->b), + next_desc); + } else { + pr_info("%s[0x%03X] %016llX %016llX %016llX %s\n", + "R ", i, + le64_to_cpu(u0->a), + le64_to_cpu(u0->b), + (u64)buffer_info->dma, + next_desc); + + if (netif_msg_pktdata(adapter) && + buffer_info->dma && buffer_info->page) { + print_hex_dump(KERN_INFO, "", + DUMP_PREFIX_ADDRESS, + 16, 1, + page_address + (buffer_info->page) + + buffer_info->page_offset, + igc_rx_bufsz(rx_ring), + true); + } + } + } + } + +exit: + return; +} + +/* igc_regs_dump - registers dump */ +void igc_regs_dump(struct igc_adapter *adapter) +{ + struct igc_hw *hw = &adapter->hw; + struct igc_reg_info *reginfo; + + /* Print Registers */ + dev_info(&adapter->pdev->dev, "Register Dump\n"); + pr_info(" Register Name Value\n"); + for (reginfo = (struct igc_reg_info *)igc_reg_info_tbl; + reginfo->name; reginfo++) { + igc_regdump(hw, reginfo); + } +} diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c index ee07011e13e9..f530fc29b074 100644 --- a/drivers/net/ethernet/intel/igc/igc_ethtool.c +++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c @@ -308,6 +308,65 @@ static void igc_get_regs(struct net_device *netdev, regs_buff[168 + i] = rd32(IGC_TXDCTL(i)); } +static void igc_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) +{ + struct igc_adapter *adapter = netdev_priv(netdev); + + wol->wolopts = 0; + + if (!(adapter->flags & IGC_FLAG_WOL_SUPPORTED)) + return; + + wol->supported = WAKE_UCAST | WAKE_MCAST | + WAKE_BCAST | WAKE_MAGIC | + WAKE_PHY; + + /* apply any specific unsupported masks here */ + switch (adapter->hw.device_id) { + default: + break; + } + + if (adapter->wol & IGC_WUFC_EX) + wol->wolopts |= WAKE_UCAST; + if (adapter->wol & IGC_WUFC_MC) + wol->wolopts |= WAKE_MCAST; + if (adapter->wol & IGC_WUFC_BC) + wol->wolopts |= WAKE_BCAST; + if (adapter->wol & IGC_WUFC_MAG) + wol->wolopts |= WAKE_MAGIC; + if (adapter->wol & IGC_WUFC_LNKC) + wol->wolopts |= WAKE_PHY; +} + +static int igc_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) +{ + struct igc_adapter *adapter = netdev_priv(netdev); + + if (wol->wolopts & (WAKE_ARP | WAKE_MAGICSECURE | WAKE_FILTER)) + return -EOPNOTSUPP; + + if (!(adapter->flags & IGC_FLAG_WOL_SUPPORTED)) + return wol->wolopts ? -EOPNOTSUPP : 0; + + /* these settings will always override what we currently have */ + adapter->wol = 0; + + if (wol->wolopts & WAKE_UCAST) + adapter->wol |= IGC_WUFC_EX; + if (wol->wolopts & WAKE_MCAST) + adapter->wol |= IGC_WUFC_MC; + if (wol->wolopts & WAKE_BCAST) + adapter->wol |= IGC_WUFC_BC; + if (wol->wolopts & WAKE_MAGIC) + adapter->wol |= IGC_WUFC_MAG; + if (wol->wolopts & WAKE_PHY) + adapter->wol |= IGC_WUFC_LNKC; + device_set_wakeup_enable(&adapter->pdev->dev, adapter->wol); + + return 0; +} + static u32 igc_get_msglevel(struct net_device *netdev) { struct igc_adapter *adapter = netdev_priv(netdev); @@ -802,27 +861,6 @@ static int igc_set_coalesce(struct net_device *netdev, struct igc_adapter *adapter = netdev_priv(netdev); int i; - if (ec->rx_max_coalesced_frames || - ec->rx_coalesce_usecs_irq || - ec->rx_max_coalesced_frames_irq || - ec->tx_max_coalesced_frames || - ec->tx_coalesce_usecs_irq || - ec->stats_block_coalesce_usecs || - ec->use_adaptive_rx_coalesce || - ec->use_adaptive_tx_coalesce || - ec->pkt_rate_low || - ec->rx_coalesce_usecs_low || - ec->rx_max_coalesced_frames_low || - ec->tx_coalesce_usecs_low || - ec->tx_max_coalesced_frames_low || - ec->pkt_rate_high || - ec->rx_coalesce_usecs_high || - ec->rx_max_coalesced_frames_high || - ec->tx_coalesce_usecs_high || - ec->tx_max_coalesced_frames_high || - ec->rate_sample_interval) - return -ENOTSUPP; - if (ec->rx_coalesce_usecs > IGC_MAX_ITR_USECS || (ec->rx_coalesce_usecs > 3 && ec->rx_coalesce_usecs < IGC_MIN_ITR_USECS) || @@ -1856,9 +1894,12 @@ static int igc_set_link_ksettings(struct net_device *netdev, } static const struct ethtool_ops igc_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_drvinfo = igc_get_drvinfo, .get_regs_len = igc_get_regs_len, .get_regs = igc_get_regs, + .get_wol = igc_get_wol, + .set_wol = igc_set_wol, .get_msglevel = igc_get_msglevel, .set_msglevel = igc_set_msglevel, .nway_reset = igc_nway_reset, diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index d9d5425fe8d9..69fa1ce1f927 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -3546,6 +3546,8 @@ static void igc_reset_task(struct work_struct *work) adapter = container_of(work, struct igc_adapter, reset_task); + igc_rings_dump(adapter); + igc_regs_dump(adapter); netdev_err(adapter->netdev, "Reset adapter\n"); igc_reinit_locked(adapter); } @@ -4029,6 +4031,9 @@ static void igc_watchdog_task(struct work_struct *work) } } if (link) { + /* Cancel scheduled suspend requests. */ + pm_runtime_resume(netdev->dev.parent); + if (!netif_carrier_ok(netdev)) { u32 ctrl; @@ -4114,6 +4119,8 @@ no_wait: return; } } + pm_schedule_suspend(netdev->dev.parent, + MSEC_PER_SEC * 5); /* also check for alternate media here */ } else if (!netif_carrier_ok(netdev) && @@ -4337,6 +4344,7 @@ request_done: static int __igc_open(struct net_device *netdev, bool resuming) { struct igc_adapter *adapter = netdev_priv(netdev); + struct pci_dev *pdev = adapter->pdev; struct igc_hw *hw = &adapter->hw; int err = 0; int i = 0; @@ -4348,6 +4356,9 @@ static int __igc_open(struct net_device *netdev, bool resuming) return -EBUSY; } + if (!resuming) + pm_runtime_get_sync(&pdev->dev); + netif_carrier_off(netdev); /* allocate transmit descriptors */ @@ -4386,6 +4397,9 @@ static int __igc_open(struct net_device *netdev, bool resuming) rd32(IGC_ICR); igc_irq_enable(adapter); + if (!resuming) + pm_runtime_put(&pdev->dev); + netif_tx_start_all_queues(netdev); /* start the watchdog. */ @@ -4404,6 +4418,8 @@ err_setup_rx: igc_free_all_tx_resources(adapter); err_setup_tx: igc_reset(adapter); + if (!resuming) + pm_runtime_put(&pdev->dev); return err; } @@ -4428,9 +4444,13 @@ static int igc_open(struct net_device *netdev) static int __igc_close(struct net_device *netdev, bool suspending) { struct igc_adapter *adapter = netdev_priv(netdev); + struct pci_dev *pdev = adapter->pdev; WARN_ON(test_bit(__IGC_RESETTING, &adapter->state)); + if (!suspending) + pm_runtime_get_sync(&pdev->dev); + igc_down(adapter); igc_release_hw_control(adapter); @@ -4440,6 +4460,9 @@ static int __igc_close(struct net_device *netdev, bool suspending) igc_free_all_tx_resources(adapter); igc_free_all_rx_resources(adapter); + if (!suspending) + pm_runtime_put_sync(&pdev->dev); + return 0; } @@ -4766,6 +4789,16 @@ static int igc_probe(struct pci_dev *pdev, hw->fc.requested_mode = igc_fc_default; hw->fc.current_mode = igc_fc_default; + /* By default, support wake on port A */ + adapter->flags |= IGC_FLAG_WOL_SUPPORTED; + + /* initialize the wol settings based on the eeprom settings */ + if (adapter->flags & IGC_FLAG_WOL_SUPPORTED) + adapter->wol |= IGC_WUFC_MAG; + + device_set_wakeup_enable(&adapter->pdev->dev, + adapter->flags & IGC_FLAG_WOL_SUPPORTED); + /* reset the hardware with the new settings */ igc_reset(adapter); @@ -4792,6 +4825,10 @@ static int igc_probe(struct pci_dev *pdev, pcie_print_link_status(pdev); netdev_info(netdev, "MAC: %pM\n", netdev->dev_addr); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + + pm_runtime_put_noidle(&pdev->dev); + return 0; err_register: @@ -4826,6 +4863,8 @@ static void igc_remove(struct pci_dev *pdev) struct net_device *netdev = pci_get_drvdata(pdev); struct igc_adapter *adapter = netdev_priv(netdev); + pm_runtime_get_noresume(&pdev->dev); + igc_ptp_stop(adapter); set_bit(__IGC_DOWN, &adapter->state); @@ -4870,6 +4909,8 @@ static int __igc_shutdown(struct pci_dev *pdev, bool *enable_wake, if (netif_running(netdev)) __igc_close(netdev, true); + igc_ptp_suspend(adapter); + igc_clear_interrupt_scheme(adapter); rtnl_unlock(); @@ -5045,6 +5086,108 @@ static void igc_shutdown(struct pci_dev *pdev) } } +/** + * igc_io_error_detected - called when PCI error is detected + * @pdev: Pointer to PCI device + * @state: The current PCI connection state + * + * This function is called after a PCI bus error affecting + * this device has been detected. + **/ +static pci_ers_result_t igc_io_error_detected(struct pci_dev *pdev, + pci_channel_state_t state) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igc_adapter *adapter = netdev_priv(netdev); + + netif_device_detach(netdev); + + if (state == pci_channel_io_perm_failure) + return PCI_ERS_RESULT_DISCONNECT; + + if (netif_running(netdev)) + igc_down(adapter); + pci_disable_device(pdev); + + /* Request a slot reset. */ + return PCI_ERS_RESULT_NEED_RESET; +} + +/** + * igc_io_slot_reset - called after the PCI bus has been reset. + * @pdev: Pointer to PCI device + * + * Restart the card from scratch, as if from a cold-boot. Implementation + * resembles the first-half of the igc_resume routine. + **/ +static pci_ers_result_t igc_io_slot_reset(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igc_adapter *adapter = netdev_priv(netdev); + struct igc_hw *hw = &adapter->hw; + pci_ers_result_t result; + + if (pci_enable_device_mem(pdev)) { + dev_err(&pdev->dev, + "Could not re-enable PCI device after reset.\n"); + result = PCI_ERS_RESULT_DISCONNECT; + } else { + pci_set_master(pdev); + pci_restore_state(pdev); + pci_save_state(pdev); + + pci_enable_wake(pdev, PCI_D3hot, 0); + pci_enable_wake(pdev, PCI_D3cold, 0); + + /* In case of PCI error, adapter loses its HW address + * so we should re-assign it here. + */ + hw->hw_addr = adapter->io_addr; + + igc_reset(adapter); + wr32(IGC_WUS, ~0); + result = PCI_ERS_RESULT_RECOVERED; + } + + return result; +} + +/** + * igc_io_resume - called when traffic can start to flow again. + * @pdev: Pointer to PCI device + * + * This callback is called when the error recovery driver tells us that + * its OK to resume normal operation. Implementation resembles the + * second-half of the igc_resume routine. + */ +static void igc_io_resume(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igc_adapter *adapter = netdev_priv(netdev); + + rtnl_lock(); + if (netif_running(netdev)) { + if (igc_open(netdev)) { + dev_err(&pdev->dev, "igc_open failed after reset\n"); + return; + } + } + + netif_device_attach(netdev); + + /* let the f/w know that the h/w is now under the control of the + * driver. + */ + igc_get_hw_control(adapter); + rtnl_unlock(); +} + +static const struct pci_error_handlers igc_err_handler = { + .error_detected = igc_io_error_detected, + .slot_reset = igc_io_slot_reset, + .resume = igc_io_resume, +}; + #ifdef CONFIG_PM static const struct dev_pm_ops igc_pm_ops = { SET_SYSTEM_SLEEP_PM_OPS(igc_suspend, igc_resume) @@ -5062,6 +5205,7 @@ static struct pci_driver igc_driver = { .driver.pm = &igc_pm_ops, #endif .shutdown = igc_shutdown, + .err_handler = &igc_err_handler, }; /** diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c index 693506587198..f99c514ad0f4 100644 --- a/drivers/net/ethernet/intel/igc/igc_ptp.c +++ b/drivers/net/ethernet/intel/igc/igc_ptp.c @@ -509,7 +509,7 @@ static void igc_ptp_tx_hwtstamp(struct igc_adapter *adapter) * This work function polls the TSYNCTXCTL valid bit to determine when a * timestamp has been taken for the current stored skb. */ -void igc_ptp_tx_work(struct work_struct *work) +static void igc_ptp_tx_work(struct work_struct *work) { struct igc_adapter *adapter = container_of(work, struct igc_adapter, ptp_tx_work); diff --git a/drivers/net/ethernet/intel/igc/igc_regs.h b/drivers/net/ethernet/intel/igc/igc_regs.h index c9029b549b90..d4af53a80f11 100644 --- a/drivers/net/ethernet/intel/igc/igc_regs.h +++ b/drivers/net/ethernet/intel/igc/igc_regs.h @@ -17,6 +17,11 @@ /* Internal Packet Buffer Size Registers */ #define IGC_RXPBS 0x02404 /* Rx Packet Buffer Size - RW */ #define IGC_TXPBS 0x03404 /* Tx Packet Buffer Size - RW */ +#define IGC_TDFH 0x03410 /* Tx Data FIFO Head - RW */ +#define IGC_TDFT 0x03418 /* Tx Data FIFO Tail - RW */ +#define IGC_TDFHS 0x03420 /* Tx Data FIFO Head Saved - RW */ +#define IGC_TDFTS 0x03428 /* Tx Data FIFO Tail Saved - RW */ +#define IGC_TDFPC 0x03430 /* Tx Data FIFO Packet Count - RW */ /* NVM Register Descriptions */ #define IGC_EERD 0x12014 /* EEprom mode read - RW */ diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h index 39e73ad60352..2833e4f041ce 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h @@ -462,7 +462,7 @@ struct ixgbe_q_vector { char name[IFNAMSIZ + 9]; /* for dynamic allocation of rings associated with this q_vector */ - struct ixgbe_ring ring[0] ____cacheline_internodealigned_in_smp; + struct ixgbe_ring ring[] ____cacheline_internodealigned_in_smp; }; #ifdef CONFIG_IXGBE_HWMON diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 7c52ae8ac005..c6bf0a50ee63 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -3444,6 +3444,7 @@ static int ixgbe_set_priv_flags(struct net_device *netdev, u32 priv_flags) } static const struct ethtool_ops ixgbe_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_drvinfo = ixgbe_get_drvinfo, .get_regs_len = ixgbe_get_regs_len, .get_regs = ixgbe_get_regs, diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c index ccd852ad62a4..ec7a11d13fdc 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c @@ -968,8 +968,7 @@ int ixgbe_fcoe_get_hbainfo(struct net_device *netdev, { struct ixgbe_adapter *adapter = netdev_priv(netdev); struct ixgbe_hw *hw = &adapter->hw; - int i, pos; - u8 buf[8]; + u64 dsn; if (!info) return -EINVAL; @@ -985,17 +984,11 @@ int ixgbe_fcoe_get_hbainfo(struct net_device *netdev, /* Serial Number */ /* Get the PCI-e Device Serial Number Capability */ - pos = pci_find_ext_capability(adapter->pdev, PCI_EXT_CAP_ID_DSN); - if (pos) { - pos += 4; - for (i = 0; i < 8; i++) - pci_read_config_byte(adapter->pdev, pos + i, &buf[i]); - + dsn = pci_get_dsn(adapter->pdev); + if (dsn) snprintf(info->serial_number, sizeof(info->serial_number), - "%02X%02X%02X%02X%02X%02X%02X%02X", - buf[7], buf[6], buf[5], buf[4], - buf[3], buf[2], buf[1], buf[0]); - } else + "%016llX", dsn); + else snprintf(info->serial_number, sizeof(info->serial_number), "Unknown"); diff --git a/drivers/net/ethernet/intel/ixgbevf/ethtool.c b/drivers/net/ethernet/intel/ixgbevf/ethtool.c index f7f309c96fa8..988fa49fa99a 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ethtool.c +++ b/drivers/net/ethernet/intel/ixgbevf/ethtool.c @@ -968,6 +968,7 @@ static int ixgbevf_set_priv_flags(struct net_device *netdev, u32 priv_flags) } static const struct ethtool_ops ixgbevf_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_drvinfo = ixgbevf_get_drvinfo, .get_regs_len = ixgbevf_get_regs_len, .get_regs = ixgbevf_get_regs, diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c index 2e4975572e9f..c97c74164c73 100644 --- a/drivers/net/ethernet/jme.c +++ b/drivers/net/ethernet/jme.c @@ -2077,12 +2077,7 @@ jme_tx_tso(struct sk_buff *skb, __le16 *mss, u8 *flags) IPPROTO_TCP, 0); } else { - struct ipv6hdr *ip6h = ipv6_hdr(skb); - - tcp_hdr(skb)->check = ~csum_ipv6_magic(&ip6h->saddr, - &ip6h->daddr, 0, - IPPROTO_TCP, - 0); + tcp_v6_gso_csum_prep(skb); } return 0; @@ -2844,6 +2839,9 @@ jme_set_eeprom(struct net_device *netdev, } static const struct ethtool_ops jme_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_drvinfo = jme_get_drvinfo, .get_regs_len = jme_get_regs_len, .get_regs = jme_get_regs, diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c index 3c8125cbc84d..81d24481b22c 100644 --- a/drivers/net/ethernet/marvell/mv643xx_eth.c +++ b/drivers/net/ethernet/marvell/mv643xx_eth.c @@ -1737,6 +1737,7 @@ static int mv643xx_eth_get_sset_count(struct net_device *dev, int sset) } static const struct ethtool_ops mv643xx_eth_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_drvinfo = mv643xx_eth_get_drvinfo, .nway_reset = phy_ethtool_nway_reset, .get_link = ethtool_op_get_link, diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 11babc79dc6c..5be61f73b6ab 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -341,6 +341,13 @@ enum { ETHTOOL_STAT_EEE_WAKEUP, ETHTOOL_STAT_SKB_ALLOC_ERR, ETHTOOL_STAT_REFILL_ERR, + ETHTOOL_XDP_REDIRECT, + ETHTOOL_XDP_PASS, + ETHTOOL_XDP_DROP, + ETHTOOL_XDP_TX, + ETHTOOL_XDP_TX_ERR, + ETHTOOL_XDP_XMIT, + ETHTOOL_XDP_XMIT_ERR, ETHTOOL_MAX_STATS, }; @@ -354,10 +361,10 @@ struct mvneta_statistic { #define T_REG_64 64 #define T_SW 1 -#define MVNETA_XDP_PASS BIT(0) -#define MVNETA_XDP_DROPPED BIT(1) -#define MVNETA_XDP_TX BIT(2) -#define MVNETA_XDP_REDIR BIT(3) +#define MVNETA_XDP_PASS 0 +#define MVNETA_XDP_DROPPED BIT(0) +#define MVNETA_XDP_TX BIT(1) +#define MVNETA_XDP_REDIR BIT(2) static const struct mvneta_statistic mvneta_statistics[] = { { 0x3000, T_REG_64, "good_octets_received", }, @@ -395,16 +402,42 @@ static const struct mvneta_statistic mvneta_statistics[] = { { ETHTOOL_STAT_EEE_WAKEUP, T_SW, "eee_wakeup_errors", }, { ETHTOOL_STAT_SKB_ALLOC_ERR, T_SW, "skb_alloc_errors", }, { ETHTOOL_STAT_REFILL_ERR, T_SW, "refill_errors", }, + { ETHTOOL_XDP_REDIRECT, T_SW, "rx_xdp_redirect", }, + { ETHTOOL_XDP_PASS, T_SW, "rx_xdp_pass", }, + { ETHTOOL_XDP_DROP, T_SW, "rx_xdp_drop", }, + { ETHTOOL_XDP_TX, T_SW, "rx_xdp_tx", }, + { ETHTOOL_XDP_TX_ERR, T_SW, "rx_xdp_tx_errors", }, + { ETHTOOL_XDP_XMIT, T_SW, "tx_xdp_xmit", }, + { ETHTOOL_XDP_XMIT_ERR, T_SW, "tx_xdp_xmit_errors", }, }; -struct mvneta_pcpu_stats { - struct u64_stats_sync syncp; +struct mvneta_stats { u64 rx_packets; u64 rx_bytes; - u64 rx_dropped; - u64 rx_errors; u64 tx_packets; u64 tx_bytes; + /* xdp */ + u64 xdp_redirect; + u64 xdp_pass; + u64 xdp_drop; + u64 xdp_xmit; + u64 xdp_xmit_err; + u64 xdp_tx; + u64 xdp_tx_err; +}; + +struct mvneta_ethtool_stats { + struct mvneta_stats ps; + u64 skb_alloc_error; + u64 refill_error; +}; + +struct mvneta_pcpu_stats { + struct u64_stats_sync syncp; + + struct mvneta_ethtool_stats es; + u64 rx_dropped; + u64 rx_errors; }; struct mvneta_pcpu_port { @@ -660,10 +693,6 @@ struct mvneta_rx_queue { /* pointer to uncomplete skb buffer */ struct sk_buff *skb; int left_size; - - /* error counters */ - u32 skb_alloc_err; - u32 refill_err; }; static enum cpuhp_state online_hpstate; @@ -748,12 +777,12 @@ mvneta_get_stats64(struct net_device *dev, cpu_stats = per_cpu_ptr(pp->stats, cpu); do { start = u64_stats_fetch_begin_irq(&cpu_stats->syncp); - rx_packets = cpu_stats->rx_packets; - rx_bytes = cpu_stats->rx_bytes; + rx_packets = cpu_stats->es.ps.rx_packets; + rx_bytes = cpu_stats->es.ps.rx_bytes; rx_dropped = cpu_stats->rx_dropped; rx_errors = cpu_stats->rx_errors; - tx_packets = cpu_stats->tx_packets; - tx_bytes = cpu_stats->tx_bytes; + tx_packets = cpu_stats->es.ps.tx_packets; + tx_bytes = cpu_stats->es.ps.tx_bytes; } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); stats->rx_packets += rx_packets; @@ -1933,7 +1962,7 @@ static void mvneta_rxq_drop_pkts(struct mvneta_port *pp, if (!data || !(rx_desc->buf_phys_addr)) continue; - page_pool_put_page(rxq->page_pool, data, false); + page_pool_put_full_page(rxq->page_pool, data, false); } if (xdp_rxq_info_is_reg(&rxq->xdp_rxq)) xdp_rxq_info_unreg(&rxq->xdp_rxq); @@ -1942,19 +1971,18 @@ static void mvneta_rxq_drop_pkts(struct mvneta_port *pp, } static void -mvneta_update_stats(struct mvneta_port *pp, u32 pkts, - u32 len, bool tx) +mvneta_update_stats(struct mvneta_port *pp, + struct mvneta_stats *ps) { struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats); u64_stats_update_begin(&stats->syncp); - if (tx) { - stats->tx_packets += pkts; - stats->tx_bytes += len; - } else { - stats->rx_packets += pkts; - stats->rx_bytes += len; - } + stats->es.ps.rx_packets += ps->rx_packets; + stats->es.ps.rx_bytes += ps->rx_bytes; + /* xdp */ + stats->es.ps.xdp_redirect += ps->xdp_redirect; + stats->es.ps.xdp_pass += ps->xdp_pass; + stats->es.ps.xdp_drop += ps->xdp_drop; u64_stats_update_end(&stats->syncp); } @@ -1969,9 +1997,15 @@ int mvneta_rx_refill_queue(struct mvneta_port *pp, struct mvneta_rx_queue *rxq) rx_desc = rxq->descs + curr_desc; if (!(rx_desc->buf_phys_addr)) { if (mvneta_rx_refill(pp, rx_desc, rxq, GFP_ATOMIC)) { + struct mvneta_pcpu_stats *stats; + pr_err("Can't refill queue %d. Done %d from %d\n", rxq->id, i, rxq->refill_num); - rxq->refill_err++; + + stats = this_cpu_ptr(pp->stats); + u64_stats_update_begin(&stats->syncp); + stats->es.refill_error++; + u64_stats_update_end(&stats->syncp); break; } } @@ -2021,7 +2055,6 @@ mvneta_xdp_submit_frame(struct mvneta_port *pp, struct mvneta_tx_queue *txq, tx_desc->buf_phys_addr = dma_addr; tx_desc->data_size = xdpf->len; - mvneta_update_stats(pp, 1, xdpf->len, true); mvneta_txq_inc_put(txq); txq->pending++; txq->count++; @@ -2032,6 +2065,7 @@ mvneta_xdp_submit_frame(struct mvneta_port *pp, struct mvneta_tx_queue *txq, static int mvneta_xdp_xmit_back(struct mvneta_port *pp, struct xdp_buff *xdp) { + struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats); struct mvneta_tx_queue *txq; struct netdev_queue *nq; struct xdp_frame *xdpf; @@ -2048,8 +2082,19 @@ mvneta_xdp_xmit_back(struct mvneta_port *pp, struct xdp_buff *xdp) __netif_tx_lock(nq, cpu); ret = mvneta_xdp_submit_frame(pp, txq, xdpf, false); - if (ret == MVNETA_XDP_TX) + if (ret == MVNETA_XDP_TX) { + u64_stats_update_begin(&stats->syncp); + stats->es.ps.tx_bytes += xdpf->len; + stats->es.ps.tx_packets++; + stats->es.ps.xdp_tx++; + u64_stats_update_end(&stats->syncp); + mvneta_txq_pend_desc_add(pp, txq, 0); + } else { + u64_stats_update_begin(&stats->syncp); + stats->es.ps.xdp_tx_err++; + u64_stats_update_end(&stats->syncp); + } __netif_tx_unlock(nq); return ret; @@ -2060,10 +2105,11 @@ mvneta_xdp_xmit(struct net_device *dev, int num_frame, struct xdp_frame **frames, u32 flags) { struct mvneta_port *pp = netdev_priv(dev); + struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats); + int i, nxmit_byte = 0, nxmit = num_frame; int cpu = smp_processor_id(); struct mvneta_tx_queue *txq; struct netdev_queue *nq; - int i, drops = 0; u32 ret; if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) @@ -2075,9 +2121,11 @@ mvneta_xdp_xmit(struct net_device *dev, int num_frame, __netif_tx_lock(nq, cpu); for (i = 0; i < num_frame; i++) { ret = mvneta_xdp_submit_frame(pp, txq, frames[i], true); - if (ret != MVNETA_XDP_TX) { + if (ret == MVNETA_XDP_TX) { + nxmit_byte += frames[i]->len; + } else { xdp_return_frame_rx_napi(frames[i]); - drops++; + nxmit--; } } @@ -2085,12 +2133,20 @@ mvneta_xdp_xmit(struct net_device *dev, int num_frame, mvneta_txq_pend_desc_add(pp, txq, 0); __netif_tx_unlock(nq); - return num_frame - drops; + u64_stats_update_begin(&stats->syncp); + stats->es.ps.tx_bytes += nxmit_byte; + stats->es.ps.tx_packets += nxmit; + stats->es.ps.xdp_xmit += nxmit; + stats->es.ps.xdp_xmit_err += num_frame - nxmit; + u64_stats_update_end(&stats->syncp); + + return nxmit; } static int mvneta_run_xdp(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, - struct bpf_prog *prog, struct xdp_buff *xdp) + struct bpf_prog *prog, struct xdp_buff *xdp, + struct mvneta_stats *stats) { unsigned int len; u32 ret, act; @@ -2100,28 +2156,29 @@ mvneta_run_xdp(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, switch (act) { case XDP_PASS: - ret = MVNETA_XDP_PASS; - break; + stats->xdp_pass++; + return MVNETA_XDP_PASS; case XDP_REDIRECT: { int err; err = xdp_do_redirect(pp->dev, xdp, prog); - if (err) { + if (unlikely(err)) { ret = MVNETA_XDP_DROPPED; - __page_pool_put_page(rxq->page_pool, - virt_to_head_page(xdp->data), - len, true); + page_pool_put_page(rxq->page_pool, + virt_to_head_page(xdp->data), len, + true); } else { ret = MVNETA_XDP_REDIR; + stats->xdp_redirect++; } break; } case XDP_TX: ret = mvneta_xdp_xmit_back(pp, xdp); if (ret != MVNETA_XDP_TX) - __page_pool_put_page(rxq->page_pool, - virt_to_head_page(xdp->data), - len, true); + page_pool_put_page(rxq->page_pool, + virt_to_head_page(xdp->data), len, + true); break; default: bpf_warn_invalid_xdp_action(act); @@ -2130,13 +2187,16 @@ mvneta_run_xdp(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, trace_xdp_exception(pp->dev, prog, act); /* fall through */ case XDP_DROP: - __page_pool_put_page(rxq->page_pool, - virt_to_head_page(xdp->data), - len, true); + page_pool_put_page(rxq->page_pool, + virt_to_head_page(xdp->data), len, true); ret = MVNETA_XDP_DROPPED; + stats->xdp_drop++; break; } + stats->rx_bytes += xdp->data_end - xdp->data; + stats->rx_packets++; + return ret; } @@ -2146,12 +2206,14 @@ mvneta_swbm_rx_frame(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, struct xdp_buff *xdp, struct bpf_prog *xdp_prog, - struct page *page, u32 *xdp_ret) + struct page *page, + struct mvneta_stats *stats) { unsigned char *data = page_address(page); int data_len = -MVNETA_MH_SIZE, len; struct net_device *dev = pp->dev; enum dma_data_direction dma_dir; + int ret = 0; if (MVNETA_SKB_SIZE(rx_desc->data_size) > PAGE_SIZE) { len = MVNETA_MAX_RX_BUF_SIZE; @@ -2175,17 +2237,9 @@ mvneta_swbm_rx_frame(struct mvneta_port *pp, xdp_set_data_meta_invalid(xdp); if (xdp_prog) { - u32 ret; - - ret = mvneta_run_xdp(pp, rxq, xdp_prog, xdp); - if (ret != MVNETA_XDP_PASS) { - mvneta_update_stats(pp, 1, - xdp->data_end - xdp->data, - false); - rx_desc->buf_phys_addr = 0; - *xdp_ret |= ret; - return ret; - } + ret = mvneta_run_xdp(pp, rxq, xdp_prog, xdp, stats); + if (ret) + goto out; } rxq->skb = build_skb(xdp->data_hard_start, PAGE_SIZE); @@ -2193,9 +2247,9 @@ mvneta_swbm_rx_frame(struct mvneta_port *pp, struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats); netdev_err(dev, "Can't allocate skb on queue %d\n", rxq->id); - rxq->skb_alloc_err++; u64_stats_update_begin(&stats->syncp); + stats->es.skb_alloc_error++; stats->rx_dropped++; u64_stats_update_end(&stats->syncp); @@ -2209,9 +2263,11 @@ mvneta_swbm_rx_frame(struct mvneta_port *pp, mvneta_rx_csum(pp, rx_desc->status, rxq->skb); rxq->left_size = rx_desc->data_size - len; + +out: rx_desc->buf_phys_addr = 0; - return 0; + return ret; } static void @@ -2252,12 +2308,11 @@ static int mvneta_rx_swbm(struct napi_struct *napi, struct mvneta_port *pp, int budget, struct mvneta_rx_queue *rxq) { - int rcvd_pkts = 0, rcvd_bytes = 0, rx_proc = 0; + int rx_proc = 0, rx_todo, refill; struct net_device *dev = pp->dev; + struct mvneta_stats ps = {}; struct bpf_prog *xdp_prog; struct xdp_buff xdp_buf; - int rx_todo, refill; - u32 xdp_ret = 0; /* Get number of received packets */ rx_todo = mvneta_rxq_busy_desc_num_get(pp, rxq); @@ -2290,7 +2345,7 @@ static int mvneta_rx_swbm(struct napi_struct *napi, } err = mvneta_swbm_rx_frame(pp, rx_desc, rxq, &xdp_buf, - xdp_prog, page, &xdp_ret); + xdp_prog, page, &ps); if (err) continue; } else { @@ -2314,8 +2369,9 @@ static int mvneta_rx_swbm(struct napi_struct *napi, rxq->skb = NULL; continue; } - rcvd_pkts++; - rcvd_bytes += rxq->skb->len; + + ps.rx_bytes += rxq->skb->len; + ps.rx_packets++; /* Linux processing */ rxq->skb->protocol = eth_type_trans(rxq->skb, dev); @@ -2327,11 +2383,11 @@ static int mvneta_rx_swbm(struct napi_struct *napi, } rcu_read_unlock(); - if (xdp_ret & MVNETA_XDP_REDIR) + if (ps.xdp_redirect) xdp_do_flush_map(); - if (rcvd_pkts) - mvneta_update_stats(pp, rcvd_pkts, rcvd_bytes, false); + if (ps.rx_packets) + mvneta_update_stats(pp, &ps); /* return some buffers to hardware queue, one at a time is too slow */ refill = mvneta_rx_refill_queue(pp, rxq); @@ -2339,7 +2395,7 @@ static int mvneta_rx_swbm(struct napi_struct *napi, /* Update rxq management counters */ mvneta_rxq_desc_num_update(pp, rxq, rx_proc, refill); - return rcvd_pkts; + return ps.rx_packets; } /* Main rx processing when using hardware buffer management */ @@ -2423,8 +2479,15 @@ err_drop_frame: /* Refill processing */ err = hwbm_pool_refill(&bm_pool->hwbm_pool, GFP_ATOMIC); if (err) { + struct mvneta_pcpu_stats *stats; + netdev_err(dev, "Linux processing - Can't refill\n"); - rxq->refill_err++; + + stats = this_cpu_ptr(pp->stats); + u64_stats_update_begin(&stats->syncp); + stats->es.refill_error++; + u64_stats_update_end(&stats->syncp); + goto err_drop_frame_ret_pool; } @@ -2454,8 +2517,14 @@ err_drop_frame: napi_gro_receive(napi, skb); } - if (rcvd_pkts) - mvneta_update_stats(pp, rcvd_pkts, rcvd_bytes, false); + if (rcvd_pkts) { + struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats); + + u64_stats_update_begin(&stats->syncp); + stats->es.ps.rx_packets += rcvd_pkts; + stats->es.ps.rx_bytes += rcvd_bytes; + u64_stats_update_end(&stats->syncp); + } /* Update rxq management counters */ mvneta_rxq_desc_num_update(pp, rxq, rx_done, rx_done); @@ -2711,6 +2780,7 @@ static netdev_tx_t mvneta_tx(struct sk_buff *skb, struct net_device *dev) out: if (frags > 0) { struct netdev_queue *nq = netdev_get_tx_queue(dev, txq_id); + struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats); netdev_tx_sent_queue(nq, len); @@ -2724,7 +2794,10 @@ out: else txq->pending += frags; - mvneta_update_stats(pp, 1, len, true); + u64_stats_update_begin(&stats->syncp); + stats->es.ps.tx_bytes += len; + stats->es.ps.tx_packets++; + u64_stats_update_end(&stats->syncp); } else { dev->stats.tx_dropped++; dev_kfree_skb_any(skb); @@ -3766,13 +3839,9 @@ static void mvneta_mac_config(struct phylink_config *config, unsigned int mode, new_clk = gmac_clk & ~MVNETA_GMAC_1MS_CLOCK_ENABLE; new_an = gmac_an & ~(MVNETA_GMAC_INBAND_AN_ENABLE | MVNETA_GMAC_INBAND_RESTART_AN | - MVNETA_GMAC_CONFIG_MII_SPEED | - MVNETA_GMAC_CONFIG_GMII_SPEED | MVNETA_GMAC_AN_SPEED_EN | MVNETA_GMAC_ADVERT_SYM_FLOW_CTRL | - MVNETA_GMAC_CONFIG_FLOW_CTRL | MVNETA_GMAC_AN_FLOW_CTRL_EN | - MVNETA_GMAC_CONFIG_FULL_DUPLEX | MVNETA_GMAC_AN_DUPLEX_EN); /* Even though it might look weird, when we're configured in @@ -3787,24 +3856,20 @@ static void mvneta_mac_config(struct phylink_config *config, unsigned int mode, if (phylink_test(state->advertising, Pause)) new_an |= MVNETA_GMAC_ADVERT_SYM_FLOW_CTRL; - if (state->pause & MLO_PAUSE_TXRX_MASK) - new_an |= MVNETA_GMAC_CONFIG_FLOW_CTRL; if (!phylink_autoneg_inband(mode)) { - /* Phy or fixed speed */ - if (state->duplex) - new_an |= MVNETA_GMAC_CONFIG_FULL_DUPLEX; - - if (state->speed == SPEED_1000 || state->speed == SPEED_2500) - new_an |= MVNETA_GMAC_CONFIG_GMII_SPEED; - else if (state->speed == SPEED_100) - new_an |= MVNETA_GMAC_CONFIG_MII_SPEED; + /* Phy or fixed speed - nothing to do, leave the + * configured speed, duplex and flow control as-is. + */ } else if (state->interface == PHY_INTERFACE_MODE_SGMII) { /* SGMII mode receives the state from the PHY */ new_ctrl2 |= MVNETA_GMAC2_INBAND_AN_ENABLE; new_clk |= MVNETA_GMAC_1MS_CLOCK_ENABLE; new_an = (new_an & ~(MVNETA_GMAC_FORCE_LINK_DOWN | - MVNETA_GMAC_FORCE_LINK_PASS)) | + MVNETA_GMAC_FORCE_LINK_PASS | + MVNETA_GMAC_CONFIG_MII_SPEED | + MVNETA_GMAC_CONFIG_GMII_SPEED | + MVNETA_GMAC_CONFIG_FULL_DUPLEX)) | MVNETA_GMAC_INBAND_AN_ENABLE | MVNETA_GMAC_AN_SPEED_EN | MVNETA_GMAC_AN_DUPLEX_EN; @@ -3813,7 +3878,8 @@ static void mvneta_mac_config(struct phylink_config *config, unsigned int mode, new_ctrl0 |= MVNETA_GMAC0_PORT_1000BASE_X; new_clk |= MVNETA_GMAC_1MS_CLOCK_ENABLE; new_an = (new_an & ~(MVNETA_GMAC_FORCE_LINK_DOWN | - MVNETA_GMAC_FORCE_LINK_PASS)) | + MVNETA_GMAC_FORCE_LINK_PASS | + MVNETA_GMAC_CONFIG_MII_SPEED)) | MVNETA_GMAC_INBAND_AN_ENABLE | MVNETA_GMAC_CONFIG_GMII_SPEED | /* The MAC only supports FD mode */ @@ -3901,9 +3967,11 @@ static void mvneta_mac_link_down(struct phylink_config *config, mvneta_set_eee(pp, false); } -static void mvneta_mac_link_up(struct phylink_config *config, unsigned int mode, - phy_interface_t interface, - struct phy_device *phy) +static void mvneta_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause) { struct net_device *ndev = to_net_dev(config->dev); struct mvneta_port *pp = netdev_priv(ndev); @@ -3911,8 +3979,36 @@ static void mvneta_mac_link_up(struct phylink_config *config, unsigned int mode, if (!phylink_autoneg_inband(mode)) { val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG); - val &= ~MVNETA_GMAC_FORCE_LINK_DOWN; + val &= ~(MVNETA_GMAC_FORCE_LINK_DOWN | + MVNETA_GMAC_CONFIG_MII_SPEED | + MVNETA_GMAC_CONFIG_GMII_SPEED | + MVNETA_GMAC_CONFIG_FLOW_CTRL | + MVNETA_GMAC_CONFIG_FULL_DUPLEX); val |= MVNETA_GMAC_FORCE_LINK_PASS; + + if (speed == SPEED_1000 || speed == SPEED_2500) + val |= MVNETA_GMAC_CONFIG_GMII_SPEED; + else if (speed == SPEED_100) + val |= MVNETA_GMAC_CONFIG_MII_SPEED; + + if (duplex == DUPLEX_FULL) + val |= MVNETA_GMAC_CONFIG_FULL_DUPLEX; + + if (tx_pause || rx_pause) + val |= MVNETA_GMAC_CONFIG_FLOW_CTRL; + + mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val); + } else { + /* When inband doesn't cover flow control or flow control is + * disabled, we need to manually configure it. This bit will + * only have effect if MVNETA_GMAC_AN_FLOW_CTRL_EN is unset. + */ + val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG); + val &= ~MVNETA_GMAC_CONFIG_FLOW_CTRL; + + if (tx_pause || rx_pause) + val |= MVNETA_GMAC_CONFIG_FLOW_CTRL; + mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val); } @@ -4419,45 +4515,112 @@ static void mvneta_ethtool_get_strings(struct net_device *netdev, u32 sset, } } +static void +mvneta_ethtool_update_pcpu_stats(struct mvneta_port *pp, + struct mvneta_ethtool_stats *es) +{ + unsigned int start; + int cpu; + + for_each_possible_cpu(cpu) { + struct mvneta_pcpu_stats *stats; + u64 skb_alloc_error; + u64 refill_error; + u64 xdp_redirect; + u64 xdp_xmit_err; + u64 xdp_tx_err; + u64 xdp_pass; + u64 xdp_drop; + u64 xdp_xmit; + u64 xdp_tx; + + stats = per_cpu_ptr(pp->stats, cpu); + do { + start = u64_stats_fetch_begin_irq(&stats->syncp); + skb_alloc_error = stats->es.skb_alloc_error; + refill_error = stats->es.refill_error; + xdp_redirect = stats->es.ps.xdp_redirect; + xdp_pass = stats->es.ps.xdp_pass; + xdp_drop = stats->es.ps.xdp_drop; + xdp_xmit = stats->es.ps.xdp_xmit; + xdp_xmit_err = stats->es.ps.xdp_xmit_err; + xdp_tx = stats->es.ps.xdp_tx; + xdp_tx_err = stats->es.ps.xdp_tx_err; + } while (u64_stats_fetch_retry_irq(&stats->syncp, start)); + + es->skb_alloc_error += skb_alloc_error; + es->refill_error += refill_error; + es->ps.xdp_redirect += xdp_redirect; + es->ps.xdp_pass += xdp_pass; + es->ps.xdp_drop += xdp_drop; + es->ps.xdp_xmit += xdp_xmit; + es->ps.xdp_xmit_err += xdp_xmit_err; + es->ps.xdp_tx += xdp_tx; + es->ps.xdp_tx_err += xdp_tx_err; + } +} + static void mvneta_ethtool_update_stats(struct mvneta_port *pp) { + struct mvneta_ethtool_stats stats = {}; const struct mvneta_statistic *s; void __iomem *base = pp->base; u32 high, low; u64 val; int i; + mvneta_ethtool_update_pcpu_stats(pp, &stats); for (i = 0, s = mvneta_statistics; s < mvneta_statistics + ARRAY_SIZE(mvneta_statistics); s++, i++) { - val = 0; - switch (s->type) { case T_REG_32: val = readl_relaxed(base + s->offset); + pp->ethtool_stats[i] += val; break; case T_REG_64: /* Docs say to read low 32-bit then high */ low = readl_relaxed(base + s->offset); high = readl_relaxed(base + s->offset + 4); val = (u64)high << 32 | low; + pp->ethtool_stats[i] += val; break; case T_SW: switch (s->offset) { case ETHTOOL_STAT_EEE_WAKEUP: val = phylink_get_eee_err(pp->phylink); + pp->ethtool_stats[i] += val; break; case ETHTOOL_STAT_SKB_ALLOC_ERR: - val = pp->rxqs[0].skb_alloc_err; + pp->ethtool_stats[i] = stats.skb_alloc_error; break; case ETHTOOL_STAT_REFILL_ERR: - val = pp->rxqs[0].refill_err; + pp->ethtool_stats[i] = stats.refill_error; + break; + case ETHTOOL_XDP_REDIRECT: + pp->ethtool_stats[i] = stats.ps.xdp_redirect; + break; + case ETHTOOL_XDP_PASS: + pp->ethtool_stats[i] = stats.ps.xdp_pass; + break; + case ETHTOOL_XDP_DROP: + pp->ethtool_stats[i] = stats.ps.xdp_drop; + break; + case ETHTOOL_XDP_TX: + pp->ethtool_stats[i] = stats.ps.xdp_tx; + break; + case ETHTOOL_XDP_TX_ERR: + pp->ethtool_stats[i] = stats.ps.xdp_tx_err; + break; + case ETHTOOL_XDP_XMIT: + pp->ethtool_stats[i] = stats.ps.xdp_xmit; + break; + case ETHTOOL_XDP_XMIT_ERR: + pp->ethtool_stats[i] = stats.ps.xdp_xmit_err; break; } break; } - - pp->ethtool_stats[i] += val; } } @@ -4674,6 +4837,8 @@ static const struct net_device_ops mvneta_netdev_ops = { }; static const struct ethtool_ops mvneta_eth_tool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .nway_reset = mvneta_ethtool_nway_reset, .get_link = ethtool_op_get_link, .set_coalesce = mvneta_ethtool_set_coalesce, diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c index 35478cba2aa5..8972cdd559e8 100644 --- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c @@ -1082,6 +1082,9 @@ static int mvpp2_port_c2_tcam_rule_add(struct mvpp2_port *port, u8 qh, ql, pmap; int index, ctx; + if (!flow_action_basic_hw_stats_check(&rule->flow->action, NULL)) + return -EOPNOTSUPP; + memset(&c2, 0, sizeof(c2)); index = mvpp2_cls_c2_port_flow_index(port, rule->loc); @@ -1305,6 +1308,9 @@ static int mvpp2_cls_rfs_parse_rule(struct mvpp2_rfs_rule *rule) struct flow_rule *flow = rule->flow; struct flow_action_entry *act; + if (!flow_action_basic_hw_stats_check(&rule->flow->action, NULL)) + return -EOPNOTSUPP; + act = &flow->action.entries[0]; if (act->id != FLOW_ACTION_QUEUE && act->id != FLOW_ACTION_DROP) return -EOPNOTSUPP; diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c index 72133cbe55d4..1fa60e985b43 100644 --- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c @@ -58,8 +58,11 @@ static struct { */ static void mvpp2_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state); -static void mvpp2_mac_link_up(struct phylink_config *config, unsigned int mode, - phy_interface_t interface, struct phy_device *phy); +static void mvpp2_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause); /* Queue modes */ #define MVPP2_QDIST_SINGLE_MODE 0 @@ -3473,8 +3476,9 @@ static void mvpp2_start_dev(struct mvpp2_port *port) .interface = port->phy_interface, }; mvpp2_mac_config(&port->phylink_config, MLO_AN_INBAND, &state); - mvpp2_mac_link_up(&port->phylink_config, MLO_AN_INBAND, - port->phy_interface, NULL); + mvpp2_mac_link_up(&port->phylink_config, NULL, + MLO_AN_INBAND, port->phy_interface, + SPEED_UNKNOWN, DUPLEX_UNKNOWN, false, false); } netif_tx_start_all_queues(port->dev); @@ -4380,6 +4384,8 @@ static const struct net_device_ops mvpp2_netdev_ops = { }; static const struct ethtool_ops mvpp2_eth_tool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .nway_reset = mvpp2_ethtool_nway_reset, .get_link = ethtool_op_get_link, .set_coalesce = mvpp2_ethtool_set_coalesce, @@ -4972,15 +4978,13 @@ static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode, old_ctrl2 = ctrl2 = readl(port->base + MVPP2_GMAC_CTRL_2_REG); old_ctrl4 = ctrl4 = readl(port->base + MVPP22_GMAC_CTRL_4_REG); - an &= ~(MVPP2_GMAC_CONFIG_MII_SPEED | MVPP2_GMAC_CONFIG_GMII_SPEED | - MVPP2_GMAC_AN_SPEED_EN | MVPP2_GMAC_FC_ADV_EN | + an &= ~(MVPP2_GMAC_AN_SPEED_EN | MVPP2_GMAC_FC_ADV_EN | MVPP2_GMAC_FC_ADV_ASM_EN | MVPP2_GMAC_FLOW_CTRL_AUTONEG | - MVPP2_GMAC_CONFIG_FULL_DUPLEX | MVPP2_GMAC_AN_DUPLEX_EN | - MVPP2_GMAC_IN_BAND_AUTONEG | MVPP2_GMAC_IN_BAND_AUTONEG_BYPASS); + MVPP2_GMAC_AN_DUPLEX_EN | MVPP2_GMAC_IN_BAND_AUTONEG | + MVPP2_GMAC_IN_BAND_AUTONEG_BYPASS); ctrl0 &= ~MVPP2_GMAC_PORT_TYPE_MASK; ctrl2 &= ~(MVPP2_GMAC_INBAND_AN_MASK | MVPP2_GMAC_PORT_RESET_MASK | MVPP2_GMAC_PCS_ENABLE_MASK); - ctrl4 &= ~(MVPP22_CTRL4_RX_FC_EN | MVPP22_CTRL4_TX_FC_EN); /* Configure port type */ if (phy_interface_mode_is_8023z(state->interface)) { @@ -5010,31 +5014,20 @@ static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode, /* Configure negotiation style */ if (!phylink_autoneg_inband(mode)) { - /* Phy or fixed speed - no in-band AN */ - if (state->duplex) - an |= MVPP2_GMAC_CONFIG_FULL_DUPLEX; - - if (state->speed == SPEED_1000 || state->speed == SPEED_2500) - an |= MVPP2_GMAC_CONFIG_GMII_SPEED; - else if (state->speed == SPEED_100) - an |= MVPP2_GMAC_CONFIG_MII_SPEED; - - if (state->pause & MLO_PAUSE_TX) - ctrl4 |= MVPP22_CTRL4_TX_FC_EN; - if (state->pause & MLO_PAUSE_RX) - ctrl4 |= MVPP22_CTRL4_RX_FC_EN; + /* Phy or fixed speed - no in-band AN, nothing to do, leave the + * configured speed, duplex and flow control as-is. + */ } else if (state->interface == PHY_INTERFACE_MODE_SGMII) { /* SGMII in-band mode receives the speed and duplex from * the PHY. Flow control information is not received. */ - an &= ~(MVPP2_GMAC_FORCE_LINK_DOWN | MVPP2_GMAC_FORCE_LINK_PASS); + an &= ~(MVPP2_GMAC_FORCE_LINK_DOWN | + MVPP2_GMAC_FORCE_LINK_PASS | + MVPP2_GMAC_CONFIG_MII_SPEED | + MVPP2_GMAC_CONFIG_GMII_SPEED | + MVPP2_GMAC_CONFIG_FULL_DUPLEX); an |= MVPP2_GMAC_IN_BAND_AUTONEG | MVPP2_GMAC_AN_SPEED_EN | MVPP2_GMAC_AN_DUPLEX_EN; - - if (state->pause & MLO_PAUSE_TX) - ctrl4 |= MVPP22_CTRL4_TX_FC_EN; - if (state->pause & MLO_PAUSE_RX) - ctrl4 |= MVPP22_CTRL4_RX_FC_EN; } else if (phy_interface_mode_is_8023z(state->interface)) { /* 1000BaseX and 2500BaseX ports cannot negotiate speed nor can * they negotiate duplex: they are always operating with a fixed @@ -5042,19 +5035,17 @@ static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode, * speed and full duplex here. */ ctrl0 |= MVPP2_GMAC_PORT_TYPE_MASK; - an &= ~(MVPP2_GMAC_FORCE_LINK_DOWN | MVPP2_GMAC_FORCE_LINK_PASS); + an &= ~(MVPP2_GMAC_FORCE_LINK_DOWN | + MVPP2_GMAC_FORCE_LINK_PASS | + MVPP2_GMAC_CONFIG_MII_SPEED | + MVPP2_GMAC_CONFIG_GMII_SPEED | + MVPP2_GMAC_CONFIG_FULL_DUPLEX); an |= MVPP2_GMAC_IN_BAND_AUTONEG | MVPP2_GMAC_CONFIG_GMII_SPEED | MVPP2_GMAC_CONFIG_FULL_DUPLEX; - if (state->pause & MLO_PAUSE_AN && state->an_enabled) { + if (state->pause & MLO_PAUSE_AN && state->an_enabled) an |= MVPP2_GMAC_FLOW_CTRL_AUTONEG; - } else { - if (state->pause & MLO_PAUSE_TX) - ctrl4 |= MVPP22_CTRL4_TX_FC_EN; - if (state->pause & MLO_PAUSE_RX) - ctrl4 |= MVPP22_CTRL4_RX_FC_EN; - } } /* Some fields of the auto-negotiation register require the port to be down when @@ -5141,25 +5132,54 @@ static void mvpp2_mac_config(struct phylink_config *config, unsigned int mode, mvpp2_port_enable(port); } -static void mvpp2_mac_link_up(struct phylink_config *config, unsigned int mode, - phy_interface_t interface, struct phy_device *phy) +static void mvpp2_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause) { struct net_device *dev = to_net_dev(config->dev); struct mvpp2_port *port = netdev_priv(dev); u32 val; - if (!phylink_autoneg_inband(mode)) { - if (mvpp2_is_xlg(interface)) { + if (mvpp2_is_xlg(interface)) { + if (!phylink_autoneg_inband(mode)) { val = readl(port->base + MVPP22_XLG_CTRL0_REG); val &= ~MVPP22_XLG_CTRL0_FORCE_LINK_DOWN; val |= MVPP22_XLG_CTRL0_FORCE_LINK_PASS; writel(val, port->base + MVPP22_XLG_CTRL0_REG); - } else { + } + } else { + if (!phylink_autoneg_inband(mode)) { val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG); - val &= ~MVPP2_GMAC_FORCE_LINK_DOWN; + val &= ~(MVPP2_GMAC_FORCE_LINK_DOWN | + MVPP2_GMAC_CONFIG_MII_SPEED | + MVPP2_GMAC_CONFIG_GMII_SPEED | + MVPP2_GMAC_CONFIG_FULL_DUPLEX); val |= MVPP2_GMAC_FORCE_LINK_PASS; + + if (speed == SPEED_1000 || speed == SPEED_2500) + val |= MVPP2_GMAC_CONFIG_GMII_SPEED; + else if (speed == SPEED_100) + val |= MVPP2_GMAC_CONFIG_MII_SPEED; + + if (duplex == DUPLEX_FULL) + val |= MVPP2_GMAC_CONFIG_FULL_DUPLEX; + writel(val, port->base + MVPP2_GMAC_AUTONEG_CONFIG); } + + /* We can always update the flow control enable bits; + * these will only be effective if flow control AN + * (MVPP2_GMAC_FLOW_CTRL_AUTONEG) is disabled. + */ + val = readl(port->base + MVPP22_GMAC_CTRL_4_REG); + val &= ~(MVPP22_CTRL4_RX_FC_EN | MVPP22_CTRL4_TX_FC_EN); + if (tx_pause) + val |= MVPP22_CTRL4_TX_FC_EN; + if (rx_pause) + val |= MVPP22_CTRL4_RX_FC_EN; + writel(val, port->base + MVPP22_GMAC_CTRL_4_REG); } mvpp2_port_enable(port); diff --git a/drivers/net/ethernet/marvell/octeontx2/Kconfig b/drivers/net/ethernet/marvell/octeontx2/Kconfig index ced514c05c97..d9dfb614daa6 100644 --- a/drivers/net/ethernet/marvell/octeontx2/Kconfig +++ b/drivers/net/ethernet/marvell/octeontx2/Kconfig @@ -33,3 +33,9 @@ config OCTEONTX2_PF depends on PCI help This driver supports Marvell's OcteonTX2 NIC physical function. + +config OCTEONTX2_VF + tristate "Marvell OcteonTX2 NIC Virtual Function driver" + depends on OCTEONTX2_PF + help + This driver supports Marvell's OcteonTX2 NIC virtual function. diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c index 5ca788691911..a4e65da8d95b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c @@ -113,7 +113,6 @@ int cgx_get_cgxcnt_max(void) return idmax + 1; } -EXPORT_SYMBOL(cgx_get_cgxcnt_max); int cgx_get_lmac_cnt(void *cgxd) { @@ -124,7 +123,6 @@ int cgx_get_lmac_cnt(void *cgxd) return cgx->lmac_count; } -EXPORT_SYMBOL(cgx_get_lmac_cnt); void *cgx_get_pdata(int cgx_id) { @@ -136,7 +134,6 @@ void *cgx_get_pdata(int cgx_id) } return NULL; } -EXPORT_SYMBOL(cgx_get_pdata); int cgx_get_cgxid(void *cgxd) { @@ -164,7 +161,6 @@ int cgx_get_link_info(void *cgxd, int lmac_id, *linfo = lmac->link_info; return 0; } -EXPORT_SYMBOL(cgx_get_link_info); static u64 mac2u64 (u8 *mac_addr) { @@ -195,7 +191,6 @@ int cgx_lmac_addr_set(u8 cgx_id, u8 lmac_id, u8 *mac_addr) return 0; } -EXPORT_SYMBOL(cgx_lmac_addr_set); u64 cgx_lmac_addr_get(u8 cgx_id, u8 lmac_id) { @@ -205,7 +200,6 @@ u64 cgx_lmac_addr_get(u8 cgx_id, u8 lmac_id) cfg = cgx_read(cgx_dev, 0, CGXX_CMRX_RX_DMAC_CAM0 + lmac_id * 0x8); return cfg & CGX_RX_DMAC_ADR_MASK; } -EXPORT_SYMBOL(cgx_lmac_addr_get); int cgx_set_pkind(void *cgxd, u8 lmac_id, int pkind) { @@ -217,7 +211,6 @@ int cgx_set_pkind(void *cgxd, u8 lmac_id, int pkind) cgx_write(cgx, lmac_id, CGXX_CMRX_RX_ID_MAP, (pkind & 0x3F)); return 0; } -EXPORT_SYMBOL(cgx_set_pkind); static inline u8 cgx_get_lmac_type(struct cgx *cgx, int lmac_id) { @@ -255,7 +248,6 @@ int cgx_lmac_internal_loopback(void *cgxd, int lmac_id, bool enable) } return 0; } -EXPORT_SYMBOL(cgx_lmac_internal_loopback); void cgx_lmac_promisc_config(int cgx_id, int lmac_id, bool enable) { @@ -289,7 +281,6 @@ void cgx_lmac_promisc_config(int cgx_id, int lmac_id, bool enable) (CGXX_CMRX_RX_DMAC_CAM0 + lmac_id * 0x8), cfg); } } -EXPORT_SYMBOL(cgx_lmac_promisc_config); /* Enable or disable forwarding received pause frames to Tx block */ void cgx_lmac_enadis_rx_pause_fwding(void *cgxd, int lmac_id, bool enable) @@ -318,7 +309,6 @@ void cgx_lmac_enadis_rx_pause_fwding(void *cgxd, int lmac_id, bool enable) cgx_write(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL, cfg); } } -EXPORT_SYMBOL(cgx_lmac_enadis_rx_pause_fwding); int cgx_get_rx_stats(void *cgxd, int lmac_id, int idx, u64 *rx_stat) { @@ -329,7 +319,6 @@ int cgx_get_rx_stats(void *cgxd, int lmac_id, int idx, u64 *rx_stat) *rx_stat = cgx_read(cgx, lmac_id, CGXX_CMRX_RX_STAT0 + (idx * 8)); return 0; } -EXPORT_SYMBOL(cgx_get_rx_stats); int cgx_get_tx_stats(void *cgxd, int lmac_id, int idx, u64 *tx_stat) { @@ -340,7 +329,6 @@ int cgx_get_tx_stats(void *cgxd, int lmac_id, int idx, u64 *tx_stat) *tx_stat = cgx_read(cgx, lmac_id, CGXX_CMRX_TX_STAT0 + (idx * 8)); return 0; } -EXPORT_SYMBOL(cgx_get_tx_stats); int cgx_lmac_rx_tx_enable(void *cgxd, int lmac_id, bool enable) { @@ -358,7 +346,6 @@ int cgx_lmac_rx_tx_enable(void *cgxd, int lmac_id, bool enable) cgx_write(cgx, lmac_id, CGXX_CMRX_CFG, cfg); return 0; } -EXPORT_SYMBOL(cgx_lmac_rx_tx_enable); int cgx_lmac_tx_enable(void *cgxd, int lmac_id, bool enable) { @@ -379,7 +366,107 @@ int cgx_lmac_tx_enable(void *cgxd, int lmac_id, bool enable) cgx_write(cgx, lmac_id, CGXX_CMRX_CFG, cfg); return !!(last & DATA_PKT_TX_EN); } -EXPORT_SYMBOL(cgx_lmac_tx_enable); + +int cgx_lmac_get_pause_frm(void *cgxd, int lmac_id, + u8 *tx_pause, u8 *rx_pause) +{ + struct cgx *cgx = cgxd; + u64 cfg; + + if (!cgx || lmac_id >= cgx->lmac_count) + return -ENODEV; + + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL); + *rx_pause = !!(cfg & CGX_SMUX_RX_FRM_CTL_CTL_BCK); + + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_TX_CTL); + *tx_pause = !!(cfg & CGX_SMUX_TX_CTL_L2P_BP_CONV); + return 0; +} + +int cgx_lmac_set_pause_frm(void *cgxd, int lmac_id, + u8 tx_pause, u8 rx_pause) +{ + struct cgx *cgx = cgxd; + u64 cfg; + + if (!cgx || lmac_id >= cgx->lmac_count) + return -ENODEV; + + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL); + cfg &= ~CGX_SMUX_RX_FRM_CTL_CTL_BCK; + cfg |= rx_pause ? CGX_SMUX_RX_FRM_CTL_CTL_BCK : 0x0; + cgx_write(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL, cfg); + + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_TX_CTL); + cfg &= ~CGX_SMUX_TX_CTL_L2P_BP_CONV; + cfg |= tx_pause ? CGX_SMUX_TX_CTL_L2P_BP_CONV : 0x0; + cgx_write(cgx, lmac_id, CGXX_SMUX_TX_CTL, cfg); + + cfg = cgx_read(cgx, 0, CGXX_CMR_RX_OVR_BP); + if (tx_pause) { + cfg &= ~CGX_CMR_RX_OVR_BP_EN(lmac_id); + } else { + cfg |= CGX_CMR_RX_OVR_BP_EN(lmac_id); + cfg &= ~CGX_CMR_RX_OVR_BP_BP(lmac_id); + } + cgx_write(cgx, 0, CGXX_CMR_RX_OVR_BP, cfg); + return 0; +} + +static void cgx_lmac_pause_frm_config(struct cgx *cgx, int lmac_id, bool enable) +{ + u64 cfg; + + if (!cgx || lmac_id >= cgx->lmac_count) + return; + if (enable) { + /* Enable receive pause frames */ + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL); + cfg |= CGX_SMUX_RX_FRM_CTL_CTL_BCK; + cgx_write(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL, cfg); + + cfg = cgx_read(cgx, lmac_id, CGXX_GMP_GMI_RXX_FRM_CTL); + cfg |= CGX_GMP_GMI_RXX_FRM_CTL_CTL_BCK; + cgx_write(cgx, lmac_id, CGXX_GMP_GMI_RXX_FRM_CTL, cfg); + + /* Enable pause frames transmission */ + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_TX_CTL); + cfg |= CGX_SMUX_TX_CTL_L2P_BP_CONV; + cgx_write(cgx, lmac_id, CGXX_SMUX_TX_CTL, cfg); + + /* Set pause time and interval */ + cgx_write(cgx, lmac_id, CGXX_SMUX_TX_PAUSE_PKT_TIME, + DEFAULT_PAUSE_TIME); + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_TX_PAUSE_PKT_INTERVAL); + cfg &= ~0xFFFFULL; + cgx_write(cgx, lmac_id, CGXX_SMUX_TX_PAUSE_PKT_INTERVAL, + cfg | (DEFAULT_PAUSE_TIME / 2)); + + cgx_write(cgx, lmac_id, CGXX_GMP_GMI_TX_PAUSE_PKT_TIME, + DEFAULT_PAUSE_TIME); + + cfg = cgx_read(cgx, lmac_id, + CGXX_GMP_GMI_TX_PAUSE_PKT_INTERVAL); + cfg &= ~0xFFFFULL; + cgx_write(cgx, lmac_id, CGXX_GMP_GMI_TX_PAUSE_PKT_INTERVAL, + cfg | (DEFAULT_PAUSE_TIME / 2)); + } else { + /* ALL pause frames received are completely ignored */ + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL); + cfg &= ~CGX_SMUX_RX_FRM_CTL_CTL_BCK; + cgx_write(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL, cfg); + + cfg = cgx_read(cgx, lmac_id, CGXX_GMP_GMI_RXX_FRM_CTL); + cfg &= ~CGX_GMP_GMI_RXX_FRM_CTL_CTL_BCK; + cgx_write(cgx, lmac_id, CGXX_GMP_GMI_RXX_FRM_CTL, cfg); + + /* Disable pause frames transmission */ + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_TX_CTL); + cfg &= ~CGX_SMUX_TX_CTL_L2P_BP_CONV; + cgx_write(cgx, lmac_id, CGXX_SMUX_TX_CTL, cfg); + } +} /* CGX Firmware interface low level support */ static int cgx_fwi_cmd_send(u64 req, u64 *resp, struct lmac *lmac) @@ -558,60 +645,6 @@ static inline bool cgx_event_is_linkevent(u64 event) return false; } -static inline int cgx_fwi_get_mkex_prfl_sz(u64 *prfl_sz, - struct cgx *cgx) -{ - u64 req = 0; - u64 resp; - int err; - - req = FIELD_SET(CMDREG_ID, CGX_CMD_GET_MKEX_PRFL_SIZE, req); - err = cgx_fwi_cmd_generic(req, &resp, cgx, 0); - if (!err) - *prfl_sz = FIELD_GET(RESP_MKEX_PRFL_SIZE, resp); - - return err; -} - -static inline int cgx_fwi_get_mkex_prfl_addr(u64 *prfl_addr, - struct cgx *cgx) -{ - u64 req = 0; - u64 resp; - int err; - - req = FIELD_SET(CMDREG_ID, CGX_CMD_GET_MKEX_PRFL_ADDR, req); - err = cgx_fwi_cmd_generic(req, &resp, cgx, 0); - if (!err) - *prfl_addr = FIELD_GET(RESP_MKEX_PRFL_ADDR, resp); - - return err; -} - -int cgx_get_mkex_prfl_info(u64 *addr, u64 *size) -{ - struct cgx *cgx_dev; - int err; - - if (!addr || !size) - return -EINVAL; - - cgx_dev = list_first_entry(&cgx_list, struct cgx, cgx_list); - if (!cgx_dev) - return -ENXIO; - - err = cgx_fwi_get_mkex_prfl_sz(size, cgx_dev); - if (err) - return -EIO; - - err = cgx_fwi_get_mkex_prfl_addr(addr, cgx_dev); - if (err) - return -EIO; - - return 0; -} -EXPORT_SYMBOL(cgx_get_mkex_prfl_info); - static irqreturn_t cgx_fwi_event_handler(int irq, void *data) { struct lmac *lmac = data; @@ -676,7 +709,6 @@ int cgx_lmac_evh_register(struct cgx_event_cb *cb, void *cgxd, int lmac_id) return 0; } -EXPORT_SYMBOL(cgx_lmac_evh_register); int cgx_lmac_evh_unregister(void *cgxd, int lmac_id) { @@ -695,7 +727,24 @@ int cgx_lmac_evh_unregister(void *cgxd, int lmac_id) return 0; } -EXPORT_SYMBOL(cgx_lmac_evh_unregister); + +int cgx_get_fwdata_base(u64 *base) +{ + u64 req = 0, resp; + struct cgx *cgx; + int err; + + cgx = list_first_entry_or_null(&cgx_list, struct cgx, cgx_list); + if (!cgx) + return -ENXIO; + + req = FIELD_SET(CMDREG_ID, CGX_CMD_GET_FWD_BASE, req); + err = cgx_fwi_cmd_generic(req, &resp, cgx, 0); + if (!err) + *base = FIELD_GET(RESP_FWD_BASE, resp); + + return err; +} static int cgx_fwi_link_change(struct cgx *cgx, int lmac_id, bool enable) { @@ -769,7 +818,6 @@ int cgx_lmac_linkup_start(void *cgxd) return 0; } -EXPORT_SYMBOL(cgx_lmac_linkup_start); static int cgx_lmac_init(struct cgx *cgx) { @@ -805,6 +853,7 @@ static int cgx_lmac_init(struct cgx *cgx) /* Add reference */ cgx->lmac_idmap[i] = lmac; + cgx_lmac_pause_frm_config(cgx, i, true); } return cgx_lmac_verify_fwi_version(cgx); @@ -823,6 +872,7 @@ static int cgx_lmac_exit(struct cgx *cgx) /* Free all lmac related resources */ for (i = 0; i < cgx->lmac_count; i++) { + cgx_lmac_pause_frm_config(cgx, i, false); lmac = cgx->lmac_idmap[i]; if (!lmac) continue; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx.h b/drivers/net/ethernet/marvell/octeontx2/af/cgx.h index 9343bf39cfac..394f96591feb 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx.h @@ -60,10 +60,20 @@ #define CGX_SMUX_RX_FRM_CTL_CTL_BCK BIT_ULL(3) #define CGXX_GMP_GMI_RXX_FRM_CTL 0x38028 #define CGX_GMP_GMI_RXX_FRM_CTL_CTL_BCK BIT_ULL(3) +#define CGXX_SMUX_TX_CTL 0x20178 +#define CGXX_SMUX_TX_PAUSE_PKT_TIME 0x20110 +#define CGXX_SMUX_TX_PAUSE_PKT_INTERVAL 0x20120 +#define CGXX_GMP_GMI_TX_PAUSE_PKT_TIME 0x38230 +#define CGXX_GMP_GMI_TX_PAUSE_PKT_INTERVAL 0x38248 +#define CGX_SMUX_TX_CTL_L2P_BP_CONV BIT_ULL(7) +#define CGXX_CMR_RX_OVR_BP 0x130 +#define CGX_CMR_RX_OVR_BP_EN(X) BIT_ULL(((X) + 8)) +#define CGX_CMR_RX_OVR_BP_BP(X) BIT_ULL(((X) + 4)) #define CGX_COMMAND_REG CGXX_SCRATCH1_REG #define CGX_EVENT_REG CGXX_SCRATCH0_REG #define CGX_CMD_TIMEOUT 2200 /* msecs */ +#define DEFAULT_PAUSE_TIME 0x7FF #define CGX_NVEC 37 #define CGX_LMAC_FWI 0 @@ -124,5 +134,9 @@ int cgx_lmac_internal_loopback(void *cgxd, int lmac_id, bool enable); int cgx_get_link_info(void *cgxd, int lmac_id, struct cgx_link_user_info *linfo); int cgx_lmac_linkup_start(void *cgxd); -int cgx_get_mkex_prfl_info(u64 *addr, u64 *size); +int cgx_get_fwdata_base(u64 *base); +int cgx_lmac_get_pause_frm(void *cgxd, int lmac_id, + u8 *tx_pause, u8 *rx_pause); +int cgx_lmac_set_pause_frm(void *cgxd, int lmac_id, + u8 tx_pause, u8 rx_pause); #endif /* CGX_H */ diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx_fw_if.h b/drivers/net/ethernet/marvell/octeontx2/af/cgx_fw_if.h index 473d9751601f..c3702fa58b6b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx_fw_if.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx_fw_if.h @@ -79,7 +79,8 @@ enum cgx_cmd_id { CGX_CMD_MODE_CHANGE, /* hot plug support */ CGX_CMD_INTF_SHUTDOWN, CGX_CMD_GET_MKEX_PRFL_SIZE, - CGX_CMD_GET_MKEX_PRFL_ADDR + CGX_CMD_GET_MKEX_PRFL_ADDR, + CGX_CMD_GET_FWD_BASE, /* get base address of shared FW data */ }; /* async event ids */ @@ -149,6 +150,11 @@ enum cgx_cmd_own { */ #define RESP_MKEX_PRFL_ADDR GENMASK_ULL(63, 9) +/* Response to cmd ID as CGX_CMD_GET_FWD_BASE with cmd status as + * CGX_STAT_SUCCESS + */ +#define RESP_FWD_BASE GENMASK_ULL(56, 9) + /* Response to cmd ID - CGX_CMD_LINK_BRING_UP/DOWN, event ID CGX_EVT_LINK_CHANGE * status can be either CGX_STAT_FAIL or CGX_STAT_SUCCESS * diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h index 8bbc1f1d81f5..6dfd0f90cd70 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h @@ -125,7 +125,7 @@ static inline struct mbox_msghdr *otx2_mbox_alloc_msg(struct otx2_mbox *mbox, M(READY, 0x001, ready, msg_req, ready_msg_rsp) \ M(ATTACH_RESOURCES, 0x002, attach_resources, rsrc_attach, msg_rsp) \ M(DETACH_RESOURCES, 0x003, detach_resources, rsrc_detach, msg_rsp) \ -M(MSIX_OFFSET, 0x004, msix_offset, msg_req, msix_offset_rsp) \ +M(MSIX_OFFSET, 0x005, msix_offset, msg_req, msix_offset_rsp) \ M(VF_FLR, 0x006, vf_flr, msg_req, msg_rsp) \ M(GET_HW_CAP, 0x008, get_hw_cap, msg_req, get_hw_cap_rsp) \ /* CGX mbox IDs (range 0x200 - 0x3FF) */ \ @@ -143,6 +143,8 @@ M(CGX_STOP_LINKEVENTS, 0x208, cgx_stop_linkevents, msg_req, msg_rsp) \ M(CGX_GET_LINKINFO, 0x209, cgx_get_linkinfo, msg_req, cgx_link_info_msg) \ M(CGX_INTLBK_ENABLE, 0x20A, cgx_intlbk_enable, msg_req, msg_rsp) \ M(CGX_INTLBK_DISABLE, 0x20B, cgx_intlbk_disable, msg_req, msg_rsp) \ +M(CGX_CFG_PAUSE_FRM, 0x20E, cgx_cfg_pause_frm, cgx_pause_frm_cfg, \ + cgx_pause_frm_cfg) \ /* NPA mbox IDs (range 0x400 - 0x5FF) */ \ M(NPA_LF_ALLOC, 0x400, npa_lf_alloc, \ npa_lf_alloc_req, npa_lf_alloc_rsp) \ @@ -211,6 +213,9 @@ M(NIX_LSO_FORMAT_CFG, 0x8011, nix_lso_format_cfg, \ nix_lso_format_cfg, \ nix_lso_format_cfg_rsp) \ M(NIX_RXVLAN_ALLOC, 0x8012, nix_rxvlan_alloc, msg_req, msg_rsp) \ +M(NIX_BP_ENABLE, 0x8016, nix_bp_enable, nix_bp_cfg_req, \ + nix_bp_cfg_rsp) \ +M(NIX_BP_DISABLE, 0x8017, nix_bp_disable, nix_bp_cfg_req, msg_rsp) \ M(NIX_GET_MAC_ADDR, 0x8018, nix_get_mac_addr, msg_req, nix_get_mac_addr_rsp) \ /* Messages initiated by AF (range 0xC00 - 0xDFF) */ @@ -251,7 +256,8 @@ enum rvu_af_status { struct ready_msg_rsp { struct mbox_msghdr hdr; - u16 sclk_feq; /* SCLK frequency */ + u16 sclk_freq; /* SCLK frequency (in MHz) */ + u16 rclk_freq; /* RCLK frequency (in MHz) */ }; /* Structure for requesting resource provisioning. @@ -342,6 +348,15 @@ struct cgx_link_info_msg { struct cgx_link_user_info link_info; }; +struct cgx_pause_frm_cfg { + struct mbox_msghdr hdr; + u8 set; + /* set = 1 if the request is to config pause frames */ + /* set = 0 if the request is to fetch pause frames config */ + u8 rx_pause; + u8 tx_pause; +}; + /* NPA mbox message formats */ /* NPA mailbox error codes @@ -676,6 +691,25 @@ struct nix_lso_format_cfg_rsp { u8 lso_format_idx; }; +struct nix_bp_cfg_req { + struct mbox_msghdr hdr; + u16 chan_base; /* Starting channel number */ + u8 chan_cnt; /* Number of channels */ + u8 bpid_per_chan; + /* bpid_per_chan = 0 assigns single bp id for range of channels */ + /* bpid_per_chan = 1 assigns separate bp id for each channel */ +}; + +/* PF can be mapped to either CGX or LBK interface, + * so maximum 64 channels are possible. + */ +#define NIX_MAX_BPID_CHAN 64 +struct nix_bp_cfg_rsp { + struct mbox_msghdr hdr; + u16 chan_bpid[NIX_MAX_BPID_CHAN]; /* Channel and bpid mapping */ + u8 chan_cnt; /* Number of channel for which bpids are assigned */ +}; + /* NPC mbox message structs */ #define NPC_MCAM_ENTRY_INVALID 0xFFFF diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c index 5c190c3ce898..557e4292c846 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c @@ -21,7 +21,6 @@ #define DRV_NAME "octeontx2-af" #define DRV_STRING "Marvell OcteonTX2 RVU Admin Function Driver" -#define DRV_VERSION "1.0" static int rvu_get_hwvf(struct rvu *rvu, int pcifunc); @@ -46,10 +45,9 @@ static const struct pci_device_id rvu_id_table[] = { { 0, } /* end of table */ }; -MODULE_AUTHOR("Marvell International Ltd."); +MODULE_AUTHOR("Sunil Goutham <sgoutham@marvell.com>"); MODULE_DESCRIPTION(DRV_STRING); MODULE_LICENSE("GPL v2"); -MODULE_VERSION(DRV_VERSION); MODULE_DEVICE_TABLE(pci, rvu_id_table); static char *mkex_profile; /* MKEX profile name */ @@ -88,13 +86,15 @@ int rvu_poll_reg(struct rvu *rvu, u64 block, u64 offset, u64 mask, bool zero) u64 reg_val; reg = rvu->afreg_base + ((block << 28) | offset); - while (time_before(jiffies, timeout)) { - reg_val = readq(reg); - if (zero && !(reg_val & mask)) - return 0; - if (!zero && (reg_val & mask)) - return 0; +again: + reg_val = readq(reg); + if (zero && !(reg_val & mask)) + return 0; + if (!zero && (reg_val & mask)) + return 0; + if (time_before(jiffies, timeout)) { usleep_range(1, 5); + goto again; } return -EBUSY; } @@ -421,6 +421,19 @@ static void rvu_check_block_implemented(struct rvu *rvu) } } +static void rvu_setup_rvum_blk_revid(struct rvu *rvu) +{ + rvu_write64(rvu, BLKADDR_RVUM, + RVU_PRIV_BLOCK_TYPEX_REV(BLKTYPE_RVUM), + RVU_BLK_RVUM_REVID); +} + +static void rvu_clear_rvum_blk_revid(struct rvu *rvu) +{ + rvu_write64(rvu, BLKADDR_RVUM, + RVU_PRIV_BLOCK_TYPEX_REV(BLKTYPE_RVUM), 0x00); +} + int rvu_lf_reset(struct rvu *rvu, struct rvu_block *block, int lf) { int err; @@ -603,7 +616,11 @@ setup_vfmsix: */ cfg = rvu_read64(rvu, BLKADDR_RVUM, RVU_PRIV_CONST); max_msix = cfg & 0xFFFFF; - phy_addr = rvu_read64(rvu, BLKADDR_RVUM, RVU_AF_MSIXTR_BASE); + if (rvu->fwdata && rvu->fwdata->msixtr_base) + phy_addr = rvu->fwdata->msixtr_base; + else + phy_addr = rvu_read64(rvu, BLKADDR_RVUM, RVU_AF_MSIXTR_BASE); + iova = dma_map_resource(rvu->dev, phy_addr, max_msix * PCI_MSIX_ENTRY_SIZE, DMA_BIDIRECTIONAL, 0); @@ -613,10 +630,18 @@ setup_vfmsix: rvu_write64(rvu, BLKADDR_RVUM, RVU_AF_MSIXTR_BASE, (u64)iova); rvu->msix_base_iova = iova; + rvu->msixtr_base_phy = phy_addr; return 0; } +static void rvu_reset_msix(struct rvu *rvu) +{ + /* Restore msixtr base register */ + rvu_write64(rvu, BLKADDR_RVUM, RVU_AF_MSIXTR_BASE, + rvu->msixtr_base_phy); +} + static void rvu_free_hw_resources(struct rvu *rvu) { struct rvu_hwinfo *hw = rvu->hw; @@ -655,9 +680,80 @@ static void rvu_free_hw_resources(struct rvu *rvu) max_msix * PCI_MSIX_ENTRY_SIZE, DMA_BIDIRECTIONAL, 0); + rvu_reset_msix(rvu); mutex_destroy(&rvu->rsrc_lock); } +static void rvu_setup_pfvf_macaddress(struct rvu *rvu) +{ + struct rvu_hwinfo *hw = rvu->hw; + int pf, vf, numvfs, hwvf; + struct rvu_pfvf *pfvf; + u64 *mac; + + for (pf = 0; pf < hw->total_pfs; pf++) { + if (!is_pf_cgxmapped(rvu, pf)) + continue; + /* Assign MAC address to PF */ + pfvf = &rvu->pf[pf]; + if (rvu->fwdata && pf < PF_MACNUM_MAX) { + mac = &rvu->fwdata->pf_macs[pf]; + if (*mac) + u64_to_ether_addr(*mac, pfvf->mac_addr); + else + eth_random_addr(pfvf->mac_addr); + } else { + eth_random_addr(pfvf->mac_addr); + } + + /* Assign MAC address to VFs */ + rvu_get_pf_numvfs(rvu, pf, &numvfs, &hwvf); + for (vf = 0; vf < numvfs; vf++, hwvf++) { + pfvf = &rvu->hwvf[hwvf]; + if (rvu->fwdata && hwvf < VF_MACNUM_MAX) { + mac = &rvu->fwdata->vf_macs[hwvf]; + if (*mac) + u64_to_ether_addr(*mac, pfvf->mac_addr); + else + eth_random_addr(pfvf->mac_addr); + } else { + eth_random_addr(pfvf->mac_addr); + } + } + } +} + +static int rvu_fwdata_init(struct rvu *rvu) +{ + u64 fwdbase; + int err; + + /* Get firmware data base address */ + err = cgx_get_fwdata_base(&fwdbase); + if (err) + goto fail; + rvu->fwdata = ioremap_wc(fwdbase, sizeof(struct rvu_fwdata)); + if (!rvu->fwdata) + goto fail; + if (!is_rvu_fwdata_valid(rvu)) { + dev_err(rvu->dev, + "Mismatch in 'fwdata' struct btw kernel and firmware\n"); + iounmap(rvu->fwdata); + rvu->fwdata = NULL; + return -EINVAL; + } + return 0; +fail: + dev_info(rvu->dev, "Unable to fetch 'fwdata' from firmware\n"); + return -EIO; +} + +static void rvu_fwdata_exit(struct rvu *rvu) +{ + if (rvu->fwdata) + iounmap(rvu->fwdata); +} + static int rvu_setup_hw_resources(struct rvu *rvu) { struct rvu_hwinfo *hw = rvu->hw; @@ -813,6 +909,8 @@ init: mutex_init(&rvu->rsrc_lock); + rvu_fwdata_init(rvu); + err = rvu_setup_msix_resources(rvu); if (err) return err; @@ -825,8 +923,10 @@ init: /* Allocate memory for block LF/slot to pcifunc mapping info */ block->fn_map = devm_kcalloc(rvu->dev, block->lf.max, sizeof(u16), GFP_KERNEL); - if (!block->fn_map) - return -ENOMEM; + if (!block->fn_map) { + err = -ENOMEM; + goto msix_err; + } /* Scan all blocks to check if low level firmware has * already provisioned any of the resources to a PF/VF. @@ -836,25 +936,36 @@ init: err = rvu_npc_init(rvu); if (err) - goto exit; + goto npc_err; err = rvu_cgx_init(rvu); if (err) - goto exit; + goto cgx_err; + + /* Assign MACs for CGX mapped functions */ + rvu_setup_pfvf_macaddress(rvu); err = rvu_npa_init(rvu); if (err) - goto cgx_err; + goto npa_err; err = rvu_nix_init(rvu); if (err) - goto cgx_err; + goto nix_err; return 0; +nix_err: + rvu_nix_freemem(rvu); +npa_err: + rvu_npa_freemem(rvu); cgx_err: rvu_cgx_exit(rvu); -exit: +npc_err: + rvu_npc_freemem(rvu); + rvu_fwdata_exit(rvu); +msix_err: + rvu_reset_msix(rvu); return err; } @@ -901,6 +1012,10 @@ int rvu_aq_alloc(struct rvu *rvu, struct admin_queue **ad_queue, int rvu_mbox_handler_ready(struct rvu *rvu, struct msg_req *req, struct ready_msg_rsp *rsp) { + if (rvu->fwdata) { + rsp->rclk_freq = rvu->fwdata->rclk; + rsp->sclk_freq = rvu->fwdata->sclk; + } return 0; } @@ -2128,6 +2243,9 @@ static int rvu_register_interrupts(struct rvu *rvu) } rvu->irq_allocated[RVU_AF_INT_VEC_PFME] = true; + /* Clear TRPEND bit for all PF */ + rvu_write64(rvu, BLKADDR_RVUM, + RVU_AF_PFTRPEND, INTR_MASK(rvu->hw->total_pfs)); /* Enable ME interrupt for all PFs*/ rvu_write64(rvu, BLKADDR_RVUM, RVU_AF_PFME_INT, INTR_MASK(rvu->hw->total_pfs)); @@ -2439,17 +2557,13 @@ static int rvu_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto err_disable_device; } - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(48)); + err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(48)); if (err) { - dev_err(dev, "Unable to set DMA mask\n"); + dev_err(dev, "DMA mask config failed, abort\n"); goto err_release_regions; } - err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(48)); - if (err) { - dev_err(dev, "Unable to set consistent DMA mask\n"); - goto err_release_regions; - } + pci_set_master(pdev); /* Map Admin function CSRs */ rvu->afreg_base = pcim_iomap(pdev, PCI_AF_REG_BAR_NUM, 0); @@ -2489,6 +2603,8 @@ static int rvu_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (err) goto err_flr; + rvu_setup_rvum_blk_revid(rvu); + /* Enable AF's VFs (if any) */ err = rvu_enable_sriov(rvu); if (err) @@ -2506,8 +2622,10 @@ err_mbox: rvu_mbox_destroy(&rvu->afpf_wq_info); err_hwsetup: rvu_cgx_exit(rvu); + rvu_fwdata_exit(rvu); rvu_reset_all_blocks(rvu); rvu_free_hw_resources(rvu); + rvu_clear_rvum_blk_revid(rvu); err_release_regions: pci_release_regions(pdev); err_disable_device: @@ -2527,11 +2645,12 @@ static void rvu_remove(struct pci_dev *pdev) rvu_unregister_interrupts(rvu); rvu_flr_wq_destroy(rvu); rvu_cgx_exit(rvu); + rvu_fwdata_exit(rvu); rvu_mbox_destroy(&rvu->afpf_wq_info); rvu_disable_sriov(rvu); rvu_reset_all_blocks(rvu); rvu_free_hw_resources(rvu); - + rvu_clear_rvum_blk_revid(rvu); pci_release_regions(pdev); pci_disable_device(pdev); pci_set_drvdata(pdev, NULL); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h index 51c206f4fe6f..dcf25a092008 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h @@ -269,6 +269,26 @@ struct mbox_wq_info { struct workqueue_struct *mbox_wq; }; +struct rvu_fwdata { +#define RVU_FWDATA_HEADER_MAGIC 0xCFDA /* Custom Firmware Data*/ +#define RVU_FWDATA_VERSION 0x0001 + u32 header_magic; + u32 version; /* version id */ + + /* MAC address */ +#define PF_MACNUM_MAX 32 +#define VF_MACNUM_MAX 256 + u64 pf_macs[PF_MACNUM_MAX]; + u64 vf_macs[VF_MACNUM_MAX]; + u64 sclk; + u64 rclk; + u64 mcam_addr; + u64 mcam_sz; + u64 msixtr_base; +#define FWDATA_RESERVED_MEM 1023 + u64 reserved[FWDATA_RESERVED_MEM]; +}; + struct rvu { void __iomem *afreg_base; void __iomem *pfreg_base; @@ -294,6 +314,7 @@ struct rvu { char *irq_name; bool *irq_allocated; dma_addr_t msix_base_iova; + u64 msixtr_base_phy; /* Register reset value */ /* CGX */ #define PF_CGXMAP_BASE 1 /* PF 0 is reserved for RVU PF */ @@ -313,6 +334,9 @@ struct rvu { char mkex_pfl_name[MKEX_NAME_LEN]; /* Configured MKEX profile name */ + /* Firmware data */ + struct rvu_fwdata *fwdata; + #ifdef CONFIG_DEBUG_FS struct rvu_debugfs rvu_dbg; #endif @@ -363,6 +387,12 @@ static inline int is_afvf(u16 pcifunc) return !(pcifunc & ~RVU_PFVF_FUNC_MASK); } +static inline bool is_rvu_fwdata_valid(struct rvu *rvu) +{ + return (rvu->fwdata->header_magic == RVU_FWDATA_HEADER_MAGIC) && + (rvu->fwdata->version == RVU_FWDATA_VERSION); +} + int rvu_alloc_bitmap(struct rsrc_bmap *rsrc); int rvu_alloc_rsrc(struct rsrc_bmap *rsrc); void rvu_free_rsrc(struct rsrc_bmap *rsrc, int id); @@ -432,7 +462,7 @@ int rvu_nix_reserve_mark_format(struct rvu *rvu, struct nix_hw *nix_hw, void rvu_nix_freemem(struct rvu *rvu); int rvu_get_nixlf_count(struct rvu *rvu); void rvu_nix_lf_teardown(struct rvu *rvu, u16 pcifunc, int blkaddr, int npalf); -int nix_get_nixlf(struct rvu *rvu, u16 pcifunc, int *nixlf); +int nix_get_nixlf(struct rvu *rvu, u16 pcifunc, int *nixlf, int *nix_blkaddr); /* NPC APIs */ int rvu_npc_init(struct rvu *rvu); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c index 11e5921c55b9..f3c82e489897 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c @@ -350,6 +350,18 @@ int rvu_cgx_exit(struct rvu *rvu) return 0; } +/* Most of the CGX configuration is restricted to the mapped PF only, + * VF's of mapped PF and other PFs are not allowed. This fn() checks + * whether a PFFUNC is permitted to do the config or not. + */ +static bool is_cgx_config_permitted(struct rvu *rvu, u16 pcifunc) +{ + if ((pcifunc & RVU_PFVF_FUNC_MASK) || + !is_pf_cgxmapped(rvu, rvu_get_pf(pcifunc))) + return false; + return true; +} + void rvu_cgx_enadis_rx_bp(struct rvu *rvu, int pf, bool enable) { u8 cgx_id, lmac_id; @@ -373,11 +385,8 @@ int rvu_cgx_config_rxtx(struct rvu *rvu, u16 pcifunc, bool start) int pf = rvu_get_pf(pcifunc); u8 cgx_id, lmac_id; - /* This msg is expected only from PFs that are mapped to CGX LMACs, - * if received from other PF/VF simply ACK, nothing to do. - */ - if ((pcifunc & RVU_PFVF_FUNC_MASK) || !is_pf_cgxmapped(rvu, pf)) - return -ENODEV; + if (!is_cgx_config_permitted(rvu, pcifunc)) + return -EPERM; rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id); @@ -409,8 +418,7 @@ int rvu_mbox_handler_cgx_stats(struct rvu *rvu, struct msg_req *req, u8 cgx_idx, lmac; void *cgxd; - if ((req->hdr.pcifunc & RVU_PFVF_FUNC_MASK) || - !is_pf_cgxmapped(rvu, pf)) + if (!is_cgx_config_permitted(rvu, req->hdr.pcifunc)) return -ENODEV; rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_idx, &lmac); @@ -477,12 +485,8 @@ int rvu_mbox_handler_cgx_promisc_enable(struct rvu *rvu, struct msg_req *req, int pf = rvu_get_pf(pcifunc); u8 cgx_id, lmac_id; - /* This msg is expected only from PFs that are mapped to CGX LMACs, - * if received from other PF/VF simply ACK, nothing to do. - */ - if ((req->hdr.pcifunc & RVU_PFVF_FUNC_MASK) || - !is_pf_cgxmapped(rvu, pf)) - return -ENODEV; + if (!is_cgx_config_permitted(rvu, req->hdr.pcifunc)) + return -EPERM; rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id); @@ -493,16 +497,11 @@ int rvu_mbox_handler_cgx_promisc_enable(struct rvu *rvu, struct msg_req *req, int rvu_mbox_handler_cgx_promisc_disable(struct rvu *rvu, struct msg_req *req, struct msg_rsp *rsp) { - u16 pcifunc = req->hdr.pcifunc; - int pf = rvu_get_pf(pcifunc); + int pf = rvu_get_pf(req->hdr.pcifunc); u8 cgx_id, lmac_id; - /* This msg is expected only from PFs that are mapped to CGX LMACs, - * if received from other PF/VF simply ACK, nothing to do. - */ - if ((req->hdr.pcifunc & RVU_PFVF_FUNC_MASK) || - !is_pf_cgxmapped(rvu, pf)) - return -ENODEV; + if (!is_cgx_config_permitted(rvu, req->hdr.pcifunc)) + return -EPERM; rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id); @@ -515,11 +514,8 @@ static int rvu_cgx_config_linkevents(struct rvu *rvu, u16 pcifunc, bool en) int pf = rvu_get_pf(pcifunc); u8 cgx_id, lmac_id; - /* This msg is expected only from PFs that are mapped to CGX LMACs, - * if received from other PF/VF simply ACK, nothing to do. - */ - if ((pcifunc & RVU_PFVF_FUNC_MASK) || !is_pf_cgxmapped(rvu, pf)) - return -ENODEV; + if (!is_cgx_config_permitted(rvu, pcifunc)) + return -EPERM; rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id); @@ -571,11 +567,8 @@ static int rvu_cgx_config_intlbk(struct rvu *rvu, u16 pcifunc, bool en) int pf = rvu_get_pf(pcifunc); u8 cgx_id, lmac_id; - /* This msg is expected only from PFs that are mapped to CGX LMACs, - * if received from other PF/VF simply ACK, nothing to do. - */ - if ((pcifunc & RVU_PFVF_FUNC_MASK) || !is_pf_cgxmapped(rvu, pf)) - return -ENODEV; + if (!is_cgx_config_permitted(rvu, pcifunc)) + return -EPERM; rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id); @@ -597,6 +590,30 @@ int rvu_mbox_handler_cgx_intlbk_disable(struct rvu *rvu, struct msg_req *req, return 0; } +int rvu_mbox_handler_cgx_cfg_pause_frm(struct rvu *rvu, + struct cgx_pause_frm_cfg *req, + struct cgx_pause_frm_cfg *rsp) +{ + int pf = rvu_get_pf(req->hdr.pcifunc); + u8 cgx_id, lmac_id; + + /* This msg is expected only from PF/VFs that are mapped to CGX LMACs, + * if received from other PF/VF simply ACK, nothing to do. + */ + if (!is_pf_cgxmapped(rvu, pf)) + return -ENODEV; + + rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id); + + if (req->set) + cgx_lmac_set_pause_frm(rvu_cgx_pdata(cgx_id, rvu), lmac_id, + req->tx_pause, req->rx_pause); + else + cgx_lmac_get_pause_frm(rvu_cgx_pdata(cgx_id, rvu), lmac_id, + &rsp->tx_pause, &rsp->rx_pause); + return 0; +} + /* Finds cumulative status of NIX rx/tx counters from LF of a PF and those * from its VFs as well. ie. NIX rx/tx counters at the CGX port level */ diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c index eb5e542424e7..36953d4f51c7 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c @@ -18,6 +18,8 @@ #include "cgx.h" static int nix_update_bcast_mce_list(struct rvu *rvu, u16 pcifunc, bool add); +static int rvu_nix_get_bpid(struct rvu *rvu, struct nix_bp_cfg_req *req, + int type, int chan_id); enum mc_tbl_sz { MC_TBL_SZ_256, @@ -90,6 +92,26 @@ int rvu_get_nixlf_count(struct rvu *rvu) return block->lf.max; } +int nix_get_nixlf(struct rvu *rvu, u16 pcifunc, int *nixlf, int *nix_blkaddr) +{ + struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, pcifunc); + struct rvu_hwinfo *hw = rvu->hw; + int blkaddr; + + blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); + if (!pfvf->nixlf || blkaddr < 0) + return NIX_AF_ERR_AF_LF_INVALID; + + *nixlf = rvu_get_lf(rvu, &hw->block[blkaddr], pcifunc, 0); + if (*nixlf < 0) + return NIX_AF_ERR_AF_LF_INVALID; + + if (nix_blkaddr) + *nix_blkaddr = blkaddr; + + return 0; +} + static void nix_mce_list_init(struct nix_mce_list *list, int max) { INIT_HLIST_HEAD(&list->head); @@ -191,6 +213,11 @@ static int nix_interface_init(struct rvu *rvu, u16 pcifunc, int type, int nixlf) pfvf->tx_chan_cnt = 1; cgx_set_pkind(rvu_cgx_pdata(cgx_id, rvu), lmac_id, pkind); rvu_npc_set_pkind(rvu, pkind, pfvf); + + /* By default we enable pause frames */ + if ((pcifunc & RVU_PFVF_FUNC_MASK) == 0) + cgx_lmac_set_pause_frm(rvu_cgx_pdata(cgx_id, rvu), + lmac_id, true, true); break; case NIX_INTF_TYPE_LBK: vf = (pcifunc & RVU_PFVF_FUNC_MASK) - 1; @@ -253,6 +280,142 @@ static void nix_interface_deinit(struct rvu *rvu, u16 pcifunc, u8 nixlf) rvu_npc_disable_mcam_entries(rvu, pcifunc, nixlf); } +int rvu_mbox_handler_nix_bp_disable(struct rvu *rvu, + struct nix_bp_cfg_req *req, + struct msg_rsp *rsp) +{ + u16 pcifunc = req->hdr.pcifunc; + struct rvu_pfvf *pfvf; + int blkaddr, pf, type; + u16 chan_base, chan; + u64 cfg; + + pf = rvu_get_pf(pcifunc); + type = is_afvf(pcifunc) ? NIX_INTF_TYPE_LBK : NIX_INTF_TYPE_CGX; + if (!is_pf_cgxmapped(rvu, pf) && type != NIX_INTF_TYPE_LBK) + return 0; + + pfvf = rvu_get_pfvf(rvu, pcifunc); + blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); + + chan_base = pfvf->rx_chan_base + req->chan_base; + for (chan = chan_base; chan < (chan_base + req->chan_cnt); chan++) { + cfg = rvu_read64(rvu, blkaddr, NIX_AF_RX_CHANX_CFG(chan)); + rvu_write64(rvu, blkaddr, NIX_AF_RX_CHANX_CFG(chan), + cfg & ~BIT_ULL(16)); + } + return 0; +} + +static int rvu_nix_get_bpid(struct rvu *rvu, struct nix_bp_cfg_req *req, + int type, int chan_id) +{ + int bpid, blkaddr, lmac_chan_cnt; + struct rvu_hwinfo *hw = rvu->hw; + u16 cgx_bpid_cnt, lbk_bpid_cnt; + struct rvu_pfvf *pfvf; + u8 cgx_id, lmac_id; + u64 cfg; + + blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, req->hdr.pcifunc); + cfg = rvu_read64(rvu, blkaddr, NIX_AF_CONST); + lmac_chan_cnt = cfg & 0xFF; + + cgx_bpid_cnt = hw->cgx_links * lmac_chan_cnt; + lbk_bpid_cnt = hw->lbk_links * ((cfg >> 16) & 0xFF); + + pfvf = rvu_get_pfvf(rvu, req->hdr.pcifunc); + + /* Backpressure IDs range division + * CGX channles are mapped to (0 - 191) BPIDs + * LBK channles are mapped to (192 - 255) BPIDs + * SDP channles are mapped to (256 - 511) BPIDs + * + * Lmac channles and bpids mapped as follows + * cgx(0)_lmac(0)_chan(0 - 15) = bpid(0 - 15) + * cgx(0)_lmac(1)_chan(0 - 15) = bpid(16 - 31) .... + * cgx(1)_lmac(0)_chan(0 - 15) = bpid(64 - 79) .... + */ + switch (type) { + case NIX_INTF_TYPE_CGX: + if ((req->chan_base + req->chan_cnt) > 15) + return -EINVAL; + rvu_get_cgx_lmac_id(pfvf->cgx_lmac, &cgx_id, &lmac_id); + /* Assign bpid based on cgx, lmac and chan id */ + bpid = (cgx_id * hw->lmac_per_cgx * lmac_chan_cnt) + + (lmac_id * lmac_chan_cnt) + req->chan_base; + + if (req->bpid_per_chan) + bpid += chan_id; + if (bpid > cgx_bpid_cnt) + return -EINVAL; + break; + + case NIX_INTF_TYPE_LBK: + if ((req->chan_base + req->chan_cnt) > 63) + return -EINVAL; + bpid = cgx_bpid_cnt + req->chan_base; + if (req->bpid_per_chan) + bpid += chan_id; + if (bpid > (cgx_bpid_cnt + lbk_bpid_cnt)) + return -EINVAL; + break; + default: + return -EINVAL; + } + return bpid; +} + +int rvu_mbox_handler_nix_bp_enable(struct rvu *rvu, + struct nix_bp_cfg_req *req, + struct nix_bp_cfg_rsp *rsp) +{ + int blkaddr, pf, type, chan_id = 0; + u16 pcifunc = req->hdr.pcifunc; + struct rvu_pfvf *pfvf; + u16 chan_base, chan; + s16 bpid, bpid_base; + u64 cfg; + + pf = rvu_get_pf(pcifunc); + type = is_afvf(pcifunc) ? NIX_INTF_TYPE_LBK : NIX_INTF_TYPE_CGX; + + /* Enable backpressure only for CGX mapped PFs and LBK interface */ + if (!is_pf_cgxmapped(rvu, pf) && type != NIX_INTF_TYPE_LBK) + return 0; + + pfvf = rvu_get_pfvf(rvu, pcifunc); + blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); + + bpid_base = rvu_nix_get_bpid(rvu, req, type, chan_id); + chan_base = pfvf->rx_chan_base + req->chan_base; + bpid = bpid_base; + + for (chan = chan_base; chan < (chan_base + req->chan_cnt); chan++) { + if (bpid < 0) { + dev_warn(rvu->dev, "Fail to enable backpressure\n"); + return -EINVAL; + } + + cfg = rvu_read64(rvu, blkaddr, NIX_AF_RX_CHANX_CFG(chan)); + rvu_write64(rvu, blkaddr, NIX_AF_RX_CHANX_CFG(chan), + cfg | (bpid & 0xFF) | BIT_ULL(16)); + chan_id++; + bpid = rvu_nix_get_bpid(rvu, req, type, chan_id); + } + + for (chan = 0; chan < req->chan_cnt; chan++) { + /* Map channel and bpid assign to it */ + rsp->chan_bpid[chan] = ((req->chan_base + chan) & 0x7F) << 10 | + (bpid_base & 0x3FF); + if (req->bpid_per_chan) + bpid_base++; + } + rsp->chan_cnt = req->chan_cnt; + + return 0; +} + static void nix_setup_lso_tso_l3(struct rvu *rvu, int blkaddr, u64 format, bool v4, u64 *fidx) { @@ -545,6 +708,11 @@ static int rvu_nix_aq_enq_inst(struct rvu *rvu, struct nix_aq_enq_req *req, */ inst.res_addr = (u64)aq->res->iova; + /* Hardware uses same aq->res->base for updating result of + * previous instruction hence wait here till it is done. + */ + spin_lock(&aq->lock); + /* Clean result + context memory */ memset(aq->res->base, 0, aq->res->entry_sz); /* Context needs to be written at RES_ADDR + 128 */ @@ -589,11 +757,10 @@ static int rvu_nix_aq_enq_inst(struct rvu *rvu, struct nix_aq_enq_req *req, break; default: rc = NIX_AF_ERR_AQ_ENQUEUE; + spin_unlock(&aq->lock); return rc; } - spin_lock(&aq->lock); - /* Submit the instruction to AQ */ rc = nix_aq_enqueue_wait(rvu, block, &inst); if (rc) { @@ -698,6 +865,8 @@ static int nix_lf_hwctx_disable(struct rvu *rvu, struct hwctx_disable_req *req) if (req->ctype == NIX_AQ_CTYPE_CQ) { aq_req.cq.ena = 0; aq_req.cq_mask.ena = 1; + aq_req.cq.bp_ena = 0; + aq_req.cq_mask.bp_ena = 1; q_cnt = pfvf->cq_ctx->qsize; bmap = pfvf->cq_bmap; } @@ -1667,13 +1836,9 @@ int rvu_mbox_handler_nix_txschq_cfg(struct rvu *rvu, req->num_regs > MAX_REGS_PER_MBOX_MSG) return NIX_AF_INVAL_TXSCHQ_CFG; - err = nix_get_nixlf(rvu, pcifunc, &nixlf); + err = nix_get_nixlf(rvu, pcifunc, &nixlf, &blkaddr); if (err) - return NIX_AF_ERR_AF_LF_INVALID; - - blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); - if (blkaddr < 0) - return NIX_AF_ERR_AF_LF_INVALID; + return err; nix_hw = get_nix_hw(rvu->hw, blkaddr); if (!nix_hw) @@ -1767,17 +1932,12 @@ int rvu_mbox_handler_nix_vtag_cfg(struct rvu *rvu, struct nix_vtag_config *req, struct msg_rsp *rsp) { - struct rvu_hwinfo *hw = rvu->hw; u16 pcifunc = req->hdr.pcifunc; int blkaddr, nixlf, err; - blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); - if (blkaddr < 0) - return NIX_AF_ERR_AF_LF_INVALID; - - nixlf = rvu_get_lf(rvu, &hw->block[blkaddr], pcifunc, 0); - if (nixlf < 0) - return NIX_AF_ERR_AF_LF_INVALID; + err = nix_get_nixlf(rvu, pcifunc, &nixlf, &blkaddr); + if (err) + return err; if (req->cfg_type) { err = nix_rx_vtag_cfg(rvu, nixlf, blkaddr, req); @@ -2119,18 +2279,13 @@ static int nix_af_mark_format_setup(struct rvu *rvu, struct nix_hw *nix_hw, int rvu_mbox_handler_nix_stats_rst(struct rvu *rvu, struct msg_req *req, struct msg_rsp *rsp) { - struct rvu_hwinfo *hw = rvu->hw; u16 pcifunc = req->hdr.pcifunc; - int i, nixlf, blkaddr; + int i, nixlf, blkaddr, err; u64 stats; - blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); - if (blkaddr < 0) - return NIX_AF_ERR_AF_LF_INVALID; - - nixlf = rvu_get_lf(rvu, &hw->block[blkaddr], pcifunc, 0); - if (nixlf < 0) - return NIX_AF_ERR_AF_LF_INVALID; + err = nix_get_nixlf(rvu, pcifunc, &nixlf, &blkaddr); + if (err) + return err; /* Get stats count supported by HW */ stats = rvu_read64(rvu, blkaddr, NIX_AF_CONST1); @@ -2418,18 +2573,14 @@ int rvu_mbox_handler_nix_rss_flowkey_cfg(struct rvu *rvu, struct nix_rss_flowkey_cfg *req, struct nix_rss_flowkey_cfg_rsp *rsp) { - struct rvu_hwinfo *hw = rvu->hw; u16 pcifunc = req->hdr.pcifunc; int alg_idx, nixlf, blkaddr; struct nix_hw *nix_hw; + int err; - blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); - if (blkaddr < 0) - return NIX_AF_ERR_AF_LF_INVALID; - - nixlf = rvu_get_lf(rvu, &hw->block[blkaddr], pcifunc, 0); - if (nixlf < 0) - return NIX_AF_ERR_AF_LF_INVALID; + err = nix_get_nixlf(rvu, pcifunc, &nixlf, &blkaddr); + if (err) + return err; nix_hw = get_nix_hw(rvu->hw, blkaddr); if (!nix_hw) @@ -2522,19 +2673,15 @@ int rvu_mbox_handler_nix_set_mac_addr(struct rvu *rvu, struct nix_set_mac_addr *req, struct msg_rsp *rsp) { - struct rvu_hwinfo *hw = rvu->hw; u16 pcifunc = req->hdr.pcifunc; + int blkaddr, nixlf, err; struct rvu_pfvf *pfvf; - int blkaddr, nixlf; - pfvf = rvu_get_pfvf(rvu, pcifunc); - blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); - if (!pfvf->nixlf || blkaddr < 0) - return NIX_AF_ERR_AF_LF_INVALID; + err = nix_get_nixlf(rvu, pcifunc, &nixlf, &blkaddr); + if (err) + return err; - nixlf = rvu_get_lf(rvu, &hw->block[blkaddr], pcifunc, 0); - if (nixlf < 0) - return NIX_AF_ERR_AF_LF_INVALID; + pfvf = rvu_get_pfvf(rvu, pcifunc); ether_addr_copy(pfvf->mac_addr, req->mac_addr); @@ -2567,19 +2714,15 @@ int rvu_mbox_handler_nix_set_rx_mode(struct rvu *rvu, struct nix_rx_mode *req, struct msg_rsp *rsp) { bool allmulti = false, disable_promisc = false; - struct rvu_hwinfo *hw = rvu->hw; u16 pcifunc = req->hdr.pcifunc; + int blkaddr, nixlf, err; struct rvu_pfvf *pfvf; - int blkaddr, nixlf; - pfvf = rvu_get_pfvf(rvu, pcifunc); - blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); - if (!pfvf->nixlf || blkaddr < 0) - return NIX_AF_ERR_AF_LF_INVALID; + err = nix_get_nixlf(rvu, pcifunc, &nixlf, &blkaddr); + if (err) + return err; - nixlf = rvu_get_lf(rvu, &hw->block[blkaddr], pcifunc, 0); - if (nixlf < 0) - return NIX_AF_ERR_AF_LF_INVALID; + pfvf = rvu_get_pfvf(rvu, pcifunc); if (req->mode & NIX_RX_MODE_PROMISC) allmulti = false; @@ -2794,22 +2937,12 @@ free_entry: int rvu_mbox_handler_nix_set_rx_cfg(struct rvu *rvu, struct nix_rx_cfg *req, struct msg_rsp *rsp) { - struct rvu_hwinfo *hw = rvu->hw; - u16 pcifunc = req->hdr.pcifunc; - struct rvu_block *block; - struct rvu_pfvf *pfvf; - int nixlf, blkaddr; + int nixlf, blkaddr, err; u64 cfg; - pfvf = rvu_get_pfvf(rvu, pcifunc); - blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); - if (!pfvf->nixlf || blkaddr < 0) - return NIX_AF_ERR_AF_LF_INVALID; - - block = &hw->block[blkaddr]; - nixlf = rvu_get_lf(rvu, block, pcifunc, 0); - if (nixlf < 0) - return NIX_AF_ERR_AF_LF_INVALID; + err = nix_get_nixlf(rvu, req->hdr.pcifunc, &nixlf, &blkaddr); + if (err) + return err; cfg = rvu_read64(rvu, blkaddr, NIX_AF_LFX_RX_CFG(nixlf)); /* Set the interface configuration */ @@ -3077,6 +3210,9 @@ int rvu_nix_init(struct rvu *rvu) /* Initialize CGX/LBK/SDP link credits, min/max pkt lengths */ nix_link_config(rvu, blkaddr); + + /* Enable Channel backpressure */ + rvu_write64(rvu, blkaddr, NIX_AF_RX_CFG, BIT_ULL(0)); } return 0; } @@ -3114,30 +3250,13 @@ void rvu_nix_freemem(struct rvu *rvu) } } -int nix_get_nixlf(struct rvu *rvu, u16 pcifunc, int *nixlf) -{ - struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, pcifunc); - struct rvu_hwinfo *hw = rvu->hw; - int blkaddr; - - blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); - if (!pfvf->nixlf || blkaddr < 0) - return NIX_AF_ERR_AF_LF_INVALID; - - *nixlf = rvu_get_lf(rvu, &hw->block[blkaddr], pcifunc, 0); - if (*nixlf < 0) - return NIX_AF_ERR_AF_LF_INVALID; - - return 0; -} - int rvu_mbox_handler_nix_lf_start_rx(struct rvu *rvu, struct msg_req *req, struct msg_rsp *rsp) { u16 pcifunc = req->hdr.pcifunc; int nixlf, err; - err = nix_get_nixlf(rvu, pcifunc, &nixlf); + err = nix_get_nixlf(rvu, pcifunc, &nixlf, NULL); if (err) return err; @@ -3152,7 +3271,7 @@ int rvu_mbox_handler_nix_lf_stop_rx(struct rvu *rvu, struct msg_req *req, u16 pcifunc = req->hdr.pcifunc; int nixlf, err; - err = nix_get_nixlf(rvu, pcifunc, &nixlf); + err = nix_get_nixlf(rvu, pcifunc, &nixlf, NULL); if (err) return err; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c index 6e7c7f459f74..67471cb2b129 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c @@ -94,6 +94,11 @@ int rvu_npa_aq_enq_inst(struct rvu *rvu, struct npa_aq_enq_req *req, */ inst.res_addr = (u64)aq->res->iova; + /* Hardware uses same aq->res->base for updating result of + * previous instruction hence wait here till it is done. + */ + spin_lock(&aq->lock); + /* Clean result + context memory */ memset(aq->res->base, 0, aq->res->entry_sz); /* Context needs to be written at RES_ADDR + 128 */ @@ -138,10 +143,10 @@ int rvu_npa_aq_enq_inst(struct rvu *rvu, struct npa_aq_enq_req *req, break; } - if (rc) + if (rc) { + spin_unlock(&aq->lock); return rc; - - spin_lock(&aq->lock); + } /* Submit the instruction to AQ */ rc = npa_aq_enqueue_wait(rvu, block, &inst); @@ -218,6 +223,8 @@ static int npa_lf_hwctx_disable(struct rvu *rvu, struct hwctx_disable_req *req) } else if (req->ctype == NPA_AQ_CTYPE_AURA) { aq_req.aura.ena = 0; aq_req.aura_mask.ena = 1; + aq_req.aura.bp_ena = 0; + aq_req.aura_mask.bp_ena = 1; cnt = pfvf->aura_ctx->qsize; bmap = pfvf->aura_bmap; } diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c index 40e431debbe9..0a214084406a 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c @@ -825,8 +825,10 @@ static void npc_load_mkex_profile(struct rvu *rvu, int blkaddr) if (!strncmp(mkex_profile, "default", MKEX_NAME_LEN)) goto load_default; - if (cgx_get_mkex_prfl_info(&prfl_addr, &prfl_sz)) + if (!rvu->fwdata) goto load_default; + prfl_addr = rvu->fwdata->mcam_addr; + prfl_sz = rvu->fwdata->mcam_sz; if (!prfl_addr || !prfl_sz) goto load_default; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h index 9d8942acc232..a3ecb5de9000 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h @@ -11,6 +11,9 @@ #ifndef RVU_STRUCT_H #define RVU_STRUCT_H +/* RVU Block revision IDs */ +#define RVU_BLK_RVUM_REVID 0x01 + /* RVU Block Address Enumeration */ enum rvu_block_addr_e { BLKADDR_RVUM = 0x0ULL, diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile index 41bf00cf5b1d..778df331c8ac 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile +++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile @@ -4,7 +4,9 @@ # obj-$(CONFIG_OCTEONTX2_PF) += octeontx2_nicpf.o +obj-$(CONFIG_OCTEONTX2_VF) += octeontx2_nicvf.o octeontx2_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o otx2_ethtool.o +octeontx2_nicvf-y := otx2_vf.o ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c index b945bd3d5d88..f1d2dea90a8c 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c @@ -49,15 +49,15 @@ void otx2_update_lmac_stats(struct otx2_nic *pfvf) if (!netif_running(pfvf->netdev)) return; - otx2_mbox_lock(&pfvf->mbox); + mutex_lock(&pfvf->mbox.lock); req = otx2_mbox_alloc_msg_cgx_stats(&pfvf->mbox); if (!req) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return; } otx2_sync_mbox_msg(&pfvf->mbox); - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); } int otx2_update_rq_stats(struct otx2_nic *pfvf, int qidx) @@ -128,6 +128,7 @@ void otx2_get_stats64(struct net_device *netdev, stats->tx_packets = dev_stats->tx_frames; stats->tx_dropped = dev_stats->tx_drops; } +EXPORT_SYMBOL(otx2_get_stats64); /* Sync MAC address with RVU AF */ static int otx2_hw_set_mac_addr(struct otx2_nic *pfvf, u8 *mac) @@ -135,17 +136,17 @@ static int otx2_hw_set_mac_addr(struct otx2_nic *pfvf, u8 *mac) struct nix_set_mac_addr *req; int err; - otx2_mbox_lock(&pfvf->mbox); + mutex_lock(&pfvf->mbox.lock); req = otx2_mbox_alloc_msg_nix_set_mac_addr(&pfvf->mbox); if (!req) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return -ENOMEM; } ether_addr_copy(req->mac_addr, mac); err = otx2_sync_mbox_msg(&pfvf->mbox); - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return err; } @@ -157,27 +158,27 @@ static int otx2_hw_get_mac_addr(struct otx2_nic *pfvf, struct msg_req *req; int err; - otx2_mbox_lock(&pfvf->mbox); + mutex_lock(&pfvf->mbox.lock); req = otx2_mbox_alloc_msg_nix_get_mac_addr(&pfvf->mbox); if (!req) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return -ENOMEM; } err = otx2_sync_mbox_msg(&pfvf->mbox); if (err) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return err; } msghdr = otx2_mbox_get_rsp(&pfvf->mbox.mbox, 0, &req->hdr); if (IS_ERR(msghdr)) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return PTR_ERR(msghdr); } rsp = (struct nix_get_mac_addr_rsp *)msghdr; ether_addr_copy(netdev->dev_addr, rsp->mac_addr); - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return 0; } @@ -197,26 +198,50 @@ int otx2_set_mac_address(struct net_device *netdev, void *p) return 0; } +EXPORT_SYMBOL(otx2_set_mac_address); int otx2_hw_set_mtu(struct otx2_nic *pfvf, int mtu) { struct nix_frs_cfg *req; int err; - otx2_mbox_lock(&pfvf->mbox); + mutex_lock(&pfvf->mbox.lock); req = otx2_mbox_alloc_msg_nix_set_hw_frs(&pfvf->mbox); if (!req) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return -ENOMEM; } - /* SMQ config limits maximum pkt size that can be transmitted */ - req->update_smq = true; pfvf->max_frs = mtu + OTX2_ETH_HLEN; req->maxlen = pfvf->max_frs; err = otx2_sync_mbox_msg(&pfvf->mbox); - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); + return err; +} + +int otx2_config_pause_frm(struct otx2_nic *pfvf) +{ + struct cgx_pause_frm_cfg *req; + int err; + + if (is_otx2_lbkvf(pfvf->pdev)) + return 0; + + mutex_lock(&pfvf->mbox.lock); + req = otx2_mbox_alloc_msg_cgx_cfg_pause_frm(&pfvf->mbox); + if (!req) { + err = -ENOMEM; + goto unlock; + } + + req->rx_pause = !!(pfvf->flags & OTX2_FLAG_RX_PAUSE_ENABLED); + req->tx_pause = !!(pfvf->flags & OTX2_FLAG_TX_PAUSE_ENABLED); + req->set = 1; + + err = otx2_sync_mbox_msg(&pfvf->mbox); +unlock: + mutex_unlock(&pfvf->mbox.lock); return err; } @@ -226,10 +251,10 @@ int otx2_set_flowkey_cfg(struct otx2_nic *pfvf) struct nix_rss_flowkey_cfg *req; int err; - otx2_mbox_lock(&pfvf->mbox); + mutex_lock(&pfvf->mbox.lock); req = otx2_mbox_alloc_msg_nix_rss_flowkey_cfg(&pfvf->mbox); if (!req) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return -ENOMEM; } req->mcam_index = -1; /* Default or reserved index */ @@ -237,7 +262,7 @@ int otx2_set_flowkey_cfg(struct otx2_nic *pfvf) req->group = DEFAULT_RSS_CONTEXT_GROUP; err = otx2_sync_mbox_msg(&pfvf->mbox); - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return err; } @@ -248,7 +273,7 @@ int otx2_set_rss_table(struct otx2_nic *pfvf) struct nix_aq_enq_req *aq; int idx, err; - otx2_mbox_lock(mbox); + mutex_lock(&mbox->lock); /* Get memory to put this msg */ for (idx = 0; idx < rss->rss_size; idx++) { aq = otx2_mbox_alloc_msg_nix_aq_enq(mbox); @@ -258,12 +283,12 @@ int otx2_set_rss_table(struct otx2_nic *pfvf) */ err = otx2_sync_mbox_msg(mbox); if (err) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); return err; } aq = otx2_mbox_alloc_msg_nix_aq_enq(mbox); if (!aq) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); return -ENOMEM; } } @@ -276,7 +301,7 @@ int otx2_set_rss_table(struct otx2_nic *pfvf) aq->op = NIX_AQ_INSTOP_INIT; } err = otx2_sync_mbox_msg(mbox); - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); return err; } @@ -394,6 +419,7 @@ void otx2_tx_timeout(struct net_device *netdev, unsigned int txq) schedule_work(&pfvf->reset_task); } +EXPORT_SYMBOL(otx2_tx_timeout); void otx2_get_mac_from_af(struct net_device *netdev) { @@ -408,6 +434,7 @@ void otx2_get_mac_from_af(struct net_device *netdev) if (!is_valid_ether_addr(netdev->dev_addr)) eth_hw_addr_random(netdev); } +EXPORT_SYMBOL(otx2_get_mac_from_af); static int otx2_get_link(struct otx2_nic *pfvf) { @@ -443,7 +470,7 @@ int otx2_txschq_config(struct otx2_nic *pfvf, int lvl) /* Set topology e.t.c configuration */ if (lvl == NIX_TXSCH_LVL_SMQ) { req->reg[0] = NIX_AF_SMQX_CFG(schq); - req->regval[0] = ((pfvf->netdev->mtu + OTX2_ETH_HLEN) << 8) | + req->regval[0] = ((OTX2_MAX_MTU + OTX2_ETH_HLEN) << 8) | OTX2_MIN_MTU; req->regval[0] |= (0x20ULL << 51) | (0x80ULL << 39) | @@ -529,17 +556,17 @@ int otx2_txschq_stop(struct otx2_nic *pfvf) struct nix_txsch_free_req *free_req; int lvl, schq, err; - otx2_mbox_lock(&pfvf->mbox); + mutex_lock(&pfvf->mbox.lock); /* Free the transmit schedulers */ free_req = otx2_mbox_alloc_msg_nix_txsch_free(&pfvf->mbox); if (!free_req) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return -ENOMEM; } free_req->flags = TXSCHQ_FREE_ALL; err = otx2_sync_mbox_msg(&pfvf->mbox); - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); /* Clear the txschq list */ for (lvl = 0; lvl < NIX_TXSCH_LVL_CNT; lvl++) { @@ -553,17 +580,19 @@ void otx2_sqb_flush(struct otx2_nic *pfvf) { int qidx, sqe_tail, sqe_head; u64 incr, *ptr, val; + int timeout = 1000; ptr = (u64 *)otx2_get_regaddr(pfvf, NIX_LF_SQ_OP_STATUS); for (qidx = 0; qidx < pfvf->hw.tx_queues; qidx++) { incr = (u64)qidx << 32; - while (1) { + while (timeout) { val = otx2_atomic64_add(incr, ptr); sqe_head = (val >> 20) & 0x3F; sqe_tail = (val >> 28) & 0x3F; if (sqe_head == sqe_tail) break; usleep_range(1, 3); + timeout--; } } } @@ -580,6 +609,7 @@ void otx2_sqb_flush(struct otx2_nic *pfvf) * RED accepts pkts if free pointers > 102 & <= 205. * Drops pkts if free pointers < 102. */ +#define RQ_BP_LVL_AURA (255 - ((85 * 256) / 100)) /* BP when 85% is full */ #define RQ_PASS_LVL_AURA (255 - ((95 * 256) / 100)) /* RED when 95% is full */ #define RQ_DROP_LVL_AURA (255 - ((99 * 256) / 100)) /* Drop when 99% is full */ @@ -741,6 +771,13 @@ static int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx) if (qidx < pfvf->hw.rx_queues) { aq->cq.drop = RQ_DROP_LVL_CQ(pfvf->hw.rq_skid, cq->cqe_cnt); aq->cq.drop_ena = 1; + + /* Enable receive CQ backpressure */ + aq->cq.bp_ena = 1; + aq->cq.bpid = pfvf->bpid[0]; + + /* Set backpressure level is same as cq pass level */ + aq->cq.bp = RQ_PASS_LVL_CQ(pfvf->hw.rq_skid, qset->rqe_cnt); } /* Fill AQ info */ @@ -951,6 +988,7 @@ void otx2_aura_pool_free(struct otx2_nic *pfvf) qmem_free(pfvf->dev, pool->fc_addr); } devm_kfree(pfvf->dev, pfvf->qset.pool); + pfvf->qset.pool = NULL; } static int otx2_aura_init(struct otx2_nic *pfvf, int aura_id, @@ -996,6 +1034,14 @@ static int otx2_aura_init(struct otx2_nic *pfvf, int aura_id, aq->aura.fc_addr = pool->fc_addr->iova; aq->aura.fc_hyst_bits = 0; /* Store count on all updates */ + /* Enable backpressure for RQ aura */ + if (aura_id < pfvf->hw.rqpool_cnt) { + aq->aura.bp_ena = 0; + aq->aura.nix0_bpid = pfvf->bpid[0]; + /* Set backpressure level for RQ's Aura */ + aq->aura.bp = RQ_BP_LVL_AURA; + } + /* Fill AQ info */ aq->ctype = NPA_AQ_CTYPE_AURA; aq->op = NPA_AQ_INSTOP_INIT; @@ -1210,10 +1256,10 @@ int otx2_detach_resources(struct mbox *mbox) { struct rsrc_detach *detach; - otx2_mbox_lock(mbox); + mutex_lock(&mbox->lock); detach = otx2_mbox_alloc_msg_detach_resources(mbox); if (!detach) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); return -ENOMEM; } @@ -1222,9 +1268,10 @@ int otx2_detach_resources(struct mbox *mbox) /* Send detach request to AF */ otx2_mbox_msg_send(&mbox->mbox, 0); - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); return 0; } +EXPORT_SYMBOL(otx2_detach_resources); int otx2_attach_npa_nix(struct otx2_nic *pfvf) { @@ -1232,11 +1279,11 @@ int otx2_attach_npa_nix(struct otx2_nic *pfvf) struct msg_req *msix; int err; - otx2_mbox_lock(&pfvf->mbox); + mutex_lock(&pfvf->mbox.lock); /* Get memory to put this msg */ attach = otx2_mbox_alloc_msg_attach_resources(&pfvf->mbox); if (!attach) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return -ENOMEM; } @@ -1246,7 +1293,7 @@ int otx2_attach_npa_nix(struct otx2_nic *pfvf) /* Send attach request to AF */ err = otx2_sync_mbox_msg(&pfvf->mbox); if (err) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return err; } @@ -1261,16 +1308,16 @@ int otx2_attach_npa_nix(struct otx2_nic *pfvf) /* Get NPA and NIX MSIX vector offsets */ msix = otx2_mbox_alloc_msg_msix_offset(&pfvf->mbox); if (!msix) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return -ENOMEM; } err = otx2_sync_mbox_msg(&pfvf->mbox); if (err) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return err; } - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); if (pfvf->hw.npa_msixoff == MSIX_VECTOR_INVALID || pfvf->hw.nix_msixoff == MSIX_VECTOR_INVALID) { @@ -1281,12 +1328,13 @@ int otx2_attach_npa_nix(struct otx2_nic *pfvf) return 0; } +EXPORT_SYMBOL(otx2_attach_npa_nix); void otx2_ctx_disable(struct mbox *mbox, int type, bool npa) { struct hwctx_disable_req *req; - otx2_mbox_lock(mbox); + mutex_lock(&mbox->lock); /* Request AQ to disable this context */ if (npa) req = otx2_mbox_alloc_msg_npa_hwctx_disable(mbox); @@ -1294,7 +1342,7 @@ void otx2_ctx_disable(struct mbox *mbox, int type, bool npa) req = otx2_mbox_alloc_msg_nix_hwctx_disable(mbox); if (!req) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); return; } @@ -1304,7 +1352,26 @@ void otx2_ctx_disable(struct mbox *mbox, int type, bool npa) dev_err(mbox->pfvf->dev, "%s failed to disable context\n", __func__); - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); +} + +int otx2_nix_config_bp(struct otx2_nic *pfvf, bool enable) +{ + struct nix_bp_cfg_req *req; + + if (enable) + req = otx2_mbox_alloc_msg_nix_bp_enable(&pfvf->mbox); + else + req = otx2_mbox_alloc_msg_nix_bp_disable(&pfvf->mbox); + + if (!req) + return -ENOMEM; + + req->chan_base = 0; + req->chan_cnt = 1; + req->bpid_per_chan = 0; + + return otx2_sync_mbox_msg(&pfvf->mbox); } /* Mbox message handlers */ @@ -1330,6 +1397,7 @@ void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf, pf->hw.txschq_list[lvl][schq] = rsp->schq_list[lvl][schq]; } +EXPORT_SYMBOL(mbox_handler_nix_txsch_alloc); void mbox_handler_npa_lf_alloc(struct otx2_nic *pfvf, struct npa_lf_alloc_rsp *rsp) @@ -1337,6 +1405,7 @@ void mbox_handler_npa_lf_alloc(struct otx2_nic *pfvf, pfvf->hw.stack_pg_ptrs = rsp->stack_pg_ptrs; pfvf->hw.stack_pg_bytes = rsp->stack_pg_bytes; } +EXPORT_SYMBOL(mbox_handler_npa_lf_alloc); void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf, struct nix_lf_alloc_rsp *rsp) @@ -1347,6 +1416,7 @@ void mbox_handler_nix_lf_alloc(struct otx2_nic *pfvf, pfvf->hw.lso_tsov4_idx = rsp->lso_tsov4_idx; pfvf->hw.lso_tsov6_idx = rsp->lso_tsov6_idx; } +EXPORT_SYMBOL(mbox_handler_nix_lf_alloc); void mbox_handler_msix_offset(struct otx2_nic *pfvf, struct msix_offset_rsp *rsp) @@ -1354,6 +1424,19 @@ void mbox_handler_msix_offset(struct otx2_nic *pfvf, pfvf->hw.npa_msixoff = rsp->npa_msixoff; pfvf->hw.nix_msixoff = rsp->nix_msixoff; } +EXPORT_SYMBOL(mbox_handler_msix_offset); + +void mbox_handler_nix_bp_enable(struct otx2_nic *pfvf, + struct nix_bp_cfg_rsp *rsp) +{ + int chan, chan_id; + + for (chan = 0; chan < rsp->chan_cnt; chan++) { + chan_id = ((rsp->chan_bpid[chan] >> 10) & 0x7F); + pfvf->bpid[chan_id] = rsp->chan_bpid[chan] & 0x3FF; + } +} +EXPORT_SYMBOL(mbox_handler_nix_bp_enable); void otx2_free_cints(struct otx2_nic *pfvf, int n) { diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h index 320f3b7bf57f..018c283a0ac4 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h @@ -20,6 +20,8 @@ /* PCI device IDs */ #define PCI_DEVID_OCTEONTX2_RVU_PF 0xA063 +#define PCI_DEVID_OCTEONTX2_RVU_VF 0xA064 +#define PCI_DEVID_OCTEONTX2_RVU_AFVF 0xA0F8 #define PCI_SUBSYS_DEVID_96XX_RVU_PFVF 0xB200 @@ -191,6 +193,17 @@ struct otx2_hw { u64 cgx_tx_stats[CGX_TX_STATS_COUNT]; }; +struct otx2_vf_config { + struct otx2_nic *pf; + struct delayed_work link_event_work; + bool intf_down; /* interface was either configured or not */ +}; + +struct flr_work { + struct work_struct work; + struct otx2_nic *pf; +}; + struct refill_work { struct delayed_work pool_refill_work; struct otx2_nic *pf; @@ -204,6 +217,8 @@ struct otx2_nic { u16 rbsize; /* Receive buffer size */ #define OTX2_FLAG_INTF_DOWN BIT_ULL(2) +#define OTX2_FLAG_RX_PAUSE_ENABLED BIT_ULL(9) +#define OTX2_FLAG_TX_PAUSE_ENABLED BIT_ULL(10) u64 flags; struct otx2_qset qset; @@ -213,14 +228,23 @@ struct otx2_nic { /* Mbox */ struct mbox mbox; + struct mbox *mbox_pfvf; struct workqueue_struct *mbox_wq; + struct workqueue_struct *mbox_pfvf_wq; + u8 total_vfs; u16 pcifunc; /* RVU PF_FUNC */ + u16 bpid[NIX_MAX_BPID_CHAN]; + struct otx2_vf_config *vf_configs; struct cgx_link_user_info linfo; u64 reset_count; struct work_struct reset_task; + struct workqueue_struct *flr_wq; + struct flr_work *flr_wrk; struct refill_work *refill_wrk; + struct workqueue_struct *otx2_wq; + struct work_struct rx_mode_work; /* Ethtool stuff */ u32 msg_enable; @@ -229,6 +253,11 @@ struct otx2_nic { int nix_blkaddr; }; +static inline bool is_otx2_lbkvf(struct pci_dev *pdev) +{ + return pdev->device == PCI_DEVID_OCTEONTX2_RVU_AFVF; +} + static inline bool is_96xx_A0(struct pci_dev *pdev) { return (pdev->revision == 0x00) && @@ -348,21 +377,6 @@ static inline void otx2_sync_mbox_bbuf(struct otx2_mbox *mbox, int devid) hw_mbase + mbox->rx_start, msg_size + msgs_offset); } -static inline void otx2_mbox_lock_init(struct mbox *mbox) -{ - mutex_init(&mbox->lock); -} - -static inline void otx2_mbox_lock(struct mbox *mbox) -{ - mutex_lock(&mbox->lock); -} - -static inline void otx2_mbox_unlock(struct mbox *mbox) -{ - mutex_unlock(&mbox->lock); -} - /* With the absence of API for 128-bit IO memory access for arm64, * implement required operations at place. */ @@ -558,6 +572,7 @@ int otx2_hw_set_mtu(struct otx2_nic *pfvf, int mtu); void otx2_tx_timeout(struct net_device *netdev, unsigned int txq); void otx2_get_mac_from_af(struct net_device *netdev); void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx); +int otx2_config_pause_frm(struct otx2_nic *pfvf); /* RVU block related APIs */ int otx2_attach_npa_nix(struct otx2_nic *pfvf); @@ -578,6 +593,7 @@ dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool, gfp_t gfp); int otx2_rxtx_enable(struct otx2_nic *pfvf, bool enable); void otx2_ctx_disable(struct mbox *mbox, int type, bool npa); +int otx2_nix_config_bp(struct otx2_nic *pfvf, bool enable); void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq); void otx2_cleanup_tx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq); @@ -598,6 +614,8 @@ void mbox_handler_nix_txsch_alloc(struct otx2_nic *pf, struct nix_txsch_alloc_rsp *rsp); void mbox_handler_cgx_stats(struct otx2_nic *pfvf, struct cgx_stats_rsp *rsp); +void mbox_handler_nix_bp_enable(struct otx2_nic *pfvf, + struct nix_bp_cfg_rsp *rsp); /* Device stats APIs */ void otx2_get_dev_stats(struct otx2_nic *pfvf); @@ -607,6 +625,7 @@ void otx2_update_lmac_stats(struct otx2_nic *pfvf); int otx2_update_rq_stats(struct otx2_nic *pfvf, int qidx); int otx2_update_sq_stats(struct otx2_nic *pfvf, int qidx); void otx2_set_ethtool_ops(struct net_device *netdev); +void otx2vf_set_ethtool_ops(struct net_device *netdev); int otx2_open(struct net_device *netdev); int otx2_stop(struct net_device *netdev); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c index 60fcf82dd8cb..d59f5a9c7273 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c @@ -17,6 +17,7 @@ #include "otx2_common.h" #define DRV_NAME "octeontx2-nicpf" +#define DRV_VF_NAME "octeontx2-nicvf" struct otx2_stat { char name[ETH_GSTRING_LEN]; @@ -63,16 +64,6 @@ static const unsigned int otx2_n_dev_stats = ARRAY_SIZE(otx2_dev_stats); static const unsigned int otx2_n_drv_stats = ARRAY_SIZE(otx2_drv_stats); static const unsigned int otx2_n_queue_stats = ARRAY_SIZE(otx2_queue_stats); -static void otx2_dev_open(struct net_device *netdev) -{ - otx2_open(netdev); -} - -static void otx2_dev_stop(struct net_device *netdev) -{ - otx2_stop(netdev); -} - static void otx2_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) { @@ -232,7 +223,7 @@ static int otx2_set_channels(struct net_device *dev, return -EINVAL; if (if_up) - otx2_dev_stop(dev); + dev->netdev_ops->ndo_stop(dev); err = otx2_set_real_num_queues(dev, channel->tx_count, channel->rx_count); @@ -245,7 +236,7 @@ static int otx2_set_channels(struct net_device *dev, fail: if (if_up) - otx2_dev_open(dev); + dev->netdev_ops->ndo_open(dev); netdev_info(dev, "Setting num Tx rings to %d, Rx rings to %d success\n", pfvf->hw.tx_queues, pfvf->hw.rx_queues); @@ -253,6 +244,51 @@ fail: return err; } +static void otx2_get_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) +{ + struct otx2_nic *pfvf = netdev_priv(netdev); + struct cgx_pause_frm_cfg *req, *rsp; + + if (is_otx2_lbkvf(pfvf->pdev)) + return; + + req = otx2_mbox_alloc_msg_cgx_cfg_pause_frm(&pfvf->mbox); + if (!req) + return; + + if (!otx2_sync_mbox_msg(&pfvf->mbox)) { + rsp = (struct cgx_pause_frm_cfg *) + otx2_mbox_get_rsp(&pfvf->mbox.mbox, 0, &req->hdr); + pause->rx_pause = rsp->rx_pause; + pause->tx_pause = rsp->tx_pause; + } +} + +static int otx2_set_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) +{ + struct otx2_nic *pfvf = netdev_priv(netdev); + + if (pause->autoneg) + return -EOPNOTSUPP; + + if (is_otx2_lbkvf(pfvf->pdev)) + return -EOPNOTSUPP; + + if (pause->rx_pause) + pfvf->flags |= OTX2_FLAG_RX_PAUSE_ENABLED; + else + pfvf->flags &= ~OTX2_FLAG_RX_PAUSE_ENABLED; + + if (pause->tx_pause) + pfvf->flags |= OTX2_FLAG_TX_PAUSE_ENABLED; + else + pfvf->flags &= ~OTX2_FLAG_TX_PAUSE_ENABLED; + + return otx2_config_pause_frm(pfvf); +} + static void otx2_get_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring) { @@ -297,14 +333,15 @@ static int otx2_set_ringparam(struct net_device *netdev, return 0; if (if_up) - otx2_dev_stop(netdev); + netdev->netdev_ops->ndo_stop(netdev); /* Assigned to the nearest possible exponent. */ qs->sqe_cnt = tx_count; qs->rqe_cnt = rx_count; if (if_up) - otx2_dev_open(netdev); + netdev->netdev_ops->ndo_open(netdev); + return 0; } @@ -329,17 +366,6 @@ static int otx2_set_coalesce(struct net_device *netdev, struct otx2_hw *hw = &pfvf->hw; int qidx; - if (ec->use_adaptive_rx_coalesce || ec->use_adaptive_tx_coalesce || - ec->rx_coalesce_usecs_irq || ec->rx_max_coalesced_frames_irq || - ec->tx_coalesce_usecs_irq || ec->tx_max_coalesced_frames_irq || - ec->stats_block_coalesce_usecs || ec->pkt_rate_low || - ec->rx_coalesce_usecs_low || ec->rx_max_coalesced_frames_low || - ec->tx_coalesce_usecs_low || ec->tx_max_coalesced_frames_low || - ec->pkt_rate_high || ec->rx_coalesce_usecs_high || - ec->rx_max_coalesced_frames_high || ec->tx_coalesce_usecs_high || - ec->tx_max_coalesced_frames_high || ec->rate_sample_interval) - return -EOPNOTSUPP; - if (!ec->rx_max_coalesced_frames || !ec->tx_max_coalesced_frames) return 0; @@ -631,10 +657,15 @@ static u32 otx2_get_link(struct net_device *netdev) { struct otx2_nic *pfvf = netdev_priv(netdev); + /* LBK link is internal and always UP */ + if (is_otx2_lbkvf(pfvf->pdev)) + return 1; return pfvf->linfo.link_up; } static const struct ethtool_ops otx2_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_link = otx2_get_link, .get_drvinfo = otx2_get_drvinfo, .get_strings = otx2_get_strings, @@ -654,9 +685,110 @@ static const struct ethtool_ops otx2_ethtool_ops = { .set_rxfh = otx2_set_rxfh, .get_msglevel = otx2_get_msglevel, .set_msglevel = otx2_set_msglevel, + .get_pauseparam = otx2_get_pauseparam, + .set_pauseparam = otx2_set_pauseparam, }; void otx2_set_ethtool_ops(struct net_device *netdev) { netdev->ethtool_ops = &otx2_ethtool_ops; } + +/* VF's ethtool APIs */ +static void otx2vf_get_drvinfo(struct net_device *netdev, + struct ethtool_drvinfo *info) +{ + struct otx2_nic *vf = netdev_priv(netdev); + + strlcpy(info->driver, DRV_VF_NAME, sizeof(info->driver)); + strlcpy(info->bus_info, pci_name(vf->pdev), sizeof(info->bus_info)); +} + +static void otx2vf_get_strings(struct net_device *netdev, u32 sset, u8 *data) +{ + struct otx2_nic *vf = netdev_priv(netdev); + int stats; + + if (sset != ETH_SS_STATS) + return; + + for (stats = 0; stats < otx2_n_dev_stats; stats++) { + memcpy(data, otx2_dev_stats[stats].name, ETH_GSTRING_LEN); + data += ETH_GSTRING_LEN; + } + + for (stats = 0; stats < otx2_n_drv_stats; stats++) { + memcpy(data, otx2_drv_stats[stats].name, ETH_GSTRING_LEN); + data += ETH_GSTRING_LEN; + } + + otx2_get_qset_strings(vf, &data, 0); + + strcpy(data, "reset_count"); + data += ETH_GSTRING_LEN; +} + +static void otx2vf_get_ethtool_stats(struct net_device *netdev, + struct ethtool_stats *stats, u64 *data) +{ + struct otx2_nic *vf = netdev_priv(netdev); + int stat; + + otx2_get_dev_stats(vf); + for (stat = 0; stat < otx2_n_dev_stats; stat++) + *(data++) = ((u64 *)&vf->hw.dev_stats) + [otx2_dev_stats[stat].index]; + + for (stat = 0; stat < otx2_n_drv_stats; stat++) + *(data++) = atomic_read(&((atomic_t *)&vf->hw.drv_stats) + [otx2_drv_stats[stat].index]); + + otx2_get_qset_stats(vf, stats, &data); + *(data++) = vf->reset_count; +} + +static int otx2vf_get_sset_count(struct net_device *netdev, int sset) +{ + struct otx2_nic *vf = netdev_priv(netdev); + int qstats_count; + + if (sset != ETH_SS_STATS) + return -EINVAL; + + qstats_count = otx2_n_queue_stats * + (vf->hw.rx_queues + vf->hw.tx_queues); + + return otx2_n_dev_stats + otx2_n_drv_stats + qstats_count + 1; +} + +static const struct ethtool_ops otx2vf_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, + .get_link = otx2_get_link, + .get_drvinfo = otx2vf_get_drvinfo, + .get_strings = otx2vf_get_strings, + .get_ethtool_stats = otx2vf_get_ethtool_stats, + .get_sset_count = otx2vf_get_sset_count, + .set_channels = otx2_set_channels, + .get_channels = otx2_get_channels, + .get_rxnfc = otx2_get_rxnfc, + .set_rxnfc = otx2_set_rxnfc, + .get_rxfh_key_size = otx2_get_rxfh_key_size, + .get_rxfh_indir_size = otx2_get_rxfh_indir_size, + .get_rxfh = otx2_get_rxfh, + .set_rxfh = otx2_set_rxfh, + .get_ringparam = otx2_get_ringparam, + .set_ringparam = otx2_set_ringparam, + .get_coalesce = otx2_get_coalesce, + .set_coalesce = otx2_set_coalesce, + .get_msglevel = otx2_get_msglevel, + .set_msglevel = otx2_set_msglevel, + .get_pauseparam = otx2_get_pauseparam, + .set_pauseparam = otx2_set_pauseparam, +}; + +void otx2vf_set_ethtool_ops(struct net_device *netdev) +{ + netdev->ethtool_ops = &otx2vf_ethtool_ops; +} +EXPORT_SYMBOL(otx2vf_set_ethtool_ops); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c index 85f9b9ba6bd5..411e5ea1031e 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c @@ -24,7 +24,6 @@ #define DRV_NAME "octeontx2-nicpf" #define DRV_STRING "Marvell OcteonTX2 NIC Physical Function Driver" -#define DRV_VERSION "1.0" /* Supported devices */ static const struct pci_device_id otx2_pf_id_table[] = { @@ -32,10 +31,9 @@ static const struct pci_device_id otx2_pf_id_table[] = { { 0, } /* end of table */ }; -MODULE_AUTHOR("Marvell International Ltd."); +MODULE_AUTHOR("Sunil Goutham <sgoutham@marvell.com>"); MODULE_DESCRIPTION(DRV_STRING); MODULE_LICENSE("GPL v2"); -MODULE_VERSION(DRV_VERSION); MODULE_DEVICE_TABLE(pci, otx2_pf_id_table); enum { @@ -61,6 +59,224 @@ static int otx2_change_mtu(struct net_device *netdev, int new_mtu) return err; } +static void otx2_disable_flr_me_intr(struct otx2_nic *pf) +{ + int irq, vfs = pf->total_vfs; + + /* Disable VFs ME interrupts */ + otx2_write64(pf, RVU_PF_VFME_INT_ENA_W1CX(0), INTR_MASK(vfs)); + irq = pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFME0); + free_irq(irq, pf); + + /* Disable VFs FLR interrupts */ + otx2_write64(pf, RVU_PF_VFFLR_INT_ENA_W1CX(0), INTR_MASK(vfs)); + irq = pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFFLR0); + free_irq(irq, pf); + + if (vfs <= 64) + return; + + otx2_write64(pf, RVU_PF_VFME_INT_ENA_W1CX(1), INTR_MASK(vfs - 64)); + irq = pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFME1); + free_irq(irq, pf); + + otx2_write64(pf, RVU_PF_VFFLR_INT_ENA_W1CX(1), INTR_MASK(vfs - 64)); + irq = pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFFLR1); + free_irq(irq, pf); +} + +static void otx2_flr_wq_destroy(struct otx2_nic *pf) +{ + if (!pf->flr_wq) + return; + destroy_workqueue(pf->flr_wq); + pf->flr_wq = NULL; + devm_kfree(pf->dev, pf->flr_wrk); +} + +static void otx2_flr_handler(struct work_struct *work) +{ + struct flr_work *flrwork = container_of(work, struct flr_work, work); + struct otx2_nic *pf = flrwork->pf; + struct mbox *mbox = &pf->mbox; + struct msg_req *req; + int vf, reg = 0; + + vf = flrwork - pf->flr_wrk; + + mutex_lock(&mbox->lock); + req = otx2_mbox_alloc_msg_vf_flr(mbox); + if (!req) { + mutex_unlock(&mbox->lock); + return; + } + req->hdr.pcifunc &= RVU_PFVF_FUNC_MASK; + req->hdr.pcifunc |= (vf + 1) & RVU_PFVF_FUNC_MASK; + + if (!otx2_sync_mbox_msg(&pf->mbox)) { + if (vf >= 64) { + reg = 1; + vf = vf - 64; + } + /* clear transcation pending bit */ + otx2_write64(pf, RVU_PF_VFTRPENDX(reg), BIT_ULL(vf)); + otx2_write64(pf, RVU_PF_VFFLR_INT_ENA_W1SX(reg), BIT_ULL(vf)); + } + + mutex_unlock(&mbox->lock); +} + +static irqreturn_t otx2_pf_flr_intr_handler(int irq, void *pf_irq) +{ + struct otx2_nic *pf = (struct otx2_nic *)pf_irq; + int reg, dev, vf, start_vf, num_reg = 1; + u64 intr; + + if (pf->total_vfs > 64) + num_reg = 2; + + for (reg = 0; reg < num_reg; reg++) { + intr = otx2_read64(pf, RVU_PF_VFFLR_INTX(reg)); + if (!intr) + continue; + start_vf = 64 * reg; + for (vf = 0; vf < 64; vf++) { + if (!(intr & BIT_ULL(vf))) + continue; + dev = vf + start_vf; + queue_work(pf->flr_wq, &pf->flr_wrk[dev].work); + /* Clear interrupt */ + otx2_write64(pf, RVU_PF_VFFLR_INTX(reg), BIT_ULL(vf)); + /* Disable the interrupt */ + otx2_write64(pf, RVU_PF_VFFLR_INT_ENA_W1CX(reg), + BIT_ULL(vf)); + } + } + return IRQ_HANDLED; +} + +static irqreturn_t otx2_pf_me_intr_handler(int irq, void *pf_irq) +{ + struct otx2_nic *pf = (struct otx2_nic *)pf_irq; + int vf, reg, num_reg = 1; + u64 intr; + + if (pf->total_vfs > 64) + num_reg = 2; + + for (reg = 0; reg < num_reg; reg++) { + intr = otx2_read64(pf, RVU_PF_VFME_INTX(reg)); + if (!intr) + continue; + for (vf = 0; vf < 64; vf++) { + if (!(intr & BIT_ULL(vf))) + continue; + /* clear trpend bit */ + otx2_write64(pf, RVU_PF_VFTRPENDX(reg), BIT_ULL(vf)); + /* clear interrupt */ + otx2_write64(pf, RVU_PF_VFME_INTX(reg), BIT_ULL(vf)); + } + } + return IRQ_HANDLED; +} + +static int otx2_register_flr_me_intr(struct otx2_nic *pf, int numvfs) +{ + struct otx2_hw *hw = &pf->hw; + char *irq_name; + int ret; + + /* Register ME interrupt handler*/ + irq_name = &hw->irq_name[RVU_PF_INT_VEC_VFME0 * NAME_SIZE]; + snprintf(irq_name, NAME_SIZE, "RVUPF%d_ME0", rvu_get_pf(pf->pcifunc)); + ret = request_irq(pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFME0), + otx2_pf_me_intr_handler, 0, irq_name, pf); + if (ret) { + dev_err(pf->dev, + "RVUPF: IRQ registration failed for ME0\n"); + } + + /* Register FLR interrupt handler */ + irq_name = &hw->irq_name[RVU_PF_INT_VEC_VFFLR0 * NAME_SIZE]; + snprintf(irq_name, NAME_SIZE, "RVUPF%d_FLR0", rvu_get_pf(pf->pcifunc)); + ret = request_irq(pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFFLR0), + otx2_pf_flr_intr_handler, 0, irq_name, pf); + if (ret) { + dev_err(pf->dev, + "RVUPF: IRQ registration failed for FLR0\n"); + return ret; + } + + if (numvfs > 64) { + irq_name = &hw->irq_name[RVU_PF_INT_VEC_VFME1 * NAME_SIZE]; + snprintf(irq_name, NAME_SIZE, "RVUPF%d_ME1", + rvu_get_pf(pf->pcifunc)); + ret = request_irq(pci_irq_vector + (pf->pdev, RVU_PF_INT_VEC_VFME1), + otx2_pf_me_intr_handler, 0, irq_name, pf); + if (ret) { + dev_err(pf->dev, + "RVUPF: IRQ registration failed for ME1\n"); + } + irq_name = &hw->irq_name[RVU_PF_INT_VEC_VFFLR1 * NAME_SIZE]; + snprintf(irq_name, NAME_SIZE, "RVUPF%d_FLR1", + rvu_get_pf(pf->pcifunc)); + ret = request_irq(pci_irq_vector + (pf->pdev, RVU_PF_INT_VEC_VFFLR1), + otx2_pf_flr_intr_handler, 0, irq_name, pf); + if (ret) { + dev_err(pf->dev, + "RVUPF: IRQ registration failed for FLR1\n"); + return ret; + } + } + + /* Enable ME interrupt for all VFs*/ + otx2_write64(pf, RVU_PF_VFME_INTX(0), INTR_MASK(numvfs)); + otx2_write64(pf, RVU_PF_VFME_INT_ENA_W1SX(0), INTR_MASK(numvfs)); + + /* Enable FLR interrupt for all VFs*/ + otx2_write64(pf, RVU_PF_VFFLR_INTX(0), INTR_MASK(numvfs)); + otx2_write64(pf, RVU_PF_VFFLR_INT_ENA_W1SX(0), INTR_MASK(numvfs)); + + if (numvfs > 64) { + numvfs -= 64; + + otx2_write64(pf, RVU_PF_VFME_INTX(1), INTR_MASK(numvfs)); + otx2_write64(pf, RVU_PF_VFME_INT_ENA_W1SX(1), + INTR_MASK(numvfs)); + + otx2_write64(pf, RVU_PF_VFFLR_INTX(1), INTR_MASK(numvfs)); + otx2_write64(pf, RVU_PF_VFFLR_INT_ENA_W1SX(1), + INTR_MASK(numvfs)); + } + return 0; +} + +static int otx2_pf_flr_init(struct otx2_nic *pf, int num_vfs) +{ + int vf; + + pf->flr_wq = alloc_workqueue("otx2_pf_flr_wq", + WQ_UNBOUND | WQ_HIGHPRI, 1); + if (!pf->flr_wq) + return -ENOMEM; + + pf->flr_wrk = devm_kcalloc(pf->dev, num_vfs, + sizeof(struct flr_work), GFP_KERNEL); + if (!pf->flr_wrk) { + destroy_workqueue(pf->flr_wq); + return -ENOMEM; + } + + for (vf = 0; vf < num_vfs; vf++) { + pf->flr_wrk[vf].pf = pf; + INIT_WORK(&pf->flr_wrk[vf].work, otx2_flr_handler); + } + + return 0; +} + static void otx2_queue_work(struct mbox *mw, struct workqueue_struct *mbox_wq, int first, int mdevs, u64 intr, int type) { @@ -115,9 +331,391 @@ static void otx2_queue_work(struct mbox *mw, struct workqueue_struct *mbox_wq, } } +static void otx2_forward_msg_pfvf(struct otx2_mbox_dev *mdev, + struct otx2_mbox *pfvf_mbox, void *bbuf_base, + int devid) +{ + struct otx2_mbox_dev *src_mdev = mdev; + int offset; + + /* Msgs are already copied, trigger VF's mbox irq */ + smp_wmb(); + + offset = pfvf_mbox->trigger | (devid << pfvf_mbox->tr_shift); + writeq(1, (void __iomem *)pfvf_mbox->reg_base + offset); + + /* Restore VF's mbox bounce buffer region address */ + src_mdev->mbase = bbuf_base; +} + +static int otx2_forward_vf_mbox_msgs(struct otx2_nic *pf, + struct otx2_mbox *src_mbox, + int dir, int vf, int num_msgs) +{ + struct otx2_mbox_dev *src_mdev, *dst_mdev; + struct mbox_hdr *mbox_hdr; + struct mbox_hdr *req_hdr; + struct mbox *dst_mbox; + int dst_size, err; + + if (dir == MBOX_DIR_PFAF) { + /* Set VF's mailbox memory as PF's bounce buffer memory, so + * that explicit copying of VF's msgs to PF=>AF mbox region + * and AF=>PF responses to VF's mbox region can be avoided. + */ + src_mdev = &src_mbox->dev[vf]; + mbox_hdr = src_mbox->hwbase + + src_mbox->rx_start + (vf * MBOX_SIZE); + + dst_mbox = &pf->mbox; + dst_size = dst_mbox->mbox.tx_size - + ALIGN(sizeof(*mbox_hdr), MBOX_MSG_ALIGN); + /* Check if msgs fit into destination area */ + if (mbox_hdr->msg_size > dst_size) + return -EINVAL; + + dst_mdev = &dst_mbox->mbox.dev[0]; + + mutex_lock(&pf->mbox.lock); + dst_mdev->mbase = src_mdev->mbase; + dst_mdev->msg_size = mbox_hdr->msg_size; + dst_mdev->num_msgs = num_msgs; + err = otx2_sync_mbox_msg(dst_mbox); + if (err) { + dev_warn(pf->dev, + "AF not responding to VF%d messages\n", vf); + /* restore PF mbase and exit */ + dst_mdev->mbase = pf->mbox.bbuf_base; + mutex_unlock(&pf->mbox.lock); + return err; + } + /* At this point, all the VF messages sent to AF are acked + * with proper responses and responses are copied to VF + * mailbox hence raise interrupt to VF. + */ + req_hdr = (struct mbox_hdr *)(dst_mdev->mbase + + dst_mbox->mbox.rx_start); + req_hdr->num_msgs = num_msgs; + + otx2_forward_msg_pfvf(dst_mdev, &pf->mbox_pfvf[0].mbox, + pf->mbox.bbuf_base, vf); + mutex_unlock(&pf->mbox.lock); + } else if (dir == MBOX_DIR_PFVF_UP) { + src_mdev = &src_mbox->dev[0]; + mbox_hdr = src_mbox->hwbase + src_mbox->rx_start; + req_hdr = (struct mbox_hdr *)(src_mdev->mbase + + src_mbox->rx_start); + req_hdr->num_msgs = num_msgs; + + dst_mbox = &pf->mbox_pfvf[0]; + dst_size = dst_mbox->mbox_up.tx_size - + ALIGN(sizeof(*mbox_hdr), MBOX_MSG_ALIGN); + /* Check if msgs fit into destination area */ + if (mbox_hdr->msg_size > dst_size) + return -EINVAL; + + dst_mdev = &dst_mbox->mbox_up.dev[vf]; + dst_mdev->mbase = src_mdev->mbase; + dst_mdev->msg_size = mbox_hdr->msg_size; + dst_mdev->num_msgs = mbox_hdr->num_msgs; + err = otx2_sync_mbox_up_msg(dst_mbox, vf); + if (err) { + dev_warn(pf->dev, + "VF%d is not responding to mailbox\n", vf); + return err; + } + } else if (dir == MBOX_DIR_VFPF_UP) { + req_hdr = (struct mbox_hdr *)(src_mbox->dev[0].mbase + + src_mbox->rx_start); + req_hdr->num_msgs = num_msgs; + otx2_forward_msg_pfvf(&pf->mbox_pfvf->mbox_up.dev[vf], + &pf->mbox.mbox_up, + pf->mbox_pfvf[vf].bbuf_base, + 0); + } + + return 0; +} + +static void otx2_pfvf_mbox_handler(struct work_struct *work) +{ + struct mbox_msghdr *msg = NULL; + int offset, vf_idx, id, err; + struct otx2_mbox_dev *mdev; + struct mbox_hdr *req_hdr; + struct otx2_mbox *mbox; + struct mbox *vf_mbox; + struct otx2_nic *pf; + + vf_mbox = container_of(work, struct mbox, mbox_wrk); + pf = vf_mbox->pfvf; + vf_idx = vf_mbox - pf->mbox_pfvf; + + mbox = &pf->mbox_pfvf[0].mbox; + mdev = &mbox->dev[vf_idx]; + req_hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start); + + offset = ALIGN(sizeof(*req_hdr), MBOX_MSG_ALIGN); + + for (id = 0; id < vf_mbox->num_msgs; id++) { + msg = (struct mbox_msghdr *)(mdev->mbase + mbox->rx_start + + offset); + + if (msg->sig != OTX2_MBOX_REQ_SIG) + goto inval_msg; + + /* Set VF's number in each of the msg */ + msg->pcifunc &= RVU_PFVF_FUNC_MASK; + msg->pcifunc |= (vf_idx + 1) & RVU_PFVF_FUNC_MASK; + offset = msg->next_msgoff; + } + err = otx2_forward_vf_mbox_msgs(pf, mbox, MBOX_DIR_PFAF, vf_idx, + vf_mbox->num_msgs); + if (err) + goto inval_msg; + return; + +inval_msg: + otx2_reply_invalid_msg(mbox, vf_idx, 0, msg->id); + otx2_mbox_msg_send(mbox, vf_idx); +} + +static void otx2_pfvf_mbox_up_handler(struct work_struct *work) +{ + struct mbox *vf_mbox = container_of(work, struct mbox, mbox_up_wrk); + struct otx2_nic *pf = vf_mbox->pfvf; + struct otx2_mbox_dev *mdev; + int offset, id, vf_idx = 0; + struct mbox_hdr *rsp_hdr; + struct mbox_msghdr *msg; + struct otx2_mbox *mbox; + + vf_idx = vf_mbox - pf->mbox_pfvf; + mbox = &pf->mbox_pfvf[0].mbox_up; + mdev = &mbox->dev[vf_idx]; + + rsp_hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start); + offset = mbox->rx_start + ALIGN(sizeof(*rsp_hdr), MBOX_MSG_ALIGN); + + for (id = 0; id < vf_mbox->up_num_msgs; id++) { + msg = mdev->mbase + offset; + + if (msg->id >= MBOX_MSG_MAX) { + dev_err(pf->dev, + "Mbox msg with unknown ID 0x%x\n", msg->id); + goto end; + } + + if (msg->sig != OTX2_MBOX_RSP_SIG) { + dev_err(pf->dev, + "Mbox msg with wrong signature %x, ID 0x%x\n", + msg->sig, msg->id); + goto end; + } + + switch (msg->id) { + case MBOX_MSG_CGX_LINK_EVENT: + break; + default: + if (msg->rc) + dev_err(pf->dev, + "Mbox msg response has err %d, ID 0x%x\n", + msg->rc, msg->id); + break; + } + +end: + offset = mbox->rx_start + msg->next_msgoff; + mdev->msgs_acked++; + } + + otx2_mbox_reset(mbox, vf_idx); +} + +static irqreturn_t otx2_pfvf_mbox_intr_handler(int irq, void *pf_irq) +{ + struct otx2_nic *pf = (struct otx2_nic *)(pf_irq); + int vfs = pf->total_vfs; + struct mbox *mbox; + u64 intr; + + mbox = pf->mbox_pfvf; + /* Handle VF interrupts */ + if (vfs > 64) { + intr = otx2_read64(pf, RVU_PF_VFPF_MBOX_INTX(1)); + otx2_write64(pf, RVU_PF_VFPF_MBOX_INTX(1), intr); + otx2_queue_work(mbox, pf->mbox_pfvf_wq, 64, vfs, intr, + TYPE_PFVF); + vfs -= 64; + } + + intr = otx2_read64(pf, RVU_PF_VFPF_MBOX_INTX(0)); + otx2_write64(pf, RVU_PF_VFPF_MBOX_INTX(0), intr); + + otx2_queue_work(mbox, pf->mbox_pfvf_wq, 0, vfs, intr, TYPE_PFVF); + + return IRQ_HANDLED; +} + +static int otx2_pfvf_mbox_init(struct otx2_nic *pf, int numvfs) +{ + void __iomem *hwbase; + struct mbox *mbox; + int err, vf; + u64 base; + + if (!numvfs) + return -EINVAL; + + pf->mbox_pfvf = devm_kcalloc(&pf->pdev->dev, numvfs, + sizeof(struct mbox), GFP_KERNEL); + if (!pf->mbox_pfvf) + return -ENOMEM; + + pf->mbox_pfvf_wq = alloc_workqueue("otx2_pfvf_mailbox", + WQ_UNBOUND | WQ_HIGHPRI | + WQ_MEM_RECLAIM, 1); + if (!pf->mbox_pfvf_wq) + return -ENOMEM; + + base = readq((void __iomem *)((u64)pf->reg_base + RVU_PF_VF_BAR4_ADDR)); + hwbase = ioremap_wc(base, MBOX_SIZE * pf->total_vfs); + + if (!hwbase) { + err = -ENOMEM; + goto free_wq; + } + + mbox = &pf->mbox_pfvf[0]; + err = otx2_mbox_init(&mbox->mbox, hwbase, pf->pdev, pf->reg_base, + MBOX_DIR_PFVF, numvfs); + if (err) + goto free_iomem; + + err = otx2_mbox_init(&mbox->mbox_up, hwbase, pf->pdev, pf->reg_base, + MBOX_DIR_PFVF_UP, numvfs); + if (err) + goto free_iomem; + + for (vf = 0; vf < numvfs; vf++) { + mbox->pfvf = pf; + INIT_WORK(&mbox->mbox_wrk, otx2_pfvf_mbox_handler); + INIT_WORK(&mbox->mbox_up_wrk, otx2_pfvf_mbox_up_handler); + mbox++; + } + + return 0; + +free_iomem: + if (hwbase) + iounmap(hwbase); +free_wq: + destroy_workqueue(pf->mbox_pfvf_wq); + return err; +} + +static void otx2_pfvf_mbox_destroy(struct otx2_nic *pf) +{ + struct mbox *mbox = &pf->mbox_pfvf[0]; + + if (!mbox) + return; + + if (pf->mbox_pfvf_wq) { + destroy_workqueue(pf->mbox_pfvf_wq); + pf->mbox_pfvf_wq = NULL; + } + + if (mbox->mbox.hwbase) + iounmap(mbox->mbox.hwbase); + + otx2_mbox_destroy(&mbox->mbox); +} + +static void otx2_enable_pfvf_mbox_intr(struct otx2_nic *pf, int numvfs) +{ + /* Clear PF <=> VF mailbox IRQ */ + otx2_write64(pf, RVU_PF_VFPF_MBOX_INTX(0), ~0ull); + otx2_write64(pf, RVU_PF_VFPF_MBOX_INTX(1), ~0ull); + + /* Enable PF <=> VF mailbox IRQ */ + otx2_write64(pf, RVU_PF_VFPF_MBOX_INT_ENA_W1SX(0), INTR_MASK(numvfs)); + if (numvfs > 64) { + numvfs -= 64; + otx2_write64(pf, RVU_PF_VFPF_MBOX_INT_ENA_W1SX(1), + INTR_MASK(numvfs)); + } +} + +static void otx2_disable_pfvf_mbox_intr(struct otx2_nic *pf, int numvfs) +{ + int vector; + + /* Disable PF <=> VF mailbox IRQ */ + otx2_write64(pf, RVU_PF_VFPF_MBOX_INT_ENA_W1CX(0), ~0ull); + otx2_write64(pf, RVU_PF_VFPF_MBOX_INT_ENA_W1CX(1), ~0ull); + + otx2_write64(pf, RVU_PF_VFPF_MBOX_INTX(0), ~0ull); + vector = pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFPF_MBOX0); + free_irq(vector, pf); + + if (numvfs > 64) { + otx2_write64(pf, RVU_PF_VFPF_MBOX_INTX(1), ~0ull); + vector = pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFPF_MBOX1); + free_irq(vector, pf); + } +} + +static int otx2_register_pfvf_mbox_intr(struct otx2_nic *pf, int numvfs) +{ + struct otx2_hw *hw = &pf->hw; + char *irq_name; + int err; + + /* Register MBOX0 interrupt handler */ + irq_name = &hw->irq_name[RVU_PF_INT_VEC_VFPF_MBOX0 * NAME_SIZE]; + if (pf->pcifunc) + snprintf(irq_name, NAME_SIZE, + "RVUPF%d_VF Mbox0", rvu_get_pf(pf->pcifunc)); + else + snprintf(irq_name, NAME_SIZE, "RVUPF_VF Mbox0"); + err = request_irq(pci_irq_vector(pf->pdev, RVU_PF_INT_VEC_VFPF_MBOX0), + otx2_pfvf_mbox_intr_handler, 0, irq_name, pf); + if (err) { + dev_err(pf->dev, + "RVUPF: IRQ registration failed for PFVF mbox0 irq\n"); + return err; + } + + if (numvfs > 64) { + /* Register MBOX1 interrupt handler */ + irq_name = &hw->irq_name[RVU_PF_INT_VEC_VFPF_MBOX1 * NAME_SIZE]; + if (pf->pcifunc) + snprintf(irq_name, NAME_SIZE, + "RVUPF%d_VF Mbox1", rvu_get_pf(pf->pcifunc)); + else + snprintf(irq_name, NAME_SIZE, "RVUPF_VF Mbox1"); + err = request_irq(pci_irq_vector(pf->pdev, + RVU_PF_INT_VEC_VFPF_MBOX1), + otx2_pfvf_mbox_intr_handler, + 0, irq_name, pf); + if (err) { + dev_err(pf->dev, + "RVUPF: IRQ registration failed for PFVF mbox1 irq\n"); + return err; + } + } + + otx2_enable_pfvf_mbox_intr(pf, numvfs); + + return 0; +} + static void otx2_process_pfaf_mbox_msg(struct otx2_nic *pf, struct mbox_msghdr *msg) { + int devid; + if (msg->id >= MBOX_MSG_MAX) { dev_err(pf->dev, "Mbox msg with unknown ID 0x%x\n", msg->id); @@ -131,6 +729,26 @@ static void otx2_process_pfaf_mbox_msg(struct otx2_nic *pf, return; } + /* message response heading VF */ + devid = msg->pcifunc & RVU_PFVF_FUNC_MASK; + if (devid) { + struct otx2_vf_config *config = &pf->vf_configs[devid - 1]; + struct delayed_work *dwork; + + switch (msg->id) { + case MBOX_MSG_NIX_LF_START_RX: + config->intf_down = false; + dwork = &config->link_event_work; + schedule_delayed_work(dwork, msecs_to_jiffies(100)); + break; + case MBOX_MSG_NIX_LF_STOP_RX: + config->intf_down = true; + break; + } + + return; + } + switch (msg->id) { case MBOX_MSG_READY: pf->pcifunc = msg->pcifunc; @@ -148,6 +766,9 @@ static void otx2_process_pfaf_mbox_msg(struct otx2_nic *pf, mbox_handler_nix_txsch_alloc(pf, (struct nix_txsch_alloc_rsp *)msg); break; + case MBOX_MSG_NIX_BP_ENABLE: + mbox_handler_nix_bp_enable(pf, (struct nix_bp_cfg_rsp *)msg); + break; case MBOX_MSG_CGX_STATS: mbox_handler_cgx_stats(pf, (struct cgx_stats_rsp *)msg); break; @@ -209,9 +830,22 @@ int otx2_mbox_up_handler_cgx_link_event(struct otx2_nic *pf, struct cgx_link_info_msg *msg, struct msg_rsp *rsp) { + int i; + /* Copy the link info sent by AF */ pf->linfo = msg->link_info; + /* notify VFs about link event */ + for (i = 0; i < pci_num_vf(pf->pdev); i++) { + struct otx2_vf_config *config = &pf->vf_configs[i]; + struct delayed_work *dwork = &config->link_event_work; + + if (config->intf_down) + continue; + + schedule_delayed_work(dwork, msecs_to_jiffies(100)); + } + /* interface has not been fully configured yet */ if (pf->flags & OTX2_FLAG_INTF_DOWN) return 0; @@ -283,6 +917,12 @@ static void otx2_pfaf_mbox_up_handler(struct work_struct *work) otx2_process_mbox_msg_up(pf, msg); offset = mbox->rx_start + msg->next_msgoff; } + if (devid) { + otx2_forward_vf_mbox_msgs(pf, &pf->mbox.mbox_up, + MBOX_DIR_PFVF_UP, devid - 1, + af_mbox->up_num_msgs); + return; + } otx2_mbox_msg_send(mbox, 0); } @@ -359,7 +999,6 @@ static void otx2_pfaf_mbox_destroy(struct otx2_nic *pf) struct mbox *mbox = &pf->mbox; if (pf->mbox_wq) { - flush_workqueue(pf->mbox_wq); destroy_workqueue(pf->mbox_wq); pf->mbox_wq = NULL; } @@ -412,7 +1051,7 @@ static int otx2_pfaf_mbox_init(struct otx2_nic *pf) INIT_WORK(&mbox->mbox_wrk, otx2_pfaf_mbox_handler); INIT_WORK(&mbox->mbox_up_wrk, otx2_pfaf_mbox_up_handler); - otx2_mbox_lock_init(&pf->mbox); + mutex_init(&mbox->lock); return 0; exit: @@ -425,19 +1064,19 @@ static int otx2_cgx_config_linkevents(struct otx2_nic *pf, bool enable) struct msg_req *msg; int err; - otx2_mbox_lock(&pf->mbox); + mutex_lock(&pf->mbox.lock); if (enable) msg = otx2_mbox_alloc_msg_cgx_start_linkevents(&pf->mbox); else msg = otx2_mbox_alloc_msg_cgx_stop_linkevents(&pf->mbox); if (!msg) { - otx2_mbox_unlock(&pf->mbox); + mutex_unlock(&pf->mbox.lock); return -ENOMEM; } err = otx2_sync_mbox_msg(&pf->mbox); - otx2_mbox_unlock(&pf->mbox); + mutex_unlock(&pf->mbox.lock); return err; } @@ -446,19 +1085,19 @@ static int otx2_cgx_config_loopback(struct otx2_nic *pf, bool enable) struct msg_req *msg; int err; - otx2_mbox_lock(&pf->mbox); + mutex_lock(&pf->mbox.lock); if (enable) msg = otx2_mbox_alloc_msg_cgx_intlbk_enable(&pf->mbox); else msg = otx2_mbox_alloc_msg_cgx_intlbk_disable(&pf->mbox); if (!msg) { - otx2_mbox_unlock(&pf->mbox); + mutex_unlock(&pf->mbox.lock); return -ENOMEM; } err = otx2_sync_mbox_msg(&pf->mbox); - otx2_mbox_unlock(&pf->mbox); + mutex_unlock(&pf->mbox.lock); return err; } @@ -480,6 +1119,7 @@ int otx2_set_real_num_queues(struct net_device *netdev, "Failed to set no of Rx queues: %d\n", rx_queues); return err; } +EXPORT_SYMBOL(otx2_set_real_num_queues); static irqreturn_t otx2_q_intr_handler(int irq, void *data) { @@ -643,7 +1283,7 @@ static int otx2_init_hw_resources(struct otx2_nic *pf) /* Get the size of receive buffers to allocate */ pf->rbsize = RCV_FRAG_LEN(pf->netdev->mtu + OTX2_ETH_HLEN); - otx2_mbox_lock(mbox); + mutex_lock(&mbox->lock); /* NPA init */ err = otx2_config_npa(pf); if (err) @@ -654,38 +1294,41 @@ static int otx2_init_hw_resources(struct otx2_nic *pf) if (err) goto err_free_npa_lf; + /* Enable backpressure */ + otx2_nix_config_bp(pf, true); + /* Init Auras and pools used by NIX RQ, for free buffer ptrs */ err = otx2_rq_aura_pool_init(pf); if (err) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); goto err_free_nix_lf; } /* Init Auras and pools used by NIX SQ, for queueing SQEs */ err = otx2_sq_aura_pool_init(pf); if (err) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); goto err_free_rq_ptrs; } err = otx2_txsch_alloc(pf); if (err) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); goto err_free_sq_ptrs; } err = otx2_config_nix_queues(pf); if (err) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); goto err_free_txsch; } for (lvl = 0; lvl < NIX_TXSCH_LVL_CNT; lvl++) { err = otx2_txschq_config(pf, lvl); if (err) { - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); goto err_free_nix_queues; } } - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); return err; err_free_nix_queues: @@ -703,7 +1346,7 @@ err_free_rq_ptrs: otx2_ctx_disable(mbox, NPA_AQ_CTYPE_AURA, true); otx2_aura_pool_free(pf); err_free_nix_lf: - otx2_mbox_lock(mbox); + mutex_lock(&mbox->lock); req = otx2_mbox_alloc_msg_nix_lf_free(mbox); if (req) { if (otx2_sync_mbox_msg(mbox)) @@ -717,7 +1360,7 @@ err_free_npa_lf: dev_err(pf->dev, "%s failed to free npalf\n", __func__); } exit: - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); return err; } @@ -737,6 +1380,12 @@ static void otx2_free_hw_resources(struct otx2_nic *pf) if (err) dev_err(pf->dev, "RVUPF: Failed to stop/free TX schedulers\n"); + mutex_lock(&mbox->lock); + /* Disable backpressure */ + if (!(pf->pcifunc & RVU_PFVF_FUNC_MASK)) + otx2_nix_config_bp(pf, false); + mutex_unlock(&mbox->lock); + /* Disable RQs */ otx2_ctx_disable(mbox, NIX_AQ_CTYPE_RQ, false); @@ -756,28 +1405,28 @@ static void otx2_free_hw_resources(struct otx2_nic *pf) otx2_free_cq_res(pf); - otx2_mbox_lock(mbox); + mutex_lock(&mbox->lock); /* Reset NIX LF */ req = otx2_mbox_alloc_msg_nix_lf_free(mbox); if (req) { if (otx2_sync_mbox_msg(mbox)) dev_err(pf->dev, "%s failed to free nixlf\n", __func__); } - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); /* Disable NPA Pool and Aura hw context */ otx2_ctx_disable(mbox, NPA_AQ_CTYPE_POOL, true); otx2_ctx_disable(mbox, NPA_AQ_CTYPE_AURA, true); otx2_aura_pool_free(pf); - otx2_mbox_lock(mbox); + mutex_lock(&mbox->lock); /* Reset NPA LF */ req = otx2_mbox_alloc_msg_npa_lf_free(mbox); if (req) { if (otx2_sync_mbox_msg(mbox)) dev_err(pf->dev, "%s failed to free npalf\n", __func__); } - otx2_mbox_unlock(mbox); + mutex_unlock(&mbox->lock); } int otx2_open(struct net_device *netdev) @@ -906,6 +1555,9 @@ int otx2_open(struct net_device *netdev) if (pf->linfo.link_up && !(pf->pcifunc & RVU_PFVF_FUNC_MASK)) otx2_handle_link_event(pf); + /* Restore pause frame settings */ + otx2_config_pause_frm(pf); + err = otx2_rxtx_enable(pf, true); if (err) goto err_free_cints; @@ -929,6 +1581,7 @@ err_free_mem: kfree(qset->napi); return err; } +EXPORT_SYMBOL(otx2_open); int otx2_stop(struct net_device *netdev) { @@ -989,6 +1642,7 @@ int otx2_stop(struct net_device *netdev) sizeof(*qset) - offsetof(struct otx2_qset, sqe_cnt)); return 0; } +EXPORT_SYMBOL(otx2_stop); static netdev_tx_t otx2_xmit(struct sk_buff *skb, struct net_device *netdev) { @@ -1025,15 +1679,23 @@ static netdev_tx_t otx2_xmit(struct sk_buff *skb, struct net_device *netdev) static void otx2_set_rx_mode(struct net_device *netdev) { struct otx2_nic *pf = netdev_priv(netdev); + + queue_work(pf->otx2_wq, &pf->rx_mode_work); +} + +static void otx2_do_set_rx_mode(struct work_struct *work) +{ + struct otx2_nic *pf = container_of(work, struct otx2_nic, rx_mode_work); + struct net_device *netdev = pf->netdev; struct nix_rx_mode *req; if (!(netdev->flags & IFF_UP)) return; - otx2_mbox_lock(&pf->mbox); + mutex_lock(&pf->mbox.lock); req = otx2_mbox_alloc_msg_nix_set_rx_mode(&pf->mbox); if (!req) { - otx2_mbox_unlock(&pf->mbox); + mutex_unlock(&pf->mbox.lock); return; } @@ -1046,7 +1708,7 @@ static void otx2_set_rx_mode(struct net_device *netdev) req->mode |= NIX_RX_MODE_ALLMULTI; otx2_sync_mbox_msg(&pf->mbox); - otx2_mbox_unlock(&pf->mbox); + mutex_unlock(&pf->mbox.lock); } static int otx2_set_features(struct net_device *netdev, @@ -1086,6 +1748,17 @@ static const struct net_device_ops otx2_netdev_ops = { .ndo_get_stats64 = otx2_get_stats64, }; +static int otx2_wq_init(struct otx2_nic *pf) +{ + pf->otx2_wq = create_singlethread_workqueue("otx2_wq"); + if (!pf->otx2_wq) + return -ENOMEM; + + INIT_WORK(&pf->rx_mode_work, otx2_do_set_rx_mode); + INIT_WORK(&pf->reset_task, otx2_reset_task); + return 0; +} + static int otx2_check_pf_usable(struct otx2_nic *nic) { u64 rev; @@ -1117,7 +1790,6 @@ static int otx2_realloc_msix_vectors(struct otx2_nic *pf) otx2_disable_mbox_intr(pf); pci_free_irq_vectors(hw->pdev); - pci_free_irq_vectors(hw->pdev); err = pci_alloc_irq_vectors(hw->pdev, num_vec, num_vec, PCI_IRQ_MSIX); if (err < 0) { dev_err(pf->dev, "%s: Failed to realloc %d IRQ vectors\n", @@ -1172,6 +1844,7 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id) pf->netdev = netdev; pf->pdev = pdev; pf->dev = dev; + pf->total_vfs = pci_sriov_get_totalvfs(pdev); pf->flags |= OTX2_FLAG_INTF_DOWN; hw = &pf->hw; @@ -1270,21 +1943,29 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id) netdev->min_mtu = OTX2_MIN_MTU; netdev->max_mtu = OTX2_MAX_MTU; - INIT_WORK(&pf->reset_task, otx2_reset_task); - err = register_netdev(netdev); if (err) { dev_err(dev, "Failed to register netdevice\n"); goto err_detach_rsrc; } + err = otx2_wq_init(pf); + if (err) + goto err_unreg_netdev; + otx2_set_ethtool_ops(netdev); /* Enable link notifications */ otx2_cgx_config_linkevents(pf, true); + /* Enable pause frames by default */ + pf->flags |= OTX2_FLAG_RX_PAUSE_ENABLED; + pf->flags |= OTX2_FLAG_TX_PAUSE_ENABLED; + return 0; +err_unreg_netdev: + unregister_netdev(netdev); err_detach_rsrc: otx2_detach_resources(&pf->mbox); err_disable_mbox_intr: @@ -1301,6 +1982,121 @@ err_release_regions: return err; } +static void otx2_vf_link_event_task(struct work_struct *work) +{ + struct otx2_vf_config *config; + struct cgx_link_info_msg *req; + struct mbox_msghdr *msghdr; + struct otx2_nic *pf; + int vf_idx; + + config = container_of(work, struct otx2_vf_config, + link_event_work.work); + vf_idx = config - config->pf->vf_configs; + pf = config->pf; + + msghdr = otx2_mbox_alloc_msg_rsp(&pf->mbox_pfvf[0].mbox_up, vf_idx, + sizeof(*req), sizeof(struct msg_rsp)); + if (!msghdr) { + dev_err(pf->dev, "Failed to create VF%d link event\n", vf_idx); + return; + } + + req = (struct cgx_link_info_msg *)msghdr; + req->hdr.id = MBOX_MSG_CGX_LINK_EVENT; + req->hdr.sig = OTX2_MBOX_REQ_SIG; + memcpy(&req->link_info, &pf->linfo, sizeof(req->link_info)); + + otx2_sync_mbox_up_msg(&pf->mbox_pfvf[0], vf_idx); +} + +static int otx2_sriov_enable(struct pci_dev *pdev, int numvfs) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct otx2_nic *pf = netdev_priv(netdev); + int ret, i; + + /* Init PF <=> VF mailbox stuff */ + ret = otx2_pfvf_mbox_init(pf, numvfs); + if (ret) + return ret; + + ret = otx2_register_pfvf_mbox_intr(pf, numvfs); + if (ret) + goto free_mbox; + + pf->vf_configs = kcalloc(numvfs, sizeof(struct otx2_vf_config), + GFP_KERNEL); + if (!pf->vf_configs) { + ret = -ENOMEM; + goto free_intr; + } + + for (i = 0; i < numvfs; i++) { + pf->vf_configs[i].pf = pf; + pf->vf_configs[i].intf_down = true; + INIT_DELAYED_WORK(&pf->vf_configs[i].link_event_work, + otx2_vf_link_event_task); + } + + ret = otx2_pf_flr_init(pf, numvfs); + if (ret) + goto free_configs; + + ret = otx2_register_flr_me_intr(pf, numvfs); + if (ret) + goto free_flr; + + ret = pci_enable_sriov(pdev, numvfs); + if (ret) + goto free_flr_intr; + + return numvfs; +free_flr_intr: + otx2_disable_flr_me_intr(pf); +free_flr: + otx2_flr_wq_destroy(pf); +free_configs: + kfree(pf->vf_configs); +free_intr: + otx2_disable_pfvf_mbox_intr(pf, numvfs); +free_mbox: + otx2_pfvf_mbox_destroy(pf); + return ret; +} + +static int otx2_sriov_disable(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct otx2_nic *pf = netdev_priv(netdev); + int numvfs = pci_num_vf(pdev); + int i; + + if (!numvfs) + return 0; + + pci_disable_sriov(pdev); + + for (i = 0; i < pci_num_vf(pdev); i++) + cancel_delayed_work_sync(&pf->vf_configs[i].link_event_work); + kfree(pf->vf_configs); + + otx2_disable_flr_me_intr(pf); + otx2_flr_wq_destroy(pf); + otx2_disable_pfvf_mbox_intr(pf, numvfs); + otx2_pfvf_mbox_destroy(pf); + + return 0; +} + +static int otx2_sriov_configure(struct pci_dev *pdev, int numvfs) +{ + if (numvfs == 0) + return otx2_sriov_disable(pdev); + else + return otx2_sriov_enable(pdev, numvfs); +} + static void otx2_remove(struct pci_dev *pdev) { struct net_device *netdev = pci_get_drvdata(pdev); @@ -1315,6 +2111,10 @@ static void otx2_remove(struct pci_dev *pdev) otx2_cgx_config_linkevents(pf, false); unregister_netdev(netdev); + otx2_sriov_disable(pf->pdev); + if (pf->otx2_wq) + destroy_workqueue(pf->otx2_wq); + otx2_detach_resources(&pf->mbox); otx2_disable_mbox_intr(pf); otx2_pfaf_mbox_destroy(pf); @@ -1331,6 +2131,7 @@ static struct pci_driver otx2_pf_driver = { .probe = otx2_probe, .shutdown = otx2_remove, .remove = otx2_remove, + .sriov_configure = otx2_sriov_configure }; static int __init otx2_rvupf_init_module(void) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h index 7963d418886a..867f646e0802 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_reg.h @@ -45,6 +45,19 @@ #define RVU_PF_MSIX_VECX_CTL(a) (0x008 | (a) << 4) #define RVU_PF_MSIX_PBAX(a) (0xF0000 | (a) << 3) +/* RVU VF registers */ +#define RVU_VF_VFPF_MBOX0 (0x00000) +#define RVU_VF_VFPF_MBOX1 (0x00008) +#define RVU_VF_VFPF_MBOXX(a) (0x00 | (a) << 3) +#define RVU_VF_INT (0x20) +#define RVU_VF_INT_W1S (0x28) +#define RVU_VF_INT_ENA_W1S (0x30) +#define RVU_VF_INT_ENA_W1C (0x38) +#define RVU_VF_BLOCK_ADDRX_DISC(a) (0x200 | (a) << 3) +#define RVU_VF_MSIX_VECX_ADDR(a) (0x000 | (a) << 4) +#define RVU_VF_MSIX_VECX_CTL(a) (0x008 | (a) << 4) +#define RVU_VF_MSIX_PBAX(a) (0xF0000 | (a) << 3) + #define RVU_FUNC_BLKADDR_SHIFT 20 #define RVU_FUNC_BLKADDR_MASK 0x1FULL diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c index bef4c20fe314..45abe0cd0e7b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c @@ -138,6 +138,25 @@ static void otx2_set_rxhash(struct otx2_nic *pfvf, skb_set_hash(skb, hash, hash_type); } +static void otx2_free_rcv_seg(struct otx2_nic *pfvf, struct nix_cqe_rx_s *cqe, + int qidx) +{ + struct nix_rx_sg_s *sg = &cqe->sg; + void *end, *start; + u64 *seg_addr; + int seg; + + start = (void *)sg; + end = start + ((cqe->parse.desc_sizem1 + 1) * 16); + while (start < end) { + sg = (struct nix_rx_sg_s *)start; + seg_addr = &sg->seg_addr; + for (seg = 0; seg < sg->segs; seg++, seg_addr++) + otx2_aura_freeptr(pfvf, qidx, *seg_addr & ~0x07ULL); + start += sizeof(*sg); + } +} + static bool otx2_check_rcv_errors(struct otx2_nic *pfvf, struct nix_cqe_rx_s *cqe, int qidx) { @@ -189,16 +208,17 @@ static bool otx2_check_rcv_errors(struct otx2_nic *pfvf, /* For now ignore all the NPC parser errors and * pass the packets to stack. */ - return false; + if (cqe->sg.segs == 1) + return false; } /* If RXALL is enabled pass on packets to stack. */ - if (cqe->sg.segs && (pfvf->netdev->features & NETIF_F_RXALL)) + if (cqe->sg.segs == 1 && (pfvf->netdev->features & NETIF_F_RXALL)) return false; /* Free buffer back to pool */ if (cqe->sg.segs) - otx2_aura_freeptr(pfvf, qidx, cqe->sg.seg_addr & ~0x07ULL); + otx2_free_rcv_seg(pfvf, cqe, qidx); return true; } @@ -210,7 +230,7 @@ static void otx2_rcv_pkt_handler(struct otx2_nic *pfvf, struct nix_rx_parse_s *parse = &cqe->parse; struct sk_buff *skb = NULL; - if (unlikely(parse->errlev || parse->errcode)) { + if (unlikely(parse->errlev || parse->errcode || cqe->sg.segs > 1)) { if (otx2_check_rcv_errors(pfvf, cqe, cq->cq_idx)) return; } @@ -284,6 +304,7 @@ static int otx2_rx_napi_handler(struct otx2_nic *pfvf, otx2_aura_freeptr(pfvf, cq->cq_idx, bufptr + OTX2_HEAD_ROOM); cq->pool_ptrs--; } + otx2_get_page(cq->rbpool); return processed_cqe; } @@ -778,6 +799,7 @@ bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq, return true; } +EXPORT_SYMBOL(otx2_sq_append_skb); void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq) { @@ -788,11 +810,15 @@ void otx2_cleanup_rx_cqes(struct otx2_nic *pfvf, struct otx2_cq_queue *cq) while ((cqe = (struct nix_cqe_rx_s *)otx2_get_next_cqe(cq))) { if (!cqe->sg.subdc) continue; + processed_cqe++; + if (cqe->sg.segs > 1) { + otx2_free_rcv_seg(pfvf, cqe, cq->cq_idx); + continue; + } iova = cqe->sg.seg_addr - OTX2_HEAD_ROOM; pa = otx2_iova_to_phys(pfvf->iommu_domain, iova); otx2_dma_unmap_page(pfvf, iova, pfvf->rbsize, DMA_FROM_DEVICE); put_page(virt_to_page(phys_to_virt(pa))); - processed_cqe++; } /* Free CQEs to HW */ @@ -831,18 +857,18 @@ int otx2_rxtx_enable(struct otx2_nic *pfvf, bool enable) struct msg_req *msg; int err; - otx2_mbox_lock(&pfvf->mbox); + mutex_lock(&pfvf->mbox.lock); if (enable) msg = otx2_mbox_alloc_msg_nix_lf_start_rx(&pfvf->mbox); else msg = otx2_mbox_alloc_msg_nix_lf_stop_rx(&pfvf->mbox); if (!msg) { - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return -ENOMEM; } err = otx2_sync_mbox_msg(&pfvf->mbox); - otx2_mbox_unlock(&pfvf->mbox); + mutex_unlock(&pfvf->mbox.lock); return err; } diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c new file mode 100644 index 000000000000..187c633a7af5 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c @@ -0,0 +1,648 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell OcteonTx2 RVU Virtual Function ethernet driver */ + +#include <linux/etherdevice.h> +#include <linux/module.h> +#include <linux/pci.h> + +#include "otx2_common.h" +#include "otx2_reg.h" + +#define DRV_NAME "octeontx2-nicvf" +#define DRV_STRING "Marvell OcteonTX2 NIC Virtual Function Driver" + +static const struct pci_device_id otx2_vf_id_table[] = { + { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_RVU_AFVF) }, + { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_RVU_VF) }, + { } +}; + +MODULE_AUTHOR("Sunil Goutham <sgoutham@marvell.com>"); +MODULE_DESCRIPTION(DRV_STRING); +MODULE_LICENSE("GPL v2"); +MODULE_DEVICE_TABLE(pci, otx2_vf_id_table); + +/* RVU VF Interrupt Vector Enumeration */ +enum { + RVU_VF_INT_VEC_MBOX = 0x0, +}; + +static void otx2vf_process_vfaf_mbox_msg(struct otx2_nic *vf, + struct mbox_msghdr *msg) +{ + if (msg->id >= MBOX_MSG_MAX) { + dev_err(vf->dev, + "Mbox msg with unknown ID %d\n", msg->id); + return; + } + + if (msg->sig != OTX2_MBOX_RSP_SIG) { + dev_err(vf->dev, + "Mbox msg with wrong signature %x, ID %d\n", + msg->sig, msg->id); + return; + } + + if (msg->rc == MBOX_MSG_INVALID) { + dev_err(vf->dev, + "PF/AF says the sent msg(s) %d were invalid\n", + msg->id); + return; + } + + switch (msg->id) { + case MBOX_MSG_READY: + vf->pcifunc = msg->pcifunc; + break; + case MBOX_MSG_MSIX_OFFSET: + mbox_handler_msix_offset(vf, (struct msix_offset_rsp *)msg); + break; + case MBOX_MSG_NPA_LF_ALLOC: + mbox_handler_npa_lf_alloc(vf, (struct npa_lf_alloc_rsp *)msg); + break; + case MBOX_MSG_NIX_LF_ALLOC: + mbox_handler_nix_lf_alloc(vf, (struct nix_lf_alloc_rsp *)msg); + break; + case MBOX_MSG_NIX_TXSCH_ALLOC: + mbox_handler_nix_txsch_alloc(vf, + (struct nix_txsch_alloc_rsp *)msg); + break; + case MBOX_MSG_NIX_BP_ENABLE: + mbox_handler_nix_bp_enable(vf, (struct nix_bp_cfg_rsp *)msg); + break; + default: + if (msg->rc) + dev_err(vf->dev, + "Mbox msg response has err %d, ID %d\n", + msg->rc, msg->id); + } +} + +static void otx2vf_vfaf_mbox_handler(struct work_struct *work) +{ + struct otx2_mbox_dev *mdev; + struct mbox_hdr *rsp_hdr; + struct mbox_msghdr *msg; + struct otx2_mbox *mbox; + struct mbox *af_mbox; + int offset, id; + + af_mbox = container_of(work, struct mbox, mbox_wrk); + mbox = &af_mbox->mbox; + mdev = &mbox->dev[0]; + rsp_hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start); + if (af_mbox->num_msgs == 0) + return; + offset = mbox->rx_start + ALIGN(sizeof(*rsp_hdr), MBOX_MSG_ALIGN); + + for (id = 0; id < af_mbox->num_msgs; id++) { + msg = (struct mbox_msghdr *)(mdev->mbase + offset); + otx2vf_process_vfaf_mbox_msg(af_mbox->pfvf, msg); + offset = mbox->rx_start + msg->next_msgoff; + mdev->msgs_acked++; + } + + otx2_mbox_reset(mbox, 0); +} + +static int otx2vf_process_mbox_msg_up(struct otx2_nic *vf, + struct mbox_msghdr *req) +{ + struct msg_rsp *rsp; + int err; + + /* Check if valid, if not reply with a invalid msg */ + if (req->sig != OTX2_MBOX_REQ_SIG) { + otx2_reply_invalid_msg(&vf->mbox.mbox_up, 0, 0, req->id); + return -ENODEV; + } + + switch (req->id) { + case MBOX_MSG_CGX_LINK_EVENT: + rsp = (struct msg_rsp *)otx2_mbox_alloc_msg( + &vf->mbox.mbox_up, 0, + sizeof(struct msg_rsp)); + if (!rsp) + return -ENOMEM; + + rsp->hdr.id = MBOX_MSG_CGX_LINK_EVENT; + rsp->hdr.sig = OTX2_MBOX_RSP_SIG; + rsp->hdr.pcifunc = 0; + rsp->hdr.rc = 0; + err = otx2_mbox_up_handler_cgx_link_event( + vf, (struct cgx_link_info_msg *)req, rsp); + return err; + default: + otx2_reply_invalid_msg(&vf->mbox.mbox_up, 0, 0, req->id); + return -ENODEV; + } + return 0; +} + +static void otx2vf_vfaf_mbox_up_handler(struct work_struct *work) +{ + struct otx2_mbox_dev *mdev; + struct mbox_hdr *rsp_hdr; + struct mbox_msghdr *msg; + struct otx2_mbox *mbox; + struct mbox *vf_mbox; + struct otx2_nic *vf; + int offset, id; + + vf_mbox = container_of(work, struct mbox, mbox_up_wrk); + vf = vf_mbox->pfvf; + mbox = &vf_mbox->mbox_up; + mdev = &mbox->dev[0]; + + rsp_hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start); + if (vf_mbox->up_num_msgs == 0) + return; + + offset = mbox->rx_start + ALIGN(sizeof(*rsp_hdr), MBOX_MSG_ALIGN); + + for (id = 0; id < vf_mbox->up_num_msgs; id++) { + msg = (struct mbox_msghdr *)(mdev->mbase + offset); + otx2vf_process_mbox_msg_up(vf, msg); + offset = mbox->rx_start + msg->next_msgoff; + } + + otx2_mbox_msg_send(mbox, 0); +} + +static irqreturn_t otx2vf_vfaf_mbox_intr_handler(int irq, void *vf_irq) +{ + struct otx2_nic *vf = (struct otx2_nic *)vf_irq; + struct otx2_mbox_dev *mdev; + struct otx2_mbox *mbox; + struct mbox_hdr *hdr; + + /* Clear the IRQ */ + otx2_write64(vf, RVU_VF_INT, BIT_ULL(0)); + + /* Read latest mbox data */ + smp_rmb(); + + /* Check for PF => VF response messages */ + mbox = &vf->mbox.mbox; + mdev = &mbox->dev[0]; + otx2_sync_mbox_bbuf(mbox, 0); + + hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start); + if (hdr->num_msgs) { + vf->mbox.num_msgs = hdr->num_msgs; + hdr->num_msgs = 0; + memset(mbox->hwbase + mbox->rx_start, 0, + ALIGN(sizeof(struct mbox_hdr), sizeof(u64))); + queue_work(vf->mbox_wq, &vf->mbox.mbox_wrk); + } + /* Check for PF => VF notification messages */ + mbox = &vf->mbox.mbox_up; + mdev = &mbox->dev[0]; + otx2_sync_mbox_bbuf(mbox, 0); + + hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start); + if (hdr->num_msgs) { + vf->mbox.up_num_msgs = hdr->num_msgs; + hdr->num_msgs = 0; + memset(mbox->hwbase + mbox->rx_start, 0, + ALIGN(sizeof(struct mbox_hdr), sizeof(u64))); + queue_work(vf->mbox_wq, &vf->mbox.mbox_up_wrk); + } + + return IRQ_HANDLED; +} + +static void otx2vf_disable_mbox_intr(struct otx2_nic *vf) +{ + int vector = pci_irq_vector(vf->pdev, RVU_VF_INT_VEC_MBOX); + + /* Disable VF => PF mailbox IRQ */ + otx2_write64(vf, RVU_VF_INT_ENA_W1C, BIT_ULL(0)); + free_irq(vector, vf); +} + +static int otx2vf_register_mbox_intr(struct otx2_nic *vf, bool probe_pf) +{ + struct otx2_hw *hw = &vf->hw; + struct msg_req *req; + char *irq_name; + int err; + + /* Register mailbox interrupt handler */ + irq_name = &hw->irq_name[RVU_VF_INT_VEC_MBOX * NAME_SIZE]; + snprintf(irq_name, NAME_SIZE, "RVUVFAF Mbox"); + err = request_irq(pci_irq_vector(vf->pdev, RVU_VF_INT_VEC_MBOX), + otx2vf_vfaf_mbox_intr_handler, 0, irq_name, vf); + if (err) { + dev_err(vf->dev, + "RVUPF: IRQ registration failed for VFAF mbox irq\n"); + return err; + } + + /* Enable mailbox interrupt for msgs coming from PF. + * First clear to avoid spurious interrupts, if any. + */ + otx2_write64(vf, RVU_VF_INT, BIT_ULL(0)); + otx2_write64(vf, RVU_VF_INT_ENA_W1S, BIT_ULL(0)); + + if (!probe_pf) + return 0; + + /* Check mailbox communication with PF */ + req = otx2_mbox_alloc_msg_ready(&vf->mbox); + if (!req) { + otx2vf_disable_mbox_intr(vf); + return -ENOMEM; + } + + err = otx2_sync_mbox_msg(&vf->mbox); + if (err) { + dev_warn(vf->dev, + "AF not responding to mailbox, deferring probe\n"); + otx2vf_disable_mbox_intr(vf); + return -EPROBE_DEFER; + } + return 0; +} + +static void otx2vf_vfaf_mbox_destroy(struct otx2_nic *vf) +{ + struct mbox *mbox = &vf->mbox; + + if (vf->mbox_wq) { + flush_workqueue(vf->mbox_wq); + destroy_workqueue(vf->mbox_wq); + vf->mbox_wq = NULL; + } + + if (mbox->mbox.hwbase) + iounmap((void __iomem *)mbox->mbox.hwbase); + + otx2_mbox_destroy(&mbox->mbox); + otx2_mbox_destroy(&mbox->mbox_up); +} + +static int otx2vf_vfaf_mbox_init(struct otx2_nic *vf) +{ + struct mbox *mbox = &vf->mbox; + void __iomem *hwbase; + int err; + + mbox->pfvf = vf; + vf->mbox_wq = alloc_workqueue("otx2_vfaf_mailbox", + WQ_UNBOUND | WQ_HIGHPRI | + WQ_MEM_RECLAIM, 1); + if (!vf->mbox_wq) + return -ENOMEM; + + /* Mailbox is a reserved memory (in RAM) region shared between + * admin function (i.e PF0) and this VF, shouldn't be mapped as + * device memory to allow unaligned accesses. + */ + hwbase = ioremap_wc(pci_resource_start(vf->pdev, PCI_MBOX_BAR_NUM), + pci_resource_len(vf->pdev, PCI_MBOX_BAR_NUM)); + if (!hwbase) { + dev_err(vf->dev, "Unable to map VFAF mailbox region\n"); + err = -ENOMEM; + goto exit; + } + + err = otx2_mbox_init(&mbox->mbox, hwbase, vf->pdev, vf->reg_base, + MBOX_DIR_VFPF, 1); + if (err) + goto exit; + + err = otx2_mbox_init(&mbox->mbox_up, hwbase, vf->pdev, vf->reg_base, + MBOX_DIR_VFPF_UP, 1); + if (err) + goto exit; + + err = otx2_mbox_bbuf_init(mbox, vf->pdev); + if (err) + goto exit; + + INIT_WORK(&mbox->mbox_wrk, otx2vf_vfaf_mbox_handler); + INIT_WORK(&mbox->mbox_up_wrk, otx2vf_vfaf_mbox_up_handler); + mutex_init(&mbox->lock); + + return 0; +exit: + destroy_workqueue(vf->mbox_wq); + return err; +} + +static int otx2vf_open(struct net_device *netdev) +{ + struct otx2_nic *vf; + int err; + + err = otx2_open(netdev); + if (err) + return err; + + /* LBKs do not receive link events so tell everyone we are up here */ + vf = netdev_priv(netdev); + if (is_otx2_lbkvf(vf->pdev)) { + pr_info("%s NIC Link is UP\n", netdev->name); + netif_carrier_on(netdev); + netif_tx_start_all_queues(netdev); + } + + return 0; +} + +static int otx2vf_stop(struct net_device *netdev) +{ + return otx2_stop(netdev); +} + +static netdev_tx_t otx2vf_xmit(struct sk_buff *skb, struct net_device *netdev) +{ + struct otx2_nic *vf = netdev_priv(netdev); + int qidx = skb_get_queue_mapping(skb); + struct otx2_snd_queue *sq; + struct netdev_queue *txq; + + sq = &vf->qset.sq[qidx]; + txq = netdev_get_tx_queue(netdev, qidx); + + if (!otx2_sq_append_skb(netdev, sq, skb, qidx)) { + netif_tx_stop_queue(txq); + + /* Check again, incase SQBs got freed up */ + smp_mb(); + if (((sq->num_sqbs - *sq->aura_fc_addr) * sq->sqe_per_sqb) + > sq->sqe_thresh) + netif_tx_wake_queue(txq); + + return NETDEV_TX_BUSY; + } + + return NETDEV_TX_OK; +} + +static int otx2vf_change_mtu(struct net_device *netdev, int new_mtu) +{ + bool if_up = netif_running(netdev); + int err = 0; + + if (if_up) + otx2vf_stop(netdev); + + netdev_info(netdev, "Changing MTU from %d to %d\n", + netdev->mtu, new_mtu); + netdev->mtu = new_mtu; + + if (if_up) + err = otx2vf_open(netdev); + + return err; +} + +static void otx2vf_reset_task(struct work_struct *work) +{ + struct otx2_nic *vf = container_of(work, struct otx2_nic, reset_task); + + rtnl_lock(); + + if (netif_running(vf->netdev)) { + otx2vf_stop(vf->netdev); + vf->reset_count++; + otx2vf_open(vf->netdev); + } + + rtnl_unlock(); +} + +static const struct net_device_ops otx2vf_netdev_ops = { + .ndo_open = otx2vf_open, + .ndo_stop = otx2vf_stop, + .ndo_start_xmit = otx2vf_xmit, + .ndo_set_mac_address = otx2_set_mac_address, + .ndo_change_mtu = otx2vf_change_mtu, + .ndo_get_stats64 = otx2_get_stats64, + .ndo_tx_timeout = otx2_tx_timeout, +}; + +static int otx2vf_realloc_msix_vectors(struct otx2_nic *vf) +{ + struct otx2_hw *hw = &vf->hw; + int num_vec, err; + + num_vec = hw->nix_msixoff; + num_vec += NIX_LF_CINT_VEC_START + hw->max_queues; + + otx2vf_disable_mbox_intr(vf); + pci_free_irq_vectors(hw->pdev); + err = pci_alloc_irq_vectors(hw->pdev, num_vec, num_vec, PCI_IRQ_MSIX); + if (err < 0) { + dev_err(vf->dev, "%s: Failed to realloc %d IRQ vectors\n", + __func__, num_vec); + return err; + } + + return otx2vf_register_mbox_intr(vf, false); +} + +static int otx2vf_probe(struct pci_dev *pdev, const struct pci_device_id *id) +{ + int num_vec = pci_msix_vec_count(pdev); + struct device *dev = &pdev->dev; + struct net_device *netdev; + struct otx2_nic *vf; + struct otx2_hw *hw; + int err, qcount; + + err = pcim_enable_device(pdev); + if (err) { + dev_err(dev, "Failed to enable PCI device\n"); + return err; + } + + err = pci_request_regions(pdev, DRV_NAME); + if (err) { + dev_err(dev, "PCI request regions failed 0x%x\n", err); + return err; + } + + err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(48)); + if (err) { + dev_err(dev, "DMA mask config failed, abort\n"); + goto err_release_regions; + } + + pci_set_master(pdev); + + qcount = num_online_cpus(); + netdev = alloc_etherdev_mqs(sizeof(*vf), qcount, qcount); + if (!netdev) { + err = -ENOMEM; + goto err_release_regions; + } + + pci_set_drvdata(pdev, netdev); + SET_NETDEV_DEV(netdev, &pdev->dev); + vf = netdev_priv(netdev); + vf->netdev = netdev; + vf->pdev = pdev; + vf->dev = dev; + vf->iommu_domain = iommu_get_domain_for_dev(dev); + + vf->flags |= OTX2_FLAG_INTF_DOWN; + hw = &vf->hw; + hw->pdev = vf->pdev; + hw->rx_queues = qcount; + hw->tx_queues = qcount; + hw->max_queues = qcount; + + hw->irq_name = devm_kmalloc_array(&hw->pdev->dev, num_vec, NAME_SIZE, + GFP_KERNEL); + if (!hw->irq_name) + goto err_free_netdev; + + hw->affinity_mask = devm_kcalloc(&hw->pdev->dev, num_vec, + sizeof(cpumask_var_t), GFP_KERNEL); + if (!hw->affinity_mask) + goto err_free_netdev; + + err = pci_alloc_irq_vectors(hw->pdev, num_vec, num_vec, PCI_IRQ_MSIX); + if (err < 0) { + dev_err(dev, "%s: Failed to alloc %d IRQ vectors\n", + __func__, num_vec); + goto err_free_netdev; + } + + vf->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0); + if (!vf->reg_base) { + dev_err(dev, "Unable to map physical function CSRs, aborting\n"); + err = -ENOMEM; + goto err_free_irq_vectors; + } + + /* Init VF <=> PF mailbox stuff */ + err = otx2vf_vfaf_mbox_init(vf); + if (err) + goto err_free_irq_vectors; + + /* Register mailbox interrupt */ + err = otx2vf_register_mbox_intr(vf, true); + if (err) + goto err_mbox_destroy; + + /* Request AF to attach NPA and LIX LFs to this AF */ + err = otx2_attach_npa_nix(vf); + if (err) + goto err_disable_mbox_intr; + + err = otx2vf_realloc_msix_vectors(vf); + if (err) + goto err_mbox_destroy; + + err = otx2_set_real_num_queues(netdev, qcount, qcount); + if (err) + goto err_detach_rsrc; + + otx2_setup_dev_hw_settings(vf); + + /* Assign default mac address */ + otx2_get_mac_from_af(netdev); + + netdev->hw_features = NETIF_F_RXCSUM | NETIF_F_IP_CSUM | + NETIF_F_IPV6_CSUM | NETIF_F_RXHASH | + NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6; + netdev->features = netdev->hw_features; + + netdev->gso_max_segs = OTX2_MAX_GSO_SEGS; + netdev->watchdog_timeo = OTX2_TX_TIMEOUT; + + netdev->netdev_ops = &otx2vf_netdev_ops; + + /* MTU range: 68 - 9190 */ + netdev->min_mtu = OTX2_MIN_MTU; + netdev->max_mtu = OTX2_MAX_MTU; + + INIT_WORK(&vf->reset_task, otx2vf_reset_task); + + /* To distinguish, for LBK VFs set netdev name explicitly */ + if (is_otx2_lbkvf(vf->pdev)) { + int n; + + n = (vf->pcifunc >> RVU_PFVF_FUNC_SHIFT) & RVU_PFVF_FUNC_MASK; + /* Need to subtract 1 to get proper VF number */ + n -= 1; + snprintf(netdev->name, sizeof(netdev->name), "lbk%d", n); + } + + err = register_netdev(netdev); + if (err) { + dev_err(dev, "Failed to register netdevice\n"); + goto err_detach_rsrc; + } + + otx2vf_set_ethtool_ops(netdev); + + /* Enable pause frames by default */ + vf->flags |= OTX2_FLAG_RX_PAUSE_ENABLED; + vf->flags |= OTX2_FLAG_TX_PAUSE_ENABLED; + + return 0; + +err_detach_rsrc: + otx2_detach_resources(&vf->mbox); +err_disable_mbox_intr: + otx2vf_disable_mbox_intr(vf); +err_mbox_destroy: + otx2vf_vfaf_mbox_destroy(vf); +err_free_irq_vectors: + pci_free_irq_vectors(hw->pdev); +err_free_netdev: + pci_set_drvdata(pdev, NULL); + free_netdev(netdev); +err_release_regions: + pci_release_regions(pdev); + return err; +} + +static void otx2vf_remove(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct otx2_nic *vf; + + if (!netdev) + return; + + vf = netdev_priv(netdev); + + otx2vf_disable_mbox_intr(vf); + + otx2_detach_resources(&vf->mbox); + otx2vf_vfaf_mbox_destroy(vf); + pci_free_irq_vectors(vf->pdev); + pci_set_drvdata(pdev, NULL); + free_netdev(netdev); + + pci_release_regions(pdev); +} + +static struct pci_driver otx2vf_driver = { + .name = DRV_NAME, + .id_table = otx2_vf_id_table, + .probe = otx2vf_probe, + .remove = otx2vf_remove, + .shutdown = otx2vf_remove, +}; + +static int __init otx2vf_init_module(void) +{ + pr_info("%s: %s\n", DRV_NAME, DRV_STRING); + + return pci_register_driver(&otx2vf_driver); +} + +static void __exit otx2vf_cleanup_module(void) +{ + pci_unregister_driver(&otx2vf_driver); +} + +module_init(otx2vf_init_module); +module_exit(otx2vf_cleanup_module); diff --git a/drivers/net/ethernet/marvell/skge.c b/drivers/net/ethernet/marvell/skge.c index 97f270d30cce..3c89206f18a7 100644 --- a/drivers/net/ethernet/marvell/skge.c +++ b/drivers/net/ethernet/marvell/skge.c @@ -876,6 +876,7 @@ static int skge_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom } static const struct ethtool_ops skge_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_drvinfo = skge_get_drvinfo, .get_regs_len = skge_get_regs_len, .get_regs = skge_get_regs, diff --git a/drivers/net/ethernet/marvell/skge.h b/drivers/net/ethernet/marvell/skge.h index 6fa7b6a34c08..6928abcec0a3 100644 --- a/drivers/net/ethernet/marvell/skge.h +++ b/drivers/net/ethernet/marvell/skge.h @@ -15,12 +15,6 @@ #define PCI_VPD_ROM_SZ 7L<<14 /* VPD ROM size 0=256, 1=512, ... */ #define PCI_REV_DESC 1<<2 /* Reverse Descriptor bytes */ -#define PCI_STATUS_ERROR_BITS (PCI_STATUS_DETECTED_PARITY | \ - PCI_STATUS_SIG_SYSTEM_ERROR | \ - PCI_STATUS_REC_MASTER_ABORT | \ - PCI_STATUS_REC_TARGET_ABORT | \ - PCI_STATUS_PARITY) - enum csr_regs { B0_RAP = 0x0000, B0_CTST = 0x0004, @@ -2426,7 +2420,7 @@ struct skge_hw { spinlock_t phy_lock; struct tasklet_struct phy_task; - char irq_name[0]; /* skge@pci:000:04:00.0 */ + char irq_name[]; /* skge@pci:000:04:00.0 */ }; enum pause_control { diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c index ebfd0ceac884..241f00716979 100644 --- a/drivers/net/ethernet/marvell/sky2.c +++ b/drivers/net/ethernet/marvell/sky2.c @@ -4400,6 +4400,10 @@ static int sky2_set_features(struct net_device *dev, netdev_features_t features) } static const struct ethtool_ops sky2_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_RX_USECS_IRQ | + ETHTOOL_COALESCE_RX_MAX_FRAMES_IRQ, .get_drvinfo = sky2_get_drvinfo, .get_wol = sky2_get_wol, .set_wol = sky2_set_wol, diff --git a/drivers/net/ethernet/marvell/sky2.h b/drivers/net/ethernet/marvell/sky2.h index b02b6523083c..b2dddd8a246c 100644 --- a/drivers/net/ethernet/marvell/sky2.h +++ b/drivers/net/ethernet/marvell/sky2.h @@ -252,12 +252,6 @@ enum { }; -#define PCI_STATUS_ERROR_BITS (PCI_STATUS_DETECTED_PARITY | \ - PCI_STATUS_SIG_SYSTEM_ERROR | \ - PCI_STATUS_REC_MASTER_ABORT | \ - PCI_STATUS_REC_TARGET_ABORT | \ - PCI_STATUS_PARITY) - enum csr_regs { B0_RAP = 0x0000, B0_CTST = 0x0004, @@ -2309,7 +2303,7 @@ struct sky2_hw { struct work_struct restart_work; wait_queue_head_t msi_wait; - char irq_name[0]; + char irq_name[]; }; static inline int sky2_is_copper(const struct sky2_hw *hw) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 8c6cfd15481c..8d28f90acfe7 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -412,9 +412,10 @@ static void mtk_mac_link_down(struct phylink_config *config, unsigned int mode, mtk_w32(mac->hw, mcr, MTK_MAC_MCR(mac->id)); } -static void mtk_mac_link_up(struct phylink_config *config, unsigned int mode, - phy_interface_t interface, - struct phy_device *phy) +static void mtk_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, bool tx_pause, bool rx_pause) { struct mtk_mac *mac = container_of(config, struct mtk_mac, phylink_config); diff --git a/drivers/net/ethernet/mellanox/mlx4/crdump.c b/drivers/net/ethernet/mellanox/mlx4/crdump.c index 64ed725aec28..73eae80e1cb7 100644 --- a/drivers/net/ethernet/mellanox/mlx4/crdump.c +++ b/drivers/net/ethernet/mellanox/mlx4/crdump.c @@ -38,8 +38,21 @@ #define CR_ENABLE_BIT_OFFSET 0xF3F04 #define MAX_NUM_OF_DUMPS_TO_STORE (8) -static const char *region_cr_space_str = "cr-space"; -static const char *region_fw_health_str = "fw-health"; +#define REGION_CR_SPACE "cr-space" +#define REGION_FW_HEALTH "fw-health" + +static const char * const region_cr_space_str = REGION_CR_SPACE; +static const char * const region_fw_health_str = REGION_FW_HEALTH; + +static const struct devlink_region_ops region_cr_space_ops = { + .name = REGION_CR_SPACE, + .destructor = &kvfree, +}; + +static const struct devlink_region_ops region_fw_health_ops = { + .name = REGION_FW_HEALTH, + .destructor = &kvfree, +}; /* Set to true in case cr enable bit was set to true before crdump */ static bool crdump_enbale_bit_set; @@ -99,7 +112,7 @@ static void mlx4_crdump_collect_crspace(struct mlx4_dev *dev, readl(cr_space + offset); err = devlink_region_snapshot_create(crdump->region_crspace, - crspace_data, id, &kvfree); + crspace_data, id); if (err) { kvfree(crspace_data); mlx4_warn(dev, "crdump: devlink create %s snapshot id %d err %d\n", @@ -138,7 +151,7 @@ static void mlx4_crdump_collect_fw_health(struct mlx4_dev *dev, readl(health_buf_start + offset); err = devlink_region_snapshot_create(crdump->region_fw_health, - health_data, id, &kvfree); + health_data, id); if (err) { kvfree(health_data); mlx4_warn(dev, "crdump: devlink create %s snapshot id %d err %d\n", @@ -159,6 +172,7 @@ int mlx4_crdump_collect(struct mlx4_dev *dev) struct pci_dev *pdev = dev->persist->pdev; unsigned long cr_res_size; u8 __iomem *cr_space; + int err; u32 id; if (!dev->caps.health_buffer_addrs) { @@ -179,15 +193,22 @@ int mlx4_crdump_collect(struct mlx4_dev *dev) return -ENODEV; } - crdump_enable_crspace_access(dev, cr_space); - /* Get the available snapshot ID for the dumps */ - id = devlink_region_snapshot_id_get(devlink); + err = devlink_region_snapshot_id_get(devlink, &id); + if (err) { + mlx4_err(dev, "crdump: devlink get snapshot id err %d\n", err); + return err; + } + + crdump_enable_crspace_access(dev, cr_space); /* Try to capture dumps */ mlx4_crdump_collect_crspace(dev, cr_space, id); mlx4_crdump_collect_fw_health(dev, cr_space, id); + /* Release reference on the snapshot id */ + devlink_region_snapshot_id_put(devlink, id); + crdump_disable_crspace_access(dev, cr_space); iounmap(cr_space); @@ -205,7 +226,7 @@ int mlx4_crdump_init(struct mlx4_dev *dev) /* Create cr-space region */ crdump->region_crspace = devlink_region_create(devlink, - region_cr_space_str, + ®ion_cr_space_ops, MAX_NUM_OF_DUMPS_TO_STORE, pci_resource_len(pdev, 0)); if (IS_ERR(crdump->region_crspace)) @@ -216,7 +237,7 @@ int mlx4_crdump_init(struct mlx4_dev *dev) /* Create fw-health region */ crdump->region_fw_health = devlink_region_create(devlink, - region_fw_health_str, + ®ion_fw_health_ops, MAX_NUM_OF_DUMPS_TO_STORE, HEALTH_BUFFER_SIZE); if (IS_ERR(crdump->region_fw_health)) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index 8bf1f08fdee2..8a5ea2543670 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -2121,6 +2121,10 @@ static int mlx4_en_set_phys_id(struct net_device *dev, } const struct ethtool_ops mlx4_en_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_TX_MAX_FRAMES_IRQ | + ETHTOOL_COALESCE_PKT_RATE_RX_USECS, .get_drvinfo = mlx4_en_get_drvinfo, .get_link_ksettings = mlx4_en_get_link_ksettings, .set_link_ksettings = mlx4_en_set_link_ksettings, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig index a1f20b205299..312e0a1ad43d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig +++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig @@ -78,6 +78,16 @@ config MLX5_ESWITCH Legacy SRIOV mode (L2 mac vlan steering based). Switchdev mode (eswitch offloads). +config MLX5_TC_CT + bool "MLX5 TC connection tracking offload support" + depends on MLX5_CORE_EN && NET_SWITCHDEV && NF_FLOW_TABLE && NET_ACT_CT && NET_TC_SKB_EXT + default y + help + Say Y here if you want to support offloading connection tracking rules + via tc ct action. + + If unsure, set to Y + config MLX5_CORE_EN_DCB bool "Data Center Bridging (DCB) Support" default y diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index d3e06cec8317..6d32915000fc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -16,7 +16,7 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ transobj.o vport.o sriov.o fs_cmd.o fs_core.o pci_irq.o \ fs_counters.o rl.o lag.o dev.o events.o wq.o lib/gid.o \ lib/devcom.o lib/pci_vsc.o lib/dm.o diag/fs_tracepoint.o \ - diag/fw_tracer.o diag/crdump.o devlink.o + diag/fw_tracer.o diag/crdump.o devlink.o diag/rsc_dump.o # # Netdev basic @@ -25,7 +25,7 @@ mlx5_core-$(CONFIG_MLX5_CORE_EN) += en_main.o en_common.o en_fs.o en_ethtool.o \ en_tx.o en_rx.o en_dim.o en_txrx.o en/xdp.o en_stats.o \ en_selftest.o en/port.o en/monitor_stats.o en/health.o \ en/reporter_tx.o en/reporter_rx.o en/params.o en/xsk/umem.o \ - en/xsk/setup.o en/xsk/rx.o en/xsk/tx.o + en/xsk/setup.o en/xsk/rx.o en/xsk/tx.o en/devlink.o # # Netdev extra @@ -34,15 +34,16 @@ mlx5_core-$(CONFIG_MLX5_EN_ARFS) += en_arfs.o mlx5_core-$(CONFIG_MLX5_EN_RXNFC) += en_fs_ethtool.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o en/port_buffer.o mlx5_core-$(CONFIG_MLX5_ESWITCH) += en_rep.o en_tc.o en/tc_tun.o lib/port_tun.o lag_mp.o \ - lib/geneve.o en/tc_tun_vxlan.o en/tc_tun_gre.o \ + lib/geneve.o en/mapping.o en/tc_tun_vxlan.o en/tc_tun_gre.o \ en/tc_tun_geneve.o diag/en_tc_tracepoint.o mlx5_core-$(CONFIG_PCI_HYPERV_INTERFACE) += en/hv_vhca_stats.o +mlx5_core-$(CONFIG_MLX5_TC_CT) += en/tc_ct.o # # Core extra # mlx5_core-$(CONFIG_MLX5_ESWITCH) += eswitch.o eswitch_offloads.o eswitch_offloads_termtbl.o \ - ecpf.o rdma.o eswitch_offloads_chains.o + ecpf.o rdma.o esw/chains.o mlx5_core-$(CONFIG_MLX5_MPFS) += lib/mpfs.o mlx5_core-$(CONFIG_VXLAN) += lib/vxlan.o mlx5_core-$(CONFIG_PTP_1588_CLOCK) += lib/clock.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/dev.c index 50862275544e..1972ddd12704 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/dev.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/dev.c @@ -193,7 +193,7 @@ bool mlx5_device_registered(struct mlx5_core_dev *dev) return found; } -int mlx5_register_device(struct mlx5_core_dev *dev) +void mlx5_register_device(struct mlx5_core_dev *dev) { struct mlx5_priv *priv = &dev->priv; struct mlx5_interface *intf; @@ -203,8 +203,6 @@ int mlx5_register_device(struct mlx5_core_dev *dev) list_for_each_entry(intf, &intf_list, list) mlx5_add_device(intf, priv); mutex_unlock(&mlx5_intf_mutex); - - return 0; } void mlx5_unregister_device(struct mlx5_core_dev *dev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c index ac108f1e5bd6..bdeb291f6b67 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c @@ -90,7 +90,8 @@ static int mlx5_devlink_reload_down(struct devlink *devlink, bool netns_change, { struct mlx5_core_dev *dev = devlink_priv(devlink); - return mlx5_unload_one(dev, false); + mlx5_unload_one(dev, false); + return 0; } static int mlx5_devlink_reload_up(struct devlink *devlink, @@ -190,11 +191,6 @@ static int mlx5_devlink_fs_mode_get(struct devlink *devlink, u32 id, return 0; } -enum mlx5_devlink_param_id { - MLX5_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX, - MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE, -}; - static int mlx5_devlink_enable_roce_validate(struct devlink *devlink, u32 id, union devlink_param_value val, struct netlink_ext_ack *extack) @@ -210,14 +206,38 @@ static int mlx5_devlink_enable_roce_validate(struct devlink *devlink, u32 id, return 0; } +#ifdef CONFIG_MLX5_ESWITCH +static int mlx5_devlink_large_group_num_validate(struct devlink *devlink, u32 id, + union devlink_param_value val, + struct netlink_ext_ack *extack) +{ + int group_num = val.vu32; + + if (group_num < 1 || group_num > 1024) { + NL_SET_ERR_MSG_MOD(extack, + "Unsupported group number, supported range is 1-1024"); + return -EOPNOTSUPP; + } + + return 0; +} +#endif + static const struct devlink_param mlx5_devlink_params[] = { - DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE, + DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_FLOW_STEERING_MODE, "flow_steering_mode", DEVLINK_PARAM_TYPE_STRING, BIT(DEVLINK_PARAM_CMODE_RUNTIME), mlx5_devlink_fs_mode_get, mlx5_devlink_fs_mode_set, mlx5_devlink_fs_mode_validate), DEVLINK_PARAM_GENERIC(ENABLE_ROCE, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), NULL, NULL, mlx5_devlink_enable_roce_validate), +#ifdef CONFIG_MLX5_ESWITCH + DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_ESW_LARGE_GROUP_NUM, + "fdb_large_groups", DEVLINK_PARAM_TYPE_U32, + BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), + NULL, NULL, + mlx5_devlink_large_group_num_validate), +#endif }; static void mlx5_devlink_set_params_init_values(struct devlink *devlink) @@ -230,13 +250,20 @@ static void mlx5_devlink_set_params_init_values(struct devlink *devlink) else strcpy(value.vstr, "smfs"); devlink_param_driverinit_value_set(devlink, - MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE, + MLX5_DEVLINK_PARAM_ID_FLOW_STEERING_MODE, value); value.vbool = MLX5_CAP_GEN(dev, roce); devlink_param_driverinit_value_set(devlink, DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE, value); + +#ifdef CONFIG_MLX5_ESWITCH + value.vu32 = ESW_OFFLOADS_DEFAULT_NUM_GROUPS; + devlink_param_driverinit_value_set(devlink, + MLX5_DEVLINK_PARAM_ID_ESW_LARGE_GROUP_NUM, + value); +#endif } int mlx5_devlink_register(struct devlink *devlink, struct device *dev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h index d0ba03774ddf..f0de327a59be 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h @@ -6,6 +6,12 @@ #include <net/devlink.h> +enum mlx5_devlink_param_id { + MLX5_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX, + MLX5_DEVLINK_PARAM_ID_FLOW_STEERING_MODE, + MLX5_DEVLINK_PARAM_ID_ESW_LARGE_GROUP_NUM, +}; + struct devlink *mlx5_devlink_alloc(void); void mlx5_devlink_free(struct devlink *devlink); int mlx5_devlink_register(struct devlink *devlink, struct device *dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c index 94d7b69a95c7..c9c9b479bda5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c @@ -188,7 +188,7 @@ static int mlx5_fw_tracer_create_mkey(struct mlx5_fw_tracer *tracer) MLX5_SET(create_mkey_in, in, translations_octword_actual_size, DIV_ROUND_UP(TRACER_BUFFER_PAGE_NUM, 2)); - mtt = (u64 *)MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt); + mtt = (__be64 *)MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt); for (i = 0 ; i < TRACER_BUFFER_PAGE_NUM ; i++) mtt[i] = cpu_to_be64(tracer->buff.dma + i * PAGE_SIZE); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.c new file mode 100644 index 000000000000..17ab7efe693d --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.c @@ -0,0 +1,286 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2019 Mellanox Technologies. */ + +#include "rsc_dump.h" +#include "lib/mlx5.h" + +#define MLX5_SGMT_TYPE(SGMT) MLX5_SGMT_TYPE_##SGMT +#define MLX5_SGMT_STR_ASSING(SGMT)[MLX5_SGMT_TYPE(SGMT)] = #SGMT +static const char *const mlx5_rsc_sgmt_name[] = { + MLX5_SGMT_STR_ASSING(HW_CQPC), + MLX5_SGMT_STR_ASSING(HW_SQPC), + MLX5_SGMT_STR_ASSING(HW_RQPC), + MLX5_SGMT_STR_ASSING(FULL_SRQC), + MLX5_SGMT_STR_ASSING(FULL_CQC), + MLX5_SGMT_STR_ASSING(FULL_EQC), + MLX5_SGMT_STR_ASSING(FULL_QPC), + MLX5_SGMT_STR_ASSING(SND_BUFF), + MLX5_SGMT_STR_ASSING(RCV_BUFF), + MLX5_SGMT_STR_ASSING(SRQ_BUFF), + MLX5_SGMT_STR_ASSING(CQ_BUFF), + MLX5_SGMT_STR_ASSING(EQ_BUFF), + MLX5_SGMT_STR_ASSING(SX_SLICE), + MLX5_SGMT_STR_ASSING(SX_SLICE_ALL), + MLX5_SGMT_STR_ASSING(RDB), + MLX5_SGMT_STR_ASSING(RX_SLICE_ALL), +}; + +struct mlx5_rsc_dump { + u32 pdn; + struct mlx5_core_mkey mkey; + u16 fw_segment_type[MLX5_SGMT_TYPE_NUM]; +}; + +struct mlx5_rsc_dump_cmd { + u64 mem_size; + u8 cmd[MLX5_ST_SZ_BYTES(resource_dump)]; +}; + +static int mlx5_rsc_dump_sgmt_get_by_name(char *name) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(mlx5_rsc_sgmt_name); i++) + if (!strcmp(name, mlx5_rsc_sgmt_name[i])) + return i; + + return -EINVAL; +} + +static void mlx5_rsc_dump_read_menu_sgmt(struct mlx5_rsc_dump *rsc_dump, struct page *page) +{ + void *data = page_address(page); + enum mlx5_sgmt_type sgmt_idx; + int num_of_items; + char *sgmt_name; + void *member; + void *menu; + int i; + + menu = MLX5_ADDR_OF(menu_resource_dump_response, data, menu); + num_of_items = MLX5_GET(resource_dump_menu_segment, menu, num_of_records); + + for (i = 0; i < num_of_items; i++) { + member = MLX5_ADDR_OF(resource_dump_menu_segment, menu, record[i]); + sgmt_name = MLX5_ADDR_OF(resource_dump_menu_record, member, segment_name); + sgmt_idx = mlx5_rsc_dump_sgmt_get_by_name(sgmt_name); + if (sgmt_idx == -EINVAL) + continue; + rsc_dump->fw_segment_type[sgmt_idx] = MLX5_GET(resource_dump_menu_record, + member, segment_type); + } +} + +static int mlx5_rsc_dump_trigger(struct mlx5_core_dev *dev, struct mlx5_rsc_dump_cmd *cmd, + struct page *page) +{ + struct mlx5_rsc_dump *rsc_dump = dev->rsc_dump; + struct device *ddev = &dev->pdev->dev; + u32 out_seq_num; + u32 in_seq_num; + dma_addr_t dma; + int err; + + dma = dma_map_page(ddev, page, 0, cmd->mem_size, DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(ddev, dma))) + return -ENOMEM; + + in_seq_num = MLX5_GET(resource_dump, cmd->cmd, seq_num); + MLX5_SET(resource_dump, cmd->cmd, mkey, rsc_dump->mkey.key); + MLX5_SET64(resource_dump, cmd->cmd, address, dma); + + err = mlx5_core_access_reg(dev, cmd->cmd, sizeof(cmd->cmd), cmd->cmd, + sizeof(cmd->cmd), MLX5_REG_RESOURCE_DUMP, 0, 1); + if (err) { + mlx5_core_err(dev, "Resource dump: Failed to access err %d\n", err); + goto out; + } + out_seq_num = MLX5_GET(resource_dump, cmd->cmd, seq_num); + if (out_seq_num && (in_seq_num + 1 != out_seq_num)) + err = -EIO; +out: + dma_unmap_page(ddev, dma, cmd->mem_size, DMA_FROM_DEVICE); + return err; +} + +struct mlx5_rsc_dump_cmd *mlx5_rsc_dump_cmd_create(struct mlx5_core_dev *dev, + struct mlx5_rsc_key *key) +{ + struct mlx5_rsc_dump_cmd *cmd; + int sgmt_type; + + if (IS_ERR_OR_NULL(dev->rsc_dump)) + return ERR_PTR(-EOPNOTSUPP); + + sgmt_type = dev->rsc_dump->fw_segment_type[key->rsc]; + if (!sgmt_type && key->rsc != MLX5_SGMT_TYPE_MENU) + return ERR_PTR(-EOPNOTSUPP); + + cmd = kzalloc(sizeof(*cmd), GFP_KERNEL); + if (!cmd) { + mlx5_core_err(dev, "Resource dump: Failed to allocate command\n"); + return ERR_PTR(-ENOMEM); + } + MLX5_SET(resource_dump, cmd->cmd, segment_type, sgmt_type); + MLX5_SET(resource_dump, cmd->cmd, index1, key->index1); + MLX5_SET(resource_dump, cmd->cmd, index2, key->index2); + MLX5_SET(resource_dump, cmd->cmd, num_of_obj1, key->num_of_obj1); + MLX5_SET(resource_dump, cmd->cmd, num_of_obj2, key->num_of_obj2); + MLX5_SET(resource_dump, cmd->cmd, size, key->size); + cmd->mem_size = key->size; + return cmd; +} + +void mlx5_rsc_dump_cmd_destroy(struct mlx5_rsc_dump_cmd *cmd) +{ + kfree(cmd); +} + +int mlx5_rsc_dump_next(struct mlx5_core_dev *dev, struct mlx5_rsc_dump_cmd *cmd, + struct page *page, int *size) +{ + bool more_dump; + int err; + + if (IS_ERR_OR_NULL(dev->rsc_dump)) + return -EOPNOTSUPP; + + err = mlx5_rsc_dump_trigger(dev, cmd, page); + if (err) { + mlx5_core_err(dev, "Resource dump: Failed to trigger dump, %d\n", err); + return err; + } + *size = MLX5_GET(resource_dump, cmd->cmd, size); + more_dump = MLX5_GET(resource_dump, cmd->cmd, more_dump); + + return more_dump; +} + +#define MLX5_RSC_DUMP_MENU_SEGMENT 0xffff +static int mlx5_rsc_dump_menu(struct mlx5_core_dev *dev) +{ + struct mlx5_rsc_dump_cmd *cmd = NULL; + struct mlx5_rsc_key key = {}; + struct page *page; + int size; + int err; + + page = alloc_page(GFP_KERNEL); + if (!page) + return -ENOMEM; + + key.rsc = MLX5_SGMT_TYPE_MENU; + key.size = PAGE_SIZE; + cmd = mlx5_rsc_dump_cmd_create(dev, &key); + if (IS_ERR(cmd)) { + err = PTR_ERR(cmd); + goto free_page; + } + MLX5_SET(resource_dump, cmd->cmd, segment_type, MLX5_RSC_DUMP_MENU_SEGMENT); + + do { + err = mlx5_rsc_dump_next(dev, cmd, page, &size); + if (err < 0) + goto destroy_cmd; + + mlx5_rsc_dump_read_menu_sgmt(dev->rsc_dump, page); + + } while (err > 0); + +destroy_cmd: + mlx5_rsc_dump_cmd_destroy(cmd); +free_page: + __free_page(page); + + return err; +} + +static int mlx5_rsc_dump_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, + struct mlx5_core_mkey *mkey) +{ + int inlen = MLX5_ST_SZ_BYTES(create_mkey_in); + void *mkc; + u32 *in; + int err; + + in = kvzalloc(inlen, GFP_KERNEL); + if (!in) + return -ENOMEM; + + mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry); + MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_PA); + MLX5_SET(mkc, mkc, lw, 1); + MLX5_SET(mkc, mkc, lr, 1); + + MLX5_SET(mkc, mkc, pd, pdn); + MLX5_SET(mkc, mkc, length64, 1); + MLX5_SET(mkc, mkc, qpn, 0xffffff); + + err = mlx5_core_create_mkey(mdev, mkey, in, inlen); + + kvfree(in); + return err; +} + +struct mlx5_rsc_dump *mlx5_rsc_dump_create(struct mlx5_core_dev *dev) +{ + struct mlx5_rsc_dump *rsc_dump; + + if (!MLX5_CAP_DEBUG(dev, resource_dump)) { + mlx5_core_dbg(dev, "Resource dump: capability not present\n"); + return NULL; + } + rsc_dump = kzalloc(sizeof(*rsc_dump), GFP_KERNEL); + if (!rsc_dump) + return ERR_PTR(-ENOMEM); + + return rsc_dump; +} + +void mlx5_rsc_dump_destroy(struct mlx5_core_dev *dev) +{ + if (IS_ERR_OR_NULL(dev->rsc_dump)) + return; + kfree(dev->rsc_dump); +} + +int mlx5_rsc_dump_init(struct mlx5_core_dev *dev) +{ + struct mlx5_rsc_dump *rsc_dump = dev->rsc_dump; + int err; + + if (IS_ERR_OR_NULL(dev->rsc_dump)) + return 0; + + err = mlx5_core_alloc_pd(dev, &rsc_dump->pdn); + if (err) { + mlx5_core_warn(dev, "Resource dump: Failed to allocate PD %d\n", err); + return err; + } + err = mlx5_rsc_dump_create_mkey(dev, rsc_dump->pdn, &rsc_dump->mkey); + if (err) { + mlx5_core_err(dev, "Resource dump: Failed to create mkey, %d\n", err); + goto free_pd; + } + err = mlx5_rsc_dump_menu(dev); + if (err) { + mlx5_core_err(dev, "Resource dump: Failed to read menu, %d\n", err); + goto destroy_mkey; + } + return err; + +destroy_mkey: + mlx5_core_destroy_mkey(dev, &rsc_dump->mkey); +free_pd: + mlx5_core_dealloc_pd(dev, rsc_dump->pdn); + return err; +} + +void mlx5_rsc_dump_cleanup(struct mlx5_core_dev *dev) +{ + if (IS_ERR_OR_NULL(dev->rsc_dump)) + return; + + mlx5_core_destroy_mkey(dev, &dev->rsc_dump->mkey); + mlx5_core_dealloc_pd(dev, dev->rsc_dump->pdn); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.h new file mode 100644 index 000000000000..148270073e71 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.h @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019 Mellanox Technologies. */ + +#ifndef __MLX5_RSC_DUMP_H +#define __MLX5_RSC_DUMP_H + +#include <linux/mlx5/driver.h> +#include "mlx5_core.h" + +enum mlx5_sgmt_type { + MLX5_SGMT_TYPE_HW_CQPC, + MLX5_SGMT_TYPE_HW_SQPC, + MLX5_SGMT_TYPE_HW_RQPC, + MLX5_SGMT_TYPE_FULL_SRQC, + MLX5_SGMT_TYPE_FULL_CQC, + MLX5_SGMT_TYPE_FULL_EQC, + MLX5_SGMT_TYPE_FULL_QPC, + MLX5_SGMT_TYPE_SND_BUFF, + MLX5_SGMT_TYPE_RCV_BUFF, + MLX5_SGMT_TYPE_SRQ_BUFF, + MLX5_SGMT_TYPE_CQ_BUFF, + MLX5_SGMT_TYPE_EQ_BUFF, + MLX5_SGMT_TYPE_SX_SLICE, + MLX5_SGMT_TYPE_SX_SLICE_ALL, + MLX5_SGMT_TYPE_RDB, + MLX5_SGMT_TYPE_RX_SLICE_ALL, + MLX5_SGMT_TYPE_MENU, + MLX5_SGMT_TYPE_TERMINATE, + + MLX5_SGMT_TYPE_NUM, /* Keep last */ +}; + +struct mlx5_rsc_key { + enum mlx5_sgmt_type rsc; + int index1; + int index2; + int num_of_obj1; + int num_of_obj2; + int size; +}; + +#define MLX5_RSC_DUMP_ALL 0xFFFF +struct mlx5_rsc_dump_cmd; +struct mlx5_rsc_dump; + +struct mlx5_rsc_dump *mlx5_rsc_dump_create(struct mlx5_core_dev *dev); +void mlx5_rsc_dump_destroy(struct mlx5_core_dev *dev); + +int mlx5_rsc_dump_init(struct mlx5_core_dev *dev); +void mlx5_rsc_dump_cleanup(struct mlx5_core_dev *dev); + +struct mlx5_rsc_dump_cmd *mlx5_rsc_dump_cmd_create(struct mlx5_core_dev *dev, + struct mlx5_rsc_key *key); +void mlx5_rsc_dump_cmd_destroy(struct mlx5_rsc_dump_cmd *cmd); + +int mlx5_rsc_dump_next(struct mlx5_core_dev *dev, struct mlx5_rsc_dump_cmd *cmd, + struct page *page, int *size); +#endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index c9606b8ab6ef..12a61bf82c14 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -204,7 +204,7 @@ struct mlx5e_tx_wqe { struct mlx5e_rx_wqe_ll { struct mlx5_wqe_srq_next_seg next; - struct mlx5_wqe_data_seg data[0]; + struct mlx5_wqe_data_seg data[]; }; struct mlx5e_rx_wqe_cyc { @@ -738,7 +738,6 @@ struct mlx5e_channel { DECLARE_BITMAP(state, MLX5E_CHANNEL_NUM_STATES); int ix; int cpu; - cpumask_var_t xps_cpumask; }; struct mlx5e_channels { @@ -814,6 +813,15 @@ struct mlx5e_xsk { bool ever_used; }; +/* Temporary storage for variables that are allocated when struct mlx5e_priv is + * initialized, and used where we can't allocate them because that functions + * must not fail. Use with care and make sure the same variable is not used + * simultaneously by multiple users. + */ +struct mlx5e_scratchpad { + cpumask_var_t cpumask; +}; + struct mlx5e_priv { /* priv data path fields - start */ struct mlx5e_txqsq *txq2sq[MLX5E_MAX_NUM_CHANNELS * MLX5E_MAX_NUM_TC]; @@ -873,10 +881,12 @@ struct mlx5e_priv { #endif struct devlink_health_reporter *tx_reporter; struct devlink_health_reporter *rx_reporter; + struct devlink_port dl_port; struct mlx5e_xsk xsk; #if IS_ENABLED(CONFIG_PCI_HYPERV_INTERFACE) struct mlx5e_hv_vhca_stats_agent stats_agent; #endif + struct mlx5e_scratchpad scratchpad; }; struct mlx5e_profile { @@ -1036,14 +1046,22 @@ int mlx5e_open_channels(struct mlx5e_priv *priv, struct mlx5e_channels *chs); void mlx5e_close_channels(struct mlx5e_channels *chs); -/* Function pointer to be used to modify WH settings while +/* Function pointer to be used to modify HW or kernel settings while * switching channels */ -typedef int (*mlx5e_fp_hw_modify)(struct mlx5e_priv *priv); +typedef int (*mlx5e_fp_preactivate)(struct mlx5e_priv *priv, void *context); +#define MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX(fn) \ +int fn##_ctx(struct mlx5e_priv *priv, void *context) \ +{ \ + return fn(priv); \ +} int mlx5e_safe_reopen_channels(struct mlx5e_priv *priv); int mlx5e_safe_switch_channels(struct mlx5e_priv *priv, struct mlx5e_channels *new_chs, - mlx5e_fp_hw_modify hw_modify); + mlx5e_fp_preactivate preactivate, + void *context); +int mlx5e_num_channels_changed(struct mlx5e_priv *priv); +int mlx5e_num_channels_changed_ctx(struct mlx5e_priv *priv, void *context); void mlx5e_activate_priv_channels(struct mlx5e_priv *priv); void mlx5e_deactivate_priv_channels(struct mlx5e_priv *priv); @@ -1124,10 +1142,10 @@ void mlx5e_update_ndo_stats(struct mlx5e_priv *priv); void mlx5e_queue_update_stats(struct mlx5e_priv *priv); int mlx5e_bits_invert(unsigned long a, int size); -typedef int (*change_hw_mtu_cb)(struct mlx5e_priv *priv); int mlx5e_set_dev_port_mtu(struct mlx5e_priv *priv); +int mlx5e_set_dev_port_mtu_ctx(struct mlx5e_priv *priv, void *context); int mlx5e_change_mtu(struct net_device *netdev, int new_mtu, - change_hw_mtu_cb set_mtu_cb); + mlx5e_fp_preactivate preactivate); /* ethtool helpers */ void mlx5e_ethtool_get_drvinfo(struct mlx5e_priv *priv, @@ -1153,6 +1171,12 @@ int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv, struct ethtool_link_ksettings *link_ksettings); int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv, const struct ethtool_link_ksettings *link_ksettings); +int mlx5e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, u8 *hfunc); +int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, const u8 *key, + const u8 hfunc); +int mlx5e_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, + u32 *rule_locs); +int mlx5e_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd); u32 mlx5e_ethtool_get_rxfh_key_size(struct mlx5e_priv *priv); u32 mlx5e_ethtool_get_rxfh_indir_size(struct mlx5e_priv *priv); int mlx5e_ethtool_get_ts_info(struct mlx5e_priv *priv, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c new file mode 100644 index 000000000000..f8b2de4b04be --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c @@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020, Mellanox Technologies inc. All rights reserved. */ + +#include "en/devlink.h" + +int mlx5e_devlink_port_register(struct mlx5e_priv *priv) +{ + struct devlink *devlink = priv_to_devlink(priv->mdev); + + if (mlx5_core_is_pf(priv->mdev)) + devlink_port_attrs_set(&priv->dl_port, + DEVLINK_PORT_FLAVOUR_PHYSICAL, + PCI_FUNC(priv->mdev->pdev->devfn), + false, 0, + NULL, 0); + else + devlink_port_attrs_set(&priv->dl_port, + DEVLINK_PORT_FLAVOUR_VIRTUAL, + 0, false, 0, NULL, 0); + + return devlink_port_register(devlink, &priv->dl_port, 1); +} + +void mlx5e_devlink_port_type_eth_set(struct mlx5e_priv *priv) +{ + devlink_port_type_eth_set(&priv->dl_port, priv->netdev); +} + +void mlx5e_devlink_port_unregister(struct mlx5e_priv *priv) +{ + devlink_port_unregister(&priv->dl_port); +} + +struct devlink_port *mlx5e_get_devlink_port(struct net_device *dev) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + + return &priv->dl_port; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.h b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.h new file mode 100644 index 000000000000..83123a801adc --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020, Mellanox Technologies inc. All rights reserved. */ + +#ifndef __MLX5E_EN_DEVLINK_H +#define __MLX5E_EN_DEVLINK_H + +#include <net/devlink.h> +#include "en.h" + +int mlx5e_devlink_port_register(struct mlx5e_priv *priv); +void mlx5e_devlink_port_unregister(struct mlx5e_priv *priv); +void mlx5e_devlink_port_type_eth_set(struct mlx5e_priv *priv); +struct devlink_port *mlx5e_get_devlink_port(struct net_device *dev); + +#endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/health.c b/drivers/net/ethernet/mellanox/mlx5/core/en/health.c index 20b907dc1e29..3a199a03d929 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/health.c @@ -3,6 +3,7 @@ #include "health.h" #include "lib/eq.h" +#include "lib/mlx5.h" int mlx5e_reporter_named_obj_nest_start(struct devlink_fmsg *fmsg, char *name) { @@ -197,10 +198,114 @@ int mlx5e_health_report(struct mlx5e_priv *priv, struct devlink_health_reporter *reporter, char *err_str, struct mlx5e_err_ctx *err_ctx) { - netdev_err(priv->netdev, err_str); + netdev_err(priv->netdev, "%s\n", err_str); if (!reporter) return err_ctx->recover(err_ctx->ctx); return devlink_health_report(reporter, err_str, err_ctx); } + +#define MLX5_HEALTH_DEVLINK_MAX_SIZE 1024 +static int mlx5e_health_rsc_fmsg_binary(struct devlink_fmsg *fmsg, + const void *value, u32 value_len) + +{ + u32 data_size; + u32 offset; + int err; + + for (offset = 0; offset < value_len; offset += data_size) { + data_size = value_len - offset; + if (data_size > MLX5_HEALTH_DEVLINK_MAX_SIZE) + data_size = MLX5_HEALTH_DEVLINK_MAX_SIZE; + err = devlink_fmsg_binary_put(fmsg, value + offset, data_size); + if (err) + break; + } + return err; +} + +int mlx5e_health_rsc_fmsg_dump(struct mlx5e_priv *priv, struct mlx5_rsc_key *key, + struct devlink_fmsg *fmsg) +{ + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_rsc_dump_cmd *cmd; + struct page *page; + int cmd_err, err; + int end_err; + int size; + + if (IS_ERR_OR_NULL(mdev->rsc_dump)) + return -EOPNOTSUPP; + + page = alloc_page(GFP_KERNEL); + if (!page) + return -ENOMEM; + + err = devlink_fmsg_binary_pair_nest_start(fmsg, "data"); + if (err) + return err; + + cmd = mlx5_rsc_dump_cmd_create(mdev, key); + if (IS_ERR(cmd)) { + err = PTR_ERR(cmd); + goto free_page; + } + + do { + cmd_err = mlx5_rsc_dump_next(mdev, cmd, page, &size); + if (cmd_err < 0) { + err = cmd_err; + goto destroy_cmd; + } + + err = mlx5e_health_rsc_fmsg_binary(fmsg, page_address(page), size); + if (err) + goto destroy_cmd; + + } while (cmd_err > 0); + +destroy_cmd: + mlx5_rsc_dump_cmd_destroy(cmd); + end_err = devlink_fmsg_binary_pair_nest_end(fmsg); + if (end_err) + err = end_err; +free_page: + __free_page(page); + return err; +} + +int mlx5e_health_queue_dump(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg, + int queue_idx, char *lbl) +{ + struct mlx5_rsc_key key = {}; + int err; + + key.rsc = MLX5_SGMT_TYPE_FULL_QPC; + key.index1 = queue_idx; + key.size = PAGE_SIZE; + key.num_of_obj1 = 1; + + err = devlink_fmsg_obj_nest_start(fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, lbl); + if (err) + return err; + + err = devlink_fmsg_u32_pair_put(fmsg, "index", queue_idx); + if (err) + return err; + + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + return devlink_fmsg_obj_nest_end(fmsg); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/health.h b/drivers/net/ethernet/mellanox/mlx5/core/en/health.h index e54f70d9af22..38f97f79ef16 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/health.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/health.h @@ -5,6 +5,7 @@ #define __MLX5E_EN_HEALTH_H #include "en.h" +#include "diag/rsc_dump.h" #define MLX5E_RX_ERR_CQE(cqe) (get_cqe_opcode(cqe) != MLX5_CQE_RESP_SEND) @@ -35,6 +36,7 @@ void mlx5e_reporter_rx_timeout(struct mlx5e_rq *rq); struct mlx5e_err_ctx { int (*recover)(void *ctx); + int (*dump)(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg, void *ctx); void *ctx; }; @@ -47,6 +49,8 @@ int mlx5e_health_report(struct mlx5e_priv *priv, int mlx5e_health_create_reporters(struct mlx5e_priv *priv); void mlx5e_health_destroy_reporters(struct mlx5e_priv *priv); void mlx5e_health_channels_update(struct mlx5e_priv *priv); - - +int mlx5e_health_rsc_fmsg_dump(struct mlx5e_priv *priv, struct mlx5_rsc_key *key, + struct devlink_fmsg *fmsg); +int mlx5e_health_queue_dump(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg, + int queue_idx, char *lbl); #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.c b/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.c new file mode 100644 index 000000000000..ea321e528749 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.c @@ -0,0 +1,218 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2018 Mellanox Technologies */ + +#include <linux/jhash.h> +#include <linux/slab.h> +#include <linux/xarray.h> +#include <linux/hashtable.h> + +#include "mapping.h" + +#define MAPPING_GRACE_PERIOD 2000 + +struct mapping_ctx { + struct xarray xarray; + DECLARE_HASHTABLE(ht, 8); + struct mutex lock; /* Guards hashtable and xarray */ + unsigned long max_id; + size_t data_size; + bool delayed_removal; + struct delayed_work dwork; + struct list_head pending_list; + spinlock_t pending_list_lock; /* Guards pending list */ +}; + +struct mapping_item { + struct rcu_head rcu; + struct list_head list; + unsigned long timeout; + struct hlist_node node; + int cnt; + u32 id; + char data[]; +}; + +int mapping_add(struct mapping_ctx *ctx, void *data, u32 *id) +{ + struct mapping_item *mi; + int err = -ENOMEM; + u32 hash_key; + + mutex_lock(&ctx->lock); + + hash_key = jhash(data, ctx->data_size, 0); + hash_for_each_possible(ctx->ht, mi, node, hash_key) { + if (!memcmp(data, mi->data, ctx->data_size)) + goto attach; + } + + mi = kzalloc(sizeof(*mi) + ctx->data_size, GFP_KERNEL); + if (!mi) + goto err_alloc; + + memcpy(mi->data, data, ctx->data_size); + hash_add(ctx->ht, &mi->node, hash_key); + + err = xa_alloc(&ctx->xarray, &mi->id, mi, XA_LIMIT(1, ctx->max_id), + GFP_KERNEL); + if (err) + goto err_assign; +attach: + ++mi->cnt; + *id = mi->id; + + mutex_unlock(&ctx->lock); + + return 0; + +err_assign: + hash_del(&mi->node); + kfree(mi); +err_alloc: + mutex_unlock(&ctx->lock); + + return err; +} + +static void mapping_remove_and_free(struct mapping_ctx *ctx, + struct mapping_item *mi) +{ + xa_erase(&ctx->xarray, mi->id); + kfree_rcu(mi, rcu); +} + +static void mapping_free_item(struct mapping_ctx *ctx, + struct mapping_item *mi) +{ + if (!ctx->delayed_removal) { + mapping_remove_and_free(ctx, mi); + return; + } + + mi->timeout = jiffies + msecs_to_jiffies(MAPPING_GRACE_PERIOD); + + spin_lock(&ctx->pending_list_lock); + list_add_tail(&mi->list, &ctx->pending_list); + spin_unlock(&ctx->pending_list_lock); + + schedule_delayed_work(&ctx->dwork, MAPPING_GRACE_PERIOD); +} + +int mapping_remove(struct mapping_ctx *ctx, u32 id) +{ + unsigned long index = id; + struct mapping_item *mi; + int err = -ENOENT; + + mutex_lock(&ctx->lock); + mi = xa_load(&ctx->xarray, index); + if (!mi) + goto out; + err = 0; + + if (--mi->cnt > 0) + goto out; + + hash_del(&mi->node); + mapping_free_item(ctx, mi); +out: + mutex_unlock(&ctx->lock); + + return err; +} + +int mapping_find(struct mapping_ctx *ctx, u32 id, void *data) +{ + unsigned long index = id; + struct mapping_item *mi; + int err = -ENOENT; + + rcu_read_lock(); + mi = xa_load(&ctx->xarray, index); + if (!mi) + goto err_find; + + memcpy(data, mi->data, ctx->data_size); + err = 0; + +err_find: + rcu_read_unlock(); + return err; +} + +static void +mapping_remove_and_free_list(struct mapping_ctx *ctx, struct list_head *list) +{ + struct mapping_item *mi; + + list_for_each_entry(mi, list, list) + mapping_remove_and_free(ctx, mi); +} + +static void mapping_work_handler(struct work_struct *work) +{ + unsigned long min_timeout = 0, now = jiffies; + struct mapping_item *mi, *next; + LIST_HEAD(pending_items); + struct mapping_ctx *ctx; + + ctx = container_of(work, struct mapping_ctx, dwork.work); + + spin_lock(&ctx->pending_list_lock); + list_for_each_entry_safe(mi, next, &ctx->pending_list, list) { + if (time_after(now, mi->timeout)) + list_move(&mi->list, &pending_items); + else if (!min_timeout || + time_before(mi->timeout, min_timeout)) + min_timeout = mi->timeout; + } + spin_unlock(&ctx->pending_list_lock); + + mapping_remove_and_free_list(ctx, &pending_items); + + if (min_timeout) + schedule_delayed_work(&ctx->dwork, abs(min_timeout - now)); +} + +static void mapping_flush_work(struct mapping_ctx *ctx) +{ + if (!ctx->delayed_removal) + return; + + cancel_delayed_work_sync(&ctx->dwork); + mapping_remove_and_free_list(ctx, &ctx->pending_list); +} + +struct mapping_ctx * +mapping_create(size_t data_size, u32 max_id, bool delayed_removal) +{ + struct mapping_ctx *ctx; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return ERR_PTR(-ENOMEM); + + ctx->max_id = max_id ? max_id : UINT_MAX; + ctx->data_size = data_size; + + if (delayed_removal) { + INIT_DELAYED_WORK(&ctx->dwork, mapping_work_handler); + INIT_LIST_HEAD(&ctx->pending_list); + spin_lock_init(&ctx->pending_list_lock); + ctx->delayed_removal = true; + } + + mutex_init(&ctx->lock); + xa_init_flags(&ctx->xarray, XA_FLAGS_ALLOC1); + + return ctx; +} + +void mapping_destroy(struct mapping_ctx *ctx) +{ + mapping_flush_work(ctx); + xa_destroy(&ctx->xarray); + mutex_destroy(&ctx->lock); + + kfree(ctx); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.h b/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.h new file mode 100644 index 000000000000..285525cc5470 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.h @@ -0,0 +1,27 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2019 Mellanox Technologies */ + +#ifndef __MLX5_MAPPING_H__ +#define __MLX5_MAPPING_H__ + +struct mapping_ctx; + +int mapping_add(struct mapping_ctx *ctx, void *data, u32 *id); +int mapping_remove(struct mapping_ctx *ctx, u32 id); +int mapping_find(struct mapping_ctx *ctx, u32 id, void *data); + +/* mapping uses an xarray to map data to ids in add(), and for find(). + * For locking, it uses a internal xarray spin lock for add()/remove(), + * find() uses rcu_read_lock(). + * Choosing delayed_removal postpones the removal of a previously mapped + * id by MAPPING_GRACE_PERIOD milliseconds. + * This is to avoid races against hardware, where we mark the packet in + * hardware with a previous id, and quick remove() and add() reusing the same + * previous id. Then find() will get the new mapping instead of the old + * which was used to mark the packet. + */ +struct mapping_ctx *mapping_create(size_t data_size, u32 max_id, + bool delayed_removal); +void mapping_destroy(struct mapping_ctx *ctx); + +#endif /* __MLX5_MAPPING_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c index fce6eccdcf8b..2c4a670c8ffd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c @@ -343,64 +343,76 @@ out: return err; } -static u32 fec_supported_speeds[] = { - 10000, - 40000, - 25000, - 50000, - 56000, - 100000 +enum mlx5e_fec_supported_link_mode { + MLX5E_FEC_SUPPORTED_LINK_MODES_10G_40G, + MLX5E_FEC_SUPPORTED_LINK_MODES_25G, + MLX5E_FEC_SUPPORTED_LINK_MODES_50G, + MLX5E_FEC_SUPPORTED_LINK_MODES_56G, + MLX5E_FEC_SUPPORTED_LINK_MODES_100G, + MLX5E_FEC_SUPPORTED_LINK_MODE_50G_1X, + MLX5E_FEC_SUPPORTED_LINK_MODE_100G_2X, + MLX5E_FEC_SUPPORTED_LINK_MODE_200G_4X, + MLX5E_FEC_SUPPORTED_LINK_MODE_400G_8X, + MLX5E_MAX_FEC_SUPPORTED_LINK_MODE, }; -#define MLX5E_FEC_SUPPORTED_SPEEDS ARRAY_SIZE(fec_supported_speeds) +#define MLX5E_FEC_FIRST_50G_PER_LANE_MODE MLX5E_FEC_SUPPORTED_LINK_MODE_50G_1X + +#define MLX5E_FEC_OVERRIDE_ADMIN_POLICY(buf, policy, write, link) \ + do { \ + u16 *_policy = &(policy); \ + u32 *_buf = buf; \ + \ + if (write) \ + MLX5_SET(pplm_reg, _buf, fec_override_admin_##link, *_policy); \ + else \ + *_policy = MLX5_GET(pplm_reg, _buf, fec_override_admin_##link); \ + } while (0) + +#define MLX5E_FEC_OVERRIDE_ADMIN_50G_POLICY(buf, policy, write, link) \ + do { \ + u16 *__policy = &(policy); \ + bool _write = (write); \ + \ + if (_write && *__policy) \ + *__policy = find_first_bit((u_long *)__policy, \ + sizeof(u16) * BITS_PER_BYTE);\ + MLX5E_FEC_OVERRIDE_ADMIN_POLICY(buf, *__policy, _write, link); \ + if (!_write && *__policy) \ + *__policy = 1 << *__policy; \ + } while (0) /* get/set FEC admin field for a given speed */ -static int mlx5e_fec_admin_field(u32 *pplm, - u8 *fec_policy, - bool write, - u32 speed) +static int mlx5e_fec_admin_field(u32 *pplm, u16 *fec_policy, bool write, + enum mlx5e_fec_supported_link_mode link_mode) { - switch (speed) { - case 10000: - case 40000: - if (!write) - *fec_policy = MLX5_GET(pplm_reg, pplm, - fec_override_admin_10g_40g); - else - MLX5_SET(pplm_reg, pplm, - fec_override_admin_10g_40g, *fec_policy); + switch (link_mode) { + case MLX5E_FEC_SUPPORTED_LINK_MODES_10G_40G: + MLX5E_FEC_OVERRIDE_ADMIN_POLICY(pplm, *fec_policy, write, 10g_40g); break; - case 25000: - if (!write) - *fec_policy = MLX5_GET(pplm_reg, pplm, - fec_override_admin_25g); - else - MLX5_SET(pplm_reg, pplm, - fec_override_admin_25g, *fec_policy); + case MLX5E_FEC_SUPPORTED_LINK_MODES_25G: + MLX5E_FEC_OVERRIDE_ADMIN_POLICY(pplm, *fec_policy, write, 25g); break; - case 50000: - if (!write) - *fec_policy = MLX5_GET(pplm_reg, pplm, - fec_override_admin_50g); - else - MLX5_SET(pplm_reg, pplm, - fec_override_admin_50g, *fec_policy); + case MLX5E_FEC_SUPPORTED_LINK_MODES_50G: + MLX5E_FEC_OVERRIDE_ADMIN_POLICY(pplm, *fec_policy, write, 50g); break; - case 56000: - if (!write) - *fec_policy = MLX5_GET(pplm_reg, pplm, - fec_override_admin_56g); - else - MLX5_SET(pplm_reg, pplm, - fec_override_admin_56g, *fec_policy); + case MLX5E_FEC_SUPPORTED_LINK_MODES_56G: + MLX5E_FEC_OVERRIDE_ADMIN_POLICY(pplm, *fec_policy, write, 56g); break; - case 100000: - if (!write) - *fec_policy = MLX5_GET(pplm_reg, pplm, - fec_override_admin_100g); - else - MLX5_SET(pplm_reg, pplm, - fec_override_admin_100g, *fec_policy); + case MLX5E_FEC_SUPPORTED_LINK_MODES_100G: + MLX5E_FEC_OVERRIDE_ADMIN_POLICY(pplm, *fec_policy, write, 100g); + break; + case MLX5E_FEC_SUPPORTED_LINK_MODE_50G_1X: + MLX5E_FEC_OVERRIDE_ADMIN_50G_POLICY(pplm, *fec_policy, write, 50g_1x); + break; + case MLX5E_FEC_SUPPORTED_LINK_MODE_100G_2X: + MLX5E_FEC_OVERRIDE_ADMIN_50G_POLICY(pplm, *fec_policy, write, 100g_2x); + break; + case MLX5E_FEC_SUPPORTED_LINK_MODE_200G_4X: + MLX5E_FEC_OVERRIDE_ADMIN_50G_POLICY(pplm, *fec_policy, write, 200g_4x); + break; + case MLX5E_FEC_SUPPORTED_LINK_MODE_400G_8X: + MLX5E_FEC_OVERRIDE_ADMIN_50G_POLICY(pplm, *fec_policy, write, 400g_8x); break; default: return -EINVAL; @@ -408,32 +420,40 @@ static int mlx5e_fec_admin_field(u32 *pplm, return 0; } +#define MLX5E_GET_FEC_OVERRIDE_CAP(buf, link) \ + MLX5_GET(pplm_reg, buf, fec_override_cap_##link) + /* returns FEC capabilities for a given speed */ -static int mlx5e_get_fec_cap_field(u32 *pplm, - u8 *fec_cap, - u32 speed) +static int mlx5e_get_fec_cap_field(u32 *pplm, u16 *fec_cap, + enum mlx5e_fec_supported_link_mode link_mode) { - switch (speed) { - case 10000: - case 40000: - *fec_cap = MLX5_GET(pplm_reg, pplm, - fec_override_cap_10g_40g); + switch (link_mode) { + case MLX5E_FEC_SUPPORTED_LINK_MODES_10G_40G: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 10g_40g); + break; + case MLX5E_FEC_SUPPORTED_LINK_MODES_25G: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 25g); + break; + case MLX5E_FEC_SUPPORTED_LINK_MODES_50G: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 50g); + break; + case MLX5E_FEC_SUPPORTED_LINK_MODES_56G: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 56g); + break; + case MLX5E_FEC_SUPPORTED_LINK_MODES_100G: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 100g); break; - case 25000: - *fec_cap = MLX5_GET(pplm_reg, pplm, - fec_override_cap_25g); + case MLX5E_FEC_SUPPORTED_LINK_MODE_50G_1X: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 50g_1x); break; - case 50000: - *fec_cap = MLX5_GET(pplm_reg, pplm, - fec_override_cap_50g); + case MLX5E_FEC_SUPPORTED_LINK_MODE_100G_2X: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 100g_2x); break; - case 56000: - *fec_cap = MLX5_GET(pplm_reg, pplm, - fec_override_cap_56g); + case MLX5E_FEC_SUPPORTED_LINK_MODE_200G_4X: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 200g_4x); break; - case 100000: - *fec_cap = MLX5_GET(pplm_reg, pplm, - fec_override_cap_100g); + case MLX5E_FEC_SUPPORTED_LINK_MODE_400G_8X: + *fec_cap = MLX5E_GET_FEC_OVERRIDE_CAP(pplm, 400g_8x); break; default: return -EINVAL; @@ -441,13 +461,14 @@ static int mlx5e_get_fec_cap_field(u32 *pplm, return 0; } -int mlx5e_get_fec_caps(struct mlx5_core_dev *dev, u8 *fec_caps) +bool mlx5e_fec_in_caps(struct mlx5_core_dev *dev, int fec_policy) { + bool fec_50g_per_lane = MLX5_CAP_PCAM_FEATURE(dev, fec_50G_per_lane_in_pplm); u32 out[MLX5_ST_SZ_DW(pplm_reg)] = {}; u32 in[MLX5_ST_SZ_DW(pplm_reg)] = {}; int sz = MLX5_ST_SZ_BYTES(pplm_reg); - u32 current_fec_speed; int err; + int i; if (!MLX5_CAP_GEN(dev, pcam_reg)) return -EOPNOTSUPP; @@ -458,23 +479,30 @@ int mlx5e_get_fec_caps(struct mlx5_core_dev *dev, u8 *fec_caps) MLX5_SET(pplm_reg, in, local_port, 1); err = mlx5_core_access_reg(dev, in, sz, out, sz, MLX5_REG_PPLM, 0, 0); if (err) - return err; + return false; - err = mlx5e_port_linkspeed(dev, ¤t_fec_speed); - if (err) - return err; + for (i = 0; i < MLX5E_MAX_FEC_SUPPORTED_LINK_MODE; i++) { + u16 fec_caps; - return mlx5e_get_fec_cap_field(out, fec_caps, current_fec_speed); + if (i >= MLX5E_FEC_FIRST_50G_PER_LANE_MODE && !fec_50g_per_lane) + break; + + mlx5e_get_fec_cap_field(out, &fec_caps, i); + if (fec_caps & fec_policy) + return true; + } + return false; } int mlx5e_get_fec_mode(struct mlx5_core_dev *dev, u32 *fec_mode_active, - u8 *fec_configured_mode) + u16 *fec_configured_mode) { + bool fec_50g_per_lane = MLX5_CAP_PCAM_FEATURE(dev, fec_50G_per_lane_in_pplm); u32 out[MLX5_ST_SZ_DW(pplm_reg)] = {}; u32 in[MLX5_ST_SZ_DW(pplm_reg)] = {}; int sz = MLX5_ST_SZ_BYTES(pplm_reg); - u32 link_speed; int err; + int i; if (!MLX5_CAP_GEN(dev, pcam_reg)) return -EOPNOTSUPP; @@ -490,24 +518,28 @@ int mlx5e_get_fec_mode(struct mlx5_core_dev *dev, u32 *fec_mode_active, *fec_mode_active = MLX5_GET(pplm_reg, out, fec_mode_active); if (!fec_configured_mode) - return 0; + goto out; - err = mlx5e_port_linkspeed(dev, &link_speed); - if (err) - return err; + *fec_configured_mode = 0; + for (i = 0; i < MLX5E_MAX_FEC_SUPPORTED_LINK_MODE; i++) { + if (i >= MLX5E_FEC_FIRST_50G_PER_LANE_MODE && !fec_50g_per_lane) + break; - return mlx5e_fec_admin_field(out, fec_configured_mode, 0, link_speed); + mlx5e_fec_admin_field(out, fec_configured_mode, 0, i); + if (*fec_configured_mode != 0) + goto out; + } +out: + return 0; } -int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u8 fec_policy) +int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u16 fec_policy) { - u8 fec_policy_nofec = BIT(MLX5E_FEC_NOFEC); - bool fec_mode_not_supp_in_speed = false; + bool fec_50g_per_lane = MLX5_CAP_PCAM_FEATURE(dev, fec_50G_per_lane_in_pplm); u32 out[MLX5_ST_SZ_DW(pplm_reg)] = {}; u32 in[MLX5_ST_SZ_DW(pplm_reg)] = {}; int sz = MLX5_ST_SZ_BYTES(pplm_reg); - u8 fec_policy_auto = 0; - u8 fec_caps = 0; + u16 fec_policy_auto = 0; int err; int i; @@ -517,6 +549,9 @@ int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u8 fec_policy) if (!MLX5_CAP_PCAM_REG(dev, pplm)) return -EOPNOTSUPP; + if (fec_policy >= (1 << MLX5E_FEC_LLRS_272_257_1) && !fec_50g_per_lane) + return -EOPNOTSUPP; + MLX5_SET(pplm_reg, in, local_port, 1); err = mlx5_core_access_reg(dev, in, sz, out, sz, MLX5_REG_PPLM, 0, 0); if (err) @@ -524,25 +559,31 @@ int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u8 fec_policy) MLX5_SET(pplm_reg, out, local_port, 1); - for (i = 0; i < MLX5E_FEC_SUPPORTED_SPEEDS; i++) { - mlx5e_get_fec_cap_field(out, &fec_caps, fec_supported_speeds[i]); - /* policy supported for link speed, or policy is auto */ - if (fec_caps & fec_policy || fec_policy == fec_policy_auto) { - mlx5e_fec_admin_field(out, &fec_policy, 1, - fec_supported_speeds[i]); - } else { - /* turn off FEC if supported. Else, leave it the same */ - if (fec_caps & fec_policy_nofec) - mlx5e_fec_admin_field(out, &fec_policy_nofec, 1, - fec_supported_speeds[i]); - fec_mode_not_supp_in_speed = true; - } - } + for (i = 0; i < MLX5E_MAX_FEC_SUPPORTED_LINK_MODE; i++) { + u16 conf_fec = fec_policy; + u16 fec_caps = 0; + + if (i >= MLX5E_FEC_FIRST_50G_PER_LANE_MODE && !fec_50g_per_lane) + break; - if (fec_mode_not_supp_in_speed) - mlx5_core_dbg(dev, - "FEC policy 0x%x is not supported for some speeds", - fec_policy); + /* RS fec in ethtool is mapped to MLX5E_FEC_RS_528_514 + * to link modes up to 25G per lane and to + * MLX5E_FEC_RS_544_514 in the new link modes based on + * 50 G per lane + */ + if (conf_fec == (1 << MLX5E_FEC_RS_528_514) && + i >= MLX5E_FEC_FIRST_50G_PER_LANE_MODE) + conf_fec = (1 << MLX5E_FEC_RS_544_514); + + mlx5e_get_fec_cap_field(out, &fec_caps, i); + + /* policy supported for link speed */ + if (fec_caps & conf_fec) + mlx5e_fec_admin_field(out, &conf_fec, 1, i); + else + /* set FEC to auto*/ + mlx5e_fec_admin_field(out, &fec_policy_auto, 1, i); + } return mlx5_core_access_reg(dev, out, sz, out, sz, MLX5_REG_PPLM, 0, 1); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.h b/drivers/net/ethernet/mellanox/mlx5/core/en/port.h index 4a7f4497692b..a2ddd446dd59 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.h @@ -60,15 +60,17 @@ int mlx5e_port_set_pbmc(struct mlx5_core_dev *mdev, void *in); int mlx5e_port_query_priority2buffer(struct mlx5_core_dev *mdev, u8 *buffer); int mlx5e_port_set_priority2buffer(struct mlx5_core_dev *mdev, u8 *buffer); -int mlx5e_get_fec_caps(struct mlx5_core_dev *dev, u8 *fec_caps); +bool mlx5e_fec_in_caps(struct mlx5_core_dev *dev, int fec_policy); int mlx5e_get_fec_mode(struct mlx5_core_dev *dev, u32 *fec_mode_active, - u8 *fec_configured_mode); -int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u8 fec_policy); + u16 *fec_configured_mode); +int mlx5e_set_fec_mode(struct mlx5_core_dev *dev, u16 fec_policy); enum { MLX5E_FEC_NOFEC, MLX5E_FEC_FIRECODE, MLX5E_FEC_RS_528_514, + MLX5E_FEC_RS_544_514 = 7, + MLX5E_FEC_LLRS_272_257_1 = 9, }; #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c index a01e2de2488f..c209579fc213 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c @@ -102,19 +102,6 @@ out: return err; } -void mlx5e_reporter_icosq_cqe_err(struct mlx5e_icosq *icosq) -{ - struct mlx5e_priv *priv = icosq->channel->priv; - char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; - struct mlx5e_err_ctx err_ctx = {}; - - err_ctx.ctx = icosq; - err_ctx.recover = mlx5e_rx_reporter_err_icosq_cqe_recover; - sprintf(err_str, "ERR CQE on ICOSQ: 0x%x", icosq->sqn); - - mlx5e_health_report(priv, priv->rx_reporter, err_str, &err_ctx); -} - static int mlx5e_rq_to_ready(struct mlx5e_rq *rq, int curr_state) { struct net_device *dev = rq->netdev; @@ -171,19 +158,6 @@ out: return err; } -void mlx5e_reporter_rq_cqe_err(struct mlx5e_rq *rq) -{ - struct mlx5e_priv *priv = rq->channel->priv; - char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; - struct mlx5e_err_ctx err_ctx = {}; - - err_ctx.ctx = rq; - err_ctx.recover = mlx5e_rx_reporter_err_rq_cqe_recover; - sprintf(err_str, "ERR CQE on RQ: 0x%x", rq->rqn); - - mlx5e_health_report(priv, priv->rx_reporter, err_str, &err_ctx); -} - static int mlx5e_rx_reporter_timeout_recover(void *ctx) { struct mlx5e_icosq *icosq; @@ -201,21 +175,6 @@ static int mlx5e_rx_reporter_timeout_recover(void *ctx) return err; } -void mlx5e_reporter_rx_timeout(struct mlx5e_rq *rq) -{ - struct mlx5e_icosq *icosq = &rq->channel->icosq; - struct mlx5e_priv *priv = rq->channel->priv; - char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; - struct mlx5e_err_ctx err_ctx = {}; - - err_ctx.ctx = rq; - err_ctx.recover = mlx5e_rx_reporter_timeout_recover; - sprintf(err_str, "RX timeout on channel: %d, ICOSQ: 0x%x RQ: 0x%x, CQ: 0x%x\n", - icosq->channel->ix, icosq->sqn, rq->rqn, rq->cq.mcq.cqn); - - mlx5e_health_report(priv, priv->rx_reporter, err_str, &err_ctx); -} - static int mlx5e_rx_reporter_recover_from_ctx(struct mlx5e_err_ctx *err_ctx) { return err_ctx->recover(err_ctx->ctx); @@ -371,10 +330,235 @@ unlock: return err; } +static int mlx5e_rx_reporter_dump_icosq(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg, + void *ctx) +{ + struct mlx5e_txqsq *icosq = ctx; + struct mlx5_rsc_key key = {}; + int err; + + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + return 0; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "SX Slice"); + if (err) + return err; + + key.size = PAGE_SIZE; + key.rsc = MLX5_SGMT_TYPE_SX_SLICE_ALL; + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "ICOSQ"); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "QPC"); + if (err) + return err; + + key.rsc = MLX5_SGMT_TYPE_FULL_QPC; + key.index1 = icosq->sqn; + key.num_of_obj1 = 1; + + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "send_buff"); + if (err) + return err; + + key.rsc = MLX5_SGMT_TYPE_SND_BUFF; + key.num_of_obj2 = MLX5_RSC_DUMP_ALL; + + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + return mlx5e_reporter_named_obj_nest_end(fmsg); +} + +static int mlx5e_rx_reporter_dump_rq(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg, + void *ctx) +{ + struct mlx5_rsc_key key = {}; + struct mlx5e_rq *rq = ctx; + int err; + + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + return 0; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "RX Slice"); + if (err) + return err; + + key.size = PAGE_SIZE; + key.rsc = MLX5_SGMT_TYPE_RX_SLICE_ALL; + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "RQ"); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "QPC"); + if (err) + return err; + + key.rsc = MLX5_SGMT_TYPE_FULL_QPC; + key.index1 = rq->rqn; + key.num_of_obj1 = 1; + + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "receive_buff"); + if (err) + return err; + + key.rsc = MLX5_SGMT_TYPE_RCV_BUFF; + key.num_of_obj2 = MLX5_RSC_DUMP_ALL; + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + return mlx5e_reporter_named_obj_nest_end(fmsg); +} + +static int mlx5e_rx_reporter_dump_all_rqs(struct mlx5e_priv *priv, + struct devlink_fmsg *fmsg) +{ + struct mlx5_rsc_key key = {}; + int i, err; + + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + return 0; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "RX Slice"); + if (err) + return err; + + key.size = PAGE_SIZE; + key.rsc = MLX5_SGMT_TYPE_RX_SLICE_ALL; + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + err = devlink_fmsg_arr_pair_nest_start(fmsg, "RQs"); + if (err) + return err; + + for (i = 0; i < priv->channels.num; i++) { + struct mlx5e_rq *rq = &priv->channels.c[i]->rq; + + err = mlx5e_health_queue_dump(priv, fmsg, rq->rqn, "RQ"); + if (err) + return err; + } + + return devlink_fmsg_arr_pair_nest_end(fmsg); +} + +static int mlx5e_rx_reporter_dump_from_ctx(struct mlx5e_priv *priv, + struct mlx5e_err_ctx *err_ctx, + struct devlink_fmsg *fmsg) +{ + return err_ctx->dump(priv, fmsg, err_ctx->ctx); +} + +static int mlx5e_rx_reporter_dump(struct devlink_health_reporter *reporter, + struct devlink_fmsg *fmsg, void *context, + struct netlink_ext_ack *extack) +{ + struct mlx5e_priv *priv = devlink_health_reporter_priv(reporter); + struct mlx5e_err_ctx *err_ctx = context; + + return err_ctx ? mlx5e_rx_reporter_dump_from_ctx(priv, err_ctx, fmsg) : + mlx5e_rx_reporter_dump_all_rqs(priv, fmsg); +} + +void mlx5e_reporter_rx_timeout(struct mlx5e_rq *rq) +{ + struct mlx5e_icosq *icosq = &rq->channel->icosq; + struct mlx5e_priv *priv = rq->channel->priv; + char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; + struct mlx5e_err_ctx err_ctx = {}; + + err_ctx.ctx = rq; + err_ctx.recover = mlx5e_rx_reporter_timeout_recover; + err_ctx.dump = mlx5e_rx_reporter_dump_rq; + snprintf(err_str, sizeof(err_str), + "RX timeout on channel: %d, ICOSQ: 0x%x RQ: 0x%x, CQ: 0x%x", + icosq->channel->ix, icosq->sqn, rq->rqn, rq->cq.mcq.cqn); + + mlx5e_health_report(priv, priv->rx_reporter, err_str, &err_ctx); +} + +void mlx5e_reporter_rq_cqe_err(struct mlx5e_rq *rq) +{ + struct mlx5e_priv *priv = rq->channel->priv; + char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; + struct mlx5e_err_ctx err_ctx = {}; + + err_ctx.ctx = rq; + err_ctx.recover = mlx5e_rx_reporter_err_rq_cqe_recover; + err_ctx.dump = mlx5e_rx_reporter_dump_rq; + snprintf(err_str, sizeof(err_str), "ERR CQE on RQ: 0x%x", rq->rqn); + + mlx5e_health_report(priv, priv->rx_reporter, err_str, &err_ctx); +} + +void mlx5e_reporter_icosq_cqe_err(struct mlx5e_icosq *icosq) +{ + struct mlx5e_priv *priv = icosq->channel->priv; + char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; + struct mlx5e_err_ctx err_ctx = {}; + + err_ctx.ctx = icosq; + err_ctx.recover = mlx5e_rx_reporter_err_icosq_cqe_recover; + err_ctx.dump = mlx5e_rx_reporter_dump_icosq; + snprintf(err_str, sizeof(err_str), "ERR CQE on ICOSQ: 0x%x", icosq->sqn); + + mlx5e_health_report(priv, priv->rx_reporter, err_str, &err_ctx); +} + static const struct devlink_health_reporter_ops mlx5_rx_reporter_ops = { .name = "rx", .recover = mlx5e_rx_reporter_recover, .diagnose = mlx5e_rx_reporter_diagnose, + .dump = mlx5e_rx_reporter_dump, }; #define MLX5E_REPORTER_RX_GRACEFUL_PERIOD 500 @@ -387,7 +571,7 @@ int mlx5e_reporter_rx_create(struct mlx5e_priv *priv) reporter = devlink_health_reporter_create(devlink, &mlx5_rx_reporter_ops, MLX5E_REPORTER_RX_GRACEFUL_PERIOD, - true, priv); + priv); if (IS_ERR(reporter)) { netdev_warn(priv->netdev, "Failed to create rx reporter, err = %ld\n", PTR_ERR(reporter)); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c index b468549e96ff..9805fc085512 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c @@ -82,19 +82,6 @@ out: return err; } -void mlx5e_reporter_tx_err_cqe(struct mlx5e_txqsq *sq) -{ - struct mlx5e_priv *priv = sq->channel->priv; - char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; - struct mlx5e_err_ctx err_ctx = {0}; - - err_ctx.ctx = sq; - err_ctx.recover = mlx5e_tx_reporter_err_cqe_recover; - sprintf(err_str, "ERR CQE on SQ: 0x%x", sq->sqn); - - mlx5e_health_report(priv, priv->tx_reporter, err_str, &err_ctx); -} - static int mlx5e_tx_reporter_timeout_recover(void *ctx) { struct mlx5_eq_comp *eq; @@ -110,22 +97,6 @@ static int mlx5e_tx_reporter_timeout_recover(void *ctx) return err; } -int mlx5e_reporter_tx_timeout(struct mlx5e_txqsq *sq) -{ - struct mlx5e_priv *priv = sq->channel->priv; - char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; - struct mlx5e_err_ctx err_ctx; - - err_ctx.ctx = sq; - err_ctx.recover = mlx5e_tx_reporter_timeout_recover; - sprintf(err_str, - "TX timeout on queue: %d, SQ: 0x%x, CQ: 0x%x, SQ Cons: 0x%x SQ Prod: 0x%x, usecs since last trans: %u\n", - sq->channel->ix, sq->sqn, sq->cq.mcq.cqn, sq->cc, sq->pc, - jiffies_to_usecs(jiffies - sq->txq->trans_start)); - - return mlx5e_health_report(priv, priv->tx_reporter, err_str, &err_ctx); -} - /* state lock cannot be grabbed within this function. * It can cause a dead lock or a read-after-free. */ @@ -275,10 +246,162 @@ unlock: return err; } +static int mlx5e_tx_reporter_dump_sq(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg, + void *ctx) +{ + struct mlx5_rsc_key key = {}; + struct mlx5e_txqsq *sq = ctx; + int err; + + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + return 0; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "SX Slice"); + if (err) + return err; + + key.size = PAGE_SIZE; + key.rsc = MLX5_SGMT_TYPE_SX_SLICE_ALL; + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "SQ"); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "QPC"); + if (err) + return err; + + key.rsc = MLX5_SGMT_TYPE_FULL_QPC; + key.index1 = sq->sqn; + key.num_of_obj1 = 1; + + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "send_buff"); + if (err) + return err; + + key.rsc = MLX5_SGMT_TYPE_SND_BUFF; + key.num_of_obj2 = MLX5_RSC_DUMP_ALL; + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + return mlx5e_reporter_named_obj_nest_end(fmsg); +} + +static int mlx5e_tx_reporter_dump_all_sqs(struct mlx5e_priv *priv, + struct devlink_fmsg *fmsg) +{ + struct mlx5_rsc_key key = {}; + int i, tc, err; + + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + return 0; + + err = mlx5e_reporter_named_obj_nest_start(fmsg, "SX Slice"); + if (err) + return err; + + key.size = PAGE_SIZE; + key.rsc = MLX5_SGMT_TYPE_SX_SLICE_ALL; + err = mlx5e_health_rsc_fmsg_dump(priv, &key, fmsg); + if (err) + return err; + + err = mlx5e_reporter_named_obj_nest_end(fmsg); + if (err) + return err; + + err = devlink_fmsg_arr_pair_nest_start(fmsg, "SQs"); + if (err) + return err; + + for (i = 0; i < priv->channels.num; i++) { + struct mlx5e_channel *c = priv->channels.c[i]; + + for (tc = 0; tc < priv->channels.params.num_tc; tc++) { + struct mlx5e_txqsq *sq = &c->sq[tc]; + + err = mlx5e_health_queue_dump(priv, fmsg, sq->sqn, "SQ"); + if (err) + return err; + } + } + return devlink_fmsg_arr_pair_nest_end(fmsg); +} + +static int mlx5e_tx_reporter_dump_from_ctx(struct mlx5e_priv *priv, + struct mlx5e_err_ctx *err_ctx, + struct devlink_fmsg *fmsg) +{ + return err_ctx->dump(priv, fmsg, err_ctx->ctx); +} + +static int mlx5e_tx_reporter_dump(struct devlink_health_reporter *reporter, + struct devlink_fmsg *fmsg, void *context, + struct netlink_ext_ack *extack) +{ + struct mlx5e_priv *priv = devlink_health_reporter_priv(reporter); + struct mlx5e_err_ctx *err_ctx = context; + + return err_ctx ? mlx5e_tx_reporter_dump_from_ctx(priv, err_ctx, fmsg) : + mlx5e_tx_reporter_dump_all_sqs(priv, fmsg); +} + +void mlx5e_reporter_tx_err_cqe(struct mlx5e_txqsq *sq) +{ + struct mlx5e_priv *priv = sq->channel->priv; + char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; + struct mlx5e_err_ctx err_ctx = {}; + + err_ctx.ctx = sq; + err_ctx.recover = mlx5e_tx_reporter_err_cqe_recover; + err_ctx.dump = mlx5e_tx_reporter_dump_sq; + snprintf(err_str, sizeof(err_str), "ERR CQE on SQ: 0x%x", sq->sqn); + + mlx5e_health_report(priv, priv->tx_reporter, err_str, &err_ctx); +} + +int mlx5e_reporter_tx_timeout(struct mlx5e_txqsq *sq) +{ + struct mlx5e_priv *priv = sq->channel->priv; + char err_str[MLX5E_REPORTER_PER_Q_MAX_LEN]; + struct mlx5e_err_ctx err_ctx = {}; + + err_ctx.ctx = sq; + err_ctx.recover = mlx5e_tx_reporter_timeout_recover; + err_ctx.dump = mlx5e_tx_reporter_dump_sq; + snprintf(err_str, sizeof(err_str), + "TX timeout on queue: %d, SQ: 0x%x, CQ: 0x%x, SQ Cons: 0x%x SQ Prod: 0x%x, usecs since last trans: %u", + sq->channel->ix, sq->sqn, sq->cq.mcq.cqn, sq->cc, sq->pc, + jiffies_to_usecs(jiffies - sq->txq->trans_start)); + + return mlx5e_health_report(priv, priv->tx_reporter, err_str, &err_ctx); +} + static const struct devlink_health_reporter_ops mlx5_tx_reporter_ops = { .name = "tx", .recover = mlx5e_tx_reporter_recover, .diagnose = mlx5e_tx_reporter_diagnose, + .dump = mlx5e_tx_reporter_dump, }; #define MLX5_REPORTER_TX_GRACEFUL_PERIOD 500 @@ -293,7 +416,7 @@ int mlx5e_reporter_tx_create(struct mlx5e_priv *priv) reporter = devlink_health_reporter_create(devlink, &mlx5_tx_reporter_ops, MLX5_REPORTER_TX_GRACEFUL_PERIOD, - true, priv); + priv); if (IS_ERR(reporter)) { netdev_warn(priv->netdev, "Failed to create tx reporter, err = %ld\n", diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c new file mode 100644 index 000000000000..ad3e3a65d403 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c @@ -0,0 +1,1369 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2019 Mellanox Technologies. */ + +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/nf_conntrack_zones.h> +#include <net/netfilter/nf_conntrack_labels.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <net/netfilter/nf_conntrack_acct.h> +#include <uapi/linux/tc_act/tc_pedit.h> +#include <net/tc_act/tc_ct.h> +#include <net/flow_offload.h> +#include <net/netfilter/nf_flow_table.h> +#include <linux/workqueue.h> + +#include "esw/chains.h" +#include "en/tc_ct.h" +#include "en.h" +#include "en_tc.h" +#include "en_rep.h" + +#define MLX5_CT_ZONE_BITS (mlx5e_tc_attr_to_reg_mappings[ZONE_TO_REG].mlen * 8) +#define MLX5_CT_ZONE_MASK GENMASK(MLX5_CT_ZONE_BITS - 1, 0) +#define MLX5_CT_STATE_ESTABLISHED_BIT BIT(1) +#define MLX5_CT_STATE_TRK_BIT BIT(2) + +#define MLX5_FTE_ID_BITS (mlx5e_tc_attr_to_reg_mappings[FTEID_TO_REG].mlen * 8) +#define MLX5_FTE_ID_MAX GENMASK(MLX5_FTE_ID_BITS - 1, 0) +#define MLX5_FTE_ID_MASK MLX5_FTE_ID_MAX + +#define ct_dbg(fmt, args...)\ + netdev_dbg(ct_priv->netdev, "ct_debug: " fmt "\n", ##args) + +struct mlx5_tc_ct_priv { + struct mlx5_eswitch *esw; + const struct net_device *netdev; + struct idr fte_ids; + struct idr tuple_ids; + struct rhashtable zone_ht; + struct mlx5_flow_table *ct; + struct mlx5_flow_table *ct_nat; + struct mlx5_flow_table *post_ct; + struct mutex control_lock; /* guards parallel adds/dels */ +}; + +struct mlx5_ct_flow { + struct mlx5_esw_flow_attr pre_ct_attr; + struct mlx5_esw_flow_attr post_ct_attr; + struct mlx5_flow_handle *pre_ct_rule; + struct mlx5_flow_handle *post_ct_rule; + struct mlx5_ct_ft *ft; + u32 fte_id; + u32 chain_mapping; +}; + +struct mlx5_ct_zone_rule { + struct mlx5_flow_handle *rule; + struct mlx5_esw_flow_attr attr; + int tupleid; + bool nat; +}; + +struct mlx5_ct_ft { + struct rhash_head node; + u16 zone; + refcount_t refcount; + struct nf_flowtable *nf_ft; + struct mlx5_tc_ct_priv *ct_priv; + struct rhashtable ct_entries_ht; + struct list_head ct_entries_list; +}; + +struct mlx5_ct_entry { + struct list_head list; + u16 zone; + struct rhash_head node; + struct flow_rule *flow_rule; + struct mlx5_fc *counter; + unsigned long lastuse; + unsigned long cookie; + unsigned long restore_cookie; + struct mlx5_ct_zone_rule zone_rules[2]; +}; + +static const struct rhashtable_params cts_ht_params = { + .head_offset = offsetof(struct mlx5_ct_entry, node), + .key_offset = offsetof(struct mlx5_ct_entry, cookie), + .key_len = sizeof(((struct mlx5_ct_entry *)0)->cookie), + .automatic_shrinking = true, + .min_size = 16 * 1024, +}; + +static const struct rhashtable_params zone_params = { + .head_offset = offsetof(struct mlx5_ct_ft, node), + .key_offset = offsetof(struct mlx5_ct_ft, zone), + .key_len = sizeof(((struct mlx5_ct_ft *)0)->zone), + .automatic_shrinking = true, +}; + +static struct mlx5_tc_ct_priv * +mlx5_tc_ct_get_ct_priv(struct mlx5e_priv *priv) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *uplink_rpriv; + + uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + uplink_priv = &uplink_rpriv->uplink_priv; + return uplink_priv->ct_priv; +} + +static int +mlx5_tc_ct_set_tuple_match(struct mlx5_flow_spec *spec, + struct flow_rule *rule) +{ + void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers); + void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers); + u16 addr_type = 0; + u8 ip_proto = 0; + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) { + struct flow_match_basic match; + + flow_rule_match_basic(rule, &match); + + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, + ntohs(match.mask->n_proto)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, + ntohs(match.key->n_proto)); + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_protocol, + match.mask->ip_proto); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_protocol, + match.key->ip_proto); + + ip_proto = match.key->ip_proto; + } + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CONTROL)) { + struct flow_match_control match; + + flow_rule_match_control(rule, &match); + addr_type = match.key->addr_type; + } + + if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) { + struct flow_match_ipv4_addrs match; + + flow_rule_match_ipv4_addrs(rule, &match); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + &match.mask->src, sizeof(match.mask->src)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + &match.key->src, sizeof(match.key->src)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + &match.mask->dst, sizeof(match.mask->dst)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + &match.key->dst, sizeof(match.key->dst)); + } + + if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) { + struct flow_match_ipv6_addrs match; + + flow_rule_match_ipv6_addrs(rule, &match); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, + src_ipv4_src_ipv6.ipv6_layout.ipv6), + &match.mask->src, sizeof(match.mask->src)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + src_ipv4_src_ipv6.ipv6_layout.ipv6), + &match.key->src, sizeof(match.key->src)); + + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, + dst_ipv4_dst_ipv6.ipv6_layout.ipv6), + &match.mask->dst, sizeof(match.mask->dst)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + dst_ipv4_dst_ipv6.ipv6_layout.ipv6), + &match.key->dst, sizeof(match.key->dst)); + } + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_PORTS)) { + struct flow_match_ports match; + + flow_rule_match_ports(rule, &match); + switch (ip_proto) { + case IPPROTO_TCP: + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + tcp_sport, ntohs(match.mask->src)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + tcp_sport, ntohs(match.key->src)); + + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + tcp_dport, ntohs(match.mask->dst)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + tcp_dport, ntohs(match.key->dst)); + break; + + case IPPROTO_UDP: + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + udp_sport, ntohs(match.mask->src)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + udp_sport, ntohs(match.key->src)); + + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + udp_dport, ntohs(match.mask->dst)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + udp_dport, ntohs(match.key->dst)); + break; + default: + break; + } + } + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_TCP)) { + struct flow_match_tcp match; + + flow_rule_match_tcp(rule, &match); + MLX5_SET(fte_match_set_lyr_2_4, headers_c, tcp_flags, + ntohs(match.mask->flags)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, tcp_flags, + ntohs(match.key->flags)); + } + + return 0; +} + +static void +mlx5_tc_ct_entry_del_rule(struct mlx5_tc_ct_priv *ct_priv, + struct mlx5_ct_entry *entry, + bool nat) +{ + struct mlx5_ct_zone_rule *zone_rule = &entry->zone_rules[nat]; + struct mlx5_esw_flow_attr *attr = &zone_rule->attr; + struct mlx5_eswitch *esw = ct_priv->esw; + + ct_dbg("Deleting ct entry rule in zone %d", entry->zone); + + mlx5_eswitch_del_offloaded_rule(esw, zone_rule->rule, attr); + mlx5_modify_header_dealloc(esw->dev, attr->modify_hdr); + idr_remove(&ct_priv->tuple_ids, zone_rule->tupleid); +} + +static void +mlx5_tc_ct_entry_del_rules(struct mlx5_tc_ct_priv *ct_priv, + struct mlx5_ct_entry *entry) +{ + mlx5_tc_ct_entry_del_rule(ct_priv, entry, true); + mlx5_tc_ct_entry_del_rule(ct_priv, entry, false); + + mlx5_fc_destroy(ct_priv->esw->dev, entry->counter); +} + +static struct flow_action_entry * +mlx5_tc_ct_get_ct_metadata_action(struct flow_rule *flow_rule) +{ + struct flow_action *flow_action = &flow_rule->action; + struct flow_action_entry *act; + int i; + + flow_action_for_each(i, act, flow_action) { + if (act->id == FLOW_ACTION_CT_METADATA) + return act; + } + + return NULL; +} + +static int +mlx5_tc_ct_entry_set_registers(struct mlx5_tc_ct_priv *ct_priv, + struct mlx5e_tc_mod_hdr_acts *mod_acts, + u8 ct_state, + u32 mark, + u32 label, + u32 tupleid) +{ + struct mlx5_eswitch *esw = ct_priv->esw; + int err; + + err = mlx5e_tc_match_to_reg_set(esw->dev, mod_acts, + CTSTATE_TO_REG, ct_state); + if (err) + return err; + + err = mlx5e_tc_match_to_reg_set(esw->dev, mod_acts, + MARK_TO_REG, mark); + if (err) + return err; + + err = mlx5e_tc_match_to_reg_set(esw->dev, mod_acts, + LABELS_TO_REG, label); + if (err) + return err; + + err = mlx5e_tc_match_to_reg_set(esw->dev, mod_acts, + TUPLEID_TO_REG, tupleid); + if (err) + return err; + + return 0; +} + +static int +mlx5_tc_ct_parse_mangle_to_mod_act(struct flow_action_entry *act, + char *modact) +{ + u32 offset = act->mangle.offset, field; + + switch (act->mangle.htype) { + case FLOW_ACT_MANGLE_HDR_TYPE_IP4: + MLX5_SET(set_action_in, modact, length, 0); + if (offset == offsetof(struct iphdr, saddr)) + field = MLX5_ACTION_IN_FIELD_OUT_SIPV4; + else if (offset == offsetof(struct iphdr, daddr)) + field = MLX5_ACTION_IN_FIELD_OUT_DIPV4; + else + return -EOPNOTSUPP; + break; + + case FLOW_ACT_MANGLE_HDR_TYPE_IP6: + MLX5_SET(set_action_in, modact, length, 0); + if (offset == offsetof(struct ipv6hdr, saddr)) + field = MLX5_ACTION_IN_FIELD_OUT_SIPV6_31_0; + else if (offset == offsetof(struct ipv6hdr, saddr) + 4) + field = MLX5_ACTION_IN_FIELD_OUT_SIPV6_63_32; + else if (offset == offsetof(struct ipv6hdr, saddr) + 8) + field = MLX5_ACTION_IN_FIELD_OUT_SIPV6_95_64; + else if (offset == offsetof(struct ipv6hdr, saddr) + 12) + field = MLX5_ACTION_IN_FIELD_OUT_SIPV6_127_96; + else if (offset == offsetof(struct ipv6hdr, daddr)) + field = MLX5_ACTION_IN_FIELD_OUT_DIPV6_31_0; + else if (offset == offsetof(struct ipv6hdr, daddr) + 4) + field = MLX5_ACTION_IN_FIELD_OUT_DIPV6_63_32; + else if (offset == offsetof(struct ipv6hdr, daddr) + 8) + field = MLX5_ACTION_IN_FIELD_OUT_DIPV6_95_64; + else if (offset == offsetof(struct ipv6hdr, daddr) + 12) + field = MLX5_ACTION_IN_FIELD_OUT_DIPV6_127_96; + else + return -EOPNOTSUPP; + break; + + case FLOW_ACT_MANGLE_HDR_TYPE_TCP: + MLX5_SET(set_action_in, modact, length, 16); + if (offset == offsetof(struct tcphdr, source)) + field = MLX5_ACTION_IN_FIELD_OUT_TCP_SPORT; + else if (offset == offsetof(struct tcphdr, dest)) + field = MLX5_ACTION_IN_FIELD_OUT_TCP_DPORT; + else + return -EOPNOTSUPP; + break; + + case FLOW_ACT_MANGLE_HDR_TYPE_UDP: + MLX5_SET(set_action_in, modact, length, 16); + if (offset == offsetof(struct udphdr, source)) + field = MLX5_ACTION_IN_FIELD_OUT_UDP_SPORT; + else if (offset == offsetof(struct udphdr, dest)) + field = MLX5_ACTION_IN_FIELD_OUT_UDP_DPORT; + else + return -EOPNOTSUPP; + break; + + default: + return -EOPNOTSUPP; + } + + MLX5_SET(set_action_in, modact, action_type, MLX5_ACTION_TYPE_SET); + MLX5_SET(set_action_in, modact, offset, 0); + MLX5_SET(set_action_in, modact, field, field); + MLX5_SET(set_action_in, modact, data, act->mangle.val); + + return 0; +} + +static int +mlx5_tc_ct_entry_create_nat(struct mlx5_tc_ct_priv *ct_priv, + struct flow_rule *flow_rule, + struct mlx5e_tc_mod_hdr_acts *mod_acts) +{ + struct flow_action *flow_action = &flow_rule->action; + struct mlx5_core_dev *mdev = ct_priv->esw->dev; + struct flow_action_entry *act; + size_t action_size; + char *modact; + int err, i; + + action_size = MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto); + + flow_action_for_each(i, act, flow_action) { + switch (act->id) { + case FLOW_ACTION_MANGLE: { + err = alloc_mod_hdr_actions(mdev, + MLX5_FLOW_NAMESPACE_FDB, + mod_acts); + if (err) + return err; + + modact = mod_acts->actions + + mod_acts->num_actions * action_size; + + err = mlx5_tc_ct_parse_mangle_to_mod_act(act, modact); + if (err) + return err; + + mod_acts->num_actions++; + } + break; + + case FLOW_ACTION_CT_METADATA: + /* Handled earlier */ + continue; + default: + return -EOPNOTSUPP; + } + } + + return 0; +} + +static int +mlx5_tc_ct_entry_create_mod_hdr(struct mlx5_tc_ct_priv *ct_priv, + struct mlx5_esw_flow_attr *attr, + struct flow_rule *flow_rule, + u32 tupleid, + bool nat) +{ + struct mlx5e_tc_mod_hdr_acts mod_acts = {}; + struct mlx5_eswitch *esw = ct_priv->esw; + struct mlx5_modify_hdr *mod_hdr; + struct flow_action_entry *meta; + int err; + + meta = mlx5_tc_ct_get_ct_metadata_action(flow_rule); + if (!meta) + return -EOPNOTSUPP; + + if (meta->ct_metadata.labels[1] || + meta->ct_metadata.labels[2] || + meta->ct_metadata.labels[3]) { + ct_dbg("Failed to offload ct entry due to unsupported label"); + return -EOPNOTSUPP; + } + + if (nat) { + err = mlx5_tc_ct_entry_create_nat(ct_priv, flow_rule, + &mod_acts); + if (err) + goto err_mapping; + } + + err = mlx5_tc_ct_entry_set_registers(ct_priv, &mod_acts, + (MLX5_CT_STATE_ESTABLISHED_BIT | + MLX5_CT_STATE_TRK_BIT), + meta->ct_metadata.mark, + meta->ct_metadata.labels[0], + tupleid); + if (err) + goto err_mapping; + + mod_hdr = mlx5_modify_header_alloc(esw->dev, MLX5_FLOW_NAMESPACE_FDB, + mod_acts.num_actions, + mod_acts.actions); + if (IS_ERR(mod_hdr)) { + err = PTR_ERR(mod_hdr); + goto err_mapping; + } + attr->modify_hdr = mod_hdr; + + dealloc_mod_hdr_actions(&mod_acts); + return 0; + +err_mapping: + dealloc_mod_hdr_actions(&mod_acts); + return err; +} + +static int +mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, + struct flow_rule *flow_rule, + struct mlx5_ct_entry *entry, + bool nat) +{ + struct mlx5_ct_zone_rule *zone_rule = &entry->zone_rules[nat]; + struct mlx5_esw_flow_attr *attr = &zone_rule->attr; + struct mlx5_eswitch *esw = ct_priv->esw; + struct mlx5_flow_spec *spec = NULL; + u32 tupleid = 1; + int err; + + zone_rule->nat = nat; + + spec = kzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) + return -ENOMEM; + + /* Get tuple unique id */ + err = idr_alloc_u32(&ct_priv->tuple_ids, zone_rule, &tupleid, + TUPLE_ID_MAX, GFP_KERNEL); + if (err) { + netdev_warn(ct_priv->netdev, + "Failed to allocate tuple id, err: %d\n", err); + goto err_idr_alloc; + } + zone_rule->tupleid = tupleid; + + err = mlx5_tc_ct_entry_create_mod_hdr(ct_priv, attr, flow_rule, + tupleid, nat); + if (err) { + ct_dbg("Failed to create ct entry mod hdr"); + goto err_mod_hdr; + } + + attr->action = MLX5_FLOW_CONTEXT_ACTION_MOD_HDR | + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | + MLX5_FLOW_CONTEXT_ACTION_COUNT; + attr->dest_chain = 0; + attr->dest_ft = ct_priv->post_ct; + attr->fdb = nat ? ct_priv->ct_nat : ct_priv->ct; + attr->outer_match_level = MLX5_MATCH_L4; + attr->counter = entry->counter; + attr->flags |= MLX5_ESW_ATTR_FLAG_NO_IN_PORT; + + mlx5_tc_ct_set_tuple_match(spec, flow_rule); + mlx5e_tc_match_to_reg_match(spec, ZONE_TO_REG, + entry->zone & MLX5_CT_ZONE_MASK, + MLX5_CT_ZONE_MASK); + + zone_rule->rule = mlx5_eswitch_add_offloaded_rule(esw, spec, attr); + if (IS_ERR(zone_rule->rule)) { + err = PTR_ERR(zone_rule->rule); + ct_dbg("Failed to add ct entry rule, nat: %d", nat); + goto err_rule; + } + + kfree(spec); + ct_dbg("Offloaded ct entry rule in zone %d", entry->zone); + + return 0; + +err_rule: + mlx5_modify_header_dealloc(esw->dev, attr->modify_hdr); +err_mod_hdr: + idr_remove(&ct_priv->tuple_ids, zone_rule->tupleid); +err_idr_alloc: + kfree(spec); + return err; +} + +static int +mlx5_tc_ct_entry_add_rules(struct mlx5_tc_ct_priv *ct_priv, + struct flow_rule *flow_rule, + struct mlx5_ct_entry *entry) +{ + struct mlx5_eswitch *esw = ct_priv->esw; + int err; + + entry->counter = mlx5_fc_create(esw->dev, true); + if (IS_ERR(entry->counter)) { + err = PTR_ERR(entry->counter); + ct_dbg("Failed to create counter for ct entry"); + return err; + } + + err = mlx5_tc_ct_entry_add_rule(ct_priv, flow_rule, entry, false); + if (err) + goto err_orig; + + err = mlx5_tc_ct_entry_add_rule(ct_priv, flow_rule, entry, true); + if (err) + goto err_nat; + + return 0; + +err_nat: + mlx5_tc_ct_entry_del_rule(ct_priv, entry, false); +err_orig: + mlx5_fc_destroy(esw->dev, entry->counter); + return err; +} + +static int +mlx5_tc_ct_block_flow_offload_add(struct mlx5_ct_ft *ft, + struct flow_cls_offload *flow) +{ + struct flow_rule *flow_rule = flow_cls_offload_flow_rule(flow); + struct mlx5_tc_ct_priv *ct_priv = ft->ct_priv; + struct flow_action_entry *meta_action; + unsigned long cookie = flow->cookie; + struct mlx5_ct_entry *entry; + int err; + + meta_action = mlx5_tc_ct_get_ct_metadata_action(flow_rule); + if (!meta_action) + return -EOPNOTSUPP; + + entry = rhashtable_lookup_fast(&ft->ct_entries_ht, &cookie, + cts_ht_params); + if (entry) + return 0; + + entry = kzalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) + return -ENOMEM; + + entry->zone = ft->zone; + entry->flow_rule = flow_rule; + entry->cookie = flow->cookie; + entry->restore_cookie = meta_action->ct_metadata.cookie; + + err = mlx5_tc_ct_entry_add_rules(ct_priv, flow_rule, entry); + if (err) + goto err_rules; + + err = rhashtable_insert_fast(&ft->ct_entries_ht, &entry->node, + cts_ht_params); + if (err) + goto err_insert; + + list_add(&entry->list, &ft->ct_entries_list); + + return 0; + +err_insert: + mlx5_tc_ct_entry_del_rules(ct_priv, entry); +err_rules: + kfree(entry); + netdev_warn(ct_priv->netdev, + "Failed to offload ct entry, err: %d\n", err); + return err; +} + +static int +mlx5_tc_ct_block_flow_offload_del(struct mlx5_ct_ft *ft, + struct flow_cls_offload *flow) +{ + unsigned long cookie = flow->cookie; + struct mlx5_ct_entry *entry; + + entry = rhashtable_lookup_fast(&ft->ct_entries_ht, &cookie, + cts_ht_params); + if (!entry) + return -ENOENT; + + mlx5_tc_ct_entry_del_rules(ft->ct_priv, entry); + WARN_ON(rhashtable_remove_fast(&ft->ct_entries_ht, + &entry->node, + cts_ht_params)); + list_del(&entry->list); + kfree(entry); + + return 0; +} + +static int +mlx5_tc_ct_block_flow_offload_stats(struct mlx5_ct_ft *ft, + struct flow_cls_offload *f) +{ + unsigned long cookie = f->cookie; + struct mlx5_ct_entry *entry; + u64 lastuse, packets, bytes; + + entry = rhashtable_lookup_fast(&ft->ct_entries_ht, &cookie, + cts_ht_params); + if (!entry) + return -ENOENT; + + mlx5_fc_query_cached(entry->counter, &bytes, &packets, &lastuse); + flow_stats_update(&f->stats, bytes, packets, lastuse, + FLOW_ACTION_HW_STATS_DELAYED); + + return 0; +} + +static int +mlx5_tc_ct_block_flow_offload(enum tc_setup_type type, void *type_data, + void *cb_priv) +{ + struct flow_cls_offload *f = type_data; + struct mlx5_ct_ft *ft = cb_priv; + + if (type != TC_SETUP_CLSFLOWER) + return -EOPNOTSUPP; + + switch (f->command) { + case FLOW_CLS_REPLACE: + return mlx5_tc_ct_block_flow_offload_add(ft, f); + case FLOW_CLS_DESTROY: + return mlx5_tc_ct_block_flow_offload_del(ft, f); + case FLOW_CLS_STATS: + return mlx5_tc_ct_block_flow_offload_stats(ft, f); + default: + break; + }; + + return -EOPNOTSUPP; +} + +int +mlx5_tc_ct_parse_match(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct flow_cls_offload *f, + struct netlink_ext_ack *extack) +{ + struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); + struct flow_dissector_key_ct *mask, *key; + bool trk, est, untrk, unest, new; + u32 ctstate = 0, ctstate_mask = 0; + u16 ct_state_on, ct_state_off; + u16 ct_state, ct_state_mask; + struct flow_match_ct match; + + if (!flow_rule_match_key(f->rule, FLOW_DISSECTOR_KEY_CT)) + return 0; + + if (!ct_priv) { + NL_SET_ERR_MSG_MOD(extack, + "offload of ct matching isn't available"); + return -EOPNOTSUPP; + } + + flow_rule_match_ct(f->rule, &match); + + key = match.key; + mask = match.mask; + + ct_state = key->ct_state; + ct_state_mask = mask->ct_state; + + if (ct_state_mask & ~(TCA_FLOWER_KEY_CT_FLAGS_TRACKED | + TCA_FLOWER_KEY_CT_FLAGS_ESTABLISHED | + TCA_FLOWER_KEY_CT_FLAGS_NEW)) { + NL_SET_ERR_MSG_MOD(extack, + "only ct_state trk, est and new are supported for offload"); + return -EOPNOTSUPP; + } + + if (mask->ct_labels[1] || mask->ct_labels[2] || mask->ct_labels[3]) { + NL_SET_ERR_MSG_MOD(extack, + "only lower 32bits of ct_labels are supported for offload"); + return -EOPNOTSUPP; + } + + ct_state_on = ct_state & ct_state_mask; + ct_state_off = (ct_state & ct_state_mask) ^ ct_state_mask; + trk = ct_state_on & TCA_FLOWER_KEY_CT_FLAGS_TRACKED; + new = ct_state_on & TCA_FLOWER_KEY_CT_FLAGS_NEW; + est = ct_state_on & TCA_FLOWER_KEY_CT_FLAGS_ESTABLISHED; + untrk = ct_state_off & TCA_FLOWER_KEY_CT_FLAGS_TRACKED; + unest = ct_state_off & TCA_FLOWER_KEY_CT_FLAGS_ESTABLISHED; + + ctstate |= trk ? MLX5_CT_STATE_TRK_BIT : 0; + ctstate |= est ? MLX5_CT_STATE_ESTABLISHED_BIT : 0; + ctstate_mask |= (untrk || trk) ? MLX5_CT_STATE_TRK_BIT : 0; + ctstate_mask |= (unest || est) ? MLX5_CT_STATE_ESTABLISHED_BIT : 0; + + if (new) { + NL_SET_ERR_MSG_MOD(extack, + "matching on ct_state +new isn't supported"); + return -EOPNOTSUPP; + } + + if (mask->ct_zone) + mlx5e_tc_match_to_reg_match(spec, ZONE_TO_REG, + key->ct_zone, MLX5_CT_ZONE_MASK); + if (ctstate_mask) + mlx5e_tc_match_to_reg_match(spec, CTSTATE_TO_REG, + ctstate, ctstate_mask); + if (mask->ct_mark) + mlx5e_tc_match_to_reg_match(spec, MARK_TO_REG, + key->ct_mark, mask->ct_mark); + if (mask->ct_labels[0]) + mlx5e_tc_match_to_reg_match(spec, LABELS_TO_REG, + key->ct_labels[0], + mask->ct_labels[0]); + + return 0; +} + +int +mlx5_tc_ct_parse_action(struct mlx5e_priv *priv, + struct mlx5_esw_flow_attr *attr, + const struct flow_action_entry *act, + struct netlink_ext_ack *extack) +{ + struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); + + if (!ct_priv) { + NL_SET_ERR_MSG_MOD(extack, + "offload of ct action isn't available"); + return -EOPNOTSUPP; + } + + attr->ct_attr.zone = act->ct.zone; + attr->ct_attr.ct_action = act->ct.action; + attr->ct_attr.nf_ft = act->ct.flow_table; + + return 0; +} + +static struct mlx5_ct_ft * +mlx5_tc_ct_add_ft_cb(struct mlx5_tc_ct_priv *ct_priv, u16 zone, + struct nf_flowtable *nf_ft) +{ + struct mlx5_ct_ft *ft; + int err; + + ft = rhashtable_lookup_fast(&ct_priv->zone_ht, &zone, zone_params); + if (ft) { + refcount_inc(&ft->refcount); + return ft; + } + + ft = kzalloc(sizeof(*ft), GFP_KERNEL); + if (!ft) + return ERR_PTR(-ENOMEM); + + ft->zone = zone; + ft->nf_ft = nf_ft; + ft->ct_priv = ct_priv; + INIT_LIST_HEAD(&ft->ct_entries_list); + refcount_set(&ft->refcount, 1); + + err = rhashtable_init(&ft->ct_entries_ht, &cts_ht_params); + if (err) + goto err_init; + + err = rhashtable_insert_fast(&ct_priv->zone_ht, &ft->node, + zone_params); + if (err) + goto err_insert; + + err = nf_flow_table_offload_add_cb(ft->nf_ft, + mlx5_tc_ct_block_flow_offload, ft); + if (err) + goto err_add_cb; + + return ft; + +err_add_cb: + rhashtable_remove_fast(&ct_priv->zone_ht, &ft->node, zone_params); +err_insert: + rhashtable_destroy(&ft->ct_entries_ht); +err_init: + kfree(ft); + return ERR_PTR(err); +} + +static void +mlx5_tc_ct_flush_ft(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_ft *ft) +{ + struct mlx5_ct_entry *entry; + + list_for_each_entry(entry, &ft->ct_entries_list, list) + mlx5_tc_ct_entry_del_rules(ft->ct_priv, entry); +} + +static void +mlx5_tc_ct_del_ft_cb(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_ft *ft) +{ + if (!refcount_dec_and_test(&ft->refcount)) + return; + + nf_flow_table_offload_del_cb(ft->nf_ft, + mlx5_tc_ct_block_flow_offload, ft); + mlx5_tc_ct_flush_ft(ct_priv, ft); + rhashtable_remove_fast(&ct_priv->zone_ht, &ft->node, zone_params); + rhashtable_destroy(&ft->ct_entries_ht); + kfree(ft); +} + +/* We translate the tc filter with CT action to the following HW model: + * + * +-------------------+ +--------------------+ +--------------+ + * + pre_ct (tc chain) +----->+ CT (nat or no nat) +--->+ post_ct +-----> + * + original match + | + tuple + zone match + | + fte_id match + | + * +-------------------+ | +--------------------+ | +--------------+ | + * v v v + * set chain miss mapping set mark original + * set fte_id set label filter + * set zone set established actions + * set tunnel_id do nat (if needed) + * do decap + */ +static int +__mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_spec *orig_spec, + struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_handle **flow_rule) +{ + struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); + bool nat = attr->ct_attr.ct_action & TCA_CT_ACT_NAT; + struct mlx5e_tc_mod_hdr_acts pre_mod_acts = {}; + struct mlx5_flow_spec *post_ct_spec = NULL; + struct mlx5_eswitch *esw = ct_priv->esw; + struct mlx5_esw_flow_attr *pre_ct_attr; + struct mlx5_modify_hdr *mod_hdr; + struct mlx5_flow_handle *rule; + struct mlx5_ct_flow *ct_flow; + int chain_mapping = 0, err; + struct mlx5_ct_ft *ft; + u32 fte_id = 1; + + post_ct_spec = kzalloc(sizeof(*post_ct_spec), GFP_KERNEL); + ct_flow = kzalloc(sizeof(*ct_flow), GFP_KERNEL); + if (!post_ct_spec || !ct_flow) { + kfree(post_ct_spec); + kfree(ct_flow); + return -ENOMEM; + } + + /* Register for CT established events */ + ft = mlx5_tc_ct_add_ft_cb(ct_priv, attr->ct_attr.zone, + attr->ct_attr.nf_ft); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + ct_dbg("Failed to register to ft callback"); + goto err_ft; + } + ct_flow->ft = ft; + + err = idr_alloc_u32(&ct_priv->fte_ids, ct_flow, &fte_id, + MLX5_FTE_ID_MAX, GFP_KERNEL); + if (err) { + netdev_warn(priv->netdev, + "Failed to allocate fte id, err: %d\n", err); + goto err_idr; + } + ct_flow->fte_id = fte_id; + + /* Base esw attributes of both rules on original rule attribute */ + pre_ct_attr = &ct_flow->pre_ct_attr; + memcpy(pre_ct_attr, attr, sizeof(*attr)); + memcpy(&ct_flow->post_ct_attr, attr, sizeof(*attr)); + + /* Modify the original rule's action to fwd and modify, leave decap */ + pre_ct_attr->action = attr->action & MLX5_FLOW_CONTEXT_ACTION_DECAP; + pre_ct_attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | + MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + + /* Write chain miss tag for miss in ct table as we + * don't go though all prios of this chain as normal tc rules + * miss. + */ + err = mlx5_esw_chains_get_chain_mapping(esw, attr->chain, + &chain_mapping); + if (err) { + ct_dbg("Failed to get chain register mapping for chain"); + goto err_get_chain; + } + ct_flow->chain_mapping = chain_mapping; + + err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, + CHAIN_TO_REG, chain_mapping); + if (err) { + ct_dbg("Failed to set chain register mapping"); + goto err_mapping; + } + + err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, ZONE_TO_REG, + attr->ct_attr.zone & + MLX5_CT_ZONE_MASK); + if (err) { + ct_dbg("Failed to set zone register mapping"); + goto err_mapping; + } + + err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, + FTEID_TO_REG, fte_id); + if (err) { + ct_dbg("Failed to set fte_id register mapping"); + goto err_mapping; + } + + /* If original flow is decap, we do it before going into ct table + * so add a rewrite for the tunnel match_id. + */ + if ((pre_ct_attr->action & MLX5_FLOW_CONTEXT_ACTION_DECAP) && + attr->chain == 0) { + u32 tun_id = mlx5e_tc_get_flow_tun_id(flow); + + err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, + TUNNEL_TO_REG, + tun_id); + if (err) { + ct_dbg("Failed to set tunnel register mapping"); + goto err_mapping; + } + } + + mod_hdr = mlx5_modify_header_alloc(esw->dev, + MLX5_FLOW_NAMESPACE_FDB, + pre_mod_acts.num_actions, + pre_mod_acts.actions); + if (IS_ERR(mod_hdr)) { + err = PTR_ERR(mod_hdr); + ct_dbg("Failed to create pre ct mod hdr"); + goto err_mapping; + } + pre_ct_attr->modify_hdr = mod_hdr; + + /* Post ct rule matches on fte_id and executes original rule's + * tc rule action + */ + mlx5e_tc_match_to_reg_match(post_ct_spec, FTEID_TO_REG, + fte_id, MLX5_FTE_ID_MASK); + + /* Put post_ct rule on post_ct fdb */ + ct_flow->post_ct_attr.chain = 0; + ct_flow->post_ct_attr.prio = 0; + ct_flow->post_ct_attr.fdb = ct_priv->post_ct; + + ct_flow->post_ct_attr.inner_match_level = MLX5_MATCH_NONE; + ct_flow->post_ct_attr.outer_match_level = MLX5_MATCH_NONE; + ct_flow->post_ct_attr.action &= ~(MLX5_FLOW_CONTEXT_ACTION_DECAP); + rule = mlx5_eswitch_add_offloaded_rule(esw, post_ct_spec, + &ct_flow->post_ct_attr); + ct_flow->post_ct_rule = rule; + if (IS_ERR(ct_flow->post_ct_rule)) { + err = PTR_ERR(ct_flow->post_ct_rule); + ct_dbg("Failed to add post ct rule"); + goto err_insert_post_ct; + } + + /* Change original rule point to ct table */ + pre_ct_attr->dest_chain = 0; + pre_ct_attr->dest_ft = nat ? ct_priv->ct_nat : ct_priv->ct; + ct_flow->pre_ct_rule = mlx5_eswitch_add_offloaded_rule(esw, + orig_spec, + pre_ct_attr); + if (IS_ERR(ct_flow->pre_ct_rule)) { + err = PTR_ERR(ct_flow->pre_ct_rule); + ct_dbg("Failed to add pre ct rule"); + goto err_insert_orig; + } + + attr->ct_attr.ct_flow = ct_flow; + *flow_rule = ct_flow->post_ct_rule; + dealloc_mod_hdr_actions(&pre_mod_acts); + kfree(post_ct_spec); + + return 0; + +err_insert_orig: + mlx5_eswitch_del_offloaded_rule(ct_priv->esw, ct_flow->post_ct_rule, + &ct_flow->post_ct_attr); +err_insert_post_ct: + mlx5_modify_header_dealloc(priv->mdev, pre_ct_attr->modify_hdr); +err_mapping: + dealloc_mod_hdr_actions(&pre_mod_acts); + mlx5_esw_chains_put_chain_mapping(esw, ct_flow->chain_mapping); +err_get_chain: + idr_remove(&ct_priv->fte_ids, fte_id); +err_idr: + mlx5_tc_ct_del_ft_cb(ct_priv, ft); +err_ft: + kfree(post_ct_spec); + kfree(ct_flow); + netdev_warn(priv->netdev, "Failed to offload ct flow, err %d\n", err); + return err; +} + +static int +__mlx5_tc_ct_flow_offload_clear(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_spec *orig_spec, + struct mlx5_esw_flow_attr *attr, + struct mlx5e_tc_mod_hdr_acts *mod_acts, + struct mlx5_flow_handle **flow_rule) +{ + struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); + struct mlx5_eswitch *esw = ct_priv->esw; + struct mlx5_esw_flow_attr *pre_ct_attr; + struct mlx5_modify_hdr *mod_hdr; + struct mlx5_flow_handle *rule; + struct mlx5_ct_flow *ct_flow; + int err; + + ct_flow = kzalloc(sizeof(*ct_flow), GFP_KERNEL); + if (!ct_flow) + return -ENOMEM; + + /* Base esw attributes on original rule attribute */ + pre_ct_attr = &ct_flow->pre_ct_attr; + memcpy(pre_ct_attr, attr, sizeof(*attr)); + + err = mlx5_tc_ct_entry_set_registers(ct_priv, mod_acts, 0, 0, 0, 0); + if (err) { + ct_dbg("Failed to set register for ct clear"); + goto err_set_registers; + } + + mod_hdr = mlx5_modify_header_alloc(esw->dev, + MLX5_FLOW_NAMESPACE_FDB, + mod_acts->num_actions, + mod_acts->actions); + if (IS_ERR(mod_hdr)) { + err = PTR_ERR(mod_hdr); + ct_dbg("Failed to add create ct clear mod hdr"); + goto err_set_registers; + } + + dealloc_mod_hdr_actions(mod_acts); + pre_ct_attr->modify_hdr = mod_hdr; + pre_ct_attr->action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + + rule = mlx5_eswitch_add_offloaded_rule(esw, orig_spec, pre_ct_attr); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + ct_dbg("Failed to add ct clear rule"); + goto err_insert; + } + + attr->ct_attr.ct_flow = ct_flow; + ct_flow->pre_ct_rule = rule; + *flow_rule = rule; + + return 0; + +err_insert: + mlx5_modify_header_dealloc(priv->mdev, mod_hdr); +err_set_registers: + netdev_warn(priv->netdev, + "Failed to offload ct clear flow, err %d\n", err); + return err; +} + +struct mlx5_flow_handle * +mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_spec *spec, + struct mlx5_esw_flow_attr *attr, + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts) +{ + bool clear_action = attr->ct_attr.ct_action & TCA_CT_ACT_CLEAR; + struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); + struct mlx5_flow_handle *rule; + int err; + + if (!ct_priv) + return ERR_PTR(-EOPNOTSUPP); + + mutex_lock(&ct_priv->control_lock); + if (clear_action) + err = __mlx5_tc_ct_flow_offload_clear(priv, flow, spec, attr, + mod_hdr_acts, &rule); + else + err = __mlx5_tc_ct_flow_offload(priv, flow, spec, attr, + &rule); + mutex_unlock(&ct_priv->control_lock); + if (err) + return ERR_PTR(err); + + return rule; +} + +static void +__mlx5_tc_ct_delete_flow(struct mlx5_tc_ct_priv *ct_priv, + struct mlx5_ct_flow *ct_flow) +{ + struct mlx5_esw_flow_attr *pre_ct_attr = &ct_flow->pre_ct_attr; + struct mlx5_eswitch *esw = ct_priv->esw; + + mlx5_eswitch_del_offloaded_rule(esw, ct_flow->pre_ct_rule, + pre_ct_attr); + mlx5_modify_header_dealloc(esw->dev, pre_ct_attr->modify_hdr); + + if (ct_flow->post_ct_rule) { + mlx5_eswitch_del_offloaded_rule(esw, ct_flow->post_ct_rule, + &ct_flow->post_ct_attr); + mlx5_esw_chains_put_chain_mapping(esw, ct_flow->chain_mapping); + idr_remove(&ct_priv->fte_ids, ct_flow->fte_id); + mlx5_tc_ct_del_ft_cb(ct_priv, ct_flow->ft); + } + + kfree(ct_flow); +} + +void +mlx5_tc_ct_delete_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, + struct mlx5_esw_flow_attr *attr) +{ + struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); + struct mlx5_ct_flow *ct_flow = attr->ct_attr.ct_flow; + + /* We are called on error to clean up stuff from parsing + * but we don't have anything for now + */ + if (!ct_flow) + return; + + mutex_lock(&ct_priv->control_lock); + __mlx5_tc_ct_delete_flow(ct_priv, ct_flow); + mutex_unlock(&ct_priv->control_lock); +} + +static int +mlx5_tc_ct_init_check_support(struct mlx5_eswitch *esw, + const char **err_msg) +{ +#if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + /* cannot restore chain ID on HW miss */ + + *err_msg = "tc skb extension missing"; + return -EOPNOTSUPP; +#endif + + if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ignore_flow_level)) { + *err_msg = "firmware level support is missing"; + return -EOPNOTSUPP; + } + + if (!mlx5_eswitch_vlan_actions_supported(esw->dev, 1)) { + /* vlan workaround should be avoided for multi chain rules. + * This is just a sanity check as pop vlan action should + * be supported by any FW that supports ignore_flow_level + */ + + *err_msg = "firmware vlan actions support is missing"; + return -EOPNOTSUPP; + } + + if (!MLX5_CAP_ESW_FLOWTABLE(esw->dev, + fdb_modify_header_fwd_to_table)) { + /* CT always writes to registers which are mod header actions. + * Therefore, mod header and goto is required + */ + + *err_msg = "firmware fwd and modify support is missing"; + return -EOPNOTSUPP; + } + + if (!mlx5_eswitch_reg_c1_loopback_enabled(esw)) { + *err_msg = "register loopback isn't supported"; + return -EOPNOTSUPP; + } + + return 0; +} + +static void +mlx5_tc_ct_init_err(struct mlx5e_rep_priv *rpriv, const char *msg, int err) +{ + if (msg) + netdev_warn(rpriv->netdev, + "tc ct offload not supported, %s, err: %d\n", + msg, err); + else + netdev_warn(rpriv->netdev, + "tc ct offload not supported, err: %d\n", + err); +} + +int +mlx5_tc_ct_init(struct mlx5_rep_uplink_priv *uplink_priv) +{ + struct mlx5_tc_ct_priv *ct_priv; + struct mlx5e_rep_priv *rpriv; + struct mlx5_eswitch *esw; + struct mlx5e_priv *priv; + const char *msg; + int err; + + rpriv = container_of(uplink_priv, struct mlx5e_rep_priv, uplink_priv); + priv = netdev_priv(rpriv->netdev); + esw = priv->mdev->priv.eswitch; + + err = mlx5_tc_ct_init_check_support(esw, &msg); + if (err) { + mlx5_tc_ct_init_err(rpriv, msg, err); + goto err_support; + } + + ct_priv = kzalloc(sizeof(*ct_priv), GFP_KERNEL); + if (!ct_priv) { + mlx5_tc_ct_init_err(rpriv, NULL, -ENOMEM); + goto err_alloc; + } + + ct_priv->esw = esw; + ct_priv->netdev = rpriv->netdev; + ct_priv->ct = mlx5_esw_chains_create_global_table(esw); + if (IS_ERR(ct_priv->ct)) { + err = PTR_ERR(ct_priv->ct); + mlx5_tc_ct_init_err(rpriv, "failed to create ct table", err); + goto err_ct_tbl; + } + + ct_priv->ct_nat = mlx5_esw_chains_create_global_table(esw); + if (IS_ERR(ct_priv->ct_nat)) { + err = PTR_ERR(ct_priv->ct_nat); + mlx5_tc_ct_init_err(rpriv, "failed to create ct nat table", + err); + goto err_ct_nat_tbl; + } + + ct_priv->post_ct = mlx5_esw_chains_create_global_table(esw); + if (IS_ERR(ct_priv->post_ct)) { + err = PTR_ERR(ct_priv->post_ct); + mlx5_tc_ct_init_err(rpriv, "failed to create post ct table", + err); + goto err_post_ct_tbl; + } + + idr_init(&ct_priv->fte_ids); + idr_init(&ct_priv->tuple_ids); + mutex_init(&ct_priv->control_lock); + rhashtable_init(&ct_priv->zone_ht, &zone_params); + + /* Done, set ct_priv to know it initializted */ + uplink_priv->ct_priv = ct_priv; + + return 0; + +err_post_ct_tbl: + mlx5_esw_chains_destroy_global_table(esw, ct_priv->ct_nat); +err_ct_nat_tbl: + mlx5_esw_chains_destroy_global_table(esw, ct_priv->ct); +err_ct_tbl: + kfree(ct_priv); +err_alloc: +err_support: + + return 0; +} + +void +mlx5_tc_ct_clean(struct mlx5_rep_uplink_priv *uplink_priv) +{ + struct mlx5_tc_ct_priv *ct_priv = uplink_priv->ct_priv; + + if (!ct_priv) + return; + + mlx5_esw_chains_destroy_global_table(ct_priv->esw, ct_priv->post_ct); + mlx5_esw_chains_destroy_global_table(ct_priv->esw, ct_priv->ct_nat); + mlx5_esw_chains_destroy_global_table(ct_priv->esw, ct_priv->ct); + + rhashtable_destroy(&ct_priv->zone_ht); + mutex_destroy(&ct_priv->control_lock); + idr_destroy(&ct_priv->tuple_ids); + idr_destroy(&ct_priv->fte_ids); + kfree(ct_priv); + + uplink_priv->ct_priv = NULL; +} + +bool +mlx5e_tc_ct_restore_flow(struct mlx5_rep_uplink_priv *uplink_priv, + struct sk_buff *skb, u32 tupleid) +{ + struct mlx5_tc_ct_priv *ct_priv = uplink_priv->ct_priv; + struct mlx5_ct_zone_rule *zone_rule; + struct mlx5_ct_entry *entry; + + if (!ct_priv || !tupleid) + return true; + + zone_rule = idr_find(&ct_priv->tuple_ids, tupleid); + if (!zone_rule) + return false; + + entry = container_of(zone_rule, struct mlx5_ct_entry, + zone_rules[zone_rule->nat]); + tcf_ct_flow_table_restore_skb(skb, entry->restore_cookie); + + return true; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h new file mode 100644 index 000000000000..091d305b633e --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h @@ -0,0 +1,180 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2018 Mellanox Technologies. */ + +#ifndef __MLX5_EN_TC_CT_H__ +#define __MLX5_EN_TC_CT_H__ + +#include <net/pkt_cls.h> +#include <linux/mlx5/fs.h> +#include <net/tc_act/tc_ct.h> + +#include "en.h" + +struct mlx5_esw_flow_attr; +struct mlx5e_tc_mod_hdr_acts; +struct mlx5_rep_uplink_priv; +struct mlx5e_tc_flow; +struct mlx5e_priv; + +struct mlx5_ct_flow; + +struct nf_flowtable; + +struct mlx5_ct_attr { + u16 zone; + u16 ct_action; + struct mlx5_ct_flow *ct_flow; + struct nf_flowtable *nf_ft; +}; + +#define zone_to_reg_ct {\ + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_2,\ + .moffset = 0,\ + .mlen = 2,\ + .soffset = MLX5_BYTE_OFF(fte_match_param,\ + misc_parameters_2.metadata_reg_c_2) + 2,\ +} + +#define ctstate_to_reg_ct {\ + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_2,\ + .moffset = 2,\ + .mlen = 2,\ + .soffset = MLX5_BYTE_OFF(fte_match_param,\ + misc_parameters_2.metadata_reg_c_2),\ +} + +#define mark_to_reg_ct {\ + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_3,\ + .moffset = 0,\ + .mlen = 4,\ + .soffset = MLX5_BYTE_OFF(fte_match_param,\ + misc_parameters_2.metadata_reg_c_3),\ +} + +#define labels_to_reg_ct {\ + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_4,\ + .moffset = 0,\ + .mlen = 4,\ + .soffset = MLX5_BYTE_OFF(fte_match_param,\ + misc_parameters_2.metadata_reg_c_4),\ +} + +#define fteid_to_reg_ct {\ + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_5,\ + .moffset = 0,\ + .mlen = 4,\ + .soffset = MLX5_BYTE_OFF(fte_match_param,\ + misc_parameters_2.metadata_reg_c_5),\ +} + +#define tupleid_to_reg_ct {\ + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_1,\ + .moffset = 0,\ + .mlen = 3,\ + .soffset = MLX5_BYTE_OFF(fte_match_param,\ + misc_parameters_2.metadata_reg_c_1),\ +} + +#define TUPLE_ID_BITS (mlx5e_tc_attr_to_reg_mappings[TUPLEID_TO_REG].mlen * 8) +#define TUPLE_ID_MAX GENMASK(TUPLE_ID_BITS - 1, 0) + +#if IS_ENABLED(CONFIG_MLX5_TC_CT) + +int +mlx5_tc_ct_init(struct mlx5_rep_uplink_priv *uplink_priv); +void +mlx5_tc_ct_clean(struct mlx5_rep_uplink_priv *uplink_priv); + +int +mlx5_tc_ct_parse_match(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct flow_cls_offload *f, + struct netlink_ext_ack *extack); +int +mlx5_tc_ct_parse_action(struct mlx5e_priv *priv, + struct mlx5_esw_flow_attr *attr, + const struct flow_action_entry *act, + struct netlink_ext_ack *extack); + +struct mlx5_flow_handle * +mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_spec *spec, + struct mlx5_esw_flow_attr *attr, + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts); +void +mlx5_tc_ct_delete_flow(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_esw_flow_attr *attr); + +bool +mlx5e_tc_ct_restore_flow(struct mlx5_rep_uplink_priv *uplink_priv, + struct sk_buff *skb, u32 tupleid); + +#else /* CONFIG_MLX5_TC_CT */ + +static inline int +mlx5_tc_ct_init(struct mlx5_rep_uplink_priv *uplink_priv) +{ + return 0; +} + +static inline void +mlx5_tc_ct_clean(struct mlx5_rep_uplink_priv *uplink_priv) +{ +} + +static inline int +mlx5_tc_ct_parse_match(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct flow_cls_offload *f, + struct netlink_ext_ack *extack) +{ + if (!flow_rule_match_key(f->rule, FLOW_DISSECTOR_KEY_CT)) + return 0; + + NL_SET_ERR_MSG_MOD(extack, "mlx5 tc ct offload isn't enabled."); + netdev_warn(priv->netdev, "mlx5 tc ct offload isn't enabled.\n"); + return -EOPNOTSUPP; +} + +static inline int +mlx5_tc_ct_parse_action(struct mlx5e_priv *priv, + struct mlx5_esw_flow_attr *attr, + const struct flow_action_entry *act, + struct netlink_ext_ack *extack) +{ + NL_SET_ERR_MSG_MOD(extack, "mlx5 tc ct offload isn't enabled."); + netdev_warn(priv->netdev, "mlx5 tc ct offload isn't enabled.\n"); + return -EOPNOTSUPP; +} + +static inline struct mlx5_flow_handle * +mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_spec *spec, + struct mlx5_esw_flow_attr *attr, + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts) +{ + return ERR_PTR(-EOPNOTSUPP); +} + +static inline void +mlx5_tc_ct_delete_flow(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_esw_flow_attr *attr) +{ +} + +static inline bool +mlx5e_tc_ct_restore_flow(struct mlx5_rep_uplink_priv *uplink_priv, + struct sk_buff *skb, u32 tupleid) +{ + if (!tupleid) + return true; + + return false; +} + +#endif /* !IS_ENABLED(CONFIG_MLX5_TC_CT) */ +#endif /* __MLX5_EN_TC_CT_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c index af4ebd2951b5..b45c3f46570b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c @@ -66,6 +66,9 @@ static int get_route_and_out_devs(struct mlx5e_priv *priv, mlx5e_is_uplink_rep(netdev_priv(*out_dev)))) return -EOPNOTSUPP; + if (mlx5e_eswitch_uplink_rep(priv->netdev) && *out_dev != priv->netdev) + return -EOPNOTSUPP; + return 0; } @@ -469,10 +472,15 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev, struct mlx5e_priv *priv, struct mlx5_flow_spec *spec, struct flow_cls_offload *f, - void *headers_c, - void *headers_v, u8 *match_level) + u8 *match_level) { struct mlx5e_tc_tunnel *tunnel = mlx5e_get_tc_tun(filter_dev); + struct flow_rule *rule = flow_cls_offload_flow_rule(f); + void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers); + void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers); + struct netlink_ext_ack *extack = f->common.extack; int err = 0; if (!tunnel) { @@ -499,6 +507,109 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev, goto out; } + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_CONTROL)) { + struct flow_match_control match; + u16 addr_type; + + flow_rule_match_enc_control(rule, &match); + addr_type = match.key->addr_type; + + /* For tunnel addr_type used same key id`s as for non-tunnel */ + if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) { + struct flow_match_ipv4_addrs match; + + flow_rule_match_enc_ipv4_addrs(rule, &match); + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + src_ipv4_src_ipv6.ipv4_layout.ipv4, + ntohl(match.mask->src)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + src_ipv4_src_ipv6.ipv4_layout.ipv4, + ntohl(match.key->src)); + + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4, + ntohl(match.mask->dst)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4, + ntohl(match.key->dst)); + + MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, + ethertype); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, + ETH_P_IP); + } else if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) { + struct flow_match_ipv6_addrs match; + + flow_rule_match_enc_ipv6_addrs(rule, &match); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, + src_ipv4_src_ipv6.ipv6_layout.ipv6), + &match.mask->src, MLX5_FLD_SZ_BYTES(ipv6_layout, + ipv6)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + src_ipv4_src_ipv6.ipv6_layout.ipv6), + &match.key->src, MLX5_FLD_SZ_BYTES(ipv6_layout, + ipv6)); + + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, + dst_ipv4_dst_ipv6.ipv6_layout.ipv6), + &match.mask->dst, MLX5_FLD_SZ_BYTES(ipv6_layout, + ipv6)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + dst_ipv4_dst_ipv6.ipv6_layout.ipv6), + &match.key->dst, MLX5_FLD_SZ_BYTES(ipv6_layout, + ipv6)); + + MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, + ethertype); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, + ETH_P_IPV6); + } + } + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IP)) { + struct flow_match_ip match; + + flow_rule_match_enc_ip(rule, &match); + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_ecn, + match.mask->tos & 0x3); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_ecn, + match.key->tos & 0x3); + + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_dscp, + match.mask->tos >> 2); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_dscp, + match.key->tos >> 2); + + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ttl_hoplimit, + match.mask->ttl); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ttl_hoplimit, + match.key->ttl); + + if (match.mask->ttl && + !MLX5_CAP_ESW_FLOWTABLE_FDB + (priv->mdev, + ft_field_support.outer_ipv4_ttl)) { + NL_SET_ERR_MSG_MOD(extack, + "Matching on TTL is not supported"); + err = -EOPNOTSUPP; + goto out; + } + } + + /* Enforce DMAC when offloading incoming tunneled flows. + * Flow counters require a match on the DMAC. + */ + MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, dmac_47_16); + MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, dmac_15_0); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, + dmac_47_16), priv->netdev->dev_addr); + + /* let software handle IP fragments */ + MLX5_SET(fte_match_set_lyr_2_4, headers_c, frag, 1); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, frag, 0); + + return 0; + out: return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h index 6f9a78c85ffd..1630f0ec3ad7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h @@ -76,8 +76,7 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev, struct mlx5e_priv *priv, struct mlx5_flow_spec *spec, struct flow_cls_offload *f, - void *headers_c, - void *headers_v, u8 *match_level); + u8 *match_level); int mlx5e_tc_tun_parse_udp_ports(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c index cf58c9637904..29626c6c9c25 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c @@ -433,7 +433,6 @@ void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv) if (!ipsec) return; - drain_workqueue(ipsec->wq); destroy_workqueue(ipsec->wq); ida_destroy(&ipsec->halloc); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c index 2c75b2752f58..014639ea06e3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c @@ -175,28 +175,20 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, struct mlx5e_tir *tir = priv->indir_tir; struct mlx5_flow_destination dest = {}; MLX5_DECLARE_FLOW_ACT(flow_act); - struct mlx5_flow_spec *spec; enum mlx5e_traffic_types tt; int err = 0; - spec = kvzalloc(sizeof(*spec), GFP_KERNEL); - if (!spec) { - err = -ENOMEM; - goto out; - } - dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; tt = arfs_get_tt(type); if (tt == -EINVAL) { netdev_err(priv->netdev, "%s: bad arfs_type: %d\n", __func__, type); - err = -EINVAL; - goto out; + return -EINVAL; } dest.tir_num = tir[tt].tirn; - arfs_t->default_rule = mlx5_add_flow_rules(arfs_t->ft.t, spec, + arfs_t->default_rule = mlx5_add_flow_rules(arfs_t->ft.t, NULL, &flow_act, &dest, 1); if (IS_ERR(arfs_t->default_rule)) { @@ -205,8 +197,7 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, netdev_err(priv->netdev, "%s: add rule failed, arfs type=%d\n", __func__, type); } -out: - kvfree(spec); + return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 01f2918063af..47874d34156b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -1098,49 +1098,59 @@ void mlx5e_dcbnl_delete_app(struct mlx5e_priv *priv) mlx5e_dcbnl_dscp_app(priv, DELETE); } -static void mlx5e_trust_update_tx_min_inline_mode(struct mlx5e_priv *priv, - struct mlx5e_params *params) +static void mlx5e_params_calc_trust_tx_min_inline_mode(struct mlx5_core_dev *mdev, + struct mlx5e_params *params, + u8 trust_state) { - mlx5_query_min_inline(priv->mdev, ¶ms->tx_min_inline_mode); - if (priv->dcbx_dp.trust_state == MLX5_QPTS_TRUST_DSCP && + mlx5_query_min_inline(mdev, ¶ms->tx_min_inline_mode); + if (trust_state == MLX5_QPTS_TRUST_DSCP && params->tx_min_inline_mode == MLX5_INLINE_MODE_L2) params->tx_min_inline_mode = MLX5_INLINE_MODE_IP; } -static void mlx5e_trust_update_sq_inline_mode(struct mlx5e_priv *priv) +static int mlx5e_update_trust_state_hw(struct mlx5e_priv *priv, void *context) +{ + u8 *trust_state = context; + int err; + + err = mlx5_set_trust_state(priv->mdev, *trust_state); + if (err) + return err; + priv->dcbx_dp.trust_state = *trust_state; + + return 0; +} + +static int mlx5e_set_trust_state(struct mlx5e_priv *priv, u8 trust_state) { struct mlx5e_channels new_channels = {}; + bool reset_channels = true; + int err = 0; mutex_lock(&priv->state_lock); new_channels.params = priv->channels.params; - mlx5e_trust_update_tx_min_inline_mode(priv, &new_channels.params); + mlx5e_params_calc_trust_tx_min_inline_mode(priv->mdev, &new_channels.params, + trust_state); if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { priv->channels.params = new_channels.params; - goto out; + reset_channels = false; } /* Skip if tx_min_inline is the same */ if (new_channels.params.tx_min_inline_mode == priv->channels.params.tx_min_inline_mode) - goto out; + reset_channels = false; - mlx5e_safe_switch_channels(priv, &new_channels, NULL); + if (reset_channels) + err = mlx5e_safe_switch_channels(priv, &new_channels, + mlx5e_update_trust_state_hw, + &trust_state); + else + err = mlx5e_update_trust_state_hw(priv, &trust_state); -out: mutex_unlock(&priv->state_lock); -} - -static int mlx5e_set_trust_state(struct mlx5e_priv *priv, u8 trust_state) -{ - int err; - - err = mlx5_set_trust_state(priv->mdev, trust_state); - if (err) - return err; - priv->dcbx_dp.trust_state = trust_state; - mlx5e_trust_update_sq_inline_mode(priv); return err; } @@ -1171,7 +1181,8 @@ static int mlx5e_trust_initialize(struct mlx5e_priv *priv) if (err) return err; - mlx5e_trust_update_tx_min_inline_mode(priv, &priv->channels.params); + mlx5e_params_calc_trust_tx_min_inline_mode(priv->mdev, &priv->channels.params, + priv->dcbx_dp.trust_state); err = mlx5_query_dscp2prio(priv->mdev, priv->dcbx_dp.dscp2prio); if (err) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index d674cb679895..6d703ddee4e2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -357,7 +357,7 @@ int mlx5e_ethtool_set_ringparam(struct mlx5e_priv *priv, goto unlock; } - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); unlock: mutex_unlock(&priv->state_lock); @@ -432,9 +432,7 @@ int mlx5e_ethtool_set_channels(struct mlx5e_priv *priv, if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { *cur_params = new_channels.params; - if (!netif_is_rxfh_configured(priv->netdev)) - mlx5e_build_default_indir_rqt(priv->rss_params.indirection_rqt, - MLX5E_INDIR_RQT_SIZE, count); + mlx5e_num_channels_changed(priv); goto out; } @@ -442,12 +440,9 @@ int mlx5e_ethtool_set_channels(struct mlx5e_priv *priv, if (arfs_enabled) mlx5e_arfs_disable(priv); - if (!netif_is_rxfh_configured(priv->netdev)) - mlx5e_build_default_indir_rqt(priv->rss_params.indirection_rqt, - MLX5E_INDIR_RQT_SIZE, count); - /* Switch to new channels, set new parameters and close old ones */ - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_safe_switch_channels(priv, &new_channels, + mlx5e_num_channels_changed_ctx, NULL); if (arfs_enabled) { int err2 = mlx5e_arfs_enable(priv); @@ -580,7 +575,7 @@ int mlx5e_ethtool_set_coalesce(struct mlx5e_priv *priv, goto out; } - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); out: mutex_unlock(&priv->state_lock); @@ -633,6 +628,8 @@ static const u32 pplm_fec_2_ethtool[] = { [MLX5E_FEC_NOFEC] = ETHTOOL_FEC_OFF, [MLX5E_FEC_FIRECODE] = ETHTOOL_FEC_BASER, [MLX5E_FEC_RS_528_514] = ETHTOOL_FEC_RS, + [MLX5E_FEC_RS_544_514] = ETHTOOL_FEC_RS, + [MLX5E_FEC_LLRS_272_257_1] = ETHTOOL_FEC_LLRS, }; static u32 pplm2ethtool_fec(u_long fec_mode, unsigned long size) @@ -650,45 +647,48 @@ static u32 pplm2ethtool_fec(u_long fec_mode, unsigned long size) return 0; } -/* we use ETHTOOL_FEC_* offset and apply it to ETHTOOL_LINK_MODE_FEC_*_BIT */ -static u32 ethtool_fec2ethtool_caps(u_long ethtool_fec_code) -{ - u32 offset; - - offset = find_first_bit(ðtool_fec_code, sizeof(u32)); - offset -= ETHTOOL_FEC_OFF_BIT; - offset += ETHTOOL_LINK_MODE_FEC_NONE_BIT; +#define MLX5E_ADVERTISE_SUPPORTED_FEC(mlx5_fec, ethtool_fec) \ + do { \ + if (mlx5e_fec_in_caps(dev, 1 << (mlx5_fec))) \ + __set_bit(ethtool_fec, \ + link_ksettings->link_modes.supported);\ + } while (0) - return offset; -} +static const u32 pplm_fec_2_ethtool_linkmodes[] = { + [MLX5E_FEC_NOFEC] = ETHTOOL_LINK_MODE_FEC_NONE_BIT, + [MLX5E_FEC_FIRECODE] = ETHTOOL_LINK_MODE_FEC_BASER_BIT, + [MLX5E_FEC_RS_528_514] = ETHTOOL_LINK_MODE_FEC_RS_BIT, + [MLX5E_FEC_RS_544_514] = ETHTOOL_LINK_MODE_FEC_RS_BIT, + [MLX5E_FEC_LLRS_272_257_1] = ETHTOOL_LINK_MODE_FEC_LLRS_BIT, +}; static int get_fec_supported_advertised(struct mlx5_core_dev *dev, struct ethtool_link_ksettings *link_ksettings) { - u_long fec_caps = 0; - u32 active_fec = 0; - u32 offset; + u_long active_fec = 0; u32 bitn; int err; - err = mlx5e_get_fec_caps(dev, (u8 *)&fec_caps); + err = mlx5e_get_fec_mode(dev, (u32 *)&active_fec, NULL); if (err) return (err == -EOPNOTSUPP) ? 0 : err; - err = mlx5e_get_fec_mode(dev, &active_fec, NULL); - if (err) - return err; - - for_each_set_bit(bitn, &fec_caps, ARRAY_SIZE(pplm_fec_2_ethtool)) { - u_long ethtool_bitmask = pplm_fec_2_ethtool[bitn]; - - offset = ethtool_fec2ethtool_caps(ethtool_bitmask); - __set_bit(offset, link_ksettings->link_modes.supported); - } - - active_fec = pplm2ethtool_fec(active_fec, sizeof(u32) * BITS_PER_BYTE); - offset = ethtool_fec2ethtool_caps(active_fec); - __set_bit(offset, link_ksettings->link_modes.advertising); + MLX5E_ADVERTISE_SUPPORTED_FEC(MLX5E_FEC_NOFEC, + ETHTOOL_LINK_MODE_FEC_NONE_BIT); + MLX5E_ADVERTISE_SUPPORTED_FEC(MLX5E_FEC_FIRECODE, + ETHTOOL_LINK_MODE_FEC_BASER_BIT); + MLX5E_ADVERTISE_SUPPORTED_FEC(MLX5E_FEC_RS_528_514, + ETHTOOL_LINK_MODE_FEC_RS_BIT); + MLX5E_ADVERTISE_SUPPORTED_FEC(MLX5E_FEC_LLRS_272_257_1, + ETHTOOL_LINK_MODE_FEC_LLRS_BIT); + + /* active fec is a bit set, find out which bit is set and + * advertise the corresponding ethtool bit + */ + bitn = find_first_bit(&active_fec, sizeof(u32) * BITS_PER_BYTE); + if (bitn < ARRAY_SIZE(pplm_fec_2_ethtool_linkmodes)) + __set_bit(pplm_fec_2_ethtool_linkmodes[bitn], + link_ksettings->link_modes.advertising); return 0; } @@ -773,6 +773,7 @@ static void ptys2ethtool_supported_advertised_port(struct ethtool_link_ksettings static void get_speed_duplex(struct net_device *netdev, u32 eth_proto_oper, bool force_legacy, + u16 data_rate_oper, struct ethtool_link_ksettings *link_ksettings) { struct mlx5e_priv *priv = netdev_priv(netdev); @@ -784,7 +785,10 @@ static void get_speed_duplex(struct net_device *netdev, speed = mlx5e_port_ptys2speed(priv->mdev, eth_proto_oper, force_legacy); if (!speed) { - speed = SPEED_UNKNOWN; + if (data_rate_oper) + speed = 100 * data_rate_oper; + else + speed = SPEED_UNKNOWN; goto out; } @@ -873,17 +877,18 @@ int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv, struct ethtool_link_ksettings *link_ksettings) { struct mlx5_core_dev *mdev = priv->mdev; - u32 out[MLX5_ST_SZ_DW(ptys_reg)] = {0}; - u32 rx_pause = 0; - u32 tx_pause = 0; - u32 eth_proto_cap; + u32 out[MLX5_ST_SZ_DW(ptys_reg)] = {}; u32 eth_proto_admin; - u32 eth_proto_lp; - u32 eth_proto_oper; u8 an_disable_admin; - u8 an_status; + u16 data_rate_oper; + u32 eth_proto_oper; + u32 eth_proto_cap; u8 connector_type; + u32 rx_pause = 0; + u32 tx_pause = 0; + u32 eth_proto_lp; bool admin_ext; + u8 an_status; bool ext; int err; @@ -917,6 +922,7 @@ int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv, an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin); an_status = MLX5_GET(ptys_reg, out, an_status); connector_type = MLX5_GET(ptys_reg, out, connector_type); + data_rate_oper = MLX5_GET(ptys_reg, out, data_rate_oper); mlx5_query_port_pause(mdev, &rx_pause, &tx_pause); @@ -927,7 +933,7 @@ int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv, get_advertising(eth_proto_admin, tx_pause, rx_pause, link_ksettings, admin_ext); get_speed_duplex(priv->netdev, eth_proto_oper, !admin_ext, - link_ksettings); + data_rate_oper, link_ksettings); eth_proto_oper = eth_proto_oper ? eth_proto_oper : eth_proto_cap; @@ -1126,8 +1132,8 @@ static u32 mlx5e_get_rxfh_indir_size(struct net_device *netdev) return mlx5e_ethtool_get_rxfh_indir_size(priv); } -static int mlx5e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, - u8 *hfunc) +int mlx5e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, + u8 *hfunc) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5e_rss_params *rss = &priv->rss_params; @@ -1146,8 +1152,8 @@ static int mlx5e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, return 0; } -static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, - const u8 *key, const u8 hfunc) +int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, + const u8 *key, const u8 hfunc) { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5e_rss_params *rss = &priv->rss_params; @@ -1511,7 +1517,7 @@ static int mlx5e_get_fecparam(struct net_device *netdev, { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; - u8 fec_configured = 0; + u16 fec_configured = 0; u32 fec_active = 0; int err; @@ -1527,7 +1533,7 @@ static int mlx5e_get_fecparam(struct net_device *netdev, return -EOPNOTSUPP; fecparam->fec = pplm2ethtool_fec((u_long)fec_configured, - sizeof(u8) * BITS_PER_BYTE); + sizeof(u16) * BITS_PER_BYTE); return 0; } @@ -1537,10 +1543,14 @@ static int mlx5e_set_fecparam(struct net_device *netdev, { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; - u8 fec_policy = 0; + u16 fec_policy = 0; int mode; int err; + if (bitmap_weight((unsigned long *)&fecparam->fec, + ETHTOOL_FEC_LLRS_BIT + 1) > 1) + return -EOPNOTSUPP; + for (mode = 0; mode < ARRAY_SIZE(pplm_fec_2_ethtool); mode++) { if (!(pplm_fec_2_ethtool[mode] & fecparam->fec)) continue; @@ -1739,7 +1749,7 @@ static int set_pflag_cqe_based_moder(struct net_device *netdev, bool enable, return 0; } - return mlx5e_safe_switch_channels(priv, &new_channels, NULL); + return mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); } static int set_pflag_tx_cqe_based_moder(struct net_device *netdev, bool enable) @@ -1772,7 +1782,7 @@ int mlx5e_modify_rx_cqe_compression_locked(struct mlx5e_priv *priv, bool new_val return 0; } - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); if (err) return err; @@ -1829,7 +1839,7 @@ static int set_pflag_rx_striding_rq(struct net_device *netdev, bool enable) return 0; } - return mlx5e_safe_switch_channels(priv, &new_channels, NULL); + return mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); } static int set_pflag_rx_no_csum_complete(struct net_device *netdev, bool enable) @@ -1873,7 +1883,7 @@ static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) return 0; } - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); return err; } @@ -1938,7 +1948,8 @@ static u32 mlx5e_get_priv_flags(struct net_device *netdev) return priv->channels.params.pflags; } -static int mlx5e_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, u32 *rule_locs) +int mlx5e_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, + u32 *rule_locs) { struct mlx5e_priv *priv = netdev_priv(dev); @@ -1955,12 +1966,15 @@ static int mlx5e_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, u return mlx5e_ethtool_get_rxnfc(dev, info, rule_locs); } -static int mlx5e_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) +int mlx5e_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) { return mlx5e_ethtool_set_rxnfc(dev, cmd); } const struct ethtool_ops mlx5e_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE, .get_drvinfo = mlx5e_get_drvinfo, .get_link = ethtool_op_get_link, .get_strings = mlx5e_get_strings, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 4ef3dc79f73c..dd7f338425eb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -63,6 +63,7 @@ #include "en/xsk/rx.h" #include "en/xsk/tx.h" #include "en/hv_vhca_stats.h" +#include "en/devlink.h" #include "lib/mlx5.h" @@ -1811,29 +1812,6 @@ static int mlx5e_set_tx_maxrate(struct net_device *dev, int index, u32 rate) return err; } -static int mlx5e_alloc_xps_cpumask(struct mlx5e_channel *c, - struct mlx5e_params *params) -{ - int num_comp_vectors = mlx5_comp_vectors_count(c->mdev); - int irq; - - if (!zalloc_cpumask_var(&c->xps_cpumask, GFP_KERNEL)) - return -ENOMEM; - - for (irq = c->ix; irq < num_comp_vectors; irq += params->num_channels) { - int cpu = cpumask_first(mlx5_comp_irq_get_affinity_mask(c->mdev, irq)); - - cpumask_set_cpu(cpu, c->xps_cpumask); - } - - return 0; -} - -static void mlx5e_free_xps_cpumask(struct mlx5e_channel *c) -{ - free_cpumask_var(c->xps_cpumask); -} - static int mlx5e_open_queues(struct mlx5e_channel *c, struct mlx5e_params *params, struct mlx5e_channel_param *cparam) @@ -1984,10 +1962,6 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->irq_desc = irq_to_desc(irq); c->lag_port = mlx5e_enumerate_lag_port(priv->mdev, ix); - err = mlx5e_alloc_xps_cpumask(c, params); - if (err) - goto err_free_channel; - netif_napi_add(netdev, &c->napi, mlx5e_napi_poll, 64); err = mlx5e_open_queues(c, params, cparam); @@ -2010,9 +1984,7 @@ err_close_queues: err_napi_del: netif_napi_del(&c->napi); - mlx5e_free_xps_cpumask(c); -err_free_channel: kvfree(c); return err; @@ -2026,7 +1998,6 @@ static void mlx5e_activate_channel(struct mlx5e_channel *c) mlx5e_activate_txqsq(&c->sq[tc]); mlx5e_activate_icosq(&c->icosq); mlx5e_activate_rq(&c->rq); - netif_set_xps_queue(c->netdev, c->xps_cpumask, c->ix); if (test_bit(MLX5E_CHANNEL_STATE_XSK, c->state)) mlx5e_activate_xsk(c); @@ -2051,7 +2022,6 @@ static void mlx5e_close_channel(struct mlx5e_channel *c) mlx5e_close_xsk(c); mlx5e_close_queues(c); netif_napi_del(&c->napi); - mlx5e_free_xps_cpumask(c); kvfree(c); } @@ -2801,6 +2771,8 @@ free_in: return err; } +static MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX(mlx5e_modify_tirs_lro); + static int mlx5e_set_mtu(struct mlx5_core_dev *mdev, struct mlx5e_params *params, u16 mtu) { @@ -2850,6 +2822,8 @@ int mlx5e_set_dev_port_mtu(struct mlx5e_priv *priv) return 0; } +MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX(mlx5e_set_dev_port_mtu); + void mlx5e_set_netdev_mtu_boundaries(struct mlx5e_priv *priv) { struct mlx5e_params *params = &priv->channels.params; @@ -2886,6 +2860,54 @@ static void mlx5e_netdev_set_tcs(struct net_device *netdev) netdev_set_tc_queue(netdev, tc, nch, 0); } +static void mlx5e_update_netdev_queues(struct mlx5e_priv *priv, u16 count) +{ + int num_txqs = count * priv->channels.params.num_tc; + int num_rxqs = count * priv->profile->rq_groups; + struct net_device *netdev = priv->netdev; + + mlx5e_netdev_set_tcs(netdev); + netif_set_real_num_tx_queues(netdev, num_txqs); + netif_set_real_num_rx_queues(netdev, num_rxqs); +} + +static void mlx5e_set_default_xps_cpumasks(struct mlx5e_priv *priv, + struct mlx5e_params *params) +{ + struct mlx5_core_dev *mdev = priv->mdev; + int num_comp_vectors, ix, irq; + + num_comp_vectors = mlx5_comp_vectors_count(mdev); + + for (ix = 0; ix < params->num_channels; ix++) { + cpumask_clear(priv->scratchpad.cpumask); + + for (irq = ix; irq < num_comp_vectors; irq += params->num_channels) { + int cpu = cpumask_first(mlx5_comp_irq_get_affinity_mask(mdev, irq)); + + cpumask_set_cpu(cpu, priv->scratchpad.cpumask); + } + + netif_set_xps_queue(priv->netdev, priv->scratchpad.cpumask, ix); + } +} + +int mlx5e_num_channels_changed(struct mlx5e_priv *priv) +{ + u16 count = priv->channels.params.num_channels; + + mlx5e_update_netdev_queues(priv, count); + mlx5e_set_default_xps_cpumasks(priv, &priv->channels.params); + + if (!netif_is_rxfh_configured(priv->netdev)) + mlx5e_build_default_indir_rqt(priv->rss_params.indirection_rqt, + MLX5E_INDIR_RQT_SIZE, count); + + return 0; +} + +MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX(mlx5e_num_channels_changed); + static void mlx5e_build_txq_maps(struct mlx5e_priv *priv) { int i, ch; @@ -2907,14 +2929,6 @@ static void mlx5e_build_txq_maps(struct mlx5e_priv *priv) void mlx5e_activate_priv_channels(struct mlx5e_priv *priv) { - int num_txqs = priv->channels.num * priv->channels.params.num_tc; - int num_rxqs = priv->channels.num * priv->profile->rq_groups; - struct net_device *netdev = priv->netdev; - - mlx5e_netdev_set_tcs(netdev); - netif_set_real_num_tx_queues(netdev, num_txqs); - netif_set_real_num_rx_queues(netdev, num_rxqs); - mlx5e_build_txq_maps(priv); mlx5e_activate_channels(&priv->channels); mlx5e_xdp_tx_enable(priv); @@ -2947,42 +2961,52 @@ void mlx5e_deactivate_priv_channels(struct mlx5e_priv *priv) mlx5e_deactivate_channels(&priv->channels); } -static void mlx5e_switch_priv_channels(struct mlx5e_priv *priv, - struct mlx5e_channels *new_chs, - mlx5e_fp_hw_modify hw_modify) +static int mlx5e_switch_priv_channels(struct mlx5e_priv *priv, + struct mlx5e_channels *new_chs, + mlx5e_fp_preactivate preactivate, + void *context) { struct net_device *netdev = priv->netdev; - int new_num_txqs; + struct mlx5e_channels old_chs; int carrier_ok; - - new_num_txqs = new_chs->num * new_chs->params.num_tc; + int err = 0; carrier_ok = netif_carrier_ok(netdev); netif_carrier_off(netdev); - if (new_num_txqs < netdev->real_num_tx_queues) - netif_set_real_num_tx_queues(netdev, new_num_txqs); - mlx5e_deactivate_priv_channels(priv); - mlx5e_close_channels(&priv->channels); + old_chs = priv->channels; priv->channels = *new_chs; - /* New channels are ready to roll, modify HW settings if needed */ - if (hw_modify) - hw_modify(priv); + /* New channels are ready to roll, call the preactivate hook if needed + * to modify HW settings or update kernel parameters. + */ + if (preactivate) { + err = preactivate(priv, context); + if (err) { + priv->channels = old_chs; + goto out; + } + } + mlx5e_close_channels(&old_chs); priv->profile->update_rx(priv); + +out: mlx5e_activate_priv_channels(priv); /* return carrier back if needed */ if (carrier_ok) netif_carrier_on(netdev); + + return err; } int mlx5e_safe_switch_channels(struct mlx5e_priv *priv, struct mlx5e_channels *new_chs, - mlx5e_fp_hw_modify hw_modify) + mlx5e_fp_preactivate preactivate, + void *context) { int err; @@ -2990,8 +3014,16 @@ int mlx5e_safe_switch_channels(struct mlx5e_priv *priv, if (err) return err; - mlx5e_switch_priv_channels(priv, new_chs, hw_modify); + err = mlx5e_switch_priv_channels(priv, new_chs, preactivate, context); + if (err) + goto err_close; + return 0; + +err_close: + mlx5e_close_channels(new_chs); + + return err; } int mlx5e_safe_reopen_channels(struct mlx5e_priv *priv) @@ -2999,7 +3031,7 @@ int mlx5e_safe_reopen_channels(struct mlx5e_priv *priv) struct mlx5e_channels new_channels = {}; new_channels.params = priv->channels.params; - return mlx5e_safe_switch_channels(priv, &new_channels, NULL); + return mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); } void mlx5e_timestamp_init(struct mlx5e_priv *priv) @@ -3448,7 +3480,8 @@ static int mlx5e_setup_tc_mqprio(struct mlx5e_priv *priv, goto out; } - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_safe_switch_channels(priv, &new_channels, + mlx5e_num_channels_changed_ctx, NULL); if (err) goto out; @@ -3661,7 +3694,8 @@ static int set_feature_lro(struct net_device *netdev, bool enable) goto out; } - err = mlx5e_safe_switch_channels(priv, &new_channels, mlx5e_modify_tirs_lro); + err = mlx5e_safe_switch_channels(priv, &new_channels, + mlx5e_modify_tirs_lro_ctx, NULL); out: mutex_unlock(&priv->state_lock); return err; @@ -3880,7 +3914,7 @@ static bool mlx5e_xsk_validate_mtu(struct net_device *netdev, } int mlx5e_change_mtu(struct net_device *netdev, int new_mtu, - change_hw_mtu_cb set_mtu_cb) + mlx5e_fp_preactivate preactivate) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5e_channels new_channels = {}; @@ -3929,13 +3963,13 @@ int mlx5e_change_mtu(struct net_device *netdev, int new_mtu, if (!reset) { params->sw_mtu = new_mtu; - if (set_mtu_cb) - set_mtu_cb(priv); + if (preactivate) + preactivate(priv, NULL); netdev->mtu = params->sw_mtu; goto out; } - err = mlx5e_safe_switch_channels(priv, &new_channels, set_mtu_cb); + err = mlx5e_safe_switch_channels(priv, &new_channels, preactivate, NULL); if (err) goto out; @@ -3948,7 +3982,7 @@ out: static int mlx5e_change_nic_mtu(struct net_device *netdev, int new_mtu) { - return mlx5e_change_mtu(netdev, new_mtu, mlx5e_set_dev_port_mtu); + return mlx5e_change_mtu(netdev, new_mtu, mlx5e_set_dev_port_mtu_ctx); } int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) @@ -4409,7 +4443,7 @@ static int mlx5e_xdp_set(struct net_device *netdev, struct bpf_prog *prog) mlx5e_set_rq_type(priv->mdev, &new_channels.params); old_prog = priv->channels.params.xdp_prog; - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); if (err) goto unlock; } else { @@ -4589,6 +4623,7 @@ const struct net_device_ops mlx5e_netdev_ops = { .ndo_set_vf_link_state = mlx5e_set_vf_link_state, .ndo_get_vf_stats = mlx5e_get_vf_stats, #endif + .ndo_get_devlink_port = mlx5e_get_devlink_port, }; static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev) @@ -4787,9 +4822,8 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, mlx5e_build_rq_params(mdev, params); /* HW LRO */ - - /* TODO: && MLX5_CAP_ETH(mdev, lro_cap) */ - if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { + if (MLX5_CAP_ETH(mdev, lro_cap) && + params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { /* No XSK params: checking the availability of striding RQ in general. */ if (!mlx5e_rx_mpwqe_is_linear_skb(mdev, params, NULL)) params->lro_en = !slow_pci_heuristic(mdev); @@ -5230,6 +5264,9 @@ int mlx5e_netdev_init(struct net_device *netdev, priv->max_nch = netdev->num_rx_queues / max_t(u8, profile->rq_groups, 1); priv->max_opened_tc = 1; + if (!alloc_cpumask_var(&priv->scratchpad.cpumask, GFP_KERNEL)) + return -ENOMEM; + mutex_init(&priv->state_lock); INIT_WORK(&priv->update_carrier_work, mlx5e_update_carrier_work); INIT_WORK(&priv->set_rx_mode_work, mlx5e_set_rx_mode_work); @@ -5238,7 +5275,7 @@ int mlx5e_netdev_init(struct net_device *netdev, priv->wq = create_singlethread_workqueue("mlx5e"); if (!priv->wq) - return -ENOMEM; + goto err_free_cpumask; /* netdev init */ netif_carrier_off(netdev); @@ -5248,11 +5285,17 @@ int mlx5e_netdev_init(struct net_device *netdev, #endif return 0; + +err_free_cpumask: + free_cpumask_var(priv->scratchpad.cpumask); + + return -ENOMEM; } void mlx5e_netdev_cleanup(struct net_device *netdev, struct mlx5e_priv *priv) { destroy_workqueue(priv->wq); + free_cpumask_var(priv->scratchpad.cpumask); } struct net_device *mlx5e_create_netdev(struct mlx5_core_dev *mdev, @@ -5287,6 +5330,7 @@ err_free_netdev: int mlx5e_attach_netdev(struct mlx5e_priv *priv) { + const bool take_rtnl = priv->netdev->reg_state == NETREG_REGISTERED; const struct mlx5e_profile *profile; int max_nch; int err; @@ -5298,10 +5342,25 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv) max_nch = mlx5e_get_max_num_channels(priv->mdev); if (priv->channels.params.num_channels > max_nch) { mlx5_core_warn(priv->mdev, "MLX5E: Reducing number of channels to %d\n", max_nch); + /* Reducing the number of channels - RXFH has to be reset, and + * mlx5e_num_channels_changed below will build the RQT. + */ + priv->netdev->priv_flags &= ~IFF_RXFH_CONFIGURED; priv->channels.params.num_channels = max_nch; - mlx5e_build_default_indir_rqt(priv->rss_params.indirection_rqt, - MLX5E_INDIR_RQT_SIZE, max_nch); } + /* 1. Set the real number of queues in the kernel the first time. + * 2. Set our default XPS cpumask. + * 3. Build the RQT. + * + * rtnl_lock is required by netif_set_real_num_*_queues in case the + * netdev has been registered by this point (if this function was called + * in the reload or resume flow). + */ + if (take_rtnl) + rtnl_lock(); + mlx5e_num_channels_changed(priv); + if (take_rtnl) + rtnl_unlock(); err = profile->init_tx(priv); if (err) @@ -5425,17 +5484,27 @@ static void *mlx5e_add(struct mlx5_core_dev *mdev) goto err_destroy_netdev; } + err = mlx5e_devlink_port_register(priv); + if (err) { + mlx5_core_err(mdev, "mlx5e_devlink_port_register failed, %d\n", err); + goto err_detach; + } + err = register_netdev(netdev); if (err) { mlx5_core_err(mdev, "register_netdev failed, %d\n", err); - goto err_detach; + goto err_devlink_port_unregister; } + mlx5e_devlink_port_type_eth_set(priv); + #ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_init_app(priv); #endif return priv; +err_devlink_port_unregister: + mlx5e_devlink_port_unregister(priv); err_detach: mlx5e_detach(mdev, priv); err_destroy_netdev: @@ -5457,6 +5526,7 @@ static void mlx5e_remove(struct mlx5_core_dev *mdev, void *vpriv) #ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_delete_app(priv); #endif + mlx5e_devlink_port_unregister(priv); unregister_netdev(priv->netdev); mlx5e_detach(mdev, vpriv); mlx5e_destroy_netdev(priv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 6ed307d7f191..2a0243e4af75 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -41,7 +41,7 @@ #include <net/ipv6_stubs.h> #include "eswitch.h" -#include "eswitch_offloads_chains.h" +#include "esw/chains.h" #include "en.h" #include "en_rep.h" #include "en_tc.h" @@ -192,7 +192,8 @@ static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(vport_rep) err = mlx5_eswitch_get_vport_stats(esw, rep->vport, &vf_stats); if (err) { - pr_warn("vport %d error %d reading stats\n", rep->vport, err); + netdev_warn(priv->netdev, "vport %d error %d reading stats\n", + rep->vport, err); return; } @@ -252,25 +253,6 @@ static int mlx5e_rep_set_ringparam(struct net_device *dev, return mlx5e_ethtool_set_ringparam(priv, param); } -static int mlx5e_replace_rep_vport_rx_rule(struct mlx5e_priv *priv, - struct mlx5_flow_destination *dest) -{ - struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5e_rep_priv *rpriv = priv->ppriv; - struct mlx5_eswitch_rep *rep = rpriv->rep; - struct mlx5_flow_handle *flow_rule; - - flow_rule = mlx5_eswitch_create_vport_rx_rule(esw, - rep->vport, - dest); - if (IS_ERR(flow_rule)) - return PTR_ERR(flow_rule); - - mlx5_del_flow_rules(rpriv->vport_rx_rule); - rpriv->vport_rx_rule = flow_rule; - return 0; -} - static void mlx5e_rep_get_channels(struct net_device *dev, struct ethtool_channels *ch) { @@ -283,33 +265,8 @@ static int mlx5e_rep_set_channels(struct net_device *dev, struct ethtool_channels *ch) { struct mlx5e_priv *priv = netdev_priv(dev); - u16 curr_channels_amount = priv->channels.params.num_channels; - u32 new_channels_amount = ch->combined_count; - struct mlx5_flow_destination new_dest; - int err = 0; - - err = mlx5e_ethtool_set_channels(priv, ch); - if (err) - return err; - if (curr_channels_amount == 1 && new_channels_amount > 1) { - new_dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - new_dest.ft = priv->fs.ttc.ft.t; - } else if (new_channels_amount == 1 && curr_channels_amount > 1) { - new_dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; - new_dest.tir_num = priv->direct_tir[0].tirn; - } else { - return 0; - } - - err = mlx5e_replace_rep_vport_rx_rule(priv, &new_dest); - if (err) { - netdev_warn(priv->netdev, "Failed to update vport rx rule, when going from (%d) channels to (%d) channels\n", - curr_channels_amount, new_channels_amount); - return err; - } - - return 0; + return mlx5e_ethtool_set_channels(priv, ch); } static int mlx5e_rep_get_coalesce(struct net_device *netdev, @@ -375,6 +332,9 @@ static int mlx5e_uplink_rep_set_link_ksettings(struct net_device *netdev, } static const struct ethtool_ops mlx5e_rep_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE, .get_drvinfo = mlx5e_rep_get_drvinfo, .get_link = ethtool_op_get_link, .get_strings = mlx5e_rep_get_strings, @@ -391,6 +351,9 @@ static const struct ethtool_ops mlx5e_rep_ethtool_ops = { }; static const struct ethtool_ops mlx5e_uplink_rep_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE, .get_drvinfo = mlx5e_uplink_rep_get_drvinfo, .get_link = ethtool_op_get_link, .get_strings = mlx5e_rep_get_strings, @@ -406,6 +369,10 @@ static const struct ethtool_ops mlx5e_uplink_rep_ethtool_ops = { .set_link_ksettings = mlx5e_uplink_rep_set_link_ksettings, .get_rxfh_key_size = mlx5e_rep_get_rxfh_key_size, .get_rxfh_indir_size = mlx5e_rep_get_rxfh_indir_size, + .get_rxfh = mlx5e_get_rxfh, + .set_rxfh = mlx5e_set_rxfh, + .get_rxnfc = mlx5e_get_rxnfc, + .set_rxnfc = mlx5e_set_rxnfc, .get_pauseparam = mlx5e_uplink_rep_get_pauseparam, .set_pauseparam = mlx5e_uplink_rep_set_pauseparam, }; @@ -727,9 +694,9 @@ static void mlx5e_rep_indr_clean_block_privs(struct mlx5e_rep_priv *rpriv) static int mlx5e_rep_indr_offload(struct net_device *netdev, struct flow_cls_offload *flower, - struct mlx5e_rep_indr_block_priv *indr_priv) + struct mlx5e_rep_indr_block_priv *indr_priv, + unsigned long flags) { - unsigned long flags = MLX5_TC_FLAG(EGRESS) | MLX5_TC_FLAG(ESW_OFFLOAD); struct mlx5e_priv *priv = netdev_priv(indr_priv->rpriv->netdev); int err = 0; @@ -750,20 +717,68 @@ mlx5e_rep_indr_offload(struct net_device *netdev, return err; } -static int mlx5e_rep_indr_setup_block_cb(enum tc_setup_type type, - void *type_data, void *indr_priv) +static int mlx5e_rep_indr_setup_tc_cb(enum tc_setup_type type, + void *type_data, void *indr_priv) { + unsigned long flags = MLX5_TC_FLAG(EGRESS) | MLX5_TC_FLAG(ESW_OFFLOAD); struct mlx5e_rep_indr_block_priv *priv = indr_priv; switch (type) { case TC_SETUP_CLSFLOWER: - return mlx5e_rep_indr_offload(priv->netdev, type_data, priv); + return mlx5e_rep_indr_offload(priv->netdev, type_data, priv, + flags); default: return -EOPNOTSUPP; } } -static void mlx5e_rep_indr_tc_block_unbind(void *cb_priv) +static int mlx5e_rep_indr_setup_ft_cb(enum tc_setup_type type, + void *type_data, void *indr_priv) +{ + struct mlx5e_rep_indr_block_priv *priv = indr_priv; + struct flow_cls_offload *f = type_data; + struct flow_cls_offload tmp; + struct mlx5e_priv *mpriv; + struct mlx5_eswitch *esw; + unsigned long flags; + int err; + + mpriv = netdev_priv(priv->rpriv->netdev); + esw = mpriv->mdev->priv.eswitch; + + flags = MLX5_TC_FLAG(EGRESS) | + MLX5_TC_FLAG(ESW_OFFLOAD) | + MLX5_TC_FLAG(FT_OFFLOAD); + + switch (type) { + case TC_SETUP_CLSFLOWER: + memcpy(&tmp, f, sizeof(*f)); + + /* Re-use tc offload path by moving the ft flow to the + * reserved ft chain. + * + * FT offload can use prio range [0, INT_MAX], so we normalize + * it to range [1, mlx5_esw_chains_get_prio_range(esw)] + * as with tc, where prio 0 isn't supported. + * + * We only support chain 0 of FT offload. + */ + if (!mlx5_esw_chains_prios_supported(esw) || + tmp.common.prio >= mlx5_esw_chains_get_prio_range(esw) || + tmp.common.chain_index) + return -EOPNOTSUPP; + + tmp.common.chain_index = mlx5_esw_chains_get_ft_chain(esw); + tmp.common.prio++; + err = mlx5e_rep_indr_offload(priv->netdev, &tmp, priv, flags); + memcpy(&f->stats, &tmp.stats, sizeof(f->stats)); + return err; + default: + return -EOPNOTSUPP; + } +} + +static void mlx5e_rep_indr_block_unbind(void *cb_priv) { struct mlx5e_rep_indr_block_priv *indr_priv = cb_priv; @@ -774,9 +789,10 @@ static void mlx5e_rep_indr_tc_block_unbind(void *cb_priv) static LIST_HEAD(mlx5e_block_cb_list); static int -mlx5e_rep_indr_setup_tc_block(struct net_device *netdev, - struct mlx5e_rep_priv *rpriv, - struct flow_block_offload *f) +mlx5e_rep_indr_setup_block(struct net_device *netdev, + struct mlx5e_rep_priv *rpriv, + struct flow_block_offload *f, + flow_setup_cb_t *setup_cb) { struct mlx5e_rep_indr_block_priv *indr_priv; struct flow_block_cb *block_cb; @@ -802,9 +818,8 @@ mlx5e_rep_indr_setup_tc_block(struct net_device *netdev, list_add(&indr_priv->list, &rpriv->uplink_priv.tc_indr_block_priv_list); - block_cb = flow_block_cb_alloc(mlx5e_rep_indr_setup_block_cb, - indr_priv, indr_priv, - mlx5e_rep_indr_tc_block_unbind); + block_cb = flow_block_cb_alloc(setup_cb, indr_priv, indr_priv, + mlx5e_rep_indr_block_unbind); if (IS_ERR(block_cb)) { list_del(&indr_priv->list); kfree(indr_priv); @@ -819,9 +834,7 @@ mlx5e_rep_indr_setup_tc_block(struct net_device *netdev, if (!indr_priv) return -ENOENT; - block_cb = flow_block_cb_lookup(f->block, - mlx5e_rep_indr_setup_block_cb, - indr_priv); + block_cb = flow_block_cb_lookup(f->block, setup_cb, indr_priv); if (!block_cb) return -ENOENT; @@ -835,13 +848,16 @@ mlx5e_rep_indr_setup_tc_block(struct net_device *netdev, } static -int mlx5e_rep_indr_setup_tc_cb(struct net_device *netdev, void *cb_priv, - enum tc_setup_type type, void *type_data) +int mlx5e_rep_indr_setup_cb(struct net_device *netdev, void *cb_priv, + enum tc_setup_type type, void *type_data) { switch (type) { case TC_SETUP_BLOCK: - return mlx5e_rep_indr_setup_tc_block(netdev, cb_priv, - type_data); + return mlx5e_rep_indr_setup_block(netdev, cb_priv, type_data, + mlx5e_rep_indr_setup_tc_cb); + case TC_SETUP_FT: + return mlx5e_rep_indr_setup_block(netdev, cb_priv, type_data, + mlx5e_rep_indr_setup_ft_cb); default: return -EOPNOTSUPP; } @@ -853,7 +869,7 @@ static int mlx5e_rep_indr_register_block(struct mlx5e_rep_priv *rpriv, int err; err = __flow_indr_block_cb_register(netdev, rpriv, - mlx5e_rep_indr_setup_tc_cb, + mlx5e_rep_indr_setup_cb, rpriv); if (err) { struct mlx5e_priv *priv = netdev_priv(rpriv->netdev); @@ -867,7 +883,7 @@ static int mlx5e_rep_indr_register_block(struct mlx5e_rep_priv *rpriv, static void mlx5e_rep_indr_unregister_block(struct mlx5e_rep_priv *rpriv, struct net_device *netdev) { - __flow_indr_block_cb_unregister(netdev, mlx5e_rep_indr_setup_tc_cb, + __flow_indr_block_cb_unregister(netdev, mlx5e_rep_indr_setup_cb, rpriv); } @@ -1279,8 +1295,7 @@ static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data, case TC_SETUP_CLSFLOWER: memcpy(&tmp, f, sizeof(*f)); - if (!mlx5_esw_chains_prios_supported(esw) || - tmp.common.chain_index) + if (!mlx5_esw_chains_prios_supported(esw)) return -EOPNOTSUPP; /* Re-use tc offload path by moving the ft flow to the @@ -1396,7 +1411,7 @@ static int mlx5e_rep_change_mtu(struct net_device *netdev, int new_mtu) static int mlx5e_uplink_rep_change_mtu(struct net_device *netdev, int new_mtu) { - return mlx5e_change_mtu(netdev, new_mtu, mlx5e_set_dev_port_mtu); + return mlx5e_change_mtu(netdev, new_mtu, mlx5e_set_dev_port_mtu_ctx); } static int mlx5e_uplink_rep_set_mac(struct net_device *netdev, void *addr) @@ -1422,7 +1437,7 @@ static int mlx5e_uplink_rep_set_vf_vlan(struct net_device *dev, int vf, u16 vlan return 0; } -static struct devlink_port *mlx5e_get_devlink_port(struct net_device *dev) +static struct devlink_port *mlx5e_rep_get_devlink_port(struct net_device *dev) { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5e_rep_priv *rpriv = priv->ppriv; @@ -1435,7 +1450,7 @@ static const struct net_device_ops mlx5e_netdev_ops_rep = { .ndo_stop = mlx5e_rep_close, .ndo_start_xmit = mlx5e_xmit, .ndo_setup_tc = mlx5e_rep_setup_tc, - .ndo_get_devlink_port = mlx5e_get_devlink_port, + .ndo_get_devlink_port = mlx5e_rep_get_devlink_port, .ndo_get_stats64 = mlx5e_rep_get_stats, .ndo_has_offload_stats = mlx5e_rep_has_offload_stats, .ndo_get_offload_stats = mlx5e_rep_get_offload_stats, @@ -1448,7 +1463,7 @@ static const struct net_device_ops mlx5e_netdev_ops_uplink_rep = { .ndo_start_xmit = mlx5e_xmit, .ndo_set_mac_address = mlx5e_uplink_rep_set_mac, .ndo_setup_tc = mlx5e_rep_setup_tc, - .ndo_get_devlink_port = mlx5e_get_devlink_port, + .ndo_get_devlink_port = mlx5e_rep_get_devlink_port, .ndo_get_stats64 = mlx5e_get_stats, .ndo_has_offload_stats = mlx5e_rep_has_offload_stats, .ndo_get_offload_stats = mlx5e_rep_get_offload_stats, @@ -1464,6 +1479,11 @@ static const struct net_device_ops mlx5e_netdev_ops_uplink_rep = { .ndo_set_features = mlx5e_set_features, }; +bool mlx5e_eswitch_uplink_rep(struct net_device *netdev) +{ + return netdev->netdev_ops == &mlx5e_netdev_ops_uplink_rep; +} + bool mlx5e_eswitch_rep(struct net_device *netdev) { if (netdev->netdev_ops == &mlx5e_netdev_ops_rep || @@ -1584,6 +1604,8 @@ static void mlx5e_cleanup_rep(struct mlx5e_priv *priv) static int mlx5e_create_rep_ttc_table(struct mlx5e_priv *priv) { + struct mlx5e_rep_priv *rpriv = priv->ppriv; + struct mlx5_eswitch_rep *rep = rpriv->rep; struct ttc_params ttc_params = {}; int tt, err; @@ -1593,6 +1615,11 @@ static int mlx5e_create_rep_ttc_table(struct mlx5e_priv *priv) /* The inner_ttc in the ttc params is intentionally not set */ ttc_params.any_tt_tirn = priv->direct_tir[0].tirn; mlx5e_set_ttc_ft_params(&ttc_params); + + if (rep->vport != MLX5_VPORT_UPLINK) + /* To give uplik rep TTC a lower level for chaining from root ft */ + ttc_params.ft_attr.level = MLX5E_TTC_FT_LEVEL + 1; + for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) ttc_params.indir_tirn[tt] = priv->indir_tir[tt].tirn; @@ -1604,6 +1631,52 @@ static int mlx5e_create_rep_ttc_table(struct mlx5e_priv *priv) return 0; } +static int mlx5e_create_rep_root_ft(struct mlx5e_priv *priv) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + struct mlx5_eswitch_rep *rep = rpriv->rep; + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_flow_namespace *ns; + int err = 0; + + if (rep->vport != MLX5_VPORT_UPLINK) { + /* non uplik reps will skip any bypass tables and go directly to + * their own ttc + */ + rpriv->root_ft = priv->fs.ttc.ft.t; + return 0; + } + + /* uplink root ft will be used to auto chain, to ethtool or ttc tables */ + ns = mlx5_get_flow_namespace(priv->mdev, MLX5_FLOW_NAMESPACE_OFFLOADS); + if (!ns) { + netdev_err(priv->netdev, "Failed to get reps offloads namespace\n"); + return -EOPNOTSUPP; + } + + ft_attr.max_fte = 0; /* Empty table, miss rule will always point to next table */ + ft_attr.prio = 1; + ft_attr.level = 1; + + rpriv->root_ft = mlx5_create_flow_table(ns, &ft_attr); + if (IS_ERR(rpriv->root_ft)) { + err = PTR_ERR(rpriv->root_ft); + rpriv->root_ft = NULL; + } + + return err; +} + +static void mlx5e_destroy_rep_root_ft(struct mlx5e_priv *priv) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + struct mlx5_eswitch_rep *rep = rpriv->rep; + + if (rep->vport != MLX5_VPORT_UPLINK) + return; + mlx5_destroy_flow_table(rpriv->root_ft); +} + static int mlx5e_create_rep_vport_rx_rule(struct mlx5e_priv *priv) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; @@ -1612,11 +1685,10 @@ static int mlx5e_create_rep_vport_rx_rule(struct mlx5e_priv *priv) struct mlx5_flow_handle *flow_rule; struct mlx5_flow_destination dest; - dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; - dest.tir_num = priv->direct_tir[0].tirn; - flow_rule = mlx5_eswitch_create_vport_rx_rule(esw, - rep->vport, - &dest); + dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest.ft = rpriv->root_ft; + + flow_rule = mlx5_eswitch_create_vport_rx_rule(esw, rep->vport, &dest); if (IS_ERR(flow_rule)) return PTR_ERR(flow_rule); rpriv->vport_rx_rule = flow_rule; @@ -1656,12 +1728,20 @@ static int mlx5e_init_rep_rx(struct mlx5e_priv *priv) if (err) goto err_destroy_direct_tirs; - err = mlx5e_create_rep_vport_rx_rule(priv); + err = mlx5e_create_rep_root_ft(priv); if (err) goto err_destroy_ttc_table; + err = mlx5e_create_rep_vport_rx_rule(priv); + if (err) + goto err_destroy_root_ft; + + mlx5e_ethtool_init_steering(priv); + return 0; +err_destroy_root_ft: + mlx5e_destroy_rep_root_ft(priv); err_destroy_ttc_table: mlx5e_destroy_ttc_table(priv, &priv->fs.ttc); err_destroy_direct_tirs: @@ -1682,6 +1762,7 @@ static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv) struct mlx5e_rep_priv *rpriv = priv->ppriv; mlx5_del_flow_rules(rpriv->vport_rx_rule); + mlx5e_destroy_rep_root_ft(priv); mlx5e_destroy_ttc_table(priv, &priv->fs.ttc); mlx5e_destroy_direct_tirs(priv, priv->direct_tir); mlx5e_destroy_indirect_tirs(priv, false); @@ -1920,7 +2001,7 @@ static const struct mlx5e_profile mlx5e_rep_profile = { .update_rx = mlx5e_update_rep_rx, .update_stats = mlx5e_update_ndo_stats, .rx_handlers.handle_rx_cqe = mlx5e_handle_rx_cqe_rep, - .rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq, + .rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq_rep, .max_tc = 1, .rq_groups = MLX5E_NUM_RQ_GROUPS(REGULAR), .stats_grps = mlx5e_rep_stats_grps, @@ -1940,7 +2021,7 @@ static const struct mlx5e_profile mlx5e_uplink_rep_profile = { .update_stats = mlx5e_update_ndo_stats, .update_carrier = mlx5e_update_carrier, .rx_handlers.handle_rx_cqe = mlx5e_handle_rx_cqe_rep, - .rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq, + .rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq_rep, .max_tc = MLX5E_MAX_NUM_TC, .rq_groups = MLX5E_NUM_RQ_GROUPS(REGULAR), .stats_grps = mlx5e_ul_rep_stats_grps, @@ -2026,8 +2107,9 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep) &mlx5e_uplink_rep_profile : &mlx5e_rep_profile; netdev = mlx5e_create_netdev(dev, profile, nch, rpriv); if (!netdev) { - pr_warn("Failed to create representor netdev for vport %d\n", - rep->vport); + mlx5_core_warn(dev, + "Failed to create representor netdev for vport %d\n", + rep->vport); kfree(rpriv); return -EINVAL; } @@ -2045,29 +2127,32 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep) err = mlx5e_attach_netdev(netdev_priv(netdev)); if (err) { - pr_warn("Failed to attach representor netdev for vport %d\n", - rep->vport); + netdev_warn(netdev, + "Failed to attach representor netdev for vport %d\n", + rep->vport); goto err_destroy_mdev_resources; } err = mlx5e_rep_neigh_init(rpriv); if (err) { - pr_warn("Failed to initialized neighbours handling for vport %d\n", - rep->vport); + netdev_warn(netdev, + "Failed to initialized neighbours handling for vport %d\n", + rep->vport); goto err_detach_netdev; } err = register_devlink_port(dev, rpriv); if (err) { - esw_warn(dev, "Failed to register devlink port %d\n", - rep->vport); + netdev_warn(netdev, "Failed to register devlink port %d\n", + rep->vport); goto err_neigh_cleanup; } err = register_netdev(netdev); if (err) { - pr_warn("Failed to register representor netdev for vport %d\n", - rep->vport); + netdev_warn(netdev, + "Failed to register representor netdev for vport %d\n", + rep->vport); goto err_devlink_cleanup; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h index 3f756d51435f..6a2337900420 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h @@ -55,6 +55,7 @@ struct mlx5e_neigh_update_table { unsigned long min_interval; /* jiffies */ }; +struct mlx5_tc_ct_priv; struct mlx5_rep_uplink_priv { /* Filters DB - instantiated by the uplink representor and shared by * the uplink's VFs @@ -81,12 +82,20 @@ struct mlx5_rep_uplink_priv { struct mutex unready_flows_lock; struct list_head unready_flows; struct work_struct reoffload_flows_work; + + /* maps tun_info to a unique id*/ + struct mapping_ctx *tunnel_mapping; + /* maps tun_enc_opts to a unique id*/ + struct mapping_ctx *tunnel_enc_opts_mapping; + + struct mlx5_tc_ct_priv *ct_priv; }; struct mlx5e_rep_priv { struct mlx5_eswitch_rep *rep; struct mlx5e_neigh_update_table neigh_update; struct net_device *netdev; + struct mlx5_flow_table *root_ft; struct mlx5_flow_handle *vport_rx_rule; struct list_head vport_sqs_list; struct mlx5_rep_uplink_priv uplink_priv; /* valid for uplink rep */ @@ -191,6 +200,8 @@ int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv); void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv); void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe); +void mlx5e_handle_rx_cqe_mpwrq_rep(struct mlx5e_rq *rq, + struct mlx5_cqe64 *cqe); int mlx5e_rep_encap_entry_attach(struct mlx5e_priv *priv, struct mlx5e_encap_entry *e); @@ -200,6 +211,7 @@ void mlx5e_rep_encap_entry_detach(struct mlx5e_priv *priv, void mlx5e_rep_queue_neigh_stats_work(struct mlx5e_priv *priv); bool mlx5e_eswitch_rep(struct net_device *netdev); +bool mlx5e_eswitch_uplink_rep(struct net_device *netdev); #else /* CONFIG_MLX5_ESWITCH */ static inline bool mlx5e_is_uplink_rep(struct mlx5e_priv *priv) { return false; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 312d4692425b..6173faf542b0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -158,7 +158,8 @@ static inline u32 mlx5e_decompress_cqes_cont(struct mlx5e_rq *rq, mlx5e_read_mini_arr_slot(wq, cqd, cqcc); mlx5e_decompress_cqe_no_hash(rq, wq, cqcc); - rq->handle_rx_cqe(rq, &cqd->title); + INDIRECT_CALL_2(rq->handle_rx_cqe, mlx5e_handle_rx_cqe_mpwrq, + mlx5e_handle_rx_cqe, rq, &cqd->title); } mlx5e_cqes_update_owner(wq, cqcc - wq->cc); wq->cc = cqcc; @@ -178,7 +179,8 @@ static inline u32 mlx5e_decompress_cqes_start(struct mlx5e_rq *rq, mlx5e_read_title_slot(rq, wq, cc); mlx5e_read_mini_arr_slot(wq, cqd, cc + 1); mlx5e_decompress_cqe(rq, wq, cc); - rq->handle_rx_cqe(rq, &cqd->title); + INDIRECT_CALL_2(rq->handle_rx_cqe, mlx5e_handle_rx_cqe_mpwrq, + mlx5e_handle_rx_cqe, rq, &cqd->title); cqd->mini_arr_idx++; return mlx5e_decompress_cqes_cont(rq, wq, 1, budget_rem) - 1; @@ -1192,6 +1194,7 @@ void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5e_rep_priv *rpriv = priv->ppriv; struct mlx5_eswitch_rep *rep = rpriv->rep; + struct mlx5e_tc_update_priv tc_priv = {}; struct mlx5_wq_cyc *wq = &rq->wqe.wq; struct mlx5e_wqe_frag_info *wi; struct sk_buff *skb; @@ -1224,13 +1227,78 @@ void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) if (rep->vlan && skb_vlan_tag_present(skb)) skb_vlan_pop(skb); + if (!mlx5e_tc_rep_update_skb(cqe, skb, &tc_priv)) + goto free_wqe; + napi_gro_receive(rq->cq.napi, skb); + mlx5_tc_rep_post_napi_receive(&tc_priv); + free_wqe: mlx5e_free_rx_wqe(rq, wi, true); wq_cyc_pop: mlx5_wq_cyc_pop(wq); } + +void mlx5e_handle_rx_cqe_mpwrq_rep(struct mlx5e_rq *rq, + struct mlx5_cqe64 *cqe) +{ + u16 cstrides = mpwrq_get_cqe_consumed_strides(cqe); + u16 wqe_id = be16_to_cpu(cqe->wqe_id); + struct mlx5e_mpw_info *wi = &rq->mpwqe.info[wqe_id]; + u16 stride_ix = mpwrq_get_cqe_stride_index(cqe); + u32 wqe_offset = stride_ix << rq->mpwqe.log_stride_sz; + u32 head_offset = wqe_offset & (PAGE_SIZE - 1); + u32 page_idx = wqe_offset >> PAGE_SHIFT; + struct mlx5e_tc_update_priv tc_priv = {}; + struct mlx5e_rx_wqe_ll *wqe; + struct mlx5_wq_ll *wq; + struct sk_buff *skb; + u16 cqe_bcnt; + + wi->consumed_strides += cstrides; + + if (unlikely(MLX5E_RX_ERR_CQE(cqe))) { + trigger_report(rq, cqe); + rq->stats->wqe_err++; + goto mpwrq_cqe_out; + } + + if (unlikely(mpwrq_is_filler_cqe(cqe))) { + struct mlx5e_rq_stats *stats = rq->stats; + + stats->mpwqe_filler_cqes++; + stats->mpwqe_filler_strides += cstrides; + goto mpwrq_cqe_out; + } + + cqe_bcnt = mpwrq_get_cqe_byte_cnt(cqe); + + skb = INDIRECT_CALL_2(rq->mpwqe.skb_from_cqe_mpwrq, + mlx5e_skb_from_cqe_mpwrq_linear, + mlx5e_skb_from_cqe_mpwrq_nonlinear, + rq, wi, cqe_bcnt, head_offset, page_idx); + if (!skb) + goto mpwrq_cqe_out; + + mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb); + + if (!mlx5e_tc_rep_update_skb(cqe, skb, &tc_priv)) + goto mpwrq_cqe_out; + + napi_gro_receive(rq->cq.napi, skb); + + mlx5_tc_rep_post_napi_receive(&tc_priv); + +mpwrq_cqe_out: + if (likely(wi->consumed_strides < rq->mpwqe.num_strides)) + return; + + wq = &rq->mpwqe.wq; + wqe = mlx5_wq_ll_get_wqe(wq, wqe_id); + mlx5e_free_rx_mpwqe(rq, wi, true); + mlx5_wq_ll_pop(wq, cqe->wqe_id, &wqe->next.next_wqe_index); +} #endif struct sk_buff * diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index ec5fc52bf572..438128dde187 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -51,14 +51,18 @@ #include "en_rep.h" #include "en_tc.h" #include "eswitch.h" -#include "eswitch_offloads_chains.h" +#include "esw/chains.h" #include "fs_core.h" #include "en/port.h" #include "en/tc_tun.h" +#include "en/mapping.h" +#include "en/tc_ct.h" #include "lib/devcom.h" #include "lib/geneve.h" #include "diag/en_tc_tracepoint.h" +#define MLX5_MH_ACT_SZ MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto) + struct mlx5_nic_flow_attr { u32 action; u32 flow_tag; @@ -84,6 +88,7 @@ enum { MLX5E_TC_FLOW_FLAG_DUP = MLX5E_TC_FLOW_BASE + 4, MLX5E_TC_FLOW_FLAG_NOT_READY = MLX5E_TC_FLOW_BASE + 5, MLX5E_TC_FLOW_FLAG_DELETED = MLX5E_TC_FLOW_BASE + 6, + MLX5E_TC_FLOW_FLAG_CT = MLX5E_TC_FLOW_BASE + 7, }; #define MLX5E_TC_MAX_SPLITS 1 @@ -134,6 +139,8 @@ struct mlx5e_tc_flow { refcount_t refcnt; struct rcu_head rcu_head; struct completion init_done; + int tunnel_id; /* the mapped tunnel id of this flow */ + union { struct mlx5_esw_flow_attr esw_attr[0]; struct mlx5_nic_flow_attr nic_attr[0]; @@ -144,15 +151,118 @@ struct mlx5e_tc_flow_parse_attr { const struct ip_tunnel_info *tun_info[MLX5_MAX_FLOW_FWD_VPORTS]; struct net_device *filter_dev; struct mlx5_flow_spec spec; - int num_mod_hdr_actions; - int max_mod_hdr_actions; - void *mod_hdr_actions; + struct mlx5e_tc_mod_hdr_acts mod_hdr_acts; int mirred_ifindex[MLX5_MAX_FLOW_FWD_VPORTS]; }; #define MLX5E_TC_TABLE_NUM_GROUPS 4 #define MLX5E_TC_TABLE_MAX_GROUP_SIZE BIT(16) +struct tunnel_match_key { + struct flow_dissector_key_control enc_control; + struct flow_dissector_key_keyid enc_key_id; + struct flow_dissector_key_ports enc_tp; + struct flow_dissector_key_ip enc_ip; + union { + struct flow_dissector_key_ipv4_addrs enc_ipv4; + struct flow_dissector_key_ipv6_addrs enc_ipv6; + }; + + int filter_ifindex; +}; + +/* Tunnel_id mapping is TUNNEL_INFO_BITS + ENC_OPTS_BITS. + * Upper TUNNEL_INFO_BITS for general tunnel info. + * Lower ENC_OPTS_BITS bits for enc_opts. + */ +#define TUNNEL_INFO_BITS 6 +#define TUNNEL_INFO_BITS_MASK GENMASK(TUNNEL_INFO_BITS - 1, 0) +#define ENC_OPTS_BITS 2 +#define ENC_OPTS_BITS_MASK GENMASK(ENC_OPTS_BITS - 1, 0) +#define TUNNEL_ID_BITS (TUNNEL_INFO_BITS + ENC_OPTS_BITS) +#define TUNNEL_ID_MASK GENMASK(TUNNEL_ID_BITS - 1, 0) + +struct mlx5e_tc_attr_to_reg_mapping mlx5e_tc_attr_to_reg_mappings[] = { + [CHAIN_TO_REG] = { + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_0, + .moffset = 0, + .mlen = 2, + }, + [TUNNEL_TO_REG] = { + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_1, + .moffset = 3, + .mlen = 1, + .soffset = MLX5_BYTE_OFF(fte_match_param, + misc_parameters_2.metadata_reg_c_1), + }, + [ZONE_TO_REG] = zone_to_reg_ct, + [CTSTATE_TO_REG] = ctstate_to_reg_ct, + [MARK_TO_REG] = mark_to_reg_ct, + [LABELS_TO_REG] = labels_to_reg_ct, + [FTEID_TO_REG] = fteid_to_reg_ct, + [TUPLEID_TO_REG] = tupleid_to_reg_ct, +}; + +static void mlx5e_put_flow_tunnel_id(struct mlx5e_tc_flow *flow); + +void +mlx5e_tc_match_to_reg_match(struct mlx5_flow_spec *spec, + enum mlx5e_tc_attr_to_reg type, + u32 data, + u32 mask) +{ + int soffset = mlx5e_tc_attr_to_reg_mappings[type].soffset; + int match_len = mlx5e_tc_attr_to_reg_mappings[type].mlen; + void *headers_c = spec->match_criteria; + void *headers_v = spec->match_value; + void *fmask, *fval; + + fmask = headers_c + soffset; + fval = headers_v + soffset; + + mask = cpu_to_be32(mask) >> (32 - (match_len * 8)); + data = cpu_to_be32(data) >> (32 - (match_len * 8)); + + memcpy(fmask, &mask, match_len); + memcpy(fval, &data, match_len); + + spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2; +} + +int +mlx5e_tc_match_to_reg_set(struct mlx5_core_dev *mdev, + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts, + enum mlx5e_tc_attr_to_reg type, + u32 data) +{ + int moffset = mlx5e_tc_attr_to_reg_mappings[type].moffset; + int mfield = mlx5e_tc_attr_to_reg_mappings[type].mfield; + int mlen = mlx5e_tc_attr_to_reg_mappings[type].mlen; + char *modact; + int err; + + err = alloc_mod_hdr_actions(mdev, MLX5_FLOW_NAMESPACE_FDB, + mod_hdr_acts); + if (err) + return err; + + modact = mod_hdr_acts->actions + + (mod_hdr_acts->num_actions * MLX5_MH_ACT_SZ); + + /* Firmware has 5bit length field and 0 means 32bits */ + if (mlen == 4) + mlen = 0; + + MLX5_SET(set_action_in, modact, action_type, MLX5_ACTION_TYPE_SET); + MLX5_SET(set_action_in, modact, field, mfield); + MLX5_SET(set_action_in, modact, offset, moffset * 8); + MLX5_SET(set_action_in, modact, length, mlen * 8); + MLX5_SET(set_action_in, modact, data, data); + mod_hdr_acts->num_actions++; + + return 0; +} + struct mlx5e_hairpin { struct mlx5_hairpin *pair; @@ -210,8 +320,6 @@ struct mlx5e_mod_hdr_entry { int compl_result; }; -#define MLX5_MH_ACT_SZ MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto) - static void mlx5e_tc_del_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow); @@ -361,10 +469,10 @@ static int mlx5e_attach_mod_hdr(struct mlx5e_priv *priv, struct mod_hdr_key key; u32 hash_key; - num_actions = parse_attr->num_mod_hdr_actions; + num_actions = parse_attr->mod_hdr_acts.num_actions; actions_size = MLX5_MH_ACT_SZ * num_actions; - key.actions = parse_attr->mod_hdr_actions; + key.actions = parse_attr->mod_hdr_acts.actions; key.num_actions = num_actions; hash_key = hash_mod_hdr_info(&key); @@ -954,7 +1062,7 @@ mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv, if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { err = mlx5e_attach_mod_hdr(priv, flow, parse_attr); flow_act.modify_hdr = attr->modify_hdr; - kfree(parse_attr->mod_hdr_actions); + dealloc_mod_hdr_actions(&parse_attr->mod_hdr_acts); if (err) return err; } @@ -1043,8 +1151,16 @@ mlx5e_tc_offload_fdb_rules(struct mlx5_eswitch *esw, struct mlx5_flow_spec *spec, struct mlx5_esw_flow_attr *attr) { + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts; struct mlx5_flow_handle *rule; + if (flow_flag_test(flow, CT)) { + mod_hdr_acts = &attr->parse_attr->mod_hdr_acts; + + return mlx5_tc_ct_flow_offload(flow->priv, flow, spec, attr, + mod_hdr_acts); + } + rule = mlx5_eswitch_add_offloaded_rule(esw, spec, attr); if (IS_ERR(rule)) return rule; @@ -1063,10 +1179,15 @@ mlx5e_tc_offload_fdb_rules(struct mlx5_eswitch *esw, static void mlx5e_tc_unoffload_fdb_rules(struct mlx5_eswitch *esw, struct mlx5e_tc_flow *flow, - struct mlx5_esw_flow_attr *attr) + struct mlx5_esw_flow_attr *attr) { flow_flag_clear(flow, OFFLOADED); + if (flow_flag_test(flow, CT)) { + mlx5_tc_ct_delete_flow(flow->priv, flow, attr); + return; + } + if (attr->split_count) mlx5_eswitch_del_fwd_rule(esw, flow->rule[1], attr); @@ -1076,17 +1197,17 @@ mlx5e_tc_unoffload_fdb_rules(struct mlx5_eswitch *esw, static struct mlx5_flow_handle * mlx5e_tc_offload_to_slow_path(struct mlx5_eswitch *esw, struct mlx5e_tc_flow *flow, - struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *slow_attr) + struct mlx5_flow_spec *spec) { + struct mlx5_esw_flow_attr slow_attr; struct mlx5_flow_handle *rule; - memcpy(slow_attr, flow->esw_attr, sizeof(*slow_attr)); - slow_attr->action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - slow_attr->split_count = 0; - slow_attr->flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; + memcpy(&slow_attr, flow->esw_attr, sizeof(slow_attr)); + slow_attr.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + slow_attr.split_count = 0; + slow_attr.flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; - rule = mlx5e_tc_offload_fdb_rules(esw, flow, spec, slow_attr); + rule = mlx5e_tc_offload_fdb_rules(esw, flow, spec, &slow_attr); if (!IS_ERR(rule)) flow_flag_set(flow, SLOW); @@ -1095,14 +1216,15 @@ mlx5e_tc_offload_to_slow_path(struct mlx5_eswitch *esw, static void mlx5e_tc_unoffload_from_slow_path(struct mlx5_eswitch *esw, - struct mlx5e_tc_flow *flow, - struct mlx5_esw_flow_attr *slow_attr) + struct mlx5e_tc_flow *flow) { - memcpy(slow_attr, flow->esw_attr, sizeof(*slow_attr)); - slow_attr->action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - slow_attr->split_count = 0; - slow_attr->flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; - mlx5e_tc_unoffload_fdb_rules(esw, flow, slow_attr); + struct mlx5_esw_flow_attr slow_attr; + + memcpy(&slow_attr, flow->esw_attr, sizeof(slow_attr)); + slow_attr.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + slow_attr.split_count = 0; + slow_attr.flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; + mlx5e_tc_unoffload_fdb_rules(esw, flow, &slow_attr); flow_flag_clear(flow, SLOW); } @@ -1173,7 +1295,8 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, int out_index; if (!mlx5_esw_chains_prios_supported(esw) && attr->prio != 1) { - NL_SET_ERR_MSG(extack, "E-switch priorities unsupported, upgrade FW"); + NL_SET_ERR_MSG_MOD(extack, + "E-switch priorities unsupported, upgrade FW"); return -EOPNOTSUPP; } @@ -1184,13 +1307,15 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, */ max_chain = mlx5_esw_chains_get_chain_range(esw); if (!mlx5e_is_ft_flow(flow) && attr->chain > max_chain) { - NL_SET_ERR_MSG(extack, "Requested chain is out of supported range"); + NL_SET_ERR_MSG_MOD(extack, + "Requested chain is out of supported range"); return -EOPNOTSUPP; } max_prio = mlx5_esw_chains_get_prio_range(esw); if (attr->prio > max_prio) { - NL_SET_ERR_MSG(extack, "Requested priority is out of supported range"); + NL_SET_ERR_MSG_MOD(extack, + "Requested priority is out of supported range"); return -EOPNOTSUPP; } @@ -1220,7 +1345,7 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { err = mlx5e_attach_mod_hdr(priv, flow, parse_attr); - kfree(parse_attr->mod_hdr_actions); + dealloc_mod_hdr_actions(&parse_attr->mod_hdr_acts); if (err) return err; } @@ -1237,14 +1362,10 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, * (1) there's no error * (2) there's an encap action and we don't have valid neigh */ - if (!encap_valid) { - /* continue with goto slow path rule instead */ - struct mlx5_esw_flow_attr slow_attr; - - flow->rule[0] = mlx5e_tc_offload_to_slow_path(esw, flow, &parse_attr->spec, &slow_attr); - } else { + if (!encap_valid) + flow->rule[0] = mlx5e_tc_offload_to_slow_path(esw, flow, &parse_attr->spec); + else flow->rule[0] = mlx5e_tc_offload_fdb_rules(esw, flow, &parse_attr->spec, attr); - } if (IS_ERR(flow->rule[0])) return PTR_ERR(flow->rule[0]); @@ -1272,9 +1393,10 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv, { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct mlx5_esw_flow_attr *attr = flow->esw_attr; - struct mlx5_esw_flow_attr slow_attr; int out_index; + mlx5e_put_flow_tunnel_id(flow); + if (flow_flag_test(flow, NOT_READY)) { remove_unready_flow(flow); kvfree(attr->parse_attr); @@ -1283,7 +1405,7 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv, if (mlx5e_is_offloaded_flow(flow)) { if (flow_flag_test(flow, SLOW)) - mlx5e_tc_unoffload_from_slow_path(esw, flow, &slow_attr); + mlx5e_tc_unoffload_from_slow_path(esw, flow); else mlx5e_tc_unoffload_fdb_rules(esw, flow, attr); } @@ -1312,7 +1434,7 @@ void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv, struct list_head *flow_list) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_esw_flow_attr slow_attr, *esw_attr; + struct mlx5_esw_flow_attr *esw_attr; struct mlx5_flow_handle *rule; struct mlx5_flow_spec *spec; struct mlx5e_tc_flow *flow; @@ -1365,7 +1487,7 @@ void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv, continue; } - mlx5e_tc_unoffload_from_slow_path(esw, flow, &slow_attr); + mlx5e_tc_unoffload_from_slow_path(esw, flow); flow->rule[0] = rule; /* was unset when slow path rule removed */ flow_flag_set(flow, OFFLOADED); @@ -1377,7 +1499,6 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, struct list_head *flow_list) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_esw_flow_attr slow_attr; struct mlx5_flow_handle *rule; struct mlx5_flow_spec *spec; struct mlx5e_tc_flow *flow; @@ -1389,7 +1510,7 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, spec = &flow->esw_attr->parse_attr->spec; /* update from encap rule to slow path rule */ - rule = mlx5e_tc_offload_to_slow_path(esw, flow, spec, &slow_attr); + rule = mlx5e_tc_offload_to_slow_path(esw, flow, spec); /* mark the flow's encap dest as non-valid */ flow->esw_attr->dests[flow->tmp_efi_index].flags &= ~MLX5_ESW_DEST_ENCAP_VALID; @@ -1664,150 +1785,272 @@ static void mlx5e_tc_del_flow(struct mlx5e_priv *priv, } } +static int flow_has_tc_fwd_action(struct flow_cls_offload *f) +{ + struct flow_rule *rule = flow_cls_offload_flow_rule(f); + struct flow_action *flow_action = &rule->action; + const struct flow_action_entry *act; + int i; -static int parse_tunnel_attr(struct mlx5e_priv *priv, - struct mlx5_flow_spec *spec, - struct flow_cls_offload *f, - struct net_device *filter_dev, u8 *match_level) + flow_action_for_each(i, act, flow_action) { + switch (act->id) { + case FLOW_ACTION_GOTO: + return true; + default: + continue; + } + } + + return false; +} + +static int +enc_opts_is_dont_care_or_full_match(struct mlx5e_priv *priv, + struct flow_dissector_key_enc_opts *opts, + struct netlink_ext_ack *extack, + bool *dont_care) +{ + struct geneve_opt *opt; + int off = 0; + + *dont_care = true; + + while (opts->len > off) { + opt = (struct geneve_opt *)&opts->data[off]; + + if (!(*dont_care) || opt->opt_class || opt->type || + memchr_inv(opt->opt_data, 0, opt->length * 4)) { + *dont_care = false; + + if (opt->opt_class != U16_MAX || + opt->type != U8_MAX || + memchr_inv(opt->opt_data, 0xFF, + opt->length * 4)) { + NL_SET_ERR_MSG(extack, + "Partial match of tunnel options in chain > 0 isn't supported"); + netdev_warn(priv->netdev, + "Partial match of tunnel options in chain > 0 isn't supported"); + return -EOPNOTSUPP; + } + } + + off += sizeof(struct geneve_opt) + opt->length * 4; + } + + return 0; +} + +#define COPY_DISSECTOR(rule, diss_key, dst)\ +({ \ + struct flow_rule *__rule = (rule);\ + typeof(dst) __dst = dst;\ +\ + memcpy(__dst,\ + skb_flow_dissector_target(__rule->match.dissector,\ + diss_key,\ + __rule->match.key),\ + sizeof(*__dst));\ +}) + +static int mlx5e_get_flow_tunnel_id(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct flow_cls_offload *f, + struct net_device *filter_dev) { - struct netlink_ext_ack *extack = f->common.extack; - void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, - outer_headers); - void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, - outer_headers); struct flow_rule *rule = flow_cls_offload_flow_rule(f); + struct netlink_ext_ack *extack = f->common.extack; + struct mlx5_esw_flow_attr *attr = flow->esw_attr; + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts; + struct flow_match_enc_opts enc_opts_match; + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *uplink_rpriv; + struct tunnel_match_key tunnel_key; + bool enc_opts_is_dont_care = true; + u32 tun_id, enc_opts_id = 0; + struct mlx5_eswitch *esw; + u32 value, mask; int err; - err = mlx5e_tc_tun_parse(filter_dev, priv, spec, f, - headers_c, headers_v, match_level); - if (err) { - NL_SET_ERR_MSG_MOD(extack, - "failed to parse tunnel attributes"); + esw = priv->mdev->priv.eswitch; + uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + uplink_priv = &uplink_rpriv->uplink_priv; + + memset(&tunnel_key, 0, sizeof(tunnel_key)); + COPY_DISSECTOR(rule, FLOW_DISSECTOR_KEY_ENC_CONTROL, + &tunnel_key.enc_control); + if (tunnel_key.enc_control.addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) + COPY_DISSECTOR(rule, FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS, + &tunnel_key.enc_ipv4); + else + COPY_DISSECTOR(rule, FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS, + &tunnel_key.enc_ipv6); + COPY_DISSECTOR(rule, FLOW_DISSECTOR_KEY_ENC_IP, &tunnel_key.enc_ip); + COPY_DISSECTOR(rule, FLOW_DISSECTOR_KEY_ENC_PORTS, + &tunnel_key.enc_tp); + COPY_DISSECTOR(rule, FLOW_DISSECTOR_KEY_ENC_KEYID, + &tunnel_key.enc_key_id); + tunnel_key.filter_ifindex = filter_dev->ifindex; + + err = mapping_add(uplink_priv->tunnel_mapping, &tunnel_key, &tun_id); + if (err) return err; - } - if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_CONTROL)) { - struct flow_match_control match; - u16 addr_type; + flow_rule_match_enc_opts(rule, &enc_opts_match); + err = enc_opts_is_dont_care_or_full_match(priv, + enc_opts_match.mask, + extack, + &enc_opts_is_dont_care); + if (err) + goto err_enc_opts; - flow_rule_match_enc_control(rule, &match); - addr_type = match.key->addr_type; + if (!enc_opts_is_dont_care) { + err = mapping_add(uplink_priv->tunnel_enc_opts_mapping, + enc_opts_match.key, &enc_opts_id); + if (err) + goto err_enc_opts; + } - /* For tunnel addr_type used same key id`s as for non-tunnel */ - if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) { - struct flow_match_ipv4_addrs match; + value = tun_id << ENC_OPTS_BITS | enc_opts_id; + mask = enc_opts_id ? TUNNEL_ID_MASK : + (TUNNEL_ID_MASK & ~ENC_OPTS_BITS_MASK); - flow_rule_match_enc_ipv4_addrs(rule, &match); - MLX5_SET(fte_match_set_lyr_2_4, headers_c, - src_ipv4_src_ipv6.ipv4_layout.ipv4, - ntohl(match.mask->src)); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, - src_ipv4_src_ipv6.ipv4_layout.ipv4, - ntohl(match.key->src)); + if (attr->chain) { + mlx5e_tc_match_to_reg_match(&attr->parse_attr->spec, + TUNNEL_TO_REG, value, mask); + } else { + mod_hdr_acts = &attr->parse_attr->mod_hdr_acts; + err = mlx5e_tc_match_to_reg_set(priv->mdev, + mod_hdr_acts, + TUNNEL_TO_REG, value); + if (err) + goto err_set; - MLX5_SET(fte_match_set_lyr_2_4, headers_c, - dst_ipv4_dst_ipv6.ipv4_layout.ipv4, - ntohl(match.mask->dst)); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, - dst_ipv4_dst_ipv6.ipv4_layout.ipv4, - ntohl(match.key->dst)); - - MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, - ethertype); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, - ETH_P_IP); - } else if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) { - struct flow_match_ipv6_addrs match; - - flow_rule_match_enc_ipv6_addrs(rule, &match); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, - src_ipv4_src_ipv6.ipv6_layout.ipv6), - &match.mask->src, MLX5_FLD_SZ_BYTES(ipv6_layout, - ipv6)); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, - src_ipv4_src_ipv6.ipv6_layout.ipv6), - &match.key->src, MLX5_FLD_SZ_BYTES(ipv6_layout, - ipv6)); - - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, - dst_ipv4_dst_ipv6.ipv6_layout.ipv6), - &match.mask->dst, MLX5_FLD_SZ_BYTES(ipv6_layout, - ipv6)); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, - dst_ipv4_dst_ipv6.ipv6_layout.ipv6), - &match.key->dst, MLX5_FLD_SZ_BYTES(ipv6_layout, - ipv6)); - - MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, - ethertype); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, - ETH_P_IPV6); - } + attr->action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; } - if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IP)) { - struct flow_match_ip match; + flow->tunnel_id = value; + return 0; - flow_rule_match_enc_ip(rule, &match); - MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_ecn, - match.mask->tos & 0x3); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_ecn, - match.key->tos & 0x3); +err_set: + if (enc_opts_id) + mapping_remove(uplink_priv->tunnel_enc_opts_mapping, + enc_opts_id); +err_enc_opts: + mapping_remove(uplink_priv->tunnel_mapping, tun_id); + return err; +} - MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_dscp, - match.mask->tos >> 2); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_dscp, - match.key->tos >> 2); +static void mlx5e_put_flow_tunnel_id(struct mlx5e_tc_flow *flow) +{ + u32 enc_opts_id = flow->tunnel_id & ENC_OPTS_BITS_MASK; + u32 tun_id = flow->tunnel_id >> ENC_OPTS_BITS; + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *uplink_rpriv; + struct mlx5_eswitch *esw; - MLX5_SET(fte_match_set_lyr_2_4, headers_c, ttl_hoplimit, - match.mask->ttl); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ttl_hoplimit, - match.key->ttl); + esw = flow->priv->mdev->priv.eswitch; + uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + uplink_priv = &uplink_rpriv->uplink_priv; + + if (tun_id) + mapping_remove(uplink_priv->tunnel_mapping, tun_id); + if (enc_opts_id) + mapping_remove(uplink_priv->tunnel_enc_opts_mapping, + enc_opts_id); +} - if (match.mask->ttl && - !MLX5_CAP_ESW_FLOWTABLE_FDB - (priv->mdev, - ft_field_support.outer_ipv4_ttl)) { +u32 mlx5e_tc_get_flow_tun_id(struct mlx5e_tc_flow *flow) +{ + return flow->tunnel_id; +} + +static int parse_tunnel_attr(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_spec *spec, + struct flow_cls_offload *f, + struct net_device *filter_dev, + u8 *match_level, + bool *match_inner) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct netlink_ext_ack *extack = f->common.extack; + bool needs_mapping, sets_mapping; + int err; + + if (!mlx5e_is_eswitch_flow(flow)) + return -EOPNOTSUPP; + + needs_mapping = !!flow->esw_attr->chain; + sets_mapping = !flow->esw_attr->chain && flow_has_tc_fwd_action(f); + *match_inner = !needs_mapping; + + if ((needs_mapping || sets_mapping) && + !mlx5_eswitch_reg_c1_loopback_enabled(esw)) { + NL_SET_ERR_MSG(extack, + "Chains on tunnel devices isn't supported without register loopback support"); + netdev_warn(priv->netdev, + "Chains on tunnel devices isn't supported without register loopback support"); + return -EOPNOTSUPP; + } + + if (!flow->esw_attr->chain) { + err = mlx5e_tc_tun_parse(filter_dev, priv, spec, f, + match_level); + if (err) { NL_SET_ERR_MSG_MOD(extack, - "Matching on TTL is not supported"); - return -EOPNOTSUPP; + "Failed to parse tunnel attributes"); + netdev_warn(priv->netdev, + "Failed to parse tunnel attributes"); + return err; } + flow->esw_attr->action |= MLX5_FLOW_CONTEXT_ACTION_DECAP; } - /* Enforce DMAC when offloading incoming tunneled flows. - * Flow counters require a match on the DMAC. - */ - MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, dmac_47_16); - MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, dmac_15_0); - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, - dmac_47_16), priv->netdev->dev_addr); + if (!needs_mapping && !sets_mapping) + return 0; - /* let software handle IP fragments */ - MLX5_SET(fte_match_set_lyr_2_4, headers_c, frag, 1); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, frag, 0); + return mlx5e_get_flow_tunnel_id(priv, flow, f, filter_dev); +} - return 0; +static void *get_match_inner_headers_criteria(struct mlx5_flow_spec *spec) +{ + return MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + inner_headers); } -static void *get_match_headers_criteria(u32 flags, - struct mlx5_flow_spec *spec) +static void *get_match_inner_headers_value(struct mlx5_flow_spec *spec) { - return (flags & MLX5_FLOW_CONTEXT_ACTION_DECAP) ? - MLX5_ADDR_OF(fte_match_param, spec->match_criteria, - inner_headers) : - MLX5_ADDR_OF(fte_match_param, spec->match_criteria, - outer_headers); + return MLX5_ADDR_OF(fte_match_param, spec->match_value, + inner_headers); +} + +static void *get_match_outer_headers_criteria(struct mlx5_flow_spec *spec) +{ + return MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers); +} + +static void *get_match_outer_headers_value(struct mlx5_flow_spec *spec) +{ + return MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers); } static void *get_match_headers_value(u32 flags, struct mlx5_flow_spec *spec) { return (flags & MLX5_FLOW_CONTEXT_ACTION_DECAP) ? - MLX5_ADDR_OF(fte_match_param, spec->match_value, - inner_headers) : - MLX5_ADDR_OF(fte_match_param, spec->match_value, - outer_headers); + get_match_inner_headers_value(spec) : + get_match_outer_headers_value(spec); +} + +static void *get_match_headers_criteria(u32 flags, + struct mlx5_flow_spec *spec) +{ + return (flags & MLX5_FLOW_CONTEXT_ACTION_DECAP) ? + get_match_inner_headers_criteria(spec) : + get_match_outer_headers_criteria(spec); } static int mlx5e_flower_parse_meta(struct net_device *filter_dev, @@ -1845,6 +2088,7 @@ static int mlx5e_flower_parse_meta(struct net_device *filter_dev, } static int __parse_cls_flower(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *spec, struct flow_cls_offload *f, struct net_device *filter_dev, @@ -1885,6 +2129,7 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, BIT(FLOW_DISSECTOR_KEY_ENC_CONTROL) | BIT(FLOW_DISSECTOR_KEY_TCP) | BIT(FLOW_DISSECTOR_KEY_IP) | + BIT(FLOW_DISSECTOR_KEY_CT) | BIT(FLOW_DISSECTOR_KEY_ENC_IP) | BIT(FLOW_DISSECTOR_KEY_ENC_OPTS))) { NL_SET_ERR_MSG_MOD(extack, "Unsupported key"); @@ -1894,18 +2139,22 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, } if (mlx5e_get_tc_tun(filter_dev)) { - if (parse_tunnel_attr(priv, spec, f, filter_dev, - outer_match_level)) - return -EOPNOTSUPP; + bool match_inner = false; - /* At this point, header pointers should point to the inner - * headers, outer header were already set by parse_tunnel_attr - */ - match_level = inner_match_level; - headers_c = get_match_headers_criteria(MLX5_FLOW_CONTEXT_ACTION_DECAP, - spec); - headers_v = get_match_headers_value(MLX5_FLOW_CONTEXT_ACTION_DECAP, - spec); + err = parse_tunnel_attr(priv, flow, spec, f, filter_dev, + outer_match_level, &match_inner); + if (err) + return err; + + if (match_inner) { + /* header pointers should point to the inner headers + * if the packet was decapsulated already. + * outer headers are set by parse_tunnel_attr. + */ + match_level = inner_match_level; + headers_c = get_match_inner_headers_criteria(spec); + headers_v = get_match_inner_headers_value(spec); + } } err = mlx5e_flower_parse_meta(filter_dev, f); @@ -2222,8 +2471,8 @@ static int parse_cls_flower(struct mlx5e_priv *priv, inner_match_level = MLX5_MATCH_NONE; outer_match_level = MLX5_MATCH_NONE; - err = __parse_cls_flower(priv, spec, f, filter_dev, &inner_match_level, - &outer_match_level); + err = __parse_cls_flower(priv, flow, spec, f, filter_dev, + &inner_match_level, &outer_match_level); non_tunnel_match_level = (inner_match_level == MLX5_MATCH_NONE) ? outer_match_level : inner_match_level; @@ -2383,25 +2632,26 @@ static struct mlx5_fields fields[] = { OFFLOAD(UDP_DPORT, 16, U16_MAX, udp.dest, 0, udp_dport), }; -/* On input attr->max_mod_hdr_actions tells how many HW actions can be parsed at - * max from the SW pedit action. On success, attr->num_mod_hdr_actions - * says how many HW actions were actually parsed. - */ -static int offload_pedit_fields(struct pedit_headers_action *hdrs, +static int offload_pedit_fields(struct mlx5e_priv *priv, + int namespace, + struct pedit_headers_action *hdrs, struct mlx5e_tc_flow_parse_attr *parse_attr, u32 *action_flags, struct netlink_ext_ack *extack) { struct pedit_headers *set_masks, *add_masks, *set_vals, *add_vals; - int i, action_size, nactions, max_actions, first, last, next_z; + int i, action_size, first, last, next_z; void *headers_c, *headers_v, *action, *vals_p; u32 *s_masks_p, *a_masks_p, s_mask, a_mask; + struct mlx5e_tc_mod_hdr_acts *mod_acts; struct mlx5_fields *f; unsigned long mask; __be32 mask_be32; __be16 mask_be16; + int err; u8 cmd; + mod_acts = &parse_attr->mod_hdr_acts; headers_c = get_match_headers_criteria(*action_flags, &parse_attr->spec); headers_v = get_match_headers_value(*action_flags, &parse_attr->spec); @@ -2411,11 +2661,6 @@ static int offload_pedit_fields(struct pedit_headers_action *hdrs, add_vals = &hdrs[1].vals; action_size = MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto); - action = parse_attr->mod_hdr_actions + - parse_attr->num_mod_hdr_actions * action_size; - - max_actions = parse_attr->max_mod_hdr_actions; - nactions = parse_attr->num_mod_hdr_actions; for (i = 0; i < ARRAY_SIZE(fields); i++) { bool skip; @@ -2441,13 +2686,6 @@ static int offload_pedit_fields(struct pedit_headers_action *hdrs, return -EOPNOTSUPP; } - if (nactions == max_actions) { - NL_SET_ERR_MSG_MOD(extack, - "too many pedit actions, can't offload"); - printk(KERN_WARNING "mlx5: parsed %d pedit actions, can't do more\n", nactions); - return -EOPNOTSUPP; - } - skip = false; if (s_mask) { void *match_mask = headers_c + f->match_offset; @@ -2495,6 +2733,18 @@ static int offload_pedit_fields(struct pedit_headers_action *hdrs, return -EOPNOTSUPP; } + err = alloc_mod_hdr_actions(priv->mdev, namespace, mod_acts); + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "too many pedit actions, can't offload"); + mlx5_core_warn(priv->mdev, + "mlx5: parsed %d pedit actions, can't do more\n", + mod_acts->num_actions); + return err; + } + + action = mod_acts->actions + + (mod_acts->num_actions * action_size); MLX5_SET(set_action_in, action, action_type, cmd); MLX5_SET(set_action_in, action, field, f->field); @@ -2517,11 +2767,9 @@ static int offload_pedit_fields(struct pedit_headers_action *hdrs, else if (f->field_bsize == 8) MLX5_SET(set_action_in, action, data, *(u8 *)vals_p >> first); - action += action_size; - nactions++; + ++mod_acts->num_actions; } - parse_attr->num_mod_hdr_actions = nactions; return 0; } @@ -2534,34 +2782,52 @@ static int mlx5e_flow_namespace_max_modify_action(struct mlx5_core_dev *mdev, return MLX5_CAP_FLOWTABLE_NIC_RX(mdev, max_modify_header_actions); } -static int alloc_mod_hdr_actions(struct mlx5e_priv *priv, - struct pedit_headers_action *hdrs, - int namespace, - struct mlx5e_tc_flow_parse_attr *parse_attr) +int alloc_mod_hdr_actions(struct mlx5_core_dev *mdev, + int namespace, + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts) { - int nkeys, action_size, max_actions; + int action_size, new_num_actions, max_hw_actions; + size_t new_sz, old_sz; + void *ret; - nkeys = hdrs[TCA_PEDIT_KEY_EX_CMD_SET].pedits + - hdrs[TCA_PEDIT_KEY_EX_CMD_ADD].pedits; - action_size = MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto); + if (mod_hdr_acts->num_actions < mod_hdr_acts->max_actions) + return 0; - max_actions = mlx5e_flow_namespace_max_modify_action(priv->mdev, namespace); - /* can get up to crazingly 16 HW actions in 32 bits pedit SW key */ - max_actions = min(max_actions, nkeys * 16); + action_size = MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto); - parse_attr->mod_hdr_actions = kcalloc(max_actions, action_size, GFP_KERNEL); - if (!parse_attr->mod_hdr_actions) + max_hw_actions = mlx5e_flow_namespace_max_modify_action(mdev, + namespace); + new_num_actions = min(max_hw_actions, + mod_hdr_acts->actions ? + mod_hdr_acts->max_actions * 2 : 1); + if (mod_hdr_acts->max_actions == new_num_actions) + return -ENOSPC; + + new_sz = action_size * new_num_actions; + old_sz = mod_hdr_acts->max_actions * action_size; + ret = krealloc(mod_hdr_acts->actions, new_sz, GFP_KERNEL); + if (!ret) return -ENOMEM; - parse_attr->max_mod_hdr_actions = max_actions; + memset(ret + old_sz, 0, new_sz - old_sz); + mod_hdr_acts->actions = ret; + mod_hdr_acts->max_actions = new_num_actions; + return 0; } +void dealloc_mod_hdr_actions(struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts) +{ + kfree(mod_hdr_acts->actions); + mod_hdr_acts->actions = NULL; + mod_hdr_acts->num_actions = 0; + mod_hdr_acts->max_actions = 0; +} + static const struct pedit_headers zero_masks = {}; static int parse_tc_pedit_action(struct mlx5e_priv *priv, const struct flow_action_entry *act, int namespace, - struct mlx5e_tc_flow_parse_attr *parse_attr, struct pedit_headers_action *hdrs, struct netlink_ext_ack *extack) { @@ -2609,13 +2875,8 @@ static int alloc_tc_pedit_action(struct mlx5e_priv *priv, int namespace, int err; u8 cmd; - if (!parse_attr->mod_hdr_actions) { - err = alloc_mod_hdr_actions(priv, hdrs, namespace, parse_attr); - if (err) - goto out_err; - } - - err = offload_pedit_fields(hdrs, parse_attr, action_flags, extack); + err = offload_pedit_fields(priv, namespace, hdrs, parse_attr, + action_flags, extack); if (err < 0) goto out_dealloc_parsed_actions; @@ -2635,8 +2896,7 @@ static int alloc_tc_pedit_action(struct mlx5e_priv *priv, int namespace, return 0; out_dealloc_parsed_actions: - kfree(parse_attr->mod_hdr_actions); -out_err: + dealloc_mod_hdr_actions(&parse_attr->mod_hdr_acts); return err; } @@ -2681,7 +2941,9 @@ struct ipv6_hoplimit_word { __u8 hop_limit; }; -static bool is_action_keys_supported(const struct flow_action_entry *act) +static int is_action_keys_supported(const struct flow_action_entry *act, + bool ct_flow, bool *modify_ip_header, + struct netlink_ext_ack *extack) { u32 mask, offset; u8 htype; @@ -2700,7 +2962,13 @@ static bool is_action_keys_supported(const struct flow_action_entry *act) if (offset != offsetof(struct iphdr, ttl) || ttl_word->protocol || ttl_word->check) { - return true; + *modify_ip_header = true; + } + + if (ct_flow && offset >= offsetof(struct iphdr, saddr)) { + NL_SET_ERR_MSG_MOD(extack, + "can't offload re-write of ipv4 address with action ct"); + return -EOPNOTSUPP; } } else if (htype == FLOW_ACT_MANGLE_HDR_TYPE_IP6) { struct ipv6_hoplimit_word *hoplimit_word = @@ -2709,15 +2977,27 @@ static bool is_action_keys_supported(const struct flow_action_entry *act) if (offset != offsetof(struct ipv6hdr, payload_len) || hoplimit_word->payload_len || hoplimit_word->nexthdr) { - return true; + *modify_ip_header = true; } + + if (ct_flow && offset >= offsetof(struct ipv6hdr, saddr)) { + NL_SET_ERR_MSG_MOD(extack, + "can't offload re-write of ipv6 address with action ct"); + return -EOPNOTSUPP; + } + } else if (ct_flow && (htype == FLOW_ACT_MANGLE_HDR_TYPE_TCP || + htype == FLOW_ACT_MANGLE_HDR_TYPE_UDP)) { + NL_SET_ERR_MSG_MOD(extack, + "can't offload re-write of transport header ports with action ct"); + return -EOPNOTSUPP; } - return false; + + return 0; } static bool modify_header_match_supported(struct mlx5_flow_spec *spec, struct flow_action *flow_action, - u32 actions, + u32 actions, bool ct_flow, struct netlink_ext_ack *extack) { const struct flow_action_entry *act; @@ -2725,7 +3005,7 @@ static bool modify_header_match_supported(struct mlx5_flow_spec *spec, void *headers_v; u16 ethertype; u8 ip_proto; - int i; + int i, err; headers_v = get_match_headers_value(actions, spec); ethertype = MLX5_GET(fte_match_set_lyr_2_4, headers_v, ethertype); @@ -2740,10 +3020,10 @@ static bool modify_header_match_supported(struct mlx5_flow_spec *spec, act->id != FLOW_ACTION_ADD) continue; - if (is_action_keys_supported(act)) { - modify_ip_header = true; - break; - } + err = is_action_keys_supported(act, ct_flow, + &modify_ip_header, extack); + if (err) + return err; } ip_proto = MLX5_GET(fte_match_set_lyr_2_4, headers_v, ip_protocol); @@ -2765,23 +3045,29 @@ static bool actions_match_supported(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack) { + bool ct_flow; u32 actions; - if (mlx5e_is_eswitch_flow(flow)) + ct_flow = flow_flag_test(flow, CT); + if (mlx5e_is_eswitch_flow(flow)) { actions = flow->esw_attr->action; - else - actions = flow->nic_attr->action; - if (flow_flag_test(flow, EGRESS) && - !((actions & MLX5_FLOW_CONTEXT_ACTION_DECAP) || - (actions & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) || - (actions & MLX5_FLOW_CONTEXT_ACTION_DROP))) - return false; + if (flow->esw_attr->split_count && ct_flow) { + /* All registers used by ct are cleared when using + * split rules. + */ + NL_SET_ERR_MSG_MOD(extack, + "Can't offload mirroring with action ct"); + return false; + } + } else { + actions = flow->nic_attr->action; + } if (actions & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) return modify_header_match_supported(&parse_attr->spec, flow_action, actions, - extack); + ct_flow, extack); return true; } @@ -2837,8 +3123,7 @@ static int add_vlan_rewrite_action(struct mlx5e_priv *priv, int namespace, return -EOPNOTSUPP; } - err = parse_tc_pedit_action(priv, &pedit_act, namespace, parse_attr, - hdrs, NULL); + err = parse_tc_pedit_action(priv, &pedit_act, namespace, hdrs, NULL); *action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; return err; @@ -2883,6 +3168,10 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, if (!flow_action_has_entries(flow_action)) return -EINVAL; + if (!flow_action_hw_stats_check(flow_action, extack, + FLOW_ACTION_HW_STATS_DELAYED_BIT)) + return -EOPNOTSUPP; + attr->flow_tag = MLX5_FS_DEFAULT_FLOW_TAG; flow_action_for_each(i, act, flow_action) { @@ -2900,7 +3189,7 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, case FLOW_ACTION_MANGLE: case FLOW_ACTION_ADD: err = parse_tc_pedit_action(priv, act, MLX5_FLOW_NAMESPACE_KERNEL, - parse_attr, hdrs, extack); + hdrs, extack); if (err) return err; @@ -2969,9 +3258,9 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, /* in case all pedit actions are skipped, remove the MOD_HDR * flag. */ - if (parse_attr->num_mod_hdr_actions == 0) { + if (parse_attr->mod_hdr_acts.num_actions == 0) { action &= ~MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; - kfree(parse_attr->mod_hdr_actions); + dealloc_mod_hdr_actions(&parse_attr->mod_hdr_acts); } } @@ -3314,6 +3603,85 @@ static bool is_duplicated_output_device(struct net_device *dev, return false; } +static int mlx5_validate_goto_chain(struct mlx5_eswitch *esw, + struct mlx5e_tc_flow *flow, + const struct flow_action_entry *act, + u32 actions, + struct netlink_ext_ack *extack) +{ + u32 max_chain = mlx5_esw_chains_get_chain_range(esw); + struct mlx5_esw_flow_attr *attr = flow->esw_attr; + bool ft_flow = mlx5e_is_ft_flow(flow); + u32 dest_chain = act->chain_index; + + if (ft_flow) { + NL_SET_ERR_MSG_MOD(extack, "Goto action is not supported"); + return -EOPNOTSUPP; + } + + if (!mlx5_esw_chains_backwards_supported(esw) && + dest_chain <= attr->chain) { + NL_SET_ERR_MSG_MOD(extack, + "Goto lower numbered chain isn't supported"); + return -EOPNOTSUPP; + } + if (dest_chain > max_chain) { + NL_SET_ERR_MSG_MOD(extack, + "Requested destination chain is out of supported range"); + return -EOPNOTSUPP; + } + + if (actions & (MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT | + MLX5_FLOW_CONTEXT_ACTION_DECAP) && + !MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, reformat_and_fwd_to_table)) { + NL_SET_ERR_MSG_MOD(extack, + "Goto chain is not allowed if action has reformat or decap"); + return -EOPNOTSUPP; + } + + return 0; +} + +static int verify_uplink_forwarding(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct net_device *out_dev, + struct netlink_ext_ack *extack) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_esw_flow_attr *attr = flow->esw_attr; + struct mlx5e_rep_priv *rep_priv; + + /* Forwarding non encapsulated traffic between + * uplink ports is allowed only if + * termination_table_raw_traffic cap is set. + * + * Input vport was stored esw_attr->in_rep. + * In LAG case, *priv* is the private data of + * uplink which may be not the input vport. + */ + rep_priv = mlx5e_rep_to_rep_priv(attr->in_rep); + + if (!(mlx5e_eswitch_uplink_rep(rep_priv->netdev) && + mlx5e_eswitch_uplink_rep(out_dev))) + return 0; + + if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, + termination_table_raw_traffic)) { + NL_SET_ERR_MSG_MOD(extack, + "devices are both uplink, can't offload forwarding"); + pr_err("devices %s %s are both uplink, can't offload forwarding\n", + priv->netdev->name, out_dev->name); + return -EOPNOTSUPP; + } else if (out_dev != rep_priv->netdev) { + NL_SET_ERR_MSG_MOD(extack, + "devices are not the same uplink, can't offload forwarding"); + pr_err("devices %s %s are both uplink but not the same, can't offload forwarding\n", + priv->netdev->name, out_dev->name); + return -EOPNOTSUPP; + } + return 0; +} + static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct flow_action *flow_action, struct mlx5e_tc_flow *flow, @@ -3328,13 +3696,17 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, int ifindexes[MLX5_MAX_FLOW_FWD_VPORTS]; bool ft_flow = mlx5e_is_ft_flow(flow); const struct flow_action_entry *act; + bool encap = false, decap = false; + u32 action = attr->action; int err, i, if_count = 0; - bool encap = false; - u32 action = 0; if (!flow_action_has_entries(flow_action)) return -EINVAL; + if (!flow_action_hw_stats_check(flow_action, extack, + FLOW_ACTION_HW_STATS_DELAYED_BIT)) + return -EOPNOTSUPP; + flow_action_for_each(i, act, flow_action) { switch (act->id) { case FLOW_ACTION_DROP: @@ -3344,7 +3716,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, case FLOW_ACTION_MANGLE: case FLOW_ACTION_ADD: err = parse_tc_pedit_action(priv, act, MLX5_FLOW_NAMESPACE_FDB, - parse_attr, hdrs, extack); + hdrs, extack); if (err) return err; @@ -3382,8 +3754,9 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, if (attr->out_count >= MLX5_MAX_FLOW_FWD_VPORTS) { NL_SET_ERR_MSG_MOD(extack, "can't support more output ports, can't offload forwarding"); - pr_err("can't support more than %d output ports, can't offload forwarding\n", - attr->out_count); + netdev_warn(priv->netdev, + "can't support more than %d output ports, can't offload forwarding\n", + attr->out_count); return -EOPNOTSUPP; } @@ -3441,11 +3814,17 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, return err; } + err = verify_uplink_forwarding(priv, flow, out_dev, extack); + if (err) + return err; + if (!mlx5e_is_valid_eswitch_fwd_dev(priv, out_dev)) { NL_SET_ERR_MSG_MOD(extack, "devices are not on same switch HW, can't offload forwarding"); - pr_err("devices %s %s not on same switch HW, can't offload forwarding\n", - priv->netdev->name, out_dev->name); + netdev_warn(priv->netdev, + "devices %s %s not on same switch HW, can't offload forwarding\n", + priv->netdev->name, + out_dev->name); return -EOPNOTSUPP; } @@ -3464,8 +3843,10 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, } else { NL_SET_ERR_MSG_MOD(extack, "devices are not on same switch HW, can't offload forwarding"); - pr_err("devices %s %s not on same switch HW, can't offload forwarding\n", - priv->netdev->name, out_dev->name); + netdev_warn(priv->netdev, + "devices %s %s not on same switch HW, can't offload forwarding\n", + priv->netdev->name, + out_dev->name); return -EINVAL; } } @@ -3507,28 +3888,24 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, attr->split_count = attr->out_count; break; case FLOW_ACTION_TUNNEL_DECAP: - action |= MLX5_FLOW_CONTEXT_ACTION_DECAP; + decap = true; break; - case FLOW_ACTION_GOTO: { - u32 dest_chain = act->chain_index; - u32 max_chain = mlx5_esw_chains_get_chain_range(esw); + case FLOW_ACTION_GOTO: + err = mlx5_validate_goto_chain(esw, flow, act, action, + extack); + if (err) + return err; - if (ft_flow) { - NL_SET_ERR_MSG_MOD(extack, "Goto action is not supported"); - return -EOPNOTSUPP; - } - if (dest_chain <= attr->chain) { - NL_SET_ERR_MSG(extack, "Goto earlier chain isn't supported"); - return -EOPNOTSUPP; - } - if (dest_chain > max_chain) { - NL_SET_ERR_MSG(extack, "Requested destination chain is out of supported range"); - return -EOPNOTSUPP; - } action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; - attr->dest_chain = dest_chain; + attr->dest_chain = act->chain_index; + break; + case FLOW_ACTION_CT: + err = mlx5_tc_ct_parse_action(priv, attr, act, extack); + if (err) + return err; + + flow_flag_set(flow, CT); break; - } default: NL_SET_ERR_MSG_MOD(extack, "The offload action is not supported"); return -EOPNOTSUPP; @@ -3557,9 +3934,9 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, * flag. we might have set split_count either by pedit or * pop/push. if there is no pop/push either, reset it too. */ - if (parse_attr->num_mod_hdr_actions == 0) { + if (parse_attr->mod_hdr_acts.num_actions == 0) { action &= ~MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; - kfree(parse_attr->mod_hdr_actions); + dealloc_mod_hdr_actions(&parse_attr->mod_hdr_acts); if (!((action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) || (action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH))) attr->split_count = 0; @@ -3571,8 +3948,25 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, return -EOPNOTSUPP; if (attr->dest_chain) { + if (decap) { + /* It can be supported if we'll create a mapping for + * the tunnel device only (without tunnel), and set + * this tunnel id with this decap flow. + * + * On restore (miss), we'll just set this saved tunnel + * device. + */ + + NL_SET_ERR_MSG(extack, + "Decap with goto isn't supported"); + netdev_warn(priv->netdev, + "Decap with goto isn't supported"); + return -EOPNOTSUPP; + } + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { - NL_SET_ERR_MSG(extack, "Mirroring goto chain rules isn't supported"); + NL_SET_ERR_MSG_MOD(extack, + "Mirroring goto chain rules isn't supported"); return -EOPNOTSUPP; } attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; @@ -3580,7 +3974,8 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, if (!(attr->action & (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | MLX5_FLOW_CONTEXT_ACTION_DROP))) { - NL_SET_ERR_MSG(extack, "Rule must have at least one forward/drop action"); + NL_SET_ERR_MSG_MOD(extack, + "Rule must have at least one forward/drop action"); return -EOPNOTSUPP; } @@ -3751,6 +4146,10 @@ __mlx5e_add_fdb_flow(struct mlx5e_priv *priv, if (err) goto err_free; + err = mlx5_tc_ct_parse_match(priv, &parse_attr->spec, f, extack); + if (err) + goto err_free; + err = mlx5e_tc_add_fdb_flow(priv, flow, extack); complete_all(&flow->init_done); if (err) { @@ -4035,7 +4434,7 @@ int mlx5e_stats_flower(struct net_device *dev, struct mlx5e_priv *priv, goto errout; } - if (mlx5e_is_offloaded_flow(flow)) { + if (mlx5e_is_offloaded_flow(flow) || flow_flag_test(flow, CT)) { counter = mlx5e_tc_get_counter(flow); if (!counter) goto errout; @@ -4069,7 +4468,8 @@ int mlx5e_stats_flower(struct net_device *dev, struct mlx5e_priv *priv, no_peer_counter: mlx5_devcom_release_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS); out: - flow_stats_update(&f->stats, bytes, packets, lastuse); + flow_stats_update(&f->stats, bytes, packets, lastuse, + FLOW_ACTION_HW_STATS_DELAYED); trace_mlx5e_stats_flower(f); errout: mlx5e_flow_put(priv, flow); @@ -4126,6 +4526,9 @@ static int scan_tc_matchall_fdb_actions(struct mlx5e_priv *priv, return -EOPNOTSUPP; } + if (!flow_action_basic_hw_stats_check(flow_action, extack)) + return -EOPNOTSUPP; + flow_action_for_each(i, act, flow_action) { switch (act->id) { case FLOW_ACTION_POLICE: @@ -4147,8 +4550,14 @@ static int scan_tc_matchall_fdb_actions(struct mlx5e_priv *priv, int mlx5e_tc_configure_matchall(struct mlx5e_priv *priv, struct tc_cls_matchall_offload *ma) { + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct netlink_ext_ack *extack = ma->common.extack; + if (!mlx5_esw_qos_enabled(esw)) { + NL_SET_ERR_MSG_MOD(extack, "QoS is not supported on this device"); + return -EOPNOTSUPP; + } + if (ma->common.prio != 1) { NL_SET_ERR_MSG_MOD(extack, "only priority 1 is supported"); return -EINVAL; @@ -4177,7 +4586,8 @@ void mlx5e_tc_stats_matchall(struct mlx5e_priv *priv, dpkts = cur_stats.rx_packets - rpriv->prev_vf_vport_stats.rx_packets; dbytes = cur_stats.rx_bytes - rpriv->prev_vf_vport_stats.rx_bytes; rpriv->prev_vf_vport_stats = cur_stats; - flow_stats_update(&ma->stats, dpkts, dbytes, jiffies); + flow_stats_update(&ma->stats, dpkts, dbytes, jiffies, + FLOW_ACTION_HW_STATS_DELAYED); } static void mlx5e_tc_hairpin_update_dead_peer(struct mlx5e_priv *priv, @@ -4295,12 +4705,63 @@ void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) int mlx5e_tc_esw_init(struct rhashtable *tc_ht) { - return rhashtable_init(tc_ht, &tc_ht_params); + const size_t sz_enc_opts = sizeof(struct flow_dissector_key_enc_opts); + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *priv; + struct mapping_ctx *mapping; + int err; + + uplink_priv = container_of(tc_ht, struct mlx5_rep_uplink_priv, tc_ht); + priv = container_of(uplink_priv, struct mlx5e_rep_priv, uplink_priv); + + err = mlx5_tc_ct_init(uplink_priv); + if (err) + goto err_ct; + + mapping = mapping_create(sizeof(struct tunnel_match_key), + TUNNEL_INFO_BITS_MASK, true); + if (IS_ERR(mapping)) { + err = PTR_ERR(mapping); + goto err_tun_mapping; + } + uplink_priv->tunnel_mapping = mapping; + + mapping = mapping_create(sz_enc_opts, ENC_OPTS_BITS_MASK, true); + if (IS_ERR(mapping)) { + err = PTR_ERR(mapping); + goto err_enc_opts_mapping; + } + uplink_priv->tunnel_enc_opts_mapping = mapping; + + err = rhashtable_init(tc_ht, &tc_ht_params); + if (err) + goto err_ht_init; + + return err; + +err_ht_init: + mapping_destroy(uplink_priv->tunnel_enc_opts_mapping); +err_enc_opts_mapping: + mapping_destroy(uplink_priv->tunnel_mapping); +err_tun_mapping: + mlx5_tc_ct_clean(uplink_priv); +err_ct: + netdev_warn(priv->netdev, + "Failed to initialize tc (eswitch), err: %d", err); + return err; } void mlx5e_tc_esw_cleanup(struct rhashtable *tc_ht) { + struct mlx5_rep_uplink_priv *uplink_priv; + rhashtable_free_and_destroy(tc_ht, _mlx5e_tc_del_flow, NULL); + + uplink_priv = container_of(tc_ht, struct mlx5_rep_uplink_priv, tc_ht); + mapping_destroy(uplink_priv->tunnel_enc_opts_mapping); + mapping_destroy(uplink_priv->tunnel_mapping); + + mlx5_tc_ct_clean(uplink_priv); } int mlx5e_tc_num_filters(struct mlx5e_priv *priv, unsigned long flags) @@ -4332,3 +4793,147 @@ void mlx5e_tc_reoffload_flows_work(struct work_struct *work) } mutex_unlock(&rpriv->unready_flows_lock); } + +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) +static bool mlx5e_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb, + struct mlx5e_tc_update_priv *tc_priv, + u32 tunnel_id) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct flow_dissector_key_enc_opts enc_opts = {}; + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *uplink_rpriv; + struct metadata_dst *tun_dst; + struct tunnel_match_key key; + u32 tun_id, enc_opts_id; + struct net_device *dev; + int err; + + enc_opts_id = tunnel_id & ENC_OPTS_BITS_MASK; + tun_id = tunnel_id >> ENC_OPTS_BITS; + + if (!tun_id) + return true; + + uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + uplink_priv = &uplink_rpriv->uplink_priv; + + err = mapping_find(uplink_priv->tunnel_mapping, tun_id, &key); + if (err) { + WARN_ON_ONCE(true); + netdev_dbg(priv->netdev, + "Couldn't find tunnel for tun_id: %d, err: %d\n", + tun_id, err); + return false; + } + + if (enc_opts_id) { + err = mapping_find(uplink_priv->tunnel_enc_opts_mapping, + enc_opts_id, &enc_opts); + if (err) { + netdev_dbg(priv->netdev, + "Couldn't find tunnel (opts) for tun_id: %d, err: %d\n", + enc_opts_id, err); + return false; + } + } + + tun_dst = tun_rx_dst(enc_opts.len); + if (!tun_dst) { + WARN_ON_ONCE(true); + return false; + } + + ip_tunnel_key_init(&tun_dst->u.tun_info.key, + key.enc_ipv4.src, key.enc_ipv4.dst, + key.enc_ip.tos, key.enc_ip.ttl, + 0, /* label */ + key.enc_tp.src, key.enc_tp.dst, + key32_to_tunnel_id(key.enc_key_id.keyid), + TUNNEL_KEY); + + if (enc_opts.len) + ip_tunnel_info_opts_set(&tun_dst->u.tun_info, enc_opts.data, + enc_opts.len, enc_opts.dst_opt_type); + + skb_dst_set(skb, (struct dst_entry *)tun_dst); + dev = dev_get_by_index(&init_net, key.filter_ifindex); + if (!dev) { + netdev_dbg(priv->netdev, + "Couldn't find tunnel device with ifindex: %d\n", + key.filter_ifindex); + return false; + } + + /* Set tun_dev so we do dev_put() after datapath */ + tc_priv->tun_dev = dev; + + skb->dev = dev; + + return true; +} +#endif /* CONFIG_NET_TC_SKB_EXT */ + +bool mlx5e_tc_rep_update_skb(struct mlx5_cqe64 *cqe, + struct sk_buff *skb, + struct mlx5e_tc_update_priv *tc_priv) +{ +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + u32 chain = 0, reg_c0, reg_c1, tunnel_id, tuple_id; + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *uplink_rpriv; + struct tc_skb_ext *tc_skb_ext; + struct mlx5_eswitch *esw; + struct mlx5e_priv *priv; + int tunnel_moffset; + int err; + + reg_c0 = (be32_to_cpu(cqe->sop_drop_qpn) & MLX5E_TC_FLOW_ID_MASK); + if (reg_c0 == MLX5_FS_DEFAULT_FLOW_TAG) + reg_c0 = 0; + reg_c1 = be32_to_cpu(cqe->imm_inval_pkey); + + if (!reg_c0) + return true; + + priv = netdev_priv(skb->dev); + esw = priv->mdev->priv.eswitch; + + err = mlx5_eswitch_get_chain_for_tag(esw, reg_c0, &chain); + if (err) { + netdev_dbg(priv->netdev, + "Couldn't find chain for chain tag: %d, err: %d\n", + reg_c0, err); + return false; + } + + if (chain) { + tc_skb_ext = skb_ext_add(skb, TC_SKB_EXT); + if (!tc_skb_ext) { + WARN_ON(1); + return false; + } + + tc_skb_ext->chain = chain; + + tuple_id = reg_c1 & TUPLE_ID_MAX; + + uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + uplink_priv = &uplink_rpriv->uplink_priv; + if (!mlx5e_tc_ct_restore_flow(uplink_priv, skb, tuple_id)) + return false; + } + + tunnel_moffset = mlx5e_tc_attr_to_reg_mappings[TUNNEL_TO_REG].moffset; + tunnel_id = reg_c1 >> (8 * tunnel_moffset); + return mlx5e_restore_tunnel(priv, skb, tc_priv, tunnel_id); +#endif /* CONFIG_NET_TC_SKB_EXT */ + + return true; +} + +void mlx5_tc_rep_post_napi_receive(struct mlx5e_tc_update_priv *tc_priv) +{ + if (tc_priv->tun_dev) + dev_put(tc_priv->tun_dev); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h index 262cdb7b69b1..abdcfa4c4e0e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h @@ -91,9 +91,63 @@ int mlx5e_tc_num_filters(struct mlx5e_priv *priv, unsigned long flags); void mlx5e_tc_reoffload_flows_work(struct work_struct *work); +enum mlx5e_tc_attr_to_reg { + CHAIN_TO_REG, + TUNNEL_TO_REG, + CTSTATE_TO_REG, + ZONE_TO_REG, + MARK_TO_REG, + LABELS_TO_REG, + FTEID_TO_REG, + TUPLEID_TO_REG, +}; + +struct mlx5e_tc_attr_to_reg_mapping { + int mfield; /* rewrite field */ + int moffset; /* offset of mfield */ + int mlen; /* bytes to rewrite/match */ + + int soffset; /* offset of spec for match */ +}; + +extern struct mlx5e_tc_attr_to_reg_mapping mlx5e_tc_attr_to_reg_mappings[]; + bool mlx5e_is_valid_eswitch_fwd_dev(struct mlx5e_priv *priv, struct net_device *out_dev); +struct mlx5e_tc_update_priv { + struct net_device *tun_dev; +}; + +bool mlx5e_tc_rep_update_skb(struct mlx5_cqe64 *cqe, struct sk_buff *skb, + struct mlx5e_tc_update_priv *tc_priv); + +void mlx5_tc_rep_post_napi_receive(struct mlx5e_tc_update_priv *tc_priv); + +struct mlx5e_tc_mod_hdr_acts { + int num_actions; + int max_actions; + void *actions; +}; + +int mlx5e_tc_match_to_reg_set(struct mlx5_core_dev *mdev, + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts, + enum mlx5e_tc_attr_to_reg type, + u32 data); + +void mlx5e_tc_match_to_reg_match(struct mlx5_flow_spec *spec, + enum mlx5e_tc_attr_to_reg type, + u32 data, + u32 mask); + +int alloc_mod_hdr_actions(struct mlx5_core_dev *mdev, + int namespace, + struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts); +void dealloc_mod_hdr_actions(struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts); + +struct mlx5e_tc_flow; +u32 mlx5e_tc_get_flow_tun_id(struct mlx5e_tc_flow *flow); + #else /* CONFIG_MLX5_ESWITCH */ static inline int mlx5e_tc_nic_init(struct mlx5e_priv *priv) { return 0; } static inline void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) {} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index ee60383adc5b..fd6b2a1898c5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -72,8 +72,8 @@ u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, { int txq_ix = netdev_pick_tx(dev, skb, NULL); struct mlx5e_priv *priv = netdev_priv(dev); - u16 num_channels; int up = 0; + int ch_ix; if (!netdev_get_num_tc(dev)) return txq_ix; @@ -86,14 +86,13 @@ u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, if (skb_vlan_tag_present(skb)) up = skb_vlan_tag_get_prio(skb); - /* txq_ix can be larger than num_channels since - * dev->num_real_tx_queues = num_channels * num_tc + /* Normalize any picked txq_ix to [0, num_channels), + * So we can return a txq_ix that matches the channel and + * packet UP. */ - num_channels = priv->channels.params.num_channels; - if (txq_ix >= num_channels) - txq_ix = priv->txq2sq[txq_ix]->ch_ix; + ch_ix = priv->txq2sq[txq_ix]->ch_ix; - return priv->channel_tc2realtxq[txq_ix][up]; + return priv->channel_tc2realtxq[ch_ix][up]; } static inline int mlx5e_skb_l2_header_offset(struct sk_buff *skb) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index 800d34ed8a96..87c49e7a164c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -31,6 +31,7 @@ */ #include <linux/irq.h> +#include <linux/indirect_call_wrapper.h> #include "en.h" #include "en/xdp.h" #include "en/xsk/rx.h" @@ -100,7 +101,10 @@ static bool mlx5e_napi_xsk_post(struct mlx5e_xdpsq *xsksq, struct mlx5e_rq *xskr busy_xsk |= mlx5e_xsk_tx(xsksq, MLX5E_TX_XSK_POLL_BUDGET); mlx5e_xsk_update_tx_wakeup(xsksq); - xsk_rx_alloc_err = xskrq->post_wqes(xskrq); + xsk_rx_alloc_err = INDIRECT_CALL_2(xskrq->post_wqes, + mlx5e_post_rx_mpwqes, + mlx5e_post_rx_wqes, + xskrq); busy_xsk |= mlx5e_xsk_update_rx_wakeup(xskrq, xsk_rx_alloc_err); return busy_xsk; @@ -143,7 +147,10 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget) mlx5e_poll_ico_cq(&c->icosq.cq); - busy |= rq->post_wqes(rq); + busy |= INDIRECT_CALL_2(rq->post_wqes, + mlx5e_post_rx_mpwqes, + mlx5e_post_rx_wqes, + rq); if (xsk_open) { mlx5e_poll_ico_cq(&c->xskicosq.cq); busy |= mlx5e_poll_xdpsq_cq(&xsksq->cq); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/esw/Makefile new file mode 100644 index 000000000000..c78512eed8d7 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/Makefile @@ -0,0 +1,2 @@ +# SPDX-License-Identifier: GPL-2.0-only +subdir-ccflags-y += -I$(src)/.. diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.c index 4276194b633f..029001040737 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.c @@ -5,23 +5,26 @@ #include <linux/mlx5/mlx5_ifc.h> #include <linux/mlx5/fs.h> -#include "eswitch_offloads_chains.h" +#include "esw/chains.h" +#include "en/mapping.h" #include "mlx5_core.h" #include "fs_core.h" #include "eswitch.h" #include "en.h" +#include "en_tc.h" #define esw_chains_priv(esw) ((esw)->fdb_table.offloads.esw_chains_priv) #define esw_chains_lock(esw) (esw_chains_priv(esw)->lock) #define esw_chains_ht(esw) (esw_chains_priv(esw)->chains_ht) +#define esw_chains_mapping(esw) (esw_chains_priv(esw)->chains_mapping) #define esw_prios_ht(esw) (esw_chains_priv(esw)->prios_ht) #define fdb_pool_left(esw) (esw_chains_priv(esw)->fdb_left) #define tc_slow_fdb(esw) ((esw)->fdb_table.offloads.slow_fdb) #define tc_end_fdb(esw) (esw_chains_priv(esw)->tc_end_fdb) #define fdb_ignore_flow_level_supported(esw) \ (MLX5_CAP_ESW_FLOWTABLE_FDB((esw)->dev, ignore_flow_level)) - -#define ESW_OFFLOADS_NUM_GROUPS 4 +#define fdb_modify_header_fwd_to_table_supported(esw) \ + (MLX5_CAP_ESW_FLOWTABLE((esw)->dev, fdb_modify_header_fwd_to_table)) /* Firmware currently has 4 pool of 4 sizes that it supports (ESW_POOLS), * and a virtual memory region of 16M (ESW_SIZE), this region is duplicated @@ -36,6 +39,7 @@ static const unsigned int ESW_POOLS[] = { 4 * 1024 * 1024, 1 * 1024 * 1024, 64 * 1024, 128 }; +#define ESW_FT_TBL_SZ (64 * 1024) struct mlx5_esw_chains_priv { struct rhashtable chains_ht; @@ -44,6 +48,7 @@ struct mlx5_esw_chains_priv { struct mutex lock; struct mlx5_flow_table *tc_end_fdb; + struct mapping_ctx *chains_mapping; int fdb_left[ARRAY_SIZE(ESW_POOLS)]; }; @@ -54,9 +59,12 @@ struct fdb_chain { u32 chain; int ref; + int id; struct mlx5_eswitch *esw; struct list_head prios_list; + struct mlx5_flow_handle *restore_rule; + struct mlx5_modify_hdr *miss_modify_hdr; }; struct fdb_prio_key { @@ -99,6 +107,12 @@ bool mlx5_esw_chains_prios_supported(struct mlx5_eswitch *esw) return esw->fdb_table.flags & ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; } +bool mlx5_esw_chains_backwards_supported(struct mlx5_eswitch *esw) +{ + return mlx5_esw_chains_prios_supported(esw) && + fdb_ignore_flow_level_supported(esw); +} + u32 mlx5_esw_chains_get_chain_range(struct mlx5_eswitch *esw) { if (!mlx5_esw_chains_prios_supported(esw)) @@ -198,7 +212,9 @@ mlx5_esw_chains_create_fdb_table(struct mlx5_eswitch *esw, ft_attr.flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT | MLX5_FLOW_TABLE_TUNNEL_EN_DECAP); - sz = mlx5_esw_chains_get_avail_sz_from_pool(esw, POOL_NEXT_SIZE); + sz = (chain == mlx5_esw_chains_get_ft_chain(esw)) ? + mlx5_esw_chains_get_avail_sz_from_pool(esw, ESW_FT_TBL_SZ) : + mlx5_esw_chains_get_avail_sz_from_pool(esw, POOL_NEXT_SIZE); if (!sz) return ERR_PTR(-ENOSPC); ft_attr.max_fte = sz; @@ -234,7 +250,7 @@ mlx5_esw_chains_create_fdb_table(struct mlx5_eswitch *esw, } ft_attr.autogroup.num_reserved_entries = 2; - ft_attr.autogroup.max_num_groups = ESW_OFFLOADS_NUM_GROUPS; + ft_attr.autogroup.max_num_groups = esw->params.large_group_num; fdb = mlx5_create_auto_grouped_flow_table(ns, &ft_attr); if (IS_ERR(fdb)) { esw_warn(esw->dev, @@ -255,6 +271,83 @@ mlx5_esw_chains_destroy_fdb_table(struct mlx5_eswitch *esw, mlx5_destroy_flow_table(fdb); } +static int +create_fdb_chain_restore(struct fdb_chain *fdb_chain) +{ + char modact[MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto)]; + struct mlx5_eswitch *esw = fdb_chain->esw; + struct mlx5_modify_hdr *mod_hdr; + u32 index; + int err; + + if (fdb_chain->chain == mlx5_esw_chains_get_ft_chain(esw) || + !mlx5_esw_chains_prios_supported(esw)) + return 0; + + err = mapping_add(esw_chains_mapping(esw), &fdb_chain->chain, &index); + if (err) + return err; + if (index == MLX5_FS_DEFAULT_FLOW_TAG) { + /* we got the special default flow tag id, so we won't know + * if we actually marked the packet with the restore rule + * we create. + * + * This case isn't possible with MLX5_FS_DEFAULT_FLOW_TAG = 0. + */ + err = mapping_add(esw_chains_mapping(esw), + &fdb_chain->chain, &index); + mapping_remove(esw_chains_mapping(esw), + MLX5_FS_DEFAULT_FLOW_TAG); + if (err) + return err; + } + + fdb_chain->id = index; + + MLX5_SET(set_action_in, modact, action_type, MLX5_ACTION_TYPE_SET); + MLX5_SET(set_action_in, modact, field, + mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].mfield); + MLX5_SET(set_action_in, modact, offset, + mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].moffset * 8); + MLX5_SET(set_action_in, modact, length, + mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].mlen * 8); + MLX5_SET(set_action_in, modact, data, fdb_chain->id); + mod_hdr = mlx5_modify_header_alloc(esw->dev, MLX5_FLOW_NAMESPACE_FDB, + 1, modact); + if (IS_ERR(mod_hdr)) { + err = PTR_ERR(mod_hdr); + goto err_mod_hdr; + } + fdb_chain->miss_modify_hdr = mod_hdr; + + fdb_chain->restore_rule = esw_add_restore_rule(esw, fdb_chain->id); + if (IS_ERR(fdb_chain->restore_rule)) { + err = PTR_ERR(fdb_chain->restore_rule); + goto err_rule; + } + + return 0; + +err_rule: + mlx5_modify_header_dealloc(esw->dev, fdb_chain->miss_modify_hdr); +err_mod_hdr: + /* Datapath can't find this mapping, so we can safely remove it */ + mapping_remove(esw_chains_mapping(esw), fdb_chain->id); + return err; +} + +static void destroy_fdb_chain_restore(struct fdb_chain *fdb_chain) +{ + struct mlx5_eswitch *esw = fdb_chain->esw; + + if (!fdb_chain->miss_modify_hdr) + return; + + mlx5_del_flow_rules(fdb_chain->restore_rule); + mlx5_modify_header_dealloc(esw->dev, fdb_chain->miss_modify_hdr); + mapping_remove(esw_chains_mapping(esw), fdb_chain->id); +} + static struct fdb_chain * mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain) { @@ -269,6 +362,10 @@ mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain) fdb_chain->chain = chain; INIT_LIST_HEAD(&fdb_chain->prios_list); + err = create_fdb_chain_restore(fdb_chain); + if (err) + goto err_restore; + err = rhashtable_insert_fast(&esw_chains_ht(esw), &fdb_chain->node, chain_params); if (err) @@ -277,6 +374,8 @@ mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain) return fdb_chain; err_insert: + destroy_fdb_chain_restore(fdb_chain); +err_restore: kvfree(fdb_chain); return ERR_PTR(err); } @@ -288,6 +387,8 @@ mlx5_esw_chains_destroy_fdb_chain(struct fdb_chain *fdb_chain) rhashtable_remove_fast(&esw_chains_ht(esw), &fdb_chain->node, chain_params); + + destroy_fdb_chain_restore(fdb_chain); kvfree(fdb_chain); } @@ -310,10 +411,11 @@ mlx5_esw_chains_get_fdb_chain(struct mlx5_eswitch *esw, u32 chain) } static struct mlx5_flow_handle * -mlx5_esw_chains_add_miss_rule(struct mlx5_flow_table *fdb, +mlx5_esw_chains_add_miss_rule(struct fdb_chain *fdb_chain, + struct mlx5_flow_table *fdb, struct mlx5_flow_table *next_fdb) { - static const struct mlx5_flow_spec spec = {}; + struct mlx5_eswitch *esw = fdb_chain->esw; struct mlx5_flow_destination dest = {}; struct mlx5_flow_act act = {}; @@ -322,7 +424,13 @@ mlx5_esw_chains_add_miss_rule(struct mlx5_flow_table *fdb, dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; dest.ft = next_fdb; - return mlx5_add_flow_rules(fdb, &spec, &act, &dest, 1); + if (next_fdb == tc_end_fdb(esw) && + mlx5_esw_chains_prios_supported(esw)) { + act.modify_hdr = fdb_chain->miss_modify_hdr; + act.action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + } + + return mlx5_add_flow_rules(fdb, NULL, &act, &dest, 1); } static int @@ -345,7 +453,8 @@ mlx5_esw_chains_update_prio_prevs(struct fdb_prio *fdb_prio, list_for_each_entry_continue_reverse(pos, &fdb_chain->prios_list, list) { - miss_rules[n] = mlx5_esw_chains_add_miss_rule(pos->fdb, + miss_rules[n] = mlx5_esw_chains_add_miss_rule(fdb_chain, + pos->fdb, next_fdb); if (IS_ERR(miss_rules[n])) { err = PTR_ERR(miss_rules[n]); @@ -459,7 +568,7 @@ mlx5_esw_chains_create_fdb_prio(struct mlx5_eswitch *esw, } /* Add miss rule to next_fdb */ - miss_rule = mlx5_esw_chains_add_miss_rule(fdb, next_fdb); + miss_rule = mlx5_esw_chains_add_miss_rule(fdb_chain, fdb, next_fdb); if (IS_ERR(miss_rule)) { err = PTR_ERR(miss_rule); goto err_miss_rule; @@ -618,12 +727,44 @@ mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw) return tc_end_fdb(esw); } +struct mlx5_flow_table * +mlx5_esw_chains_create_global_table(struct mlx5_eswitch *esw) +{ + u32 chain, prio, level; + int err; + + if (!fdb_ignore_flow_level_supported(esw)) { + err = -EOPNOTSUPP; + + esw_warn(esw->dev, + "Couldn't create global flow table, ignore_flow_level not supported."); + goto err_ignore; + } + + chain = mlx5_esw_chains_get_chain_range(esw), + prio = mlx5_esw_chains_get_prio_range(esw); + level = mlx5_esw_chains_get_level_range(esw); + + return mlx5_esw_chains_create_fdb_table(esw, chain, prio, level); + +err_ignore: + return ERR_PTR(err); +} + +void +mlx5_esw_chains_destroy_global_table(struct mlx5_eswitch *esw, + struct mlx5_flow_table *ft) +{ + mlx5_esw_chains_destroy_fdb_table(esw, ft); +} + static int mlx5_esw_chains_init(struct mlx5_eswitch *esw) { struct mlx5_esw_chains_priv *chains_priv; struct mlx5_core_dev *dev = esw->dev; u32 max_flow_counter, fdb_max; + struct mapping_ctx *mapping; int err; chains_priv = kzalloc(sizeof(*chains_priv), GFP_KERNEL); @@ -637,7 +778,7 @@ mlx5_esw_chains_init(struct mlx5_eswitch *esw) esw_debug(dev, "Init esw offloads chains, max counters(%d), groups(%d), max flow table size(%d)\n", - max_flow_counter, ESW_OFFLOADS_NUM_GROUPS, fdb_max); + max_flow_counter, esw->params.large_group_num, fdb_max); mlx5_esw_chains_init_sz_pool(esw); @@ -645,6 +786,16 @@ mlx5_esw_chains_init(struct mlx5_eswitch *esw) esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) { esw->fdb_table.flags &= ~ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; esw_warn(dev, "Tc chains and priorities offload aren't supported, update firmware if needed\n"); + } else if (!mlx5_eswitch_reg_c1_loopback_enabled(esw)) { + esw->fdb_table.flags &= ~ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; + esw_warn(dev, "Tc chains and priorities offload aren't supported\n"); + } else if (!fdb_modify_header_fwd_to_table_supported(esw)) { + /* Disabled when ttl workaround is needed, e.g + * when ESWITCH_IPV4_TTL_MODIFY_ENABLE = true in mlxconfig + */ + esw_warn(dev, + "Tc chains and priorities offload aren't supported, check firmware version, or mlxconfig settings\n"); + esw->fdb_table.flags &= ~ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; } else { esw->fdb_table.flags |= ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; esw_info(dev, "Supported tc offload range - chains: %u, prios: %u\n", @@ -660,10 +811,20 @@ mlx5_esw_chains_init(struct mlx5_eswitch *esw) if (err) goto init_prios_ht_err; + mapping = mapping_create(sizeof(u32), esw_get_max_restore_tag(esw), + true); + if (IS_ERR(mapping)) { + err = PTR_ERR(mapping); + goto mapping_err; + } + esw_chains_mapping(esw) = mapping; + mutex_init(&esw_chains_lock(esw)); return 0; +mapping_err: + rhashtable_destroy(&esw_prios_ht(esw)); init_prios_ht_err: rhashtable_destroy(&esw_chains_ht(esw)); init_chains_ht_err: @@ -675,6 +836,7 @@ static void mlx5_esw_chains_cleanup(struct mlx5_eswitch *esw) { mutex_destroy(&esw_chains_lock(esw)); + mapping_destroy(esw_chains_mapping(esw)); rhashtable_destroy(&esw_prios_ht(esw)); rhashtable_destroy(&esw_chains_ht(esw)); @@ -704,12 +866,9 @@ mlx5_esw_chains_open(struct mlx5_eswitch *esw) /* Open level 1 for split rules now if prios isn't supported */ if (!mlx5_esw_chains_prios_supported(esw)) { - ft = mlx5_esw_chains_get_table(esw, 0, 1, 1); - - if (IS_ERR(ft)) { - err = PTR_ERR(ft); + err = mlx5_esw_vport_tbl_get(esw); + if (err) goto level_1_err; - } } return 0; @@ -725,7 +884,7 @@ static void mlx5_esw_chains_close(struct mlx5_eswitch *esw) { if (!mlx5_esw_chains_prios_supported(esw)) - mlx5_esw_chains_put_table(esw, 0, 1, 1); + mlx5_esw_vport_tbl_put(esw); mlx5_esw_chains_put_table(esw, 0, 1, 0); mlx5_esw_chains_put_table(esw, mlx5_esw_chains_get_ft_chain(esw), 1, 0); } @@ -756,3 +915,30 @@ mlx5_esw_chains_destroy(struct mlx5_eswitch *esw) mlx5_esw_chains_close(esw); mlx5_esw_chains_cleanup(esw); } + +int +mlx5_esw_chains_get_chain_mapping(struct mlx5_eswitch *esw, u32 chain, + u32 *chain_mapping) +{ + return mapping_add(esw_chains_mapping(esw), &chain, chain_mapping); +} + +int +mlx5_esw_chains_put_chain_mapping(struct mlx5_eswitch *esw, u32 chain_mapping) +{ + return mapping_remove(esw_chains_mapping(esw), chain_mapping); +} + +int mlx5_eswitch_get_chain_for_tag(struct mlx5_eswitch *esw, u32 tag, + u32 *chain) +{ + int err; + + err = mapping_find(esw_chains_mapping(esw), tag, chain); + if (err) { + esw_warn(esw->dev, "Can't find chain for tag: %d\n", tag); + return -ENOENT; + } + + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h index 2e13097fe348..f8c4239846ea 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h @@ -4,8 +4,12 @@ #ifndef __ML5_ESW_CHAINS_H__ #define __ML5_ESW_CHAINS_H__ +#include "eswitch.h" + bool mlx5_esw_chains_prios_supported(struct mlx5_eswitch *esw); +bool +mlx5_esw_chains_backwards_supported(struct mlx5_eswitch *esw); u32 mlx5_esw_chains_get_prio_range(struct mlx5_eswitch *esw); u32 @@ -23,8 +27,23 @@ mlx5_esw_chains_put_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, struct mlx5_flow_table * mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw); +struct mlx5_flow_table * +mlx5_esw_chains_create_global_table(struct mlx5_eswitch *esw); +void +mlx5_esw_chains_destroy_global_table(struct mlx5_eswitch *esw, + struct mlx5_flow_table *ft); + +int +mlx5_esw_chains_get_chain_mapping(struct mlx5_eswitch *esw, u32 chain, + u32 *chain_mapping); +int +mlx5_esw_chains_put_chain_mapping(struct mlx5_eswitch *esw, + u32 chain_mapping); + int mlx5_esw_chains_create(struct mlx5_eswitch *esw); void mlx5_esw_chains_destroy(struct mlx5_eswitch *esw); -#endif /* __ML5_ESW_CHAINS_H__ */ +int +mlx5_eswitch_get_chain_for_tag(struct mlx5_eswitch *esw, u32 tag, u32 *chain); +#endif /* __ML5_ESW_CHAINS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index e49acd0c5da5..7f618a443bfd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -39,6 +39,7 @@ #include "lib/eq.h" #include "eswitch.h" #include "fs_core.h" +#include "devlink.h" #include "ecpf.h" enum { @@ -1333,7 +1334,6 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, goto out; } - memset(spec, 0, sizeof(*spec)); flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; /* Attach drop flow counter */ @@ -1345,7 +1345,7 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, dest_num++; } vport->ingress.legacy.drop_rule = - mlx5_add_flow_rules(vport->ingress.acl, spec, + mlx5_add_flow_rules(vport->ingress.acl, NULL, &flow_act, dst, dest_num); if (IS_ERR(vport->ingress.legacy.drop_rule)) { err = PTR_ERR(vport->ingress.legacy.drop_rule); @@ -1408,7 +1408,6 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, struct mlx5_flow_destination drop_ctr_dst = {0}; struct mlx5_flow_destination *dst = NULL; struct mlx5_flow_act flow_act = {0}; - struct mlx5_flow_spec *spec; int dest_num = 0; int err = 0; @@ -1437,11 +1436,6 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, if (err) return err; - /* Drop others rule (star rule) */ - spec = kvzalloc(sizeof(*spec), GFP_KERNEL); - if (!spec) - goto out; - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; /* Attach egress drop flow counter */ @@ -1453,7 +1447,7 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, dest_num++; } vport->egress.legacy.drop_rule = - mlx5_add_flow_rules(vport->egress.acl, spec, + mlx5_add_flow_rules(vport->egress.acl, NULL, &flow_act, dst, dest_num); if (IS_ERR(vport->egress.legacy.drop_rule)) { err = PTR_ERR(vport->egress.legacy.drop_rule); @@ -1462,8 +1456,7 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, vport->vport, err); vport->egress.legacy.drop_rule = NULL; } -out: - kvfree(spec); + return err; } @@ -1669,34 +1662,6 @@ static void node_guid_gen_from_mac(u64 *node_guid, u8 mac[ETH_ALEN]) ((u8 *)node_guid)[0] = mac[5]; } -static void esw_apply_vport_conf(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - u16 vport_num = vport->vport; - int flags; - - if (mlx5_esw_is_manager_vport(esw, vport_num)) - return; - - mlx5_modify_vport_admin_state(esw->dev, - MLX5_VPORT_STATE_OP_MOD_ESW_VPORT, - vport_num, 1, - vport->info.link_state); - - /* Host PF has its own mac/guid. */ - if (vport_num) { - mlx5_modify_nic_vport_mac_address(esw->dev, vport_num, - vport->info.mac); - mlx5_modify_nic_vport_node_guid(esw->dev, vport_num, - vport->info.node_guid); - } - - flags = (vport->info.vlan || vport->info.qos) ? - SET_VLAN_STRIP | SET_VLAN_INSERT : 0; - modify_esw_vport_cvlan(esw->dev, vport_num, vport->info.vlan, vport->info.qos, - flags); -} - static int esw_vport_create_legacy_acl_tables(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { @@ -1706,8 +1671,7 @@ static int esw_vport_create_legacy_acl_tables(struct mlx5_eswitch *esw, if (mlx5_esw_is_manager_vport(esw, vport->vport)) return 0; - if (!mlx5_esw_is_manager_vport(esw, vport->vport) && - MLX5_CAP_ESW_INGRESS_ACL(esw->dev, flow_counter)) { + if (MLX5_CAP_ESW_INGRESS_ACL(esw->dev, flow_counter)) { vport->ingress.legacy.drop_counter = mlx5_fc_create(esw->dev, false); if (IS_ERR(vport->ingress.legacy.drop_counter)) { esw_warn(esw->dev, @@ -1721,8 +1685,7 @@ static int esw_vport_create_legacy_acl_tables(struct mlx5_eswitch *esw, if (ret) goto ingress_err; - if (!mlx5_esw_is_manager_vport(esw, vport->vport) && - MLX5_CAP_ESW_EGRESS_ACL(esw->dev, flow_counter)) { + if (MLX5_CAP_ESW_EGRESS_ACL(esw->dev, flow_counter)) { vport->egress.legacy.drop_counter = mlx5_fc_create(esw->dev, false); if (IS_ERR(vport->egress.legacy.drop_counter)) { esw_warn(esw->dev, @@ -1783,29 +1746,75 @@ static void esw_vport_cleanup_acl(struct mlx5_eswitch *esw, esw_vport_destroy_offloads_acl_tables(esw, vport); } -static int esw_enable_vport(struct mlx5_eswitch *esw, struct mlx5_vport *vport, - enum mlx5_eswitch_vport_event enabled_events) +static int esw_vport_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) +{ + u16 vport_num = vport->vport; + int flags; + int err; + + err = esw_vport_setup_acl(esw, vport); + if (err) + return err; + + /* Attach vport to the eswitch rate limiter */ + esw_vport_enable_qos(esw, vport, vport->info.max_rate, vport->qos.bw_share); + + if (mlx5_esw_is_manager_vport(esw, vport_num)) + return 0; + + mlx5_modify_vport_admin_state(esw->dev, + MLX5_VPORT_STATE_OP_MOD_ESW_VPORT, + vport_num, 1, + vport->info.link_state); + + /* Host PF has its own mac/guid. */ + if (vport_num) { + mlx5_modify_nic_vport_mac_address(esw->dev, vport_num, + vport->info.mac); + mlx5_modify_nic_vport_node_guid(esw->dev, vport_num, + vport->info.node_guid); + } + + flags = (vport->info.vlan || vport->info.qos) ? + SET_VLAN_STRIP | SET_VLAN_INSERT : 0; + modify_esw_vport_cvlan(esw->dev, vport_num, vport->info.vlan, + vport->info.qos, flags); + + return 0; +} + +/* Don't cleanup vport->info, it's needed to restore vport configuration */ +static void esw_vport_cleanup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { u16 vport_num = vport->vport; + + if (!mlx5_esw_is_manager_vport(esw, vport_num)) + mlx5_modify_vport_admin_state(esw->dev, + MLX5_VPORT_STATE_OP_MOD_ESW_VPORT, + vport_num, 1, + MLX5_VPORT_ADMIN_STATE_DOWN); + + esw_vport_disable_qos(esw, vport); + esw_vport_cleanup_acl(esw, vport); +} + +static int esw_enable_vport(struct mlx5_eswitch *esw, u16 vport_num, + enum mlx5_eswitch_vport_event enabled_events) +{ + struct mlx5_vport *vport; int ret; + vport = mlx5_eswitch_get_vport(esw, vport_num); + mutex_lock(&esw->state_lock); WARN_ON(vport->enabled); esw_debug(esw->dev, "Enabling VPORT(%d)\n", vport_num); - /* Restore old vport configuration */ - esw_apply_vport_conf(esw, vport); - - ret = esw_vport_setup_acl(esw, vport); + ret = esw_vport_setup(esw, vport); if (ret) goto done; - /* Attach vport to the eswitch rate limiter */ - if (esw_vport_enable_qos(esw, vport, vport->info.max_rate, - vport->qos.bw_share)) - esw_warn(esw->dev, "Failed to attach vport %d to eswitch rate limiter", vport_num); - /* Sync with current vport context */ vport->enabled_events = enabled_events; vport->enabled = true; @@ -1826,10 +1835,11 @@ done: return ret; } -static void esw_disable_vport(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) +static void esw_disable_vport(struct mlx5_eswitch *esw, u16 vport_num) { - u16 vport_num = vport->vport; + struct mlx5_vport *vport; + + vport = mlx5_eswitch_get_vport(esw, vport_num); mutex_lock(&esw->state_lock); if (!vport->enabled) @@ -1847,16 +1857,7 @@ static void esw_disable_vport(struct mlx5_eswitch *esw, */ esw_vport_change_handle_locked(vport); vport->enabled_events = 0; - esw_vport_disable_qos(esw, vport); - - if (!mlx5_esw_is_manager_vport(esw, vport->vport) && - esw->mode == MLX5_ESWITCH_LEGACY) - mlx5_modify_vport_admin_state(esw->dev, - MLX5_VPORT_STATE_OP_MOD_ESW_VPORT, - vport_num, 1, - MLX5_VPORT_ADMIN_STATE_DOWN); - - esw_vport_cleanup_acl(esw, vport); + esw_vport_cleanup(esw, vport); esw->enabled_vports--; done: @@ -1944,6 +1945,59 @@ static void mlx5_eswitch_clear_vf_vports_info(struct mlx5_eswitch *esw) /* Public E-Switch API */ #define ESW_ALLOWED(esw) ((esw) && MLX5_ESWITCH_MANAGER((esw)->dev)) +int mlx5_eswitch_load_vport(struct mlx5_eswitch *esw, u16 vport_num, + enum mlx5_eswitch_vport_event enabled_events) +{ + int err; + + err = esw_enable_vport(esw, vport_num, enabled_events); + if (err) + return err; + + err = esw_offloads_load_rep(esw, vport_num); + if (err) + goto err_rep; + + return err; + +err_rep: + esw_disable_vport(esw, vport_num); + return err; +} + +void mlx5_eswitch_unload_vport(struct mlx5_eswitch *esw, u16 vport_num) +{ + esw_offloads_unload_rep(esw, vport_num); + esw_disable_vport(esw, vport_num); +} + +void mlx5_eswitch_unload_vf_vports(struct mlx5_eswitch *esw, u16 num_vfs) +{ + int i; + + mlx5_esw_for_each_vf_vport_num_reverse(esw, i, num_vfs) + mlx5_eswitch_unload_vport(esw, i); +} + +int mlx5_eswitch_load_vf_vports(struct mlx5_eswitch *esw, u16 num_vfs, + enum mlx5_eswitch_vport_event enabled_events) +{ + int err; + int i; + + mlx5_esw_for_each_vf_vport_num(esw, i, num_vfs) { + err = mlx5_eswitch_load_vport(esw, i, enabled_events); + if (err) + goto vf_err; + } + + return 0; + +vf_err: + mlx5_eswitch_unload_vf_vports(esw, i - 1); + return err; +} + /* mlx5_eswitch_enable_pf_vf_vports() enables vports of PF, ECPF and VFs * whichever are present on the eswitch. */ @@ -1951,46 +2005,33 @@ int mlx5_eswitch_enable_pf_vf_vports(struct mlx5_eswitch *esw, enum mlx5_eswitch_vport_event enabled_events) { - struct mlx5_vport *vport; - int num_vfs; int ret; - int i; /* Enable PF vport */ - vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_PF); - ret = esw_enable_vport(esw, vport, enabled_events); + ret = mlx5_eswitch_load_vport(esw, MLX5_VPORT_PF, enabled_events); if (ret) return ret; /* Enable ECPF vport */ if (mlx5_ecpf_vport_exists(esw->dev)) { - vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_ECPF); - ret = esw_enable_vport(esw, vport, enabled_events); + ret = mlx5_eswitch_load_vport(esw, MLX5_VPORT_ECPF, enabled_events); if (ret) goto ecpf_err; } /* Enable VF vports */ - mlx5_esw_for_each_vf_vport(esw, i, vport, esw->esw_funcs.num_vfs) { - ret = esw_enable_vport(esw, vport, enabled_events); - if (ret) - goto vf_err; - } + ret = mlx5_eswitch_load_vf_vports(esw, esw->esw_funcs.num_vfs, + enabled_events); + if (ret) + goto vf_err; return 0; vf_err: - num_vfs = i - 1; - mlx5_esw_for_each_vf_vport_reverse(esw, i, vport, num_vfs) - esw_disable_vport(esw, vport); - - if (mlx5_ecpf_vport_exists(esw->dev)) { - vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_ECPF); - esw_disable_vport(esw, vport); - } + if (mlx5_ecpf_vport_exists(esw->dev)) + mlx5_eswitch_unload_vport(esw, MLX5_VPORT_ECPF); ecpf_err: - vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_PF); - esw_disable_vport(esw, vport); + mlx5_eswitch_unload_vport(esw, MLX5_VPORT_PF); return ret; } @@ -1999,19 +2040,81 @@ ecpf_err: */ void mlx5_eswitch_disable_pf_vf_vports(struct mlx5_eswitch *esw) { - struct mlx5_vport *vport; - int i; + mlx5_eswitch_unload_vf_vports(esw, esw->esw_funcs.num_vfs); - mlx5_esw_for_all_vports_reverse(esw, i, vport) - esw_disable_vport(esw, vport); + if (mlx5_ecpf_vport_exists(esw->dev)) + mlx5_eswitch_unload_vport(esw, MLX5_VPORT_ECPF); + + mlx5_eswitch_unload_vport(esw, MLX5_VPORT_PF); } -int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int mode) +static void mlx5_eswitch_get_devlink_param(struct mlx5_eswitch *esw) { + struct devlink *devlink = priv_to_devlink(esw->dev); + union devlink_param_value val; int err; - if (!ESW_ALLOWED(esw) || - !MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ft_support)) { + err = devlink_param_driverinit_value_get(devlink, + MLX5_DEVLINK_PARAM_ID_ESW_LARGE_GROUP_NUM, + &val); + if (!err) { + esw->params.large_group_num = val.vu32; + } else { + esw_warn(esw->dev, + "Devlink can't get param fdb_large_groups, uses default (%d).\n", + ESW_OFFLOADS_DEFAULT_NUM_GROUPS); + esw->params.large_group_num = ESW_OFFLOADS_DEFAULT_NUM_GROUPS; + } +} + +static void +mlx5_eswitch_update_num_of_vfs(struct mlx5_eswitch *esw, int num_vfs) +{ + const u32 *out; + + WARN_ON_ONCE(esw->mode != MLX5_ESWITCH_NONE); + + if (num_vfs < 0) + return; + + if (!mlx5_core_is_ecpf_esw_manager(esw->dev)) { + esw->esw_funcs.num_vfs = num_vfs; + return; + } + + out = mlx5_esw_query_functions(esw->dev); + if (IS_ERR(out)) + return; + + esw->esw_funcs.num_vfs = MLX5_GET(query_esw_functions_out, out, + host_params_context.host_num_of_vfs); + kvfree(out); +} + +/** + * mlx5_eswitch_enable_locked - Enable eswitch + * @esw: Pointer to eswitch + * @mode: Eswitch mode to enable + * @num_vfs: Enable eswitch for given number of VFs. This is optional. + * Valid value are 0, > 0 and MLX5_ESWITCH_IGNORE_NUM_VFS. + * Caller should pass num_vfs > 0 when enabling eswitch for + * vf vports. Caller should pass num_vfs = 0, when eswitch + * is enabled without sriov VFs or when caller + * is unaware of the sriov state of the host PF on ECPF based + * eswitch. Caller should pass < 0 when num_vfs should be + * completely ignored. This is typically the case when eswitch + * is enabled without sriov regardless of PF/ECPF system. + * mlx5_eswitch_enable_locked() Enables eswitch in either legacy or offloads + * mode. If num_vfs >=0 is provided, it setup VF related eswitch vports. + * It returns 0 on success or error code on failure. + */ +int mlx5_eswitch_enable_locked(struct mlx5_eswitch *esw, int mode, int num_vfs) +{ + int err; + + lockdep_assert_held(&esw->mode_lock); + + if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ft_support)) { esw_warn(esw->dev, "FDB is not supported, aborting ...\n"); return -EOPNOTSUPP; } @@ -2022,6 +2125,10 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int mode) if (!MLX5_CAP_ESW_EGRESS_ACL(esw->dev, ft_support)) esw_warn(esw->dev, "engress ACL is not supported by FW\n"); + mlx5_eswitch_get_devlink_param(esw); + + mlx5_eswitch_update_num_of_vfs(esw, num_vfs); + esw_create_tsar(esw); esw->mode = mode; @@ -2058,11 +2165,34 @@ abort: return err; } -void mlx5_eswitch_disable(struct mlx5_eswitch *esw, bool clear_vf) +/** + * mlx5_eswitch_enable - Enable eswitch + * @esw: Pointer to eswitch + * @num_vfs: Enable eswitch swich for given number of VFs. + * Caller must pass num_vfs > 0 when enabling eswitch for + * vf vports. + * mlx5_eswitch_enable() returns 0 on success or error code on failure. + */ +int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs) +{ + int ret; + + if (!ESW_ALLOWED(esw)) + return 0; + + mutex_lock(&esw->mode_lock); + ret = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_LEGACY, num_vfs); + mutex_unlock(&esw->mode_lock); + return ret; +} + +void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw, bool clear_vf) { int old_mode; - if (!ESW_ALLOWED(esw) || esw->mode == MLX5_ESWITCH_NONE) + lockdep_assert_held_write(&esw->mode_lock); + + if (esw->mode == MLX5_ESWITCH_NONE) return; esw_info(esw->dev, "Disable: mode(%s), nvfs(%d), active vports(%d)\n", @@ -2091,6 +2221,16 @@ void mlx5_eswitch_disable(struct mlx5_eswitch *esw, bool clear_vf) mlx5_eswitch_clear_vf_vports_info(esw); } +void mlx5_eswitch_disable(struct mlx5_eswitch *esw, bool clear_vf) +{ + if (!ESW_ALLOWED(esw)) + return; + + mutex_lock(&esw->mode_lock); + mlx5_eswitch_disable_locked(esw, clear_vf); + mutex_unlock(&esw->mode_lock); +} + int mlx5_eswitch_init(struct mlx5_core_dev *dev) { struct mlx5_eswitch *esw; @@ -2142,6 +2282,7 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) hash_init(esw->offloads.mod_hdr.hlist); atomic64_set(&esw->offloads.num_flows, 0); mutex_init(&esw->state_lock); + mutex_init(&esw->mode_lock); mlx5_esw_for_all_vports(esw, i, vport) { vport->vport = mlx5_eswitch_index_to_vport_num(esw, i); @@ -2176,6 +2317,8 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) esw->dev->priv.eswitch = NULL; destroy_workqueue(esw->work_queue); esw_offloads_cleanup_reps(esw); + mutex_destroy(&esw->mode_lock); + mutex_destroy(&esw->state_lock); mutex_destroy(&esw->offloads.mod_hdr.lock); mutex_destroy(&esw->offloads.encap_tbl_lock); kfree(esw->vports); @@ -2410,12 +2553,11 @@ static int _mlx5_eswitch_set_vepa_locked(struct mlx5_eswitch *esw, } /* Star rule to forward all traffic to uplink vport */ - memset(spec, 0, sizeof(*spec)); memset(&dest, 0, sizeof(dest)); dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; dest.vport.num = MLX5_VPORT_UPLINK; flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - flow_rule = mlx5_add_flow_rules(esw->fdb_table.legacy.vepa_fdb, spec, + flow_rule = mlx5_add_flow_rules(esw->fdb_table.legacy.vepa_fdb, NULL, &flow_act, &dest, 1); if (IS_ERR(flow_rule)) { err = PTR_ERR(flow_rule); @@ -2600,9 +2742,13 @@ static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev, u64 bytes = 0; int err = 0; - if (!vport->enabled || esw->mode != MLX5_ESWITCH_LEGACY) + if (esw->mode != MLX5_ESWITCH_LEGACY) return 0; + mutex_lock(&esw->state_lock); + if (!vport->enabled) + goto unlock; + if (vport->egress.legacy.drop_counter) mlx5_fc_query(dev, vport->egress.legacy.drop_counter, &stats->rx_dropped, &bytes); @@ -2613,20 +2759,22 @@ static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev, if (!MLX5_CAP_GEN(dev, receive_discard_vport_down) && !MLX5_CAP_GEN(dev, transmit_discard_vport_down)) - return 0; + goto unlock; err = mlx5_query_vport_down_stats(dev, vport->vport, 1, &rx_discard_vport_down, &tx_discard_vport_down); if (err) - return err; + goto unlock; if (MLX5_CAP_GEN(dev, receive_discard_vport_down)) stats->rx_dropped += rx_discard_vport_down; if (MLX5_CAP_GEN(dev, transmit_discard_vport_down)) stats->tx_dropped += tx_discard_vport_down; - return 0; +unlock: + mutex_unlock(&esw->state_lock); + return err; } int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, @@ -2742,22 +2890,4 @@ bool mlx5_esw_multipath_prereq(struct mlx5_core_dev *dev0, dev1->priv.eswitch->mode == MLX5_ESWITCH_OFFLOADS); } -void mlx5_eswitch_update_num_of_vfs(struct mlx5_eswitch *esw, const int num_vfs) -{ - const u32 *out; - - WARN_ON_ONCE(esw->mode != MLX5_ESWITCH_NONE); - - if (!mlx5_core_is_ecpf_esw_manager(esw->dev)) { - esw->esw_funcs.num_vfs = num_vfs; - return; - } - - out = mlx5_esw_query_functions(esw->dev); - if (IS_ERR(out)) - return; - esw->esw_funcs.num_vfs = MLX5_GET(query_esw_functions_out, out, - host_params_context.host_num_of_vfs); - kvfree(out); -} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 4472710ccc9c..39f42f985fbd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -42,6 +42,7 @@ #include <linux/mlx5/vport.h> #include <linux/mlx5/fs.h> #include "lib/mpfs.h" +#include "en/tc_ct.h" #define FDB_TC_MAX_CHAIN 3 #define FDB_FT_CHAIN (FDB_TC_MAX_CHAIN + 1) @@ -55,6 +56,8 @@ #ifdef CONFIG_MLX5_ESWITCH +#define ESW_OFFLOADS_DEFAULT_NUM_GROUPS 15 + #define MLX5_MAX_UC_PER_VPORT(dev) \ (1 << MLX5_CAP_GEN(dev, log_max_current_uc_list)) @@ -183,12 +186,22 @@ struct mlx5_eswitch_fdb { int vlan_push_pop_refcount; struct mlx5_esw_chains_priv *esw_chains_priv; + struct { + DECLARE_HASHTABLE(table, 8); + /* Protects vports.table */ + struct mutex lock; + } vports; + } offloads; }; u32 flags; }; struct mlx5_esw_offload { + struct mlx5_flow_table *ft_offloads_restore; + struct mlx5_flow_group *restore_group; + struct mlx5_modify_hdr *restore_copy_hdr_id; + struct mlx5_flow_table *ft_offloads; struct mlx5_flow_group *vport_rx_group; struct mlx5_eswitch_rep *vport_reps; @@ -224,6 +237,7 @@ struct mlx5_esw_functions { enum { MLX5_ESWITCH_VPORT_MATCH_METADATA = BIT(0), + MLX5_ESWITCH_REG_C1_LOOPBACK_ENABLED = BIT(1), }; struct mlx5_eswitch { @@ -244,6 +258,11 @@ struct mlx5_eswitch { */ struct mutex state_lock; + /* Protects eswitch mode change that occurs via one or more + * user commands, i.e. sriov state change, devlink commands. + */ + struct mutex mode_lock; + struct { bool enabled; u32 root_tsar_id; @@ -255,6 +274,9 @@ struct mlx5_eswitch { u16 manager_vport; u16 first_host_vport; struct mlx5_esw_functions esw_funcs; + struct { + u32 large_group_num; + } params; }; void esw_offloads_disable(struct mlx5_eswitch *esw); @@ -279,7 +301,11 @@ int mlx5_esw_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num, /* E-Switch API */ int mlx5_eswitch_init(struct mlx5_core_dev *dev); void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw); -int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int mode); + +#define MLX5_ESWITCH_IGNORE_NUM_VFS (-1) +int mlx5_eswitch_enable_locked(struct mlx5_eswitch *esw, int mode, int num_vfs); +int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs); +void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw, bool clear_vf); void mlx5_eswitch_disable(struct mlx5_eswitch *esw, bool clear_vf); int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, u16 vport, u8 mac[ETH_ALEN]); @@ -315,6 +341,7 @@ struct mlx5_termtbl_handle; bool mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw, + struct mlx5_esw_flow_attr *attr, struct mlx5_flow_act *flow_act, struct mlx5_flow_spec *spec); @@ -375,6 +402,8 @@ enum { enum { MLX5_ESW_ATTR_FLAG_VLAN_HANDLED = BIT(0), MLX5_ESW_ATTR_FLAG_SLOW_PATH = BIT(1), + MLX5_ESW_ATTR_FLAG_NO_IN_PORT = BIT(2), + MLX5_ESW_ATTR_FLAG_HAIRPIN = BIT(3), }; struct mlx5_esw_flow_attr { @@ -405,6 +434,9 @@ struct mlx5_esw_flow_attr { u16 prio; u32 dest_chain; u32 flags; + struct mlx5_flow_table *fdb; + struct mlx5_flow_table *dest_ft; + struct mlx5_ct_attr ct_attr; struct mlx5e_tc_flow_parse_attr *parse_attr; }; @@ -414,7 +446,6 @@ int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode); int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode, struct netlink_ext_ack *extack); int mlx5_devlink_eswitch_inline_mode_get(struct devlink *devlink, u8 *mode); -int mlx5_eswitch_inline_mode_get(struct mlx5_eswitch *esw, u8 *mode); int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink, enum devlink_eswitch_encap_mode encap, struct netlink_ext_ack *extack); @@ -433,6 +464,11 @@ int mlx5_esw_create_vport_egress_acl_vlan(struct mlx5_eswitch *esw, struct mlx5_vport *vport, u16 vlan_id, u32 flow_action); +static inline bool mlx5_esw_qos_enabled(struct mlx5_eswitch *esw) +{ + return esw->qos.enabled; +} + static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev, u8 vlan_depth) { @@ -608,7 +644,6 @@ mlx5_eswitch_get_vport(struct mlx5_eswitch *esw, u16 vport_num); bool mlx5_eswitch_is_vf_vport(const struct mlx5_eswitch *esw, u16 vport_num); -void mlx5_eswitch_update_num_of_vfs(struct mlx5_eswitch *esw, const int num_vfs); int mlx5_esw_funcs_changed_handler(struct notifier_block *nb, unsigned long type, void *data); int @@ -623,11 +658,30 @@ void esw_vport_destroy_offloads_acl_tables(struct mlx5_eswitch *esw, struct mlx5_vport *vport); +int mlx5_esw_vport_tbl_get(struct mlx5_eswitch *esw); +void mlx5_esw_vport_tbl_put(struct mlx5_eswitch *esw); + +struct mlx5_flow_handle * +esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag); +u32 +esw_get_max_restore_tag(struct mlx5_eswitch *esw); + +int esw_offloads_load_rep(struct mlx5_eswitch *esw, u16 vport_num); +void esw_offloads_unload_rep(struct mlx5_eswitch *esw, u16 vport_num); + +int mlx5_eswitch_load_vport(struct mlx5_eswitch *esw, u16 vport_num, + enum mlx5_eswitch_vport_event enabled_events); +void mlx5_eswitch_unload_vport(struct mlx5_eswitch *esw, u16 vport_num); + +int mlx5_eswitch_load_vf_vports(struct mlx5_eswitch *esw, u16 num_vfs, + enum mlx5_eswitch_vport_event enabled_events); +void mlx5_eswitch_unload_vf_vports(struct mlx5_eswitch *esw, u16 num_vfs); + #else /* CONFIG_MLX5_ESWITCH */ /* eswitch API stubs */ static inline int mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; } static inline void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) {} -static inline int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int mode) { return 0; } +static inline int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs) { return 0; } static inline void mlx5_eswitch_disable(struct mlx5_eswitch *esw, bool clear_vf) {} static inline bool mlx5_esw_lag_prereq(struct mlx5_core_dev *dev0, struct mlx5_core_dev *dev1) { return true; } static inline bool mlx5_eswitch_is_funcs_handler(struct mlx5_core_dev *dev) { return false; } @@ -636,8 +690,11 @@ static inline const u32 *mlx5_esw_query_functions(struct mlx5_core_dev *dev) return ERR_PTR(-EOPNOTSUPP); } -static inline void mlx5_eswitch_update_num_of_vfs(struct mlx5_eswitch *esw, const int num_vfs) {} - +static inline struct mlx5_flow_handle * +esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag) +{ + return ERR_PTR(-EOPNOTSUPP); +} #endif /* CONFIG_MLX5_ESWITCH */ #endif /* __MLX5_ESWITCH_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 1a57b2bd74b8..f171eb2234b0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -37,7 +37,7 @@ #include <linux/mlx5/fs.h> #include "mlx5_core.h" #include "eswitch.h" -#include "eswitch_offloads_chains.h" +#include "esw/chains.h" #include "rdma.h" #include "en.h" #include "fs_core.h" @@ -50,6 +50,181 @@ #define MLX5_ESW_MISS_FLOWS (2) #define UPLINK_REP_INDEX 0 +/* Per vport tables */ + +#define MLX5_ESW_VPORT_TABLE_SIZE 128 + +/* This struct is used as a key to the hash table and we need it to be packed + * so hash result is consistent + */ +struct mlx5_vport_key { + u32 chain; + u16 prio; + u16 vport; + u16 vhca_id; +} __packed; + +struct mlx5_vport_table { + struct hlist_node hlist; + struct mlx5_flow_table *fdb; + u32 num_rules; + struct mlx5_vport_key key; +}; + +#define MLX5_ESW_VPORT_TBL_NUM_GROUPS 4 + +static struct mlx5_flow_table * +esw_vport_tbl_create(struct mlx5_eswitch *esw, struct mlx5_flow_namespace *ns) +{ + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_flow_table *fdb; + + ft_attr.autogroup.max_num_groups = MLX5_ESW_VPORT_TBL_NUM_GROUPS; + ft_attr.max_fte = MLX5_ESW_VPORT_TABLE_SIZE; + ft_attr.prio = FDB_PER_VPORT; + fdb = mlx5_create_auto_grouped_flow_table(ns, &ft_attr); + if (IS_ERR(fdb)) { + esw_warn(esw->dev, "Failed to create per vport FDB Table err %ld\n", + PTR_ERR(fdb)); + } + + return fdb; +} + +static u32 flow_attr_to_vport_key(struct mlx5_eswitch *esw, + struct mlx5_esw_flow_attr *attr, + struct mlx5_vport_key *key) +{ + key->vport = attr->in_rep->vport; + key->chain = attr->chain; + key->prio = attr->prio; + key->vhca_id = MLX5_CAP_GEN(esw->dev, vhca_id); + return jhash(key, sizeof(*key), 0); +} + +/* caller must hold vports.lock */ +static struct mlx5_vport_table * +esw_vport_tbl_lookup(struct mlx5_eswitch *esw, struct mlx5_vport_key *skey, u32 key) +{ + struct mlx5_vport_table *e; + + hash_for_each_possible(esw->fdb_table.offloads.vports.table, e, hlist, key) + if (!memcmp(&e->key, skey, sizeof(*skey))) + return e; + + return NULL; +} + +static void +esw_vport_tbl_put(struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *attr) +{ + struct mlx5_vport_table *e; + struct mlx5_vport_key key; + u32 hkey; + + mutex_lock(&esw->fdb_table.offloads.vports.lock); + hkey = flow_attr_to_vport_key(esw, attr, &key); + e = esw_vport_tbl_lookup(esw, &key, hkey); + if (!e || --e->num_rules) + goto out; + + hash_del(&e->hlist); + mlx5_destroy_flow_table(e->fdb); + kfree(e); +out: + mutex_unlock(&esw->fdb_table.offloads.vports.lock); +} + +static struct mlx5_flow_table * +esw_vport_tbl_get(struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *attr) +{ + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *fdb; + struct mlx5_vport_table *e; + struct mlx5_vport_key skey; + u32 hkey; + + mutex_lock(&esw->fdb_table.offloads.vports.lock); + hkey = flow_attr_to_vport_key(esw, attr, &skey); + e = esw_vport_tbl_lookup(esw, &skey, hkey); + if (e) { + e->num_rules++; + goto out; + } + + e = kzalloc(sizeof(*e), GFP_KERNEL); + if (!e) { + fdb = ERR_PTR(-ENOMEM); + goto err_alloc; + } + + ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB); + if (!ns) { + esw_warn(dev, "Failed to get FDB namespace\n"); + fdb = ERR_PTR(-ENOENT); + goto err_ns; + } + + fdb = esw_vport_tbl_create(esw, ns); + if (IS_ERR(fdb)) + goto err_ns; + + e->fdb = fdb; + e->num_rules = 1; + e->key = skey; + hash_add(esw->fdb_table.offloads.vports.table, &e->hlist, hkey); +out: + mutex_unlock(&esw->fdb_table.offloads.vports.lock); + return e->fdb; + +err_ns: + kfree(e); +err_alloc: + mutex_unlock(&esw->fdb_table.offloads.vports.lock); + return fdb; +} + +int mlx5_esw_vport_tbl_get(struct mlx5_eswitch *esw) +{ + struct mlx5_esw_flow_attr attr = {}; + struct mlx5_eswitch_rep rep = {}; + struct mlx5_flow_table *fdb; + struct mlx5_vport *vport; + int i; + + attr.prio = 1; + attr.in_rep = &rep; + mlx5_esw_for_all_vports(esw, i, vport) { + attr.in_rep->vport = vport->vport; + fdb = esw_vport_tbl_get(esw, &attr); + if (IS_ERR(fdb)) + goto out; + } + return 0; + +out: + mlx5_esw_vport_tbl_put(esw); + return PTR_ERR(fdb); +} + +void mlx5_esw_vport_tbl_put(struct mlx5_eswitch *esw) +{ + struct mlx5_esw_flow_attr attr = {}; + struct mlx5_eswitch_rep rep = {}; + struct mlx5_vport *vport; + int i; + + attr.prio = 1; + attr.in_rep = &rep; + mlx5_esw_for_all_vports(esw, i, vport) { + attr.in_rep->vport = vport->vport; + esw_vport_tbl_put(esw, &attr); + } +} + +/* End: Per vport tables */ + static struct mlx5_eswitch_rep *mlx5_eswitch_get_rep(struct mlx5_eswitch *esw, u16 vport_num) { @@ -85,7 +260,8 @@ mlx5_eswitch_set_rule_source_port(struct mlx5_eswitch *esw, attr->in_rep->vport)); misc2 = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_2); - MLX5_SET_TO_ONES(fte_match_set_misc2, misc2, metadata_reg_c_0); + MLX5_SET(fte_match_set_misc2, misc2, metadata_reg_c_0, + mlx5_eswitch_get_vport_metadata_mask()); spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2; misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters); @@ -124,6 +300,7 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, bool split = !!(attr->split_count); struct mlx5_flow_handle *rule; struct mlx5_flow_table *fdb; + bool hairpin = false; int j, i = 0; if (esw->mode != MLX5_ESWITCH_OFFLOADS) @@ -148,7 +325,12 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { struct mlx5_flow_table *ft; - if (attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH) { + if (attr->dest_ft) { + flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; + dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest[i].ft = attr->dest_ft; + i++; + } else if (attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH) { flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; dest[i].ft = mlx5_esw_chains_get_tc_end_ft(esw); @@ -191,8 +373,6 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, i++; } - mlx5_eswitch_set_rule_source_port(esw, spec, attr); - if (attr->outer_match_level != MLX5_MATCH_NONE) spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS; if (attr->inner_match_level != MLX5_MATCH_NONE) @@ -201,27 +381,45 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) flow_act.modify_hdr = attr->modify_hdr; - fdb = mlx5_esw_chains_get_table(esw, attr->chain, attr->prio, - !!split); + if (split) { + fdb = esw_vport_tbl_get(esw, attr); + } else { + if (attr->chain || attr->prio) + fdb = mlx5_esw_chains_get_table(esw, attr->chain, + attr->prio, 0); + else + fdb = attr->fdb; + + if (!(attr->flags & MLX5_ESW_ATTR_FLAG_NO_IN_PORT)) + mlx5_eswitch_set_rule_source_port(esw, spec, attr); + } if (IS_ERR(fdb)) { rule = ERR_CAST(fdb); goto err_esw_get; } - if (mlx5_eswitch_termtbl_required(esw, &flow_act, spec)) + if (mlx5_eswitch_termtbl_required(esw, attr, &flow_act, spec)) { rule = mlx5_eswitch_add_termtbl_rule(esw, fdb, spec, attr, &flow_act, dest, i); - else + hairpin = true; + } else { rule = mlx5_add_flow_rules(fdb, spec, &flow_act, dest, i); + } if (IS_ERR(rule)) goto err_add_rule; else atomic64_inc(&esw->offloads.num_flows); + if (hairpin) + attr->flags |= MLX5_ESW_ATTR_FLAG_HAIRPIN; + return rule; err_add_rule: - mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, !!split); + if (split) + esw_vport_tbl_put(esw, attr); + else if (attr->chain || attr->prio) + mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 0); err_esw_get: if (!(attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH) && attr->dest_chain) mlx5_esw_chains_put_table(esw, attr->dest_chain, 1, 0); @@ -247,7 +445,7 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw, goto err_get_fast; } - fwd_fdb = mlx5_esw_chains_get_table(esw, attr->chain, attr->prio, 1); + fwd_fdb = esw_vport_tbl_get(esw, attr); if (IS_ERR(fwd_fdb)) { rule = ERR_CAST(fwd_fdb); goto err_get_fwd; @@ -285,7 +483,7 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw, return rule; add_err: - mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 1); + esw_vport_tbl_put(esw, attr); err_get_fwd: mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 0); err_get_fast: @@ -303,20 +501,25 @@ __mlx5_eswitch_del_rule(struct mlx5_eswitch *esw, mlx5_del_flow_rules(rule); - /* unref the term table */ - for (i = 0; i < MLX5_MAX_FLOW_FWD_VPORTS; i++) { - if (attr->dests[i].termtbl) - mlx5_eswitch_termtbl_put(esw, attr->dests[i].termtbl); + if (attr->flags & MLX5_ESW_ATTR_FLAG_HAIRPIN) { + /* unref the term table */ + for (i = 0; i < MLX5_MAX_FLOW_FWD_VPORTS; i++) { + if (attr->dests[i].termtbl) + mlx5_eswitch_termtbl_put(esw, attr->dests[i].termtbl); + } } atomic64_dec(&esw->offloads.num_flows); if (fwd_rule) { - mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 1); + esw_vport_tbl_put(esw, attr); mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 0); } else { - mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, - !!split); + if (split) + esw_vport_tbl_put(esw, attr); + else if (attr->chain || attr->prio) + mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, + 0); if (attr->dest_chain) mlx5_esw_chains_put_table(esw, attr->dest_chain, 1, 0); } @@ -578,14 +781,21 @@ void mlx5_eswitch_del_send_to_vport_rule(struct mlx5_flow_handle *rule) mlx5_del_flow_rules(rule); } +static bool mlx5_eswitch_reg_c1_loopback_supported(struct mlx5_eswitch *esw) +{ + return MLX5_CAP_ESW_FLOWTABLE(esw->dev, fdb_to_vport_reg_c_id) & + MLX5_FDB_TO_VPORT_REG_C_1; +} + static int esw_set_passing_vport_metadata(struct mlx5_eswitch *esw, bool enable) { u32 out[MLX5_ST_SZ_DW(query_esw_vport_context_out)] = {}; u32 in[MLX5_ST_SZ_DW(modify_esw_vport_context_in)] = {}; - u8 fdb_to_vport_reg_c_id; + u8 curr, wanted; int err; - if (!mlx5_eswitch_vport_match_metadata_enabled(esw)) + if (!mlx5_eswitch_reg_c1_loopback_supported(esw) && + !mlx5_eswitch_vport_match_metadata_enabled(esw)) return 0; err = mlx5_eswitch_query_esw_vport_context(esw->dev, 0, false, @@ -593,22 +803,33 @@ static int esw_set_passing_vport_metadata(struct mlx5_eswitch *esw, bool enable) if (err) return err; - fdb_to_vport_reg_c_id = MLX5_GET(query_esw_vport_context_out, out, - esw_vport_context.fdb_to_vport_reg_c_id); + curr = MLX5_GET(query_esw_vport_context_out, out, + esw_vport_context.fdb_to_vport_reg_c_id); + wanted = MLX5_FDB_TO_VPORT_REG_C_0; + if (mlx5_eswitch_reg_c1_loopback_supported(esw)) + wanted |= MLX5_FDB_TO_VPORT_REG_C_1; if (enable) - fdb_to_vport_reg_c_id |= MLX5_FDB_TO_VPORT_REG_C_0; + curr |= wanted; else - fdb_to_vport_reg_c_id &= ~MLX5_FDB_TO_VPORT_REG_C_0; + curr &= ~wanted; MLX5_SET(modify_esw_vport_context_in, in, - esw_vport_context.fdb_to_vport_reg_c_id, fdb_to_vport_reg_c_id); + esw_vport_context.fdb_to_vport_reg_c_id, curr); MLX5_SET(modify_esw_vport_context_in, in, field_select.fdb_to_vport_reg_c_id, 1); - return mlx5_eswitch_modify_esw_vport_context(esw->dev, 0, false, - in, sizeof(in)); + err = mlx5_eswitch_modify_esw_vport_context(esw->dev, 0, false, in, + sizeof(in)); + if (!err) { + if (enable && (curr & MLX5_FDB_TO_VPORT_REG_C_1)) + esw->flags |= MLX5_ESWITCH_REG_C1_LOOPBACK_ENABLED; + else + esw->flags &= ~MLX5_ESWITCH_REG_C1_LOOPBACK_ENABLED; + } + + return err; } static void peer_miss_rules_setup(struct mlx5_eswitch *esw, @@ -621,7 +842,8 @@ static void peer_miss_rules_setup(struct mlx5_eswitch *esw, if (mlx5_eswitch_vport_match_metadata_enabled(esw)) { misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_2); - MLX5_SET_TO_ONES(fte_match_set_misc2, misc, metadata_reg_c_0); + MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0, + mlx5_eswitch_get_vport_metadata_mask()); spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2; } else { @@ -836,6 +1058,59 @@ out: return err; } +struct mlx5_flow_handle * +esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag) +{ + struct mlx5_flow_act flow_act = { .flags = FLOW_ACT_NO_APPEND, }; + struct mlx5_flow_table *ft = esw->offloads.ft_offloads_restore; + struct mlx5_flow_context *flow_context; + struct mlx5_flow_handle *flow_rule; + struct mlx5_flow_destination dest; + struct mlx5_flow_spec *spec; + void *misc; + + if (!mlx5_eswitch_reg_c1_loopback_supported(esw)) + return ERR_PTR(-EOPNOTSUPP); + + spec = kzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) + return ERR_PTR(-ENOMEM); + + misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + misc_parameters_2); + MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0, + ESW_CHAIN_TAG_METADATA_MASK); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, + misc_parameters_2); + MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0, tag); + spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2; + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | + MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + flow_act.modify_hdr = esw->offloads.restore_copy_hdr_id; + + flow_context = &spec->flow_context; + flow_context->flags |= FLOW_CONTEXT_HAS_TAG; + flow_context->flow_tag = tag; + dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest.ft = esw->offloads.ft_offloads; + + flow_rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1); + kfree(spec); + + if (IS_ERR(flow_rule)) + esw_warn(esw->dev, + "Failed to create restore rule for tag: %d, err(%d)\n", + tag, (int)PTR_ERR(flow_rule)); + + return flow_rule; +} + +u32 +esw_get_max_restore_tag(struct mlx5_eswitch *esw) +{ + return ESW_CHAIN_TAG_METADATA_MASK; +} + #define MAX_PF_SQ 256 #define MAX_SQ_NVPORTS 32 @@ -851,8 +1126,9 @@ static void esw_set_flow_group_source_port(struct mlx5_eswitch *esw, match_criteria_enable, MLX5_MATCH_MISC_PARAMETERS_2); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, - misc_parameters_2.metadata_reg_c_0); + MLX5_SET(fte_match_param, match_criteria, + misc_parameters_2.metadata_reg_c_0, + mlx5_eswitch_get_vport_metadata_mask()); } else { MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, @@ -1057,6 +1333,7 @@ static int esw_create_offloads_table(struct mlx5_eswitch *esw, int nvports) } ft_attr.max_fte = nvports + MLX5_ESW_MISS_FLOWS; + ft_attr.prio = 1; ft_offloads = mlx5_create_flow_table(ns, &ft_attr); if (IS_ERR(ft_offloads)) { @@ -1134,7 +1411,8 @@ mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, u16 vport, mlx5_eswitch_get_vport_metadata_for_match(esw, vport)); misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_2); - MLX5_SET_TO_ONES(fte_match_set_misc2, misc, metadata_reg_c_0); + MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0, + mlx5_eswitch_get_vport_metadata_mask()); spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2; } else { @@ -1160,6 +1438,148 @@ out: return flow_rule; } + +static int mlx5_eswitch_inline_mode_get(const struct mlx5_eswitch *esw, u8 *mode) +{ + u8 prev_mlx5_mode, mlx5_mode = MLX5_INLINE_MODE_L2; + struct mlx5_core_dev *dev = esw->dev; + int vport; + + if (!MLX5_CAP_GEN(dev, vport_group_manager)) + return -EOPNOTSUPP; + + if (esw->mode == MLX5_ESWITCH_NONE) + return -EOPNOTSUPP; + + switch (MLX5_CAP_ETH(dev, wqe_inline_mode)) { + case MLX5_CAP_INLINE_MODE_NOT_REQUIRED: + mlx5_mode = MLX5_INLINE_MODE_NONE; + goto out; + case MLX5_CAP_INLINE_MODE_L2: + mlx5_mode = MLX5_INLINE_MODE_L2; + goto out; + case MLX5_CAP_INLINE_MODE_VPORT_CONTEXT: + goto query_vports; + } + +query_vports: + mlx5_query_nic_vport_min_inline(dev, esw->first_host_vport, &prev_mlx5_mode); + mlx5_esw_for_each_host_func_vport(esw, vport, esw->esw_funcs.num_vfs) { + mlx5_query_nic_vport_min_inline(dev, vport, &mlx5_mode); + if (prev_mlx5_mode != mlx5_mode) + return -EINVAL; + prev_mlx5_mode = mlx5_mode; + } + +out: + *mode = mlx5_mode; + return 0; +} + +static void esw_destroy_restore_table(struct mlx5_eswitch *esw) +{ + struct mlx5_esw_offload *offloads = &esw->offloads; + + if (!mlx5_eswitch_reg_c1_loopback_supported(esw)) + return; + + mlx5_modify_header_dealloc(esw->dev, offloads->restore_copy_hdr_id); + mlx5_destroy_flow_group(offloads->restore_group); + mlx5_destroy_flow_table(offloads->ft_offloads_restore); +} + +static int esw_create_restore_table(struct mlx5_eswitch *esw) +{ + u8 modact[MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto)] = {}; + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_namespace *ns; + struct mlx5_modify_hdr *mod_hdr; + void *match_criteria, *misc; + struct mlx5_flow_table *ft; + struct mlx5_flow_group *g; + u32 *flow_group_in; + int err = 0; + + if (!mlx5_eswitch_reg_c1_loopback_supported(esw)) + return 0; + + ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_OFFLOADS); + if (!ns) { + esw_warn(esw->dev, "Failed to get offloads flow namespace\n"); + return -EOPNOTSUPP; + } + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) { + err = -ENOMEM; + goto out_free; + } + + ft_attr.max_fte = 1 << ESW_CHAIN_TAG_METADATA_BITS; + ft = mlx5_create_flow_table(ns, &ft_attr); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + esw_warn(esw->dev, "Failed to create restore table, err %d\n", + err); + goto out_free; + } + + memset(flow_group_in, 0, inlen); + match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, + match_criteria); + misc = MLX5_ADDR_OF(fte_match_param, match_criteria, + misc_parameters_2); + + MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0, + ESW_CHAIN_TAG_METADATA_MASK); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, + ft_attr.max_fte - 1); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_MISC_PARAMETERS_2); + g = mlx5_create_flow_group(ft, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "Failed to create restore flow group, err: %d\n", + err); + goto err_group; + } + + MLX5_SET(copy_action_in, modact, action_type, MLX5_ACTION_TYPE_COPY); + MLX5_SET(copy_action_in, modact, src_field, + MLX5_ACTION_IN_FIELD_METADATA_REG_C_1); + MLX5_SET(copy_action_in, modact, dst_field, + MLX5_ACTION_IN_FIELD_METADATA_REG_B); + mod_hdr = mlx5_modify_header_alloc(esw->dev, + MLX5_FLOW_NAMESPACE_KERNEL, 1, + modact); + if (IS_ERR(mod_hdr)) { + esw_warn(dev, "Failed to create restore mod header, err: %d\n", + err); + err = PTR_ERR(mod_hdr); + goto err_mod_hdr; + } + + esw->offloads.ft_offloads_restore = ft; + esw->offloads.restore_group = g; + esw->offloads.restore_copy_hdr_id = mod_hdr; + + kvfree(flow_group_in); + + return 0; + +err_mod_hdr: + mlx5_destroy_flow_group(g); +err_group: + mlx5_destroy_flow_table(ft); +out_free: + kvfree(flow_group_in); + + return err; +} + static int esw_offloads_start(struct mlx5_eswitch *esw, struct netlink_ext_ack *extack) { @@ -1172,13 +1592,14 @@ static int esw_offloads_start(struct mlx5_eswitch *esw, return -EINVAL; } - mlx5_eswitch_disable(esw, false); - mlx5_eswitch_update_num_of_vfs(esw, esw->dev->priv.sriov.num_vfs); - err = mlx5_eswitch_enable(esw, MLX5_ESWITCH_OFFLOADS); + mlx5_eswitch_disable_locked(esw, false); + err = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_OFFLOADS, + esw->dev->priv.sriov.num_vfs); if (err) { NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch to offloads"); - err1 = mlx5_eswitch_enable(esw, MLX5_ESWITCH_LEGACY); + err1 = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_LEGACY, + MLX5_ESWITCH_IGNORE_NUM_VFS); if (err1) { NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch back to legacy"); @@ -1233,187 +1654,66 @@ static void __esw_offloads_unload_rep(struct mlx5_eswitch *esw, esw->offloads.rep_ops[rep_type]->unload(rep); } -static void __unload_reps_special_vport(struct mlx5_eswitch *esw, u8 rep_type) -{ - struct mlx5_eswitch_rep *rep; - - if (mlx5_ecpf_vport_exists(esw->dev)) { - rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_ECPF); - __esw_offloads_unload_rep(esw, rep, rep_type); - } - - if (mlx5_core_is_ecpf_esw_manager(esw->dev)) { - rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_PF); - __esw_offloads_unload_rep(esw, rep, rep_type); - } - - rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK); - __esw_offloads_unload_rep(esw, rep, rep_type); -} - -static void __unload_reps_vf_vport(struct mlx5_eswitch *esw, int nvports, - u8 rep_type) +static void __unload_reps_all_vport(struct mlx5_eswitch *esw, u8 rep_type) { struct mlx5_eswitch_rep *rep; int i; - mlx5_esw_for_each_vf_rep_reverse(esw, i, rep, nvports) + mlx5_esw_for_each_vf_rep_reverse(esw, i, rep, esw->esw_funcs.num_vfs) __esw_offloads_unload_rep(esw, rep, rep_type); -} - -static void esw_offloads_unload_vf_reps(struct mlx5_eswitch *esw, int nvports) -{ - u8 rep_type = NUM_REP_TYPES; - - while (rep_type-- > 0) - __unload_reps_vf_vport(esw, nvports, rep_type); -} - -static void __unload_reps_all_vport(struct mlx5_eswitch *esw, u8 rep_type) -{ - __unload_reps_vf_vport(esw, esw->esw_funcs.num_vfs, rep_type); - - /* Special vports must be the last to unload. */ - __unload_reps_special_vport(esw, rep_type); -} - -static void esw_offloads_unload_all_reps(struct mlx5_eswitch *esw) -{ - u8 rep_type = NUM_REP_TYPES; - - while (rep_type-- > 0) - __unload_reps_all_vport(esw, rep_type); -} - -static int __esw_offloads_load_rep(struct mlx5_eswitch *esw, - struct mlx5_eswitch_rep *rep, u8 rep_type) -{ - int err = 0; - - if (atomic_cmpxchg(&rep->rep_data[rep_type].state, - REP_REGISTERED, REP_LOADED) == REP_REGISTERED) { - err = esw->offloads.rep_ops[rep_type]->load(esw->dev, rep); - if (err) - atomic_set(&rep->rep_data[rep_type].state, - REP_REGISTERED); - } - - return err; -} - -static int __load_reps_special_vport(struct mlx5_eswitch *esw, u8 rep_type) -{ - struct mlx5_eswitch_rep *rep; - int err; - - rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK); - err = __esw_offloads_load_rep(esw, rep, rep_type); - if (err) - return err; - - if (mlx5_core_is_ecpf_esw_manager(esw->dev)) { - rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_PF); - err = __esw_offloads_load_rep(esw, rep, rep_type); - if (err) - goto err_pf; - } if (mlx5_ecpf_vport_exists(esw->dev)) { rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_ECPF); - err = __esw_offloads_load_rep(esw, rep, rep_type); - if (err) - goto err_ecpf; + __esw_offloads_unload_rep(esw, rep, rep_type); } - return 0; - -err_ecpf: if (mlx5_core_is_ecpf_esw_manager(esw->dev)) { rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_PF); __esw_offloads_unload_rep(esw, rep, rep_type); } -err_pf: rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK); __esw_offloads_unload_rep(esw, rep, rep_type); - return err; } -static int __load_reps_vf_vport(struct mlx5_eswitch *esw, int nvports, - u8 rep_type) +int esw_offloads_load_rep(struct mlx5_eswitch *esw, u16 vport_num) { struct mlx5_eswitch_rep *rep; - int err, i; - - mlx5_esw_for_each_vf_rep(esw, i, rep, nvports) { - err = __esw_offloads_load_rep(esw, rep, rep_type); - if (err) - goto err_vf; - } - - return 0; - -err_vf: - __unload_reps_vf_vport(esw, --i, rep_type); - return err; -} - -static int __load_reps_all_vport(struct mlx5_eswitch *esw, u8 rep_type) -{ + int rep_type; int err; - /* Special vports must be loaded first, uplink rep creates mdev resource. */ - err = __load_reps_special_vport(esw, rep_type); - if (err) - return err; + if (esw->mode != MLX5_ESWITCH_OFFLOADS) + return 0; - err = __load_reps_vf_vport(esw, esw->esw_funcs.num_vfs, rep_type); - if (err) - goto err_vfs; + rep = mlx5_eswitch_get_rep(esw, vport_num); + for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) + if (atomic_cmpxchg(&rep->rep_data[rep_type].state, + REP_REGISTERED, REP_LOADED) == REP_REGISTERED) { + err = esw->offloads.rep_ops[rep_type]->load(esw->dev, rep); + if (err) + goto err_reps; + } return 0; -err_vfs: - __unload_reps_special_vport(esw, rep_type); - return err; -} - -static int esw_offloads_load_vf_reps(struct mlx5_eswitch *esw, int nvports) -{ - u8 rep_type = 0; - int err; - - for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) { - err = __load_reps_vf_vport(esw, nvports, rep_type); - if (err) - goto err_reps; - } - - return err; - err_reps: - while (rep_type-- > 0) - __unload_reps_vf_vport(esw, nvports, rep_type); + atomic_set(&rep->rep_data[rep_type].state, REP_REGISTERED); + for (--rep_type; rep_type >= 0; rep_type--) + __esw_offloads_unload_rep(esw, rep, rep_type); return err; } -static int esw_offloads_load_all_reps(struct mlx5_eswitch *esw) +void esw_offloads_unload_rep(struct mlx5_eswitch *esw, u16 vport_num) { - u8 rep_type = 0; - int err; - - for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) { - err = __load_reps_all_vport(esw, rep_type); - if (err) - goto err_reps; - } + struct mlx5_eswitch_rep *rep; + int rep_type; - return err; + if (esw->mode != MLX5_ESWITCH_OFFLOADS) + return; -err_reps: - while (rep_type-- > 0) - __unload_reps_all_vport(esw, rep_type); - return err; + rep = mlx5_eswitch_get_rep(esw, vport_num); + for (rep_type = NUM_REP_TYPES - 1; rep_type >= 0; rep_type--) + __esw_offloads_unload_rep(esw, rep, rep_type); } #define ESW_OFFLOADS_DEVCOM_PAIR (0) @@ -1601,14 +1901,21 @@ static int esw_vport_add_ingress_acl_modify_metadata(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { u8 action[MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto)] = {}; - static const struct mlx5_flow_spec spec = {}; struct mlx5_flow_act flow_act = {}; int err = 0; + u32 key; + + key = mlx5_eswitch_get_vport_metadata_for_match(esw, vport->vport); + key >>= ESW_SOURCE_PORT_METADATA_OFFSET; MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET); - MLX5_SET(set_action_in, action, field, MLX5_ACTION_IN_FIELD_METADATA_REG_C_0); - MLX5_SET(set_action_in, action, data, - mlx5_eswitch_get_vport_metadata_for_match(esw, vport->vport)); + MLX5_SET(set_action_in, action, field, + MLX5_ACTION_IN_FIELD_METADATA_REG_C_0); + MLX5_SET(set_action_in, action, data, key); + MLX5_SET(set_action_in, action, offset, + ESW_SOURCE_PORT_METADATA_OFFSET); + MLX5_SET(set_action_in, action, length, + ESW_SOURCE_PORT_METADATA_BITS); vport->ingress.offloads.modify_metadata = mlx5_modify_header_alloc(esw->dev, MLX5_FLOW_NAMESPACE_ESW_INGRESS, @@ -1625,7 +1932,7 @@ static int esw_vport_add_ingress_acl_modify_metadata(struct mlx5_eswitch *esw, flow_act.modify_hdr = vport->ingress.offloads.modify_metadata; vport->ingress.offloads.modify_metadata_rule = mlx5_add_flow_rules(vport->ingress.acl, - &spec, &flow_act, NULL, 0); + NULL, &flow_act, NULL, 0); if (IS_ERR(vport->ingress.offloads.modify_metadata_rule)) { err = PTR_ERR(vport->ingress.offloads.modify_metadata_rule); esw_warn(esw->dev, @@ -1837,6 +2144,18 @@ esw_check_vport_match_metadata_supported(const struct mlx5_eswitch *esw) return true; } +static bool +esw_check_vport_match_metadata_mandatory(const struct mlx5_eswitch *esw) +{ + return mlx5_core_mp_enabled(esw->dev); +} + +static bool esw_use_vport_metadata(const struct mlx5_eswitch *esw) +{ + return esw_check_vport_match_metadata_mandatory(esw) && + esw_check_vport_match_metadata_supported(esw); +} + int esw_vport_create_offloads_acl_tables(struct mlx5_eswitch *esw, struct mlx5_vport *vport) @@ -1875,7 +2194,7 @@ static int esw_create_uplink_offloads_acl_tables(struct mlx5_eswitch *esw) struct mlx5_vport *vport; int err; - if (esw_check_vport_match_metadata_supported(esw)) + if (esw_use_vport_metadata(esw)) esw->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA; vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_UPLINK); @@ -1911,27 +2230,34 @@ static int esw_offloads_steering_init(struct mlx5_eswitch *esw) if (err) return err; - err = esw_create_offloads_fdb_tables(esw, total_vports); + err = esw_create_offloads_table(esw, total_vports); if (err) - goto create_fdb_err; + goto create_offloads_err; - err = esw_create_offloads_table(esw, total_vports); + err = esw_create_restore_table(esw); if (err) - goto create_ft_err; + goto create_restore_err; + + err = esw_create_offloads_fdb_tables(esw, total_vports); + if (err) + goto create_fdb_err; err = esw_create_vport_rx_group(esw, total_vports); if (err) goto create_fg_err; + mutex_init(&esw->fdb_table.offloads.vports.lock); + hash_init(esw->fdb_table.offloads.vports.table); + return 0; create_fg_err: - esw_destroy_offloads_table(esw); - -create_ft_err: esw_destroy_offloads_fdb_tables(esw); - create_fdb_err: + esw_destroy_restore_table(esw); +create_restore_err: + esw_destroy_offloads_table(esw); +create_offloads_err: esw_destroy_uplink_offloads_acl_tables(esw); return err; @@ -1939,9 +2265,11 @@ create_fdb_err: static void esw_offloads_steering_cleanup(struct mlx5_eswitch *esw) { + mutex_destroy(&esw->fdb_table.offloads.vports.lock); esw_destroy_vport_rx_group(esw); - esw_destroy_offloads_table(esw); esw_destroy_offloads_fdb_tables(esw); + esw_destroy_restore_table(esw); + esw_destroy_offloads_table(esw); esw_destroy_uplink_offloads_acl_tables(esw); } @@ -1961,11 +2289,12 @@ esw_vfs_changed_event_handler(struct mlx5_eswitch *esw, const u32 *out) /* Number of VFs can only change from "0 to x" or "x to 0". */ if (esw->esw_funcs.num_vfs > 0) { - esw_offloads_unload_vf_reps(esw, esw->esw_funcs.num_vfs); + mlx5_eswitch_unload_vf_vports(esw, esw->esw_funcs.num_vfs); } else { int err; - err = esw_offloads_load_vf_reps(esw, new_num_vfs); + err = mlx5_eswitch_load_vf_vports(esw, new_num_vfs, + MLX5_VPORT_UC_ADDR_CHANGE); if (err) return; } @@ -2023,40 +2352,43 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) else esw->offloads.encap = DEVLINK_ESWITCH_ENCAP_MODE_NONE; + mutex_init(&esw->offloads.termtbl_mutex); mlx5_rdma_enable_roce(esw->dev); - err = esw_offloads_steering_init(esw); - if (err) - goto err_steering_init; err = esw_set_passing_vport_metadata(esw, true); if (err) goto err_vport_metadata; + err = esw_offloads_steering_init(esw); + if (err) + goto err_steering_init; + /* Representor will control the vport link state */ mlx5_esw_for_each_vf_vport(esw, i, vport, esw->esw_funcs.num_vfs) vport->info.link_state = MLX5_VPORT_ADMIN_STATE_DOWN; - err = mlx5_eswitch_enable_pf_vf_vports(esw, MLX5_VPORT_UC_ADDR_CHANGE); + /* Uplink vport rep must load first. */ + err = esw_offloads_load_rep(esw, MLX5_VPORT_UPLINK); if (err) - goto err_vports; + goto err_uplink; - err = esw_offloads_load_all_reps(esw); + err = mlx5_eswitch_enable_pf_vf_vports(esw, MLX5_VPORT_UC_ADDR_CHANGE); if (err) - goto err_reps; + goto err_vports; esw_offloads_devcom_init(esw); - mutex_init(&esw->offloads.termtbl_mutex); return 0; -err_reps: - mlx5_eswitch_disable_pf_vf_vports(esw); err_vports: + esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK); +err_uplink: esw_set_passing_vport_metadata(esw, false); -err_vport_metadata: - esw_offloads_steering_cleanup(esw); err_steering_init: + esw_offloads_steering_cleanup(esw); +err_vport_metadata: mlx5_rdma_disable_roce(esw->dev); + mutex_destroy(&esw->offloads.termtbl_mutex); return err; } @@ -2065,11 +2397,13 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw, { int err, err1; - mlx5_eswitch_disable(esw, false); - err = mlx5_eswitch_enable(esw, MLX5_ESWITCH_LEGACY); + mlx5_eswitch_disable_locked(esw, false); + err = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_LEGACY, + MLX5_ESWITCH_IGNORE_NUM_VFS); if (err) { NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch to legacy"); - err1 = mlx5_eswitch_enable(esw, MLX5_ESWITCH_OFFLOADS); + err1 = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_OFFLOADS, + MLX5_ESWITCH_IGNORE_NUM_VFS); if (err1) { NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch back to offloads"); @@ -2082,11 +2416,12 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw, void esw_offloads_disable(struct mlx5_eswitch *esw) { esw_offloads_devcom_cleanup(esw); - esw_offloads_unload_all_reps(esw); mlx5_eswitch_disable_pf_vf_vports(esw); + esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK); esw_set_passing_vport_metadata(esw, false); esw_offloads_steering_cleanup(esw); mlx5_rdma_disable_roce(esw->dev); + mutex_destroy(&esw->offloads.termtbl_mutex); esw->offloads.encap = DEVLINK_ESWITCH_ENCAP_MODE_NONE; } @@ -2166,60 +2501,82 @@ static int esw_inline_mode_to_devlink(u8 mlx5_mode, u8 *mode) return 0; } -static int mlx5_devlink_eswitch_check(struct devlink *devlink) +static int mlx5_eswitch_check(const struct mlx5_core_dev *dev) { - struct mlx5_core_dev *dev = devlink_priv(devlink); - if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) return -EOPNOTSUPP; if(!MLX5_ESWITCH_MANAGER(dev)) return -EPERM; - if (dev->priv.eswitch->mode == MLX5_ESWITCH_NONE && - !mlx5_core_is_ecpf_esw_manager(dev)) - return -EOPNOTSUPP; - return 0; } +static int eswitch_devlink_esw_mode_check(const struct mlx5_eswitch *esw) +{ + /* devlink commands in NONE eswitch mode are currently supported only + * on ECPF. + */ + return (esw->mode == MLX5_ESWITCH_NONE && + !mlx5_core_is_ecpf_esw_manager(esw->dev)) ? -EOPNOTSUPP : 0; +} + int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode, struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); + struct mlx5_eswitch *esw = dev->priv.eswitch; u16 cur_mlx5_mode, mlx5_mode = 0; int err; - err = mlx5_devlink_eswitch_check(devlink); + err = mlx5_eswitch_check(dev); if (err) return err; - cur_mlx5_mode = dev->priv.eswitch->mode; - if (esw_mode_from_devlink(mode, &mlx5_mode)) return -EINVAL; + mutex_lock(&esw->mode_lock); + err = eswitch_devlink_esw_mode_check(esw); + if (err) + goto unlock; + + cur_mlx5_mode = esw->mode; + if (cur_mlx5_mode == mlx5_mode) - return 0; + goto unlock; if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV) - return esw_offloads_start(dev->priv.eswitch, extack); + err = esw_offloads_start(esw, extack); else if (mode == DEVLINK_ESWITCH_MODE_LEGACY) - return esw_offloads_stop(dev->priv.eswitch, extack); + err = esw_offloads_stop(esw, extack); else - return -EINVAL; + err = -EINVAL; + +unlock: + mutex_unlock(&esw->mode_lock); + return err; } int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode) { struct mlx5_core_dev *dev = devlink_priv(devlink); + struct mlx5_eswitch *esw = dev->priv.eswitch; int err; - err = mlx5_devlink_eswitch_check(devlink); + err = mlx5_eswitch_check(dev); if (err) return err; - return esw_mode_to_devlink(dev->priv.eswitch->mode, mode); + mutex_lock(&esw->mode_lock); + err = eswitch_devlink_esw_mode_check(dev->priv.eswitch); + if (err) + goto unlock; + + err = esw_mode_to_devlink(esw->mode, mode); +unlock: + mutex_unlock(&esw->mode_lock); + return err; } int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode, @@ -2230,18 +2587,24 @@ int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode, int err, vport, num_vport; u8 mlx5_mode; - err = mlx5_devlink_eswitch_check(devlink); + err = mlx5_eswitch_check(dev); if (err) return err; + mutex_lock(&esw->mode_lock); + err = eswitch_devlink_esw_mode_check(esw); + if (err) + goto out; + switch (MLX5_CAP_ETH(dev, wqe_inline_mode)) { case MLX5_CAP_INLINE_MODE_NOT_REQUIRED: if (mode == DEVLINK_ESWITCH_INLINE_MODE_NONE) - return 0; + goto out; /* fall through */ case MLX5_CAP_INLINE_MODE_L2: NL_SET_ERR_MSG_MOD(extack, "Inline mode can't be set"); - return -EOPNOTSUPP; + err = -EOPNOTSUPP; + goto out; case MLX5_CAP_INLINE_MODE_VPORT_CONTEXT: break; } @@ -2249,7 +2612,8 @@ int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode, if (atomic64_read(&esw->offloads.num_flows) > 0) { NL_SET_ERR_MSG_MOD(extack, "Can't set inline mode when flows are configured"); - return -EOPNOTSUPP; + err = -EOPNOTSUPP; + goto out; } err = esw_inline_mode_from_devlink(mode, &mlx5_mode); @@ -2266,6 +2630,7 @@ int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode, } esw->offloads.inline_mode = mlx5_mode; + mutex_unlock(&esw->mode_lock); return 0; revert_inline_mode: @@ -2275,6 +2640,7 @@ revert_inline_mode: vport, esw->offloads.inline_mode); out: + mutex_unlock(&esw->mode_lock); return err; } @@ -2284,48 +2650,19 @@ int mlx5_devlink_eswitch_inline_mode_get(struct devlink *devlink, u8 *mode) struct mlx5_eswitch *esw = dev->priv.eswitch; int err; - err = mlx5_devlink_eswitch_check(devlink); + err = mlx5_eswitch_check(dev); if (err) return err; - return esw_inline_mode_to_devlink(esw->offloads.inline_mode, mode); -} - -int mlx5_eswitch_inline_mode_get(struct mlx5_eswitch *esw, u8 *mode) -{ - u8 prev_mlx5_mode, mlx5_mode = MLX5_INLINE_MODE_L2; - struct mlx5_core_dev *dev = esw->dev; - int vport; - - if (!MLX5_CAP_GEN(dev, vport_group_manager)) - return -EOPNOTSUPP; - - if (esw->mode == MLX5_ESWITCH_NONE) - return -EOPNOTSUPP; - - switch (MLX5_CAP_ETH(dev, wqe_inline_mode)) { - case MLX5_CAP_INLINE_MODE_NOT_REQUIRED: - mlx5_mode = MLX5_INLINE_MODE_NONE; - goto out; - case MLX5_CAP_INLINE_MODE_L2: - mlx5_mode = MLX5_INLINE_MODE_L2; - goto out; - case MLX5_CAP_INLINE_MODE_VPORT_CONTEXT: - goto query_vports; - } - -query_vports: - mlx5_query_nic_vport_min_inline(dev, esw->first_host_vport, &prev_mlx5_mode); - mlx5_esw_for_each_host_func_vport(esw, vport, esw->esw_funcs.num_vfs) { - mlx5_query_nic_vport_min_inline(dev, vport, &mlx5_mode); - if (prev_mlx5_mode != mlx5_mode) - return -EINVAL; - prev_mlx5_mode = mlx5_mode; - } + mutex_lock(&esw->mode_lock); + err = eswitch_devlink_esw_mode_check(esw); + if (err) + goto unlock; -out: - *mode = mlx5_mode; - return 0; + err = esw_inline_mode_to_devlink(esw->offloads.inline_mode, mode); +unlock: + mutex_unlock(&esw->mode_lock); + return err; } int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink, @@ -2336,30 +2673,40 @@ int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink, struct mlx5_eswitch *esw = dev->priv.eswitch; int err; - err = mlx5_devlink_eswitch_check(devlink); + err = mlx5_eswitch_check(dev); if (err) return err; + mutex_lock(&esw->mode_lock); + err = eswitch_devlink_esw_mode_check(esw); + if (err) + goto unlock; + if (encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE && (!MLX5_CAP_ESW_FLOWTABLE_FDB(dev, reformat) || - !MLX5_CAP_ESW_FLOWTABLE_FDB(dev, decap))) - return -EOPNOTSUPP; + !MLX5_CAP_ESW_FLOWTABLE_FDB(dev, decap))) { + err = -EOPNOTSUPP; + goto unlock; + } - if (encap && encap != DEVLINK_ESWITCH_ENCAP_MODE_BASIC) - return -EOPNOTSUPP; + if (encap && encap != DEVLINK_ESWITCH_ENCAP_MODE_BASIC) { + err = -EOPNOTSUPP; + goto unlock; + } if (esw->mode == MLX5_ESWITCH_LEGACY) { esw->offloads.encap = encap; - return 0; + goto unlock; } if (esw->offloads.encap == encap) - return 0; + goto unlock; if (atomic64_read(&esw->offloads.num_flows) > 0) { NL_SET_ERR_MSG_MOD(extack, "Can't set encapsulation when flows are configured"); - return -EOPNOTSUPP; + err = -EOPNOTSUPP; + goto unlock; } esw_destroy_offloads_fdb_tables(esw); @@ -2375,6 +2722,8 @@ int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink, (void)esw_create_offloads_fdb_tables(esw, esw->nvports); } +unlock: + mutex_unlock(&esw->mode_lock); return err; } @@ -2385,14 +2734,36 @@ int mlx5_devlink_eswitch_encap_mode_get(struct devlink *devlink, struct mlx5_eswitch *esw = dev->priv.eswitch; int err; - err = mlx5_devlink_eswitch_check(devlink); + err = mlx5_eswitch_check(dev); if (err) return err; + mutex_lock(&esw->mode_lock); + err = eswitch_devlink_esw_mode_check(esw); + if (err) + goto unlock; + *encap = esw->offloads.encap; +unlock: + mutex_unlock(&esw->mode_lock); return 0; } +static bool +mlx5_eswitch_vport_has_rep(const struct mlx5_eswitch *esw, u16 vport_num) +{ + /* Currently, only ECPF based device has representor for host PF. */ + if (vport_num == MLX5_VPORT_PF && + !mlx5_core_is_ecpf_esw_manager(esw->dev)) + return false; + + if (vport_num == MLX5_VPORT_ECPF && + !mlx5_ecpf_vport_exists(esw->dev)) + return false; + + return true; +} + void mlx5_eswitch_register_vport_reps(struct mlx5_eswitch *esw, const struct mlx5_eswitch_rep_ops *ops, u8 rep_type) @@ -2403,8 +2774,10 @@ void mlx5_eswitch_register_vport_reps(struct mlx5_eswitch *esw, esw->offloads.rep_ops[rep_type] = ops; mlx5_esw_for_all_reps(esw, i, rep) { - rep_data = &rep->rep_data[rep_type]; - atomic_set(&rep_data->state, REP_REGISTERED); + if (likely(mlx5_eswitch_vport_has_rep(esw, i))) { + rep_data = &rep->rep_data[rep_type]; + atomic_set(&rep_data->state, REP_REGISTERED); + } } } EXPORT_SYMBOL(mlx5_eswitch_register_vport_reps); @@ -2464,15 +2837,53 @@ bool mlx5_eswitch_is_vf_vport(const struct mlx5_eswitch *esw, u16 vport_num) vport_num <= esw->dev->priv.sriov.max_vfs; } +bool mlx5_eswitch_reg_c1_loopback_enabled(const struct mlx5_eswitch *esw) +{ + return !!(esw->flags & MLX5_ESWITCH_REG_C1_LOOPBACK_ENABLED); +} +EXPORT_SYMBOL(mlx5_eswitch_reg_c1_loopback_enabled); + bool mlx5_eswitch_vport_match_metadata_enabled(const struct mlx5_eswitch *esw) { return !!(esw->flags & MLX5_ESWITCH_VPORT_MATCH_METADATA); } EXPORT_SYMBOL(mlx5_eswitch_vport_match_metadata_enabled); -u32 mlx5_eswitch_get_vport_metadata_for_match(const struct mlx5_eswitch *esw, +u32 mlx5_eswitch_get_vport_metadata_for_match(struct mlx5_eswitch *esw, u16 vport_num) { - return ((MLX5_CAP_GEN(esw->dev, vhca_id) & 0xffff) << 16) | vport_num; + u32 vport_num_mask = GENMASK(ESW_VPORT_BITS - 1, 0); + u32 vhca_id_mask = GENMASK(ESW_VHCA_ID_BITS - 1, 0); + u32 vhca_id = MLX5_CAP_GEN(esw->dev, vhca_id); + u32 val; + + /* Make sure the vhca_id fits the ESW_VHCA_ID_BITS */ + WARN_ON_ONCE(vhca_id >= BIT(ESW_VHCA_ID_BITS)); + + /* Trim vhca_id to ESW_VHCA_ID_BITS */ + vhca_id &= vhca_id_mask; + + /* Make sure pf and ecpf map to end of ESW_VPORT_BITS range so they + * don't overlap with VF numbers, and themselves, after trimming. + */ + WARN_ON_ONCE((MLX5_VPORT_UPLINK & vport_num_mask) < + vport_num_mask - 1); + WARN_ON_ONCE((MLX5_VPORT_ECPF & vport_num_mask) < + vport_num_mask - 1); + WARN_ON_ONCE((MLX5_VPORT_UPLINK & vport_num_mask) == + (MLX5_VPORT_ECPF & vport_num_mask)); + + /* Make sure that the VF vport_num fits ESW_VPORT_BITS and don't + * overlap with pf and ecpf. + */ + if (vport_num != MLX5_VPORT_UPLINK && + vport_num != MLX5_VPORT_ECPF) + WARN_ON_ONCE(vport_num >= vport_num_mask - 1); + + /* We can now trim vport_num to ESW_VPORT_BITS */ + vport_num &= vport_num_mask; + + val = (vhca_id << ESW_VPORT_BITS) | vport_num; + return val << (32 - ESW_SOURCE_PORT_METADATA_BITS); } EXPORT_SYMBOL(mlx5_eswitch_get_vport_metadata_for_match); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c index dc08ed9339ab..17a0d2bc102b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c @@ -3,6 +3,7 @@ #include <linux/mlx5/fs.h> #include "eswitch.h" +#include "fs_core.h" struct mlx5_termtbl_handle { struct hlist_node termtbl_hlist; @@ -28,6 +29,10 @@ mlx5_eswitch_termtbl_hash(struct mlx5_flow_act *flow_act, sizeof(dest->vport.num), hash); hash = jhash((const void *)&dest->vport.vhca_id, sizeof(dest->vport.num), hash); + if (dest->vport.pkt_reformat) + hash = jhash(dest->vport.pkt_reformat, + sizeof(*dest->vport.pkt_reformat), + hash); return hash; } @@ -37,11 +42,19 @@ mlx5_eswitch_termtbl_cmp(struct mlx5_flow_act *flow_act1, struct mlx5_flow_act *flow_act2, struct mlx5_flow_destination *dest2) { - return flow_act1->action != flow_act2->action || - dest1->vport.num != dest2->vport.num || - dest1->vport.vhca_id != dest2->vport.vhca_id || - memcmp(&flow_act1->vlan, &flow_act2->vlan, - sizeof(flow_act1->vlan)); + int ret; + + ret = flow_act1->action != flow_act2->action || + dest1->vport.num != dest2->vport.num || + dest1->vport.vhca_id != dest2->vport.vhca_id || + memcmp(&flow_act1->vlan, &flow_act2->vlan, + sizeof(flow_act1->vlan)); + if (ret) + return ret; + + return dest1->vport.pkt_reformat && dest2->vport.pkt_reformat ? + memcmp(dest1->vport.pkt_reformat, dest2->vport.pkt_reformat, + sizeof(*dest1->vport.pkt_reformat)) : 0; } static int @@ -49,7 +62,6 @@ mlx5_eswitch_termtbl_create(struct mlx5_core_dev *dev, struct mlx5_termtbl_handle *tt, struct mlx5_flow_act *flow_act) { - static const struct mlx5_flow_spec spec = {}; struct mlx5_flow_table_attr ft_attr = {}; struct mlx5_flow_namespace *root_ns; int err; @@ -63,7 +75,8 @@ mlx5_eswitch_termtbl_create(struct mlx5_core_dev *dev, /* As this is the terminating action then the termination table is the * same prio as the slow path */ - ft_attr.flags = MLX5_FLOW_TABLE_TERMINATION; + ft_attr.flags = MLX5_FLOW_TABLE_TERMINATION | + MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT; ft_attr.prio = FDB_SLOW_PATH; ft_attr.max_fte = 1; ft_attr.autogroup.max_num_groups = 1; @@ -73,9 +86,8 @@ mlx5_eswitch_termtbl_create(struct mlx5_core_dev *dev, return -EOPNOTSUPP; } - tt->rule = mlx5_add_flow_rules(tt->termtbl, &spec, flow_act, + tt->rule = mlx5_add_flow_rules(tt->termtbl, NULL, flow_act, &tt->dest, 1); - if (IS_ERR(tt->rule)) { esw_warn(dev, "Failed to create termination table rule\n"); goto add_flow_err; @@ -93,7 +105,8 @@ add_flow_err: static struct mlx5_termtbl_handle * mlx5_eswitch_termtbl_get_create(struct mlx5_eswitch *esw, struct mlx5_flow_act *flow_act, - struct mlx5_flow_destination *dest) + struct mlx5_flow_destination *dest, + struct mlx5_esw_flow_attr *attr) { struct mlx5_termtbl_handle *tt; bool found = false; @@ -101,7 +114,6 @@ mlx5_eswitch_termtbl_get_create(struct mlx5_eswitch *esw, int err; mutex_lock(&esw->offloads.termtbl_mutex); - hash_key = mlx5_eswitch_termtbl_hash(flow_act, dest); hash_for_each_possible(esw->offloads.termtbl_tbl, tt, termtbl_hlist, hash_key) { @@ -123,6 +135,7 @@ mlx5_eswitch_termtbl_get_create(struct mlx5_eswitch *esw, tt->dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; tt->dest.vport.num = dest->vport.num; tt->dest.vport.vhca_id = dest->vport.vhca_id; + tt->dest.vport.flags = dest->vport.flags; memcpy(&tt->flow_act, flow_act, sizeof(*flow_act)); err = mlx5_eswitch_termtbl_create(esw->dev, tt, flow_act); @@ -157,31 +170,50 @@ mlx5_eswitch_termtbl_put(struct mlx5_eswitch *esw, } } +static bool mlx5_eswitch_termtbl_is_encap_reformat(struct mlx5_pkt_reformat *rt) +{ + switch (rt->reformat_type) { + case MLX5_REFORMAT_TYPE_L2_TO_VXLAN: + case MLX5_REFORMAT_TYPE_L2_TO_NVGRE: + case MLX5_REFORMAT_TYPE_L2_TO_L2_TUNNEL: + case MLX5_REFORMAT_TYPE_L2_TO_L3_TUNNEL: + return true; + default: + return false; + } +} + static void mlx5_eswitch_termtbl_actions_move(struct mlx5_flow_act *src, struct mlx5_flow_act *dst) { - if (!(src->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH)) - return; - - src->action &= ~MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH; - dst->action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH; - memcpy(&dst->vlan[0], &src->vlan[0], sizeof(src->vlan[0])); - memset(&src->vlan[0], 0, sizeof(src->vlan[0])); - - if (!(src->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2)) - return; + if (src->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) { + src->action &= ~MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH; + dst->action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH; + memcpy(&dst->vlan[0], &src->vlan[0], sizeof(src->vlan[0])); + memset(&src->vlan[0], 0, sizeof(src->vlan[0])); + + if (src->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2) { + src->action &= ~MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2; + dst->action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2; + memcpy(&dst->vlan[1], &src->vlan[1], sizeof(src->vlan[1])); + memset(&src->vlan[1], 0, sizeof(src->vlan[1])); + } + } - src->action &= ~MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2; - dst->action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2; - memcpy(&dst->vlan[1], &src->vlan[1], sizeof(src->vlan[1])); - memset(&src->vlan[1], 0, sizeof(src->vlan[1])); + if (src->action & MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT && + mlx5_eswitch_termtbl_is_encap_reformat(src->pkt_reformat)) { + src->action &= ~MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT; + dst->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT; + dst->pkt_reformat = src->pkt_reformat; + src->pkt_reformat = NULL; + } } static bool mlx5_eswitch_offload_is_uplink_port(const struct mlx5_eswitch *esw, const struct mlx5_flow_spec *spec) { - u32 port_mask, port_value; + u16 port_mask, port_value; if (MLX5_CAP_ESW_FLOWTABLE(esw->dev, flow_source)) return spec->flow_context.flow_source == @@ -191,20 +223,32 @@ static bool mlx5_eswitch_offload_is_uplink_port(const struct mlx5_eswitch *esw, misc_parameters.source_port); port_value = MLX5_GET(fte_match_param, spec->match_value, misc_parameters.source_port); - return (port_mask & port_value & 0xffff) == MLX5_VPORT_UPLINK; + return (port_mask & port_value) == MLX5_VPORT_UPLINK; } bool mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw, + struct mlx5_esw_flow_attr *attr, struct mlx5_flow_act *flow_act, struct mlx5_flow_spec *spec) { - if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, termination_table)) + int i; + + if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, termination_table) || + attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH || + !mlx5_eswitch_offload_is_uplink_port(esw, spec)) return false; /* push vlan on RX */ - return (flow_act->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) && - mlx5_eswitch_offload_is_uplink_port(esw, spec); + if (flow_act->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) + return true; + + /* hairpin */ + for (i = attr->split_count; i < attr->out_count; i++) + if (attr->dests[i].rep->vport == MLX5_VPORT_UPLINK) + return true; + + return false; } struct mlx5_flow_handle * @@ -234,7 +278,7 @@ mlx5_eswitch_add_termtbl_rule(struct mlx5_eswitch *esw, /* get the terminating table for the action list */ tt = mlx5_eswitch_termtbl_get_create(esw, &term_tbl_act, - &dest[i]); + &dest[i], attr); if (IS_ERR(tt)) { esw_warn(esw->dev, "Failed to create termination table\n"); goto revert_changes; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c index 4c61d25d2e88..b794888fa3ba 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c @@ -57,7 +57,7 @@ struct mlx5_fpga_ipsec_cmd_context { struct completion complete; struct mlx5_fpga_device *dev; struct list_head list; /* Item in pending_cmds */ - u8 command[0]; + u8 command[]; }; struct mlx5_fpga_esp_xfrm; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 9dc24241dc91..62ce2b9417ab 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -110,9 +110,9 @@ #define ANCHOR_NUM_PRIOS 1 #define ANCHOR_MIN_LEVEL (BY_PASS_MIN_LEVEL + 1) -#define OFFLOADS_MAX_FT 1 -#define OFFLOADS_NUM_PRIOS 1 -#define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + 1) +#define OFFLOADS_MAX_FT 2 +#define OFFLOADS_NUM_PRIOS 2 +#define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + OFFLOADS_NUM_PRIOS) #define LAG_PRIO_NUM_LEVELS 1 #define LAG_NUM_PRIOS 1 @@ -145,7 +145,7 @@ static struct init_tree_node { ADD_NS(MLX5_FLOW_TABLE_MISS_ACTION_DEF, ADD_MULTIPLE_PRIO(LAG_NUM_PRIOS, LAG_PRIO_NUM_LEVELS))), - ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, {}, + ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, FS_CHAINING_CAPS, ADD_NS(MLX5_FLOW_TABLE_MISS_ACTION_DEF, ADD_MULTIPLE_PRIO(OFFLOADS_NUM_PRIOS, OFFLOADS_MAX_FT))), @@ -1322,7 +1322,7 @@ add_rule_fte(struct fs_fte *fte, fte->node.active = true; fte->status |= FS_FTE_STATUS_EXISTING; - atomic_inc(&fte->node.version); + atomic_inc(&fg->node.version); out: return handle; @@ -1577,28 +1577,19 @@ struct match_list { struct mlx5_flow_group *g; }; -struct match_list_head { - struct list_head list; - struct match_list first; -}; - -static void free_match_list(struct match_list_head *head, bool ft_locked) +static void free_match_list(struct match_list *head, bool ft_locked) { - if (!list_empty(&head->list)) { - struct match_list *iter, *match_tmp; + struct match_list *iter, *match_tmp; - list_del(&head->first.list); - tree_put_node(&head->first.g->node, ft_locked); - list_for_each_entry_safe(iter, match_tmp, &head->list, - list) { - tree_put_node(&iter->g->node, ft_locked); - list_del(&iter->list); - kfree(iter); - } + list_for_each_entry_safe(iter, match_tmp, &head->list, + list) { + tree_put_node(&iter->g->node, ft_locked); + list_del(&iter->list); + kfree(iter); } } -static int build_match_list(struct match_list_head *match_head, +static int build_match_list(struct match_list *match_head, struct mlx5_flow_table *ft, const struct mlx5_flow_spec *spec, bool ft_locked) @@ -1615,14 +1606,8 @@ static int build_match_list(struct match_list_head *match_head, rhl_for_each_entry_rcu(g, tmp, list, hash) { struct match_list *curr_match; - if (likely(list_empty(&match_head->list))) { - if (!tree_get_node(&g->node)) - continue; - match_head->first.g = g; - list_add_tail(&match_head->first.list, - &match_head->list); + if (unlikely(!tree_get_node(&g->node))) continue; - } curr_match = kmalloc(sizeof(*curr_match), GFP_ATOMIC); if (!curr_match) { @@ -1630,10 +1615,6 @@ static int build_match_list(struct match_list_head *match_head, err = -ENOMEM; goto out; } - if (!tree_get_node(&g->node)) { - kfree(curr_match); - continue; - } curr_match->g = g; list_add_tail(&curr_match->list, &match_head->list); } @@ -1699,7 +1680,7 @@ try_add_to_existing_fg(struct mlx5_flow_table *ft, struct match_list *iter; bool take_write = false; struct fs_fte *fte; - u64 version; + u64 version = 0; int err; fte = alloc_fte(ft, spec, flow_act); @@ -1707,10 +1688,12 @@ try_add_to_existing_fg(struct mlx5_flow_table *ft, return ERR_PTR(-ENOMEM); search_again_locked: - version = matched_fgs_get_version(match_head); if (flow_act->flags & FLOW_ACT_NO_APPEND) goto skip_search; - /* Try to find a fg that already contains a matching fte */ + version = matched_fgs_get_version(match_head); + /* Try to find an fte with identical match value and attempt update its + * action. + */ list_for_each_entry(iter, match_head, list) { struct fs_fte *fte_tmp; @@ -1738,10 +1721,12 @@ skip_search: goto out; } - /* Check the fgs version, for case the new FTE with the - * same values was added while the fgs weren't locked + /* Check the fgs version. If version have changed it could be that an + * FTE with the same match value was added while the fgs weren't + * locked. */ - if (version != matched_fgs_get_version(match_head)) { + if (!(flow_act->flags & FLOW_ACT_NO_APPEND) && + version != matched_fgs_get_version(match_head)) { take_write = true; goto search_again_locked; } @@ -1785,9 +1770,9 @@ _mlx5_add_flow_rules(struct mlx5_flow_table *ft, { struct mlx5_flow_steering *steering = get_steering(&ft->node); - struct mlx5_flow_group *g; struct mlx5_flow_handle *rule; - struct match_list_head match_head; + struct match_list match_head; + struct mlx5_flow_group *g; bool take_write = false; struct fs_fte *fte; int version; @@ -1892,12 +1877,16 @@ mlx5_add_flow_rules(struct mlx5_flow_table *ft, int num_dest) { struct mlx5_flow_root_namespace *root = find_root(&ft->node); + static const struct mlx5_flow_spec zero_spec = {}; struct mlx5_flow_destination gen_dest = {}; struct mlx5_flow_table *next_ft = NULL; struct mlx5_flow_handle *handle = NULL; u32 sw_action = flow_act->action; struct fs_prio *prio; + if (!spec) + spec = &zero_spec; + fs_get_obj(prio, ft->node.parent); if (flow_act->action == MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO) { if (!fwd_next_prio_supported(ft)) @@ -2700,6 +2689,17 @@ static int init_fdb_root_ns(struct mlx5_flow_steering *steering) goto out_err; } + /* We put this priority last, knowing that nothing will get here + * unless explicitly forwarded to. This is possible because the + * slow path tables have catch all rules and nothing gets passed + * those tables. + */ + maj_prio = fs_create_prio(&steering->fdb_root_ns->ns, FDB_PER_VPORT, 1); + if (IS_ERR(maj_prio)) { + err = PTR_ERR(maj_prio); + goto out_err; + } + set_prio_attrs(steering->fdb_root_ns); return 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c index ab69effb056d..f43caefd07a1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c @@ -470,7 +470,7 @@ struct mlx5_fc_bulk { u32 base_id; int bulk_len; unsigned long *bitmask; - struct mlx5_fc fcs[0]; + struct mlx5_fc fcs[]; }; static void mlx5_fc_init(struct mlx5_fc *counter, struct mlx5_fc_bulk *bulk, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index 909a7f284614..90e3d0233101 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -613,6 +613,44 @@ static void mlx5_fsm_release(struct mlxfw_dev *mlxfw_dev, u32 fwhandle) fwhandle, 0); } +#define MLX5_FSM_REACTIVATE_TOUT 5000 /* msecs */ +static int mlx5_fsm_reactivate(struct mlxfw_dev *mlxfw_dev, u8 *status) +{ + unsigned long exp_time = jiffies + msecs_to_jiffies(MLX5_FSM_REACTIVATE_TOUT); + struct mlx5_mlxfw_dev *mlx5_mlxfw_dev = + container_of(mlxfw_dev, struct mlx5_mlxfw_dev, mlxfw_dev); + struct mlx5_core_dev *dev = mlx5_mlxfw_dev->mlx5_core_dev; + u32 out[MLX5_ST_SZ_DW(mirc_reg)]; + u32 in[MLX5_ST_SZ_DW(mirc_reg)]; + int err; + + if (!MLX5_CAP_MCAM_REG2(dev, mirc)) + return -EOPNOTSUPP; + + memset(in, 0, sizeof(in)); + + err = mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_MIRC, 0, 1); + if (err) + return err; + + do { + memset(out, 0, sizeof(out)); + err = mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_MIRC, 0, 0); + if (err) + return err; + + *status = MLX5_GET(mirc_reg, out, status_code); + if (*status != MLXFW_FSM_REACTIVATE_STATUS_BUSY) + return 0; + + msleep(20); + } while (time_before(jiffies, exp_time)); + + return 0; +} + static const struct mlxfw_dev_ops mlx5_mlxfw_dev_ops = { .component_query = mlx5_component_query, .fsm_lock = mlx5_fsm_lock, @@ -620,6 +658,7 @@ static const struct mlxfw_dev_ops mlx5_mlxfw_dev_ops = { .fsm_block_download = mlx5_fsm_block_download, .fsm_component_verify = mlx5_fsm_component_verify, .fsm_activate = mlx5_fsm_activate, + .fsm_reactivate = mlx5_fsm_reactivate, .fsm_query_state = mlx5_fsm_query_state, .fsm_cancel = mlx5_fsm_cancel, .fsm_release = mlx5_fsm_release @@ -634,6 +673,7 @@ int mlx5_firmware_flash(struct mlx5_core_dev *dev, .ops = &mlx5_mlxfw_dev_ops, .psid = dev->board_id, .psid_size = strlen(dev->board_id), + .devlink = priv_to_devlink(dev), }, .mlx5_core_dev = dev }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c index d9f4e8c59c1f..fa1665caac46 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c @@ -627,7 +627,7 @@ static void mlx5_fw_reporters_create(struct mlx5_core_dev *dev) health->fw_reporter = devlink_health_reporter_create(devlink, &mlx5_fw_reporter_ops, - 0, false, dev); + 0, dev); if (IS_ERR(health->fw_reporter)) mlx5_core_warn(dev, "Failed to create fw reporter, err = %ld\n", PTR_ERR(health->fw_reporter)); @@ -636,7 +636,7 @@ static void mlx5_fw_reporters_create(struct mlx5_core_dev *dev) devlink_health_reporter_create(devlink, &mlx5_fw_fatal_reporter_ops, MLX5_REPORTER_FW_GRACEFUL_PERIOD, - true, dev); + dev); if (IS_ERR(health->fw_fatal_reporter)) mlx5_core_warn(dev, "Failed to create fw fatal reporter, err = %ld\n", PTR_ERR(health->fw_fatal_reporter)); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c index 90cb50fe17fd..1eef66ee849e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c @@ -235,6 +235,9 @@ static int mlx5i_get_link_ksettings(struct net_device *netdev, } const struct ethtool_ops mlx5i_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE, .get_drvinfo = mlx5i_get_drvinfo, .get_strings = mlx5i_get_strings, .get_sset_count = mlx5i_get_sset_count, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 56078b23f1a0..673aaa815f57 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -483,7 +483,7 @@ static int mlx5i_change_mtu(struct net_device *netdev, int new_mtu) new_channels.params = *params; new_channels.params.sw_mtu = new_mtu; - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); if (err) goto out; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h index c87962cab921..de7e01a027bb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h @@ -56,7 +56,7 @@ struct mlx5i_priv { u32 qkey; u16 pkey_index; struct mlx5i_pkey_qpn_ht *qpn_htbl; - char *mlx5e_priv[0]; + char *mlx5e_priv[]; }; int mlx5i_create_tis(struct mlx5_core_dev *mdev, u32 underlay_qpn, u32 *tisn); @@ -107,7 +107,7 @@ struct mlx5i_tx_wqe { struct mlx5_wqe_datagram_seg datagram; struct mlx5_wqe_eth_pad pad; struct mlx5_wqe_eth_seg eth; - struct mlx5_wqe_data_seg data[0]; + struct mlx5_wqe_data_seg data[]; }; static inline void mlx5i_sq_fetch_wqe(struct mlx5e_txqsq *sq, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c index 416676c35b1f..e9089a793632 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c @@ -93,9 +93,8 @@ static void mlx5_lag_set_port_affinity(struct mlx5_lag *ldev, static void mlx5_lag_fib_event_flush(struct notifier_block *nb) { struct lag_mp *mp = container_of(nb, struct lag_mp, fib_nb); - struct mlx5_lag *ldev = container_of(mp, struct mlx5_lag, lag_mp); - flush_workqueue(ldev->wq); + flush_workqueue(mp->wq); } struct mlx5_fib_event_work { @@ -293,7 +292,7 @@ static int mlx5_lag_fib_event(struct notifier_block *nb, return NOTIFY_DONE; } - queue_work(ldev->wq, &fib_work->work); + queue_work(mp->wq, &fib_work->work); return NOTIFY_DONE; } @@ -306,11 +305,17 @@ int mlx5_lag_mp_init(struct mlx5_lag *ldev) if (mp->fib_nb.notifier_call) return 0; + mp->wq = create_singlethread_workqueue("mlx5_lag_mp"); + if (!mp->wq) + return -ENOMEM; + mp->fib_nb.notifier_call = mlx5_lag_fib_event; err = register_fib_notifier(&init_net, &mp->fib_nb, mlx5_lag_fib_event_flush, NULL); - if (err) + if (err) { + destroy_workqueue(mp->wq); mp->fib_nb.notifier_call = NULL; + } return err; } @@ -323,5 +328,6 @@ void mlx5_lag_mp_cleanup(struct mlx5_lag *ldev) return; unregister_fib_notifier(&init_net, &mp->fib_nb); + destroy_workqueue(mp->wq); mp->fib_nb.notifier_call = NULL; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.h b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.h index 79be89e9c7a4..258ac7b2964e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.h @@ -16,6 +16,7 @@ enum mlx5_lag_port_affinity { struct lag_mp { struct notifier_block fib_nb; struct fib_info *mfi; /* used in tracking fib events */ + struct workqueue_struct *wq; }; #ifdef CONFIG_MLX5_ESWITCH diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/crypto.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/crypto.c index 3fc575d1c3ec..dcea87ec5977 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/crypto.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/crypto.c @@ -42,7 +42,7 @@ int mlx5_create_encryption_key(struct mlx5_core_dev *mdev, MLX5_SET(encryption_key_obj, obj, key_size, general_obj_key_size); MLX5_SET(encryption_key_obj, obj, key_type, - MLX5_GENERAL_OBJECT_TYPE_ENCRYPTION_KEY_TYPE_DEK); + MLX5_GENERAL_OBJECT_TYPE_ENCRYPTION_KEY_TYPE_TLS); MLX5_SET(general_obj_in_cmd_hdr, in, opcode, MLX5_CMD_OP_CREATE_GENERAL_OBJECT); MLX5_SET(general_obj_in_cmd_hdr, in, obj_type, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/dm.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/dm.c index e065c2f68f5a..6cbccba56f70 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/dm.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/dm.c @@ -21,7 +21,7 @@ struct mlx5_dm *mlx5_dm_create(struct mlx5_core_dev *dev) struct mlx5_dm *dm; if (!(MLX5_CAP_GEN_64(dev, general_obj_types) & MLX5_GENERAL_OBJ_TYPES_CAP_SW_ICM)) - return 0; + return NULL; dm = kzalloc(sizeof(*dm), GFP_KERNEL); if (!dm) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index f554cfddcf4e..7af4210c1b96 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -70,6 +70,7 @@ #include "diag/fw_tracer.h" #include "ecpf.h" #include "lib/hv_vhca.h" +#include "diag/rsc_dump.h" MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>"); MODULE_DESCRIPTION("Mellanox 5th generation network adapters (ConnectX series) core driver"); @@ -880,6 +881,7 @@ static int mlx5_init_once(struct mlx5_core_dev *dev) dev->tracer = mlx5_fw_tracer_create(dev); dev->hv_vhca = mlx5_hv_vhca_create(dev); + dev->rsc_dump = mlx5_rsc_dump_create(dev); return 0; @@ -909,6 +911,7 @@ err_devcom: static void mlx5_cleanup_once(struct mlx5_core_dev *dev) { + mlx5_rsc_dump_destroy(dev); mlx5_hv_vhca_destroy(dev->hv_vhca); mlx5_fw_tracer_destroy(dev->tracer); mlx5_dm_cleanup(dev); @@ -1079,6 +1082,12 @@ static int mlx5_load(struct mlx5_core_dev *dev) mlx5_hv_vhca_init(dev->hv_vhca); + err = mlx5_rsc_dump_init(dev); + if (err) { + mlx5_core_err(dev, "Failed to init Resource dump\n"); + goto err_rsc_dump; + } + err = mlx5_fpga_device_start(dev); if (err) { mlx5_core_err(dev, "fpga device start failed %d\n", err); @@ -1134,6 +1143,8 @@ err_tls_start: err_ipsec_start: mlx5_fpga_device_stop(dev); err_fpga_start: + mlx5_rsc_dump_cleanup(dev); +err_rsc_dump: mlx5_hv_vhca_cleanup(dev->hv_vhca); mlx5_fw_tracer_cleanup(dev->tracer); err_fw_tracer: @@ -1155,6 +1166,7 @@ static void mlx5_unload(struct mlx5_core_dev *dev) mlx5_accel_ipsec_cleanup(dev); mlx5_accel_tls_cleanup(dev); mlx5_fpga_device_stop(dev); + mlx5_rsc_dump_cleanup(dev); mlx5_hv_vhca_cleanup(dev->hv_vhca); mlx5_fw_tracer_cleanup(dev->tracer); mlx5_eq_table_destroy(dev); @@ -1199,15 +1211,10 @@ int mlx5_load_one(struct mlx5_core_dev *dev, bool boot) goto err_devlink_reg; } - if (mlx5_device_registered(dev)) { + if (mlx5_device_registered(dev)) mlx5_attach_device(dev); - } else { - err = mlx5_register_device(dev); - if (err) { - mlx5_core_err(dev, "register device failed %d\n", err); - goto err_reg_dev; - } - } + else + mlx5_register_device(dev); set_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state); out: @@ -1215,9 +1222,6 @@ out: return err; -err_reg_dev: - if (boot) - mlx5_devlink_unregister(priv_to_devlink(dev)); err_devlink_reg: mlx5_unload(dev); err_load: @@ -1231,7 +1235,7 @@ function_teardown: return err; } -int mlx5_unload_one(struct mlx5_core_dev *dev, bool cleanup) +void mlx5_unload_one(struct mlx5_core_dev *dev, bool cleanup) { if (cleanup) { mlx5_unregister_device(dev); @@ -1260,7 +1264,6 @@ int mlx5_unload_one(struct mlx5_core_dev *dev, bool cleanup) mlx5_function_teardown(dev, cleanup); out: mutex_unlock(&dev->intf_state_mutex); - return 0; } static int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx) @@ -1282,7 +1285,6 @@ static int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx) mutex_init(&priv->alloc_mutex); mutex_init(&priv->pgdir_mutex); INIT_LIST_HEAD(&priv->pgdir_list); - spin_lock_init(&priv->mkey_lock); priv->dbg_root = debugfs_create_dir(dev_name(dev->device), mlx5_debugfs_root); @@ -1381,12 +1383,7 @@ static void remove_one(struct pci_dev *pdev) mlx5_crdump_disable(dev); mlx5_devlink_unregister(devlink); - if (mlx5_unload_one(dev, true)) { - mlx5_core_err(dev, "mlx5_unload_one failed\n"); - mlx5_health_flush(dev); - return; - } - + mlx5_unload_one(dev, true); mlx5_pci_close(dev); mlx5_mdev_uninit(dev); mlx5_devlink_free(devlink); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index da67b28d6e23..a8fb43a85d1d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -182,7 +182,7 @@ void mlx5_remove_device(struct mlx5_interface *intf, struct mlx5_priv *priv); void mlx5_attach_device(struct mlx5_core_dev *dev); void mlx5_detach_device(struct mlx5_core_dev *dev); bool mlx5_device_registered(struct mlx5_core_dev *dev); -int mlx5_register_device(struct mlx5_core_dev *dev); +void mlx5_register_device(struct mlx5_core_dev *dev); void mlx5_unregister_device(struct mlx5_core_dev *dev); void mlx5_add_dev_by_protocol(struct mlx5_core_dev *dev, int protocol); void mlx5_remove_dev_by_protocol(struct mlx5_core_dev *dev, int protocol); @@ -244,6 +244,6 @@ enum { u8 mlx5_get_nic_state(struct mlx5_core_dev *dev); void mlx5_set_nic_state(struct mlx5_core_dev *dev, u8 state); -int mlx5_unload_one(struct mlx5_core_dev *dev, bool cleanup); +void mlx5_unload_one(struct mlx5_core_dev *dev, bool cleanup); int mlx5_load_one(struct mlx5_core_dev *dev, bool boot); #endif /* __MLX5_CORE_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mr.c b/drivers/net/ethernet/mellanox/mlx5/core/mr.c index 42cc3c7ac5b6..366f2cbfc6db 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/mr.c @@ -36,54 +36,31 @@ #include <linux/mlx5/cmd.h> #include "mlx5_core.h" -int mlx5_core_create_mkey_cb(struct mlx5_core_dev *dev, - struct mlx5_core_mkey *mkey, - struct mlx5_async_ctx *async_ctx, u32 *in, - int inlen, u32 *out, int outlen, - mlx5_async_cbk_t callback, - struct mlx5_async_work *context) +int mlx5_core_create_mkey(struct mlx5_core_dev *dev, + struct mlx5_core_mkey *mkey, + u32 *in, int inlen) { u32 lout[MLX5_ST_SZ_DW(create_mkey_out)] = {0}; u32 mkey_index; void *mkc; int err; - u8 key; - - spin_lock_irq(&dev->priv.mkey_lock); - key = dev->priv.mkey_key++; - spin_unlock_irq(&dev->priv.mkey_lock); - mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry); MLX5_SET(create_mkey_in, in, opcode, MLX5_CMD_OP_CREATE_MKEY); - MLX5_SET(mkc, mkc, mkey_7_0, key); - - if (callback) - return mlx5_cmd_exec_cb(async_ctx, in, inlen, out, outlen, - callback, context); err = mlx5_cmd_exec(dev, in, inlen, lout, sizeof(lout)); if (err) return err; + mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry); mkey_index = MLX5_GET(create_mkey_out, lout, mkey_index); mkey->iova = MLX5_GET64(mkc, mkc, start_addr); mkey->size = MLX5_GET64(mkc, mkc, len); - mkey->key = mlx5_idx_to_mkey(mkey_index) | key; + mkey->key |= mlx5_idx_to_mkey(mkey_index); mkey->pd = MLX5_GET(mkc, mkc, pd); - mlx5_core_dbg(dev, "out 0x%x, key 0x%x, mkey 0x%x\n", - mkey_index, key, mkey->key); + mlx5_core_dbg(dev, "out 0x%x, mkey 0x%x\n", mkey_index, mkey->key); return 0; } -EXPORT_SYMBOL(mlx5_core_create_mkey_cb); - -int mlx5_core_create_mkey(struct mlx5_core_dev *dev, - struct mlx5_core_mkey *mkey, - u32 *in, int inlen) -{ - return mlx5_core_create_mkey_cb(dev, mkey, NULL, in, inlen, - NULL, 0, NULL, NULL); -} EXPORT_SYMBOL(mlx5_core_create_mkey); int mlx5_core_destroy_mkey(struct mlx5_core_dev *dev, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rl.c b/drivers/net/ethernet/mellanox/mlx5/core/rl.c index 01c380425f9d..f3b29d9ade1f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/rl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/rl.c @@ -101,22 +101,39 @@ int mlx5_destroy_scheduling_element_cmd(struct mlx5_core_dev *dev, u8 hierarchy, return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); } +static bool mlx5_rl_are_equal_raw(struct mlx5_rl_entry *entry, void *rl_in, + u16 uid) +{ + return (!memcmp(entry->rl_raw, rl_in, sizeof(entry->rl_raw)) && + entry->uid == uid); +} + /* Finds an entry where we can register the given rate * If the rate already exists, return the entry where it is registered, * otherwise return the first available entry. * If the table is full, return NULL */ static struct mlx5_rl_entry *find_rl_entry(struct mlx5_rl_table *table, - struct mlx5_rate_limit *rl) + void *rl_in, u16 uid, bool dedicated) { struct mlx5_rl_entry *ret_entry = NULL; bool empty_found = false; int i; for (i = 0; i < table->max_size; i++) { - if (mlx5_rl_are_equal(&table->rl_entry[i].rl, rl)) - return &table->rl_entry[i]; - if (!empty_found && !table->rl_entry[i].rl.rate) { + if (dedicated) { + if (!table->rl_entry[i].refcount) + return &table->rl_entry[i]; + continue; + } + + if (table->rl_entry[i].refcount) { + if (table->rl_entry[i].dedicated) + continue; + if (mlx5_rl_are_equal_raw(&table->rl_entry[i], rl_in, + uid)) + return &table->rl_entry[i]; + } else if (!empty_found) { empty_found = true; ret_entry = &table->rl_entry[i]; } @@ -126,18 +143,19 @@ static struct mlx5_rl_entry *find_rl_entry(struct mlx5_rl_table *table, } static int mlx5_set_pp_rate_limit_cmd(struct mlx5_core_dev *dev, - u16 index, - struct mlx5_rate_limit *rl) + struct mlx5_rl_entry *entry, bool set) { - u32 in[MLX5_ST_SZ_DW(set_pp_rate_limit_in)] = {0}; - u32 out[MLX5_ST_SZ_DW(set_pp_rate_limit_out)] = {0}; + u32 in[MLX5_ST_SZ_DW(set_pp_rate_limit_in)] = {}; + u32 out[MLX5_ST_SZ_DW(set_pp_rate_limit_out)] = {}; + void *pp_context; + pp_context = MLX5_ADDR_OF(set_pp_rate_limit_in, in, ctx); MLX5_SET(set_pp_rate_limit_in, in, opcode, MLX5_CMD_OP_SET_PP_RATE_LIMIT); - MLX5_SET(set_pp_rate_limit_in, in, rate_limit_index, index); - MLX5_SET(set_pp_rate_limit_in, in, rate_limit, rl->rate); - MLX5_SET(set_pp_rate_limit_in, in, burst_upper_bound, rl->max_burst_sz); - MLX5_SET(set_pp_rate_limit_in, in, typical_packet_size, rl->typical_pkt_sz); + MLX5_SET(set_pp_rate_limit_in, in, uid, entry->uid); + MLX5_SET(set_pp_rate_limit_in, in, rate_limit_index, entry->index); + if (set) + memcpy(pp_context, entry->rl_raw, sizeof(entry->rl_raw)); return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); } @@ -158,23 +176,25 @@ bool mlx5_rl_are_equal(struct mlx5_rate_limit *rl_0, } EXPORT_SYMBOL(mlx5_rl_are_equal); -int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u16 *index, - struct mlx5_rate_limit *rl) +int mlx5_rl_add_rate_raw(struct mlx5_core_dev *dev, void *rl_in, u16 uid, + bool dedicated_entry, u16 *index) { struct mlx5_rl_table *table = &dev->priv.rl_table; struct mlx5_rl_entry *entry; int err = 0; + u32 rate; + rate = MLX5_GET(set_pp_rate_limit_context, rl_in, rate_limit); mutex_lock(&table->rl_lock); - if (!rl->rate || !mlx5_rl_is_in_range(dev, rl->rate)) { + if (!rate || !mlx5_rl_is_in_range(dev, rate)) { mlx5_core_err(dev, "Invalid rate: %u, should be %u to %u\n", - rl->rate, table->min_rate, table->max_rate); + rate, table->min_rate, table->max_rate); err = -EINVAL; goto out; } - entry = find_rl_entry(table, rl); + entry = find_rl_entry(table, rl_in, uid, dedicated_entry); if (!entry) { mlx5_core_err(dev, "Max number of %u rates reached\n", table->max_size); @@ -185,16 +205,24 @@ int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u16 *index, /* rate already configured */ entry->refcount++; } else { + memcpy(entry->rl_raw, rl_in, sizeof(entry->rl_raw)); + entry->uid = uid; /* new rate limit */ - err = mlx5_set_pp_rate_limit_cmd(dev, entry->index, rl); + err = mlx5_set_pp_rate_limit_cmd(dev, entry, true); if (err) { - mlx5_core_err(dev, "Failed configuring rate limit(err %d): rate %u, max_burst_sz %u, typical_pkt_sz %u\n", - err, rl->rate, rl->max_burst_sz, - rl->typical_pkt_sz); + mlx5_core_err( + dev, + "Failed configuring rate limit(err %d): rate %u, max_burst_sz %u, typical_pkt_sz %u\n", + err, rate, + MLX5_GET(set_pp_rate_limit_context, rl_in, + burst_upper_bound), + MLX5_GET(set_pp_rate_limit_context, rl_in, + typical_packet_size)); goto out; } - entry->rl = *rl; + entry->refcount = 1; + entry->dedicated = dedicated_entry; } *index = entry->index; @@ -202,20 +230,61 @@ out: mutex_unlock(&table->rl_lock); return err; } +EXPORT_SYMBOL(mlx5_rl_add_rate_raw); + +void mlx5_rl_remove_rate_raw(struct mlx5_core_dev *dev, u16 index) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + struct mlx5_rl_entry *entry; + + mutex_lock(&table->rl_lock); + entry = &table->rl_entry[index - 1]; + entry->refcount--; + if (!entry->refcount) + /* need to remove rate */ + mlx5_set_pp_rate_limit_cmd(dev, entry, false); + mutex_unlock(&table->rl_lock); +} +EXPORT_SYMBOL(mlx5_rl_remove_rate_raw); + +int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u16 *index, + struct mlx5_rate_limit *rl) +{ + u8 rl_raw[MLX5_ST_SZ_BYTES(set_pp_rate_limit_context)] = {}; + + MLX5_SET(set_pp_rate_limit_context, rl_raw, rate_limit, rl->rate); + MLX5_SET(set_pp_rate_limit_context, rl_raw, burst_upper_bound, + rl->max_burst_sz); + MLX5_SET(set_pp_rate_limit_context, rl_raw, typical_packet_size, + rl->typical_pkt_sz); + + return mlx5_rl_add_rate_raw(dev, rl_raw, + MLX5_CAP_QOS(dev, packet_pacing_uid) ? + MLX5_SHARED_RESOURCE_UID : 0, + false, index); +} EXPORT_SYMBOL(mlx5_rl_add_rate); void mlx5_rl_remove_rate(struct mlx5_core_dev *dev, struct mlx5_rate_limit *rl) { + u8 rl_raw[MLX5_ST_SZ_BYTES(set_pp_rate_limit_context)] = {}; struct mlx5_rl_table *table = &dev->priv.rl_table; struct mlx5_rl_entry *entry = NULL; - struct mlx5_rate_limit reset_rl = {0}; /* 0 is a reserved value for unlimited rate */ if (rl->rate == 0) return; + MLX5_SET(set_pp_rate_limit_context, rl_raw, rate_limit, rl->rate); + MLX5_SET(set_pp_rate_limit_context, rl_raw, burst_upper_bound, + rl->max_burst_sz); + MLX5_SET(set_pp_rate_limit_context, rl_raw, typical_packet_size, + rl->typical_pkt_sz); + mutex_lock(&table->rl_lock); - entry = find_rl_entry(table, rl); + entry = find_rl_entry(table, rl_raw, + MLX5_CAP_QOS(dev, packet_pacing_uid) ? + MLX5_SHARED_RESOURCE_UID : 0, false); if (!entry || !entry->refcount) { mlx5_core_warn(dev, "Rate %u, max_burst_sz %u typical_pkt_sz %u are not configured\n", rl->rate, rl->max_burst_sz, rl->typical_pkt_sz); @@ -223,11 +292,9 @@ void mlx5_rl_remove_rate(struct mlx5_core_dev *dev, struct mlx5_rate_limit *rl) } entry->refcount--; - if (!entry->refcount) { + if (!entry->refcount) /* need to remove rate */ - mlx5_set_pp_rate_limit_cmd(dev, entry->index, &reset_rl); - entry->rl = reset_rl; - } + mlx5_set_pp_rate_limit_cmd(dev, entry, false); out: mutex_unlock(&table->rl_lock); @@ -273,14 +340,13 @@ int mlx5_init_rl_table(struct mlx5_core_dev *dev) void mlx5_cleanup_rl_table(struct mlx5_core_dev *dev) { struct mlx5_rl_table *table = &dev->priv.rl_table; - struct mlx5_rate_limit rl = {0}; int i; /* Clear all configured rates */ for (i = 0; i < table->max_size; i++) - if (table->rl_entry[i].rl.rate) - mlx5_set_pp_rate_limit_cmd(dev, table->rl_entry[i].index, - &rl); + if (table->rl_entry[i].refcount) + mlx5_set_pp_rate_limit_cmd(dev, &table->rl_entry[i], + false); kfree(dev->priv.rl_table.rl_entry); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c index 03f037811f1d..3094d20297a9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c @@ -77,8 +77,7 @@ static int mlx5_device_enable_sriov(struct mlx5_core_dev *dev, int num_vfs) if (!MLX5_ESWITCH_MANAGER(dev)) goto enable_vfs_hca; - mlx5_eswitch_update_num_of_vfs(dev->priv.eswitch, num_vfs); - err = mlx5_eswitch_enable(dev->priv.eswitch, MLX5_ESWITCH_LEGACY); + err = mlx5_eswitch_enable(dev->priv.eswitch, num_vfs); if (err) { mlx5_core_warn(dev, "failed to enable eswitch SRIOV (%d)\n", err); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c index 2d93228ff633..554811de4c9d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c @@ -672,7 +672,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher, dest_action = action; if (!action->dest_tbl.is_fw_tbl) { if (action->dest_tbl.tbl->dmn != dmn) { - mlx5dr_dbg(dmn, + mlx5dr_err(dmn, "Destination table belongs to a different domain\n"); goto out_invalid_arg; } @@ -703,7 +703,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher, action->dest_tbl.fw_tbl.rx_icm_addr = output.sw_owner_icm_root_0; } else { - mlx5dr_dbg(dmn, + mlx5dr_err(dmn, "Failed mlx5_cmd_query_flow_table ret: %d\n", ret); return ret; @@ -772,7 +772,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher, /* Check action duplication */ if (++action_type_set[action_type] > max_actions_type) { - mlx5dr_dbg(dmn, "Action type %d supports only max %d time(s)\n", + mlx5dr_err(dmn, "Action type %d supports only max %d time(s)\n", action_type, max_actions_type); goto out_invalid_arg; } @@ -781,7 +781,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher, if (dr_action_validate_and_get_next_state(action_domain, action_type, &state)) { - mlx5dr_dbg(dmn, "Invalid action sequence provided\n"); + mlx5dr_err(dmn, "Invalid action sequence provided\n"); return -EOPNOTSUPP; } } @@ -797,7 +797,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher, rx_rule && recalc_cs_required && dest_action) { ret = dr_action_handle_cs_recalc(dmn, dest_action, &attr.final_icm_addr); if (ret) { - mlx5dr_dbg(dmn, + mlx5dr_err(dmn, "Failed to handle checksum recalculation err %d\n", ret); return ret; @@ -964,6 +964,24 @@ struct mlx5dr_action *mlx5dr_action_create_drop(void) } struct mlx5dr_action * +mlx5dr_action_create_dest_table_num(struct mlx5dr_domain *dmn, u32 table_num) +{ + struct mlx5dr_action *action; + + action = dr_action_create_generic(DR_ACTION_TYP_FT); + if (!action) + return NULL; + + action->dest_tbl.is_fw_tbl = true; + action->dest_tbl.fw_tbl.dmn = dmn; + action->dest_tbl.fw_tbl.id = table_num; + action->dest_tbl.fw_tbl.type = FS_FT_FDB; + refcount_inc(&dmn->refcount); + + return action; +} + +struct mlx5dr_action * mlx5dr_action_create_dest_table(struct mlx5dr_table *tbl) { struct mlx5dr_action *action; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c index a9da961d4d2f..48b6358b6845 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c @@ -59,7 +59,7 @@ static int dr_domain_init_resources(struct mlx5dr_domain *dmn) ret = mlx5_core_alloc_pd(dmn->mdev, &dmn->pdn); if (ret) { - mlx5dr_dbg(dmn, "Couldn't allocate PD\n"); + mlx5dr_err(dmn, "Couldn't allocate PD, ret: %d", ret); return ret; } @@ -192,7 +192,7 @@ static int dr_domain_query_fdb_caps(struct mlx5_core_dev *mdev, ret = dr_domain_query_vports(dmn); if (ret) { - mlx5dr_dbg(dmn, "Failed to query vports caps\n"); + mlx5dr_err(dmn, "Failed to query vports caps (err: %d)", ret); goto free_vports_caps; } @@ -213,7 +213,7 @@ static int dr_domain_caps_init(struct mlx5_core_dev *mdev, int ret; if (MLX5_CAP_GEN(mdev, port_type) != MLX5_CAP_PORT_TYPE_ETH) { - mlx5dr_dbg(dmn, "Failed to allocate domain, bad link type\n"); + mlx5dr_err(dmn, "Failed to allocate domain, bad link type\n"); return -EOPNOTSUPP; } @@ -257,7 +257,7 @@ static int dr_domain_caps_init(struct mlx5_core_dev *mdev, dmn->info.tx.ste_type = MLX5DR_STE_TYPE_TX; vport_cap = mlx5dr_get_vport_cap(&dmn->info.caps, 0); if (!vport_cap) { - mlx5dr_dbg(dmn, "Failed to get esw manager vport\n"); + mlx5dr_err(dmn, "Failed to get esw manager vport\n"); return -ENOENT; } @@ -268,7 +268,7 @@ static int dr_domain_caps_init(struct mlx5_core_dev *mdev, dmn->info.tx.drop_icm_addr = dmn->info.caps.esw_tx_drop_address; break; default: - mlx5dr_dbg(dmn, "Invalid domain\n"); + mlx5dr_err(dmn, "Invalid domain\n"); ret = -EINVAL; break; } @@ -300,7 +300,7 @@ mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type) mutex_init(&dmn->mutex); if (dr_domain_caps_init(mdev, dmn)) { - mlx5dr_dbg(dmn, "Failed init domain, no caps\n"); + mlx5dr_err(dmn, "Failed init domain, no caps\n"); goto free_domain; } @@ -348,8 +348,11 @@ int mlx5dr_domain_sync(struct mlx5dr_domain *dmn, u32 flags) mutex_lock(&dmn->mutex); ret = mlx5dr_send_ring_force_drain(dmn); mutex_unlock(&dmn->mutex); - if (ret) + if (ret) { + mlx5dr_err(dmn, "Force drain failed flags: %d, ret: %d\n", + flags, ret); return ret; + } } if (flags & MLX5DR_DOMAIN_SYNC_FLAGS_HW) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c index d7c7467e2d53..30d2d7376f56 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c @@ -468,7 +468,7 @@ mlx5dr_icm_alloc_chunk(struct mlx5dr_icm_pool *pool, err = mlx5dr_cmd_sync_steering(pool->dmn->mdev); if (err) { dr_icm_chill_buckets_abort(pool, bucket, buckets); - mlx5dr_dbg(pool->dmn, "Sync_steering failed\n"); + mlx5dr_err(pool->dmn, "Sync_steering failed\n"); chunk = NULL; goto out; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c index c6dbd856df94..a95938874798 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c @@ -388,14 +388,14 @@ static int dr_matcher_set_ste_builders(struct mlx5dr_matcher *matcher, mlx5dr_ste_build_empty_always_hit(&sb[idx++], rx); if (idx == 0) { - mlx5dr_dbg(dmn, "Cannot generate any valid rules from mask\n"); + mlx5dr_err(dmn, "Cannot generate any valid rules from mask\n"); return -EINVAL; } /* Check that all mask fields were consumed */ for (i = 0; i < sizeof(struct mlx5dr_match_param); i++) { if (((u8 *)&mask)[i] != 0) { - mlx5dr_info(dmn, "Mask contains unsupported parameters\n"); + mlx5dr_err(dmn, "Mask contains unsupported parameters\n"); return -EOPNOTSUPP; } } @@ -563,7 +563,7 @@ static int dr_matcher_set_all_ste_builders(struct mlx5dr_matcher *matcher, dr_matcher_set_ste_builders(matcher, nic_matcher, DR_RULE_IPV6, DR_RULE_IPV6); if (!nic_matcher->ste_builder) { - mlx5dr_dbg(dmn, "Cannot generate IPv4 or IPv6 rules with given mask\n"); + mlx5dr_err(dmn, "Cannot generate IPv4 or IPv6 rules with given mask\n"); return -EINVAL; } @@ -634,13 +634,13 @@ static int dr_matcher_init(struct mlx5dr_matcher *matcher, int ret; if (matcher->match_criteria >= DR_MATCHER_CRITERIA_MAX) { - mlx5dr_info(dmn, "Invalid match criteria attribute\n"); + mlx5dr_err(dmn, "Invalid match criteria attribute\n"); return -EINVAL; } if (mask) { if (mask->match_sz > sizeof(struct mlx5dr_match_param)) { - mlx5dr_info(dmn, "Invalid match size attribute\n"); + mlx5dr_err(dmn, "Invalid match size attribute\n"); return -EINVAL; } mlx5dr_ste_copy_param(matcher->match_criteria, @@ -671,7 +671,7 @@ static int dr_matcher_init(struct mlx5dr_matcher *matcher, struct mlx5dr_matcher * mlx5dr_matcher_create(struct mlx5dr_table *tbl, - u16 priority, + u32 priority, u8 match_criteria_enable, struct mlx5dr_match_parameters *mask) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c index e4cff7abb348..cce3ee7a6614 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c @@ -826,8 +826,8 @@ again: ste_location, send_ste_list); if (!new_htbl) { mlx5dr_htbl_put(cur_htbl); - mlx5dr_info(dmn, "failed creating rehash table, htbl-log_size: %d\n", - cur_htbl->chunk_size); + mlx5dr_err(dmn, "Failed creating rehash table, htbl-log_size: %d\n", + cur_htbl->chunk_size); } else { cur_htbl = new_htbl; } @@ -877,7 +877,7 @@ static bool dr_rule_verify(struct mlx5dr_matcher *matcher, if (!value_size || (value_size > sizeof(struct mlx5dr_match_param) || (value_size % sizeof(u32)))) { - mlx5dr_dbg(matcher->tbl->dmn, "Rule parameters length is incorrect\n"); + mlx5dr_err(matcher->tbl->dmn, "Rule parameters length is incorrect\n"); return false; } @@ -888,7 +888,7 @@ static bool dr_rule_verify(struct mlx5dr_matcher *matcher, e_idx = min(s_idx + sizeof(param->outer), value_size); if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) { - mlx5dr_dbg(matcher->tbl->dmn, "Rule outer parameters contains a value not specified by mask\n"); + mlx5dr_err(matcher->tbl->dmn, "Rule outer parameters contains a value not specified by mask\n"); return false; } } @@ -898,7 +898,7 @@ static bool dr_rule_verify(struct mlx5dr_matcher *matcher, e_idx = min(s_idx + sizeof(param->misc), value_size); if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) { - mlx5dr_dbg(matcher->tbl->dmn, "Rule misc parameters contains a value not specified by mask\n"); + mlx5dr_err(matcher->tbl->dmn, "Rule misc parameters contains a value not specified by mask\n"); return false; } } @@ -908,7 +908,7 @@ static bool dr_rule_verify(struct mlx5dr_matcher *matcher, e_idx = min(s_idx + sizeof(param->inner), value_size); if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) { - mlx5dr_dbg(matcher->tbl->dmn, "Rule inner parameters contains a value not specified by mask\n"); + mlx5dr_err(matcher->tbl->dmn, "Rule inner parameters contains a value not specified by mask\n"); return false; } } @@ -918,7 +918,7 @@ static bool dr_rule_verify(struct mlx5dr_matcher *matcher, e_idx = min(s_idx + sizeof(param->misc2), value_size); if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) { - mlx5dr_dbg(matcher->tbl->dmn, "Rule misc2 parameters contains a value not specified by mask\n"); + mlx5dr_err(matcher->tbl->dmn, "Rule misc2 parameters contains a value not specified by mask\n"); return false; } } @@ -928,7 +928,7 @@ static bool dr_rule_verify(struct mlx5dr_matcher *matcher, e_idx = min(s_idx + sizeof(param->misc3), value_size); if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) { - mlx5dr_dbg(matcher->tbl->dmn, "Rule misc3 parameters contains a value not specified by mask\n"); + mlx5dr_err(matcher->tbl->dmn, "Rule misc3 parameters contains a value not specified by mask\n"); return false; } } @@ -1221,7 +1221,7 @@ remove_action_members: dr_rule_remove_action_members(rule); free_rule: kfree(rule); - mlx5dr_info(dmn, "Failed creating rule\n"); + mlx5dr_err(dmn, "Failed creating rule\n"); return NULL; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c index 095ec7b1399d..c0ab9cf74929 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c @@ -136,7 +136,7 @@ static struct mlx5dr_qp *dr_create_rc_qp(struct mlx5_core_dev *mdev, err = mlx5_wq_qp_create(mdev, &wqp, temp_qpc, &dr_qp->wq, &dr_qp->wq_ctrl); if (err) { - mlx5_core_info(mdev, "Can't create QP WQ\n"); + mlx5_core_warn(mdev, "Can't create QP WQ\n"); goto err_wq; } @@ -652,8 +652,10 @@ static int dr_prepare_qp_to_rts(struct mlx5dr_domain *dmn) /* Init */ ret = dr_modify_qp_rst2init(dmn->mdev, dr_qp, port); - if (ret) + if (ret) { + mlx5dr_err(dmn, "Failed modify QP rst2init\n"); return ret; + } /* RTR */ ret = mlx5dr_cmd_query_gid(dmn->mdev, port, gid_index, &rtr_attr.dgid_attr); @@ -668,8 +670,10 @@ static int dr_prepare_qp_to_rts(struct mlx5dr_domain *dmn) rtr_attr.udp_src_port = dmn->info.caps.roce_min_src_udp; ret = dr_cmd_modify_qp_init2rtr(dmn->mdev, dr_qp, &rtr_attr); - if (ret) + if (ret) { + mlx5dr_err(dmn, "Failed modify QP init2rtr\n"); return ret; + } /* RTS */ rts_attr.timeout = 14; @@ -677,8 +681,10 @@ static int dr_prepare_qp_to_rts(struct mlx5dr_domain *dmn) rts_attr.rnr_retry = 7; ret = dr_cmd_modify_qp_rtr2rts(dmn->mdev, dr_qp, &rts_attr); - if (ret) + if (ret) { + mlx5dr_err(dmn, "Failed modify QP rtr2rts\n"); return ret; + } return 0; } @@ -862,6 +868,7 @@ int mlx5dr_send_ring_alloc(struct mlx5dr_domain *dmn) cq_size = QUEUE_SIZE + 1; dmn->send_ring->cq = dr_create_cq(dmn->mdev, dmn->uar, cq_size); if (!dmn->send_ring->cq) { + mlx5dr_err(dmn, "Failed creating CQ\n"); ret = -ENOMEM; goto free_send_ring; } @@ -873,6 +880,7 @@ int mlx5dr_send_ring_alloc(struct mlx5dr_domain *dmn) dmn->send_ring->qp = dr_create_rc_qp(dmn->mdev, &init_attr); if (!dmn->send_ring->qp) { + mlx5dr_err(dmn, "Failed creating QP\n"); ret = -ENOMEM; goto clean_cq; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c index aade62a9ee5c..c0e3a1e7389d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c @@ -728,7 +728,7 @@ int mlx5dr_ste_build_pre_check(struct mlx5dr_domain *dmn, { if (!value && (match_criteria & DR_MATCHER_CRITERIA_MISC)) { if (mask->misc.source_port && mask->misc.source_port != 0xffff) { - mlx5dr_dbg(dmn, "Partial mask source_port is not supported\n"); + mlx5dr_err(dmn, "Partial mask source_port is not supported\n"); return -EINVAL; } } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c index 14ce2d7dbb66..c2fe48d7b75a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c @@ -128,16 +128,20 @@ static int dr_table_init_nic(struct mlx5dr_domain *dmn, DR_CHUNK_SIZE_1, MLX5DR_STE_LU_TYPE_DONT_CARE, 0); - if (!nic_tbl->s_anchor) + if (!nic_tbl->s_anchor) { + mlx5dr_err(dmn, "Failed allocating htbl\n"); return -ENOMEM; + } info.type = CONNECT_MISS; info.miss_icm_addr = nic_dmn->default_icm_addr; ret = mlx5dr_ste_htbl_init_and_postsend(dmn, nic_dmn, nic_tbl->s_anchor, &info, true); - if (ret) + if (ret) { + mlx5dr_err(dmn, "Failed int and send htbl\n"); goto free_s_anchor; + } mlx5dr_htbl_get(nic_tbl->s_anchor); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h index dffe35145d19..3fa739951b34 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h @@ -705,7 +705,7 @@ struct mlx5dr_matcher { struct mlx5dr_matcher_rx_tx rx; struct mlx5dr_matcher_rx_tx tx; struct list_head matcher_list; - u16 prio; + u32 prio; struct mlx5dr_match_param mask; u8 match_criteria; refcount_t refcount; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c index c2027192e21e..3b3f5b9d4f95 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c @@ -140,7 +140,7 @@ static int mlx5_cmd_dr_create_flow_group(struct mlx5_flow_root_namespace *ns, struct mlx5_flow_group *fg) { struct mlx5dr_matcher *matcher; - u16 priority = MLX5_GET(create_flow_group_in, in, + u32 priority = MLX5_GET(create_flow_group_in, in, start_flow_index); u8 match_criteria_enable = MLX5_GET(create_flow_group_in, in, @@ -384,6 +384,7 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns, if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { list_for_each_entry(dst, &fte->node.children, node.list) { enum mlx5_flow_destination_type type = dst->dest_attr.type; + u32 ft_id; if (num_actions == MLX5_FLOW_CONTEXT_ACTION_MAX || num_term_actions >= MLX5_FLOW_CONTEXT_ACTION_MAX) { @@ -420,6 +421,17 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns, num_term_actions++; break; + case MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE_NUM: + ft_id = dst->dest_attr.ft_num; + tmp_action = mlx5dr_action_create_dest_table_num(domain, + ft_id); + if (!tmp_action) { + err = -ENOMEM; + goto free_actions; + } + fs_dr_actions[fs_dr_num_actions++] = tmp_action; + term_actions[num_term_actions++].dest = tmp_action; + break; default: err = -EOPNOTSUPP; goto free_actions; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h index e1edc9c247b7..7deaca9ade3b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h @@ -38,8 +38,6 @@ struct mlx5dr_action_dest { struct mlx5dr_action *reformat; }; -#ifdef CONFIG_MLX5_SW_STEERING - struct mlx5dr_domain * mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type); @@ -59,7 +57,7 @@ u32 mlx5dr_table_get_id(struct mlx5dr_table *table); struct mlx5dr_matcher * mlx5dr_matcher_create(struct mlx5dr_table *table, - u16 priority, + u32 priority, u8 match_criteria_enable, struct mlx5dr_match_parameters *mask); @@ -77,6 +75,9 @@ int mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl, struct mlx5dr_action *action); struct mlx5dr_action * +mlx5dr_action_create_dest_table_num(struct mlx5dr_domain *dmn, u32 table_num); + +struct mlx5dr_action * mlx5dr_action_create_dest_table(struct mlx5dr_table *table); struct mlx5dr_action * @@ -125,103 +126,4 @@ mlx5dr_is_supported(struct mlx5_core_dev *dev) return MLX5_CAP_ESW_FLOWTABLE_FDB(dev, sw_owner); } -#else /* CONFIG_MLX5_SW_STEERING */ - -static inline struct mlx5dr_domain * -mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type) { return NULL; } - -static inline int -mlx5dr_domain_destroy(struct mlx5dr_domain *domain) { return 0; } - -static inline int -mlx5dr_domain_sync(struct mlx5dr_domain *domain, u32 flags) { return 0; } - -static inline void -mlx5dr_domain_set_peer(struct mlx5dr_domain *dmn, - struct mlx5dr_domain *peer_dmn) { } - -static inline struct mlx5dr_table * -mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level, u32 flags) { return NULL; } - -static inline int -mlx5dr_table_destroy(struct mlx5dr_table *table) { return 0; } - -static inline u32 -mlx5dr_table_get_id(struct mlx5dr_table *table) { return 0; } - -static inline struct mlx5dr_matcher * -mlx5dr_matcher_create(struct mlx5dr_table *table, - u16 priority, - u8 match_criteria_enable, - struct mlx5dr_match_parameters *mask) { return NULL; } - -static inline int -mlx5dr_matcher_destroy(struct mlx5dr_matcher *matcher) { return 0; } - -static inline struct mlx5dr_rule * -mlx5dr_rule_create(struct mlx5dr_matcher *matcher, - struct mlx5dr_match_parameters *value, - size_t num_actions, - struct mlx5dr_action *actions[]) { return NULL; } - -static inline int -mlx5dr_rule_destroy(struct mlx5dr_rule *rule) { return 0; } - -static inline int -mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl, - struct mlx5dr_action *action) { return 0; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_dest_table(struct mlx5dr_table *table) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_dest_flow_fw_table(struct mlx5dr_domain *domain, - struct mlx5_flow_table *ft) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_dest_vport(struct mlx5dr_domain *domain, - u32 vport, u8 vhca_id_valid, - u16 vhca_id) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn, - struct mlx5dr_action_dest *dests, - u32 num_of_dests) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_drop(void) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_tag(u32 tag_value) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_flow_counter(u32 counter_id) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_packet_reformat(struct mlx5dr_domain *dmn, - enum mlx5dr_action_reformat_type reformat_type, - size_t data_sz, - void *data) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_modify_header(struct mlx5dr_domain *domain, - u32 flags, - size_t actions_sz, - __be64 actions[]) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_pop_vlan(void) { return NULL; } - -static inline struct mlx5dr_action * -mlx5dr_action_create_push_vlan(struct mlx5dr_domain *domain, - __be32 vlan_hdr) { return NULL; } - -static inline int -mlx5dr_action_destroy(struct mlx5dr_action *action) { return 0; } - -static inline bool -mlx5dr_is_supported(struct mlx5_core_dev *dev) { return false; } - -#endif /* CONFIG_MLX5_SW_STEERING */ - #endif /* _MLX5DR_H_ */ diff --git a/drivers/net/ethernet/mellanox/mlxfw/Kconfig b/drivers/net/ethernet/mellanox/mlxfw/Kconfig index 0367f835a846..5b604501f33e 100644 --- a/drivers/net/ethernet/mellanox/mlxfw/Kconfig +++ b/drivers/net/ethernet/mellanox/mlxfw/Kconfig @@ -12,3 +12,4 @@ config MLXFW To compile this driver as a module, choose M here: the module will be called mlxfw. select XZ_DEC + select NET_DEVLINK diff --git a/drivers/net/ethernet/mellanox/mlxfw/mlxfw.h b/drivers/net/ethernet/mellanox/mlxfw/mlxfw.h index c50e74ab02c4..7654841a05c2 100644 --- a/drivers/net/ethernet/mellanox/mlxfw/mlxfw.h +++ b/drivers/net/ethernet/mellanox/mlxfw/mlxfw.h @@ -6,6 +6,30 @@ #include <linux/firmware.h> #include <linux/netlink.h> +#include <linux/device.h> +#include <net/devlink.h> + +struct mlxfw_dev { + const struct mlxfw_dev_ops *ops; + const char *psid; + u16 psid_size; + struct devlink *devlink; +}; + +static inline +struct device *mlxfw_dev_dev(struct mlxfw_dev *mlxfw_dev) +{ + return mlxfw_dev->devlink->dev; +} + +#define MLXFW_PRFX "mlxfw: " + +#define mlxfw_info(mlxfw_dev, fmt, ...) \ + dev_info(mlxfw_dev_dev(mlxfw_dev), MLXFW_PRFX fmt, ## __VA_ARGS__) +#define mlxfw_err(mlxfw_dev, fmt, ...) \ + dev_err(mlxfw_dev_dev(mlxfw_dev), MLXFW_PRFX fmt, ## __VA_ARGS__) +#define mlxfw_dbg(mlxfw_dev, fmt, ...) \ + dev_dbg(mlxfw_dev_dev(mlxfw_dev), MLXFW_PRFX fmt, ## __VA_ARGS__) enum mlxfw_fsm_state { MLXFW_FSM_STATE_IDLE, @@ -31,7 +55,19 @@ enum mlxfw_fsm_state_err { MLXFW_FSM_STATE_ERR_MAX, }; -struct mlxfw_dev; +enum mlxfw_fsm_reactivate_status { + MLXFW_FSM_REACTIVATE_STATUS_OK, + MLXFW_FSM_REACTIVATE_STATUS_BUSY, + MLXFW_FSM_REACTIVATE_STATUS_PROHIBITED_FW_VER_ERR, + MLXFW_FSM_REACTIVATE_STATUS_FIRST_PAGE_COPY_FAILED, + MLXFW_FSM_REACTIVATE_STATUS_FIRST_PAGE_ERASE_FAILED, + MLXFW_FSM_REACTIVATE_STATUS_FIRST_PAGE_RESTORE_FAILED, + MLXFW_FSM_REACTIVATE_STATUS_CANDIDATE_FW_DEACTIVATION_FAILED, + MLXFW_FSM_REACTIVATE_STATUS_FW_ALREADY_ACTIVATED, + MLXFW_FSM_REACTIVATE_STATUS_ERR_DEVICE_RESET_REQUIRED, + MLXFW_FSM_REACTIVATE_STATUS_ERR_FW_PROGRAMMING_NEEDED, + MLXFW_FSM_REACTIVATE_STATUS_MAX, +}; struct mlxfw_dev_ops { int (*component_query)(struct mlxfw_dev *mlxfw_dev, u16 component_index, @@ -51,6 +87,8 @@ struct mlxfw_dev_ops { int (*fsm_activate)(struct mlxfw_dev *mlxfw_dev, u32 fwhandle); + int (*fsm_reactivate)(struct mlxfw_dev *mlxfw_dev, u8 *status); + int (*fsm_query_state)(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, enum mlxfw_fsm_state *fsm_state, enum mlxfw_fsm_state_err *fsm_state_err); @@ -58,16 +96,6 @@ struct mlxfw_dev_ops { void (*fsm_cancel)(struct mlxfw_dev *mlxfw_dev, u32 fwhandle); void (*fsm_release)(struct mlxfw_dev *mlxfw_dev, u32 fwhandle); - - void (*status_notify)(struct mlxfw_dev *mlxfw_dev, - const char *msg, const char *comp_name, - u32 done_bytes, u32 total_bytes); -}; - -struct mlxfw_dev { - const struct mlxfw_dev_ops *ops; - const char *psid; - u16 psid_size; }; #if IS_REACHABLE(CONFIG_MLXFW) diff --git a/drivers/net/ethernet/mellanox/mlxfw/mlxfw_fsm.c b/drivers/net/ethernet/mellanox/mlxfw/mlxfw_fsm.c index 29e95d0a6ad1..046a0cb82ed8 100644 --- a/drivers/net/ethernet/mellanox/mlxfw/mlxfw_fsm.c +++ b/drivers/net/ethernet/mellanox/mlxfw/mlxfw_fsm.c @@ -16,38 +16,70 @@ (MLXFW_FSM_STATE_WAIT_TIMEOUT_MS / MLXFW_FSM_STATE_WAIT_CYCLE_MS) #define MLXFW_FSM_MAX_COMPONENT_SIZE (10 * (1 << 20)) -static const char * const mlxfw_fsm_state_err_str[] = { - [MLXFW_FSM_STATE_ERR_ERROR] = - "general error", - [MLXFW_FSM_STATE_ERR_REJECTED_DIGEST_ERR] = - "component hash mismatch", - [MLXFW_FSM_STATE_ERR_REJECTED_NOT_APPLICABLE] = - "component not applicable", - [MLXFW_FSM_STATE_ERR_REJECTED_UNKNOWN_KEY] = - "unknown key", - [MLXFW_FSM_STATE_ERR_REJECTED_AUTH_FAILED] = - "authentication failed", - [MLXFW_FSM_STATE_ERR_REJECTED_UNSIGNED] = - "component was not signed", - [MLXFW_FSM_STATE_ERR_REJECTED_KEY_NOT_APPLICABLE] = - "key not applicable", - [MLXFW_FSM_STATE_ERR_REJECTED_BAD_FORMAT] = - "bad format", - [MLXFW_FSM_STATE_ERR_BLOCKED_PENDING_RESET] = - "pending reset", - [MLXFW_FSM_STATE_ERR_MAX] = - "unknown error" +static const int mlxfw_fsm_state_errno[] = { + [MLXFW_FSM_STATE_ERR_ERROR] = -EIO, + [MLXFW_FSM_STATE_ERR_REJECTED_DIGEST_ERR] = -EBADMSG, + [MLXFW_FSM_STATE_ERR_REJECTED_NOT_APPLICABLE] = -ENOENT, + [MLXFW_FSM_STATE_ERR_REJECTED_UNKNOWN_KEY] = -ENOKEY, + [MLXFW_FSM_STATE_ERR_REJECTED_AUTH_FAILED] = -EACCES, + [MLXFW_FSM_STATE_ERR_REJECTED_UNSIGNED] = -EKEYREVOKED, + [MLXFW_FSM_STATE_ERR_REJECTED_KEY_NOT_APPLICABLE] = -EKEYREJECTED, + [MLXFW_FSM_STATE_ERR_REJECTED_BAD_FORMAT] = -ENOEXEC, + [MLXFW_FSM_STATE_ERR_BLOCKED_PENDING_RESET] = -EBUSY, + [MLXFW_FSM_STATE_ERR_MAX] = -EINVAL }; -static void mlxfw_status_notify(struct mlxfw_dev *mlxfw_dev, - const char *msg, const char *comp_name, - u32 done_bytes, u32 total_bytes) +#define MLXFW_ERR_PRFX "Firmware flash failed: " +#define MLXFW_ERR_MSG(fwdev, extack, msg, err) do { \ + mlxfw_err(fwdev, "%s, err (%d)\n", MLXFW_ERR_PRFX msg, err); \ + NL_SET_ERR_MSG_MOD(extack, MLXFW_ERR_PRFX msg); \ +} while (0) + +static int mlxfw_fsm_state_err(struct mlxfw_dev *mlxfw_dev, + struct netlink_ext_ack *extack, + enum mlxfw_fsm_state_err err) { - if (!mlxfw_dev->ops->status_notify) - return; - mlxfw_dev->ops->status_notify(mlxfw_dev, msg, comp_name, - done_bytes, total_bytes); -} + enum mlxfw_fsm_state_err fsm_state_err; + + fsm_state_err = min_t(enum mlxfw_fsm_state_err, err, + MLXFW_FSM_STATE_ERR_MAX); + + switch (fsm_state_err) { + case MLXFW_FSM_STATE_ERR_ERROR: + MLXFW_ERR_MSG(mlxfw_dev, extack, "general error", err); + break; + case MLXFW_FSM_STATE_ERR_REJECTED_DIGEST_ERR: + MLXFW_ERR_MSG(mlxfw_dev, extack, "component hash mismatch", err); + break; + case MLXFW_FSM_STATE_ERR_REJECTED_NOT_APPLICABLE: + MLXFW_ERR_MSG(mlxfw_dev, extack, "component not applicable", err); + break; + case MLXFW_FSM_STATE_ERR_REJECTED_UNKNOWN_KEY: + MLXFW_ERR_MSG(mlxfw_dev, extack, "unknown key", err); + break; + case MLXFW_FSM_STATE_ERR_REJECTED_AUTH_FAILED: + MLXFW_ERR_MSG(mlxfw_dev, extack, "authentication failed", err); + break; + case MLXFW_FSM_STATE_ERR_REJECTED_UNSIGNED: + MLXFW_ERR_MSG(mlxfw_dev, extack, "component was not signed", err); + break; + case MLXFW_FSM_STATE_ERR_REJECTED_KEY_NOT_APPLICABLE: + MLXFW_ERR_MSG(mlxfw_dev, extack, "key not applicable", err); + break; + case MLXFW_FSM_STATE_ERR_REJECTED_BAD_FORMAT: + MLXFW_ERR_MSG(mlxfw_dev, extack, "bad format", err); + break; + case MLXFW_FSM_STATE_ERR_BLOCKED_PENDING_RESET: + MLXFW_ERR_MSG(mlxfw_dev, extack, "pending reset", err); + break; + case MLXFW_FSM_STATE_ERR_OK: /* fall through */ + case MLXFW_FSM_STATE_ERR_MAX: + MLXFW_ERR_MSG(mlxfw_dev, extack, "unknown error", err); + break; + }; + + return mlxfw_fsm_state_errno[fsm_state_err]; +}; static int mlxfw_fsm_state_wait(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, enum mlxfw_fsm_state fsm_state, @@ -62,21 +94,18 @@ static int mlxfw_fsm_state_wait(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, retry: err = mlxfw_dev->ops->fsm_query_state(mlxfw_dev, fwhandle, &curr_fsm_state, &fsm_state_err); - if (err) + if (err) { + MLXFW_ERR_MSG(mlxfw_dev, extack, "FSM state query failed", err); return err; - - if (fsm_state_err != MLXFW_FSM_STATE_ERR_OK) { - fsm_state_err = min_t(enum mlxfw_fsm_state_err, - fsm_state_err, MLXFW_FSM_STATE_ERR_MAX); - pr_err("Firmware flash failed: %s\n", - mlxfw_fsm_state_err_str[fsm_state_err]); - NL_SET_ERR_MSG_MOD(extack, "Firmware flash failed"); - return -EINVAL; } + + if (fsm_state_err != MLXFW_FSM_STATE_ERR_OK) + return mlxfw_fsm_state_err(mlxfw_dev, extack, fsm_state_err); + if (curr_fsm_state != fsm_state) { if (--times == 0) { - pr_err("Timeout reached on FSM state change"); - NL_SET_ERR_MSG_MOD(extack, "Timeout reached on FSM state change"); + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Timeout reached on FSM state change", -ETIMEDOUT); return -ETIMEDOUT; } msleep(MLXFW_FSM_STATE_WAIT_CYCLE_MS); @@ -85,6 +114,92 @@ retry: return 0; } +static int +mlxfw_fsm_reactivate_err(struct mlxfw_dev *mlxfw_dev, + struct netlink_ext_ack *extack, u8 err) +{ + enum mlxfw_fsm_reactivate_status status; + +#define MXFW_REACT_PRFX "Reactivate FSM: " +#define MLXFW_REACT_ERR(msg, err) \ + MLXFW_ERR_MSG(mlxfw_dev, extack, MXFW_REACT_PRFX msg, err) + + status = min_t(enum mlxfw_fsm_reactivate_status, err, + MLXFW_FSM_REACTIVATE_STATUS_MAX); + + switch (status) { + case MLXFW_FSM_REACTIVATE_STATUS_BUSY: + MLXFW_REACT_ERR("busy", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_PROHIBITED_FW_VER_ERR: + MLXFW_REACT_ERR("prohibited fw ver", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_FIRST_PAGE_COPY_FAILED: + MLXFW_REACT_ERR("first page copy failed", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_FIRST_PAGE_ERASE_FAILED: + MLXFW_REACT_ERR("first page erase failed", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_FIRST_PAGE_RESTORE_FAILED: + MLXFW_REACT_ERR("first page restore failed", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_CANDIDATE_FW_DEACTIVATION_FAILED: + MLXFW_REACT_ERR("candidate fw deactivation failed", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_ERR_DEVICE_RESET_REQUIRED: + MLXFW_REACT_ERR("device reset required", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_ERR_FW_PROGRAMMING_NEEDED: + MLXFW_REACT_ERR("fw programming needed", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_FW_ALREADY_ACTIVATED: + MLXFW_REACT_ERR("fw already activated", err); + break; + case MLXFW_FSM_REACTIVATE_STATUS_OK: /* fall through */ + case MLXFW_FSM_REACTIVATE_STATUS_MAX: + MLXFW_REACT_ERR("unexpected error", err); + break; + }; + return -EREMOTEIO; +}; + +static int mlxfw_fsm_reactivate(struct mlxfw_dev *mlxfw_dev, + struct netlink_ext_ack *extack, + bool *supported) +{ + u8 status; + int err; + + if (!mlxfw_dev->ops->fsm_reactivate) + return 0; + + err = mlxfw_dev->ops->fsm_reactivate(mlxfw_dev, &status); + if (err == -EOPNOTSUPP) { + *supported = false; + return 0; + } + + if (err) { + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Could not reactivate firmware flash", err); + return err; + } + + if (status == MLXFW_FSM_REACTIVATE_STATUS_OK || + status == MLXFW_FSM_REACTIVATE_STATUS_FW_ALREADY_ACTIVATED) + return 0; + + return mlxfw_fsm_reactivate_err(mlxfw_dev, extack, status); +} + +static void mlxfw_status_notify(struct mlxfw_dev *mlxfw_dev, + const char *msg, const char *comp_name, + u32 done_bytes, u32 total_bytes) +{ + devlink_flash_update_status_notify(mlxfw_dev->devlink, msg, comp_name, + done_bytes, total_bytes); +} + #define MLXFW_ALIGN_DOWN(x, align_bits) ((x) & ~((1 << (align_bits)) - 1)) #define MLXFW_ALIGN_UP(x, align_bits) \ MLXFW_ALIGN_DOWN((x) + ((1 << (align_bits)) - 1), (align_bits)) @@ -92,6 +207,7 @@ retry: static int mlxfw_flash_component(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, struct mlxfw_mfa2_component *comp, + bool reactivate_supp, struct netlink_ext_ack *extack) { u16 comp_max_write_size; @@ -108,34 +224,43 @@ static int mlxfw_flash_component(struct mlxfw_dev *mlxfw_dev, err = mlxfw_dev->ops->component_query(mlxfw_dev, comp->index, &comp_max_size, &comp_align_bits, &comp_max_write_size); - if (err) + if (err) { + MLXFW_ERR_MSG(mlxfw_dev, extack, "FSM component query failed", err); return err; + } comp_max_size = min_t(u32, comp_max_size, MLXFW_FSM_MAX_COMPONENT_SIZE); if (comp->data_size > comp_max_size) { - pr_err("Component %d is of size %d which is bigger than limit %d\n", - comp->index, comp->data_size, comp_max_size); - NL_SET_ERR_MSG_MOD(extack, "Component is bigger than limit"); + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Component size is bigger than limit", -EINVAL); return -EINVAL; } comp_max_write_size = MLXFW_ALIGN_DOWN(comp_max_write_size, comp_align_bits); - pr_debug("Component update\n"); + mlxfw_dbg(mlxfw_dev, "Component update\n"); mlxfw_status_notify(mlxfw_dev, "Updating component", comp_name, 0, 0); err = mlxfw_dev->ops->fsm_component_update(mlxfw_dev, fwhandle, comp->index, comp->data_size); - if (err) + if (err) { + if (!reactivate_supp) + MLXFW_ERR_MSG(mlxfw_dev, extack, + "FSM component update failed, FW reactivate is not supported", + err); + else + MLXFW_ERR_MSG(mlxfw_dev, extack, + "FSM component update failed", err); return err; + } err = mlxfw_fsm_state_wait(mlxfw_dev, fwhandle, MLXFW_FSM_STATE_DOWNLOAD, extack); if (err) goto err_out; - pr_debug("Component download\n"); + mlxfw_dbg(mlxfw_dev, "Component download\n"); mlxfw_status_notify(mlxfw_dev, "Downloading component", comp_name, 0, comp->data_size); for (offset = 0; @@ -147,19 +272,25 @@ static int mlxfw_flash_component(struct mlxfw_dev *mlxfw_dev, err = mlxfw_dev->ops->fsm_block_download(mlxfw_dev, fwhandle, block_ptr, block_size, offset); - if (err) + if (err) { + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Component download failed", err); goto err_out; + } mlxfw_status_notify(mlxfw_dev, "Downloading component", comp_name, offset + block_size, comp->data_size); } - pr_debug("Component verify\n"); + mlxfw_dbg(mlxfw_dev, "Component verify\n"); mlxfw_status_notify(mlxfw_dev, "Verifying component", comp_name, 0, 0); err = mlxfw_dev->ops->fsm_component_verify(mlxfw_dev, fwhandle, comp->index); - if (err) + if (err) { + MLXFW_ERR_MSG(mlxfw_dev, extack, + "FSM component verify failed", err); goto err_out; + } err = mlxfw_fsm_state_wait(mlxfw_dev, fwhandle, MLXFW_FSM_STATE_LOCKED, extack); @@ -174,6 +305,7 @@ err_out: static int mlxfw_flash_components(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, struct mlxfw_mfa2_file *mfa2_file, + bool reactivate_supp, struct netlink_ext_ack *extack) { u32 component_count; @@ -184,8 +316,8 @@ static int mlxfw_flash_components(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, mlxfw_dev->psid_size, &component_count); if (err) { - pr_err("Could not find device PSID in MFA2 file\n"); - NL_SET_ERR_MSG_MOD(extack, "Could not find device PSID in MFA2 file"); + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Could not find device PSID in MFA2 file", err); return err; } @@ -194,11 +326,17 @@ static int mlxfw_flash_components(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, comp = mlxfw_mfa2_file_component_get(mfa2_file, mlxfw_dev->psid, mlxfw_dev->psid_size, i); - if (IS_ERR(comp)) - return PTR_ERR(comp); + if (IS_ERR(comp)) { + err = PTR_ERR(comp); + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Failed to get MFA2 component", err); + return err; + } - pr_info("Flashing component type %d\n", comp->index); - err = mlxfw_flash_component(mlxfw_dev, fwhandle, comp, extack); + mlxfw_info(mlxfw_dev, "Flashing component type %d\n", + comp->index); + err = mlxfw_flash_component(mlxfw_dev, fwhandle, comp, + reactivate_supp, extack); mlxfw_mfa2_file_component_put(comp); if (err) return err; @@ -211,26 +349,32 @@ int mlxfw_firmware_flash(struct mlxfw_dev *mlxfw_dev, struct netlink_ext_ack *extack) { struct mlxfw_mfa2_file *mfa2_file; + bool reactivate_supp = true; u32 fwhandle; int err; if (!mlxfw_mfa2_check(firmware)) { - pr_err("Firmware file is not MFA2\n"); - NL_SET_ERR_MSG_MOD(extack, "Firmware file is not MFA2"); + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Firmware file is not MFA2", -EINVAL); return -EINVAL; } mfa2_file = mlxfw_mfa2_file_init(firmware); - if (IS_ERR(mfa2_file)) - return PTR_ERR(mfa2_file); + if (IS_ERR(mfa2_file)) { + err = PTR_ERR(mfa2_file); + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Failed to initialize MFA2 firmware file", err); + return err; + } - pr_info("Initialize firmware flash process\n"); + mlxfw_info(mlxfw_dev, "Initialize firmware flash process\n"); + devlink_flash_update_begin_notify(mlxfw_dev->devlink); mlxfw_status_notify(mlxfw_dev, "Initializing firmware flash process", NULL, 0, 0); err = mlxfw_dev->ops->fsm_lock(mlxfw_dev, &fwhandle); if (err) { - pr_err("Could not lock the firmware FSM\n"); - NL_SET_ERR_MSG_MOD(extack, "Could not lock the firmware FSM"); + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Could not lock the firmware FSM", err); goto err_fsm_lock; } @@ -239,16 +383,26 @@ int mlxfw_firmware_flash(struct mlxfw_dev *mlxfw_dev, if (err) goto err_state_wait_idle_to_locked; - err = mlxfw_flash_components(mlxfw_dev, fwhandle, mfa2_file, extack); + err = mlxfw_fsm_reactivate(mlxfw_dev, extack, &reactivate_supp); + if (err) + goto err_fsm_reactivate; + + err = mlxfw_fsm_state_wait(mlxfw_dev, fwhandle, + MLXFW_FSM_STATE_LOCKED, extack); + if (err) + goto err_state_wait_reactivate_to_locked; + + err = mlxfw_flash_components(mlxfw_dev, fwhandle, mfa2_file, + reactivate_supp, extack); if (err) goto err_flash_components; - pr_debug("Activate image\n"); + mlxfw_dbg(mlxfw_dev, "Activate image\n"); mlxfw_status_notify(mlxfw_dev, "Activating image", NULL, 0, 0); err = mlxfw_dev->ops->fsm_activate(mlxfw_dev, fwhandle); if (err) { - pr_err("Could not activate the downloaded image\n"); - NL_SET_ERR_MSG_MOD(extack, "Could not activate the downloaded image"); + MLXFW_ERR_MSG(mlxfw_dev, extack, + "Could not activate the downloaded image", err); goto err_fsm_activate; } @@ -257,21 +411,25 @@ int mlxfw_firmware_flash(struct mlxfw_dev *mlxfw_dev, if (err) goto err_state_wait_activate_to_locked; - pr_debug("Handle release\n"); + mlxfw_dbg(mlxfw_dev, "Handle release\n"); mlxfw_dev->ops->fsm_release(mlxfw_dev, fwhandle); - pr_info("Firmware flash done.\n"); + mlxfw_info(mlxfw_dev, "Firmware flash done\n"); mlxfw_status_notify(mlxfw_dev, "Firmware flash done", NULL, 0, 0); mlxfw_mfa2_file_fini(mfa2_file); + devlink_flash_update_end_notify(mlxfw_dev->devlink); return 0; err_state_wait_activate_to_locked: err_fsm_activate: err_flash_components: +err_state_wait_reactivate_to_locked: +err_fsm_reactivate: err_state_wait_idle_to_locked: mlxfw_dev->ops->fsm_release(mlxfw_dev, fwhandle); err_fsm_lock: mlxfw_mfa2_file_fini(mfa2_file); + devlink_flash_update_end_notify(mlxfw_dev->devlink); return err; } EXPORT_SYMBOL(mlxfw_firmware_flash); diff --git a/drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2.c b/drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2.c index 79057af4fe99..5d9ddf36fb4e 100644 --- a/drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2.c +++ b/drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2.c @@ -496,7 +496,7 @@ mlxfw_mfa2_file_component_tlv_get(const struct mlxfw_mfa2_file *mfa2_file, struct mlxfw_mfa2_comp_data { struct mlxfw_mfa2_component comp; - u8 buff[0]; + u8 buff[]; }; static const struct mlxfw_mfa2_tlv_component_descriptor * diff --git a/drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2_tlv.h b/drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2_tlv.h index 33c971190bba..2014a5de5a01 100644 --- a/drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2_tlv.h +++ b/drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2_tlv.h @@ -11,7 +11,7 @@ struct mlxfw_mfa2_tlv { u8 version; u8 type; __be16 len; - u8 data[0]; + u8 data[]; } __packed; static inline const struct mlxfw_mfa2_tlv * diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index e9f791c43f20..e9ccd333f61d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -82,7 +82,7 @@ struct mlxsw_core { struct mlxsw_core_port *ports; unsigned int max_ports; bool fw_flash_in_progress; - unsigned long driver_priv[0]; + unsigned long driver_priv[]; /* driver_priv has to be always the last item */ }; @@ -142,6 +142,7 @@ struct mlxsw_rx_listener_item { struct list_head list; struct mlxsw_rx_listener rxl; void *priv; + bool enabled; }; struct mlxsw_event_listener_item { @@ -1197,6 +1198,72 @@ mlxsw_devlink_trap_group_init(struct devlink *devlink, return mlxsw_driver->trap_group_init(mlxsw_core, group); } +static int +mlxsw_devlink_trap_group_set(struct devlink *devlink, + const struct devlink_trap_group *group, + const struct devlink_trap_policer *policer) +{ + struct mlxsw_core *mlxsw_core = devlink_priv(devlink); + struct mlxsw_driver *mlxsw_driver = mlxsw_core->driver; + + if (!mlxsw_driver->trap_group_set) + return -EOPNOTSUPP; + return mlxsw_driver->trap_group_set(mlxsw_core, group, policer); +} + +static int +mlxsw_devlink_trap_policer_init(struct devlink *devlink, + const struct devlink_trap_policer *policer) +{ + struct mlxsw_core *mlxsw_core = devlink_priv(devlink); + struct mlxsw_driver *mlxsw_driver = mlxsw_core->driver; + + if (!mlxsw_driver->trap_policer_init) + return -EOPNOTSUPP; + return mlxsw_driver->trap_policer_init(mlxsw_core, policer); +} + +static void +mlxsw_devlink_trap_policer_fini(struct devlink *devlink, + const struct devlink_trap_policer *policer) +{ + struct mlxsw_core *mlxsw_core = devlink_priv(devlink); + struct mlxsw_driver *mlxsw_driver = mlxsw_core->driver; + + if (!mlxsw_driver->trap_policer_fini) + return; + mlxsw_driver->trap_policer_fini(mlxsw_core, policer); +} + +static int +mlxsw_devlink_trap_policer_set(struct devlink *devlink, + const struct devlink_trap_policer *policer, + u64 rate, u64 burst, + struct netlink_ext_ack *extack) +{ + struct mlxsw_core *mlxsw_core = devlink_priv(devlink); + struct mlxsw_driver *mlxsw_driver = mlxsw_core->driver; + + if (!mlxsw_driver->trap_policer_set) + return -EOPNOTSUPP; + return mlxsw_driver->trap_policer_set(mlxsw_core, policer, rate, burst, + extack); +} + +static int +mlxsw_devlink_trap_policer_counter_get(struct devlink *devlink, + const struct devlink_trap_policer *policer, + u64 *p_drops) +{ + struct mlxsw_core *mlxsw_core = devlink_priv(devlink); + struct mlxsw_driver *mlxsw_driver = mlxsw_core->driver; + + if (!mlxsw_driver->trap_policer_counter_get) + return -EOPNOTSUPP; + return mlxsw_driver->trap_policer_counter_get(mlxsw_core, policer, + p_drops); +} + static const struct devlink_ops mlxsw_devlink_ops = { .reload_down = mlxsw_devlink_core_bus_device_reload_down, .reload_up = mlxsw_devlink_core_bus_device_reload_up, @@ -1219,6 +1286,11 @@ static const struct devlink_ops mlxsw_devlink_ops = { .trap_fini = mlxsw_devlink_trap_fini, .trap_action_set = mlxsw_devlink_trap_action_set, .trap_group_init = mlxsw_devlink_trap_group_init, + .trap_group_set = mlxsw_devlink_trap_group_set, + .trap_policer_init = mlxsw_devlink_trap_policer_init, + .trap_policer_fini = mlxsw_devlink_trap_policer_fini, + .trap_policer_set = mlxsw_devlink_trap_policer_set, + .trap_policer_counter_get = mlxsw_devlink_trap_policer_counter_get, }; static int @@ -1457,14 +1529,12 @@ static bool __is_rx_listener_equal(const struct mlxsw_rx_listener *rxl_a, static struct mlxsw_rx_listener_item * __find_rx_listener_item(struct mlxsw_core *mlxsw_core, - const struct mlxsw_rx_listener *rxl, - void *priv) + const struct mlxsw_rx_listener *rxl) { struct mlxsw_rx_listener_item *rxl_item; list_for_each_entry(rxl_item, &mlxsw_core->rx_listener_list, list) { - if (__is_rx_listener_equal(&rxl_item->rxl, rxl) && - rxl_item->priv == priv) + if (__is_rx_listener_equal(&rxl_item->rxl, rxl)) return rxl_item; } return NULL; @@ -1472,11 +1542,11 @@ __find_rx_listener_item(struct mlxsw_core *mlxsw_core, int mlxsw_core_rx_listener_register(struct mlxsw_core *mlxsw_core, const struct mlxsw_rx_listener *rxl, - void *priv) + void *priv, bool enabled) { struct mlxsw_rx_listener_item *rxl_item; - rxl_item = __find_rx_listener_item(mlxsw_core, rxl, priv); + rxl_item = __find_rx_listener_item(mlxsw_core, rxl); if (rxl_item) return -EEXIST; rxl_item = kmalloc(sizeof(*rxl_item), GFP_KERNEL); @@ -1484,6 +1554,7 @@ int mlxsw_core_rx_listener_register(struct mlxsw_core *mlxsw_core, return -ENOMEM; rxl_item->rxl = *rxl; rxl_item->priv = priv; + rxl_item->enabled = enabled; list_add_rcu(&rxl_item->list, &mlxsw_core->rx_listener_list); return 0; @@ -1491,12 +1562,11 @@ int mlxsw_core_rx_listener_register(struct mlxsw_core *mlxsw_core, EXPORT_SYMBOL(mlxsw_core_rx_listener_register); void mlxsw_core_rx_listener_unregister(struct mlxsw_core *mlxsw_core, - const struct mlxsw_rx_listener *rxl, - void *priv) + const struct mlxsw_rx_listener *rxl) { struct mlxsw_rx_listener_item *rxl_item; - rxl_item = __find_rx_listener_item(mlxsw_core, rxl, priv); + rxl_item = __find_rx_listener_item(mlxsw_core, rxl); if (!rxl_item) return; list_del_rcu(&rxl_item->list); @@ -1505,6 +1575,19 @@ void mlxsw_core_rx_listener_unregister(struct mlxsw_core *mlxsw_core, } EXPORT_SYMBOL(mlxsw_core_rx_listener_unregister); +static void +mlxsw_core_rx_listener_state_set(struct mlxsw_core *mlxsw_core, + const struct mlxsw_rx_listener *rxl, + bool enabled) +{ + struct mlxsw_rx_listener_item *rxl_item; + + rxl_item = __find_rx_listener_item(mlxsw_core, rxl); + if (WARN_ON(!rxl_item)) + return; + rxl_item->enabled = enabled; +} + static void mlxsw_core_event_listener_func(struct sk_buff *skb, u8 local_port, void *priv) { @@ -1534,14 +1617,12 @@ static bool __is_event_listener_equal(const struct mlxsw_event_listener *el_a, static struct mlxsw_event_listener_item * __find_event_listener_item(struct mlxsw_core *mlxsw_core, - const struct mlxsw_event_listener *el, - void *priv) + const struct mlxsw_event_listener *el) { struct mlxsw_event_listener_item *el_item; list_for_each_entry(el_item, &mlxsw_core->event_listener_list, list) { - if (__is_event_listener_equal(&el_item->el, el) && - el_item->priv == priv) + if (__is_event_listener_equal(&el_item->el, el)) return el_item; } return NULL; @@ -1559,7 +1640,7 @@ int mlxsw_core_event_listener_register(struct mlxsw_core *mlxsw_core, .trap_id = el->trap_id, }; - el_item = __find_event_listener_item(mlxsw_core, el, priv); + el_item = __find_event_listener_item(mlxsw_core, el); if (el_item) return -EEXIST; el_item = kmalloc(sizeof(*el_item), GFP_KERNEL); @@ -1568,7 +1649,7 @@ int mlxsw_core_event_listener_register(struct mlxsw_core *mlxsw_core, el_item->el = *el; el_item->priv = priv; - err = mlxsw_core_rx_listener_register(mlxsw_core, &rxl, el_item); + err = mlxsw_core_rx_listener_register(mlxsw_core, &rxl, el_item, true); if (err) goto err_rx_listener_register; @@ -1586,8 +1667,7 @@ err_rx_listener_register: EXPORT_SYMBOL(mlxsw_core_event_listener_register); void mlxsw_core_event_listener_unregister(struct mlxsw_core *mlxsw_core, - const struct mlxsw_event_listener *el, - void *priv) + const struct mlxsw_event_listener *el) { struct mlxsw_event_listener_item *el_item; const struct mlxsw_rx_listener rxl = { @@ -1596,10 +1676,10 @@ void mlxsw_core_event_listener_unregister(struct mlxsw_core *mlxsw_core, .trap_id = el->trap_id, }; - el_item = __find_event_listener_item(mlxsw_core, el, priv); + el_item = __find_event_listener_item(mlxsw_core, el); if (!el_item) return; - mlxsw_core_rx_listener_unregister(mlxsw_core, &rxl, el_item); + mlxsw_core_rx_listener_unregister(mlxsw_core, &rxl); list_del(&el_item->list); kfree(el_item); } @@ -1607,16 +1687,18 @@ EXPORT_SYMBOL(mlxsw_core_event_listener_unregister); static int mlxsw_core_listener_register(struct mlxsw_core *mlxsw_core, const struct mlxsw_listener *listener, - void *priv) + void *priv, bool enabled) { - if (listener->is_event) + if (listener->is_event) { + WARN_ON(!enabled); return mlxsw_core_event_listener_register(mlxsw_core, - &listener->u.event_listener, + &listener->event_listener, priv); - else + } else { return mlxsw_core_rx_listener_register(mlxsw_core, - &listener->u.rx_listener, - priv); + &listener->rx_listener, + priv, enabled); + } } static void mlxsw_core_listener_unregister(struct mlxsw_core *mlxsw_core, @@ -1625,26 +1707,31 @@ static void mlxsw_core_listener_unregister(struct mlxsw_core *mlxsw_core, { if (listener->is_event) mlxsw_core_event_listener_unregister(mlxsw_core, - &listener->u.event_listener, - priv); + &listener->event_listener); else mlxsw_core_rx_listener_unregister(mlxsw_core, - &listener->u.rx_listener, - priv); + &listener->rx_listener); } int mlxsw_core_trap_register(struct mlxsw_core *mlxsw_core, const struct mlxsw_listener *listener, void *priv) { + enum mlxsw_reg_htgt_trap_group trap_group; + enum mlxsw_reg_hpkt_action action; char hpkt_pl[MLXSW_REG_HPKT_LEN]; int err; - err = mlxsw_core_listener_register(mlxsw_core, listener, priv); + err = mlxsw_core_listener_register(mlxsw_core, listener, priv, + listener->enabled_on_register); if (err) return err; - mlxsw_reg_hpkt_pack(hpkt_pl, listener->action, listener->trap_id, - listener->trap_group, listener->is_ctrl); + action = listener->enabled_on_register ? listener->en_action : + listener->dis_action; + trap_group = listener->enabled_on_register ? listener->en_trap_group : + listener->dis_trap_group; + mlxsw_reg_hpkt_pack(hpkt_pl, action, listener->trap_id, + trap_group, listener->is_ctrl); err = mlxsw_reg_write(mlxsw_core, MLXSW_REG(hpkt), hpkt_pl); if (err) goto err_trap_set; @@ -1664,8 +1751,8 @@ void mlxsw_core_trap_unregister(struct mlxsw_core *mlxsw_core, char hpkt_pl[MLXSW_REG_HPKT_LEN]; if (!listener->is_event) { - mlxsw_reg_hpkt_pack(hpkt_pl, listener->unreg_action, - listener->trap_id, listener->trap_group, + mlxsw_reg_hpkt_pack(hpkt_pl, listener->dis_action, + listener->trap_id, listener->dis_trap_group, listener->is_ctrl); mlxsw_reg_write(mlxsw_core, MLXSW_REG(hpkt), hpkt_pl); } @@ -1674,17 +1761,33 @@ void mlxsw_core_trap_unregister(struct mlxsw_core *mlxsw_core, } EXPORT_SYMBOL(mlxsw_core_trap_unregister); -int mlxsw_core_trap_action_set(struct mlxsw_core *mlxsw_core, - const struct mlxsw_listener *listener, - enum mlxsw_reg_hpkt_action action) +int mlxsw_core_trap_state_set(struct mlxsw_core *mlxsw_core, + const struct mlxsw_listener *listener, + bool enabled) { + enum mlxsw_reg_htgt_trap_group trap_group; + enum mlxsw_reg_hpkt_action action; char hpkt_pl[MLXSW_REG_HPKT_LEN]; + int err; + + /* Not supported for event listener */ + if (WARN_ON(listener->is_event)) + return -EINVAL; + action = enabled ? listener->en_action : listener->dis_action; + trap_group = enabled ? listener->en_trap_group : + listener->dis_trap_group; mlxsw_reg_hpkt_pack(hpkt_pl, action, listener->trap_id, - listener->trap_group, listener->is_ctrl); - return mlxsw_reg_write(mlxsw_core, MLXSW_REG(hpkt), hpkt_pl); + trap_group, listener->is_ctrl); + err = mlxsw_reg_write(mlxsw_core, MLXSW_REG(hpkt), hpkt_pl); + if (err) + return err; + + mlxsw_core_rx_listener_state_set(mlxsw_core, &listener->rx_listener, + enabled); + return 0; } -EXPORT_SYMBOL(mlxsw_core_trap_action_set); +EXPORT_SYMBOL(mlxsw_core_trap_state_set); static u64 mlxsw_core_tid_get(struct mlxsw_core *mlxsw_core) { @@ -1942,7 +2045,8 @@ void mlxsw_core_skb_receive(struct mlxsw_core *mlxsw_core, struct sk_buff *skb, if ((rxl->local_port == MLXSW_PORT_DONT_CARE || rxl->local_port == local_port) && rxl->trap_id == rx_info->trap_id) { - found = true; + if (rxl_item->enabled) + found = true; break; } } @@ -2168,13 +2272,22 @@ int mlxsw_core_module_max_width(struct mlxsw_core *mlxsw_core, u8 module) /* Here we need to get the module width according to the module type. */ switch (module_type) { + case MLXSW_REG_PMTM_MODULE_TYPE_C2C8X: /* fall through */ + case MLXSW_REG_PMTM_MODULE_TYPE_QSFP_DD: /* fall through */ + case MLXSW_REG_PMTM_MODULE_TYPE_OSFP: + return 8; + case MLXSW_REG_PMTM_MODULE_TYPE_C2C4X: /* fall through */ case MLXSW_REG_PMTM_MODULE_TYPE_BP_4X: /* fall through */ - case MLXSW_REG_PMTM_MODULE_TYPE_BP_QSFP: + case MLXSW_REG_PMTM_MODULE_TYPE_QSFP: return 4; - case MLXSW_REG_PMTM_MODULE_TYPE_BP_2X: + case MLXSW_REG_PMTM_MODULE_TYPE_C2C2X: /* fall through */ + case MLXSW_REG_PMTM_MODULE_TYPE_BP_2X: /* fall through */ + case MLXSW_REG_PMTM_MODULE_TYPE_SFP_DD: /* fall through */ + case MLXSW_REG_PMTM_MODULE_TYPE_DSFP: return 2; - case MLXSW_REG_PMTM_MODULE_TYPE_BP_SFP: /* fall through */ - case MLXSW_REG_PMTM_MODULE_TYPE_BP_1X: + case MLXSW_REG_PMTM_MODULE_TYPE_C2C1X: /* fall through */ + case MLXSW_REG_PMTM_MODULE_TYPE_BP_1X: /* fall through */ + case MLXSW_REG_PMTM_MODULE_TYPE_SFP: return 1; default: return -EINVAL; diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h index 543476a2e503..22b0dfa7cfae 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core.h @@ -62,7 +62,6 @@ struct mlxsw_rx_listener { void (*func)(struct sk_buff *skb, u8 local_port, void *priv); u8 local_port; u16 trap_id; - enum mlxsw_reg_hpkt_action action; }; struct mlxsw_event_listener { @@ -76,58 +75,71 @@ struct mlxsw_listener { union { struct mlxsw_rx_listener rx_listener; struct mlxsw_event_listener event_listener; - } u; - enum mlxsw_reg_hpkt_action action; - enum mlxsw_reg_hpkt_action unreg_action; - u8 trap_group; - bool is_ctrl; /* should go via control buffer or not */ - bool is_event; + }; + enum mlxsw_reg_hpkt_action en_action; /* Action when enabled */ + enum mlxsw_reg_hpkt_action dis_action; /* Action when disabled */ + u8 en_trap_group; /* Trap group when enabled */ + u8 dis_trap_group; /* Trap group when disabled */ + u8 is_ctrl:1, /* should go via control buffer or not */ + is_event:1, + enabled_on_register:1; /* Trap should be enabled when listener + * is registered. + */ }; -#define MLXSW_RXL(_func, _trap_id, _action, _is_ctrl, _trap_group, \ - _unreg_action) \ - { \ - .trap_id = MLXSW_TRAP_ID_##_trap_id, \ - .u.rx_listener = \ - { \ - .func = _func, \ - .local_port = MLXSW_PORT_DONT_CARE, \ - .trap_id = MLXSW_TRAP_ID_##_trap_id, \ - }, \ - .action = MLXSW_REG_HPKT_ACTION_##_action, \ - .unreg_action = MLXSW_REG_HPKT_ACTION_##_unreg_action, \ - .trap_group = MLXSW_REG_HTGT_TRAP_GROUP_##_trap_group, \ - .is_ctrl = _is_ctrl, \ - .is_event = false, \ +#define __MLXSW_RXL(_func, _trap_id, _en_action, _is_ctrl, _en_trap_group, \ + _dis_action, _enabled_on_register, _dis_trap_group) \ + { \ + .trap_id = MLXSW_TRAP_ID_##_trap_id, \ + .rx_listener = \ + { \ + .func = _func, \ + .local_port = MLXSW_PORT_DONT_CARE, \ + .trap_id = MLXSW_TRAP_ID_##_trap_id, \ + }, \ + .en_action = MLXSW_REG_HPKT_ACTION_##_en_action, \ + .dis_action = MLXSW_REG_HPKT_ACTION_##_dis_action, \ + .en_trap_group = MLXSW_REG_HTGT_TRAP_GROUP_##_en_trap_group, \ + .dis_trap_group = MLXSW_REG_HTGT_TRAP_GROUP_##_dis_trap_group, \ + .is_ctrl = _is_ctrl, \ + .enabled_on_register = _enabled_on_register, \ } -#define MLXSW_EVENTL(_func, _trap_id, _trap_group) \ - { \ - .trap_id = MLXSW_TRAP_ID_##_trap_id, \ - .u.event_listener = \ - { \ - .func = _func, \ - .trap_id = MLXSW_TRAP_ID_##_trap_id, \ - }, \ - .action = MLXSW_REG_HPKT_ACTION_TRAP_TO_CPU, \ - .trap_group = MLXSW_REG_HTGT_TRAP_GROUP_##_trap_group, \ - .is_ctrl = false, \ - .is_event = true, \ +#define MLXSW_RXL(_func, _trap_id, _en_action, _is_ctrl, _trap_group, \ + _dis_action) \ + __MLXSW_RXL(_func, _trap_id, _en_action, _is_ctrl, _trap_group, \ + _dis_action, true, _trap_group) + +#define MLXSW_RXL_DIS(_func, _trap_id, _en_action, _is_ctrl, _en_trap_group, \ + _dis_action, _dis_trap_group) \ + __MLXSW_RXL(_func, _trap_id, _en_action, _is_ctrl, _en_trap_group, \ + _dis_action, false, _dis_trap_group) + +#define MLXSW_EVENTL(_func, _trap_id, _trap_group) \ + { \ + .trap_id = MLXSW_TRAP_ID_##_trap_id, \ + .event_listener = \ + { \ + .func = _func, \ + .trap_id = MLXSW_TRAP_ID_##_trap_id, \ + }, \ + .en_action = MLXSW_REG_HPKT_ACTION_TRAP_TO_CPU, \ + .en_trap_group = MLXSW_REG_HTGT_TRAP_GROUP_##_trap_group, \ + .is_event = true, \ + .enabled_on_register = true, \ } int mlxsw_core_rx_listener_register(struct mlxsw_core *mlxsw_core, const struct mlxsw_rx_listener *rxl, - void *priv); + void *priv, bool enabled); void mlxsw_core_rx_listener_unregister(struct mlxsw_core *mlxsw_core, - const struct mlxsw_rx_listener *rxl, - void *priv); + const struct mlxsw_rx_listener *rxl); int mlxsw_core_event_listener_register(struct mlxsw_core *mlxsw_core, const struct mlxsw_event_listener *el, void *priv); void mlxsw_core_event_listener_unregister(struct mlxsw_core *mlxsw_core, - const struct mlxsw_event_listener *el, - void *priv); + const struct mlxsw_event_listener *el); int mlxsw_core_trap_register(struct mlxsw_core *mlxsw_core, const struct mlxsw_listener *listener, @@ -135,9 +147,9 @@ int mlxsw_core_trap_register(struct mlxsw_core *mlxsw_core, void mlxsw_core_trap_unregister(struct mlxsw_core *mlxsw_core, const struct mlxsw_listener *listener, void *priv); -int mlxsw_core_trap_action_set(struct mlxsw_core *mlxsw_core, - const struct mlxsw_listener *listener, - enum mlxsw_reg_hpkt_action action); +int mlxsw_core_trap_state_set(struct mlxsw_core *mlxsw_core, + const struct mlxsw_listener *listener, + bool enabled); typedef void mlxsw_reg_trans_cb_t(struct mlxsw_core *mlxsw_core, char *payload, size_t payload_len, unsigned long cb_priv); @@ -315,6 +327,20 @@ struct mlxsw_driver { enum devlink_trap_action action); int (*trap_group_init)(struct mlxsw_core *mlxsw_core, const struct devlink_trap_group *group); + int (*trap_group_set)(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_group *group, + const struct devlink_trap_policer *policer); + int (*trap_policer_init)(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer); + void (*trap_policer_fini)(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer); + int (*trap_policer_set)(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer, + u64 rate, u64 burst, + struct netlink_ext_ack *extack); + int (*trap_policer_counter_get)(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer, + u64 *p_drops); void (*txhdr_construct)(struct sk_buff *skb, const struct mlxsw_tx_info *tx_info); int (*resources_register)(struct mlxsw_core *mlxsw_core); @@ -461,7 +487,10 @@ enum mlxsw_devlink_param_id { }; struct mlxsw_skb_cb { - struct mlxsw_tx_info tx_info; + union { + struct mlxsw_tx_info tx_info; + u32 cookie_index; /* Only used during receive */ + }; }; static inline struct mlxsw_skb_cb *mlxsw_skb_cb(struct sk_buff *skb) diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c index c51b2adfc1e1..70a104e728f6 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c @@ -7,6 +7,9 @@ #include <linux/errno.h> #include <linux/rhashtable.h> #include <linux/list.h> +#include <linux/idr.h> +#include <linux/refcount.h> +#include <net/flow_offload.h> #include "item.h" #include "trap.h" @@ -63,6 +66,8 @@ struct mlxsw_afa { void *ops_priv; struct rhashtable set_ht; struct rhashtable fwd_entry_ht; + struct rhashtable cookie_ht; + struct idr cookie_idr; }; #define MLXSW_AFA_SET_LEN 0xA8 @@ -121,6 +126,55 @@ static const struct rhashtable_params mlxsw_afa_fwd_entry_ht_params = { .automatic_shrinking = true, }; +struct mlxsw_afa_cookie { + struct rhash_head ht_node; + refcount_t ref_count; + struct rcu_head rcu; + u32 cookie_index; + struct flow_action_cookie fa_cookie; +}; + +static u32 mlxsw_afa_cookie_hash(const struct flow_action_cookie *fa_cookie, + u32 seed) +{ + return jhash2((u32 *) fa_cookie->cookie, + fa_cookie->cookie_len / sizeof(u32), seed); +} + +static u32 mlxsw_afa_cookie_key_hashfn(const void *data, u32 len, u32 seed) +{ + const struct flow_action_cookie *fa_cookie = data; + + return mlxsw_afa_cookie_hash(fa_cookie, seed); +} + +static u32 mlxsw_afa_cookie_obj_hashfn(const void *data, u32 len, u32 seed) +{ + const struct mlxsw_afa_cookie *cookie = data; + + return mlxsw_afa_cookie_hash(&cookie->fa_cookie, seed); +} + +static int mlxsw_afa_cookie_obj_cmpfn(struct rhashtable_compare_arg *arg, + const void *obj) +{ + const struct flow_action_cookie *fa_cookie = arg->key; + const struct mlxsw_afa_cookie *cookie = obj; + + if (cookie->fa_cookie.cookie_len == fa_cookie->cookie_len) + return memcmp(cookie->fa_cookie.cookie, fa_cookie->cookie, + fa_cookie->cookie_len); + return 1; +} + +static const struct rhashtable_params mlxsw_afa_cookie_ht_params = { + .head_offset = offsetof(struct mlxsw_afa_cookie, ht_node), + .hashfn = mlxsw_afa_cookie_key_hashfn, + .obj_hashfn = mlxsw_afa_cookie_obj_hashfn, + .obj_cmpfn = mlxsw_afa_cookie_obj_cmpfn, + .automatic_shrinking = true, +}; + struct mlxsw_afa *mlxsw_afa_create(unsigned int max_acts_per_set, const struct mlxsw_afa_ops *ops, void *ops_priv) @@ -138,11 +192,18 @@ struct mlxsw_afa *mlxsw_afa_create(unsigned int max_acts_per_set, &mlxsw_afa_fwd_entry_ht_params); if (err) goto err_fwd_entry_rhashtable_init; + err = rhashtable_init(&mlxsw_afa->cookie_ht, + &mlxsw_afa_cookie_ht_params); + if (err) + goto err_cookie_rhashtable_init; + idr_init(&mlxsw_afa->cookie_idr); mlxsw_afa->max_acts_per_set = max_acts_per_set; mlxsw_afa->ops = ops; mlxsw_afa->ops_priv = ops_priv; return mlxsw_afa; +err_cookie_rhashtable_init: + rhashtable_destroy(&mlxsw_afa->fwd_entry_ht); err_fwd_entry_rhashtable_init: rhashtable_destroy(&mlxsw_afa->set_ht); err_set_rhashtable_init: @@ -153,6 +214,9 @@ EXPORT_SYMBOL(mlxsw_afa_create); void mlxsw_afa_destroy(struct mlxsw_afa *mlxsw_afa) { + WARN_ON(!idr_is_empty(&mlxsw_afa->cookie_idr)); + idr_destroy(&mlxsw_afa->cookie_idr); + rhashtable_destroy(&mlxsw_afa->cookie_ht); rhashtable_destroy(&mlxsw_afa->fwd_entry_ht); rhashtable_destroy(&mlxsw_afa->set_ht); kfree(mlxsw_afa); @@ -627,6 +691,151 @@ err_counter_index_get: return ERR_PTR(err); } +/* 20 bits is a maximum that hardware can handle in trap with userdef action + * and carry along with the trapped packet. + */ +#define MLXSW_AFA_COOKIE_INDEX_BITS 20 +#define MLXSW_AFA_COOKIE_INDEX_MAX ((1 << MLXSW_AFA_COOKIE_INDEX_BITS) - 1) + +static struct mlxsw_afa_cookie * +mlxsw_afa_cookie_create(struct mlxsw_afa *mlxsw_afa, + const struct flow_action_cookie *fa_cookie) +{ + struct mlxsw_afa_cookie *cookie; + u32 cookie_index; + int err; + + cookie = kzalloc(sizeof(*cookie) + fa_cookie->cookie_len, GFP_KERNEL); + if (!cookie) + return ERR_PTR(-ENOMEM); + refcount_set(&cookie->ref_count, 1); + memcpy(&cookie->fa_cookie, fa_cookie, + sizeof(*fa_cookie) + fa_cookie->cookie_len); + + err = rhashtable_insert_fast(&mlxsw_afa->cookie_ht, &cookie->ht_node, + mlxsw_afa_cookie_ht_params); + if (err) + goto err_rhashtable_insert; + + /* Start cookie indexes with 1. Leave the 0 index unused. Packets + * that come from the HW which are not dropped by drop-with-cookie + * action are going to pass cookie_index 0 to lookup. + */ + cookie_index = 1; + err = idr_alloc_u32(&mlxsw_afa->cookie_idr, cookie, &cookie_index, + MLXSW_AFA_COOKIE_INDEX_MAX, GFP_KERNEL); + if (err) + goto err_idr_alloc; + cookie->cookie_index = cookie_index; + return cookie; + +err_idr_alloc: + rhashtable_remove_fast(&mlxsw_afa->cookie_ht, &cookie->ht_node, + mlxsw_afa_cookie_ht_params); +err_rhashtable_insert: + kfree(cookie); + return ERR_PTR(err); +} + +static void mlxsw_afa_cookie_destroy(struct mlxsw_afa *mlxsw_afa, + struct mlxsw_afa_cookie *cookie) +{ + idr_remove(&mlxsw_afa->cookie_idr, cookie->cookie_index); + rhashtable_remove_fast(&mlxsw_afa->cookie_ht, &cookie->ht_node, + mlxsw_afa_cookie_ht_params); + kfree_rcu(cookie, rcu); +} + +static struct mlxsw_afa_cookie * +mlxsw_afa_cookie_get(struct mlxsw_afa *mlxsw_afa, + const struct flow_action_cookie *fa_cookie) +{ + struct mlxsw_afa_cookie *cookie; + + cookie = rhashtable_lookup_fast(&mlxsw_afa->cookie_ht, fa_cookie, + mlxsw_afa_cookie_ht_params); + if (cookie) { + refcount_inc(&cookie->ref_count); + return cookie; + } + return mlxsw_afa_cookie_create(mlxsw_afa, fa_cookie); +} + +static void mlxsw_afa_cookie_put(struct mlxsw_afa *mlxsw_afa, + struct mlxsw_afa_cookie *cookie) +{ + if (!refcount_dec_and_test(&cookie->ref_count)) + return; + mlxsw_afa_cookie_destroy(mlxsw_afa, cookie); +} + +/* RCU read lock must be held */ +const struct flow_action_cookie * +mlxsw_afa_cookie_lookup(struct mlxsw_afa *mlxsw_afa, u32 cookie_index) +{ + struct mlxsw_afa_cookie *cookie; + + /* 0 index means no cookie */ + if (!cookie_index) + return NULL; + cookie = idr_find(&mlxsw_afa->cookie_idr, cookie_index); + if (!cookie) + return NULL; + return &cookie->fa_cookie; +} +EXPORT_SYMBOL(mlxsw_afa_cookie_lookup); + +struct mlxsw_afa_cookie_ref { + struct mlxsw_afa_resource resource; + struct mlxsw_afa_cookie *cookie; +}; + +static void +mlxsw_afa_cookie_ref_destroy(struct mlxsw_afa_block *block, + struct mlxsw_afa_cookie_ref *cookie_ref) +{ + mlxsw_afa_resource_del(&cookie_ref->resource); + mlxsw_afa_cookie_put(block->afa, cookie_ref->cookie); + kfree(cookie_ref); +} + +static void +mlxsw_afa_cookie_ref_destructor(struct mlxsw_afa_block *block, + struct mlxsw_afa_resource *resource) +{ + struct mlxsw_afa_cookie_ref *cookie_ref; + + cookie_ref = container_of(resource, struct mlxsw_afa_cookie_ref, + resource); + mlxsw_afa_cookie_ref_destroy(block, cookie_ref); +} + +static struct mlxsw_afa_cookie_ref * +mlxsw_afa_cookie_ref_create(struct mlxsw_afa_block *block, + const struct flow_action_cookie *fa_cookie) +{ + struct mlxsw_afa_cookie_ref *cookie_ref; + struct mlxsw_afa_cookie *cookie; + int err; + + cookie_ref = kzalloc(sizeof(*cookie_ref), GFP_KERNEL); + if (!cookie_ref) + return ERR_PTR(-ENOMEM); + cookie = mlxsw_afa_cookie_get(block->afa, fa_cookie); + if (IS_ERR(cookie)) { + err = PTR_ERR(cookie); + goto err_cookie_get; + } + cookie_ref->cookie = cookie; + cookie_ref->resource.destructor = mlxsw_afa_cookie_ref_destructor; + mlxsw_afa_resource_add(block, &cookie_ref->resource); + return cookie_ref; + +err_cookie_get: + kfree(cookie_ref); + return ERR_PTR(err); +} + #define MLXSW_AFA_ONE_ACTION_LEN 32 #define MLXSW_AFA_PAYLOAD_OFFSET 4 @@ -747,97 +956,170 @@ int mlxsw_afa_block_append_vlan_modify(struct mlxsw_afa_block *block, } EXPORT_SYMBOL(mlxsw_afa_block_append_vlan_modify); -/* Trap / Discard Action - * --------------------- - * The Trap / Discard action enables trapping / mirroring packets to the CPU +/* Trap Action / Trap With Userdef Action + * -------------------------------------- + * The Trap action enables trapping / mirroring packets to the CPU * as well as discarding packets. * The ACL Trap / Discard separates the forward/discard control from CPU * trap control. In addition, the Trap / Discard action enables activating * SPAN (port mirroring). + * + * The Trap with userdef action action has the same functionality as + * the Trap action with addition of user defined value that can be set + * and used by higher layer applications. */ -#define MLXSW_AFA_TRAPDISC_CODE 0x03 -#define MLXSW_AFA_TRAPDISC_SIZE 1 +#define MLXSW_AFA_TRAP_CODE 0x03 +#define MLXSW_AFA_TRAP_SIZE 1 -enum mlxsw_afa_trapdisc_trap_action { - MLXSW_AFA_TRAPDISC_TRAP_ACTION_NOP = 0, - MLXSW_AFA_TRAPDISC_TRAP_ACTION_TRAP = 2, +#define MLXSW_AFA_TRAPWU_CODE 0x04 +#define MLXSW_AFA_TRAPWU_SIZE 2 + +enum mlxsw_afa_trap_trap_action { + MLXSW_AFA_TRAP_TRAP_ACTION_NOP = 0, + MLXSW_AFA_TRAP_TRAP_ACTION_TRAP = 2, }; -/* afa_trapdisc_trap_action +/* afa_trap_trap_action * Trap Action. */ -MLXSW_ITEM32(afa, trapdisc, trap_action, 0x00, 24, 4); +MLXSW_ITEM32(afa, trap, trap_action, 0x00, 24, 4); -enum mlxsw_afa_trapdisc_forward_action { - MLXSW_AFA_TRAPDISC_FORWARD_ACTION_FORWARD = 1, - MLXSW_AFA_TRAPDISC_FORWARD_ACTION_DISCARD = 3, +enum mlxsw_afa_trap_forward_action { + MLXSW_AFA_TRAP_FORWARD_ACTION_FORWARD = 1, + MLXSW_AFA_TRAP_FORWARD_ACTION_DISCARD = 3, }; -/* afa_trapdisc_forward_action +/* afa_trap_forward_action * Forward Action. */ -MLXSW_ITEM32(afa, trapdisc, forward_action, 0x00, 0, 4); +MLXSW_ITEM32(afa, trap, forward_action, 0x00, 0, 4); -/* afa_trapdisc_trap_id +/* afa_trap_trap_id * Trap ID to configure. */ -MLXSW_ITEM32(afa, trapdisc, trap_id, 0x04, 0, 9); +MLXSW_ITEM32(afa, trap, trap_id, 0x04, 0, 9); -/* afa_trapdisc_mirror_agent +/* afa_trap_mirror_agent * Mirror agent. */ -MLXSW_ITEM32(afa, trapdisc, mirror_agent, 0x08, 29, 3); +MLXSW_ITEM32(afa, trap, mirror_agent, 0x08, 29, 3); -/* afa_trapdisc_mirror_enable +/* afa_trap_mirror_enable * Mirror enable. */ -MLXSW_ITEM32(afa, trapdisc, mirror_enable, 0x08, 24, 1); +MLXSW_ITEM32(afa, trap, mirror_enable, 0x08, 24, 1); + +/* user_def_val + * Value for the SW usage. Can be used to pass information of which + * rule has caused a trap. This may be overwritten by later traps. + * This field does a set on the packet's user_def_val only if this + * is the first trap_id or if the trap_id has replaced the previous + * packet's trap_id. + */ +MLXSW_ITEM32(afa, trap, user_def_val, 0x0C, 0, 20); static inline void -mlxsw_afa_trapdisc_pack(char *payload, - enum mlxsw_afa_trapdisc_trap_action trap_action, - enum mlxsw_afa_trapdisc_forward_action forward_action, - u16 trap_id) +mlxsw_afa_trap_pack(char *payload, + enum mlxsw_afa_trap_trap_action trap_action, + enum mlxsw_afa_trap_forward_action forward_action, + u16 trap_id) { - mlxsw_afa_trapdisc_trap_action_set(payload, trap_action); - mlxsw_afa_trapdisc_forward_action_set(payload, forward_action); - mlxsw_afa_trapdisc_trap_id_set(payload, trap_id); + mlxsw_afa_trap_trap_action_set(payload, trap_action); + mlxsw_afa_trap_forward_action_set(payload, forward_action); + mlxsw_afa_trap_trap_id_set(payload, trap_id); } static inline void -mlxsw_afa_trapdisc_mirror_pack(char *payload, bool mirror_enable, - u8 mirror_agent) +mlxsw_afa_trapwu_pack(char *payload, + enum mlxsw_afa_trap_trap_action trap_action, + enum mlxsw_afa_trap_forward_action forward_action, + u16 trap_id, u32 user_def_val) { - mlxsw_afa_trapdisc_mirror_enable_set(payload, mirror_enable); - mlxsw_afa_trapdisc_mirror_agent_set(payload, mirror_agent); + mlxsw_afa_trap_pack(payload, trap_action, forward_action, trap_id); + mlxsw_afa_trap_user_def_val_set(payload, user_def_val); } -int mlxsw_afa_block_append_drop(struct mlxsw_afa_block *block) +static inline void +mlxsw_afa_trap_mirror_pack(char *payload, bool mirror_enable, + u8 mirror_agent) { - char *act = mlxsw_afa_block_append_action(block, - MLXSW_AFA_TRAPDISC_CODE, - MLXSW_AFA_TRAPDISC_SIZE); + mlxsw_afa_trap_mirror_enable_set(payload, mirror_enable); + mlxsw_afa_trap_mirror_agent_set(payload, mirror_agent); +} + +static int mlxsw_afa_block_append_drop_plain(struct mlxsw_afa_block *block, + bool ingress) +{ + char *act = mlxsw_afa_block_append_action(block, MLXSW_AFA_TRAP_CODE, + MLXSW_AFA_TRAP_SIZE); if (IS_ERR(act)) return PTR_ERR(act); - mlxsw_afa_trapdisc_pack(act, MLXSW_AFA_TRAPDISC_TRAP_ACTION_NOP, - MLXSW_AFA_TRAPDISC_FORWARD_ACTION_DISCARD, 0); + mlxsw_afa_trap_pack(act, MLXSW_AFA_TRAP_TRAP_ACTION_TRAP, + MLXSW_AFA_TRAP_FORWARD_ACTION_DISCARD, + ingress ? MLXSW_TRAP_ID_DISCARD_INGRESS_ACL : + MLXSW_TRAP_ID_DISCARD_EGRESS_ACL); return 0; } + +static int +mlxsw_afa_block_append_drop_with_cookie(struct mlxsw_afa_block *block, + bool ingress, + const struct flow_action_cookie *fa_cookie, + struct netlink_ext_ack *extack) +{ + struct mlxsw_afa_cookie_ref *cookie_ref; + u32 cookie_index; + char *act; + int err; + + cookie_ref = mlxsw_afa_cookie_ref_create(block, fa_cookie); + if (IS_ERR(cookie_ref)) { + NL_SET_ERR_MSG_MOD(extack, "Cannot create cookie for drop action"); + return PTR_ERR(cookie_ref); + } + cookie_index = cookie_ref->cookie->cookie_index; + + act = mlxsw_afa_block_append_action(block, MLXSW_AFA_TRAPWU_CODE, + MLXSW_AFA_TRAPWU_SIZE); + if (IS_ERR(act)) { + NL_SET_ERR_MSG_MOD(extack, "Cannot append drop with cookie action"); + err = PTR_ERR(act); + goto err_append_action; + } + mlxsw_afa_trapwu_pack(act, MLXSW_AFA_TRAP_TRAP_ACTION_TRAP, + MLXSW_AFA_TRAP_FORWARD_ACTION_DISCARD, + ingress ? MLXSW_TRAP_ID_DISCARD_INGRESS_ACL : + MLXSW_TRAP_ID_DISCARD_EGRESS_ACL, + cookie_index); + return 0; + +err_append_action: + mlxsw_afa_cookie_ref_destroy(block, cookie_ref); + return err; +} + +int mlxsw_afa_block_append_drop(struct mlxsw_afa_block *block, bool ingress, + const struct flow_action_cookie *fa_cookie, + struct netlink_ext_ack *extack) +{ + return fa_cookie ? + mlxsw_afa_block_append_drop_with_cookie(block, ingress, + fa_cookie, extack) : + mlxsw_afa_block_append_drop_plain(block, ingress); +} EXPORT_SYMBOL(mlxsw_afa_block_append_drop); int mlxsw_afa_block_append_trap(struct mlxsw_afa_block *block, u16 trap_id) { - char *act = mlxsw_afa_block_append_action(block, - MLXSW_AFA_TRAPDISC_CODE, - MLXSW_AFA_TRAPDISC_SIZE); + char *act = mlxsw_afa_block_append_action(block, MLXSW_AFA_TRAP_CODE, + MLXSW_AFA_TRAP_SIZE); if (IS_ERR(act)) return PTR_ERR(act); - mlxsw_afa_trapdisc_pack(act, MLXSW_AFA_TRAPDISC_TRAP_ACTION_TRAP, - MLXSW_AFA_TRAPDISC_FORWARD_ACTION_DISCARD, - trap_id); + mlxsw_afa_trap_pack(act, MLXSW_AFA_TRAP_TRAP_ACTION_TRAP, + MLXSW_AFA_TRAP_FORWARD_ACTION_DISCARD, trap_id); return 0; } EXPORT_SYMBOL(mlxsw_afa_block_append_trap); @@ -845,15 +1127,13 @@ EXPORT_SYMBOL(mlxsw_afa_block_append_trap); int mlxsw_afa_block_append_trap_and_forward(struct mlxsw_afa_block *block, u16 trap_id) { - char *act = mlxsw_afa_block_append_action(block, - MLXSW_AFA_TRAPDISC_CODE, - MLXSW_AFA_TRAPDISC_SIZE); + char *act = mlxsw_afa_block_append_action(block, MLXSW_AFA_TRAP_CODE, + MLXSW_AFA_TRAP_SIZE); if (IS_ERR(act)) return PTR_ERR(act); - mlxsw_afa_trapdisc_pack(act, MLXSW_AFA_TRAPDISC_TRAP_ACTION_TRAP, - MLXSW_AFA_TRAPDISC_FORWARD_ACTION_FORWARD, - trap_id); + mlxsw_afa_trap_pack(act, MLXSW_AFA_TRAP_TRAP_ACTION_TRAP, + MLXSW_AFA_TRAP_FORWARD_ACTION_FORWARD, trap_id); return 0; } EXPORT_SYMBOL(mlxsw_afa_block_append_trap_and_forward); @@ -920,13 +1200,13 @@ mlxsw_afa_block_append_allocated_mirror(struct mlxsw_afa_block *block, u8 mirror_agent) { char *act = mlxsw_afa_block_append_action(block, - MLXSW_AFA_TRAPDISC_CODE, - MLXSW_AFA_TRAPDISC_SIZE); + MLXSW_AFA_TRAP_CODE, + MLXSW_AFA_TRAP_SIZE); if (IS_ERR(act)) return PTR_ERR(act); - mlxsw_afa_trapdisc_pack(act, MLXSW_AFA_TRAPDISC_TRAP_ACTION_NOP, - MLXSW_AFA_TRAPDISC_FORWARD_ACTION_FORWARD, 0); - mlxsw_afa_trapdisc_mirror_pack(act, true, mirror_agent); + mlxsw_afa_trap_pack(act, MLXSW_AFA_TRAP_TRAP_ACTION_NOP, + MLXSW_AFA_TRAP_FORWARD_ACTION_FORWARD, 0); + mlxsw_afa_trap_mirror_pack(act, true, mirror_agent); return 0; } @@ -958,6 +1238,179 @@ err_append_allocated_mirror: } EXPORT_SYMBOL(mlxsw_afa_block_append_mirror); +/* QoS Action + * ---------- + * The QOS_ACTION is used for manipulating the QoS attributes of a packet. It + * can be used to change the DCSP, ECN, Color and Switch Priority of the packet. + * Note that PCP field can be changed using the VLAN action. + */ + +#define MLXSW_AFA_QOS_CODE 0x06 +#define MLXSW_AFA_QOS_SIZE 1 + +enum mlxsw_afa_qos_ecn_cmd { + /* Do nothing */ + MLXSW_AFA_QOS_ECN_CMD_NOP, + /* Set ECN to afa_qos_ecn */ + MLXSW_AFA_QOS_ECN_CMD_SET, +}; + +/* afa_qos_ecn_cmd + */ +MLXSW_ITEM32(afa, qos, ecn_cmd, 0x04, 29, 3); + +/* afa_qos_ecn + * ECN value. + */ +MLXSW_ITEM32(afa, qos, ecn, 0x04, 24, 2); + +enum mlxsw_afa_qos_dscp_cmd { + /* Do nothing */ + MLXSW_AFA_QOS_DSCP_CMD_NOP, + /* Set DSCP 3 LSB bits according to dscp[2:0] */ + MLXSW_AFA_QOS_DSCP_CMD_SET_3LSB, + /* Set DSCP 3 MSB bits according to dscp[5:3] */ + MLXSW_AFA_QOS_DSCP_CMD_SET_3MSB, + /* Set DSCP 6 bits according to dscp[5:0] */ + MLXSW_AFA_QOS_DSCP_CMD_SET_ALL, +}; + +/* afa_qos_dscp_cmd + * DSCP command. + */ +MLXSW_ITEM32(afa, qos, dscp_cmd, 0x04, 14, 2); + +/* afa_qos_dscp + * DSCP value. + */ +MLXSW_ITEM32(afa, qos, dscp, 0x04, 0, 6); + +enum mlxsw_afa_qos_switch_prio_cmd { + /* Do nothing */ + MLXSW_AFA_QOS_SWITCH_PRIO_CMD_NOP, + /* Set Switch Priority to afa_qos_switch_prio */ + MLXSW_AFA_QOS_SWITCH_PRIO_CMD_SET, +}; + +/* afa_qos_switch_prio_cmd + */ +MLXSW_ITEM32(afa, qos, switch_prio_cmd, 0x08, 14, 2); + +/* afa_qos_switch_prio + * Switch Priority. + */ +MLXSW_ITEM32(afa, qos, switch_prio, 0x08, 0, 4); + +enum mlxsw_afa_qos_dscp_rw { + MLXSW_AFA_QOS_DSCP_RW_PRESERVE, + MLXSW_AFA_QOS_DSCP_RW_SET, + MLXSW_AFA_QOS_DSCP_RW_CLEAR, +}; + +/* afa_qos_dscp_rw + * DSCP Re-write Enable. Controlling the rewrite_enable for DSCP. + */ +MLXSW_ITEM32(afa, qos, dscp_rw, 0x0C, 30, 2); + +static inline void +mlxsw_afa_qos_ecn_pack(char *payload, + enum mlxsw_afa_qos_ecn_cmd ecn_cmd, u8 ecn) +{ + mlxsw_afa_qos_ecn_cmd_set(payload, ecn_cmd); + mlxsw_afa_qos_ecn_set(payload, ecn); +} + +static inline void +mlxsw_afa_qos_dscp_pack(char *payload, + enum mlxsw_afa_qos_dscp_cmd dscp_cmd, u8 dscp) +{ + mlxsw_afa_qos_dscp_cmd_set(payload, dscp_cmd); + mlxsw_afa_qos_dscp_set(payload, dscp); +} + +static inline void +mlxsw_afa_qos_switch_prio_pack(char *payload, + enum mlxsw_afa_qos_switch_prio_cmd prio_cmd, + u8 prio) +{ + mlxsw_afa_qos_switch_prio_cmd_set(payload, prio_cmd); + mlxsw_afa_qos_switch_prio_set(payload, prio); +} + +static int __mlxsw_afa_block_append_qos_dsfield(struct mlxsw_afa_block *block, + bool set_dscp, u8 dscp, + bool set_ecn, u8 ecn, + struct netlink_ext_ack *extack) +{ + char *act = mlxsw_afa_block_append_action(block, + MLXSW_AFA_QOS_CODE, + MLXSW_AFA_QOS_SIZE); + + if (IS_ERR(act)) { + NL_SET_ERR_MSG_MOD(extack, "Cannot append QOS action"); + return PTR_ERR(act); + } + + if (set_ecn) + mlxsw_afa_qos_ecn_pack(act, MLXSW_AFA_QOS_ECN_CMD_SET, ecn); + if (set_dscp) { + mlxsw_afa_qos_dscp_pack(act, MLXSW_AFA_QOS_DSCP_CMD_SET_ALL, + dscp); + mlxsw_afa_qos_dscp_rw_set(act, MLXSW_AFA_QOS_DSCP_RW_CLEAR); + } + + return 0; +} + +int mlxsw_afa_block_append_qos_dsfield(struct mlxsw_afa_block *block, + u8 dsfield, + struct netlink_ext_ack *extack) +{ + return __mlxsw_afa_block_append_qos_dsfield(block, + true, dsfield >> 2, + true, dsfield & 0x03, + extack); +} +EXPORT_SYMBOL(mlxsw_afa_block_append_qos_dsfield); + +int mlxsw_afa_block_append_qos_dscp(struct mlxsw_afa_block *block, + u8 dscp, struct netlink_ext_ack *extack) +{ + return __mlxsw_afa_block_append_qos_dsfield(block, + true, dscp, + false, 0, + extack); +} +EXPORT_SYMBOL(mlxsw_afa_block_append_qos_dscp); + +int mlxsw_afa_block_append_qos_ecn(struct mlxsw_afa_block *block, + u8 ecn, struct netlink_ext_ack *extack) +{ + return __mlxsw_afa_block_append_qos_dsfield(block, + false, 0, + true, ecn, + extack); +} +EXPORT_SYMBOL(mlxsw_afa_block_append_qos_ecn); + +int mlxsw_afa_block_append_qos_switch_prio(struct mlxsw_afa_block *block, + u8 prio, + struct netlink_ext_ack *extack) +{ + char *act = mlxsw_afa_block_append_action(block, + MLXSW_AFA_QOS_CODE, + MLXSW_AFA_QOS_SIZE); + + if (IS_ERR(act)) { + NL_SET_ERR_MSG_MOD(extack, "Cannot append QOS action"); + return PTR_ERR(act); + } + mlxsw_afa_qos_switch_prio_pack(act, MLXSW_AFA_QOS_SWITCH_PRIO_CMD_SET, + prio); + return 0; +} +EXPORT_SYMBOL(mlxsw_afa_block_append_qos_switch_prio); + /* Forwarding Action * ----------------- * Forwarding Action can be used to implement Policy Based Switching (PBS) diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h index 0e3a59dda12e..8c2705e16ef7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h @@ -6,6 +6,7 @@ #include <linux/types.h> #include <linux/netdevice.h> +#include <net/flow_offload.h> struct mlxsw_afa; struct mlxsw_afa_block; @@ -42,7 +43,11 @@ int mlxsw_afa_block_activity_get(struct mlxsw_afa_block *block, bool *activity); int mlxsw_afa_block_continue(struct mlxsw_afa_block *block); int mlxsw_afa_block_jump(struct mlxsw_afa_block *block, u16 group_id); int mlxsw_afa_block_terminate(struct mlxsw_afa_block *block); -int mlxsw_afa_block_append_drop(struct mlxsw_afa_block *block); +const struct flow_action_cookie * +mlxsw_afa_cookie_lookup(struct mlxsw_afa *mlxsw_afa, u32 cookie_index); +int mlxsw_afa_block_append_drop(struct mlxsw_afa_block *block, bool ingress, + const struct flow_action_cookie *fa_cookie, + struct netlink_ext_ack *extack); int mlxsw_afa_block_append_trap(struct mlxsw_afa_block *block, u16 trap_id); int mlxsw_afa_block_append_trap_and_forward(struct mlxsw_afa_block *block, u16 trap_id); @@ -57,6 +62,16 @@ int mlxsw_afa_block_append_fwd(struct mlxsw_afa_block *block, int mlxsw_afa_block_append_vlan_modify(struct mlxsw_afa_block *block, u16 vid, u8 pcp, u8 et, struct netlink_ext_ack *extack); +int mlxsw_afa_block_append_qos_switch_prio(struct mlxsw_afa_block *block, + u8 prio, + struct netlink_ext_ack *extack); +int mlxsw_afa_block_append_qos_dsfield(struct mlxsw_afa_block *block, + u8 dsfield, + struct netlink_ext_ack *extack); +int mlxsw_afa_block_append_qos_dscp(struct mlxsw_afa_block *block, + u8 dscp, struct netlink_ext_ack *extack); +int mlxsw_afa_block_append_qos_ecn(struct mlxsw_afa_block *block, + u8 ecn, struct netlink_ext_ack *extack); int mlxsw_afa_block_append_allocated_counter(struct mlxsw_afa_block *block, u32 counter_index); int mlxsw_afa_block_append_counter(struct mlxsw_afa_block *block, diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c index feb4672a5ac0..9f6905fa6b47 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c @@ -9,6 +9,41 @@ #include "item.h" #include "core_acl_flex_keys.h" +/* For the purpose of the driver, define an internal storage scratchpad + * that will be used to store key/mask values. For each defined element type + * define an internal storage geometry. + * + * When adding new elements, MLXSW_AFK_ELEMENT_STORAGE_SIZE must be increased + * accordingly. + */ +static const struct mlxsw_afk_element_info mlxsw_afk_element_infos[] = { + MLXSW_AFK_ELEMENT_INFO_U32(SRC_SYS_PORT, 0x00, 16, 16), + MLXSW_AFK_ELEMENT_INFO_BUF(DMAC_32_47, 0x04, 2), + MLXSW_AFK_ELEMENT_INFO_BUF(DMAC_0_31, 0x06, 4), + MLXSW_AFK_ELEMENT_INFO_BUF(SMAC_32_47, 0x0A, 2), + MLXSW_AFK_ELEMENT_INFO_BUF(SMAC_0_31, 0x0C, 4), + MLXSW_AFK_ELEMENT_INFO_U32(ETHERTYPE, 0x00, 0, 16), + MLXSW_AFK_ELEMENT_INFO_U32(IP_PROTO, 0x10, 0, 8), + MLXSW_AFK_ELEMENT_INFO_U32(VID, 0x10, 8, 12), + MLXSW_AFK_ELEMENT_INFO_U32(PCP, 0x10, 20, 3), + MLXSW_AFK_ELEMENT_INFO_U32(TCP_FLAGS, 0x10, 23, 9), + MLXSW_AFK_ELEMENT_INFO_U32(DST_L4_PORT, 0x14, 0, 16), + MLXSW_AFK_ELEMENT_INFO_U32(SRC_L4_PORT, 0x14, 16, 16), + MLXSW_AFK_ELEMENT_INFO_U32(IP_TTL_, 0x18, 0, 8), + MLXSW_AFK_ELEMENT_INFO_U32(IP_ECN, 0x18, 9, 2), + MLXSW_AFK_ELEMENT_INFO_U32(IP_DSCP, 0x18, 11, 6), + MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_8_10, 0x18, 17, 3), + MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_0_7, 0x18, 20, 8), + MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_96_127, 0x20, 4), + MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_64_95, 0x24, 4), + MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_32_63, 0x28, 4), + MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_0_31, 0x2C, 4), + MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_96_127, 0x30, 4), + MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_64_95, 0x34, 4), + MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_32_63, 0x38, 4), + MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_0_31, 0x3C, 4), +}; + struct mlxsw_afk { struct list_head key_info_list; unsigned int max_blocks; @@ -26,13 +61,15 @@ static bool mlxsw_afk_blocks_check(struct mlxsw_afk *mlxsw_afk) const struct mlxsw_afk_block *block = &mlxsw_afk->blocks[i]; for (j = 0; j < block->instances_count; j++) { + const struct mlxsw_afk_element_info *elinfo; struct mlxsw_afk_element_inst *elinst; elinst = &block->instances[j]; - if (elinst->type != elinst->info->type || + elinfo = &mlxsw_afk_element_infos[elinst->element]; + if (elinst->type != elinfo->type || (!elinst->avoid_size_check && elinst->item.size.bits != - elinst->info->item.size.bits)) + elinfo->item.size.bits)) return false; } } @@ -72,7 +109,7 @@ struct mlxsw_afk_key_info { * is index inside "blocks" */ struct mlxsw_afk_element_usage elusage; - const struct mlxsw_afk_block *blocks[0]; + const struct mlxsw_afk_block *blocks[]; }; static bool @@ -116,7 +153,7 @@ static void mlxsw_afk_picker_count_hits(struct mlxsw_afk *mlxsw_afk, struct mlxsw_afk_element_inst *elinst; elinst = &block->instances[j]; - if (elinst->info->element == element) { + if (elinst->element == element) { __set_bit(element, picker->hits[i].element); picker->hits[i].total++; } @@ -301,7 +338,7 @@ mlxsw_afk_block_elinst_get(const struct mlxsw_afk_block *block, struct mlxsw_afk_element_inst *elinst; elinst = &block->instances[i]; - if (elinst->info->element == element) + if (elinst->element == element) return elinst; } return NULL; @@ -409,9 +446,12 @@ static void mlxsw_sp_afk_encode_one(const struct mlxsw_afk_element_inst *elinst, char *output, char *storage, int u32_diff) { - const struct mlxsw_item *storage_item = &elinst->info->item; const struct mlxsw_item *output_item = &elinst->item; + const struct mlxsw_afk_element_info *elinfo; + const struct mlxsw_item *storage_item; + elinfo = &mlxsw_afk_element_infos[elinst->element]; + storage_item = &elinfo->item; if (elinst->type == MLXSW_AFK_ELEMENT_TYPE_U32) mlxsw_sp_afk_encode_u32(storage_item, output_item, storage, output, u32_diff); diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h index cb229b55ecc4..a47a17c04c62 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h @@ -69,42 +69,10 @@ struct mlxsw_afk_element_info { MLXSW_AFK_ELEMENT_INFO(MLXSW_AFK_ELEMENT_TYPE_BUF, \ _element, _offset, 0, _size) -/* For the purpose of the driver, define an internal storage scratchpad - * that will be used to store key/mask values. For each defined element type - * define an internal storage geometry. - */ -static const struct mlxsw_afk_element_info mlxsw_afk_element_infos[] = { - MLXSW_AFK_ELEMENT_INFO_U32(SRC_SYS_PORT, 0x00, 16, 16), - MLXSW_AFK_ELEMENT_INFO_BUF(DMAC_32_47, 0x04, 2), - MLXSW_AFK_ELEMENT_INFO_BUF(DMAC_0_31, 0x06, 4), - MLXSW_AFK_ELEMENT_INFO_BUF(SMAC_32_47, 0x0A, 2), - MLXSW_AFK_ELEMENT_INFO_BUF(SMAC_0_31, 0x0C, 4), - MLXSW_AFK_ELEMENT_INFO_U32(ETHERTYPE, 0x00, 0, 16), - MLXSW_AFK_ELEMENT_INFO_U32(IP_PROTO, 0x10, 0, 8), - MLXSW_AFK_ELEMENT_INFO_U32(VID, 0x10, 8, 12), - MLXSW_AFK_ELEMENT_INFO_U32(PCP, 0x10, 20, 3), - MLXSW_AFK_ELEMENT_INFO_U32(TCP_FLAGS, 0x10, 23, 9), - MLXSW_AFK_ELEMENT_INFO_U32(DST_L4_PORT, 0x14, 0, 16), - MLXSW_AFK_ELEMENT_INFO_U32(SRC_L4_PORT, 0x14, 16, 16), - MLXSW_AFK_ELEMENT_INFO_U32(IP_TTL_, 0x18, 0, 8), - MLXSW_AFK_ELEMENT_INFO_U32(IP_ECN, 0x18, 9, 2), - MLXSW_AFK_ELEMENT_INFO_U32(IP_DSCP, 0x18, 11, 6), - MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_8_10, 0x18, 17, 3), - MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_0_7, 0x18, 20, 8), - MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_96_127, 0x20, 4), - MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_64_95, 0x24, 4), - MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_32_63, 0x28, 4), - MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_0_31, 0x2C, 4), - MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_96_127, 0x30, 4), - MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_64_95, 0x34, 4), - MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_32_63, 0x38, 4), - MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_0_31, 0x3C, 4), -}; - #define MLXSW_AFK_ELEMENT_STORAGE_SIZE 0x40 struct mlxsw_afk_element_inst { /* element instance in actual block */ - const struct mlxsw_afk_element_info *info; + enum mlxsw_afk_element element; enum mlxsw_afk_element_type type; struct mlxsw_item item; /* element geometry in block */ int u32_key_diff; /* in case value needs to be adjusted before write @@ -116,7 +84,7 @@ struct mlxsw_afk_element_inst { /* element instance in actual block */ #define MLXSW_AFK_ELEMENT_INST(_type, _element, _offset, \ _shift, _size, _u32_key_diff, _avoid_size_check) \ { \ - .info = &mlxsw_afk_element_infos[MLXSW_AFK_ELEMENT_##_element], \ + .element = MLXSW_AFK_ELEMENT_##_element, \ .type = _type, \ .item = { \ .offset = _offset, \ diff --git a/drivers/net/ethernet/mellanox/mlxsw/i2c.c b/drivers/net/ethernet/mellanox/mlxsw/i2c.c index 34566eb62c47..939b692ffc33 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/i2c.c +++ b/drivers/net/ethernet/mellanox/mlxsw/i2c.c @@ -53,6 +53,7 @@ /** * struct mlxsw_i2c - device private data: + * @cmd: command attributes; * @cmd.mb_size_in: input mailbox size; * @cmd.mb_off_in: input mailbox offset in register space; * @cmd.mb_size_out: output mailbox size; diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index e9ded1a6e131..fd0e97de44e7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -575,6 +575,15 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci, rx_info.trap_id = mlxsw_pci_cqe_trap_id_get(cqe); + if (rx_info.trap_id == MLXSW_TRAP_ID_DISCARD_INGRESS_ACL || + rx_info.trap_id == MLXSW_TRAP_ID_DISCARD_EGRESS_ACL) { + u32 cookie_index = 0; + + if (mlxsw_pci->max_cqe_ver >= MLXSW_PCI_CQE_V2) + cookie_index = mlxsw_pci_cqe2_user_def_val_orig_pkt_len_get(cqe); + mlxsw_skb_cb(skb)->cookie_index = cookie_index; + } + byte_count = mlxsw_pci_cqe_byte_count_get(cqe); if (mlxsw_pci_cqe_crc_get(cqe_v, cqe)) byte_count -= ETH_FCS_LEN; diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h b/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h index 43fa8c85b5d9..32c7cabfb261 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h +++ b/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h @@ -25,8 +25,6 @@ #define MLXSW_PCI_CIR_CTRL_STATUS_SHIFT 24 #define MLXSW_PCI_CIR_TIMEOUT_MSECS 1000 -#define MLXSW_PCI_SW_RESET 0xF0010 -#define MLXSW_PCI_SW_RESET_RST_BIT BIT(0) #define MLXSW_PCI_SW_RESET_TIMEOUT_MSECS 900000 #define MLXSW_PCI_SW_RESET_WAIT_MSECS 200 #define MLXSW_PCI_FW_READY 0xA1844 @@ -210,6 +208,11 @@ MLXSW_ITEM32(pci, cqe0, dqn, 0x0C, 1, 5); MLXSW_ITEM32(pci, cqe12, dqn, 0x0C, 1, 6); mlxsw_pci_cqe_item_helpers(dqn, 0, 12, 12); +/* pci_cqe_user_def_val_orig_pkt_len + * When trap_id is an ACL: User defined value from policy engine action. + */ +MLXSW_ITEM32(pci, cqe2, user_def_val_orig_pkt_len, 0x14, 0, 20); + /* pci_cqe_owner * Ownership bit. */ diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index e05d1d1be2fd..9b39b8e70519 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -621,7 +621,7 @@ static inline void mlxsw_reg_sfn_pack(char *payload) { MLXSW_REG_ZERO(sfn, payload); mlxsw_reg_sfn_swid_set(payload, 0); - mlxsw_reg_sfn_end_set(payload, 1); + mlxsw_reg_sfn_end_set(payload, 0); mlxsw_reg_sfn_num_rec_set(payload, MLXSW_REG_SFN_REC_MAX_COUNT); } @@ -3296,6 +3296,12 @@ MLXSW_ITEM32(reg, qpcr, g, 0x00, 14, 2); */ MLXSW_ITEM32(reg, qpcr, pid, 0x00, 0, 14); +/* reg_qpcr_clear_counter + * Clear counters. + * Access: OP + */ +MLXSW_ITEM32(reg, qpcr, clear_counter, 0x04, 31, 1); + /* reg_qpcr_color_aware * Is the policer aware of colors. * Must be 0 (unaware) for cpu port. @@ -3393,6 +3399,17 @@ enum mlxsw_reg_qpcr_action { */ MLXSW_ITEM32(reg, qpcr, violate_action, 0x18, 0, 4); +/* reg_qpcr_violate_count + * Counts the number of times violate_action happened on this PID. + * Access: RW + */ +MLXSW_ITEM64(reg, qpcr, violate_count, 0x20, 0, 64); + +#define MLXSW_REG_QPCR_LOWEST_CIR 1 +#define MLXSW_REG_QPCR_HIGHEST_CIR (2 * 1000 * 1000 * 1000) /* 2Gpps */ +#define MLXSW_REG_QPCR_LOWEST_CBS 4 +#define MLXSW_REG_QPCR_HIGHEST_CBS 24 + static inline void mlxsw_reg_qpcr_pack(char *payload, u16 pid, enum mlxsw_reg_qpcr_ir_units ir_units, bool bytes, u32 cir, u16 cbs) @@ -5440,15 +5457,29 @@ enum mlxsw_reg_pmtm_module_type { /* Backplane with 4 lanes */ MLXSW_REG_PMTM_MODULE_TYPE_BP_4X, /* QSFP */ - MLXSW_REG_PMTM_MODULE_TYPE_BP_QSFP, + MLXSW_REG_PMTM_MODULE_TYPE_QSFP, /* SFP */ - MLXSW_REG_PMTM_MODULE_TYPE_BP_SFP, + MLXSW_REG_PMTM_MODULE_TYPE_SFP, /* Backplane with single lane */ MLXSW_REG_PMTM_MODULE_TYPE_BP_1X = 4, /* Backplane with two lane */ MLXSW_REG_PMTM_MODULE_TYPE_BP_2X = 8, - /* Chip2Chip */ - MLXSW_REG_PMTM_MODULE_TYPE_C2C = 10, + /* Chip2Chip4x */ + MLXSW_REG_PMTM_MODULE_TYPE_C2C4X = 10, + /* Chip2Chip2x */ + MLXSW_REG_PMTM_MODULE_TYPE_C2C2X, + /* Chip2Chip1x */ + MLXSW_REG_PMTM_MODULE_TYPE_C2C1X, + /* QSFP-DD */ + MLXSW_REG_PMTM_MODULE_TYPE_QSFP_DD = 14, + /* OSFP */ + MLXSW_REG_PMTM_MODULE_TYPE_OSFP, + /* SFP-DD */ + MLXSW_REG_PMTM_MODULE_TYPE_SFP_DD, + /* DSFP */ + MLXSW_REG_PMTM_MODULE_TYPE_DSFP, + /* Chip2Chip8x */ + MLXSW_REG_PMTM_MODULE_TYPE_C2C8X, }; /* reg_pmtm_module_type @@ -5506,12 +5537,10 @@ enum mlxsw_reg_htgt_trap_group { MLXSW_REG_HTGT_TRAP_GROUP_SP_PIM, MLXSW_REG_HTGT_TRAP_GROUP_SP_MULTICAST, MLXSW_REG_HTGT_TRAP_GROUP_SP_ARP, - MLXSW_REG_HTGT_TRAP_GROUP_SP_HOST_MISS, MLXSW_REG_HTGT_TRAP_GROUP_SP_ROUTER_EXP, MLXSW_REG_HTGT_TRAP_GROUP_SP_REMOTE_ROUTE, MLXSW_REG_HTGT_TRAP_GROUP_SP_IP2ME, MLXSW_REG_HTGT_TRAP_GROUP_SP_DHCP, - MLXSW_REG_HTGT_TRAP_GROUP_SP_RPF, MLXSW_REG_HTGT_TRAP_GROUP_SP_EVENT, MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_MLD, MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_ND, @@ -5526,9 +5555,11 @@ enum mlxsw_reg_htgt_trap_group { enum mlxsw_reg_htgt_discard_trap_group { MLXSW_REG_HTGT_DISCARD_TRAP_GROUP_BASE = MLXSW_REG_HTGT_TRAP_GROUP_MAX, + MLXSW_REG_HTGT_TRAP_GROUP_SP_DUMMY, MLXSW_REG_HTGT_TRAP_GROUP_SP_L2_DISCARDS, MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_DISCARDS, MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS, + MLXSW_REG_HTGT_TRAP_GROUP_SP_ACL_DISCARDS, }; /* reg_htgt_trap_group diff --git a/drivers/net/ethernet/mellanox/mlxsw/resources.h b/drivers/net/ethernet/mellanox/mlxsw/resources.h index 6534184cb942..d62496ef299c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/resources.h +++ b/drivers/net/ethernet/mellanox/mlxsw/resources.h @@ -18,6 +18,7 @@ enum mlxsw_res_id { MLXSW_RES_ID_CQE_V1, MLXSW_RES_ID_CQE_V2, MLXSW_RES_ID_COUNTER_POOL_SIZE, + MLXSW_RES_ID_COUNTER_BANK_SIZE, MLXSW_RES_ID_MAX_SPAN, MLXSW_RES_ID_COUNTER_SIZE_PACKETS_BYTES, MLXSW_RES_ID_COUNTER_SIZE_ROUTER_BASIC, @@ -75,6 +76,7 @@ static u16 mlxsw_res_ids[] = { [MLXSW_RES_ID_CQE_V1] = 0x2211, [MLXSW_RES_ID_CQE_V2] = 0x2212, [MLXSW_RES_ID_COUNTER_POOL_SIZE] = 0x2410, + [MLXSW_RES_ID_COUNTER_BANK_SIZE] = 0x2411, [MLXSW_RES_ID_MAX_SPAN] = 0x2420, [MLXSW_RES_ID_COUNTER_SIZE_PACKETS_BYTES] = 0x2443, [MLXSW_RES_ID_COUNTER_SIZE_ROUTER_BASIC] = 0x2449, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 7358b5bc7eb6..24ca8d5bc564 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -43,6 +43,7 @@ #include "spectrum_acl_flex_actions.h" #include "spectrum_span.h" #include "spectrum_ptp.h" +#include "spectrum_trap.h" #include "../mlxfw/mlxfw.h" #define MLXSW_SP1_FWREV_MAJOR 13 @@ -347,19 +348,6 @@ static void mlxsw_sp_fsm_release(struct mlxfw_dev *mlxfw_dev, u32 fwhandle) mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); } -static void mlxsw_sp_status_notify(struct mlxfw_dev *mlxfw_dev, - const char *msg, const char *comp_name, - u32 done_bytes, u32 total_bytes) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - - devlink_flash_update_status_notify(priv_to_devlink(mlxsw_sp->core), - msg, comp_name, - done_bytes, total_bytes); -} - static const struct mlxfw_dev_ops mlxsw_sp_mlxfw_dev_ops = { .component_query = mlxsw_sp_component_query, .fsm_lock = mlxsw_sp_fsm_lock, @@ -370,7 +358,6 @@ static const struct mlxfw_dev_ops mlxsw_sp_mlxfw_dev_ops = { .fsm_query_state = mlxsw_sp_fsm_query_state, .fsm_cancel = mlxsw_sp_fsm_cancel, .fsm_release = mlxsw_sp_fsm_release, - .status_notify = mlxsw_sp_status_notify, }; static int mlxsw_sp_firmware_flash(struct mlxsw_sp *mlxsw_sp, @@ -382,16 +369,15 @@ static int mlxsw_sp_firmware_flash(struct mlxsw_sp *mlxsw_sp, .ops = &mlxsw_sp_mlxfw_dev_ops, .psid = mlxsw_sp->bus_info->psid, .psid_size = strlen(mlxsw_sp->bus_info->psid), + .devlink = priv_to_devlink(mlxsw_sp->core), }, .mlxsw_sp = mlxsw_sp }; int err; mlxsw_core_fw_flash_start(mlxsw_sp->core); - devlink_flash_update_begin_notify(priv_to_devlink(mlxsw_sp->core)); err = mlxfw_firmware_flash(&mlxsw_sp_mlxfw_dev.mlxfw_dev, firmware, extack); - devlink_flash_update_end_notify(priv_to_devlink(mlxsw_sp->core)); mlxsw_core_fw_flash_end(mlxsw_sp->core); return err; @@ -1798,6 +1784,8 @@ static int mlxsw_sp_setup_tc(struct net_device *dev, enum tc_setup_type type, return mlxsw_sp_setup_tc_ets(mlxsw_sp_port, type_data); case TC_SETUP_QDISC_TBF: return mlxsw_sp_setup_tc_tbf(mlxsw_sp_port, type_data); + case TC_SETUP_QDISC_FIFO: + return mlxsw_sp_setup_tc_fifo(mlxsw_sp_port, type_data); default: return -EOPNOTSUPP; } @@ -2243,6 +2231,15 @@ static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_rfc_3635_stats[] = { #define MLXSW_SP_PORT_HW_RFC_3635_STATS_LEN \ ARRAY_SIZE(mlxsw_sp_port_hw_rfc_3635_stats) +static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_ext_stats[] = { + { + .str = "ecn_marked", + .getter = mlxsw_reg_ppcnt_ecn_marked_get, + }, +}; + +#define MLXSW_SP_PORT_HW_EXT_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_hw_ext_stats) + static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_discard_stats[] = { { .str = "discard_ingress_general", @@ -2352,6 +2349,7 @@ static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_tc_stats[] = { MLXSW_SP_PORT_HW_RFC_2863_STATS_LEN + \ MLXSW_SP_PORT_HW_RFC_2819_STATS_LEN + \ MLXSW_SP_PORT_HW_RFC_3635_STATS_LEN + \ + MLXSW_SP_PORT_HW_EXT_STATS_LEN + \ MLXSW_SP_PORT_HW_DISCARD_STATS_LEN + \ (MLXSW_SP_PORT_HW_PRIO_STATS_LEN * \ IEEE_8021QAZ_MAX_TCS) + \ @@ -2413,6 +2411,12 @@ static void mlxsw_sp_port_get_strings(struct net_device *dev, p += ETH_GSTRING_LEN; } + for (i = 0; i < MLXSW_SP_PORT_HW_EXT_STATS_LEN; i++) { + memcpy(p, mlxsw_sp_port_hw_ext_stats[i].str, + ETH_GSTRING_LEN); + p += ETH_GSTRING_LEN; + } + for (i = 0; i < MLXSW_SP_PORT_HW_DISCARD_STATS_LEN; i++) { memcpy(p, mlxsw_sp_port_hw_discard_stats[i].str, ETH_GSTRING_LEN); @@ -2474,6 +2478,10 @@ mlxsw_sp_get_hw_stats_by_group(struct mlxsw_sp_port_hw_stats **p_hw_stats, *p_hw_stats = mlxsw_sp_port_hw_rfc_3635_stats; *p_len = MLXSW_SP_PORT_HW_RFC_3635_STATS_LEN; break; + case MLXSW_REG_PPCNT_EXT_CNT: + *p_hw_stats = mlxsw_sp_port_hw_ext_stats; + *p_len = MLXSW_SP_PORT_HW_EXT_STATS_LEN; + break; case MLXSW_REG_PPCNT_DISCARD_CNT: *p_hw_stats = mlxsw_sp_port_hw_discard_stats; *p_len = MLXSW_SP_PORT_HW_DISCARD_STATS_LEN; @@ -2543,6 +2551,11 @@ static void mlxsw_sp_port_get_stats(struct net_device *dev, data, data_index); data_index += MLXSW_SP_PORT_HW_RFC_3635_STATS_LEN; + /* Extended Counters */ + __mlxsw_sp_port_get_stats(dev, MLXSW_REG_PPCNT_EXT_CNT, 0, + data, data_index); + data_index += MLXSW_SP_PORT_HW_EXT_STATS_LEN; + /* Discard Counters */ __mlxsw_sp_port_get_stats(dev, MLXSW_REG_PPCNT_DISCARD_CNT, 0, data, data_index); @@ -2788,27 +2801,6 @@ static u32 mlxsw_sp1_to_ptys_speed(struct mlxsw_sp *mlxsw_sp, u8 width, return ptys_proto; } -static u32 -mlxsw_sp1_to_ptys_upper_speed(struct mlxsw_sp *mlxsw_sp, u32 upper_speed) -{ - u32 ptys_proto = 0; - int i; - - for (i = 0; i < MLXSW_SP1_PORT_LINK_MODE_LEN; i++) { - if (mlxsw_sp1_port_link_mode[i].speed <= upper_speed) - ptys_proto |= mlxsw_sp1_port_link_mode[i].mask; - } - return ptys_proto; -} - -static int -mlxsw_sp1_port_speed_base(struct mlxsw_sp *mlxsw_sp, u8 local_port, - u32 *base_speed) -{ - *base_speed = MLXSW_SP_PORT_BASE_SPEED_25G; - return 0; -} - static void mlxsw_sp1_reg_ptys_eth_pack(struct mlxsw_sp *mlxsw_sp, char *payload, u8 local_port, u32 proto_admin, bool autoneg) @@ -2833,8 +2825,6 @@ mlxsw_sp1_port_type_speed_ops = { .from_ptys_speed_duplex = mlxsw_sp1_from_ptys_speed_duplex, .to_ptys_advert_link = mlxsw_sp1_to_ptys_advert_link, .to_ptys_speed = mlxsw_sp1_to_ptys_speed, - .to_ptys_upper_speed = mlxsw_sp1_to_ptys_upper_speed, - .port_speed_base = mlxsw_sp1_port_speed_base, .reg_ptys_eth_pack = mlxsw_sp1_reg_ptys_eth_pack, .reg_ptys_eth_unpack = mlxsw_sp1_reg_ptys_eth_unpack, }; @@ -3235,51 +3225,6 @@ static u32 mlxsw_sp2_to_ptys_speed(struct mlxsw_sp *mlxsw_sp, return ptys_proto; } -static u32 -mlxsw_sp2_to_ptys_upper_speed(struct mlxsw_sp *mlxsw_sp, u32 upper_speed) -{ - u32 ptys_proto = 0; - int i; - - for (i = 0; i < MLXSW_SP2_PORT_LINK_MODE_LEN; i++) { - if (mlxsw_sp2_port_link_mode[i].speed <= upper_speed) - ptys_proto |= mlxsw_sp2_port_link_mode[i].mask; - } - return ptys_proto; -} - -static int -mlxsw_sp2_port_speed_base(struct mlxsw_sp *mlxsw_sp, u8 local_port, - u32 *base_speed) -{ - char ptys_pl[MLXSW_REG_PTYS_LEN]; - u32 eth_proto_cap; - int err; - - /* In Spectrum-2, the speed of 1x can change from port to port, so query - * it from firmware. - */ - mlxsw_reg_ptys_ext_eth_pack(ptys_pl, local_port, 0, false); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); - if (err) - return err; - mlxsw_reg_ptys_ext_eth_unpack(ptys_pl, ð_proto_cap, NULL, NULL); - - if (eth_proto_cap & - MLXSW_REG_PTYS_EXT_ETH_SPEED_50GAUI_1_LAUI_1_50GBASE_CR_KR) { - *base_speed = MLXSW_SP_PORT_BASE_SPEED_50G; - return 0; - } - - if (eth_proto_cap & - MLXSW_REG_PTYS_EXT_ETH_SPEED_25GAUI_1_25GBASE_CR_KR) { - *base_speed = MLXSW_SP_PORT_BASE_SPEED_25G; - return 0; - } - - return -EIO; -} - static void mlxsw_sp2_reg_ptys_eth_pack(struct mlxsw_sp *mlxsw_sp, char *payload, u8 local_port, u32 proto_admin, @@ -3305,8 +3250,6 @@ mlxsw_sp2_port_type_speed_ops = { .from_ptys_speed_duplex = mlxsw_sp2_from_ptys_speed_duplex, .to_ptys_advert_link = mlxsw_sp2_to_ptys_advert_link, .to_ptys_speed = mlxsw_sp2_to_ptys_speed, - .to_ptys_upper_speed = mlxsw_sp2_to_ptys_upper_speed, - .port_speed_base = mlxsw_sp2_port_speed_base, .reg_ptys_eth_pack = mlxsw_sp2_reg_ptys_eth_pack, .reg_ptys_eth_unpack = mlxsw_sp2_reg_ptys_eth_unpack, }; @@ -3520,24 +3463,24 @@ static int mlxsw_sp_port_speed_by_width_set(struct mlxsw_sp_port *mlxsw_sp_port) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + u32 eth_proto_cap, eth_proto_admin, eth_proto_oper; const struct mlxsw_sp_port_type_speed_ops *ops; char ptys_pl[MLXSW_REG_PTYS_LEN]; - u32 eth_proto_admin; - u32 upper_speed; - u32 base_speed; int err; ops = mlxsw_sp->port_type_speed_ops; - err = ops->port_speed_base(mlxsw_sp, mlxsw_sp_port->local_port, - &base_speed); + /* Set advertised speeds to supported speeds. */ + ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl, mlxsw_sp_port->local_port, + 0, false); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); if (err) return err; - upper_speed = base_speed * mlxsw_sp_port->mapping.width; - eth_proto_admin = ops->to_ptys_upper_speed(mlxsw_sp, upper_speed); + ops->reg_ptys_eth_unpack(mlxsw_sp, ptys_pl, ð_proto_cap, + ð_proto_admin, ð_proto_oper); ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl, mlxsw_sp_port->local_port, - eth_proto_admin, mlxsw_sp_port->link.autoneg); + eth_proto_cap, mlxsw_sp_port->link.autoneg); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); } @@ -4614,6 +4557,7 @@ static const struct mlxsw_listener mlxsw_sp1_listener[] = { static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core) { + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); char qpcr_pl[MLXSW_REG_QPCR_LEN]; enum mlxsw_reg_qpcr_ir_units ir_units; int max_cpu_policers; @@ -4636,7 +4580,6 @@ static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core) case MLXSW_REG_HTGT_TRAP_GROUP_SP_LLDP: case MLXSW_REG_HTGT_TRAP_GROUP_SP_OSPF: case MLXSW_REG_HTGT_TRAP_GROUP_SP_PIM: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_RPF: case MLXSW_REG_HTGT_TRAP_GROUP_SP_LBERROR: rate = 128; burst_size = 7; @@ -4649,7 +4592,6 @@ static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core) case MLXSW_REG_HTGT_TRAP_GROUP_SP_BGP: case MLXSW_REG_HTGT_TRAP_GROUP_SP_ARP: case MLXSW_REG_HTGT_TRAP_GROUP_SP_DHCP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_HOST_MISS: case MLXSW_REG_HTGT_TRAP_GROUP_SP_ROUTER_EXP: case MLXSW_REG_HTGT_TRAP_GROUP_SP_REMOTE_ROUTE: case MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_ND: @@ -4677,6 +4619,7 @@ static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core) continue; } + __set_bit(i, mlxsw_sp->trap->policers_usage); mlxsw_reg_qpcr_pack(qpcr_pl, i, ir_units, is_bytes, rate, burst_size); err = mlxsw_reg_write(mlxsw_core, MLXSW_REG(qpcr), qpcr_pl); @@ -4729,19 +4672,20 @@ static int mlxsw_sp_trap_groups_set(struct mlxsw_core *mlxsw_core) break; case MLXSW_REG_HTGT_TRAP_GROUP_SP_ARP: case MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_ND: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_RPF: case MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP1: priority = 2; tc = 2; break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_HOST_MISS: case MLXSW_REG_HTGT_TRAP_GROUP_SP_ROUTER_EXP: case MLXSW_REG_HTGT_TRAP_GROUP_SP_REMOTE_ROUTE: case MLXSW_REG_HTGT_TRAP_GROUP_SP_MULTICAST: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_LBERROR: priority = 1; tc = 1; break; + case MLXSW_REG_HTGT_TRAP_GROUP_SP_LBERROR: + priority = 0; + tc = 1; + break; case MLXSW_REG_HTGT_TRAP_GROUP_SP_EVENT: priority = MLXSW_REG_HTGT_DEFAULT_PRIORITY; tc = MLXSW_REG_HTGT_DEFAULT_TC; @@ -4805,20 +4749,32 @@ static void mlxsw_sp_traps_unregister(struct mlxsw_sp *mlxsw_sp, static int mlxsw_sp_traps_init(struct mlxsw_sp *mlxsw_sp) { + struct mlxsw_sp_trap *trap; + u64 max_policers; int err; + if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, MAX_CPU_POLICERS)) + return -EIO; + max_policers = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_CPU_POLICERS); + trap = kzalloc(struct_size(trap, policers_usage, + BITS_TO_LONGS(max_policers)), GFP_KERNEL); + if (!trap) + return -ENOMEM; + trap->max_policers = max_policers; + mlxsw_sp->trap = trap; + err = mlxsw_sp_cpu_policers_set(mlxsw_sp->core); if (err) - return err; + goto err_cpu_policers_set; err = mlxsw_sp_trap_groups_set(mlxsw_sp->core); if (err) - return err; + goto err_trap_groups_set; err = mlxsw_sp_traps_register(mlxsw_sp, mlxsw_sp_listener, ARRAY_SIZE(mlxsw_sp_listener)); if (err) - return err; + goto err_traps_register; err = mlxsw_sp_traps_register(mlxsw_sp, mlxsw_sp->listeners, mlxsw_sp->listeners_count); @@ -4830,6 +4786,10 @@ static int mlxsw_sp_traps_init(struct mlxsw_sp *mlxsw_sp) err_extra_traps_init: mlxsw_sp_traps_unregister(mlxsw_sp, mlxsw_sp_listener, ARRAY_SIZE(mlxsw_sp_listener)); +err_traps_register: +err_trap_groups_set: +err_cpu_policers_set: + kfree(trap); return err; } @@ -4839,6 +4799,7 @@ static void mlxsw_sp_traps_fini(struct mlxsw_sp *mlxsw_sp) mlxsw_sp->listeners_count); mlxsw_sp_traps_unregister(mlxsw_sp, mlxsw_sp_listener, ARRAY_SIZE(mlxsw_sp_listener)); + kfree(mlxsw_sp->trap); } #define MLXSW_SP_LAG_SEED_INIT 0xcafecafe @@ -4935,16 +4896,35 @@ static const struct mlxsw_sp_span_ops mlxsw_sp1_span_ops = { }; #define MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR 38 +#define MLXSW_SP3_SPAN_EG_MIRROR_BUFFER_FACTOR 50 + +static u32 __mlxsw_sp_span_buffsize_get(int mtu, u32 speed, u32 buffer_factor) +{ + return 3 * mtu + buffer_factor * speed / 1000; +} static u32 mlxsw_sp2_span_buffsize_get(int mtu, u32 speed) { - return 3 * mtu + MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR * speed / 1000; + int factor = MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR; + + return __mlxsw_sp_span_buffsize_get(mtu, speed, factor); } static const struct mlxsw_sp_span_ops mlxsw_sp2_span_ops = { .buffsize_get = mlxsw_sp2_span_buffsize_get, }; +static u32 mlxsw_sp3_span_buffsize_get(int mtu, u32 speed) +{ + int factor = MLXSW_SP3_SPAN_EG_MIRROR_BUFFER_FACTOR; + + return __mlxsw_sp_span_buffsize_get(mtu, speed, factor); +} + +static const struct mlxsw_sp_span_ops mlxsw_sp3_span_ops = { + .buffsize_get = mlxsw_sp3_span_buffsize_get, +}; + u32 mlxsw_sp_span_buffsize_get(struct mlxsw_sp *mlxsw_sp, int mtu, u32 speed) { u32 buffsize = mlxsw_sp->span_ops->buffsize_get(speed, mtu); @@ -5223,7 +5203,7 @@ static int mlxsw_sp3_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->sb_vals = &mlxsw_sp2_sb_vals; mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops; mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops; - mlxsw_sp->span_ops = &mlxsw_sp2_span_ops; + mlxsw_sp->span_ops = &mlxsw_sp3_span_ops; mlxsw_sp->lowest_shaper_bs = MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP3; return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack); @@ -5460,8 +5440,13 @@ static int mlxsw_sp1_resources_register(struct mlxsw_core *mlxsw_core) if (err) goto err_resources_span_register; + err = mlxsw_sp_counter_resources_register(mlxsw_core); + if (err) + goto err_resources_counter_register; + return 0; +err_resources_counter_register: err_resources_span_register: devlink_resources_unregister(priv_to_devlink(mlxsw_core), NULL); return err; @@ -5479,8 +5464,13 @@ static int mlxsw_sp2_resources_register(struct mlxsw_core *mlxsw_core) if (err) goto err_resources_span_register; + err = mlxsw_sp_counter_resources_register(mlxsw_core); + if (err) + goto err_resources_counter_register; + return 0; +err_resources_counter_register: err_resources_span_register: devlink_resources_unregister(priv_to_devlink(mlxsw_core), NULL); return err; @@ -5684,6 +5674,11 @@ static struct mlxsw_driver mlxsw_sp1_driver = { .trap_fini = mlxsw_sp_trap_fini, .trap_action_set = mlxsw_sp_trap_action_set, .trap_group_init = mlxsw_sp_trap_group_init, + .trap_group_set = mlxsw_sp_trap_group_set, + .trap_policer_init = mlxsw_sp_trap_policer_init, + .trap_policer_fini = mlxsw_sp_trap_policer_fini, + .trap_policer_set = mlxsw_sp_trap_policer_set, + .trap_policer_counter_get = mlxsw_sp_trap_policer_counter_get, .txhdr_construct = mlxsw_sp_txhdr_construct, .resources_register = mlxsw_sp1_resources_register, .kvd_sizes_get = mlxsw_sp_kvd_sizes_get, @@ -5718,6 +5713,11 @@ static struct mlxsw_driver mlxsw_sp2_driver = { .trap_fini = mlxsw_sp_trap_fini, .trap_action_set = mlxsw_sp_trap_action_set, .trap_group_init = mlxsw_sp_trap_group_init, + .trap_group_set = mlxsw_sp_trap_group_set, + .trap_policer_init = mlxsw_sp_trap_policer_init, + .trap_policer_fini = mlxsw_sp_trap_policer_fini, + .trap_policer_set = mlxsw_sp_trap_policer_set, + .trap_policer_counter_get = mlxsw_sp_trap_policer_counter_get, .txhdr_construct = mlxsw_sp_txhdr_construct, .resources_register = mlxsw_sp2_resources_register, .params_register = mlxsw_sp2_params_register, @@ -5751,6 +5751,11 @@ static struct mlxsw_driver mlxsw_sp3_driver = { .trap_fini = mlxsw_sp_trap_fini, .trap_action_set = mlxsw_sp_trap_action_set, .trap_group_init = mlxsw_sp_trap_group_init, + .trap_group_set = mlxsw_sp_trap_group_set, + .trap_policer_init = mlxsw_sp_trap_policer_init, + .trap_policer_fini = mlxsw_sp_trap_policer_fini, + .trap_policer_set = mlxsw_sp_trap_policer_set, + .trap_policer_counter_get = mlxsw_sp_trap_policer_counter_get, .txhdr_construct = mlxsw_sp_txhdr_construct, .resources_register = mlxsw_sp2_resources_register, .params_register = mlxsw_sp2_params_register, @@ -6316,7 +6321,7 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *lower_dev, return -EINVAL; } if (netif_is_macvlan(upper_dev) && - !mlxsw_sp_rif_find_by_dev(mlxsw_sp, lower_dev)) { + !mlxsw_sp_rif_exists(mlxsw_sp, lower_dev)) { NL_SET_ERR_MSG_MOD(extack, "macvlan is only supported on top of router interfaces"); return -EOPNOTSUPP; } @@ -6472,7 +6477,7 @@ static int mlxsw_sp_netdevice_port_vlan_event(struct net_device *vlan_dev, return -EINVAL; } if (netif_is_macvlan(upper_dev) && - !mlxsw_sp_rif_find_by_dev(mlxsw_sp, vlan_dev)) { + !mlxsw_sp_rif_exists(mlxsw_sp, vlan_dev)) { NL_SET_ERR_MSG_MOD(extack, "macvlan is only supported on top of router interfaces"); return -EOPNOTSUPP; } @@ -6549,7 +6554,7 @@ static int mlxsw_sp_netdevice_bridge_vlan_event(struct net_device *vlan_dev, if (!info->linking) break; if (netif_is_macvlan(upper_dev) && - !mlxsw_sp_rif_find_by_dev(mlxsw_sp, vlan_dev)) { + !mlxsw_sp_rif_exists(mlxsw_sp, vlan_dev)) { NL_SET_ERR_MSG_MOD(extack, "macvlan is only supported on top of router interfaces"); return -EOPNOTSUPP; } @@ -6609,7 +6614,7 @@ static int mlxsw_sp_netdevice_bridge_event(struct net_device *br_dev, if (!info->linking) break; if (netif_is_macvlan(upper_dev) && - !mlxsw_sp_rif_find_by_dev(mlxsw_sp, br_dev)) { + !mlxsw_sp_rif_exists(mlxsw_sp, br_dev)) { NL_SET_ERR_MSG_MOD(extack, "macvlan is only supported on top of router interfaces"); return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index a0f1f9dceec5..ca56e72cb4b7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -19,6 +19,7 @@ #include <net/pkt_cls.h> #include <net/red.h> #include <net/vxlan.h> +#include <net/flow_offload.h> #include "port.h" #include "core.h" @@ -32,9 +33,6 @@ #define MLXSW_SP_MID_MAX 7000 -#define MLXSW_SP_PORT_BASE_SPEED_25G 25000 /* Mb/s */ -#define MLXSW_SP_PORT_BASE_SPEED_50G 50000 /* Mb/s */ - #define MLXSW_SP_KVD_LINEAR_SIZE 98304 /* entries */ #define MLXSW_SP_KVD_GRANULARITY 128 @@ -48,6 +46,10 @@ #define MLXSW_SP_RESOURCE_NAME_SPAN "span_agents" +#define MLXSW_SP_RESOURCE_NAME_COUNTERS "counters" +#define MLXSW_SP_RESOURCE_NAME_COUNTERS_FLOW "flow" +#define MLXSW_SP_RESOURCE_NAME_COUNTERS_RIF "rif" + enum mlxsw_sp_resource_id { MLXSW_SP_RESOURCE_KVD = 1, MLXSW_SP_RESOURCE_KVD_LINEAR, @@ -57,6 +59,9 @@ enum mlxsw_sp_resource_id { MLXSW_SP_RESOURCE_KVD_LINEAR_CHUNKS, MLXSW_SP_RESOURCE_KVD_LINEAR_LARGE_CHUNKS, MLXSW_SP_RESOURCE_SPAN, + MLXSW_SP_RESOURCE_COUNTERS, + MLXSW_SP_RESOURCE_COUNTERS_FLOW, + MLXSW_SP_RESOURCE_COUNTERS_RIF, }; struct mlxsw_sp_port; @@ -141,6 +146,7 @@ struct mlxsw_sp_port_type_speed_ops; struct mlxsw_sp_ptp_state; struct mlxsw_sp_ptp_ops; struct mlxsw_sp_span_ops; +struct mlxsw_sp_qdisc_state; struct mlxsw_sp_port_mapping { u8 module; @@ -168,12 +174,9 @@ struct mlxsw_sp { struct notifier_block netdevice_nb; struct mlxsw_sp_ptp_clock *clock; struct mlxsw_sp_ptp_state *ptp_state; - struct mlxsw_sp_counter_pool *counter_pool; - struct { - struct mlxsw_sp_span_entry *entries; - int entries_count; - } span; + struct mlxsw_sp_span *span; + struct mlxsw_sp_trap *trap; const struct mlxsw_fw_rev *req_rev; const char *fw_filename; const struct mlxsw_sp_kvdl_ops *kvdl_ops; @@ -282,8 +285,7 @@ struct mlxsw_sp_port { struct mlxsw_sp_port_sample *sample; struct list_head vlans_list; struct mlxsw_sp_port_vlan *default_vlan; - struct mlxsw_sp_qdisc *root_qdisc; - struct mlxsw_sp_qdisc *tclass_qdiscs; + struct mlxsw_sp_qdisc_state *qdisc; unsigned acl_rule_count; struct mlxsw_sp_acl_block *ing_acl_block; struct mlxsw_sp_acl_block *eg_acl_block; @@ -313,9 +315,6 @@ struct mlxsw_sp_port_type_speed_ops { u32 (*to_ptys_advert_link)(struct mlxsw_sp *mlxsw_sp, u8 width, const struct ethtool_link_ksettings *cmd); u32 (*to_ptys_speed)(struct mlxsw_sp *mlxsw_sp, u8 width, u32 speed); - u32 (*to_ptys_upper_speed)(struct mlxsw_sp *mlxsw_sp, u32 upper_speed); - int (*port_speed_base)(struct mlxsw_sp *mlxsw_sp, u8 local_port, - u32 *base_speed); void (*reg_ptys_eth_pack)(struct mlxsw_sp *mlxsw_sp, char *payload, u8 local_port, u32 proto_admin, bool autoneg); void (*reg_ptys_eth_unpack)(struct mlxsw_sp *mlxsw_sp, char *payload, @@ -468,10 +467,6 @@ int mlxsw_sp_bridge_vxlan_join(struct mlxsw_sp *mlxsw_sp, struct netlink_ext_ack *extack); void mlxsw_sp_bridge_vxlan_leave(struct mlxsw_sp *mlxsw_sp, const struct net_device *vxlan_dev); -struct mlxsw_sp_fid *mlxsw_sp_bridge_fid_get(struct mlxsw_sp *mlxsw_sp, - const struct net_device *br_dev, - u16 vid, - struct netlink_ext_ack *extack); extern struct notifier_block mlxsw_sp_switchdev_notifier; /* spectrum.c */ @@ -556,7 +551,7 @@ int mlxsw_sp_netdevice_vrf_event(struct net_device *l3_dev, unsigned long event, struct netdev_notifier_changeupper_info *info); bool mlxsw_sp_netdev_is_ipip_ol(const struct mlxsw_sp *mlxsw_sp, const struct net_device *dev); -bool mlxsw_sp_netdev_is_ipip_ul(const struct mlxsw_sp *mlxsw_sp, +bool mlxsw_sp_netdev_is_ipip_ul(struct mlxsw_sp *mlxsw_sp, const struct net_device *dev); int mlxsw_sp_netdevice_ipip_ol_event(struct mlxsw_sp *mlxsw_sp, struct net_device *l3_dev, @@ -571,10 +566,10 @@ void mlxsw_sp_port_vlan_router_leave(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan); void mlxsw_sp_rif_destroy_by_dev(struct mlxsw_sp *mlxsw_sp, struct net_device *dev); -struct mlxsw_sp_rif *mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp, - const struct net_device *dev); +bool mlxsw_sp_rif_exists(struct mlxsw_sp *mlxsw_sp, + const struct net_device *dev); +u16 mlxsw_sp_rif_vid(struct mlxsw_sp *mlxsw_sp, const struct net_device *dev); u8 mlxsw_sp_router_port(const struct mlxsw_sp *mlxsw_sp); -struct mlxsw_sp_fid *mlxsw_sp_rif_fid(const struct mlxsw_sp_rif *rif); int mlxsw_sp_router_nve_promote_decap(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id, enum mlxsw_sp_l3proto ul_proto, const union mlxsw_sp_l3addr *ul_sip, @@ -653,7 +648,9 @@ struct mlxsw_sp_acl_rule_info { struct mlxsw_afk_element_values values; struct mlxsw_afa_block *act_block; u8 action_created:1, - egress_bind_blocker:1; + ingress_bind_blocker:1, + egress_bind_blocker:1, + counter_valid:1; unsigned int counter_index; }; @@ -672,16 +669,20 @@ struct mlxsw_sp_acl_block { struct mlxsw_sp *mlxsw_sp; unsigned int rule_count; unsigned int disable_count; + unsigned int ingress_blocker_rule_count; unsigned int egress_blocker_rule_count; + unsigned int ingress_binding_count; + unsigned int egress_binding_count; struct net *net; }; struct mlxsw_afk *mlxsw_sp_acl_afk(struct mlxsw_sp_acl *acl); struct mlxsw_sp *mlxsw_sp_acl_block_mlxsw_sp(struct mlxsw_sp_acl_block *block); -unsigned int mlxsw_sp_acl_block_rule_count(struct mlxsw_sp_acl_block *block); +unsigned int +mlxsw_sp_acl_block_rule_count(const struct mlxsw_sp_acl_block *block); void mlxsw_sp_acl_block_disable_inc(struct mlxsw_sp_acl_block *block); void mlxsw_sp_acl_block_disable_dec(struct mlxsw_sp_acl_block *block); -bool mlxsw_sp_acl_block_disabled(struct mlxsw_sp_acl_block *block); +bool mlxsw_sp_acl_block_disabled(const struct mlxsw_sp_acl_block *block); struct mlxsw_sp_acl_block *mlxsw_sp_acl_block_create(struct mlxsw_sp *mlxsw_sp, struct net *net); void mlxsw_sp_acl_block_destroy(struct mlxsw_sp_acl_block *block); @@ -694,7 +695,9 @@ int mlxsw_sp_acl_block_unbind(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_block *block, struct mlxsw_sp_port *mlxsw_sp_port, bool ingress); -bool mlxsw_sp_acl_block_is_egress_bound(struct mlxsw_sp_acl_block *block); +bool mlxsw_sp_acl_block_is_egress_bound(const struct mlxsw_sp_acl_block *block); +bool mlxsw_sp_acl_block_is_ingress_bound(const struct mlxsw_sp_acl_block *block); +bool mlxsw_sp_acl_block_is_mixed_bound(const struct mlxsw_sp_acl_block *block); struct mlxsw_sp_acl_ruleset * mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_block *block, u32 chain_index, @@ -726,7 +729,10 @@ int mlxsw_sp_acl_rulei_act_continue(struct mlxsw_sp_acl_rule_info *rulei); int mlxsw_sp_acl_rulei_act_jump(struct mlxsw_sp_acl_rule_info *rulei, u16 group_id); int mlxsw_sp_acl_rulei_act_terminate(struct mlxsw_sp_acl_rule_info *rulei); -int mlxsw_sp_acl_rulei_act_drop(struct mlxsw_sp_acl_rule_info *rulei); +int mlxsw_sp_acl_rulei_act_drop(struct mlxsw_sp_acl_rule_info *rulei, + bool ingress, + const struct flow_action_cookie *fa_cookie, + struct netlink_ext_ack *extack); int mlxsw_sp_acl_rulei_act_trap(struct mlxsw_sp_acl_rule_info *rulei); int mlxsw_sp_acl_rulei_act_mirror(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule_info *rulei, @@ -741,6 +747,14 @@ int mlxsw_sp_acl_rulei_act_vlan(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule_info *rulei, u32 action, u16 vid, u16 proto, u8 prio, struct netlink_ext_ack *extack); +int mlxsw_sp_acl_rulei_act_priority(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_acl_rule_info *rulei, + u32 prio, struct netlink_ext_ack *extack); +int mlxsw_sp_acl_rulei_act_mangle(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_acl_rule_info *rulei, + enum flow_action_mangle_base htype, + u32 offset, u32 mask, u32 val, + struct netlink_ext_ack *extack); int mlxsw_sp_acl_rulei_act_count(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule_info *rulei, struct netlink_ext_ack *extack); @@ -773,10 +787,17 @@ struct mlxsw_sp_acl_rule_info * mlxsw_sp_acl_rule_rulei(struct mlxsw_sp_acl_rule *rule); int mlxsw_sp_acl_rule_get_stats(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule *rule, - u64 *packets, u64 *bytes, u64 *last_use); + u64 *packets, u64 *bytes, u64 *last_use, + enum flow_action_hw_stats *used_hw_stats); struct mlxsw_sp_fid *mlxsw_sp_acl_dummy_fid(struct mlxsw_sp *mlxsw_sp); +static inline const struct flow_action_cookie * +mlxsw_sp_acl_act_cookie_lookup(struct mlxsw_sp *mlxsw_sp, u32 cookie_index) +{ + return mlxsw_afa_cookie_lookup(mlxsw_sp->afa, cookie_index); +} + int mlxsw_sp_acl_init(struct mlxsw_sp *mlxsw_sp); void mlxsw_sp_acl_fini(struct mlxsw_sp *mlxsw_sp); u32 mlxsw_sp_acl_region_rehash_intrvl_get(struct mlxsw_sp *mlxsw_sp); @@ -864,6 +885,8 @@ int mlxsw_sp_setup_tc_ets(struct mlxsw_sp_port *mlxsw_sp_port, struct tc_ets_qopt_offload *p); int mlxsw_sp_setup_tc_tbf(struct mlxsw_sp_port *mlxsw_sp_port, struct tc_tbf_qopt_offload *p); +int mlxsw_sp_setup_tc_fifo(struct mlxsw_sp_port *mlxsw_sp_port, + struct tc_fifo_qopt_offload *p); /* spectrum_fid.c */ bool mlxsw_sp_fid_is_dummy(struct mlxsw_sp *mlxsw_sp, u16 fid_index); @@ -974,9 +997,6 @@ void mlxsw_sp_nve_flood_ip_del(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *fid, enum mlxsw_sp_l3proto proto, union mlxsw_sp_l3addr *addr); -u32 mlxsw_sp_nve_decap_tunnel_index_get(const struct mlxsw_sp *mlxsw_sp); -bool mlxsw_sp_nve_ipv4_route_is_decap(const struct mlxsw_sp *mlxsw_sp, - u32 tb_id, __be32 addr); int mlxsw_sp_nve_fid_enable(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *fid, struct mlxsw_sp_nve_params *params, struct netlink_ext_ack *extack); @@ -1003,6 +1023,22 @@ int mlxsw_sp_trap_action_set(struct mlxsw_core *mlxsw_core, enum devlink_trap_action action); int mlxsw_sp_trap_group_init(struct mlxsw_core *mlxsw_core, const struct devlink_trap_group *group); +int mlxsw_sp_trap_group_set(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_group *group, + const struct devlink_trap_policer *policer); +int +mlxsw_sp_trap_policer_init(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer); +void mlxsw_sp_trap_policer_fini(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer); +int +mlxsw_sp_trap_policer_set(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer, + u64 rate, u64 burst, struct netlink_ext_ack *extack); +int +mlxsw_sp_trap_policer_counter_get(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer, + u64 *p_drops); static inline struct net *mlxsw_sp_net(struct mlxsw_sp *mlxsw_sp) { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum1_kvdl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum1_kvdl.c index 09ee0a807747..a9fff8adc75e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum1_kvdl.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum1_kvdl.c @@ -60,7 +60,7 @@ static const struct mlxsw_sp1_kvdl_part_info mlxsw_sp1_kvdl_parts_info[] = { struct mlxsw_sp1_kvdl_part { struct mlxsw_sp1_kvdl_part_info info; - unsigned long usage[0]; /* Entries */ + unsigned long usage[]; /* Entries */ }; struct mlxsw_sp1_kvdl { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum2_kvdl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum2_kvdl.c index 8d14770766b4..3a73d654017f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum2_kvdl.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum2_kvdl.c @@ -45,7 +45,7 @@ struct mlxsw_sp2_kvdl_part { unsigned int usage_bit_count; unsigned int indexes_per_usage_bit; unsigned int last_allocated_bit; - unsigned long usage[0]; /* Usage bits */ + unsigned long usage[]; /* Usage bits */ }; struct mlxsw_sp2_kvdl { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c index 3d3cca596116..67ee880a8727 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c @@ -58,7 +58,7 @@ struct mlxsw_sp_acl_ruleset { struct mlxsw_sp_acl_ruleset_ht_key ht_key; struct rhashtable rule_ht; unsigned int ref_count; - unsigned long priv[0]; + unsigned long priv[]; /* priv has to be always the last item */ }; @@ -71,7 +71,7 @@ struct mlxsw_sp_acl_rule { u64 last_used; u64 last_packets; u64 last_bytes; - unsigned long priv[0]; + unsigned long priv[]; /* priv has to be always the last item */ }; @@ -99,7 +99,8 @@ struct mlxsw_sp *mlxsw_sp_acl_block_mlxsw_sp(struct mlxsw_sp_acl_block *block) return block->mlxsw_sp; } -unsigned int mlxsw_sp_acl_block_rule_count(struct mlxsw_sp_acl_block *block) +unsigned int +mlxsw_sp_acl_block_rule_count(const struct mlxsw_sp_acl_block *block) { return block ? block->rule_count : 0; } @@ -116,20 +117,24 @@ void mlxsw_sp_acl_block_disable_dec(struct mlxsw_sp_acl_block *block) block->disable_count--; } -bool mlxsw_sp_acl_block_disabled(struct mlxsw_sp_acl_block *block) +bool mlxsw_sp_acl_block_disabled(const struct mlxsw_sp_acl_block *block) { return block->disable_count; } -bool mlxsw_sp_acl_block_is_egress_bound(struct mlxsw_sp_acl_block *block) +bool mlxsw_sp_acl_block_is_egress_bound(const struct mlxsw_sp_acl_block *block) { - struct mlxsw_sp_acl_block_binding *binding; + return block->egress_binding_count; +} - list_for_each_entry(binding, &block->binding_list, list) { - if (!binding->ingress) - return true; - } - return false; +bool mlxsw_sp_acl_block_is_ingress_bound(const struct mlxsw_sp_acl_block *block) +{ + return block->ingress_binding_count; +} + +bool mlxsw_sp_acl_block_is_mixed_bound(const struct mlxsw_sp_acl_block *block) +{ + return block->ingress_binding_count && block->egress_binding_count; } static bool @@ -163,7 +168,8 @@ mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp, binding->mlxsw_sp_port, binding->ingress); } -static bool mlxsw_sp_acl_ruleset_block_bound(struct mlxsw_sp_acl_block *block) +static bool +mlxsw_sp_acl_ruleset_block_bound(const struct mlxsw_sp_acl_block *block) { return block->ruleset_zero; } @@ -250,6 +256,11 @@ int mlxsw_sp_acl_block_bind(struct mlxsw_sp *mlxsw_sp, if (WARN_ON(mlxsw_sp_acl_block_lookup(block, mlxsw_sp_port, ingress))) return -EEXIST; + if (ingress && block->ingress_blocker_rule_count) { + NL_SET_ERR_MSG_MOD(extack, "Block cannot be bound to ingress because it contains unsupported rules"); + return -EOPNOTSUPP; + } + if (!ingress && block->egress_blocker_rule_count) { NL_SET_ERR_MSG_MOD(extack, "Block cannot be bound to egress because it contains unsupported rules"); return -EOPNOTSUPP; @@ -267,6 +278,10 @@ int mlxsw_sp_acl_block_bind(struct mlxsw_sp *mlxsw_sp, goto err_ruleset_bind; } + if (ingress) + block->ingress_binding_count++; + else + block->egress_binding_count++; list_add(&binding->list, &block->binding_list); return 0; @@ -288,6 +303,11 @@ int mlxsw_sp_acl_block_unbind(struct mlxsw_sp *mlxsw_sp, list_del(&binding->list); + if (ingress) + block->ingress_binding_count--; + else + block->egress_binding_count--; + if (mlxsw_sp_acl_ruleset_block_bound(block)) mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, block, binding); @@ -515,9 +535,13 @@ int mlxsw_sp_acl_rulei_act_terminate(struct mlxsw_sp_acl_rule_info *rulei) return mlxsw_afa_block_terminate(rulei->act_block); } -int mlxsw_sp_acl_rulei_act_drop(struct mlxsw_sp_acl_rule_info *rulei) +int mlxsw_sp_acl_rulei_act_drop(struct mlxsw_sp_acl_rule_info *rulei, + bool ingress, + const struct flow_action_cookie *fa_cookie, + struct netlink_ext_ack *extack) { - return mlxsw_afa_block_append_drop(rulei->act_block); + return mlxsw_afa_block_append_drop(rulei->act_block, ingress, + fa_cookie, extack); } int mlxsw_sp_acl_rulei_act_trap(struct mlxsw_sp_acl_rule_info *rulei) @@ -614,12 +638,126 @@ int mlxsw_sp_acl_rulei_act_vlan(struct mlxsw_sp *mlxsw_sp, } } +int mlxsw_sp_acl_rulei_act_priority(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_acl_rule_info *rulei, + u32 prio, struct netlink_ext_ack *extack) +{ + /* Even though both Linux and Spectrum switches support 16 priorities, + * spectrum_qdisc only processes the first eight priomap elements, and + * the DCB and PFC features are tied to 8 priorities as well. Therefore + * bounce attempts to prioritize packets to higher priorities. + */ + if (prio >= IEEE_8021QAZ_MAX_TCS) { + NL_SET_ERR_MSG_MOD(extack, "Only priorities 0..7 are supported"); + return -EINVAL; + } + return mlxsw_afa_block_append_qos_switch_prio(rulei->act_block, prio, + extack); +} + +enum mlxsw_sp_acl_mangle_field { + MLXSW_SP_ACL_MANGLE_FIELD_IP_DSFIELD, + MLXSW_SP_ACL_MANGLE_FIELD_IP_DSCP, + MLXSW_SP_ACL_MANGLE_FIELD_IP_ECN, +}; + +struct mlxsw_sp_acl_mangle_action { + enum flow_action_mangle_base htype; + /* Offset is u32-aligned. */ + u32 offset; + /* Mask bits are unset for the modified field. */ + u32 mask; + /* Shift required to extract the set value. */ + u32 shift; + enum mlxsw_sp_acl_mangle_field field; +}; + +#define MLXSW_SP_ACL_MANGLE_ACTION(_htype, _offset, _mask, _shift, _field) \ + { \ + .htype = _htype, \ + .offset = _offset, \ + .mask = _mask, \ + .shift = _shift, \ + .field = MLXSW_SP_ACL_MANGLE_FIELD_##_field, \ + } + +#define MLXSW_SP_ACL_MANGLE_ACTION_IP4(_offset, _mask, _shift, _field) \ + MLXSW_SP_ACL_MANGLE_ACTION(FLOW_ACT_MANGLE_HDR_TYPE_IP4, \ + _offset, _mask, _shift, _field) + +#define MLXSW_SP_ACL_MANGLE_ACTION_IP6(_offset, _mask, _shift, _field) \ + MLXSW_SP_ACL_MANGLE_ACTION(FLOW_ACT_MANGLE_HDR_TYPE_IP6, \ + _offset, _mask, _shift, _field) + +static struct mlxsw_sp_acl_mangle_action mlxsw_sp_acl_mangle_actions[] = { + MLXSW_SP_ACL_MANGLE_ACTION_IP4(0, 0xff00ffff, 16, IP_DSFIELD), + MLXSW_SP_ACL_MANGLE_ACTION_IP4(0, 0xff03ffff, 18, IP_DSCP), + MLXSW_SP_ACL_MANGLE_ACTION_IP4(0, 0xfffcffff, 16, IP_ECN), + MLXSW_SP_ACL_MANGLE_ACTION_IP6(0, 0xf00fffff, 20, IP_DSFIELD), + MLXSW_SP_ACL_MANGLE_ACTION_IP6(0, 0xf03fffff, 22, IP_DSCP), + MLXSW_SP_ACL_MANGLE_ACTION_IP6(0, 0xffcfffff, 20, IP_ECN), +}; + +static int +mlxsw_sp_acl_rulei_act_mangle_field(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_acl_rule_info *rulei, + struct mlxsw_sp_acl_mangle_action *mact, + u32 val, struct netlink_ext_ack *extack) +{ + switch (mact->field) { + case MLXSW_SP_ACL_MANGLE_FIELD_IP_DSFIELD: + return mlxsw_afa_block_append_qos_dsfield(rulei->act_block, + val, extack); + case MLXSW_SP_ACL_MANGLE_FIELD_IP_DSCP: + return mlxsw_afa_block_append_qos_dscp(rulei->act_block, + val, extack); + case MLXSW_SP_ACL_MANGLE_FIELD_IP_ECN: + return mlxsw_afa_block_append_qos_ecn(rulei->act_block, + val, extack); + } + + /* We shouldn't have gotten a match in the first place! */ + WARN_ONCE(1, "Unhandled mangle field"); + return -EINVAL; +} + +int mlxsw_sp_acl_rulei_act_mangle(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_acl_rule_info *rulei, + enum flow_action_mangle_base htype, + u32 offset, u32 mask, u32 val, + struct netlink_ext_ack *extack) +{ + struct mlxsw_sp_acl_mangle_action *mact; + size_t i; + + for (i = 0; i < ARRAY_SIZE(mlxsw_sp_acl_mangle_actions); ++i) { + mact = &mlxsw_sp_acl_mangle_actions[i]; + if (mact->htype == htype && + mact->offset == offset && + mact->mask == mask) { + val >>= mact->shift; + return mlxsw_sp_acl_rulei_act_mangle_field(mlxsw_sp, + rulei, mact, + val, extack); + } + } + + NL_SET_ERR_MSG_MOD(extack, "Unsupported mangle field"); + return -EINVAL; +} + int mlxsw_sp_acl_rulei_act_count(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule_info *rulei, struct netlink_ext_ack *extack) { - return mlxsw_afa_block_append_counter(rulei->act_block, - &rulei->counter_index, extack); + int err; + + err = mlxsw_afa_block_append_counter(rulei->act_block, + &rulei->counter_index, extack); + if (err) + return err; + rulei->counter_valid = true; + return 0; } int mlxsw_sp_acl_rulei_act_fid_set(struct mlxsw_sp *mlxsw_sp, @@ -707,6 +845,7 @@ int mlxsw_sp_acl_rule_add(struct mlxsw_sp *mlxsw_sp, list_add_tail(&rule->list, &mlxsw_sp->acl->rules); mutex_unlock(&mlxsw_sp->acl->rules_lock); block->rule_count++; + block->ingress_blocker_rule_count += rule->rulei->ingress_bind_blocker; block->egress_blocker_rule_count += rule->rulei->egress_bind_blocker; return 0; @@ -726,6 +865,7 @@ void mlxsw_sp_acl_rule_del(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_block *block = ruleset->ht_key.block; block->egress_blocker_rule_count -= rule->rulei->egress_bind_blocker; + block->ingress_blocker_rule_count -= rule->rulei->ingress_bind_blocker; ruleset->ht_key.block->rule_count--; mutex_lock(&mlxsw_sp->acl->rules_lock); list_del(&rule->list); @@ -827,20 +967,24 @@ static void mlxsw_sp_acl_rule_activity_update_work(struct work_struct *work) int mlxsw_sp_acl_rule_get_stats(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule *rule, - u64 *packets, u64 *bytes, u64 *last_use) + u64 *packets, u64 *bytes, u64 *last_use, + enum flow_action_hw_stats *used_hw_stats) { struct mlxsw_sp_acl_rule_info *rulei; - u64 current_packets; - u64 current_bytes; + u64 current_packets = 0; + u64 current_bytes = 0; int err; rulei = mlxsw_sp_acl_rule_rulei(rule); - err = mlxsw_sp_flow_counter_get(mlxsw_sp, rulei->counter_index, - ¤t_packets, ¤t_bytes); - if (err) - return err; - + if (rulei->counter_valid) { + err = mlxsw_sp_flow_counter_get(mlxsw_sp, rulei->counter_index, + ¤t_packets, + ¤t_bytes); + if (err) + return err; + *used_hw_stats = FLOW_ACTION_HW_STATS_IMMEDIATE; + } *packets = current_packets - rule->last_packets; *bytes = current_bytes - rule->last_bytes; *last_use = rule->last_used; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_bloom_filter.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_bloom_filter.c index 3a2de13fcb68..dbd3bebf11ec 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_bloom_filter.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_bloom_filter.c @@ -13,7 +13,7 @@ struct mlxsw_sp_acl_bf { struct mutex lock; /* Protects Bloom Filter updates. */ unsigned int bank_size; - refcount_t refcnt[0]; + refcount_t refcnt[]; }; /* Bloom filter uses a crc-16 hash over chunks of data which contain 4 key diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c index e993159e8e4c..430da69003d8 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c @@ -224,7 +224,7 @@ struct mlxsw_sp_acl_tcam_vchunk; struct mlxsw_sp_acl_tcam_chunk { struct mlxsw_sp_acl_tcam_vchunk *vchunk; struct mlxsw_sp_acl_tcam_region *region; - unsigned long priv[0]; + unsigned long priv[]; /* priv has to be always the last item */ }; @@ -243,7 +243,7 @@ struct mlxsw_sp_acl_tcam_vchunk { struct mlxsw_sp_acl_tcam_entry { struct mlxsw_sp_acl_tcam_ventry *ventry; struct mlxsw_sp_acl_tcam_chunk *chunk; - unsigned long priv[0]; + unsigned long priv[]; /* priv has to be always the last item */ }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h index 5965913565a5..96437992b102 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h @@ -20,7 +20,7 @@ struct mlxsw_sp_acl_tcam { struct mutex lock; /* guards vregion list */ struct list_head vregion_list; u32 vregion_rehash_intrvl; /* ms */ - unsigned long priv[0]; + unsigned long priv[]; /* priv has to be always the last item */ }; @@ -86,7 +86,7 @@ struct mlxsw_sp_acl_tcam_region { char tcam_region_info[MLXSW_REG_PXXX_TCAM_REGION_INFO_LEN]; struct mlxsw_afk_key_info *key_info; struct mlxsw_sp *mlxsw_sp; - unsigned long priv[0]; + unsigned long priv[]; /* priv has to be always the last item */ }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.c index 83c2e1e5f216..7974982533b5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.c @@ -3,92 +3,147 @@ #include <linux/kernel.h> #include <linux/bitops.h> +#include <linux/spinlock.h> #include "spectrum_cnt.h" -#define MLXSW_SP_COUNTER_POOL_BANK_SIZE 4096 - struct mlxsw_sp_counter_sub_pool { + u64 size; unsigned int base_index; - unsigned int size; + enum mlxsw_res_id entry_size_res_id; + const char *resource_name; /* devlink resource name */ + u64 resource_id; /* devlink resource id */ unsigned int entry_size; unsigned int bank_count; + atomic_t active_entries_count; }; struct mlxsw_sp_counter_pool { - unsigned int pool_size; + u64 pool_size; unsigned long *usage; /* Usage bitmap */ - struct mlxsw_sp_counter_sub_pool *sub_pools; + spinlock_t counter_pool_lock; /* Protects counter pool allocations */ + atomic_t active_entries_count; + unsigned int sub_pools_count; + struct mlxsw_sp_counter_sub_pool sub_pools[]; }; -static struct mlxsw_sp_counter_sub_pool mlxsw_sp_counter_sub_pools[] = { +static const struct mlxsw_sp_counter_sub_pool mlxsw_sp_counter_sub_pools[] = { [MLXSW_SP_COUNTER_SUB_POOL_FLOW] = { + .entry_size_res_id = MLXSW_RES_ID_COUNTER_SIZE_PACKETS_BYTES, + .resource_name = MLXSW_SP_RESOURCE_NAME_COUNTERS_FLOW, + .resource_id = MLXSW_SP_RESOURCE_COUNTERS_FLOW, .bank_count = 6, }, [MLXSW_SP_COUNTER_SUB_POOL_RIF] = { + .entry_size_res_id = MLXSW_RES_ID_COUNTER_SIZE_ROUTER_BASIC, + .resource_name = MLXSW_SP_RESOURCE_NAME_COUNTERS_RIF, + .resource_id = MLXSW_SP_RESOURCE_COUNTERS_RIF, .bank_count = 2, } }; -static int mlxsw_sp_counter_pool_validate(struct mlxsw_sp *mlxsw_sp) +static u64 mlxsw_sp_counter_sub_pool_occ_get(void *priv) +{ + const struct mlxsw_sp_counter_sub_pool *sub_pool = priv; + + return atomic_read(&sub_pool->active_entries_count); +} + +static int mlxsw_sp_counter_sub_pools_init(struct mlxsw_sp *mlxsw_sp) { - unsigned int total_bank_config = 0; - unsigned int pool_size; + struct mlxsw_sp_counter_pool *pool = mlxsw_sp->counter_pool; + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + struct mlxsw_sp_counter_sub_pool *sub_pool; + unsigned int base_index = 0; + enum mlxsw_res_id res_id; + int err; int i; - pool_size = MLXSW_CORE_RES_GET(mlxsw_sp->core, COUNTER_POOL_SIZE); - /* Check config is valid, no bank over subscription */ - for (i = 0; i < ARRAY_SIZE(mlxsw_sp_counter_sub_pools); i++) - total_bank_config += mlxsw_sp_counter_sub_pools[i].bank_count; - if (total_bank_config > pool_size / MLXSW_SP_COUNTER_POOL_BANK_SIZE + 1) - return -EINVAL; + for (i = 0; i < pool->sub_pools_count; i++) { + sub_pool = &pool->sub_pools[i]; + res_id = sub_pool->entry_size_res_id; + + if (!mlxsw_core_res_valid(mlxsw_sp->core, res_id)) + return -EIO; + sub_pool->entry_size = mlxsw_core_res_get(mlxsw_sp->core, + res_id); + err = devlink_resource_size_get(devlink, + sub_pool->resource_id, + &sub_pool->size); + if (err) + goto err_resource_size_get; + + devlink_resource_occ_get_register(devlink, + sub_pool->resource_id, + mlxsw_sp_counter_sub_pool_occ_get, + sub_pool); + + sub_pool->base_index = base_index; + base_index += sub_pool->size; + atomic_set(&sub_pool->active_entries_count, 0); + } return 0; + +err_resource_size_get: + for (i--; i >= 0; i--) { + sub_pool = &pool->sub_pools[i]; + + devlink_resource_occ_get_unregister(devlink, + sub_pool->resource_id); + } + return err; } -static int mlxsw_sp_counter_sub_pools_prepare(struct mlxsw_sp *mlxsw_sp) +static void mlxsw_sp_counter_sub_pools_fini(struct mlxsw_sp *mlxsw_sp) { + struct mlxsw_sp_counter_pool *pool = mlxsw_sp->counter_pool; + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); struct mlxsw_sp_counter_sub_pool *sub_pool; + int i; - /* Prepare generic flow pool*/ - sub_pool = &mlxsw_sp_counter_sub_pools[MLXSW_SP_COUNTER_SUB_POOL_FLOW]; - if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, COUNTER_SIZE_PACKETS_BYTES)) - return -EIO; - sub_pool->entry_size = MLXSW_CORE_RES_GET(mlxsw_sp->core, - COUNTER_SIZE_PACKETS_BYTES); - /* Prepare erif pool*/ - sub_pool = &mlxsw_sp_counter_sub_pools[MLXSW_SP_COUNTER_SUB_POOL_RIF]; - if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, COUNTER_SIZE_ROUTER_BASIC)) - return -EIO; - sub_pool->entry_size = MLXSW_CORE_RES_GET(mlxsw_sp->core, - COUNTER_SIZE_ROUTER_BASIC); - return 0; + for (i = 0; i < pool->sub_pools_count; i++) { + sub_pool = &pool->sub_pools[i]; + + WARN_ON(atomic_read(&sub_pool->active_entries_count)); + devlink_resource_occ_get_unregister(devlink, + sub_pool->resource_id); + } +} + +static u64 mlxsw_sp_counter_pool_occ_get(void *priv) +{ + const struct mlxsw_sp_counter_pool *pool = priv; + + return atomic_read(&pool->active_entries_count); } int mlxsw_sp_counter_pool_init(struct mlxsw_sp *mlxsw_sp) { + unsigned int sub_pools_count = ARRAY_SIZE(mlxsw_sp_counter_sub_pools); + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); struct mlxsw_sp_counter_sub_pool *sub_pool; struct mlxsw_sp_counter_pool *pool; - unsigned int base_index; unsigned int map_size; - int i; int err; - if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, COUNTER_POOL_SIZE)) - return -EIO; - - err = mlxsw_sp_counter_pool_validate(mlxsw_sp); - if (err) - return err; - - err = mlxsw_sp_counter_sub_pools_prepare(mlxsw_sp); - if (err) - return err; - - pool = kzalloc(sizeof(*pool), GFP_KERNEL); + pool = kzalloc(struct_size(pool, sub_pools, sub_pools_count), + GFP_KERNEL); if (!pool) return -ENOMEM; + mlxsw_sp->counter_pool = pool; + memcpy(pool->sub_pools, mlxsw_sp_counter_sub_pools, + sub_pools_count * sizeof(*sub_pool)); + pool->sub_pools_count = sub_pools_count; + spin_lock_init(&pool->counter_pool_lock); + atomic_set(&pool->active_entries_count, 0); + + err = devlink_resource_size_get(devlink, MLXSW_SP_RESOURCE_COUNTERS, + &pool->pool_size); + if (err) + goto err_pool_resource_size_get; + devlink_resource_occ_get_register(devlink, MLXSW_SP_RESOURCE_COUNTERS, + mlxsw_sp_counter_pool_occ_get, pool); - pool->pool_size = MLXSW_CORE_RES_GET(mlxsw_sp->core, COUNTER_POOL_SIZE); map_size = BITS_TO_LONGS(pool->pool_size) * sizeof(unsigned long); pool->usage = kzalloc(map_size, GFP_KERNEL); @@ -97,26 +152,18 @@ int mlxsw_sp_counter_pool_init(struct mlxsw_sp *mlxsw_sp) goto err_usage_alloc; } - pool->sub_pools = mlxsw_sp_counter_sub_pools; - /* Allocation is based on bank count which should be - * specified for each sub pool statically. - */ - base_index = 0; - for (i = 0; i < ARRAY_SIZE(mlxsw_sp_counter_sub_pools); i++) { - sub_pool = &pool->sub_pools[i]; - sub_pool->size = sub_pool->bank_count * - MLXSW_SP_COUNTER_POOL_BANK_SIZE; - sub_pool->base_index = base_index; - base_index += sub_pool->size; - /* The last bank can't be fully used */ - if (sub_pool->base_index + sub_pool->size > pool->pool_size) - sub_pool->size = pool->pool_size - sub_pool->base_index; - } + err = mlxsw_sp_counter_sub_pools_init(mlxsw_sp); + if (err) + goto err_sub_pools_init; - mlxsw_sp->counter_pool = pool; return 0; +err_sub_pools_init: + kfree(pool->usage); err_usage_alloc: + devlink_resource_occ_get_unregister(devlink, + MLXSW_SP_RESOURCE_COUNTERS); +err_pool_resource_size_get: kfree(pool); return err; } @@ -124,10 +171,15 @@ err_usage_alloc: void mlxsw_sp_counter_pool_fini(struct mlxsw_sp *mlxsw_sp) { struct mlxsw_sp_counter_pool *pool = mlxsw_sp->counter_pool; + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + mlxsw_sp_counter_sub_pools_fini(mlxsw_sp); WARN_ON(find_first_bit(pool->usage, pool->pool_size) != pool->pool_size); + WARN_ON(atomic_read(&pool->active_entries_count)); kfree(pool->usage); + devlink_resource_occ_get_unregister(devlink, + MLXSW_SP_RESOURCE_COUNTERS); kfree(pool); } @@ -139,25 +191,37 @@ int mlxsw_sp_counter_alloc(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_counter_sub_pool *sub_pool; unsigned int entry_index; unsigned int stop_index; - int i; + int i, err; - sub_pool = &mlxsw_sp_counter_sub_pools[sub_pool_id]; + sub_pool = &pool->sub_pools[sub_pool_id]; stop_index = sub_pool->base_index + sub_pool->size; entry_index = sub_pool->base_index; + spin_lock(&pool->counter_pool_lock); entry_index = find_next_zero_bit(pool->usage, stop_index, entry_index); - if (entry_index == stop_index) - return -ENOBUFS; + if (entry_index == stop_index) { + err = -ENOBUFS; + goto err_alloc; + } /* The sub-pools can contain non-integer number of entries * so we must check for overflow */ - if (entry_index + sub_pool->entry_size > stop_index) - return -ENOBUFS; + if (entry_index + sub_pool->entry_size > stop_index) { + err = -ENOBUFS; + goto err_alloc; + } for (i = 0; i < sub_pool->entry_size; i++) __set_bit(entry_index + i, pool->usage); + spin_unlock(&pool->counter_pool_lock); *p_counter_index = entry_index; + atomic_add(sub_pool->entry_size, &sub_pool->active_entries_count); + atomic_add(sub_pool->entry_size, &pool->active_entries_count); return 0; + +err_alloc: + spin_unlock(&pool->counter_pool_lock); + return err; } void mlxsw_sp_counter_free(struct mlxsw_sp *mlxsw_sp, @@ -170,7 +234,77 @@ void mlxsw_sp_counter_free(struct mlxsw_sp *mlxsw_sp, if (WARN_ON(counter_index >= pool->pool_size)) return; - sub_pool = &mlxsw_sp_counter_sub_pools[sub_pool_id]; + sub_pool = &pool->sub_pools[sub_pool_id]; + spin_lock(&pool->counter_pool_lock); for (i = 0; i < sub_pool->entry_size; i++) __clear_bit(counter_index + i, pool->usage); + spin_unlock(&pool->counter_pool_lock); + atomic_sub(sub_pool->entry_size, &sub_pool->active_entries_count); + atomic_sub(sub_pool->entry_size, &pool->active_entries_count); +} + +int mlxsw_sp_counter_resources_register(struct mlxsw_core *mlxsw_core) +{ + static struct devlink_resource_size_params size_params; + struct devlink *devlink = priv_to_devlink(mlxsw_core); + const struct mlxsw_sp_counter_sub_pool *sub_pool; + unsigned int total_bank_config; + u64 sub_pool_size; + u64 base_index; + u64 pool_size; + u64 bank_size; + int err; + int i; + + if (!MLXSW_CORE_RES_VALID(mlxsw_core, COUNTER_POOL_SIZE) || + !MLXSW_CORE_RES_VALID(mlxsw_core, COUNTER_BANK_SIZE)) + return -EIO; + + pool_size = MLXSW_CORE_RES_GET(mlxsw_core, COUNTER_POOL_SIZE); + bank_size = MLXSW_CORE_RES_GET(mlxsw_core, COUNTER_BANK_SIZE); + + devlink_resource_size_params_init(&size_params, pool_size, + pool_size, bank_size, + DEVLINK_RESOURCE_UNIT_ENTRY); + err = devlink_resource_register(devlink, + MLXSW_SP_RESOURCE_NAME_COUNTERS, + pool_size, + MLXSW_SP_RESOURCE_COUNTERS, + DEVLINK_RESOURCE_ID_PARENT_TOP, + &size_params); + if (err) + return err; + + /* Allocation is based on bank count which should be + * specified for each sub pool statically. + */ + total_bank_config = 0; + base_index = 0; + for (i = 0; i < ARRAY_SIZE(mlxsw_sp_counter_sub_pools); i++) { + sub_pool = &mlxsw_sp_counter_sub_pools[i]; + sub_pool_size = sub_pool->bank_count * bank_size; + /* The last bank can't be fully used */ + if (base_index + sub_pool_size > pool_size) + sub_pool_size = pool_size - base_index; + base_index += sub_pool_size; + + devlink_resource_size_params_init(&size_params, sub_pool_size, + sub_pool_size, bank_size, + DEVLINK_RESOURCE_UNIT_ENTRY); + err = devlink_resource_register(devlink, + sub_pool->resource_name, + sub_pool_size, + sub_pool->resource_id, + MLXSW_SP_RESOURCE_COUNTERS, + &size_params); + if (err) + return err; + total_bank_config += sub_pool->bank_count; + } + + /* Check config is valid, no bank over subscription */ + if (WARN_ON(total_bank_config > div64_u64(pool_size, bank_size) + 1)) + return -EINVAL; + + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.h index 81465e267b10..a68d931090dd 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_cnt.h @@ -4,6 +4,7 @@ #ifndef _MLXSW_SPECTRUM_CNT_H #define _MLXSW_SPECTRUM_CNT_H +#include "core.h" #include "spectrum.h" enum mlxsw_sp_counter_sub_pool_id { @@ -19,5 +20,6 @@ void mlxsw_sp_counter_free(struct mlxsw_sp *mlxsw_sp, unsigned int counter_index); int mlxsw_sp_counter_pool_init(struct mlxsw_sp *mlxsw_sp); void mlxsw_sp_counter_pool_fini(struct mlxsw_sp *mlxsw_sp); +int mlxsw_sp_counter_resources_register(struct mlxsw_core *mlxsw_core); #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c index 2dc0978428e6..daf029931b5f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c @@ -2,6 +2,7 @@ /* Copyright (c) 2017-2018 Mellanox Technologies. All rights reserved */ #include <linux/kernel.h> +#include <linux/mutex.h> #include <net/devlink.h> #include "spectrum.h" @@ -210,7 +211,7 @@ mlxsw_sp_dpipe_table_erif_entries_dump(void *priv, bool counters_enabled, return err; rif_count = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); i = 0; start_again: err = devlink_dpipe_entry_ctx_prepare(dump_ctx); @@ -241,14 +242,14 @@ start_again: devlink_dpipe_entry_ctx_close(dump_ctx); if (i != rif_count) goto start_again; - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); devlink_dpipe_entry_clear(&entry); return 0; err_entry_append: err_entry_get: err_ctx_prepare: - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); devlink_dpipe_entry_clear(&entry); return err; } @@ -258,7 +259,7 @@ static int mlxsw_sp_dpipe_table_erif_counters_update(void *priv, bool enable) struct mlxsw_sp *mlxsw_sp = priv; int i; - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); i++) { struct mlxsw_sp_rif *rif = mlxsw_sp_rif_by_index(mlxsw_sp, i); @@ -271,7 +272,7 @@ static int mlxsw_sp_dpipe_table_erif_counters_update(void *priv, bool enable) mlxsw_sp_rif_counter_free(mlxsw_sp, rif, MLXSW_SP_RIF_COUNTER_EGRESS); } - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); return 0; } @@ -546,7 +547,7 @@ mlxsw_sp_dpipe_table_host_entries_get(struct mlxsw_sp *mlxsw_sp, int i, j; int err; - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); i = 0; rif_count = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); start_again: @@ -602,12 +603,12 @@ out: if (i != rif_count) goto start_again; - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); return 0; err_ctx_prepare: err_entry_append: - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); return err; } @@ -662,7 +663,7 @@ mlxsw_sp_dpipe_table_host_counters_update(struct mlxsw_sp *mlxsw_sp, { int i; - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); i++) { struct mlxsw_sp_rif *rif = mlxsw_sp_rif_by_index(mlxsw_sp, i); struct mlxsw_sp_neigh_entry *neigh_entry; @@ -684,7 +685,7 @@ mlxsw_sp_dpipe_table_host_counters_update(struct mlxsw_sp *mlxsw_sp, enable); } } - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); } static int mlxsw_sp_dpipe_table_host4_counters_update(void *priv, bool enable) @@ -701,7 +702,7 @@ mlxsw_sp_dpipe_table_host_size_get(struct mlxsw_sp *mlxsw_sp, int type) u64 size = 0; int i; - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); i++) { struct mlxsw_sp_rif *rif = mlxsw_sp_rif_by_index(mlxsw_sp, i); struct mlxsw_sp_neigh_entry *neigh_entry; @@ -721,7 +722,7 @@ mlxsw_sp_dpipe_table_host_size_get(struct mlxsw_sp *mlxsw_sp, int type) size++; } } - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); return size; } @@ -1093,7 +1094,7 @@ mlxsw_sp_dpipe_table_adj_entries_get(struct mlxsw_sp *mlxsw_sp, int j; int err; - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); nh_count_max = mlxsw_sp_dpipe_table_adj_size(mlxsw_sp); start_again: err = devlink_dpipe_entry_ctx_prepare(dump_ctx); @@ -1130,13 +1131,13 @@ skip: devlink_dpipe_entry_ctx_close(dump_ctx); if (nh_count != nh_count_max) goto start_again; - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); return 0; err_ctx_prepare: err_entry_append: - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); return err; } @@ -1206,9 +1207,9 @@ mlxsw_sp_dpipe_table_adj_size_get(void *priv) struct mlxsw_sp *mlxsw_sp = priv; u64 size; - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); size = mlxsw_sp_dpipe_table_adj_size(mlxsw_sp); - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); return size; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c index 8df3cb21baa6..004c42274e48 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_fid.c @@ -8,6 +8,7 @@ #include <linux/netdevice.h> #include <linux/rhashtable.h> #include <linux/rtnetlink.h> +#include <linux/refcount.h> #include "spectrum.h" #include "reg.h" @@ -24,7 +25,7 @@ struct mlxsw_sp_fid_core { struct mlxsw_sp_fid { struct list_head list; struct mlxsw_sp_rif *rif; - unsigned int ref_count; + refcount_t ref_count; u16 fid_index; struct mlxsw_sp_fid_family *fid_family; struct rhash_head ht_node; @@ -149,7 +150,7 @@ struct mlxsw_sp_fid *mlxsw_sp_fid_lookup_by_index(struct mlxsw_sp *mlxsw_sp, fid = rhashtable_lookup_fast(&mlxsw_sp->fid_core->fid_ht, &fid_index, mlxsw_sp_fid_ht_params); if (fid) - fid->ref_count++; + refcount_inc(&fid->ref_count); return fid; } @@ -183,7 +184,7 @@ struct mlxsw_sp_fid *mlxsw_sp_fid_lookup_by_vni(struct mlxsw_sp *mlxsw_sp, fid = rhashtable_lookup_fast(&mlxsw_sp->fid_core->vni_ht, &vni, mlxsw_sp_fid_vni_ht_params); if (fid) - fid->ref_count++; + refcount_inc(&fid->ref_count); return fid; } @@ -437,16 +438,6 @@ static int mlxsw_sp_fid_vni_op(struct mlxsw_sp *mlxsw_sp, u16 fid_index, return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); } -static int mlxsw_sp_fid_vid_map(struct mlxsw_sp *mlxsw_sp, u16 fid_index, - u16 vid, bool valid) -{ - enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_VID_TO_FID; - char svfa_pl[MLXSW_REG_SVFA_LEN]; - - mlxsw_reg_svfa_pack(svfa_pl, 0, mt, valid, fid_index, vid); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svfa), svfa_pl); -} - static int __mlxsw_sp_fid_port_vid_map(struct mlxsw_sp *mlxsw_sp, u16 fid_index, u8 local_port, u16 vid, bool valid) { @@ -457,140 +448,6 @@ static int __mlxsw_sp_fid_port_vid_map(struct mlxsw_sp *mlxsw_sp, u16 fid_index, return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svfa), svfa_pl); } -static int mlxsw_sp_fid_8021q_configure(struct mlxsw_sp_fid *fid) -{ - struct mlxsw_sp *mlxsw_sp = fid->fid_family->mlxsw_sp; - struct mlxsw_sp_fid_8021q *fid_8021q; - int err; - - err = mlxsw_sp_fid_op(mlxsw_sp, fid->fid_index, fid->fid_index, true); - if (err) - return err; - - fid_8021q = mlxsw_sp_fid_8021q_fid(fid); - err = mlxsw_sp_fid_vid_map(mlxsw_sp, fid->fid_index, fid_8021q->vid, - true); - if (err) - goto err_fid_map; - - return 0; - -err_fid_map: - mlxsw_sp_fid_op(mlxsw_sp, fid->fid_index, 0, false); - return err; -} - -static void mlxsw_sp_fid_8021q_deconfigure(struct mlxsw_sp_fid *fid) -{ - struct mlxsw_sp *mlxsw_sp = fid->fid_family->mlxsw_sp; - struct mlxsw_sp_fid_8021q *fid_8021q; - - fid_8021q = mlxsw_sp_fid_8021q_fid(fid); - mlxsw_sp_fid_vid_map(mlxsw_sp, fid->fid_index, fid_8021q->vid, false); - mlxsw_sp_fid_op(mlxsw_sp, fid->fid_index, 0, false); -} - -static int mlxsw_sp_fid_8021q_index_alloc(struct mlxsw_sp_fid *fid, - const void *arg, u16 *p_fid_index) -{ - struct mlxsw_sp_fid_family *fid_family = fid->fid_family; - u16 vid = *(u16 *) arg; - - /* Use 1:1 mapping for simplicity although not a must */ - if (vid < fid_family->start_index || vid > fid_family->end_index) - return -EINVAL; - *p_fid_index = vid; - - return 0; -} - -static bool -mlxsw_sp_fid_8021q_compare(const struct mlxsw_sp_fid *fid, const void *arg) -{ - u16 vid = *(u16 *) arg; - - return mlxsw_sp_fid_8021q_fid(fid)->vid == vid; -} - -static u16 mlxsw_sp_fid_8021q_flood_index(const struct mlxsw_sp_fid *fid) -{ - return fid->fid_index; -} - -static int mlxsw_sp_fid_8021q_port_vid_map(struct mlxsw_sp_fid *fid, - struct mlxsw_sp_port *mlxsw_sp_port, - u16 vid) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - u8 local_port = mlxsw_sp_port->local_port; - - /* In case there are no {Port, VID} => FID mappings on the port, - * we can use the global VID => FID mapping we created when the - * FID was configured. - */ - if (mlxsw_sp->fid_core->port_fid_mappings[local_port] == 0) - return 0; - return __mlxsw_sp_fid_port_vid_map(mlxsw_sp, fid->fid_index, local_port, - vid, true); -} - -static void -mlxsw_sp_fid_8021q_port_vid_unmap(struct mlxsw_sp_fid *fid, - struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - u8 local_port = mlxsw_sp_port->local_port; - - if (mlxsw_sp->fid_core->port_fid_mappings[local_port] == 0) - return; - __mlxsw_sp_fid_port_vid_map(mlxsw_sp, fid->fid_index, local_port, vid, - false); -} - -static const struct mlxsw_sp_fid_ops mlxsw_sp_fid_8021q_ops = { - .setup = mlxsw_sp_fid_8021q_setup, - .configure = mlxsw_sp_fid_8021q_configure, - .deconfigure = mlxsw_sp_fid_8021q_deconfigure, - .index_alloc = mlxsw_sp_fid_8021q_index_alloc, - .compare = mlxsw_sp_fid_8021q_compare, - .flood_index = mlxsw_sp_fid_8021q_flood_index, - .port_vid_map = mlxsw_sp_fid_8021q_port_vid_map, - .port_vid_unmap = mlxsw_sp_fid_8021q_port_vid_unmap, -}; - -static const struct mlxsw_sp_flood_table mlxsw_sp_fid_8021q_flood_tables[] = { - { - .packet_type = MLXSW_SP_FLOOD_TYPE_UC, - .bridge_type = MLXSW_REG_SFGC_BRIDGE_TYPE_1Q_FID, - .table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFSET, - .table_index = 0, - }, - { - .packet_type = MLXSW_SP_FLOOD_TYPE_MC, - .bridge_type = MLXSW_REG_SFGC_BRIDGE_TYPE_1Q_FID, - .table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFSET, - .table_index = 1, - }, - { - .packet_type = MLXSW_SP_FLOOD_TYPE_BC, - .bridge_type = MLXSW_REG_SFGC_BRIDGE_TYPE_1Q_FID, - .table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFSET, - .table_index = 2, - }, -}; - -/* Range and flood configuration must match mlxsw_config_profile */ -static const struct mlxsw_sp_fid_family mlxsw_sp_fid_8021q_family = { - .type = MLXSW_SP_FID_TYPE_8021Q, - .fid_size = sizeof(struct mlxsw_sp_fid_8021q), - .start_index = 1, - .end_index = VLAN_VID_MASK, - .flood_tables = mlxsw_sp_fid_8021q_flood_tables, - .nr_flood_tables = ARRAY_SIZE(mlxsw_sp_fid_8021q_flood_tables), - .rif_type = MLXSW_SP_RIF_TYPE_VLAN, - .ops = &mlxsw_sp_fid_8021q_ops, -}; - static struct mlxsw_sp_fid_8021d * mlxsw_sp_fid_8021d_fid(const struct mlxsw_sp_fid *fid) { @@ -845,6 +702,14 @@ static const struct mlxsw_sp_fid_family mlxsw_sp_fid_8021d_family = { .lag_vid_valid = 1, }; +static bool +mlxsw_sp_fid_8021q_compare(const struct mlxsw_sp_fid *fid, const void *arg) +{ + u16 vid = *(u16 *) arg; + + return mlxsw_sp_fid_8021q_fid(fid)->vid == vid; +} + static void mlxsw_sp_fid_8021q_fdb_clear_offload(const struct mlxsw_sp_fid *fid, const struct net_device *nve_dev) @@ -1030,7 +895,7 @@ static struct mlxsw_sp_fid *mlxsw_sp_fid_lookup(struct mlxsw_sp *mlxsw_sp, list_for_each_entry(fid, &fid_family->fids_list, list) { if (!fid->fid_family->ops->compare(fid, arg)) continue; - fid->ref_count++; + refcount_inc(&fid->ref_count); return fid; } @@ -1075,7 +940,7 @@ static struct mlxsw_sp_fid *mlxsw_sp_fid_get(struct mlxsw_sp *mlxsw_sp, goto err_rhashtable_insert; list_add(&fid->list, &fid_family->fids_list); - fid->ref_count++; + refcount_set(&fid->ref_count, 1); return fid; err_rhashtable_insert: @@ -1093,7 +958,7 @@ void mlxsw_sp_fid_put(struct mlxsw_sp_fid *fid) struct mlxsw_sp_fid_family *fid_family = fid->fid_family; struct mlxsw_sp *mlxsw_sp = fid_family->mlxsw_sp; - if (--fid->ref_count != 0) + if (!refcount_dec_and_test(&fid->ref_count)) return; list_del(&fid->list); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c index b607919c8ad0..2f76908cae73 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c @@ -26,11 +26,20 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, if (!flow_action_has_entries(flow_action)) return 0; + if (!flow_action_mixed_hw_stats_check(flow_action, extack)) + return -EOPNOTSUPP; - /* Count action is inserted first */ - err = mlxsw_sp_acl_rulei_act_count(mlxsw_sp, rulei, extack); - if (err) - return err; + act = flow_action_first_entry_get(flow_action); + if (act->hw_stats == FLOW_ACTION_HW_STATS_ANY || + act->hw_stats == FLOW_ACTION_HW_STATS_IMMEDIATE) { + /* Count action is inserted first */ + err = mlxsw_sp_acl_rulei_act_count(mlxsw_sp, rulei, extack); + if (err) + return err; + } else if (act->hw_stats != FLOW_ACTION_HW_STATS_DISABLED) { + NL_SET_ERR_MSG_MOD(extack, "Unsupported action HW stats type"); + return -EOPNOTSUPP; + } flow_action_for_each(i, act, flow_action) { switch (act->id) { @@ -41,12 +50,30 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, return err; } break; - case FLOW_ACTION_DROP: - err = mlxsw_sp_acl_rulei_act_drop(rulei); + case FLOW_ACTION_DROP: { + bool ingress; + + if (mlxsw_sp_acl_block_is_mixed_bound(block)) { + NL_SET_ERR_MSG_MOD(extack, "Drop action is not supported when block is bound to ingress and egress"); + return -EOPNOTSUPP; + } + ingress = mlxsw_sp_acl_block_is_ingress_bound(block); + err = mlxsw_sp_acl_rulei_act_drop(rulei, ingress, + act->cookie, extack); if (err) { NL_SET_ERR_MSG_MOD(extack, "Cannot append drop action"); return err; } + + /* Forbid block with this rulei to be bound + * to ingress/egress in future. Ingress rule is + * a blocker for egress and vice versa. + */ + if (ingress) + rulei->egress_bind_blocker = 1; + else + rulei->ingress_bind_blocker = 1; + } break; case FLOW_ACTION_TRAP: err = mlxsw_sp_acl_rulei_act_trap(rulei); @@ -127,6 +154,25 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, act->id, vid, proto, prio, extack); } + case FLOW_ACTION_PRIORITY: + return mlxsw_sp_acl_rulei_act_priority(mlxsw_sp, rulei, + act->priority, + extack); + case FLOW_ACTION_MANGLE: { + enum flow_action_mangle_base htype = act->mangle.htype; + __be32 be_mask = (__force __be32) act->mangle.mask; + __be32 be_val = (__force __be32) act->mangle.val; + u32 offset = act->mangle.offset; + u32 mask = be32_to_cpu(be_mask); + u32 val = be32_to_cpu(be_val); + + err = mlxsw_sp_acl_rulei_act_mangle(mlxsw_sp, rulei, + htype, offset, + mask, val, extack); + if (err) + return err; + break; + } default: NL_SET_ERR_MSG_MOD(extack, "Unsupported action"); dev_err(mlxsw_sp->bus_info->dev, "Unsupported action\n"); @@ -525,6 +571,7 @@ int mlxsw_sp_flower_stats(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_block *block, struct flow_cls_offload *f) { + enum flow_action_hw_stats used_hw_stats = FLOW_ACTION_HW_STATS_DISABLED; struct mlxsw_sp_acl_ruleset *ruleset; struct mlxsw_sp_acl_rule *rule; u64 packets; @@ -543,11 +590,11 @@ int mlxsw_sp_flower_stats(struct mlxsw_sp *mlxsw_sp, return -EINVAL; err = mlxsw_sp_acl_rule_get_stats(mlxsw_sp, rule, &packets, &bytes, - &lastuse); + &lastuse, &used_hw_stats); if (err) goto err_rule_get_stats; - flow_stats_update(&f->stats, bytes, packets, lastuse); + flow_stats_update(&f->stats, bytes, packets, lastuse, used_hw_stats); mlxsw_sp_acl_ruleset_put(mlxsw_sp, ruleset); return 0; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c index 1e4cdee7bcd7..20d72f1c0cee 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c @@ -2,13 +2,15 @@ /* Copyright (c) 2016-2018 Mellanox Technologies. All rights reserved */ #include <linux/kernel.h> +#include <linux/mutex.h> #include <linux/slab.h> #include "spectrum.h" struct mlxsw_sp_kvdl { const struct mlxsw_sp_kvdl_ops *kvdl_ops; - unsigned long priv[0]; + struct mutex kvdl_lock; /* Protects kvdl allocations */ + unsigned long priv[]; /* priv has to be always the last item */ }; @@ -22,6 +24,7 @@ int mlxsw_sp_kvdl_init(struct mlxsw_sp *mlxsw_sp) GFP_KERNEL); if (!kvdl) return -ENOMEM; + mutex_init(&kvdl->kvdl_lock); kvdl->kvdl_ops = kvdl_ops; mlxsw_sp->kvdl = kvdl; @@ -31,6 +34,7 @@ int mlxsw_sp_kvdl_init(struct mlxsw_sp *mlxsw_sp) return 0; err_init: + mutex_destroy(&kvdl->kvdl_lock); kfree(kvdl); return err; } @@ -40,6 +44,7 @@ void mlxsw_sp_kvdl_fini(struct mlxsw_sp *mlxsw_sp) struct mlxsw_sp_kvdl *kvdl = mlxsw_sp->kvdl; kvdl->kvdl_ops->fini(mlxsw_sp, kvdl->priv); + mutex_destroy(&kvdl->kvdl_lock); kfree(kvdl); } @@ -48,9 +53,14 @@ int mlxsw_sp_kvdl_alloc(struct mlxsw_sp *mlxsw_sp, unsigned int entry_count, u32 *p_entry_index) { struct mlxsw_sp_kvdl *kvdl = mlxsw_sp->kvdl; + int err; + + mutex_lock(&kvdl->kvdl_lock); + err = kvdl->kvdl_ops->alloc(mlxsw_sp, kvdl->priv, type, + entry_count, p_entry_index); + mutex_unlock(&kvdl->kvdl_lock); - return kvdl->kvdl_ops->alloc(mlxsw_sp, kvdl->priv, type, - entry_count, p_entry_index); + return err; } void mlxsw_sp_kvdl_free(struct mlxsw_sp *mlxsw_sp, @@ -59,8 +69,10 @@ void mlxsw_sp_kvdl_free(struct mlxsw_sp *mlxsw_sp, { struct mlxsw_sp_kvdl *kvdl = mlxsw_sp->kvdl; + mutex_lock(&kvdl->kvdl_lock); kvdl->kvdl_ops->free(mlxsw_sp, kvdl->priv, type, entry_count, entry_index); + mutex_unlock(&kvdl->kvdl_lock); } int mlxsw_sp_kvdl_alloc_count_query(struct mlxsw_sp *mlxsw_sp, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c index 336e5ecc68f8..47eb751a2570 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 /* Copyright (c) 2017-2018 Mellanox Technologies. All rights reserved */ +#include <linux/mutex.h> #include <linux/rhashtable.h> #include <net/ipv6.h> @@ -12,6 +13,7 @@ struct mlxsw_sp_mr { void *catchall_route_priv; struct delayed_work stats_update_dw; struct list_head table_list; + struct mutex table_list_lock; /* Protects table_list */ #define MLXSW_SP_MR_ROUTES_COUNTER_UPDATE_INTERVAL 5000 /* ms */ unsigned long priv[0]; /* priv has to be always the last item */ @@ -66,9 +68,10 @@ struct mlxsw_sp_mr_table { u32 vr_id; struct mlxsw_sp_mr_vif vifs[MAXVIFS]; struct list_head route_list; + struct mutex route_list_lock; /* Protects route_list */ struct rhashtable route_ht; const struct mlxsw_sp_mr_table_ops *ops; - char catchall_route_priv[0]; + char catchall_route_priv[]; /* catchall_route_priv has to be always the last item */ }; @@ -370,11 +373,13 @@ static void mlxsw_sp_mr_mfc_offload_update(struct mlxsw_sp_mr_route *mr_route) static void __mlxsw_sp_mr_route_del(struct mlxsw_sp_mr_table *mr_table, struct mlxsw_sp_mr_route *mr_route) { + WARN_ON_ONCE(!mutex_is_locked(&mr_table->route_list_lock)); + mlxsw_sp_mr_mfc_offload_set(mr_route, false); - mlxsw_sp_mr_route_erase(mr_table, mr_route); rhashtable_remove_fast(&mr_table->route_ht, &mr_route->ht_node, mlxsw_sp_mr_route_ht_params); list_del(&mr_route->node); + mlxsw_sp_mr_route_erase(mr_table, mr_route); mlxsw_sp_mr_route_destroy(mr_table, mr_route); } @@ -415,19 +420,21 @@ int mlxsw_sp_mr_route_add(struct mlxsw_sp_mr_table *mr_table, goto err_duplicate_route; } + /* Write the route to the hardware */ + err = mlxsw_sp_mr_route_write(mr_table, mr_route, replace); + if (err) + goto err_mr_route_write; + /* Put it in the table data-structures */ + mutex_lock(&mr_table->route_list_lock); list_add_tail(&mr_route->node, &mr_table->route_list); + mutex_unlock(&mr_table->route_list_lock); err = rhashtable_insert_fast(&mr_table->route_ht, &mr_route->ht_node, mlxsw_sp_mr_route_ht_params); if (err) goto err_rhashtable_insert; - /* Write the route to the hardware */ - err = mlxsw_sp_mr_route_write(mr_table, mr_route, replace); - if (err) - goto err_mr_route_write; - /* Destroy the original route */ if (replace) { rhashtable_remove_fast(&mr_table->route_ht, @@ -440,11 +447,12 @@ int mlxsw_sp_mr_route_add(struct mlxsw_sp_mr_table *mr_table, mlxsw_sp_mr_mfc_offload_update(mr_route); return 0; -err_mr_route_write: - rhashtable_remove_fast(&mr_table->route_ht, &mr_route->ht_node, - mlxsw_sp_mr_route_ht_params); err_rhashtable_insert: + mutex_lock(&mr_table->route_list_lock); list_del(&mr_route->node); + mutex_unlock(&mr_table->route_list_lock); + mlxsw_sp_mr_route_erase(mr_table, mr_route); +err_mr_route_write: err_no_orig_route: err_duplicate_route: mlxsw_sp_mr_route_destroy(mr_table, mr_route); @@ -460,8 +468,11 @@ void mlxsw_sp_mr_route_del(struct mlxsw_sp_mr_table *mr_table, mr_table->ops->key_create(mr_table, &key, mfc); mr_route = rhashtable_lookup_fast(&mr_table->route_ht, &key, mlxsw_sp_mr_route_ht_params); - if (mr_route) + if (mr_route) { + mutex_lock(&mr_table->route_list_lock); __mlxsw_sp_mr_route_del(mr_table, mr_route); + mutex_unlock(&mr_table->route_list_lock); + } } /* Should be called after the VIF struct is updated */ @@ -910,6 +921,7 @@ struct mlxsw_sp_mr_table *mlxsw_sp_mr_table_create(struct mlxsw_sp *mlxsw_sp, mr_table->proto = proto; mr_table->ops = &mlxsw_sp_mr_table_ops_arr[proto]; INIT_LIST_HEAD(&mr_table->route_list); + mutex_init(&mr_table->route_list_lock); err = rhashtable_init(&mr_table->route_ht, &mlxsw_sp_mr_route_ht_params); @@ -927,12 +939,15 @@ struct mlxsw_sp_mr_table *mlxsw_sp_mr_table_create(struct mlxsw_sp *mlxsw_sp, &catchall_route_params); if (err) goto err_ops_route_create; + mutex_lock(&mr->table_list_lock); list_add_tail(&mr_table->node, &mr->table_list); + mutex_unlock(&mr->table_list_lock); return mr_table; err_ops_route_create: rhashtable_destroy(&mr_table->route_ht); err_route_rhashtable_init: + mutex_destroy(&mr_table->route_list_lock); kfree(mr_table); return ERR_PTR(err); } @@ -943,10 +958,13 @@ void mlxsw_sp_mr_table_destroy(struct mlxsw_sp_mr_table *mr_table) struct mlxsw_sp_mr *mr = mlxsw_sp->mr; WARN_ON(!mlxsw_sp_mr_table_empty(mr_table)); + mutex_lock(&mr->table_list_lock); list_del(&mr_table->node); + mutex_unlock(&mr->table_list_lock); mr->mr_ops->route_destroy(mlxsw_sp, mr->priv, &mr_table->catchall_route_priv); rhashtable_destroy(&mr_table->route_ht); + mutex_destroy(&mr_table->route_list_lock); kfree(mr_table); } @@ -955,8 +973,10 @@ void mlxsw_sp_mr_table_flush(struct mlxsw_sp_mr_table *mr_table) struct mlxsw_sp_mr_route *mr_route, *tmp; int i; + mutex_lock(&mr_table->route_list_lock); list_for_each_entry_safe(mr_route, tmp, &mr_table->route_list, node) __mlxsw_sp_mr_route_del(mr_table, mr_route); + mutex_unlock(&mr_table->route_list_lock); for (i = 0; i < MAXVIFS; i++) { mr_table->vifs[i].dev = NULL; @@ -1000,12 +1020,15 @@ static void mlxsw_sp_mr_stats_update(struct work_struct *work) struct mlxsw_sp_mr_route *mr_route; unsigned long interval; - rtnl_lock(); - list_for_each_entry(mr_table, &mr->table_list, node) + mutex_lock(&mr->table_list_lock); + list_for_each_entry(mr_table, &mr->table_list, node) { + mutex_lock(&mr_table->route_list_lock); list_for_each_entry(mr_route, &mr_table->route_list, node) mlxsw_sp_mr_route_stats_update(mr_table->mlxsw_sp, mr_route); - rtnl_unlock(); + mutex_unlock(&mr_table->route_list_lock); + } + mutex_unlock(&mr->table_list_lock); interval = msecs_to_jiffies(MLXSW_SP_MR_ROUTES_COUNTER_UPDATE_INTERVAL); mlxsw_core_schedule_dw(&mr->stats_update_dw, interval); @@ -1024,6 +1047,7 @@ int mlxsw_sp_mr_init(struct mlxsw_sp *mlxsw_sp, mr->mr_ops = mr_ops; mlxsw_sp->mr = mr; INIT_LIST_HEAD(&mr->table_list); + mutex_init(&mr->table_list_lock); err = mr_ops->init(mlxsw_sp, mr->priv); if (err) @@ -1035,6 +1059,7 @@ int mlxsw_sp_mr_init(struct mlxsw_sp *mlxsw_sp, mlxsw_core_schedule_dw(&mr->stats_update_dw, interval); return 0; err: + mutex_destroy(&mr->table_list_lock); kfree(mr); return err; } @@ -1045,5 +1070,6 @@ void mlxsw_sp_mr_fini(struct mlxsw_sp *mlxsw_sp) cancel_delayed_work_sync(&mr->stats_update_dw); mr->mr_ops->fini(mlxsw_sp, mr->priv); + mutex_destroy(&mr->table_list_lock); kfree(mr); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_nve.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_nve.c index 2153bcc4b585..54d3e7dcd303 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_nve.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_nve.c @@ -67,7 +67,7 @@ struct mlxsw_sp_nve_mc_record { struct mlxsw_sp_nve_mc_list *mc_list; const struct mlxsw_sp_nve_mc_record_ops *ops; u32 kvdl_index; - struct mlxsw_sp_nve_mc_entry entries[0]; + struct mlxsw_sp_nve_mc_entry entries[]; }; struct mlxsw_sp_nve_mc_list { @@ -713,27 +713,6 @@ static void mlxsw_sp_nve_flood_ip_flush(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_nve_mc_list_put(mlxsw_sp, mc_list); } -u32 mlxsw_sp_nve_decap_tunnel_index_get(const struct mlxsw_sp *mlxsw_sp) -{ - WARN_ON(mlxsw_sp->nve->num_nve_tunnels == 0); - - return mlxsw_sp->nve->tunnel_index; -} - -bool mlxsw_sp_nve_ipv4_route_is_decap(const struct mlxsw_sp *mlxsw_sp, - u32 tb_id, __be32 addr) -{ - struct mlxsw_sp_nve *nve = mlxsw_sp->nve; - struct mlxsw_sp_nve_config *config = &nve->config; - - if (nve->num_nve_tunnels && - config->ul_proto == MLXSW_SP_L3_PROTO_IPV4 && - config->ul_sip.addr4 == addr && config->ul_tb_id == tb_id) - return true; - - return false; -} - static int mlxsw_sp_nve_tunnel_init(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nve_config *config) { @@ -744,6 +723,8 @@ static int mlxsw_sp_nve_tunnel_init(struct mlxsw_sp *mlxsw_sp, if (nve->num_nve_tunnels++ != 0) return 0; + nve->config = *config; + err = mlxsw_sp_kvdl_alloc(mlxsw_sp, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ, 1, &nve->tunnel_index); if (err) @@ -760,6 +741,7 @@ err_ops_init: mlxsw_sp_kvdl_free(mlxsw_sp, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ, 1, nve->tunnel_index); err_kvdl_alloc: + memset(&nve->config, 0, sizeof(nve->config)); nve->num_nve_tunnels--; return err; } @@ -840,8 +822,6 @@ int mlxsw_sp_nve_fid_enable(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *fid, goto err_fid_vni_set; } - nve->config = config; - err = ops->fdb_replay(params->dev, params->vni, extack); if (err) goto err_fdb_replay; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c index 34f7c3501b08..9650562fc0ef 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c @@ -922,6 +922,8 @@ static int mlxsw_sp_ptp_get_message_types(const struct hwtstamp_config *config, case HWTSTAMP_TX_ONESTEP_SYNC: case HWTSTAMP_TX_ONESTEP_P2P: return -ERANGE; + default: + return -EINVAL; } switch (rx_filter) { @@ -952,6 +954,8 @@ static int mlxsw_sp_ptp_get_message_types(const struct hwtstamp_config *config, case HWTSTAMP_FILTER_SOME: case HWTSTAMP_FILTER_NTP_ALL: return -ERANGE; + default: + return -EINVAL; } *p_ing_types = ing_types; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c index 02526c53d4f5..670a43fe2a00 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c @@ -20,14 +20,17 @@ enum mlxsw_sp_qdisc_type { MLXSW_SP_QDISC_PRIO, MLXSW_SP_QDISC_ETS, MLXSW_SP_QDISC_TBF, + MLXSW_SP_QDISC_FIFO, }; +struct mlxsw_sp_qdisc; + struct mlxsw_sp_qdisc_ops { enum mlxsw_sp_qdisc_type type; int (*check_params)(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, void *params); - int (*replace)(struct mlxsw_sp_port *mlxsw_sp_port, + int (*replace)(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, void *params); int (*destroy)(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc); @@ -64,6 +67,25 @@ struct mlxsw_sp_qdisc { struct mlxsw_sp_qdisc_ops *ops; }; +struct mlxsw_sp_qdisc_state { + struct mlxsw_sp_qdisc root_qdisc; + struct mlxsw_sp_qdisc tclass_qdiscs[IEEE_8021QAZ_MAX_TCS]; + + /* When a PRIO or ETS are added, the invisible FIFOs in their bands are + * created first. When notifications for these FIFOs arrive, it is not + * known what qdisc their parent handle refers to. It could be a + * newly-created PRIO that will replace the currently-offloaded one, or + * it could be e.g. a RED that will be attached below it. + * + * As the notifications start to arrive, use them to note what the + * future parent handle is, and keep track of which child FIFOs were + * seen. Then when the parent is known, retroactively offload those + * FIFOs. + */ + u32 future_handle; + bool future_fifos[IEEE_8021QAZ_MAX_TCS]; +}; + static bool mlxsw_sp_qdisc_compare(struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, u32 handle, enum mlxsw_sp_qdisc_type type) @@ -77,36 +99,38 @@ static struct mlxsw_sp_qdisc * mlxsw_sp_qdisc_find(struct mlxsw_sp_port *mlxsw_sp_port, u32 parent, bool root_only) { + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; int tclass, child_index; if (parent == TC_H_ROOT) - return mlxsw_sp_port->root_qdisc; + return &qdisc_state->root_qdisc; - if (root_only || !mlxsw_sp_port->root_qdisc || - !mlxsw_sp_port->root_qdisc->ops || - TC_H_MAJ(parent) != mlxsw_sp_port->root_qdisc->handle || + if (root_only || !qdisc_state || + !qdisc_state->root_qdisc.ops || + TC_H_MAJ(parent) != qdisc_state->root_qdisc.handle || TC_H_MIN(parent) > IEEE_8021QAZ_MAX_TCS) return NULL; child_index = TC_H_MIN(parent); tclass = MLXSW_SP_PRIO_CHILD_TO_TCLASS(child_index); - return &mlxsw_sp_port->tclass_qdiscs[tclass]; + return &qdisc_state->tclass_qdiscs[tclass]; } static struct mlxsw_sp_qdisc * mlxsw_sp_qdisc_find_by_handle(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle) { + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; int i; - if (mlxsw_sp_port->root_qdisc->handle == handle) - return mlxsw_sp_port->root_qdisc; + if (qdisc_state->root_qdisc.handle == handle) + return &qdisc_state->root_qdisc; - if (mlxsw_sp_port->root_qdisc->handle == TC_H_UNSPEC) + if (qdisc_state->root_qdisc.handle == TC_H_UNSPEC) return NULL; for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) - if (mlxsw_sp_port->tclass_qdiscs[i].handle == handle) - return &mlxsw_sp_port->tclass_qdiscs[i]; + if (qdisc_state->tclass_qdiscs[i].handle == handle) + return &qdisc_state->tclass_qdiscs[i]; return NULL; } @@ -147,11 +171,15 @@ mlxsw_sp_qdisc_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, if (err) goto err_bad_param; - err = ops->replace(mlxsw_sp_port, mlxsw_sp_qdisc, params); + err = ops->replace(mlxsw_sp_port, handle, mlxsw_sp_qdisc, params); if (err) goto err_config; - if (mlxsw_sp_qdisc->handle != handle) { + /* Check if the Qdisc changed. That includes a situation where an + * invisible Qdisc replaces another one, or is being added for the + * first time. + */ + if (mlxsw_sp_qdisc->handle != handle || handle == TC_H_UNSPEC) { mlxsw_sp_qdisc->ops = ops; if (ops->clean_stats) ops->clean_stats(mlxsw_sp_port, mlxsw_sp_qdisc); @@ -295,7 +323,7 @@ mlxsw_sp_qdisc_get_tc_stats(struct mlxsw_sp_port *mlxsw_sp_port, static int mlxsw_sp_tclass_congestion_enable(struct mlxsw_sp_port *mlxsw_sp_port, int tclass_num, u32 min, u32 max, - u32 probability, bool is_ecn) + u32 probability, bool is_wred, bool is_ecn) { char cwtpm_cmd[MLXSW_REG_CWTPM_LEN]; char cwtp_cmd[MLXSW_REG_CWTP_LEN]; @@ -313,7 +341,7 @@ mlxsw_sp_tclass_congestion_enable(struct mlxsw_sp_port *mlxsw_sp_port, return err; mlxsw_reg_cwtpm_pack(cwtpm_cmd, mlxsw_sp_port->local_port, tclass_num, - MLXSW_REG_CWTP_DEFAULT_PROFILE, true, is_ecn); + MLXSW_REG_CWTP_DEFAULT_PROFILE, is_wred, is_ecn); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(cwtpm), cwtpm_cmd); } @@ -347,7 +375,6 @@ mlxsw_sp_setup_tc_qdisc_red_clean_stats(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_qdisc->prio_bitmap, &stats_base->tx_packets, &stats_base->tx_bytes); - red_base->prob_mark = xstats->ecn; red_base->prob_drop = xstats->wred_drop[tclass_num]; red_base->pdrop = mlxsw_sp_xstats_tail_drop(xstats, tclass_num); @@ -361,7 +388,8 @@ static int mlxsw_sp_qdisc_red_destroy(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) { - struct mlxsw_sp_qdisc *root_qdisc = mlxsw_sp_port->root_qdisc; + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; + struct mlxsw_sp_qdisc *root_qdisc = &qdisc_state->root_qdisc; if (root_qdisc != mlxsw_sp_qdisc) root_qdisc->stats_base.backlog -= @@ -400,7 +428,7 @@ mlxsw_sp_qdisc_red_check_params(struct mlxsw_sp_port *mlxsw_sp_port, } static int -mlxsw_sp_qdisc_red_replace(struct mlxsw_sp_port *mlxsw_sp_port, +mlxsw_sp_qdisc_red_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, void *params) { @@ -417,8 +445,9 @@ mlxsw_sp_qdisc_red_replace(struct mlxsw_sp_port *mlxsw_sp_port, prob = DIV_ROUND_UP(prob, 1 << 16); min = mlxsw_sp_bytes_cells(mlxsw_sp, p->min); max = mlxsw_sp_bytes_cells(mlxsw_sp, p->max); - return mlxsw_sp_tclass_congestion_enable(mlxsw_sp_port, tclass_num, min, - max, prob, p->is_ecn); + return mlxsw_sp_tclass_congestion_enable(mlxsw_sp_port, tclass_num, + min, max, prob, + !p->is_nodrop, p->is_ecn); } static void @@ -453,22 +482,19 @@ mlxsw_sp_qdisc_get_red_xstats(struct mlxsw_sp_port *mlxsw_sp_port, u8 tclass_num = mlxsw_sp_qdisc->tclass_num; struct mlxsw_sp_port_xstats *xstats; struct red_stats *res = xstats_ptr; - int early_drops, marks, pdrops; + int early_drops, pdrops; xstats = &mlxsw_sp_port->periodic_hw_stats.xstats; early_drops = xstats->wred_drop[tclass_num] - xstats_base->prob_drop; - marks = xstats->ecn - xstats_base->prob_mark; pdrops = mlxsw_sp_xstats_tail_drop(xstats, tclass_num) - xstats_base->pdrop; res->pdrop += pdrops; res->prob_drop += early_drops; - res->prob_mark += marks; xstats_base->pdrop += pdrops; xstats_base->prob_drop += early_drops; - xstats_base->prob_mark += marks; return 0; } @@ -486,8 +512,7 @@ mlxsw_sp_qdisc_get_red_stats(struct mlxsw_sp_port *mlxsw_sp_port, stats_base = &mlxsw_sp_qdisc->stats_base; mlxsw_sp_qdisc_get_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc, stats_ptr); - overlimits = xstats->wred_drop[tclass_num] + xstats->ecn - - stats_base->overlimits; + overlimits = xstats->wred_drop[tclass_num] - stats_base->overlimits; stats_ptr->qstats->overlimits += overlimits; stats_base->overlimits += overlimits; @@ -564,7 +589,8 @@ static int mlxsw_sp_qdisc_tbf_destroy(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) { - struct mlxsw_sp_qdisc *root_qdisc = mlxsw_sp_port->root_qdisc; + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; + struct mlxsw_sp_qdisc *root_qdisc = &qdisc_state->root_qdisc; if (root_qdisc != mlxsw_sp_qdisc) root_qdisc->stats_base.backlog -= @@ -651,7 +677,7 @@ mlxsw_sp_qdisc_tbf_check_params(struct mlxsw_sp_port *mlxsw_sp_port, } static int -mlxsw_sp_qdisc_tbf_replace(struct mlxsw_sp_port *mlxsw_sp_port, +mlxsw_sp_qdisc_tbf_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, void *params) { @@ -740,8 +766,121 @@ int mlxsw_sp_setup_tc_tbf(struct mlxsw_sp_port *mlxsw_sp_port, } static int +mlxsw_sp_qdisc_fifo_destroy(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) +{ + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; + struct mlxsw_sp_qdisc *root_qdisc = &qdisc_state->root_qdisc; + + if (root_qdisc != mlxsw_sp_qdisc) + root_qdisc->stats_base.backlog -= + mlxsw_sp_qdisc->stats_base.backlog; + return 0; +} + +static int +mlxsw_sp_qdisc_fifo_check_params(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + return 0; +} + +static int +mlxsw_sp_qdisc_fifo_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + return 0; +} + +static int +mlxsw_sp_qdisc_get_fifo_stats(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + struct tc_qopt_offload_stats *stats_ptr) +{ + mlxsw_sp_qdisc_get_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc, + stats_ptr); + return 0; +} + +static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_fifo = { + .type = MLXSW_SP_QDISC_FIFO, + .check_params = mlxsw_sp_qdisc_fifo_check_params, + .replace = mlxsw_sp_qdisc_fifo_replace, + .destroy = mlxsw_sp_qdisc_fifo_destroy, + .get_stats = mlxsw_sp_qdisc_get_fifo_stats, + .clean_stats = mlxsw_sp_setup_tc_qdisc_leaf_clean_stats, +}; + +int mlxsw_sp_setup_tc_fifo(struct mlxsw_sp_port *mlxsw_sp_port, + struct tc_fifo_qopt_offload *p) +{ + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc; + int tclass, child_index; + u32 parent_handle; + + /* Invisible FIFOs are tracked in future_handle and future_fifos. Make + * sure that not more than one qdisc is created for a port at a time. + * RTNL is a simple proxy for that. + */ + ASSERT_RTNL(); + + mlxsw_sp_qdisc = mlxsw_sp_qdisc_find(mlxsw_sp_port, p->parent, false); + if (!mlxsw_sp_qdisc && p->handle == TC_H_UNSPEC) { + parent_handle = TC_H_MAJ(p->parent); + if (parent_handle != qdisc_state->future_handle) { + /* This notifications is for a different Qdisc than + * previously. Wipe the future cache. + */ + memset(qdisc_state->future_fifos, 0, + sizeof(qdisc_state->future_fifos)); + qdisc_state->future_handle = parent_handle; + } + + child_index = TC_H_MIN(p->parent); + tclass = MLXSW_SP_PRIO_CHILD_TO_TCLASS(child_index); + if (tclass < IEEE_8021QAZ_MAX_TCS) { + if (p->command == TC_FIFO_REPLACE) + qdisc_state->future_fifos[tclass] = true; + else if (p->command == TC_FIFO_DESTROY) + qdisc_state->future_fifos[tclass] = false; + } + } + if (!mlxsw_sp_qdisc) + return -EOPNOTSUPP; + + if (p->command == TC_FIFO_REPLACE) { + return mlxsw_sp_qdisc_replace(mlxsw_sp_port, p->handle, + mlxsw_sp_qdisc, + &mlxsw_sp_qdisc_ops_fifo, NULL); + } + + if (!mlxsw_sp_qdisc_compare(mlxsw_sp_qdisc, p->handle, + MLXSW_SP_QDISC_FIFO)) + return -EOPNOTSUPP; + + switch (p->command) { + case TC_FIFO_DESTROY: + if (p->handle == mlxsw_sp_qdisc->handle) + return mlxsw_sp_qdisc_destroy(mlxsw_sp_port, + mlxsw_sp_qdisc); + return 0; + case TC_FIFO_STATS: + return mlxsw_sp_qdisc_get_stats(mlxsw_sp_port, mlxsw_sp_qdisc, + &p->stats); + case TC_FIFO_REPLACE: /* Handled above. */ + break; + } + + return -EOPNOTSUPP; +} + +static int __mlxsw_sp_qdisc_ets_destroy(struct mlxsw_sp_port *mlxsw_sp_port) { + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; int i; for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { @@ -751,8 +890,8 @@ __mlxsw_sp_qdisc_ets_destroy(struct mlxsw_sp_port *mlxsw_sp_port) MLXSW_REG_QEEC_HR_SUBGROUP, i, 0, false, 0); mlxsw_sp_qdisc_destroy(mlxsw_sp_port, - &mlxsw_sp_port->tclass_qdiscs[i]); - mlxsw_sp_port->tclass_qdiscs[i].prio_bitmap = 0; + &qdisc_state->tclass_qdiscs[i]); + qdisc_state->tclass_qdiscs[i].prio_bitmap = 0; } return 0; @@ -785,12 +924,13 @@ mlxsw_sp_qdisc_prio_check_params(struct mlxsw_sp_port *mlxsw_sp_port, } static int -__mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port, +__mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, unsigned int nbands, const unsigned int *quanta, const unsigned int *weights, const u8 *priomap) { + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; struct mlxsw_sp_qdisc *child_qdisc; int tclass, i, band, backlog; u8 old_priomap; @@ -798,7 +938,7 @@ __mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port, for (band = 0; band < nbands; band++) { tclass = MLXSW_SP_PRIO_BAND_TO_TCLASS(band); - child_qdisc = &mlxsw_sp_port->tclass_qdiscs[tclass]; + child_qdisc = &qdisc_state->tclass_qdiscs[tclass]; old_priomap = child_qdisc->prio_bitmap; child_qdisc->prio_bitmap = 0; @@ -827,28 +967,41 @@ __mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port, child_qdisc); child_qdisc->stats_base.backlog = backlog; } + + if (handle == qdisc_state->future_handle && + qdisc_state->future_fifos[tclass]) { + err = mlxsw_sp_qdisc_replace(mlxsw_sp_port, TC_H_UNSPEC, + child_qdisc, + &mlxsw_sp_qdisc_ops_fifo, + NULL); + if (err) + return err; + } } for (; band < IEEE_8021QAZ_MAX_TCS; band++) { tclass = MLXSW_SP_PRIO_BAND_TO_TCLASS(band); - child_qdisc = &mlxsw_sp_port->tclass_qdiscs[tclass]; + child_qdisc = &qdisc_state->tclass_qdiscs[tclass]; child_qdisc->prio_bitmap = 0; mlxsw_sp_qdisc_destroy(mlxsw_sp_port, child_qdisc); mlxsw_sp_port_ets_set(mlxsw_sp_port, MLXSW_REG_QEEC_HR_SUBGROUP, tclass, 0, false, 0); } + + qdisc_state->future_handle = TC_H_UNSPEC; + memset(qdisc_state->future_fifos, 0, sizeof(qdisc_state->future_fifos)); return 0; } static int -mlxsw_sp_qdisc_prio_replace(struct mlxsw_sp_port *mlxsw_sp_port, +mlxsw_sp_qdisc_prio_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, void *params) { struct tc_prio_qopt_offload_params *p = params; unsigned int zeroes[TCQ_ETS_MAX_BANDS] = {0}; - return __mlxsw_sp_qdisc_ets_replace(mlxsw_sp_port, p->bands, + return __mlxsw_sp_qdisc_ets_replace(mlxsw_sp_port, handle, p->bands, zeroes, zeroes, p->priomap); } @@ -880,6 +1033,7 @@ mlxsw_sp_qdisc_get_prio_stats(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, struct tc_qopt_offload_stats *stats_ptr) { + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; struct mlxsw_sp_qdisc *tc_qdisc; u64 tx_packets = 0; u64 tx_bytes = 0; @@ -888,7 +1042,7 @@ mlxsw_sp_qdisc_get_prio_stats(struct mlxsw_sp_port *mlxsw_sp_port, int i; for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { - tc_qdisc = &mlxsw_sp_port->tclass_qdiscs[i]; + tc_qdisc = &qdisc_state->tclass_qdiscs[i]; mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, tc_qdisc, &tx_bytes, &tx_packets, &drops, &backlog); @@ -946,13 +1100,13 @@ mlxsw_sp_qdisc_ets_check_params(struct mlxsw_sp_port *mlxsw_sp_port, } static int -mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port, +mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, void *params) { struct tc_ets_qopt_offload_replace_params *p = params; - return __mlxsw_sp_qdisc_ets_replace(mlxsw_sp_port, p->bands, + return __mlxsw_sp_qdisc_ets_replace(mlxsw_sp_port, handle, p->bands, p->quanta, p->weights, p->priomap); } @@ -1014,11 +1168,12 @@ __mlxsw_sp_qdisc_ets_graft(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, u8 band, u32 child_handle) { + struct mlxsw_sp_qdisc_state *qdisc_state = mlxsw_sp_port->qdisc; int tclass_num = MLXSW_SP_PRIO_BAND_TO_TCLASS(band); struct mlxsw_sp_qdisc *old_qdisc; if (band < IEEE_8021QAZ_MAX_TCS && - mlxsw_sp_port->tclass_qdiscs[tclass_num].handle == child_handle) + qdisc_state->tclass_qdiscs[tclass_num].handle == child_handle) return 0; if (!child_handle) { @@ -1037,7 +1192,7 @@ __mlxsw_sp_qdisc_ets_graft(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_qdisc_destroy(mlxsw_sp_port, old_qdisc); mlxsw_sp_qdisc_destroy(mlxsw_sp_port, - &mlxsw_sp_port->tclass_qdiscs[tclass_num]); + &qdisc_state->tclass_qdiscs[tclass_num]); return -EOPNOTSUPP; } @@ -1119,37 +1274,23 @@ int mlxsw_sp_setup_tc_ets(struct mlxsw_sp_port *mlxsw_sp_port, int mlxsw_sp_tc_qdisc_init(struct mlxsw_sp_port *mlxsw_sp_port) { - struct mlxsw_sp_qdisc *mlxsw_sp_qdisc; + struct mlxsw_sp_qdisc_state *qdisc_state; int i; - mlxsw_sp_qdisc = kzalloc(sizeof(*mlxsw_sp_qdisc), GFP_KERNEL); - if (!mlxsw_sp_qdisc) - goto err_root_qdisc_init; - - mlxsw_sp_port->root_qdisc = mlxsw_sp_qdisc; - mlxsw_sp_port->root_qdisc->prio_bitmap = 0xff; - mlxsw_sp_port->root_qdisc->tclass_num = MLXSW_SP_PORT_DEFAULT_TCLASS; + qdisc_state = kzalloc(sizeof(*qdisc_state), GFP_KERNEL); + if (!qdisc_state) + return -ENOMEM; - mlxsw_sp_qdisc = kcalloc(IEEE_8021QAZ_MAX_TCS, - sizeof(*mlxsw_sp_qdisc), - GFP_KERNEL); - if (!mlxsw_sp_qdisc) - goto err_tclass_qdiscs_init; - - mlxsw_sp_port->tclass_qdiscs = mlxsw_sp_qdisc; + qdisc_state->root_qdisc.prio_bitmap = 0xff; + qdisc_state->root_qdisc.tclass_num = MLXSW_SP_PORT_DEFAULT_TCLASS; for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) - mlxsw_sp_port->tclass_qdiscs[i].tclass_num = i; + qdisc_state->tclass_qdiscs[i].tclass_num = i; + mlxsw_sp_port->qdisc = qdisc_state; return 0; - -err_tclass_qdiscs_init: - kfree(mlxsw_sp_port->root_qdisc); -err_root_qdisc_init: - return -ENOMEM; } void mlxsw_sp_tc_qdisc_fini(struct mlxsw_sp_port *mlxsw_sp_port) { - kfree(mlxsw_sp_port->tclass_qdiscs); - kfree(mlxsw_sp_port->root_qdisc); + kfree(mlxsw_sp_port->qdisc); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 4a77b511ead2..d5bca1be3ef5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -17,6 +17,7 @@ #include <linux/refcount.h> #include <linux/jhash.h> #include <linux/net_namespace.h> +#include <linux/mutex.h> #include <net/netevent.h> #include <net/neighbour.h> #include <net/arp.h> @@ -48,39 +49,6 @@ struct mlxsw_sp_vr; struct mlxsw_sp_lpm_tree; struct mlxsw_sp_rif_ops; -struct mlxsw_sp_router { - struct mlxsw_sp *mlxsw_sp; - struct mlxsw_sp_rif **rifs; - struct mlxsw_sp_vr *vrs; - struct rhashtable neigh_ht; - struct rhashtable nexthop_group_ht; - struct rhashtable nexthop_ht; - struct list_head nexthop_list; - struct { - /* One tree for each protocol: IPv4 and IPv6 */ - struct mlxsw_sp_lpm_tree *proto_trees[2]; - struct mlxsw_sp_lpm_tree *trees; - unsigned int tree_count; - } lpm; - struct { - struct delayed_work dw; - unsigned long interval; /* ms */ - } neighs_update; - struct delayed_work nexthop_probe_dw; -#define MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL 5000 /* ms */ - struct list_head nexthop_neighs_list; - struct list_head ipip_list; - bool aborted; - struct notifier_block fib_nb; - struct notifier_block netevent_nb; - struct notifier_block inetaddr_nb; - struct notifier_block inet6addr_nb; - const struct mlxsw_sp_rif_ops **rif_ops_arr; - const struct mlxsw_sp_ipip_ops **ipip_ops_arr; - u32 adj_discard_index; - bool adj_discard_index_valid; -}; - struct mlxsw_sp_rif { struct list_head nexthop_list; struct list_head neigh_list; @@ -145,6 +113,9 @@ struct mlxsw_sp_rif_ops { void (*fdb_del)(struct mlxsw_sp_rif *rif, const char *mac); }; +static struct mlxsw_sp_rif * +mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp, + const struct net_device *dev); static void mlxsw_sp_rif_destroy(struct mlxsw_sp_rif *rif); static void mlxsw_sp_lpm_tree_hold(struct mlxsw_sp_lpm_tree *lpm_tree); static void mlxsw_sp_lpm_tree_put(struct mlxsw_sp *mlxsw_sp, @@ -760,13 +731,18 @@ int mlxsw_sp_router_tb_id_vr_id(struct mlxsw_sp *mlxsw_sp, u32 tb_id, u16 *vr_id) { struct mlxsw_sp_vr *vr; + int err = 0; + mutex_lock(&mlxsw_sp->router->lock); vr = mlxsw_sp_vr_find(mlxsw_sp, tb_id); - if (!vr) - return -ESRCH; + if (!vr) { + err = -ESRCH; + goto out; + } *vr_id = vr->id; - - return 0; +out: + mutex_unlock(&mlxsw_sp->router->lock); + return err; } static struct mlxsw_sp_fib *mlxsw_sp_vr_fib(const struct mlxsw_sp_vr *vr, @@ -988,17 +964,23 @@ __mlxsw_sp_ipip_netdev_ul_dev_get(const struct net_device *ol_dev) struct ip_tunnel *tun = netdev_priv(ol_dev); struct net *net = dev_net(ol_dev); - return __dev_get_by_index(net, tun->parms.link); + return dev_get_by_index_rcu(net, tun->parms.link); } u32 mlxsw_sp_ipip_dev_ul_tb_id(const struct net_device *ol_dev) { - struct net_device *d = __mlxsw_sp_ipip_netdev_ul_dev_get(ol_dev); + struct net_device *d; + u32 tb_id; + rcu_read_lock(); + d = __mlxsw_sp_ipip_netdev_ul_dev_get(ol_dev); if (d) - return l3mdev_fib_table(d) ? : RT_TABLE_MAIN; + tb_id = l3mdev_fib_table(d) ? : RT_TABLE_MAIN; else - return RT_TABLE_MAIN; + tb_id = RT_TABLE_MAIN; + rcu_read_unlock(); + + return tb_id; } static struct mlxsw_sp_rif * @@ -1230,7 +1212,7 @@ mlxsw_sp_ipip_entry_find_decap(struct mlxsw_sp *mlxsw_sp, saddr_len = 4; saddr_prefix_len = 32; break; - case MLXSW_SP_L3_PROTO_IPV6: + default: WARN_ON(1); return NULL; } @@ -1355,8 +1337,12 @@ mlxsw_sp_ipip_entry_find_by_ul_dev(const struct mlxsw_sp *mlxsw_sp, ipip_list_node); list_for_each_entry_continue(ipip_entry, &mlxsw_sp->router->ipip_list, ipip_list_node) { - struct net_device *ipip_ul_dev = - __mlxsw_sp_ipip_netdev_ul_dev_get(ipip_entry->ol_dev); + struct net_device *ol_dev = ipip_entry->ol_dev; + struct net_device *ipip_ul_dev; + + rcu_read_lock(); + ipip_ul_dev = __mlxsw_sp_ipip_netdev_ul_dev_get(ol_dev); + rcu_read_unlock(); if (ipip_ul_dev == ul_dev) return ipip_entry; @@ -1365,10 +1351,16 @@ mlxsw_sp_ipip_entry_find_by_ul_dev(const struct mlxsw_sp *mlxsw_sp, return NULL; } -bool mlxsw_sp_netdev_is_ipip_ul(const struct mlxsw_sp *mlxsw_sp, +bool mlxsw_sp_netdev_is_ipip_ul(struct mlxsw_sp *mlxsw_sp, const struct net_device *dev) { - return mlxsw_sp_ipip_entry_find_by_ul_dev(mlxsw_sp, dev, NULL); + bool is_ipip_ul; + + mutex_lock(&mlxsw_sp->router->lock); + is_ipip_ul = mlxsw_sp_ipip_entry_find_by_ul_dev(mlxsw_sp, dev, NULL); + mutex_unlock(&mlxsw_sp->router->lock); + + return is_ipip_ul; } static bool mlxsw_sp_netdevice_ipip_can_offload(struct mlxsw_sp *mlxsw_sp, @@ -1388,9 +1380,9 @@ static bool mlxsw_sp_netdevice_ipip_can_offload(struct mlxsw_sp *mlxsw_sp, static int mlxsw_sp_netdevice_ipip_ol_reg_event(struct mlxsw_sp *mlxsw_sp, struct net_device *ol_dev) { + enum mlxsw_sp_ipip_type ipipt = MLXSW_SP_IPIP_TYPE_MAX; struct mlxsw_sp_ipip_entry *ipip_entry; enum mlxsw_sp_l3proto ul_proto; - enum mlxsw_sp_ipip_type ipipt; union mlxsw_sp_l3addr saddr; u32 ul_tb_id; @@ -1543,13 +1535,17 @@ static void mlxsw_sp_nexthop_rif_update(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_rif *rif); /** - * Update the offload related to an IPIP entry. This always updates decap, and - * in addition to that it also: - * @recreate_loopback: recreates the associated loopback RIF - * @keep_encap: updates next hops that use the tunnel netdevice. This is only + * __mlxsw_sp_ipip_entry_update_tunnel - Update offload related to IPIP entry. + * @mlxsw_sp: mlxsw_sp. + * @ipip_entry: IPIP entry. + * @recreate_loopback: Recreates the associated loopback RIF. + * @keep_encap: Updates next hops that use the tunnel netdevice. This is only * relevant when recreate_loopback is true. - * @update_nexthops: updates next hops, keeping the current loopback RIF. This + * @update_nexthops: Updates next hops, keeping the current loopback RIF. This * is only relevant when recreate_loopback is false. + * @extack: extack. + * + * Return: Non-zero value on failure. */ int __mlxsw_sp_ipip_entry_update_tunnel(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_ipip_entry *ipip_entry, @@ -1722,9 +1718,12 @@ static void mlxsw_sp_ipip_demote_tunnel_by_ul_netdev(struct mlxsw_sp *mlxsw_sp, list_for_each_entry_safe(ipip_entry, tmp, &mlxsw_sp->router->ipip_list, ipip_list_node) { - struct net_device *ipip_ul_dev = - __mlxsw_sp_ipip_netdev_ul_dev_get(ipip_entry->ol_dev); + struct net_device *ol_dev = ipip_entry->ol_dev; + struct net_device *ipip_ul_dev; + rcu_read_lock(); + ipip_ul_dev = __mlxsw_sp_ipip_netdev_ul_dev_get(ol_dev); + rcu_read_unlock(); if (ipip_ul_dev == ul_dev) mlxsw_sp_ipip_entry_demote_tunnel(mlxsw_sp, ipip_entry); } @@ -1737,35 +1736,41 @@ int mlxsw_sp_netdevice_ipip_ol_event(struct mlxsw_sp *mlxsw_sp, { struct netdev_notifier_changeupper_info *chup; struct netlink_ext_ack *extack; + int err = 0; + mutex_lock(&mlxsw_sp->router->lock); switch (event) { case NETDEV_REGISTER: - return mlxsw_sp_netdevice_ipip_ol_reg_event(mlxsw_sp, ol_dev); + err = mlxsw_sp_netdevice_ipip_ol_reg_event(mlxsw_sp, ol_dev); + break; case NETDEV_UNREGISTER: mlxsw_sp_netdevice_ipip_ol_unreg_event(mlxsw_sp, ol_dev); - return 0; + break; case NETDEV_UP: mlxsw_sp_netdevice_ipip_ol_up_event(mlxsw_sp, ol_dev); - return 0; + break; case NETDEV_DOWN: mlxsw_sp_netdevice_ipip_ol_down_event(mlxsw_sp, ol_dev); - return 0; + break; case NETDEV_CHANGEUPPER: chup = container_of(info, typeof(*chup), info); extack = info->extack; if (netif_is_l3_master(chup->upper_dev)) - return mlxsw_sp_netdevice_ipip_ol_vrf_event(mlxsw_sp, - ol_dev, - extack); - return 0; + err = mlxsw_sp_netdevice_ipip_ol_vrf_event(mlxsw_sp, + ol_dev, + extack); + break; case NETDEV_CHANGE: extack = info->extack; - return mlxsw_sp_netdevice_ipip_ol_change_event(mlxsw_sp, - ol_dev, extack); + err = mlxsw_sp_netdevice_ipip_ol_change_event(mlxsw_sp, + ol_dev, extack); + break; case NETDEV_CHANGEMTU: - return mlxsw_sp_netdevice_ipip_ol_update_mtu(mlxsw_sp, ol_dev); + err = mlxsw_sp_netdevice_ipip_ol_update_mtu(mlxsw_sp, ol_dev); + break; } - return 0; + mutex_unlock(&mlxsw_sp->router->lock); + return err; } static int @@ -1809,8 +1814,9 @@ mlxsw_sp_netdevice_ipip_ul_event(struct mlxsw_sp *mlxsw_sp, struct netdev_notifier_info *info) { struct mlxsw_sp_ipip_entry *ipip_entry = NULL; - int err; + int err = 0; + mutex_lock(&mlxsw_sp->router->lock); while ((ipip_entry = mlxsw_sp_ipip_entry_find_by_ul_dev(mlxsw_sp, ul_dev, ipip_entry))) { @@ -1823,7 +1829,7 @@ mlxsw_sp_netdevice_ipip_ul_event(struct mlxsw_sp *mlxsw_sp, if (err) { mlxsw_sp_ipip_demote_tunnel_by_ul_netdev(mlxsw_sp, ul_dev); - return err; + break; } if (demote_this) { @@ -1840,8 +1846,9 @@ mlxsw_sp_netdevice_ipip_ul_event(struct mlxsw_sp *mlxsw_sp, ipip_entry = prev; } } + mutex_unlock(&mlxsw_sp->router->lock); - return 0; + return err; } int mlxsw_sp_router_nve_promote_decap(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id, @@ -1850,8 +1857,22 @@ int mlxsw_sp_router_nve_promote_decap(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id, u32 tunnel_index) { enum mlxsw_sp_fib_entry_type type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP; + struct mlxsw_sp_router *router = mlxsw_sp->router; struct mlxsw_sp_fib_entry *fib_entry; - int err; + int err = 0; + + mutex_lock(&mlxsw_sp->router->lock); + + if (WARN_ON_ONCE(router->nve_decap_config.valid)) { + err = -EINVAL; + goto out; + } + + router->nve_decap_config.ul_tb_id = ul_tb_id; + router->nve_decap_config.tunnel_index = tunnel_index; + router->nve_decap_config.ul_proto = ul_proto; + router->nve_decap_config.ul_sip = *ul_sip; + router->nve_decap_config.valid = true; /* It is valid to create a tunnel with a local IP and only later * assign this IP address to a local interface @@ -1860,7 +1881,7 @@ int mlxsw_sp_router_nve_promote_decap(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id, ul_proto, ul_sip, type); if (!fib_entry) - return 0; + goto out; fib_entry->decap.tunnel_index = tunnel_index; fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP; @@ -1869,11 +1890,13 @@ int mlxsw_sp_router_nve_promote_decap(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id, if (err) goto err_fib_entry_update; - return 0; + goto out; err_fib_entry_update: fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP; mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); +out: + mutex_unlock(&mlxsw_sp->router->lock); return err; } @@ -1882,16 +1905,40 @@ void mlxsw_sp_router_nve_demote_decap(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id, const union mlxsw_sp_l3addr *ul_sip) { enum mlxsw_sp_fib_entry_type type = MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP; + struct mlxsw_sp_router *router = mlxsw_sp->router; struct mlxsw_sp_fib_entry *fib_entry; + mutex_lock(&mlxsw_sp->router->lock); + + if (WARN_ON_ONCE(!router->nve_decap_config.valid)) + goto out; + + router->nve_decap_config.valid = false; + fib_entry = mlxsw_sp_router_ip2me_fib_entry_find(mlxsw_sp, ul_tb_id, ul_proto, ul_sip, type); if (!fib_entry) - return; + goto out; fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP; mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); +out: + mutex_unlock(&mlxsw_sp->router->lock); +} + +static bool mlxsw_sp_router_nve_is_decap(struct mlxsw_sp *mlxsw_sp, + u32 ul_tb_id, + enum mlxsw_sp_l3proto ul_proto, + const union mlxsw_sp_l3addr *ul_sip) +{ + struct mlxsw_sp_router *router = mlxsw_sp->router; + + return router->nve_decap_config.valid && + router->nve_decap_config.ul_tb_id == ul_tb_id && + router->nve_decap_config.ul_proto == ul_proto && + !memcmp(&router->nve_decap_config.ul_sip, ul_sip, + sizeof(*ul_sip)); } struct mlxsw_sp_neigh_key { @@ -2264,10 +2311,8 @@ __mlxsw_sp_router_neighs_update_rauhtd(struct mlxsw_sp *mlxsw_sp, int i, num_rec; int err; - /* Make sure the neighbour's netdev isn't removed in the - * process. - */ - rtnl_lock(); + /* Ensure the RIF we read from the device does not change mid-dump. */ + mutex_lock(&mlxsw_sp->router->lock); do { mlxsw_reg_rauhtd_pack(rauhtd_pl, type); err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(rauhtd), @@ -2281,7 +2326,7 @@ __mlxsw_sp_router_neighs_update_rauhtd(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_router_neigh_rec_process(mlxsw_sp, rauhtd_pl, i); } while (mlxsw_sp_router_rauhtd_is_full(rauhtd_pl)); - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); return err; } @@ -2312,15 +2357,14 @@ static void mlxsw_sp_router_neighs_update_nh(struct mlxsw_sp *mlxsw_sp) { struct mlxsw_sp_neigh_entry *neigh_entry; - /* Take RTNL mutex here to prevent lists from changes */ - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); list_for_each_entry(neigh_entry, &mlxsw_sp->router->nexthop_neighs_list, nexthop_neighs_list_node) /* If this neigh have nexthops, make the kernel think this neigh * is active regardless of the traffic. */ neigh_event_send(neigh_entry->key.n, NULL); - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); } static void @@ -2360,15 +2404,13 @@ static void mlxsw_sp_router_probe_unresolved_nexthops(struct work_struct *work) * the nexthop wouldn't get offloaded until the neighbor is resolved * but it wouldn't get resolved ever in case traffic is flowing in HW * using different nexthop. - * - * Take RTNL mutex here to prevent lists from changes. */ - rtnl_lock(); + mutex_lock(&router->lock); list_for_each_entry(neigh_entry, &router->nexthop_neighs_list, nexthop_neighs_list_node) if (!neigh_entry->connected) neigh_event_send(neigh_entry->key.n, NULL); - rtnl_unlock(); + mutex_unlock(&router->lock); mlxsw_core_schedule_dw(&router->nexthop_probe_dw, MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL); @@ -2506,7 +2548,7 @@ static void mlxsw_sp_router_neigh_event_work(struct work_struct *work) dead = n->dead; read_unlock_bh(&n->lock); - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); mlxsw_sp_span_respin(mlxsw_sp); entry_connected = nud_state & NUD_VALID && !dead; @@ -2528,7 +2570,7 @@ static void mlxsw_sp_router_neigh_event_work(struct work_struct *work) mlxsw_sp_neigh_entry_destroy(mlxsw_sp, neigh_entry); out: - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); neigh_release(n); kfree(net_work); } @@ -3189,7 +3231,6 @@ mlxsw_sp_nexthop_group_update(struct mlxsw_sp *mlxsw_sp, u32 adj_index = nh_grp->adj_index; /* base */ struct mlxsw_sp_nexthop *nh; int i; - int err; for (i = 0; i < nh_grp->count; i++) { nh = &nh_grp->nexthops[i]; @@ -3200,6 +3241,8 @@ mlxsw_sp_nexthop_group_update(struct mlxsw_sp *mlxsw_sp, } if (nh->update || reallocate) { + int err = 0; + switch (nh->type) { case MLXSW_SP_NEXTHOP_TYPE_ETH: err = mlxsw_sp_nexthop_update @@ -3711,9 +3754,15 @@ static void mlxsw_sp_nexthop_neigh_fini(struct mlxsw_sp *mlxsw_sp, static bool mlxsw_sp_ipip_netdev_ul_up(struct net_device *ol_dev) { - struct net_device *ul_dev = __mlxsw_sp_ipip_netdev_ul_dev_get(ol_dev); + struct net_device *ul_dev; + bool is_up; + + rcu_read_lock(); + ul_dev = __mlxsw_sp_ipip_netdev_ul_dev_get(ol_dev); + is_up = ul_dev ? (ul_dev->flags & IFF_UP) : true; + rcu_read_unlock(); - return ul_dev ? (ul_dev->flags & IFF_UP) : true; + return is_up; } static void mlxsw_sp_nexthop_ipip_init(struct mlxsw_sp *mlxsw_sp, @@ -3840,10 +3889,14 @@ static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp, if (!dev) return 0; - in_dev = __in_dev_get_rtnl(dev); + rcu_read_lock(); + in_dev = __in_dev_get_rcu(dev); if (in_dev && IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) && - fib_nh->fib_nh_flags & RTNH_F_LINKDOWN) + fib_nh->fib_nh_flags & RTNH_F_LINKDOWN) { + rcu_read_unlock(); return 0; + } + rcu_read_unlock(); err = mlxsw_sp_nexthop4_type_init(mlxsw_sp, nh, fib_nh); if (err) @@ -4473,6 +4526,7 @@ mlxsw_sp_fib4_entry_type_set(struct mlxsw_sp *mlxsw_sp, { struct net_device *dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev; union mlxsw_sp_l3addr dip = { .addr4 = htonl(fen_info->dst) }; + struct mlxsw_sp_router *router = mlxsw_sp->router; u32 tb_id = mlxsw_sp_fix_tb_id(fen_info->tb_id); struct mlxsw_sp_ipip_entry *ipip_entry; struct fib_info *fi = fen_info->fi; @@ -4487,12 +4541,13 @@ mlxsw_sp_fib4_entry_type_set(struct mlxsw_sp *mlxsw_sp, fib_entry, ipip_entry); } - if (mlxsw_sp_nve_ipv4_route_is_decap(mlxsw_sp, tb_id, - dip.addr4)) { - u32 t_index; + if (mlxsw_sp_router_nve_is_decap(mlxsw_sp, tb_id, + MLXSW_SP_L3_PROTO_IPV4, + &dip)) { + u32 tunnel_index; - t_index = mlxsw_sp_nve_decap_tunnel_index_get(mlxsw_sp); - fib_entry->decap.tunnel_index = t_index; + tunnel_index = router->nve_decap_config.tunnel_index; + fib_entry->decap.tunnel_index = tunnel_index; fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP; return 0; } @@ -5923,8 +5978,7 @@ static void mlxsw_sp_router_fib4_event_work(struct work_struct *work) struct mlxsw_sp *mlxsw_sp = fib_work->mlxsw_sp; int err; - /* Protect internal structures from changes */ - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); mlxsw_sp_span_respin(mlxsw_sp); switch (fib_work->event) { @@ -5946,7 +6000,7 @@ static void mlxsw_sp_router_fib4_event_work(struct work_struct *work) fib_info_put(fib_work->fnh_info.fib_nh->nh_parent); break; } - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); kfree(fib_work); } @@ -5957,7 +6011,7 @@ static void mlxsw_sp_router_fib6_event_work(struct work_struct *work) struct mlxsw_sp *mlxsw_sp = fib_work->mlxsw_sp; int err; - rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); mlxsw_sp_span_respin(mlxsw_sp); switch (fib_work->event) { @@ -5984,7 +6038,7 @@ static void mlxsw_sp_router_fib6_event_work(struct work_struct *work) mlxsw_sp_router_fib6_work_fini(&fib_work->fib6_work); break; } - rtnl_unlock(); + mutex_unlock(&mlxsw_sp->router->lock); kfree(fib_work); } @@ -5997,6 +6051,7 @@ static void mlxsw_sp_router_fibmr_event_work(struct work_struct *work) int err; rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); switch (fib_work->event) { case FIB_EVENT_ENTRY_REPLACE: /* fall through */ case FIB_EVENT_ENTRY_ADD: @@ -6025,6 +6080,7 @@ static void mlxsw_sp_router_fibmr_event_work(struct work_struct *work) dev_put(fib_work->ven_info.dev); break; } + mutex_unlock(&mlxsw_sp->router->lock); rtnl_unlock(); kfree(fib_work); } @@ -6233,7 +6289,7 @@ err_fib_event: return NOTIFY_BAD; } -struct mlxsw_sp_rif * +static struct mlxsw_sp_rif * mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp, const struct net_device *dev) { @@ -6247,6 +6303,41 @@ mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp, return NULL; } +bool mlxsw_sp_rif_exists(struct mlxsw_sp *mlxsw_sp, + const struct net_device *dev) +{ + struct mlxsw_sp_rif *rif; + + mutex_lock(&mlxsw_sp->router->lock); + rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); + mutex_unlock(&mlxsw_sp->router->lock); + + return rif; +} + +u16 mlxsw_sp_rif_vid(struct mlxsw_sp *mlxsw_sp, const struct net_device *dev) +{ + struct mlxsw_sp_rif *rif; + u16 vid = 0; + + mutex_lock(&mlxsw_sp->router->lock); + rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); + if (!rif) + goto out; + + /* We only return the VID for VLAN RIFs. Otherwise we return an + * invalid value (0). + */ + if (rif->ops->type != MLXSW_SP_RIF_TYPE_VLAN) + goto out; + + vid = mlxsw_sp_fid_8021q_vid(rif->fid); + +out: + mutex_unlock(&mlxsw_sp->router->lock); + return vid; +} + static int mlxsw_sp_router_rif_disable(struct mlxsw_sp *mlxsw_sp, u16 rif) { char ritr_pl[MLXSW_REG_RITR_LEN]; @@ -6281,7 +6372,8 @@ mlxsw_sp_rif_should_config(struct mlxsw_sp_rif *rif, struct net_device *dev, case NETDEV_UP: return rif == NULL; case NETDEV_DOWN: - idev = __in_dev_get_rtnl(dev); + rcu_read_lock(); + idev = __in_dev_get_rcu(dev); if (idev && idev->ifa_list) addr_list_empty = false; @@ -6289,6 +6381,7 @@ mlxsw_sp_rif_should_config(struct mlxsw_sp_rif *rif, struct net_device *dev, if (addr_list_empty && inet6_dev && !list_empty(&inet6_dev->addr_list)) addr_list_empty = false; + rcu_read_unlock(); /* macvlans do not have a RIF, but rather piggy back on the * RIF of their lower device. @@ -6411,11 +6504,6 @@ const struct net_device *mlxsw_sp_rif_dev(const struct mlxsw_sp_rif *rif) return rif->dev; } -struct mlxsw_sp_fid *mlxsw_sp_rif_fid(const struct mlxsw_sp_rif *rif) -{ - return rif->fid; -} - static struct mlxsw_sp_rif * mlxsw_sp_rif_create(struct mlxsw_sp *mlxsw_sp, const struct mlxsw_sp_rif_params *params, @@ -6528,10 +6616,13 @@ void mlxsw_sp_rif_destroy_by_dev(struct mlxsw_sp *mlxsw_sp, { struct mlxsw_sp_rif *rif; + mutex_lock(&mlxsw_sp->router->lock); rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); if (!rif) - return; + goto out; mlxsw_sp_rif_destroy(rif); +out: + mutex_unlock(&mlxsw_sp->router->lock); } static void @@ -6631,8 +6722,8 @@ err_fid_port_vid_map: return err; } -void -mlxsw_sp_port_vlan_router_leave(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan) +static void +__mlxsw_sp_port_vlan_router_leave(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan) { struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp_port_vlan->mlxsw_sp_port; struct mlxsw_sp_fid *fid = mlxsw_sp_port_vlan->fid; @@ -6650,6 +6741,16 @@ mlxsw_sp_port_vlan_router_leave(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan) mlxsw_sp_rif_subport_put(rif); } +void +mlxsw_sp_port_vlan_router_leave(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port_vlan->mlxsw_sp_port->mlxsw_sp; + + mutex_lock(&mlxsw_sp->router->lock); + __mlxsw_sp_port_vlan_router_leave(mlxsw_sp_port_vlan); + mutex_unlock(&mlxsw_sp->router->lock); +} + static int mlxsw_sp_inetaddr_port_vlan_event(struct net_device *l3_dev, struct net_device *port_dev, unsigned long event, u16 vid, @@ -6667,7 +6768,7 @@ static int mlxsw_sp_inetaddr_port_vlan_event(struct net_device *l3_dev, return mlxsw_sp_port_vlan_router_join(mlxsw_sp_port_vlan, l3_dev, extack); case NETDEV_DOWN: - mlxsw_sp_port_vlan_router_leave(mlxsw_sp_port_vlan); + __mlxsw_sp_port_vlan_router_leave(mlxsw_sp_port_vlan); break; } @@ -6848,8 +6949,8 @@ err_rif_vrrp_add: return err; } -void mlxsw_sp_rif_macvlan_del(struct mlxsw_sp *mlxsw_sp, - const struct net_device *macvlan_dev) +static void __mlxsw_sp_rif_macvlan_del(struct mlxsw_sp *mlxsw_sp, + const struct net_device *macvlan_dev) { struct macvlan_dev *vlan = netdev_priv(macvlan_dev); struct mlxsw_sp_rif *rif; @@ -6866,6 +6967,14 @@ void mlxsw_sp_rif_macvlan_del(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_fid_index(rif->fid), false); } +void mlxsw_sp_rif_macvlan_del(struct mlxsw_sp *mlxsw_sp, + const struct net_device *macvlan_dev) +{ + mutex_lock(&mlxsw_sp->router->lock); + __mlxsw_sp_rif_macvlan_del(mlxsw_sp, macvlan_dev); + mutex_unlock(&mlxsw_sp->router->lock); +} + static int mlxsw_sp_inetaddr_macvlan_event(struct mlxsw_sp *mlxsw_sp, struct net_device *macvlan_dev, unsigned long event, @@ -6875,7 +6984,7 @@ static int mlxsw_sp_inetaddr_macvlan_event(struct mlxsw_sp *mlxsw_sp, case NETDEV_UP: return mlxsw_sp_rif_macvlan_add(mlxsw_sp, macvlan_dev, extack); case NETDEV_DOWN: - mlxsw_sp_rif_macvlan_del(mlxsw_sp, macvlan_dev); + __mlxsw_sp_rif_macvlan_del(mlxsw_sp, macvlan_dev); break; } @@ -6945,15 +7054,17 @@ static int mlxsw_sp_inetaddr_event(struct notifier_block *nb, /* NETDEV_UP event is handled by mlxsw_sp_inetaddr_valid_event */ if (event == NETDEV_UP) - goto out; + return NOTIFY_DONE; router = container_of(nb, struct mlxsw_sp_router, inetaddr_nb); + mutex_lock(&router->lock); rif = mlxsw_sp_rif_find_by_dev(router->mlxsw_sp, dev); if (!mlxsw_sp_rif_should_config(rif, dev, event)) goto out; err = __mlxsw_sp_inetaddr_event(router->mlxsw_sp, dev, event, NULL); out: + mutex_unlock(&router->lock); return notifier_from_errno(err); } @@ -6968,8 +7079,9 @@ int mlxsw_sp_inetaddr_valid_event(struct notifier_block *unused, mlxsw_sp = mlxsw_sp_lower_get(dev); if (!mlxsw_sp) - goto out; + return NOTIFY_DONE; + mutex_lock(&mlxsw_sp->router->lock); rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); if (!mlxsw_sp_rif_should_config(rif, dev, event)) goto out; @@ -6981,6 +7093,7 @@ int mlxsw_sp_inetaddr_valid_event(struct notifier_block *unused, err = __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, ivi->extack); out: + mutex_unlock(&mlxsw_sp->router->lock); return notifier_from_errno(err); } @@ -7001,6 +7114,7 @@ static void mlxsw_sp_inet6addr_event_work(struct work_struct *work) struct mlxsw_sp_rif *rif; rtnl_lock(); + mutex_lock(&mlxsw_sp->router->lock); rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); if (!mlxsw_sp_rif_should_config(rif, dev, event)) @@ -7008,6 +7122,7 @@ static void mlxsw_sp_inet6addr_event_work(struct work_struct *work) __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, NULL); out: + mutex_unlock(&mlxsw_sp->router->lock); rtnl_unlock(); dev_put(dev); kfree(inet6addr_work); @@ -7052,8 +7167,9 @@ int mlxsw_sp_inet6addr_valid_event(struct notifier_block *unused, mlxsw_sp = mlxsw_sp_lower_get(dev); if (!mlxsw_sp) - goto out; + return NOTIFY_DONE; + mutex_lock(&mlxsw_sp->router->lock); rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); if (!mlxsw_sp_rif_should_config(rif, dev, event)) goto out; @@ -7065,6 +7181,7 @@ int mlxsw_sp_inet6addr_valid_event(struct notifier_block *unused, err = __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, i6vi->extack); out: + mutex_unlock(&mlxsw_sp->router->lock); return notifier_from_errno(err); } @@ -7151,24 +7268,30 @@ int mlxsw_sp_netdevice_router_port_event(struct net_device *dev, { struct mlxsw_sp *mlxsw_sp; struct mlxsw_sp_rif *rif; + int err = 0; mlxsw_sp = mlxsw_sp_lower_get(dev); if (!mlxsw_sp) return 0; + mutex_lock(&mlxsw_sp->router->lock); rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); if (!rif) - return 0; + goto out; switch (event) { case NETDEV_CHANGEMTU: /* fall through */ case NETDEV_CHANGEADDR: - return mlxsw_sp_router_port_change_event(mlxsw_sp, rif); + err = mlxsw_sp_router_port_change_event(mlxsw_sp, rif); + break; case NETDEV_PRE_CHANGEADDR: - return mlxsw_sp_router_port_pre_changeaddr_event(rif, ptr); + err = mlxsw_sp_router_port_pre_changeaddr_event(rif, ptr); + break; } - return 0; +out: + mutex_unlock(&mlxsw_sp->router->lock); + return err; } static int mlxsw_sp_port_vrf_join(struct mlxsw_sp *mlxsw_sp, @@ -7211,9 +7334,10 @@ int mlxsw_sp_netdevice_vrf_event(struct net_device *l3_dev, unsigned long event, if (!mlxsw_sp || netif_is_macvlan(l3_dev)) return 0; + mutex_lock(&mlxsw_sp->router->lock); switch (event) { case NETDEV_PRECHANGEUPPER: - return 0; + break; case NETDEV_CHANGEUPPER: if (info->linking) { struct netlink_ext_ack *extack; @@ -7225,6 +7349,7 @@ int mlxsw_sp_netdevice_vrf_event(struct net_device *l3_dev, unsigned long event, } break; } + mutex_unlock(&mlxsw_sp->router->lock); return err; } @@ -7351,13 +7476,14 @@ u8 mlxsw_sp_router_port(const struct mlxsw_sp *mlxsw_sp) return mlxsw_core_max_ports(mlxsw_sp->core) + 1; } -static int mlxsw_sp_rif_vlan_configure(struct mlxsw_sp_rif *rif) +static int mlxsw_sp_rif_fid_configure(struct mlxsw_sp_rif *rif) { struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; - u16 vid = mlxsw_sp_fid_8021q_vid(rif->fid); + u16 fid_index = mlxsw_sp_fid_index(rif->fid); int err; - err = mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_VLAN_IF, vid, true); + err = mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_FID_IF, fid_index, + true); if (err) return err; @@ -7386,13 +7512,13 @@ err_fid_bc_flood_set: mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_MC, mlxsw_sp_router_port(mlxsw_sp), false); err_fid_mc_flood_set: - mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_VLAN_IF, vid, false); + mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_FID_IF, fid_index, false); return err; } -static void mlxsw_sp_rif_vlan_deconfigure(struct mlxsw_sp_rif *rif) +static void mlxsw_sp_rif_fid_deconfigure(struct mlxsw_sp_rif *rif) { - u16 vid = mlxsw_sp_fid_8021q_vid(rif->fid); + u16 fid_index = mlxsw_sp_fid_index(rif->fid); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; struct mlxsw_sp_fid *fid = rif->fid; @@ -7404,10 +7530,41 @@ static void mlxsw_sp_rif_vlan_deconfigure(struct mlxsw_sp_rif *rif) mlxsw_sp_router_port(mlxsw_sp), false); mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_MC, mlxsw_sp_router_port(mlxsw_sp), false); - mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_VLAN_IF, vid, false); + mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_FID_IF, fid_index, false); } static struct mlxsw_sp_fid * +mlxsw_sp_rif_fid_fid_get(struct mlxsw_sp_rif *rif, + struct netlink_ext_ack *extack) +{ + return mlxsw_sp_fid_8021d_get(rif->mlxsw_sp, rif->dev->ifindex); +} + +static void mlxsw_sp_rif_fid_fdb_del(struct mlxsw_sp_rif *rif, const char *mac) +{ + struct switchdev_notifier_fdb_info info; + struct net_device *dev; + + dev = br_fdb_find_port(rif->dev, mac, 0); + if (!dev) + return; + + info.addr = mac; + info.vid = 0; + call_switchdev_notifiers(SWITCHDEV_FDB_DEL_TO_BRIDGE, dev, &info.info, + NULL); +} + +static const struct mlxsw_sp_rif_ops mlxsw_sp_rif_fid_ops = { + .type = MLXSW_SP_RIF_TYPE_FID, + .rif_size = sizeof(struct mlxsw_sp_rif), + .configure = mlxsw_sp_rif_fid_configure, + .deconfigure = mlxsw_sp_rif_fid_deconfigure, + .fid_get = mlxsw_sp_rif_fid_fid_get, + .fdb_del = mlxsw_sp_rif_fid_fdb_del, +}; + +static struct mlxsw_sp_fid * mlxsw_sp_rif_vlan_fid_get(struct mlxsw_sp_rif *rif, struct netlink_ext_ack *extack) { @@ -7428,7 +7585,7 @@ mlxsw_sp_rif_vlan_fid_get(struct mlxsw_sp_rif *rif, } } - return mlxsw_sp_bridge_fid_get(rif->mlxsw_sp, br_dev, vid, extack); + return mlxsw_sp_fid_8021q_get(rif->mlxsw_sp, vid); } static void mlxsw_sp_rif_vlan_fdb_del(struct mlxsw_sp_rif *rif, const char *mac) @@ -7449,103 +7606,6 @@ static void mlxsw_sp_rif_vlan_fdb_del(struct mlxsw_sp_rif *rif, const char *mac) NULL); } -static const struct mlxsw_sp_rif_ops mlxsw_sp_rif_vlan_ops = { - .type = MLXSW_SP_RIF_TYPE_VLAN, - .rif_size = sizeof(struct mlxsw_sp_rif), - .configure = mlxsw_sp_rif_vlan_configure, - .deconfigure = mlxsw_sp_rif_vlan_deconfigure, - .fid_get = mlxsw_sp_rif_vlan_fid_get, - .fdb_del = mlxsw_sp_rif_vlan_fdb_del, -}; - -static int mlxsw_sp_rif_fid_configure(struct mlxsw_sp_rif *rif) -{ - struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; - u16 fid_index = mlxsw_sp_fid_index(rif->fid); - int err; - - err = mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_FID_IF, fid_index, - true); - if (err) - return err; - - err = mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_MC, - mlxsw_sp_router_port(mlxsw_sp), true); - if (err) - goto err_fid_mc_flood_set; - - err = mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_BC, - mlxsw_sp_router_port(mlxsw_sp), true); - if (err) - goto err_fid_bc_flood_set; - - err = mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, - mlxsw_sp_fid_index(rif->fid), true); - if (err) - goto err_rif_fdb_op; - - mlxsw_sp_fid_rif_set(rif->fid, rif); - return 0; - -err_rif_fdb_op: - mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_BC, - mlxsw_sp_router_port(mlxsw_sp), false); -err_fid_bc_flood_set: - mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_MC, - mlxsw_sp_router_port(mlxsw_sp), false); -err_fid_mc_flood_set: - mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_FID_IF, fid_index, false); - return err; -} - -static void mlxsw_sp_rif_fid_deconfigure(struct mlxsw_sp_rif *rif) -{ - u16 fid_index = mlxsw_sp_fid_index(rif->fid); - struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; - struct mlxsw_sp_fid *fid = rif->fid; - - mlxsw_sp_fid_rif_set(fid, NULL); - mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, - mlxsw_sp_fid_index(fid), false); - mlxsw_sp_rif_macvlan_flush(rif); - mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_BC, - mlxsw_sp_router_port(mlxsw_sp), false); - mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_MC, - mlxsw_sp_router_port(mlxsw_sp), false); - mlxsw_sp_rif_vlan_fid_op(rif, MLXSW_REG_RITR_FID_IF, fid_index, false); -} - -static struct mlxsw_sp_fid * -mlxsw_sp_rif_fid_fid_get(struct mlxsw_sp_rif *rif, - struct netlink_ext_ack *extack) -{ - return mlxsw_sp_bridge_fid_get(rif->mlxsw_sp, rif->dev, 0, extack); -} - -static void mlxsw_sp_rif_fid_fdb_del(struct mlxsw_sp_rif *rif, const char *mac) -{ - struct switchdev_notifier_fdb_info info; - struct net_device *dev; - - dev = br_fdb_find_port(rif->dev, mac, 0); - if (!dev) - return; - - info.addr = mac; - info.vid = 0; - call_switchdev_notifiers(SWITCHDEV_FDB_DEL_TO_BRIDGE, dev, &info.info, - NULL); -} - -static const struct mlxsw_sp_rif_ops mlxsw_sp_rif_fid_ops = { - .type = MLXSW_SP_RIF_TYPE_FID, - .rif_size = sizeof(struct mlxsw_sp_rif), - .configure = mlxsw_sp_rif_fid_configure, - .deconfigure = mlxsw_sp_rif_fid_deconfigure, - .fid_get = mlxsw_sp_rif_fid_fid_get, - .fdb_del = mlxsw_sp_rif_fid_fdb_del, -}; - static const struct mlxsw_sp_rif_ops mlxsw_sp_rif_vlan_emu_ops = { .type = MLXSW_SP_RIF_TYPE_VLAN, .rif_size = sizeof(struct mlxsw_sp_rif), @@ -7733,28 +7793,32 @@ int mlxsw_sp_router_ul_rif_get(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id, u16 *ul_rif_index) { struct mlxsw_sp_rif *ul_rif; + int err = 0; - ASSERT_RTNL(); - + mutex_lock(&mlxsw_sp->router->lock); ul_rif = mlxsw_sp_ul_rif_get(mlxsw_sp, ul_tb_id, NULL); - if (IS_ERR(ul_rif)) - return PTR_ERR(ul_rif); + if (IS_ERR(ul_rif)) { + err = PTR_ERR(ul_rif); + goto out; + } *ul_rif_index = ul_rif->rif_index; - - return 0; +out: + mutex_unlock(&mlxsw_sp->router->lock); + return err; } void mlxsw_sp_router_ul_rif_put(struct mlxsw_sp *mlxsw_sp, u16 ul_rif_index) { struct mlxsw_sp_rif *ul_rif; - ASSERT_RTNL(); - + mutex_lock(&mlxsw_sp->router->lock); ul_rif = mlxsw_sp->router->rifs[ul_rif_index]; if (WARN_ON(!ul_rif)) - return; + goto out; mlxsw_sp_ul_rif_put(ul_rif); +out: + mutex_unlock(&mlxsw_sp->router->lock); } static int @@ -8004,6 +8068,7 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp, router = kzalloc(sizeof(*mlxsw_sp->router), GFP_KERNEL); if (!router) return -ENOMEM; + mutex_init(&router->lock); mlxsw_sp->router = router; router->mlxsw_sp = mlxsw_sp; @@ -8107,6 +8172,7 @@ err_router_init: err_register_inet6addr_notifier: unregister_inetaddr_notifier(&router->inetaddr_nb); err_register_inetaddr_notifier: + mutex_destroy(&mlxsw_sp->router->lock); kfree(mlxsw_sp->router); return err; } @@ -8127,5 +8193,6 @@ void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) __mlxsw_sp_router_fini(mlxsw_sp); unregister_inet6addr_notifier(&mlxsw_sp->router->inet6addr_nb); unregister_inetaddr_notifier(&mlxsw_sp->router->inetaddr_nb); + mutex_destroy(&mlxsw_sp->router->lock); kfree(mlxsw_sp->router); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h index c9b94f435cdd..8418dc3ae967 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h @@ -7,6 +7,49 @@ #include "spectrum.h" #include "reg.h" +struct mlxsw_sp_router_nve_decap { + u32 ul_tb_id; + u32 tunnel_index; + enum mlxsw_sp_l3proto ul_proto; + union mlxsw_sp_l3addr ul_sip; + u8 valid:1; +}; + +struct mlxsw_sp_router { + struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_rif **rifs; + struct mlxsw_sp_vr *vrs; + struct rhashtable neigh_ht; + struct rhashtable nexthop_group_ht; + struct rhashtable nexthop_ht; + struct list_head nexthop_list; + struct { + /* One tree for each protocol: IPv4 and IPv6 */ + struct mlxsw_sp_lpm_tree *proto_trees[2]; + struct mlxsw_sp_lpm_tree *trees; + unsigned int tree_count; + } lpm; + struct { + struct delayed_work dw; + unsigned long interval; /* ms */ + } neighs_update; + struct delayed_work nexthop_probe_dw; +#define MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL 5000 /* ms */ + struct list_head nexthop_neighs_list; + struct list_head ipip_list; + bool aborted; + struct notifier_block fib_nb; + struct notifier_block netevent_nb; + struct notifier_block inetaddr_nb; + struct notifier_block inet6addr_nb; + const struct mlxsw_sp_rif_ops **rif_ops_arr; + const struct mlxsw_sp_ipip_ops **ipip_ops_arr; + u32 adj_discard_index; + bool adj_discard_index_valid; + struct mlxsw_sp_router_nve_decap nve_decap_config; + struct mutex lock; /* Protects shared router resources */ +}; + struct mlxsw_sp_rif_ipip_lb; struct mlxsw_sp_rif_ipip_lb_config { enum mlxsw_reg_ritr_loopback_ipip_type lb_ipipt; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c index 0cdd7954a085..9fb2e9d93929 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c @@ -3,6 +3,8 @@ #include <linux/if_bridge.h> #include <linux/list.h> +#include <linux/rtnetlink.h> +#include <linux/workqueue.h> #include <net/arp.h> #include <net/gre.h> #include <net/lag.h> @@ -14,38 +16,43 @@ #include "spectrum_span.h" #include "spectrum_switchdev.h" +struct mlxsw_sp_span { + struct work_struct work; + struct mlxsw_sp *mlxsw_sp; + atomic_t active_entries_count; + int entries_count; + struct mlxsw_sp_span_entry entries[0]; +}; + +static void mlxsw_sp_span_respin_work(struct work_struct *work); + static u64 mlxsw_sp_span_occ_get(void *priv) { const struct mlxsw_sp *mlxsw_sp = priv; - u64 occ = 0; - int i; - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { - if (mlxsw_sp->span.entries[i].ref_count) - occ++; - } - - return occ; + return atomic_read(&mlxsw_sp->span->active_entries_count); } int mlxsw_sp_span_init(struct mlxsw_sp *mlxsw_sp) { struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); - int i; + struct mlxsw_sp_span *span; + int i, entries_count; if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, MAX_SPAN)) return -EIO; - mlxsw_sp->span.entries_count = MLXSW_CORE_RES_GET(mlxsw_sp->core, - MAX_SPAN); - mlxsw_sp->span.entries = kcalloc(mlxsw_sp->span.entries_count, - sizeof(struct mlxsw_sp_span_entry), - GFP_KERNEL); - if (!mlxsw_sp->span.entries) + entries_count = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_SPAN); + span = kzalloc(struct_size(span, entries, entries_count), GFP_KERNEL); + if (!span) return -ENOMEM; + span->entries_count = entries_count; + atomic_set(&span->active_entries_count, 0); + span->mlxsw_sp = mlxsw_sp; + mlxsw_sp->span = span; - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; INIT_LIST_HEAD(&curr->bound_ports_list); curr->id = i; @@ -53,6 +60,7 @@ int mlxsw_sp_span_init(struct mlxsw_sp *mlxsw_sp) devlink_resource_occ_get_register(devlink, MLXSW_SP_RESOURCE_SPAN, mlxsw_sp_span_occ_get, mlxsw_sp); + INIT_WORK(&span->work, mlxsw_sp_span_respin_work); return 0; } @@ -62,14 +70,15 @@ void mlxsw_sp_span_fini(struct mlxsw_sp *mlxsw_sp) struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); int i; + cancel_work_sync(&mlxsw_sp->span->work); devlink_resource_occ_get_unregister(devlink, MLXSW_SP_RESOURCE_SPAN); - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; WARN_ON_ONCE(!list_empty(&curr->bound_ports_list)); } - kfree(mlxsw_sp->span.entries); + kfree(mlxsw_sp->span); } static int @@ -645,15 +654,16 @@ mlxsw_sp_span_entry_create(struct mlxsw_sp *mlxsw_sp, int i; /* find a free entry to use */ - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { - if (!mlxsw_sp->span.entries[i].ref_count) { - span_entry = &mlxsw_sp->span.entries[i]; + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { + if (!mlxsw_sp->span->entries[i].ref_count) { + span_entry = &mlxsw_sp->span->entries[i]; break; } } if (!span_entry) return NULL; + atomic_inc(&mlxsw_sp->span->active_entries_count); span_entry->ops = ops; span_entry->ref_count = 1; span_entry->to_dev = to_dev; @@ -662,9 +672,11 @@ mlxsw_sp_span_entry_create(struct mlxsw_sp *mlxsw_sp, return span_entry; } -static void mlxsw_sp_span_entry_destroy(struct mlxsw_sp_span_entry *span_entry) +static void mlxsw_sp_span_entry_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_span_entry *span_entry) { mlxsw_sp_span_entry_deconfigure(span_entry); + atomic_dec(&mlxsw_sp->span->active_entries_count); } struct mlxsw_sp_span_entry * @@ -673,8 +685,8 @@ mlxsw_sp_span_entry_find_by_port(struct mlxsw_sp *mlxsw_sp, { int i; - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; if (curr->ref_count && curr->to_dev == to_dev) return curr; @@ -694,8 +706,8 @@ mlxsw_sp_span_entry_find_by_id(struct mlxsw_sp *mlxsw_sp, int span_id) { int i; - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; if (curr->ref_count && curr->id == span_id) return curr; @@ -726,7 +738,7 @@ static int mlxsw_sp_span_entry_put(struct mlxsw_sp *mlxsw_sp, { WARN_ON(!span_entry->ref_count); if (--span_entry->ref_count == 0) - mlxsw_sp_span_entry_destroy(span_entry); + mlxsw_sp_span_entry_destroy(mlxsw_sp, span_entry); return 0; } @@ -736,8 +748,8 @@ static bool mlxsw_sp_span_is_egress_mirror(struct mlxsw_sp_port *port) struct mlxsw_sp_span_inspected_port *p; int i; - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; list_for_each_entry(p, &curr->bound_ports_list, list) if (p->local_port == port->local_port && @@ -842,9 +854,9 @@ mlxsw_sp_span_inspected_port_add(struct mlxsw_sp_port *port, * so if a binding is requested, check for conflicts. */ if (bind) - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { struct mlxsw_sp_span_entry *curr = - &mlxsw_sp->span.entries[i]; + &mlxsw_sp->span->entries[i]; if (mlxsw_sp_span_entry_bound_port_find(curr, type, port, bind)) @@ -988,14 +1000,18 @@ void mlxsw_sp_span_mirror_del(struct mlxsw_sp_port *from, int span_id, mlxsw_sp_span_inspected_port_del(from, span_entry, type, bind); } -void mlxsw_sp_span_respin(struct mlxsw_sp *mlxsw_sp) +static void mlxsw_sp_span_respin_work(struct work_struct *work) { - int i; - int err; + struct mlxsw_sp_span *span; + struct mlxsw_sp *mlxsw_sp; + int i, err; - ASSERT_RTNL(); - for (i = 0; i < mlxsw_sp->span.entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + span = container_of(work, struct mlxsw_sp_span, work); + mlxsw_sp = span->mlxsw_sp; + + rtnl_lock(); + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; struct mlxsw_sp_span_parms sparms = {NULL}; if (!curr->ref_count) @@ -1010,4 +1026,12 @@ void mlxsw_sp_span_respin(struct mlxsw_sp *mlxsw_sp) mlxsw_sp_span_entry_configure(mlxsw_sp, curr, sparms); } } + rtnl_unlock(); +} + +void mlxsw_sp_span_respin(struct mlxsw_sp *mlxsw_sp) +{ + if (atomic_read(&mlxsw_sp->span->active_entries_count) == 0) + return; + mlxsw_core_schedule_work(&mlxsw_sp->span->work); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index a3af171c6358..a26162b08b7d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -153,16 +153,64 @@ static void mlxsw_sp_bridge_device_rifs_destroy(struct mlxsw_sp *mlxsw_sp, mlxsw_sp); } +static int mlxsw_sp_bridge_device_vxlan_init(struct mlxsw_sp_bridge *bridge, + struct net_device *br_dev, + struct netlink_ext_ack *extack) +{ + struct net_device *dev, *stop_dev; + struct list_head *iter; + int err; + + netdev_for_each_lower_dev(br_dev, dev, iter) { + if (netif_is_vxlan(dev) && netif_running(dev)) { + err = mlxsw_sp_bridge_vxlan_join(bridge->mlxsw_sp, + br_dev, dev, 0, + extack); + if (err) { + stop_dev = dev; + goto err_vxlan_join; + } + } + } + + return 0; + +err_vxlan_join: + netdev_for_each_lower_dev(br_dev, dev, iter) { + if (netif_is_vxlan(dev) && netif_running(dev)) { + if (stop_dev == dev) + break; + mlxsw_sp_bridge_vxlan_leave(bridge->mlxsw_sp, dev); + } + } + return err; +} + +static void mlxsw_sp_bridge_device_vxlan_fini(struct mlxsw_sp_bridge *bridge, + struct net_device *br_dev) +{ + struct net_device *dev; + struct list_head *iter; + + netdev_for_each_lower_dev(br_dev, dev, iter) { + if (netif_is_vxlan(dev) && netif_running(dev)) + mlxsw_sp_bridge_vxlan_leave(bridge->mlxsw_sp, dev); + } +} + static struct mlxsw_sp_bridge_device * mlxsw_sp_bridge_device_create(struct mlxsw_sp_bridge *bridge, - struct net_device *br_dev) + struct net_device *br_dev, + struct netlink_ext_ack *extack) { struct device *dev = bridge->mlxsw_sp->bus_info->dev; struct mlxsw_sp_bridge_device *bridge_device; bool vlan_enabled = br_vlan_enabled(br_dev); + int err; if (vlan_enabled && bridge->vlan_enabled_exists) { dev_err(dev, "Only one VLAN-aware bridge is supported\n"); + NL_SET_ERR_MSG_MOD(extack, "Only one VLAN-aware bridge is supported"); return ERR_PTR(-EINVAL); } @@ -184,13 +232,29 @@ mlxsw_sp_bridge_device_create(struct mlxsw_sp_bridge *bridge, INIT_LIST_HEAD(&bridge_device->mids_list); list_add(&bridge_device->list, &bridge->bridges_list); + /* It is possible we already have VXLAN devices enslaved to the bridge. + * In which case, we need to replay their configuration as if they were + * just now enslaved to the bridge. + */ + err = mlxsw_sp_bridge_device_vxlan_init(bridge, br_dev, extack); + if (err) + goto err_vxlan_init; + return bridge_device; + +err_vxlan_init: + list_del(&bridge_device->list); + if (bridge_device->vlan_enabled) + bridge->vlan_enabled_exists = false; + kfree(bridge_device); + return ERR_PTR(err); } static void mlxsw_sp_bridge_device_destroy(struct mlxsw_sp_bridge *bridge, struct mlxsw_sp_bridge_device *bridge_device) { + mlxsw_sp_bridge_device_vxlan_fini(bridge, bridge_device->dev); mlxsw_sp_bridge_device_rifs_destroy(bridge->mlxsw_sp, bridge_device->dev); list_del(&bridge_device->list); @@ -203,7 +267,8 @@ mlxsw_sp_bridge_device_destroy(struct mlxsw_sp_bridge *bridge, static struct mlxsw_sp_bridge_device * mlxsw_sp_bridge_device_get(struct mlxsw_sp_bridge *bridge, - struct net_device *br_dev) + struct net_device *br_dev, + struct netlink_ext_ack *extack) { struct mlxsw_sp_bridge_device *bridge_device; @@ -211,7 +276,7 @@ mlxsw_sp_bridge_device_get(struct mlxsw_sp_bridge *bridge, if (bridge_device) return bridge_device; - return mlxsw_sp_bridge_device_create(bridge, br_dev); + return mlxsw_sp_bridge_device_create(bridge, br_dev, extack); } static void @@ -292,7 +357,8 @@ mlxsw_sp_bridge_port_destroy(struct mlxsw_sp_bridge_port *bridge_port) static struct mlxsw_sp_bridge_port * mlxsw_sp_bridge_port_get(struct mlxsw_sp_bridge *bridge, - struct net_device *brport_dev) + struct net_device *brport_dev, + struct netlink_ext_ack *extack) { struct net_device *br_dev = netdev_master_upper_dev_get(brport_dev); struct mlxsw_sp_bridge_device *bridge_device; @@ -305,7 +371,7 @@ mlxsw_sp_bridge_port_get(struct mlxsw_sp_bridge *bridge, return bridge_port; } - bridge_device = mlxsw_sp_bridge_device_get(bridge, br_dev); + bridge_device = mlxsw_sp_bridge_device_get(bridge, br_dev, extack); if (IS_ERR(bridge_device)) return ERR_CAST(bridge_device); @@ -1000,7 +1066,7 @@ mlxsw_sp_port_vlan_bridge_join(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan, &bridge_vlan->port_vlan_list); mlxsw_sp_bridge_port_get(mlxsw_sp_port->mlxsw_sp->bridge, - bridge_port->dev); + bridge_port->dev, extack); mlxsw_sp_port_vlan->bridge_port = bridge_port; return 0; @@ -1107,16 +1173,12 @@ mlxsw_sp_br_ban_rif_pvid_change(struct mlxsw_sp *mlxsw_sp, const struct net_device *br_dev, const struct switchdev_obj_port_vlan *vlan) { - struct mlxsw_sp_rif *rif; - struct mlxsw_sp_fid *fid; u16 pvid; u16 vid; - rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, br_dev); - if (!rif) + pvid = mlxsw_sp_rif_vid(mlxsw_sp, br_dev); + if (!pvid) return 0; - fid = mlxsw_sp_rif_fid(rif); - pvid = mlxsw_sp_fid_8021q_vid(fid); for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { if (vlan->flags & BRIDGE_VLAN_INFO_PVID) { @@ -1712,36 +1774,6 @@ mlxsw_sp_port_mrouter_update_mdb(struct mlxsw_sp_port *mlxsw_sp_port, } } -struct mlxsw_sp_span_respin_work { - struct work_struct work; - struct mlxsw_sp *mlxsw_sp; -}; - -static void mlxsw_sp_span_respin_work(struct work_struct *work) -{ - struct mlxsw_sp_span_respin_work *respin_work = - container_of(work, struct mlxsw_sp_span_respin_work, work); - - rtnl_lock(); - mlxsw_sp_span_respin(respin_work->mlxsw_sp); - rtnl_unlock(); - kfree(respin_work); -} - -static void mlxsw_sp_span_respin_schedule(struct mlxsw_sp *mlxsw_sp) -{ - struct mlxsw_sp_span_respin_work *respin_work; - - respin_work = kzalloc(sizeof(*respin_work), GFP_ATOMIC); - if (!respin_work) - return; - - INIT_WORK(&respin_work->work, mlxsw_sp_span_respin_work); - respin_work->mlxsw_sp = mlxsw_sp; - - mlxsw_core_schedule_work(&respin_work->work); -} - static int mlxsw_sp_port_obj_add(struct net_device *dev, const struct switchdev_obj *obj, struct switchdev_trans *trans, @@ -1763,7 +1795,7 @@ static int mlxsw_sp_port_obj_add(struct net_device *dev, * call for later, so that the respin logic sees the * updated bridge state. */ - mlxsw_sp_span_respin_schedule(mlxsw_sp_port->mlxsw_sp); + mlxsw_sp_span_respin(mlxsw_sp_port->mlxsw_sp); } break; case SWITCHDEV_OBJ_ID_PORT_MDB: @@ -1916,7 +1948,7 @@ static int mlxsw_sp_port_obj_del(struct net_device *dev, break; } - mlxsw_sp_span_respin_schedule(mlxsw_sp_port->mlxsw_sp); + mlxsw_sp_span_respin(mlxsw_sp_port->mlxsw_sp); return err; } @@ -1990,12 +2022,11 @@ mlxsw_sp_bridge_8021q_vxlan_join(struct mlxsw_sp_bridge_device *bridge_device, return err; } - /* If no other port is member in the VLAN, then the FID does not exist. - * NVE will be enabled on the FID once a port joins the VLAN - */ - fid = mlxsw_sp_fid_8021q_lookup(mlxsw_sp, vid); - if (!fid) - return 0; + fid = mlxsw_sp_fid_8021q_get(mlxsw_sp, vid); + if (IS_ERR(fid)) { + NL_SET_ERR_MSG_MOD(extack, "Failed to create 802.1Q FID"); + return PTR_ERR(fid); + } if (mlxsw_sp_fid_vni_is_set(fid)) { NL_SET_ERR_MSG_MOD(extack, "VNI is already set on FID"); @@ -2007,11 +2038,6 @@ mlxsw_sp_bridge_8021q_vxlan_join(struct mlxsw_sp_bridge_device *bridge_device, if (err) goto err_nve_fid_enable; - /* The tunnel port does not hold a reference on the FID. Only - * local ports and the router port - */ - mlxsw_sp_fid_put(fid); - return 0; err_nve_fid_enable: @@ -2048,38 +2074,8 @@ mlxsw_sp_bridge_8021q_fid_get(struct mlxsw_sp_bridge_device *bridge_device, u16 vid, struct netlink_ext_ack *extack) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(bridge_device->dev); - struct net_device *vxlan_dev; - struct mlxsw_sp_fid *fid; - int err; - - fid = mlxsw_sp_fid_8021q_get(mlxsw_sp, vid); - if (IS_ERR(fid)) - return fid; - - if (mlxsw_sp_fid_vni_is_set(fid)) - return fid; - - /* Find the VxLAN device that has the specified VLAN configured as - * PVID and egress untagged. There can be at most one such device - */ - vxlan_dev = mlxsw_sp_bridge_8021q_vxlan_dev_find(bridge_device->dev, - vid); - if (!vxlan_dev) - return fid; - - if (!netif_running(vxlan_dev)) - return fid; - - err = mlxsw_sp_bridge_8021q_vxlan_join(bridge_device, vxlan_dev, vid, - extack); - if (err) - goto err_vxlan_join; - - return fid; -err_vxlan_join: - mlxsw_sp_fid_put(fid); - return ERR_PTR(err); + return mlxsw_sp_fid_8021q_get(mlxsw_sp, vid); } static struct mlxsw_sp_fid * @@ -2184,9 +2180,9 @@ mlxsw_sp_bridge_8021d_vxlan_join(struct mlxsw_sp_bridge_device *bridge_device, struct mlxsw_sp_fid *fid; int err; - fid = mlxsw_sp_fid_8021d_lookup(mlxsw_sp, bridge_device->dev->ifindex); - if (!fid) { - NL_SET_ERR_MSG_MOD(extack, "Did not find a corresponding FID"); + fid = mlxsw_sp_fid_8021d_get(mlxsw_sp, bridge_device->dev->ifindex); + if (IS_ERR(fid)) { + NL_SET_ERR_MSG_MOD(extack, "Failed to create 802.1D FID"); return -EINVAL; } @@ -2200,11 +2196,6 @@ mlxsw_sp_bridge_8021d_vxlan_join(struct mlxsw_sp_bridge_device *bridge_device, if (err) goto err_nve_fid_enable; - /* The tunnel port does not hold a reference on the FID. Only - * local ports and the router port - */ - mlxsw_sp_fid_put(fid); - return 0; err_nve_fid_enable: @@ -2218,34 +2209,8 @@ mlxsw_sp_bridge_8021d_fid_get(struct mlxsw_sp_bridge_device *bridge_device, u16 vid, struct netlink_ext_ack *extack) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(bridge_device->dev); - struct net_device *vxlan_dev; - struct mlxsw_sp_fid *fid; - int err; - fid = mlxsw_sp_fid_8021d_get(mlxsw_sp, bridge_device->dev->ifindex); - if (IS_ERR(fid)) - return fid; - - if (mlxsw_sp_fid_vni_is_set(fid)) - return fid; - - vxlan_dev = mlxsw_sp_bridge_vxlan_dev_find(bridge_device->dev); - if (!vxlan_dev) - return fid; - - if (!netif_running(vxlan_dev)) - return fid; - - err = mlxsw_sp_bridge_8021d_vxlan_join(bridge_device, vxlan_dev, 0, - extack); - if (err) - goto err_vxlan_join; - - return fid; - -err_vxlan_join: - mlxsw_sp_fid_put(fid); - return ERR_PTR(err); + return mlxsw_sp_fid_8021d_get(mlxsw_sp, bridge_device->dev->ifindex); } static struct mlxsw_sp_fid * @@ -2287,7 +2252,8 @@ int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_bridge_port *bridge_port; int err; - bridge_port = mlxsw_sp_bridge_port_get(mlxsw_sp->bridge, brport_dev); + bridge_port = mlxsw_sp_bridge_port_get(mlxsw_sp->bridge, brport_dev, + extack); if (IS_ERR(bridge_port)) return PTR_ERR(bridge_port); bridge_device = bridge_port->bridge_device; @@ -2351,21 +2317,11 @@ void mlxsw_sp_bridge_vxlan_leave(struct mlxsw_sp *mlxsw_sp, return; mlxsw_sp_nve_fid_disable(mlxsw_sp, fid); + /* Drop both the reference we just took during lookup and the reference + * the VXLAN device took. + */ + mlxsw_sp_fid_put(fid); mlxsw_sp_fid_put(fid); -} - -struct mlxsw_sp_fid *mlxsw_sp_bridge_fid_get(struct mlxsw_sp *mlxsw_sp, - const struct net_device *br_dev, - u16 vid, - struct netlink_ext_ack *extack) -{ - struct mlxsw_sp_bridge_device *bridge_device; - - bridge_device = mlxsw_sp_bridge_device_find(mlxsw_sp->bridge, br_dev); - if (WARN_ON(!bridge_device)) - return ERR_PTR(-EINVAL); - - return bridge_device->ops->fid_get(bridge_device, vid, extack); } static void @@ -2718,19 +2674,24 @@ static void mlxsw_sp_fdb_notify_rec_process(struct mlxsw_sp *mlxsw_sp, } } -static void mlxsw_sp_fdb_notify_work_schedule(struct mlxsw_sp *mlxsw_sp) +static void mlxsw_sp_fdb_notify_work_schedule(struct mlxsw_sp *mlxsw_sp, + bool no_delay) { struct mlxsw_sp_bridge *bridge = mlxsw_sp->bridge; + unsigned int interval = no_delay ? 0 : bridge->fdb_notify.interval; mlxsw_core_schedule_dw(&bridge->fdb_notify.dw, - msecs_to_jiffies(bridge->fdb_notify.interval)); + msecs_to_jiffies(interval)); } +#define MLXSW_SP_FDB_SFN_QUERIES_PER_SESSION 10 + static void mlxsw_sp_fdb_notify_work(struct work_struct *work) { struct mlxsw_sp_bridge *bridge; struct mlxsw_sp *mlxsw_sp; char *sfn_pl; + int queries; u8 num_rec; int i; int err; @@ -2743,20 +2704,26 @@ static void mlxsw_sp_fdb_notify_work(struct work_struct *work) mlxsw_sp = bridge->mlxsw_sp; rtnl_lock(); - mlxsw_reg_sfn_pack(sfn_pl); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(sfn), sfn_pl); - if (err) { - dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to get FDB notifications\n"); - goto out; + queries = MLXSW_SP_FDB_SFN_QUERIES_PER_SESSION; + while (queries > 0) { + mlxsw_reg_sfn_pack(sfn_pl); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(sfn), sfn_pl); + if (err) { + dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to get FDB notifications\n"); + goto out; + } + num_rec = mlxsw_reg_sfn_num_rec_get(sfn_pl); + for (i = 0; i < num_rec; i++) + mlxsw_sp_fdb_notify_rec_process(mlxsw_sp, sfn_pl, i); + if (num_rec != MLXSW_REG_SFN_REC_MAX_COUNT) + goto out; + queries--; } - num_rec = mlxsw_reg_sfn_num_rec_get(sfn_pl); - for (i = 0; i < num_rec; i++) - mlxsw_sp_fdb_notify_rec_process(mlxsw_sp, sfn_pl, i); out: rtnl_unlock(); kfree(sfn_pl); - mlxsw_sp_fdb_notify_work_schedule(mlxsw_sp); + mlxsw_sp_fdb_notify_work_schedule(mlxsw_sp, !queries); } struct mlxsw_sp_switchdev_event_work { @@ -3502,7 +3469,7 @@ static int mlxsw_sp_fdb_init(struct mlxsw_sp *mlxsw_sp) INIT_DELAYED_WORK(&bridge->fdb_notify.dw, mlxsw_sp_fdb_notify_work); bridge->fdb_notify.interval = MLXSW_SP_DEFAULT_LEARNING_INTERVAL; - mlxsw_sp_fdb_notify_work_schedule(mlxsw_sp); + mlxsw_sp_fdb_notify_work_schedule(mlxsw_sp, false); return 0; err_register_switchdev_blocking_notifier: diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c index 60205aa3f6a5..9096ffd89e50 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c @@ -1,13 +1,16 @@ // SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 /* Copyright (c) 2019 Mellanox Technologies. All rights reserved */ +#include <linux/bitops.h> #include <linux/kernel.h> +#include <linux/netlink.h> #include <net/devlink.h> #include <uapi/linux/devlink.h> #include "core.h" #include "reg.h" #include "spectrum.h" +#include "spectrum_trap.h" /* All driver-specific traps must be documented in * Documentation/networking/devlink/mlxsw.rst @@ -25,36 +28,166 @@ enum { #define MLXSW_SP_TRAP_METADATA DEVLINK_TRAP_METADATA_TYPE_F_IN_PORT +static int mlxsw_sp_rx_listener(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb, + u8 local_port, + struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp_port_pcpu_stats *pcpu_stats; + + if (unlikely(!mlxsw_sp_port)) { + dev_warn_ratelimited(mlxsw_sp->bus_info->dev, "Port %d: skb received for non-existent port\n", + local_port); + kfree_skb(skb); + return -EINVAL; + } + + skb->dev = mlxsw_sp_port->dev; + + pcpu_stats = this_cpu_ptr(mlxsw_sp_port->pcpu_stats); + u64_stats_update_begin(&pcpu_stats->syncp); + pcpu_stats->rx_packets++; + pcpu_stats->rx_bytes += skb->len; + u64_stats_update_end(&pcpu_stats->syncp); + + skb->protocol = eth_type_trans(skb, skb->dev); + + return 0; +} + static void mlxsw_sp_rx_drop_listener(struct sk_buff *skb, u8 local_port, - void *priv); + void *trap_ctx) +{ + struct devlink_port *in_devlink_port; + struct mlxsw_sp_port *mlxsw_sp_port; + struct mlxsw_sp *mlxsw_sp; + struct devlink *devlink; + int err; + + mlxsw_sp = devlink_trap_ctx_priv(trap_ctx); + mlxsw_sp_port = mlxsw_sp->ports[local_port]; + + err = mlxsw_sp_rx_listener(mlxsw_sp, skb, local_port, mlxsw_sp_port); + if (err) + return; + + devlink = priv_to_devlink(mlxsw_sp->core); + in_devlink_port = mlxsw_core_port_devlink_port_get(mlxsw_sp->core, + local_port); + skb_push(skb, ETH_HLEN); + devlink_trap_report(devlink, skb, trap_ctx, in_devlink_port, NULL); + consume_skb(skb); +} + +static void mlxsw_sp_rx_acl_drop_listener(struct sk_buff *skb, u8 local_port, + void *trap_ctx) +{ + u32 cookie_index = mlxsw_skb_cb(skb)->cookie_index; + const struct flow_action_cookie *fa_cookie; + struct devlink_port *in_devlink_port; + struct mlxsw_sp_port *mlxsw_sp_port; + struct mlxsw_sp *mlxsw_sp; + struct devlink *devlink; + int err; + + mlxsw_sp = devlink_trap_ctx_priv(trap_ctx); + mlxsw_sp_port = mlxsw_sp->ports[local_port]; + + err = mlxsw_sp_rx_listener(mlxsw_sp, skb, local_port, mlxsw_sp_port); + if (err) + return; + + devlink = priv_to_devlink(mlxsw_sp->core); + in_devlink_port = mlxsw_core_port_devlink_port_get(mlxsw_sp->core, + local_port); + skb_push(skb, ETH_HLEN); + rcu_read_lock(); + fa_cookie = mlxsw_sp_acl_act_cookie_lookup(mlxsw_sp, cookie_index); + devlink_trap_report(devlink, skb, trap_ctx, in_devlink_port, fa_cookie); + rcu_read_unlock(); + consume_skb(skb); +} + static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, - void *trap_ctx); + void *trap_ctx) +{ + struct devlink_port *in_devlink_port; + struct mlxsw_sp_port *mlxsw_sp_port; + struct mlxsw_sp *mlxsw_sp; + struct devlink *devlink; + int err; + + mlxsw_sp = devlink_trap_ctx_priv(trap_ctx); + mlxsw_sp_port = mlxsw_sp->ports[local_port]; + + err = mlxsw_sp_rx_listener(mlxsw_sp, skb, local_port, mlxsw_sp_port); + if (err) + return; + + devlink = priv_to_devlink(mlxsw_sp->core); + in_devlink_port = mlxsw_core_port_devlink_port_get(mlxsw_sp->core, + local_port); + skb_push(skb, ETH_HLEN); + devlink_trap_report(devlink, skb, trap_ctx, in_devlink_port, NULL); + skb_pull(skb, ETH_HLEN); + skb->offload_fwd_mark = 1; + netif_receive_skb(skb); +} #define MLXSW_SP_TRAP_DROP(_id, _group_id) \ DEVLINK_TRAP_GENERIC(DROP, DROP, _id, \ - DEVLINK_TRAP_GROUP_GENERIC(_group_id), \ + DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, \ MLXSW_SP_TRAP_METADATA) +#define MLXSW_SP_TRAP_DROP_EXT(_id, _group_id, _metadata) \ + DEVLINK_TRAP_GENERIC(DROP, DROP, _id, \ + DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, \ + MLXSW_SP_TRAP_METADATA | (_metadata)) + #define MLXSW_SP_TRAP_DRIVER_DROP(_id, _group_id) \ DEVLINK_TRAP_DRIVER(DROP, DROP, DEVLINK_MLXSW_TRAP_ID_##_id, \ DEVLINK_MLXSW_TRAP_NAME_##_id, \ - DEVLINK_TRAP_GROUP_GENERIC(_group_id), \ + DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, \ MLXSW_SP_TRAP_METADATA) #define MLXSW_SP_TRAP_EXCEPTION(_id, _group_id) \ DEVLINK_TRAP_GENERIC(EXCEPTION, TRAP, _id, \ - DEVLINK_TRAP_GROUP_GENERIC(_group_id), \ + DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, \ MLXSW_SP_TRAP_METADATA) #define MLXSW_SP_RXL_DISCARD(_id, _group_id) \ - MLXSW_RXL(mlxsw_sp_rx_drop_listener, DISCARD_##_id, SET_FW_DEFAULT, \ - false, SP_##_group_id, DISCARD) + MLXSW_RXL_DIS(mlxsw_sp_rx_drop_listener, DISCARD_##_id, \ + TRAP_EXCEPTION_TO_CPU, false, SP_##_group_id, \ + SET_FW_DEFAULT, SP_##_group_id) + +#define MLXSW_SP_RXL_ACL_DISCARD(_id, _en_group_id, _dis_group_id) \ + MLXSW_RXL_DIS(mlxsw_sp_rx_acl_drop_listener, DISCARD_##_id, \ + TRAP_EXCEPTION_TO_CPU, false, SP_##_en_group_id, \ + SET_FW_DEFAULT, SP_##_dis_group_id) #define MLXSW_SP_RXL_EXCEPTION(_id, _group_id, _action) \ MLXSW_RXL(mlxsw_sp_rx_exception_listener, _id, \ - _action, false, SP_##_group_id, DISCARD) + _action, false, SP_##_group_id, SET_FW_DEFAULT) + +#define MLXSW_SP_TRAP_POLICER(_id, _rate, _burst) \ + DEVLINK_TRAP_POLICER(_id, _rate, _burst, \ + MLXSW_REG_QPCR_HIGHEST_CIR, \ + MLXSW_REG_QPCR_LOWEST_CIR, \ + 1 << MLXSW_REG_QPCR_HIGHEST_CBS, \ + 1 << MLXSW_REG_QPCR_LOWEST_CBS) + +/* Ordered by policer identifier */ +static const struct devlink_trap_policer mlxsw_sp_trap_policers_arr[] = { + MLXSW_SP_TRAP_POLICER(1, 10 * 1024, 128), +}; -static struct devlink_trap mlxsw_sp_traps_arr[] = { +static const struct devlink_trap_group mlxsw_sp_trap_groups_arr[] = { + DEVLINK_TRAP_GROUP_GENERIC(L2_DROPS, 1), + DEVLINK_TRAP_GROUP_GENERIC(L3_DROPS, 1), + DEVLINK_TRAP_GROUP_GENERIC(TUNNEL_DROPS, 1), + DEVLINK_TRAP_GROUP_GENERIC(ACL_DROPS, 1), +}; + +static const struct devlink_trap mlxsw_sp_traps_arr[] = { MLXSW_SP_TRAP_DROP(SMAC_MC, L2_DROPS), MLXSW_SP_TRAP_DROP(VLAN_TAG_MISMATCH, L2_DROPS), MLXSW_SP_TRAP_DROP(INGRESS_VLAN_FILTER, L2_DROPS), @@ -83,9 +216,13 @@ static struct devlink_trap mlxsw_sp_traps_arr[] = { MLXSW_SP_TRAP_DROP(NON_ROUTABLE, L3_DROPS), MLXSW_SP_TRAP_EXCEPTION(DECAP_ERROR, TUNNEL_DROPS), MLXSW_SP_TRAP_DROP(OVERLAY_SMAC_MC, TUNNEL_DROPS), + MLXSW_SP_TRAP_DROP_EXT(INGRESS_FLOW_ACTION_DROP, ACL_DROPS, + DEVLINK_TRAP_METADATA_TYPE_F_FA_COOKIE), + MLXSW_SP_TRAP_DROP_EXT(EGRESS_FLOW_ACTION_DROP, ACL_DROPS, + DEVLINK_TRAP_METADATA_TYPE_F_FA_COOKIE), }; -static struct mlxsw_listener mlxsw_sp_listeners_arr[] = { +static const struct mlxsw_listener mlxsw_sp_listeners_arr[] = { MLXSW_SP_RXL_DISCARD(ING_PACKET_SMAC_MC, L2_DISCARDS), MLXSW_SP_RXL_DISCARD(ING_SWITCH_VTAG_ALLOW, L2_DISCARDS), MLXSW_SP_RXL_DISCARD(ING_SWITCH_VLAN, L2_DISCARDS), @@ -103,34 +240,37 @@ static struct mlxsw_listener mlxsw_sp_listeners_arr[] = { MLXSW_SP_RXL_DISCARD(ING_ROUTER_IPV4_SIP_BC, L3_DISCARDS), MLXSW_SP_RXL_DISCARD(IPV6_MC_DIP_RESERVED_SCOPE, L3_DISCARDS), MLXSW_SP_RXL_DISCARD(IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE, L3_DISCARDS), - MLXSW_SP_RXL_EXCEPTION(MTUERROR, ROUTER_EXP, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(TTLERROR, ROUTER_EXP, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(RPF, RPF, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(RTR_INGRESS1, REMOTE_ROUTE, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(HOST_MISS_IPV4, HOST_MISS, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(HOST_MISS_IPV6, HOST_MISS, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER3, REMOTE_ROUTE, + MLXSW_SP_RXL_EXCEPTION(MTUERROR, L3_DISCARDS, TRAP_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(TTLERROR, L3_DISCARDS, TRAP_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(RPF, L3_DISCARDS, TRAP_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(RTR_INGRESS1, L3_DISCARDS, TRAP_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(HOST_MISS_IPV4, L3_DISCARDS, TRAP_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(HOST_MISS_IPV6, L3_DISCARDS, TRAP_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER3, L3_DISCARDS, TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM4, ROUTER_EXP, + MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM4, L3_DISCARDS, TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM6, ROUTER_EXP, + MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM6, L3_DISCARDS, TRAP_EXCEPTION_TO_CPU), MLXSW_SP_RXL_DISCARD(ROUTER_IRIF_EN, L3_DISCARDS), MLXSW_SP_RXL_DISCARD(ROUTER_ERIF_EN, L3_DISCARDS), MLXSW_SP_RXL_DISCARD(NON_ROUTABLE, L3_DISCARDS), - MLXSW_SP_RXL_EXCEPTION(DECAP_ECN0, ROUTER_EXP, TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(IPIP_DECAP_ERROR, ROUTER_EXP, + MLXSW_SP_RXL_EXCEPTION(DECAP_ECN0, TUNNEL_DISCARDS, + TRAP_EXCEPTION_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(IPIP_DECAP_ERROR, TUNNEL_DISCARDS, TRAP_EXCEPTION_TO_CPU), MLXSW_SP_RXL_EXCEPTION(DISCARD_DEC_PKT, TUNNEL_DISCARDS, TRAP_EXCEPTION_TO_CPU), MLXSW_SP_RXL_DISCARD(OVERLAY_SMAC_MC, TUNNEL_DISCARDS), + MLXSW_SP_RXL_ACL_DISCARD(INGRESS_ACL, ACL_DISCARDS, DUMMY), + MLXSW_SP_RXL_ACL_DISCARD(EGRESS_ACL, ACL_DISCARDS, DUMMY), }; /* Mapping between hardware trap and devlink trap. Multiple hardware traps can * be mapped to the same devlink trap. Order is according to * 'mlxsw_sp_listeners_arr'. */ -static u16 mlxsw_sp_listener_devlink_map[] = { +static const u16 mlxsw_sp_listener_devlink_map[] = { DEVLINK_TRAP_GENERIC_ID_SMAC_MC, DEVLINK_TRAP_GENERIC_ID_VLAN_TAG_MISMATCH, DEVLINK_TRAP_GENERIC_ID_INGRESS_VLAN_FILTER, @@ -164,99 +304,168 @@ static u16 mlxsw_sp_listener_devlink_map[] = { DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR, DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR, DEVLINK_TRAP_GENERIC_ID_OVERLAY_SMAC_MC, + DEVLINK_TRAP_GENERIC_ID_INGRESS_FLOW_ACTION_DROP, + DEVLINK_TRAP_GENERIC_ID_EGRESS_FLOW_ACTION_DROP, }; -static int mlxsw_sp_rx_listener(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb, - u8 local_port, - struct mlxsw_sp_port *mlxsw_sp_port) +#define MLXSW_SP_THIN_POLICER_ID (MLXSW_REG_HTGT_TRAP_GROUP_MAX + 1) + +static struct mlxsw_sp_trap_policer_item * +mlxsw_sp_trap_policer_item_lookup(struct mlxsw_sp *mlxsw_sp, u32 id) { - struct mlxsw_sp_port_pcpu_stats *pcpu_stats; + struct mlxsw_sp_trap_policer_item *policer_item; + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; - if (unlikely(!mlxsw_sp_port)) { - dev_warn_ratelimited(mlxsw_sp->bus_info->dev, "Port %d: skb received for non-existent port\n", - local_port); - kfree_skb(skb); - return -EINVAL; + list_for_each_entry(policer_item, &trap->policer_item_list, list) { + if (policer_item->id == id) + return policer_item; } - skb->dev = mlxsw_sp_port->dev; + return NULL; +} - pcpu_stats = this_cpu_ptr(mlxsw_sp_port->pcpu_stats); - u64_stats_update_begin(&pcpu_stats->syncp); - pcpu_stats->rx_packets++; - pcpu_stats->rx_bytes += skb->len; - u64_stats_update_end(&pcpu_stats->syncp); +static int mlxsw_sp_trap_cpu_policers_set(struct mlxsw_sp *mlxsw_sp) +{ + char qpcr_pl[MLXSW_REG_QPCR_LEN]; - skb->protocol = eth_type_trans(skb, skb->dev); + /* The purpose of "thin" policer is to drop as many packets + * as possible. The dummy group is using it. + */ + __set_bit(MLXSW_SP_THIN_POLICER_ID, mlxsw_sp->trap->policers_usage); + mlxsw_reg_qpcr_pack(qpcr_pl, MLXSW_SP_THIN_POLICER_ID, + MLXSW_REG_QPCR_IR_UNITS_M, false, 1, 4); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qpcr), qpcr_pl); +} - return 0; +static int mlxsw_sp_trap_dummy_group_init(struct mlxsw_sp *mlxsw_sp) +{ + char htgt_pl[MLXSW_REG_HTGT_LEN]; + + mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_SP_DUMMY, + MLXSW_SP_THIN_POLICER_ID, 0, 1); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(htgt), htgt_pl); } -static void mlxsw_sp_rx_drop_listener(struct sk_buff *skb, u8 local_port, - void *trap_ctx) +static int mlxsw_sp_trap_policers_init(struct mlxsw_sp *mlxsw_sp) { - struct devlink_port *in_devlink_port; - struct mlxsw_sp_port *mlxsw_sp_port; - struct mlxsw_sp *mlxsw_sp; - struct devlink *devlink; + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + u64 free_policers = 0; + u32 last_id = 0; + int err, i; - mlxsw_sp = devlink_trap_ctx_priv(trap_ctx); - mlxsw_sp_port = mlxsw_sp->ports[local_port]; + for_each_clear_bit(i, trap->policers_usage, trap->max_policers) + free_policers++; - if (mlxsw_sp_rx_listener(mlxsw_sp, skb, local_port, mlxsw_sp_port)) - return; + if (ARRAY_SIZE(mlxsw_sp_trap_policers_arr) > free_policers) { + dev_err(mlxsw_sp->bus_info->dev, "Exceeded number of supported packet trap policers\n"); + return -ENOBUFS; + } - devlink = priv_to_devlink(mlxsw_sp->core); - in_devlink_port = mlxsw_core_port_devlink_port_get(mlxsw_sp->core, - local_port); - skb_push(skb, ETH_HLEN); - devlink_trap_report(devlink, skb, trap_ctx, in_devlink_port); - consume_skb(skb); -} + trap->policers_arr = kcalloc(free_policers, + sizeof(struct devlink_trap_policer), + GFP_KERNEL); + if (!trap->policers_arr) + return -ENOMEM; + + trap->policers_count = free_policers; + + for (i = 0; i < free_policers; i++) { + const struct devlink_trap_policer *policer; + + if (i < ARRAY_SIZE(mlxsw_sp_trap_policers_arr)) { + policer = &mlxsw_sp_trap_policers_arr[i]; + trap->policers_arr[i] = *policer; + last_id = policer->id; + } else { + /* Use parameters set for first policer and override + * relevant ones. + */ + policer = &mlxsw_sp_trap_policers_arr[0]; + trap->policers_arr[i] = *policer; + trap->policers_arr[i].id = ++last_id; + trap->policers_arr[i].init_rate = 1; + trap->policers_arr[i].init_burst = 16; + } + } -static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, - void *trap_ctx) -{ - struct devlink_port *in_devlink_port; - struct mlxsw_sp_port *mlxsw_sp_port; - struct mlxsw_sp *mlxsw_sp; - struct devlink *devlink; + INIT_LIST_HEAD(&trap->policer_item_list); - mlxsw_sp = devlink_trap_ctx_priv(trap_ctx); - mlxsw_sp_port = mlxsw_sp->ports[local_port]; + err = devlink_trap_policers_register(devlink, trap->policers_arr, + trap->policers_count); + if (err) + goto err_trap_policers_register; - if (mlxsw_sp_rx_listener(mlxsw_sp, skb, local_port, mlxsw_sp_port)) - return; + return 0; - devlink = priv_to_devlink(mlxsw_sp->core); - in_devlink_port = mlxsw_core_port_devlink_port_get(mlxsw_sp->core, - local_port); - skb_push(skb, ETH_HLEN); - devlink_trap_report(devlink, skb, trap_ctx, in_devlink_port); - skb_pull(skb, ETH_HLEN); - skb->offload_fwd_mark = 1; - netif_receive_skb(skb); +err_trap_policers_register: + kfree(trap->policers_arr); + return err; +} + +static void mlxsw_sp_trap_policers_fini(struct mlxsw_sp *mlxsw_sp) +{ + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + + devlink_trap_policers_unregister(devlink, trap->policers_arr, + trap->policers_count); + WARN_ON(!list_empty(&trap->policer_item_list)); + kfree(trap->policers_arr); } int mlxsw_sp_devlink_traps_init(struct mlxsw_sp *mlxsw_sp) { + size_t groups_count = ARRAY_SIZE(mlxsw_sp_trap_groups_arr); struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + int err; + + err = mlxsw_sp_trap_cpu_policers_set(mlxsw_sp); + if (err) + return err; + + err = mlxsw_sp_trap_dummy_group_init(mlxsw_sp); + if (err) + return err; if (WARN_ON(ARRAY_SIZE(mlxsw_sp_listener_devlink_map) != ARRAY_SIZE(mlxsw_sp_listeners_arr))) return -EINVAL; - return devlink_traps_register(devlink, mlxsw_sp_traps_arr, - ARRAY_SIZE(mlxsw_sp_traps_arr), - mlxsw_sp); + err = mlxsw_sp_trap_policers_init(mlxsw_sp); + if (err) + return err; + + err = devlink_trap_groups_register(devlink, mlxsw_sp_trap_groups_arr, + groups_count); + if (err) + goto err_trap_groups_register; + + err = devlink_traps_register(devlink, mlxsw_sp_traps_arr, + ARRAY_SIZE(mlxsw_sp_traps_arr), mlxsw_sp); + if (err) + goto err_traps_register; + + return 0; + +err_traps_register: + devlink_trap_groups_unregister(devlink, mlxsw_sp_trap_groups_arr, + groups_count); +err_trap_groups_register: + mlxsw_sp_trap_policers_fini(mlxsw_sp); + return err; } void mlxsw_sp_devlink_traps_fini(struct mlxsw_sp *mlxsw_sp) { + size_t groups_count = ARRAY_SIZE(mlxsw_sp_trap_groups_arr); struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); devlink_traps_unregister(devlink, mlxsw_sp_traps_arr, ARRAY_SIZE(mlxsw_sp_traps_arr)); + devlink_trap_groups_unregister(devlink, mlxsw_sp_trap_groups_arr, + groups_count); + mlxsw_sp_trap_policers_fini(mlxsw_sp); } int mlxsw_sp_trap_init(struct mlxsw_core *mlxsw_core, @@ -265,7 +474,7 @@ int mlxsw_sp_trap_init(struct mlxsw_core *mlxsw_core, int i; for (i = 0; i < ARRAY_SIZE(mlxsw_sp_listener_devlink_map); i++) { - struct mlxsw_listener *listener; + const struct mlxsw_listener *listener; int err; if (mlxsw_sp_listener_devlink_map[i] != trap->id) @@ -286,7 +495,7 @@ void mlxsw_sp_trap_fini(struct mlxsw_core *mlxsw_core, int i; for (i = 0; i < ARRAY_SIZE(mlxsw_sp_listener_devlink_map); i++) { - struct mlxsw_listener *listener; + const struct mlxsw_listener *listener; if (mlxsw_sp_listener_devlink_map[i] != trap->id) continue; @@ -303,27 +512,24 @@ int mlxsw_sp_trap_action_set(struct mlxsw_core *mlxsw_core, int i; for (i = 0; i < ARRAY_SIZE(mlxsw_sp_listener_devlink_map); i++) { - enum mlxsw_reg_hpkt_action hw_action; - struct mlxsw_listener *listener; + const struct mlxsw_listener *listener; + bool enabled; int err; if (mlxsw_sp_listener_devlink_map[i] != trap->id) continue; listener = &mlxsw_sp_listeners_arr[i]; - switch (action) { case DEVLINK_TRAP_ACTION_DROP: - hw_action = MLXSW_REG_HPKT_ACTION_SET_FW_DEFAULT; + enabled = false; break; case DEVLINK_TRAP_ACTION_TRAP: - hw_action = MLXSW_REG_HPKT_ACTION_TRAP_EXCEPTION_TO_CPU; + enabled = true; break; default: return -EINVAL; } - - err = mlxsw_core_trap_action_set(mlxsw_core, listener, - hw_action); + err = mlxsw_core_trap_state_set(mlxsw_core, listener, enabled); if (err) return err; } @@ -331,62 +537,34 @@ int mlxsw_sp_trap_action_set(struct mlxsw_core *mlxsw_core, return 0; } -#define MLXSW_SP_DISCARD_POLICER_ID (MLXSW_REG_HTGT_TRAP_GROUP_MAX + 1) - -static int -mlxsw_sp_trap_group_policer_init(struct mlxsw_sp *mlxsw_sp, - const struct devlink_trap_group *group) -{ - enum mlxsw_reg_qpcr_ir_units ir_units; - char qpcr_pl[MLXSW_REG_QPCR_LEN]; - u16 policer_id; - u8 burst_size; - bool is_bytes; - u32 rate; - - switch (group->id) { - case DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS: /* fall through */ - case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS: /* fall through */ - case DEVLINK_TRAP_GROUP_GENERIC_ID_TUNNEL_DROPS: - policer_id = MLXSW_SP_DISCARD_POLICER_ID; - ir_units = MLXSW_REG_QPCR_IR_UNITS_M; - is_bytes = false; - rate = 10 * 1024; /* 10Kpps */ - burst_size = 7; - break; - default: - return -EINVAL; - } - - mlxsw_reg_qpcr_pack(qpcr_pl, policer_id, ir_units, is_bytes, rate, - burst_size); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qpcr), qpcr_pl); -} - static int -__mlxsw_sp_trap_group_init(struct mlxsw_sp *mlxsw_sp, - const struct devlink_trap_group *group) +__mlxsw_sp_trap_group_init(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_group *group, + u32 policer_id) { + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + u16 hw_policer_id = MLXSW_REG_HTGT_INVALID_POLICER; char htgt_pl[MLXSW_REG_HTGT_LEN]; u8 priority, tc, group_id; - u16 policer_id; switch (group->id) { case DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS: group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_L2_DISCARDS; - policer_id = MLXSW_SP_DISCARD_POLICER_ID; priority = 0; tc = 1; break; case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS: group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_DISCARDS; - policer_id = MLXSW_SP_DISCARD_POLICER_ID; priority = 0; tc = 1; break; case DEVLINK_TRAP_GROUP_GENERIC_ID_TUNNEL_DROPS: group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS; - policer_id = MLXSW_SP_DISCARD_POLICER_ID; + priority = 0; + tc = 1; + break; + case DEVLINK_TRAP_GROUP_GENERIC_ID_ACL_DROPS: + group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_ACL_DISCARDS; priority = 0; tc = 1; break; @@ -394,23 +572,179 @@ __mlxsw_sp_trap_group_init(struct mlxsw_sp *mlxsw_sp, return -EINVAL; } - mlxsw_reg_htgt_pack(htgt_pl, group_id, policer_id, priority, tc); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(htgt), htgt_pl); + if (policer_id) { + struct mlxsw_sp_trap_policer_item *policer_item; + + policer_item = mlxsw_sp_trap_policer_item_lookup(mlxsw_sp, + policer_id); + if (WARN_ON(!policer_item)) + return -EINVAL; + hw_policer_id = policer_item->hw_id; + } + + mlxsw_reg_htgt_pack(htgt_pl, group_id, hw_policer_id, priority, tc); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(htgt), htgt_pl); } int mlxsw_sp_trap_group_init(struct mlxsw_core *mlxsw_core, const struct devlink_trap_group *group) { - struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + return __mlxsw_sp_trap_group_init(mlxsw_core, group, + group->init_policer_id); +} + +int mlxsw_sp_trap_group_set(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_group *group, + const struct devlink_trap_policer *policer) +{ + u32 policer_id = policer ? policer->id : 0; + + return __mlxsw_sp_trap_group_init(mlxsw_core, group, policer_id); +} + +static struct mlxsw_sp_trap_policer_item * +mlxsw_sp_trap_policer_item_init(struct mlxsw_sp *mlxsw_sp, u32 id) +{ + struct mlxsw_sp_trap_policer_item *policer_item; + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + u16 hw_id; + + /* We should be able to allocate a policer because the number of + * policers we registered with devlink is in according with the number + * of available policers. + */ + hw_id = find_first_zero_bit(trap->policers_usage, trap->max_policers); + if (WARN_ON(hw_id == trap->max_policers)) + return ERR_PTR(-ENOBUFS); + + policer_item = kzalloc(sizeof(*policer_item), GFP_KERNEL); + if (!policer_item) + return ERR_PTR(-ENOMEM); + + __set_bit(hw_id, trap->policers_usage); + policer_item->hw_id = hw_id; + policer_item->id = id; + list_add_tail(&policer_item->list, &trap->policer_item_list); + + return policer_item; +} + +static void +mlxsw_sp_trap_policer_item_fini(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_trap_policer_item *policer_item) +{ + list_del(&policer_item->list); + __clear_bit(policer_item->hw_id, mlxsw_sp->trap->policers_usage); + kfree(policer_item); +} + +static int mlxsw_sp_trap_policer_bs(u64 burst, u8 *p_burst_size, + struct netlink_ext_ack *extack) +{ + int bs = fls64(burst) - 1; + + if (burst != (1 << bs)) { + NL_SET_ERR_MSG_MOD(extack, "Policer burst size is not power of two"); + return -EINVAL; + } + + *p_burst_size = bs; + + return 0; +} + +static int __mlxsw_sp_trap_policer_set(struct mlxsw_sp *mlxsw_sp, u16 hw_id, + u64 rate, u64 burst, bool clear_counter, + struct netlink_ext_ack *extack) +{ + char qpcr_pl[MLXSW_REG_QPCR_LEN]; + u8 burst_size; int err; - err = mlxsw_sp_trap_group_policer_init(mlxsw_sp, group); + err = mlxsw_sp_trap_policer_bs(burst, &burst_size, extack); if (err) return err; - err = __mlxsw_sp_trap_group_init(mlxsw_sp, group); + mlxsw_reg_qpcr_pack(qpcr_pl, hw_id, MLXSW_REG_QPCR_IR_UNITS_M, false, + rate, burst_size); + mlxsw_reg_qpcr_clear_counter_set(qpcr_pl, clear_counter); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qpcr), qpcr_pl); +} + +int mlxsw_sp_trap_policer_init(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + struct mlxsw_sp_trap_policer_item *policer_item; + int err; + + policer_item = mlxsw_sp_trap_policer_item_init(mlxsw_sp, policer->id); + if (IS_ERR(policer_item)) + return PTR_ERR(policer_item); + + err = __mlxsw_sp_trap_policer_set(mlxsw_sp, policer_item->hw_id, + policer->init_rate, + policer->init_burst, true, NULL); + if (err) + goto err_trap_policer_set; + + return 0; + +err_trap_policer_set: + mlxsw_sp_trap_policer_item_fini(mlxsw_sp, policer_item); + return err; +} + +void mlxsw_sp_trap_policer_fini(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + struct mlxsw_sp_trap_policer_item *policer_item; + + policer_item = mlxsw_sp_trap_policer_item_lookup(mlxsw_sp, policer->id); + if (WARN_ON(!policer_item)) + return; + + mlxsw_sp_trap_policer_item_fini(mlxsw_sp, policer_item); +} + +int mlxsw_sp_trap_policer_set(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer, + u64 rate, u64 burst, + struct netlink_ext_ack *extack) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + struct mlxsw_sp_trap_policer_item *policer_item; + + policer_item = mlxsw_sp_trap_policer_item_lookup(mlxsw_sp, policer->id); + if (WARN_ON(!policer_item)) + return -EINVAL; + + return __mlxsw_sp_trap_policer_set(mlxsw_sp, policer_item->hw_id, + rate, burst, false, extack); +} + +int +mlxsw_sp_trap_policer_counter_get(struct mlxsw_core *mlxsw_core, + const struct devlink_trap_policer *policer, + u64 *p_drops) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + struct mlxsw_sp_trap_policer_item *policer_item; + char qpcr_pl[MLXSW_REG_QPCR_LEN]; + int err; + + policer_item = mlxsw_sp_trap_policer_item_lookup(mlxsw_sp, policer->id); + if (WARN_ON(!policer_item)) + return -EINVAL; + + mlxsw_reg_qpcr_pack(qpcr_pl, policer_item->hw_id, + MLXSW_REG_QPCR_IR_UNITS_M, false, 0, 0); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(qpcr), qpcr_pl); if (err) return err; + *p_drops = mlxsw_reg_qpcr_violate_count_get(qpcr_pl); + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.h new file mode 100644 index 000000000000..8c54897ba173 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.h @@ -0,0 +1,24 @@ +/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ +/* Copyright (c) 2020 Mellanox Technologies. All rights reserved */ + +#ifndef _MLXSW_SPECTRUM_TRAP_H +#define _MLXSW_SPECTRUM_TRAP_H + +#include <linux/list.h> +#include <net/devlink.h> + +struct mlxsw_sp_trap { + struct devlink_trap_policer *policers_arr; /* Registered policers */ + u64 policers_count; /* Number of registered policers */ + struct list_head policer_item_list; + u64 max_policers; + unsigned long policers_usage[]; /* Usage bitmap */ +}; + +struct mlxsw_sp_trap_policer_item { + u16 hw_id; + u32 id; + struct list_head list; /* Member of policer_item_list */ +}; + +#endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c index f0e98ec8f1ee..90535820b559 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c +++ b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c @@ -180,7 +180,7 @@ static int mlxsw_sx_port_oper_status_get(struct mlxsw_sx_port *mlxsw_sx_port, if (err) return err; oper_status = mlxsw_reg_paos_oper_status_get(paos_pl); - *p_is_up = oper_status == MLXSW_PORT_ADMIN_STATUS_UP ? true : false; + *p_is_up = oper_status == MLXSW_PORT_ADMIN_STATUS_UP; return 0; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h index 12e1fa998d42..eaa521b7561b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/trap.h +++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h @@ -102,6 +102,8 @@ enum { MLXSW_TRAP_ID_ACL1 = 0x1C1, /* Multicast trap used for routes with trap-and-forward action */ MLXSW_TRAP_ID_ACL2 = 0x1C2, + MLXSW_TRAP_ID_DISCARD_INGRESS_ACL = 0x1C3, + MLXSW_TRAP_ID_DISCARD_EGRESS_ACL = 0x1C4, MLXSW_TRAP_ID_MAX = 0x1FF }; diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c index d1444ba36e10..4fe6aedca22f 100644 --- a/drivers/net/ethernet/micrel/ksz884x.c +++ b/drivers/net/ethernet/micrel/ksz884x.c @@ -5694,7 +5694,7 @@ static void dev_set_promiscuous(struct net_device *dev, struct dev_priv *priv, * from the bridge. */ if ((hw->features & STP_SUPPORT) && !promiscuous && - (dev->priv_flags & IFF_BRIDGE_PORT)) { + netif_is_bridge_port(dev)) { struct ksz_switch *sw = hw->ksz_switch; int port = priv->port.first_port; diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c index d3b7373c5961..b4731df186f4 100644 --- a/drivers/net/ethernet/mscc/ocelot.c +++ b/drivers/net/ethernet/mscc/ocelot.c @@ -442,8 +442,23 @@ void ocelot_adjust_link(struct ocelot *ocelot, int port, ocelot_port_writel(ocelot_port, DEV_MAC_MODE_CFG_FDX_ENA | mode, DEV_MAC_MODE_CFG); - if (ocelot->ops->pcs_init) - ocelot->ops->pcs_init(ocelot, port); + /* Disable HDX fast control */ + ocelot_port_writel(ocelot_port, DEV_PORT_MISC_HDX_FAST_DIS, + DEV_PORT_MISC); + + /* SGMII only for now */ + ocelot_port_writel(ocelot_port, PCS1G_MODE_CFG_SGMII_MODE_ENA, + PCS1G_MODE_CFG); + ocelot_port_writel(ocelot_port, PCS1G_SD_CFG_SD_SEL, PCS1G_SD_CFG); + + /* Enable PCS */ + ocelot_port_writel(ocelot_port, PCS1G_CFG_PCS_ENA, PCS1G_CFG); + + /* No aneg on SGMII */ + ocelot_port_writel(ocelot_port, 0, PCS1G_ANEG_CFG); + + /* No loopback */ + ocelot_port_writel(ocelot_port, 0, PCS1G_LB_CFG); /* Enable MAC module */ ocelot_port_writel(ocelot_port, DEV_MAC_ENA_CFG_RX_ENA | @@ -1398,7 +1413,7 @@ void ocelot_bridge_stp_state_set(struct ocelot *ocelot, int port, u8 state) * a source for the other ports. */ for (p = 0; p < ocelot->num_phys_ports; p++) { - if (p == ocelot->cpu || (ocelot->bridge_fwd_mask & BIT(p))) { + if (ocelot->bridge_fwd_mask & BIT(p)) { unsigned long mask = ocelot->bridge_fwd_mask & ~BIT(p); for (i = 0; i < ocelot->num_phys_ports; i++) { @@ -1413,18 +1428,10 @@ void ocelot_bridge_stp_state_set(struct ocelot *ocelot, int port, u8 state) } } - /* Avoid the NPI port from looping back to itself */ - if (p != ocelot->cpu) - mask |= BIT(ocelot->cpu); - ocelot_write_rix(ocelot, mask, ANA_PGID_PGID, PGID_SRC + p); } else { - /* Only the CPU port, this is compatible with link - * aggregation. - */ - ocelot_write_rix(ocelot, - BIT(ocelot->cpu), + ocelot_write_rix(ocelot, 0, ANA_PGID_PGID, PGID_SRC + p); } } @@ -2178,13 +2185,25 @@ static int ocelot_init_timestamp(struct ocelot *ocelot) /* Configure the maximum SDU (L2 payload) on RX to the value specified in @sdu. * The length of VLAN tags is accounted for automatically via DEV_MAC_TAGS_CFG. + * In the special case that it's the NPI port that we're configuring, the + * length of the tag and optional prefix needs to be accounted for privately, + * in order to be able to sustain communication at the requested @sdu. */ -static void ocelot_port_set_maxlen(struct ocelot *ocelot, int port, size_t sdu) +void ocelot_port_set_maxlen(struct ocelot *ocelot, int port, size_t sdu) { struct ocelot_port *ocelot_port = ocelot->ports[port]; int maxlen = sdu + ETH_HLEN + ETH_FCS_LEN; int atop_wm; + if (port == ocelot->npi) { + maxlen += OCELOT_TAG_LEN; + + if (ocelot->inj_prefix == OCELOT_TAG_PREFIX_SHORT) + maxlen += OCELOT_SHORT_PREFIX_LEN; + else if (ocelot->inj_prefix == OCELOT_TAG_PREFIX_LONG) + maxlen += OCELOT_LONG_PREFIX_LEN; + } + ocelot_port_writel(ocelot_port, maxlen, DEV_MAC_MAXLEN_CFG); /* Set Pause WM hysteresis @@ -2202,6 +2221,24 @@ static void ocelot_port_set_maxlen(struct ocelot *ocelot, int port, size_t sdu) SYS_ATOP, port); ocelot_write(ocelot, ocelot_wm_enc(atop_wm), SYS_ATOP_TOT_CFG); } +EXPORT_SYMBOL(ocelot_port_set_maxlen); + +int ocelot_get_max_mtu(struct ocelot *ocelot, int port) +{ + int max_mtu = 65535 - ETH_HLEN - ETH_FCS_LEN; + + if (port == ocelot->npi) { + max_mtu -= OCELOT_TAG_LEN; + + if (ocelot->inj_prefix == OCELOT_TAG_PREFIX_SHORT) + max_mtu -= OCELOT_SHORT_PREFIX_LEN; + else if (ocelot->inj_prefix == OCELOT_TAG_PREFIX_LONG) + max_mtu -= OCELOT_LONG_PREFIX_LEN; + } + + return max_mtu; +} +EXPORT_SYMBOL(ocelot_get_max_mtu); void ocelot_init_port(struct ocelot *ocelot, int port) { @@ -2299,42 +2336,57 @@ int ocelot_probe_port(struct ocelot *ocelot, u8 port, } EXPORT_SYMBOL(ocelot_probe_port); -void ocelot_set_cpu_port(struct ocelot *ocelot, int cpu, - enum ocelot_tag_prefix injection, - enum ocelot_tag_prefix extraction) +/* Configure and enable the CPU port module, which is a set of queues. + * If @npi contains a valid port index, the CPU port module is connected + * to the Node Processor Interface (NPI). This is the mode through which + * frames can be injected from and extracted to an external CPU, + * over Ethernet. + */ +void ocelot_configure_cpu(struct ocelot *ocelot, int npi, + enum ocelot_tag_prefix injection, + enum ocelot_tag_prefix extraction) { - /* Configure and enable the CPU port. */ + int cpu = ocelot->num_phys_ports; + + ocelot->npi = npi; + ocelot->inj_prefix = injection; + ocelot->xtr_prefix = extraction; + + /* The unicast destination PGID for the CPU port module is unused */ ocelot_write_rix(ocelot, 0, ANA_PGID_PGID, cpu); + /* Instead set up a multicast destination PGID for traffic copied to + * the CPU. Whitelisted MAC addresses like the port netdevice MAC + * addresses will be copied to the CPU via this PGID. + */ ocelot_write_rix(ocelot, BIT(cpu), ANA_PGID_PGID, PGID_CPU); ocelot_write_gix(ocelot, ANA_PORT_PORT_CFG_RECV_ENA | ANA_PORT_PORT_CFG_PORTID_VAL(cpu), ANA_PORT_PORT_CFG, cpu); - /* If the CPU port is a physical port, set up the port in Node - * Processor Interface (NPI) mode. This is the mode through which - * frames can be injected from and extracted to an external CPU. - * Only one port can be an NPI at the same time. - */ - if (cpu < ocelot->num_phys_ports) { - int sdu = ETH_DATA_LEN + OCELOT_TAG_LEN; - + if (npi >= 0 && npi < ocelot->num_phys_ports) { ocelot_write(ocelot, QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK_M | - QSYS_EXT_CPU_CFG_EXT_CPU_PORT(cpu), + QSYS_EXT_CPU_CFG_EXT_CPU_PORT(npi), QSYS_EXT_CPU_CFG); - if (injection == OCELOT_TAG_PREFIX_SHORT) - sdu += OCELOT_SHORT_PREFIX_LEN; - else if (injection == OCELOT_TAG_PREFIX_LONG) - sdu += OCELOT_LONG_PREFIX_LEN; - - ocelot_port_set_maxlen(ocelot, cpu, sdu); + /* Enable NPI port */ + ocelot_write_rix(ocelot, + QSYS_SWITCH_PORT_MODE_INGRESS_DROP_MODE | + QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG(1) | + QSYS_SWITCH_PORT_MODE_PORT_ENA, + QSYS_SWITCH_PORT_MODE, npi); + /* NPI port Injection/Extraction configuration */ + ocelot_write_rix(ocelot, + SYS_PORT_MODE_INCL_XTR_HDR(extraction) | + SYS_PORT_MODE_INCL_INJ_HDR(injection), + SYS_PORT_MODE, npi); } - /* CPU port Injection/Extraction configuration */ + /* Enable CPU port module */ ocelot_write_rix(ocelot, QSYS_SWITCH_PORT_MODE_INGRESS_DROP_MODE | QSYS_SWITCH_PORT_MODE_SCH_NEXT_CFG(1) | QSYS_SWITCH_PORT_MODE_PORT_ENA, QSYS_SWITCH_PORT_MODE, cpu); + /* CPU port Injection/Extraction configuration */ ocelot_write_rix(ocelot, SYS_PORT_MODE_INCL_XTR_HDR(extraction) | SYS_PORT_MODE_INCL_INJ_HDR(injection), SYS_PORT_MODE, cpu); @@ -2344,10 +2396,8 @@ void ocelot_set_cpu_port(struct ocelot *ocelot, int cpu, ANA_PORT_VLAN_CFG_VLAN_AWARE_ENA | ANA_PORT_VLAN_CFG_VLAN_POP_CNT(1), ANA_PORT_VLAN_CFG, cpu); - - ocelot->cpu = cpu; } -EXPORT_SYMBOL(ocelot_set_cpu_port); +EXPORT_SYMBOL(ocelot_configure_cpu); int ocelot_init(struct ocelot *ocelot) { @@ -2499,7 +2549,6 @@ void ocelot_deinit(struct ocelot *ocelot) cancel_delayed_work(&ocelot->stats_work); destroy_workqueue(ocelot->stats_queue); mutex_destroy(&ocelot->stats_lock); - ocelot_ace_deinit(); if (ocelot->ptp_clock) ptp_clock_unregister(ocelot->ptp_clock); diff --git a/drivers/net/ethernet/mscc/ocelot.h b/drivers/net/ethernet/mscc/ocelot.h index 04372ba72fec..e34ef8380eb3 100644 --- a/drivers/net/ethernet/mscc/ocelot.h +++ b/drivers/net/ethernet/mscc/ocelot.h @@ -28,16 +28,6 @@ #include "ocelot_tc.h" #include "ocelot_ptp.h" -#define PGID_AGGR 64 -#define PGID_SRC 80 - -/* Reserved PGIDs */ -#define PGID_CPU (PGID_AGGR - 5) -#define PGID_UC (PGID_AGGR - 4) -#define PGID_MC (PGID_AGGR - 3) -#define PGID_MCIPV4 (PGID_AGGR - 2) -#define PGID_MCIPV6 (PGID_AGGR - 1) - #define OCELOT_BUFFER_CELL_SZ 60 #define OCELOT_STATS_CHECK_DELAY (2 * HZ) diff --git a/drivers/net/ethernet/mscc/ocelot_ace.c b/drivers/net/ethernet/mscc/ocelot_ace.c index 86fc6e6b46dd..3bd286044480 100644 --- a/drivers/net/ethernet/mscc/ocelot_ace.c +++ b/drivers/net/ethernet/mscc/ocelot_ace.c @@ -6,60 +6,13 @@ #include <linux/iopoll.h> #include <linux/proc_fs.h> +#include <soc/mscc/ocelot_vcap.h> +#include "ocelot_police.h" #include "ocelot_ace.h" -#include "ocelot_vcap.h" #include "ocelot_s2.h" #define OCELOT_POLICER_DISCARD 0x17f - -static struct ocelot_acl_block *acl_block; - -struct vcap_props { - const char *name; /* Symbolic name */ - u16 tg_width; /* Type-group width (in bits) */ - u16 sw_count; /* Sub word count */ - u16 entry_count; /* Entry count */ - u16 entry_words; /* Number of entry words */ - u16 entry_width; /* Entry width (in bits) */ - u16 action_count; /* Action count */ - u16 action_words; /* Number of action words */ - u16 action_width; /* Action width (in bits) */ - u16 action_type_width; /* Action type width (in bits) */ - struct { - u16 width; /* Action type width (in bits) */ - u16 count; /* Action type sub word count */ - } action_table[2]; - u16 counter_words; /* Number of counter words */ - u16 counter_width; /* Counter width (in bits) */ -}; - #define ENTRY_WIDTH 32 -#define BITS_TO_32BIT(x) (1 + (((x) - 1) / ENTRY_WIDTH)) - -static const struct vcap_props vcap_is2 = { - .name = "IS2", - .tg_width = 2, - .sw_count = 4, - .entry_count = VCAP_IS2_CNT, - .entry_words = BITS_TO_32BIT(VCAP_IS2_ENTRY_WIDTH), - .entry_width = VCAP_IS2_ENTRY_WIDTH, - .action_count = (VCAP_IS2_CNT + VCAP_PORT_CNT + 2), - .action_words = BITS_TO_32BIT(VCAP_IS2_ACTION_WIDTH), - .action_width = (VCAP_IS2_ACTION_WIDTH), - .action_type_width = 1, - .action_table = { - { - .width = (IS2_AO_ACL_ID + IS2_AL_ACL_ID), - .count = 2 - }, - { - .width = 6, - .count = 4 - }, - }, - .counter_words = BITS_TO_32BIT(4 * ENTRY_WIDTH), - .counter_width = ENTRY_WIDTH, -}; enum vcap_sel { VCAP_SEL_ENTRY = 0x1, @@ -95,18 +48,20 @@ struct vcap_data { u32 tg_mask; /* Current type-group mask */ }; -static u32 vcap_s2_read_update_ctrl(struct ocelot *oc) +static u32 vcap_s2_read_update_ctrl(struct ocelot *ocelot) { - return ocelot_read(oc, S2_CORE_UPDATE_CTRL); + return ocelot_read(ocelot, S2_CORE_UPDATE_CTRL); } -static void vcap_cmd(struct ocelot *oc, u16 ix, int cmd, int sel) +static void vcap_cmd(struct ocelot *ocelot, u16 ix, int cmd, int sel) { + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + u32 value = (S2_CORE_UPDATE_CTRL_UPDATE_CMD(cmd) | S2_CORE_UPDATE_CTRL_UPDATE_ADDR(ix) | S2_CORE_UPDATE_CTRL_UPDATE_SHOT); - if ((sel & VCAP_SEL_ENTRY) && ix >= vcap_is2.entry_count) + if ((sel & VCAP_SEL_ENTRY) && ix >= vcap_is2->entry_count) return; if (!(sel & VCAP_SEL_ENTRY)) @@ -118,83 +73,101 @@ static void vcap_cmd(struct ocelot *oc, u16 ix, int cmd, int sel) if (!(sel & VCAP_SEL_COUNTER)) value |= S2_CORE_UPDATE_CTRL_UPDATE_CNT_DIS; - ocelot_write(oc, value, S2_CORE_UPDATE_CTRL); - readx_poll_timeout(vcap_s2_read_update_ctrl, oc, value, + ocelot_write(ocelot, value, S2_CORE_UPDATE_CTRL); + readx_poll_timeout(vcap_s2_read_update_ctrl, ocelot, value, (value & S2_CORE_UPDATE_CTRL_UPDATE_SHOT) == 0, 10, 100000); } /* Convert from 0-based row to VCAP entry row and run command */ -static void vcap_row_cmd(struct ocelot *oc, u32 row, int cmd, int sel) +static void vcap_row_cmd(struct ocelot *ocelot, u32 row, int cmd, int sel) { - vcap_cmd(oc, vcap_is2.entry_count - row - 1, cmd, sel); + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + + vcap_cmd(ocelot, vcap_is2->entry_count - row - 1, cmd, sel); } -static void vcap_entry2cache(struct ocelot *oc, struct vcap_data *data) +static void vcap_entry2cache(struct ocelot *ocelot, struct vcap_data *data) { - u32 i; + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + u32 entry_words, i; - for (i = 0; i < vcap_is2.entry_words; i++) { - ocelot_write_rix(oc, data->entry[i], S2_CACHE_ENTRY_DAT, i); - ocelot_write_rix(oc, ~data->mask[i], S2_CACHE_MASK_DAT, i); + entry_words = DIV_ROUND_UP(vcap_is2->entry_width, ENTRY_WIDTH); + + for (i = 0; i < entry_words; i++) { + ocelot_write_rix(ocelot, data->entry[i], S2_CACHE_ENTRY_DAT, i); + ocelot_write_rix(ocelot, ~data->mask[i], S2_CACHE_MASK_DAT, i); } - ocelot_write(oc, data->tg, S2_CACHE_TG_DAT); + ocelot_write(ocelot, data->tg, S2_CACHE_TG_DAT); } -static void vcap_cache2entry(struct ocelot *oc, struct vcap_data *data) +static void vcap_cache2entry(struct ocelot *ocelot, struct vcap_data *data) { - u32 i; + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + u32 entry_words, i; + + entry_words = DIV_ROUND_UP(vcap_is2->entry_width, ENTRY_WIDTH); - for (i = 0; i < vcap_is2.entry_words; i++) { - data->entry[i] = ocelot_read_rix(oc, S2_CACHE_ENTRY_DAT, i); + for (i = 0; i < entry_words; i++) { + data->entry[i] = ocelot_read_rix(ocelot, S2_CACHE_ENTRY_DAT, i); // Invert mask - data->mask[i] = ~ocelot_read_rix(oc, S2_CACHE_MASK_DAT, i); + data->mask[i] = ~ocelot_read_rix(ocelot, S2_CACHE_MASK_DAT, i); } - data->tg = ocelot_read(oc, S2_CACHE_TG_DAT); + data->tg = ocelot_read(ocelot, S2_CACHE_TG_DAT); } -static void vcap_action2cache(struct ocelot *oc, struct vcap_data *data) +static void vcap_action2cache(struct ocelot *ocelot, struct vcap_data *data) { - u32 i, width, mask; + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + u32 action_words, i, width, mask; /* Encode action type */ - width = vcap_is2.action_type_width; + width = vcap_is2->action_type_width; if (width) { mask = GENMASK(width, 0); data->action[0] = ((data->action[0] & ~mask) | data->type); } - for (i = 0; i < vcap_is2.action_words; i++) - ocelot_write_rix(oc, data->action[i], S2_CACHE_ACTION_DAT, i); + action_words = DIV_ROUND_UP(vcap_is2->action_width, ENTRY_WIDTH); - for (i = 0; i < vcap_is2.counter_words; i++) - ocelot_write_rix(oc, data->counter[i], S2_CACHE_CNT_DAT, i); + for (i = 0; i < action_words; i++) + ocelot_write_rix(ocelot, data->action[i], S2_CACHE_ACTION_DAT, + i); + + for (i = 0; i < vcap_is2->counter_words; i++) + ocelot_write_rix(ocelot, data->counter[i], S2_CACHE_CNT_DAT, i); } -static void vcap_cache2action(struct ocelot *oc, struct vcap_data *data) +static void vcap_cache2action(struct ocelot *ocelot, struct vcap_data *data) { - u32 i, width; + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + u32 action_words, i, width; + + action_words = DIV_ROUND_UP(vcap_is2->action_width, ENTRY_WIDTH); - for (i = 0; i < vcap_is2.action_words; i++) - data->action[i] = ocelot_read_rix(oc, S2_CACHE_ACTION_DAT, i); + for (i = 0; i < action_words; i++) + data->action[i] = ocelot_read_rix(ocelot, S2_CACHE_ACTION_DAT, + i); - for (i = 0; i < vcap_is2.counter_words; i++) - data->counter[i] = ocelot_read_rix(oc, S2_CACHE_CNT_DAT, i); + for (i = 0; i < vcap_is2->counter_words; i++) + data->counter[i] = ocelot_read_rix(ocelot, S2_CACHE_CNT_DAT, i); /* Extract action type */ - width = vcap_is2.action_type_width; + width = vcap_is2->action_type_width; data->type = (width ? (data->action[0] & GENMASK(width, 0)) : 0); } /* Calculate offsets for entry */ -static void is2_data_get(struct vcap_data *data, int ix) +static void is2_data_get(struct ocelot *ocelot, struct vcap_data *data, int ix) { - u32 i, col, offset, count, cnt, base, width = vcap_is2.tg_width; + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + u32 i, col, offset, count, cnt, base; + u32 width = vcap_is2->tg_width; count = (data->tg_sw == VCAP_TG_HALF ? 2 : 4); col = (ix % 2); - cnt = (vcap_is2.sw_count / count); - base = (vcap_is2.sw_count - col * cnt - cnt); + cnt = (vcap_is2->sw_count / count); + base = (vcap_is2->sw_count - col * cnt - cnt); data->tg_value = 0; data->tg_mask = 0; for (i = 0; i < cnt; i++) { @@ -205,13 +178,13 @@ static void is2_data_get(struct vcap_data *data, int ix) /* Calculate key/action/counter offsets */ col = (count - col - 1); - data->key_offset = (base * vcap_is2.entry_width) / vcap_is2.sw_count; - data->counter_offset = (cnt * col * vcap_is2.counter_width); + data->key_offset = (base * vcap_is2->entry_width) / vcap_is2->sw_count; + data->counter_offset = (cnt * col * vcap_is2->counter_width); i = data->type; - width = vcap_is2.action_table[i].width; - cnt = vcap_is2.action_table[i].count; + width = vcap_is2->action_table[i].width; + cnt = vcap_is2->action_table[i].count; data->action_offset = - (((cnt * col * width) / count) + vcap_is2.action_type_width); + (((cnt * col * width) / count) + vcap_is2->action_type_width); } static void vcap_data_set(u32 *data, u32 offset, u32 len, u32 value) @@ -242,22 +215,39 @@ static u32 vcap_data_get(u32 *data, u32 offset, u32 len) return value; } -static void vcap_key_set(struct vcap_data *data, u32 offset, u32 width, - u32 value, u32 mask) +static void vcap_key_field_set(struct vcap_data *data, u32 offset, u32 width, + u32 value, u32 mask) { vcap_data_set(data->entry, offset + data->key_offset, width, value); vcap_data_set(data->mask, offset + data->key_offset, width, mask); } -static void vcap_key_bytes_set(struct vcap_data *data, u32 offset, u8 *val, - u8 *msk, u32 count) +static void vcap_key_set(struct ocelot *ocelot, struct vcap_data *data, + enum vcap_is2_half_key_field field, + u32 value, u32 mask) { + u32 offset = ocelot->vcap_is2_keys[field].offset; + u32 length = ocelot->vcap_is2_keys[field].length; + + vcap_key_field_set(data, offset, length, value, mask); +} + +static void vcap_key_bytes_set(struct ocelot *ocelot, struct vcap_data *data, + enum vcap_is2_half_key_field field, + u8 *val, u8 *msk) +{ + u32 offset = ocelot->vcap_is2_keys[field].offset; + u32 count = ocelot->vcap_is2_keys[field].length; u32 i, j, n = 0, value = 0, mask = 0; + WARN_ON(count % 8); + /* Data wider than 32 bits are split up in chunks of maximum 32 bits. * The 32 LSB of the data are written to the 32 MSB of the TCAM. */ - offset += (count * 8); + offset += count; + count /= 8; + for (i = 0; i < count; i++) { j = (count - i - 1); value += (val[j] << n); @@ -265,7 +255,7 @@ static void vcap_key_bytes_set(struct vcap_data *data, u32 offset, u8 *val, n += 8; if (n == ENTRY_WIDTH || (i + 1) == count) { offset -= n; - vcap_key_set(data, offset, n, value, mask); + vcap_key_field_set(data, offset, n, value, mask); n = 0; value = 0; mask = 0; @@ -273,55 +263,71 @@ static void vcap_key_bytes_set(struct vcap_data *data, u32 offset, u8 *val, } } -static void vcap_key_l4_port_set(struct vcap_data *data, u32 offset, +static void vcap_key_l4_port_set(struct ocelot *ocelot, struct vcap_data *data, + enum vcap_is2_half_key_field field, struct ocelot_vcap_udp_tcp *port) { - vcap_key_set(data, offset, 16, port->value, port->mask); + u32 offset = ocelot->vcap_is2_keys[field].offset; + u32 length = ocelot->vcap_is2_keys[field].length; + + WARN_ON(length != 16); + + vcap_key_field_set(data, offset, length, port->value, port->mask); } -static void vcap_key_bit_set(struct vcap_data *data, u32 offset, +static void vcap_key_bit_set(struct ocelot *ocelot, struct vcap_data *data, + enum vcap_is2_half_key_field field, enum ocelot_vcap_bit val) { - vcap_key_set(data, offset, 1, val == OCELOT_VCAP_BIT_1 ? 1 : 0, - val == OCELOT_VCAP_BIT_ANY ? 0 : 1); -} + u32 offset = ocelot->vcap_is2_keys[field].offset; + u32 length = ocelot->vcap_is2_keys[field].length; + u32 value = (val == OCELOT_VCAP_BIT_1 ? 1 : 0); + u32 msk = (val == OCELOT_VCAP_BIT_ANY ? 0 : 1); -#define VCAP_KEY_SET(fld, val, msk) \ - vcap_key_set(&data, IS2_HKO_##fld, IS2_HKL_##fld, val, msk) -#define VCAP_KEY_ANY_SET(fld) \ - vcap_key_set(&data, IS2_HKO_##fld, IS2_HKL_##fld, 0, 0) -#define VCAP_KEY_BIT_SET(fld, val) vcap_key_bit_set(&data, IS2_HKO_##fld, val) -#define VCAP_KEY_BYTES_SET(fld, val, msk) \ - vcap_key_bytes_set(&data, IS2_HKO_##fld, val, msk, IS2_HKL_##fld / 8) + WARN_ON(length != 1); -static void vcap_action_set(struct vcap_data *data, u32 offset, u32 width, - u32 value) -{ - vcap_data_set(data->action, offset + data->action_offset, width, value); + vcap_key_field_set(data, offset, length, value, msk); } -#define VCAP_ACT_SET(fld, val) \ - vcap_action_set(data, IS2_AO_##fld, IS2_AL_##fld, val) +static void vcap_action_set(struct ocelot *ocelot, struct vcap_data *data, + enum vcap_is2_action_field field, u32 value) +{ + int offset = ocelot->vcap_is2_actions[field].offset; + int length = ocelot->vcap_is2_actions[field].length; + + vcap_data_set(data->action, offset + data->action_offset, length, + value); +} -static void is2_action_set(struct vcap_data *data, - enum ocelot_ace_action action) +static void is2_action_set(struct ocelot *ocelot, struct vcap_data *data, + struct ocelot_ace_rule *ace) { - switch (action) { + switch (ace->action) { case OCELOT_ACL_ACTION_DROP: - VCAP_ACT_SET(PORT_MASK, 0x0); - VCAP_ACT_SET(MASK_MODE, 0x1); - VCAP_ACT_SET(POLICE_ENA, 0x1); - VCAP_ACT_SET(POLICE_IDX, OCELOT_POLICER_DISCARD); - VCAP_ACT_SET(CPU_QU_NUM, 0x0); - VCAP_ACT_SET(CPU_COPY_ENA, 0x0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_PORT_MASK, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_MASK_MODE, 1); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_ENA, 1); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_IDX, + OCELOT_POLICER_DISCARD); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_QU_NUM, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_COPY_ENA, 0); break; case OCELOT_ACL_ACTION_TRAP: - VCAP_ACT_SET(PORT_MASK, 0x0); - VCAP_ACT_SET(MASK_MODE, 0x1); - VCAP_ACT_SET(POLICE_ENA, 0x0); - VCAP_ACT_SET(POLICE_IDX, 0x0); - VCAP_ACT_SET(CPU_QU_NUM, 0x0); - VCAP_ACT_SET(CPU_COPY_ENA, 0x1); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_PORT_MASK, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_MASK_MODE, 1); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_ENA, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_IDX, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_QU_NUM, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_COPY_ENA, 1); + break; + case OCELOT_ACL_ACTION_POLICE: + vcap_action_set(ocelot, data, VCAP_IS2_ACT_PORT_MASK, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_MASK_MODE, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_ENA, 1); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_IDX, + ace->pol_ix); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_QU_NUM, 0); + vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_COPY_ENA, 0); break; } } @@ -329,6 +335,7 @@ static void is2_action_set(struct vcap_data *data, static void is2_entry_set(struct ocelot *ocelot, int ix, struct ocelot_ace_rule *ace) { + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; u32 val, msk, type, type_mask = 0xf, i, count; struct ocelot_ace_vlan *tag = &ace->vlan; struct ocelot_vcap_u64 payload; @@ -344,60 +351,76 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, vcap_cache2action(ocelot, &data); data.tg_sw = VCAP_TG_HALF; - is2_data_get(&data, ix); + is2_data_get(ocelot, &data, ix); data.tg = (data.tg & ~data.tg_mask); if (ace->prio != 0) data.tg |= data.tg_value; data.type = IS2_ACTION_TYPE_NORMAL; - VCAP_KEY_ANY_SET(PAG); - VCAP_KEY_SET(IGR_PORT_MASK, 0, ~BIT(ace->chip_port)); - VCAP_KEY_BIT_SET(FIRST, OCELOT_VCAP_BIT_1); - VCAP_KEY_BIT_SET(HOST_MATCH, OCELOT_VCAP_BIT_ANY); - VCAP_KEY_BIT_SET(L2_MC, ace->dmac_mc); - VCAP_KEY_BIT_SET(L2_BC, ace->dmac_bc); - VCAP_KEY_BIT_SET(VLAN_TAGGED, tag->tagged); - VCAP_KEY_SET(VID, tag->vid.value, tag->vid.mask); - VCAP_KEY_SET(PCP, tag->pcp.value[0], tag->pcp.mask[0]); - VCAP_KEY_BIT_SET(DEI, tag->dei); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_PAG, 0, 0); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_IGR_PORT_MASK, 0, + ~ace->ingress_port_mask); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_FIRST, OCELOT_VCAP_BIT_1); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_HOST_MATCH, + OCELOT_VCAP_BIT_ANY); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L2_MC, ace->dmac_mc); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L2_BC, ace->dmac_bc); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_VLAN_TAGGED, tag->tagged); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_VID, + tag->vid.value, tag->vid.mask); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_PCP, + tag->pcp.value[0], tag->pcp.mask[0]); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_DEI, tag->dei); switch (ace->type) { case OCELOT_ACE_TYPE_ETYPE: { struct ocelot_ace_frame_etype *etype = &ace->frame.etype; type = IS2_TYPE_ETYPE; - VCAP_KEY_BYTES_SET(L2_DMAC, etype->dmac.value, - etype->dmac.mask); - VCAP_KEY_BYTES_SET(L2_SMAC, etype->smac.value, - etype->smac.mask); - VCAP_KEY_BYTES_SET(MAC_ETYPE_ETYPE, etype->etype.value, - etype->etype.mask); - VCAP_KEY_ANY_SET(MAC_ETYPE_L2_PAYLOAD); // Clear unused bits - vcap_key_bytes_set(&data, IS2_HKO_MAC_ETYPE_L2_PAYLOAD, - etype->data.value, etype->data.mask, 2); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_DMAC, + etype->dmac.value, etype->dmac.mask); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_SMAC, + etype->smac.value, etype->smac.mask); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_MAC_ETYPE_ETYPE, + etype->etype.value, etype->etype.mask); + /* Clear unused bits */ + vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD0, + 0, 0); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD1, + 0, 0); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD2, + 0, 0); + vcap_key_bytes_set(ocelot, &data, + VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD0, + etype->data.value, etype->data.mask); break; } case OCELOT_ACE_TYPE_LLC: { struct ocelot_ace_frame_llc *llc = &ace->frame.llc; type = IS2_TYPE_LLC; - VCAP_KEY_BYTES_SET(L2_DMAC, llc->dmac.value, llc->dmac.mask); - VCAP_KEY_BYTES_SET(L2_SMAC, llc->smac.value, llc->smac.mask); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_DMAC, + llc->dmac.value, llc->dmac.mask); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_SMAC, + llc->smac.value, llc->smac.mask); for (i = 0; i < 4; i++) { payload.value[i] = llc->llc.value[i]; payload.mask[i] = llc->llc.mask[i]; } - VCAP_KEY_BYTES_SET(MAC_LLC_L2_LLC, payload.value, payload.mask); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_MAC_LLC_L2_LLC, + payload.value, payload.mask); break; } case OCELOT_ACE_TYPE_SNAP: { struct ocelot_ace_frame_snap *snap = &ace->frame.snap; type = IS2_TYPE_SNAP; - VCAP_KEY_BYTES_SET(L2_DMAC, snap->dmac.value, snap->dmac.mask); - VCAP_KEY_BYTES_SET(L2_SMAC, snap->smac.value, snap->smac.mask); - VCAP_KEY_BYTES_SET(MAC_SNAP_L2_SNAP, + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_DMAC, + snap->dmac.value, snap->dmac.mask); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_SMAC, + snap->smac.value, snap->smac.mask); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_MAC_SNAP_L2_SNAP, ace->frame.snap.snap.value, ace->frame.snap.snap.mask); break; @@ -406,26 +429,42 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, struct ocelot_ace_frame_arp *arp = &ace->frame.arp; type = IS2_TYPE_ARP; - VCAP_KEY_BYTES_SET(MAC_ARP_L2_SMAC, arp->smac.value, - arp->smac.mask); - VCAP_KEY_BIT_SET(MAC_ARP_ARP_ADDR_SPACE_OK, arp->ethernet); - VCAP_KEY_BIT_SET(MAC_ARP_ARP_PROTO_SPACE_OK, arp->ip); - VCAP_KEY_BIT_SET(MAC_ARP_ARP_LEN_OK, arp->length); - VCAP_KEY_BIT_SET(MAC_ARP_ARP_TGT_MATCH, arp->dmac_match); - VCAP_KEY_BIT_SET(MAC_ARP_ARP_SENDER_MATCH, arp->smac_match); - VCAP_KEY_BIT_SET(MAC_ARP_ARP_OPCODE_UNKNOWN, arp->unknown); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_MAC_ARP_SMAC, + arp->smac.value, arp->smac.mask); + vcap_key_bit_set(ocelot, &data, + VCAP_IS2_HK_MAC_ARP_ADDR_SPACE_OK, + arp->ethernet); + vcap_key_bit_set(ocelot, &data, + VCAP_IS2_HK_MAC_ARP_PROTO_SPACE_OK, + arp->ip); + vcap_key_bit_set(ocelot, &data, + VCAP_IS2_HK_MAC_ARP_LEN_OK, + arp->length); + vcap_key_bit_set(ocelot, &data, + VCAP_IS2_HK_MAC_ARP_TARGET_MATCH, + arp->dmac_match); + vcap_key_bit_set(ocelot, &data, + VCAP_IS2_HK_MAC_ARP_SENDER_MATCH, + arp->smac_match); + vcap_key_bit_set(ocelot, &data, + VCAP_IS2_HK_MAC_ARP_OPCODE_UNKNOWN, + arp->unknown); /* OPCODE is inverse, bit 0 is reply flag, bit 1 is RARP flag */ val = ((arp->req == OCELOT_VCAP_BIT_0 ? 1 : 0) | (arp->arp == OCELOT_VCAP_BIT_0 ? 2 : 0)); msk = ((arp->req == OCELOT_VCAP_BIT_ANY ? 0 : 1) | (arp->arp == OCELOT_VCAP_BIT_ANY ? 0 : 2)); - VCAP_KEY_SET(MAC_ARP_ARP_OPCODE, val, msk); - vcap_key_bytes_set(&data, IS2_HKO_MAC_ARP_L3_IP4_DIP, - arp->dip.value.addr, arp->dip.mask.addr, 4); - vcap_key_bytes_set(&data, IS2_HKO_MAC_ARP_L3_IP4_SIP, - arp->sip.value.addr, arp->sip.mask.addr, 4); - VCAP_KEY_ANY_SET(MAC_ARP_DIP_EQ_SIP); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ARP_OPCODE, + val, msk); + vcap_key_bytes_set(ocelot, &data, + VCAP_IS2_HK_MAC_ARP_L3_IP4_DIP, + arp->dip.value.addr, arp->dip.mask.addr); + vcap_key_bytes_set(ocelot, &data, + VCAP_IS2_HK_MAC_ARP_L3_IP4_SIP, + arp->sip.value.addr, arp->sip.mask.addr); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ARP_DIP_EQ_SIP, + 0, 0); break; } case OCELOT_ACE_TYPE_IPV4: @@ -493,18 +532,23 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, seq_zero = ipv6->seq_zero; } - VCAP_KEY_BIT_SET(IP4, + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_IP4, ipv4 ? OCELOT_VCAP_BIT_1 : OCELOT_VCAP_BIT_0); - VCAP_KEY_BIT_SET(L3_FRAGMENT, fragment); - VCAP_KEY_ANY_SET(L3_FRAG_OFS_GT0); - VCAP_KEY_BIT_SET(L3_OPTIONS, options); - VCAP_KEY_BIT_SET(L3_TTL_GT0, ttl); - VCAP_KEY_BYTES_SET(L3_TOS, ds.value, ds.mask); - vcap_key_bytes_set(&data, IS2_HKO_L3_IP4_DIP, dip.value.addr, - dip.mask.addr, 4); - vcap_key_bytes_set(&data, IS2_HKO_L3_IP4_SIP, sip.value.addr, - sip.mask.addr, 4); - VCAP_KEY_BIT_SET(DIP_EQ_SIP, sip_eq_dip); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L3_FRAGMENT, + fragment); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_L3_FRAG_OFS_GT0, 0, 0); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L3_OPTIONS, + options); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_IP4_L3_TTL_GT0, + ttl); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L3_TOS, + ds.value, ds.mask); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L3_IP4_DIP, + dip.value.addr, dip.mask.addr); + vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L3_IP4_SIP, + sip.value.addr, sip.mask.addr); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_DIP_EQ_SIP, + sip_eq_dip); val = proto.value[0]; msk = proto.mask[0]; type = IS2_TYPE_IP_UDP_TCP; @@ -512,25 +556,34 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, /* UDP/TCP protocol match */ tcp = (val == 6 ? OCELOT_VCAP_BIT_1 : OCELOT_VCAP_BIT_0); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_TCP, tcp); - vcap_key_l4_port_set(&data, - IS2_HKO_IP4_TCP_UDP_L4_DPORT, - dport); - vcap_key_l4_port_set(&data, - IS2_HKO_IP4_TCP_UDP_L4_SPORT, - sport); - VCAP_KEY_ANY_SET(IP4_TCP_UDP_L4_RNG); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_SPORT_EQ_DPORT, + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_TCP, tcp); + vcap_key_l4_port_set(ocelot, &data, + VCAP_IS2_HK_L4_DPORT, dport); + vcap_key_l4_port_set(ocelot, &data, + VCAP_IS2_HK_L4_SPORT, sport); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_L4_RNG, 0, 0); + vcap_key_bit_set(ocelot, &data, + VCAP_IS2_HK_L4_SPORT_EQ_DPORT, sport_eq_dport); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_SEQUENCE_EQ0, seq_zero); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_L4_FIN, tcp_fin); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_L4_SYN, tcp_syn); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_L4_RST, tcp_rst); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_L4_PSH, tcp_psh); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_L4_ACK, tcp_ack); - VCAP_KEY_BIT_SET(IP4_TCP_UDP_L4_URG, tcp_urg); - VCAP_KEY_ANY_SET(IP4_TCP_UDP_L4_1588_DOM); - VCAP_KEY_ANY_SET(IP4_TCP_UDP_L4_1588_VER); + vcap_key_bit_set(ocelot, &data, + VCAP_IS2_HK_L4_SEQUENCE_EQ0, + seq_zero); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_FIN, + tcp_fin); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_SYN, + tcp_syn); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_RST, + tcp_rst); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_PSH, + tcp_psh); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_ACK, + tcp_ack); + vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_URG, + tcp_urg); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_L4_1588_DOM, + 0, 0); + vcap_key_set(ocelot, &data, VCAP_IS2_HK_L4_1588_VER, + 0, 0); } else { if (msk == 0) { /* Any IP protocol match */ @@ -543,10 +596,12 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, payload.mask[i] = ip_data->mask[i]; } } - VCAP_KEY_BYTES_SET(IP4_OTHER_L3_PROTO, proto.value, - proto.mask); - VCAP_KEY_BYTES_SET(IP4_OTHER_L3_PAYLOAD, payload.value, - payload.mask); + vcap_key_bytes_set(ocelot, &data, + VCAP_IS2_HK_IP4_L3_PROTO, + proto.value, proto.mask); + vcap_key_bytes_set(ocelot, &data, + VCAP_IS2_HK_L3_PAYLOAD, + payload.value, payload.mask); } break; } @@ -554,19 +609,21 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, default: type = 0; type_mask = 0; - count = (vcap_is2.entry_width / 2); - for (i = (IS2_HKO_PCP + IS2_HKL_PCP); i < count; - i += ENTRY_WIDTH) { - /* Clear entry data */ - vcap_key_set(&data, i, min(32u, count - i), 0, 0); + count = vcap_is2->entry_width / 2; + /* Iterate over the non-common part of the key and + * clear entry data + */ + for (i = ocelot->vcap_is2_keys[VCAP_IS2_HK_L2_DMAC].offset; + i < count; i += ENTRY_WIDTH) { + vcap_key_field_set(&data, i, min(32u, count - i), 0, 0); } break; } - VCAP_KEY_SET(TYPE, type, type_mask); - is2_action_set(&data, ace->action); - vcap_data_set(data.counter, data.counter_offset, vcap_is2.counter_width, - ace->stats.pkts); + vcap_key_set(ocelot, &data, VCAP_IS2_TYPE, type, type_mask); + is2_action_set(ocelot, &data, ace); + vcap_data_set(data.counter, data.counter_offset, + vcap_is2->counter_width, ace->stats.pkts); /* Write row */ vcap_entry2cache(ocelot, &data); @@ -574,29 +631,37 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, vcap_row_cmd(ocelot, row, VCAP_CMD_WRITE, VCAP_SEL_ALL); } -static void is2_entry_get(struct ocelot_ace_rule *rule, int ix) +static void is2_entry_get(struct ocelot *ocelot, struct ocelot_ace_rule *rule, + int ix) { - struct ocelot *op = rule->port->ocelot; + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; struct vcap_data data; int row = (ix / 2); u32 cnt; - vcap_row_cmd(op, row, VCAP_CMD_READ, VCAP_SEL_COUNTER); - vcap_cache2action(op, &data); + vcap_row_cmd(ocelot, row, VCAP_CMD_READ, VCAP_SEL_COUNTER); + vcap_cache2action(ocelot, &data); data.tg_sw = VCAP_TG_HALF; - is2_data_get(&data, ix); + is2_data_get(ocelot, &data, ix); cnt = vcap_data_get(data.counter, data.counter_offset, - vcap_is2.counter_width); + vcap_is2->counter_width); rule->stats.pkts = cnt; } -static void ocelot_ace_rule_add(struct ocelot_acl_block *block, +static void ocelot_ace_rule_add(struct ocelot *ocelot, + struct ocelot_acl_block *block, struct ocelot_ace_rule *rule) { struct ocelot_ace_rule *tmp; struct list_head *pos, *n; + if (rule->action == OCELOT_ACL_ACTION_POLICE) { + block->pol_lpr--; + rule->pol_ix = block->pol_lpr; + ocelot_ace_policer_add(ocelot, rule->pol_ix, &rule->pol); + } + block->count++; if (list_empty(&block->rules)) { @@ -641,29 +706,57 @@ ocelot_ace_rule_get_rule_index(struct ocelot_acl_block *block, int index) return NULL; } -int ocelot_ace_rule_offload_add(struct ocelot_ace_rule *rule) +int ocelot_ace_rule_offload_add(struct ocelot *ocelot, + struct ocelot_ace_rule *rule) { + struct ocelot_acl_block *block = &ocelot->acl_block; struct ocelot_ace_rule *ace; int i, index; /* Add rule to the linked list */ - ocelot_ace_rule_add(acl_block, rule); + ocelot_ace_rule_add(ocelot, block, rule); /* Get the index of the inserted rule */ - index = ocelot_ace_rule_get_index_id(acl_block, rule); + index = ocelot_ace_rule_get_index_id(block, rule); /* Move down the rules to make place for the new rule */ - for (i = acl_block->count - 1; i > index; i--) { - ace = ocelot_ace_rule_get_rule_index(acl_block, i); - is2_entry_set(rule->port->ocelot, i, ace); + for (i = block->count - 1; i > index; i--) { + ace = ocelot_ace_rule_get_rule_index(block, i); + is2_entry_set(ocelot, i, ace); } /* Now insert the new rule */ - is2_entry_set(rule->port->ocelot, index, rule); + is2_entry_set(ocelot, index, rule); return 0; } -static void ocelot_ace_rule_del(struct ocelot_acl_block *block, +static void ocelot_ace_police_del(struct ocelot *ocelot, + struct ocelot_acl_block *block, + u32 ix) +{ + struct ocelot_ace_rule *ace; + int index = -1; + + if (ix < block->pol_lpr) + return; + + list_for_each_entry(ace, &block->rules, list) { + index++; + if (ace->action == OCELOT_ACL_ACTION_POLICE && + ace->pol_ix < ix) { + ace->pol_ix += 1; + ocelot_ace_policer_add(ocelot, ace->pol_ix, + &ace->pol); + is2_entry_set(ocelot, index, ace); + } + } + + ocelot_ace_policer_del(ocelot, block->pol_lpr); + block->pol_lpr++; +} + +static void ocelot_ace_rule_del(struct ocelot *ocelot, + struct ocelot_acl_block *block, struct ocelot_ace_rule *rule) { struct ocelot_ace_rule *tmp; @@ -672,6 +765,10 @@ static void ocelot_ace_rule_del(struct ocelot_acl_block *block, list_for_each_safe(pos, q, &block->rules) { tmp = list_entry(pos, struct ocelot_ace_rule, list); if (tmp->id == rule->id) { + if (tmp->action == OCELOT_ACL_ACTION_POLICE) + ocelot_ace_police_del(ocelot, block, + tmp->pol_ix); + list_del(pos); kfree(tmp); } @@ -680,8 +777,10 @@ static void ocelot_ace_rule_del(struct ocelot_acl_block *block, block->count--; } -int ocelot_ace_rule_offload_del(struct ocelot_ace_rule *rule) +int ocelot_ace_rule_offload_del(struct ocelot *ocelot, + struct ocelot_ace_rule *rule) { + struct ocelot_acl_block *block = &ocelot->acl_block; struct ocelot_ace_rule del_ace; struct ocelot_ace_rule *ace; int i, index; @@ -689,70 +788,55 @@ int ocelot_ace_rule_offload_del(struct ocelot_ace_rule *rule) memset(&del_ace, 0, sizeof(del_ace)); /* Gets index of the rule */ - index = ocelot_ace_rule_get_index_id(acl_block, rule); + index = ocelot_ace_rule_get_index_id(block, rule); /* Delete rule */ - ocelot_ace_rule_del(acl_block, rule); + ocelot_ace_rule_del(ocelot, block, rule); /* Move up all the blocks over the deleted rule */ - for (i = index; i < acl_block->count; i++) { - ace = ocelot_ace_rule_get_rule_index(acl_block, i); - is2_entry_set(rule->port->ocelot, i, ace); + for (i = index; i < block->count; i++) { + ace = ocelot_ace_rule_get_rule_index(block, i); + is2_entry_set(ocelot, i, ace); } /* Now delete the last rule, because it is duplicated */ - is2_entry_set(rule->port->ocelot, acl_block->count, &del_ace); + is2_entry_set(ocelot, block->count, &del_ace); return 0; } -int ocelot_ace_rule_stats_update(struct ocelot_ace_rule *rule) +int ocelot_ace_rule_stats_update(struct ocelot *ocelot, + struct ocelot_ace_rule *rule) { + struct ocelot_acl_block *block = &ocelot->acl_block; struct ocelot_ace_rule *tmp; int index; - index = ocelot_ace_rule_get_index_id(acl_block, rule); - is2_entry_get(rule, index); + index = ocelot_ace_rule_get_index_id(block, rule); + is2_entry_get(ocelot, rule, index); /* After we get the result we need to clear the counters */ - tmp = ocelot_ace_rule_get_rule_index(acl_block, index); + tmp = ocelot_ace_rule_get_rule_index(block, index); tmp->stats.pkts = 0; - is2_entry_set(rule->port->ocelot, index, tmp); + is2_entry_set(ocelot, index, tmp); return 0; } -static struct ocelot_acl_block *ocelot_acl_block_create(struct ocelot *ocelot) -{ - struct ocelot_acl_block *block; - - block = kzalloc(sizeof(*block), GFP_KERNEL); - if (!block) - return NULL; - - INIT_LIST_HEAD(&block->rules); - block->count = 0; - block->ocelot = ocelot; - - return block; -} - -static void ocelot_acl_block_destroy(struct ocelot_acl_block *block) -{ - kfree(block); -} - int ocelot_ace_init(struct ocelot *ocelot) { + const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + struct ocelot_acl_block *block = &ocelot->acl_block; struct vcap_data data; memset(&data, 0, sizeof(data)); + vcap_entry2cache(ocelot, &data); - ocelot_write(ocelot, vcap_is2.entry_count, S2_CORE_MV_CFG); + ocelot_write(ocelot, vcap_is2->entry_count, S2_CORE_MV_CFG); vcap_cmd(ocelot, 0, VCAP_CMD_INITIALIZE, VCAP_SEL_ENTRY); vcap_action2cache(ocelot, &data); - ocelot_write(ocelot, vcap_is2.action_count, S2_CORE_MV_CFG); + ocelot_write(ocelot, vcap_is2->action_count, S2_CORE_MV_CFG); vcap_cmd(ocelot, 0, VCAP_CMD_INITIALIZE, VCAP_SEL_ACTION | VCAP_SEL_COUNTER); @@ -771,12 +855,9 @@ int ocelot_ace_init(struct ocelot *ocelot) ocelot_write_gix(ocelot, 0x3fffff, ANA_POL_CIR_STATE, OCELOT_POLICER_DISCARD); - acl_block = ocelot_acl_block_create(ocelot); + block->pol_lpr = OCELOT_POLICER_DISCARD - 1; - return 0; -} + INIT_LIST_HEAD(&ocelot->acl_block.rules); -void ocelot_ace_deinit(void) -{ - ocelot_acl_block_destroy(acl_block); + return 0; } diff --git a/drivers/net/ethernet/mscc/ocelot_ace.h b/drivers/net/ethernet/mscc/ocelot_ace.h index c08e3e8482e7..29d22c566786 100644 --- a/drivers/net/ethernet/mscc/ocelot_ace.h +++ b/drivers/net/ethernet/mscc/ocelot_ace.h @@ -7,6 +7,7 @@ #define _MSCC_OCELOT_ACE_H_ #include "ocelot.h" +#include "ocelot_police.h" #include <net/sch_generic.h> #include <net/pkt_cls.h> @@ -176,6 +177,7 @@ struct ocelot_ace_frame_ipv6 { enum ocelot_ace_action { OCELOT_ACL_ACTION_DROP, OCELOT_ACL_ACTION_TRAP, + OCELOT_ACL_ACTION_POLICE, }; struct ocelot_ace_stats { @@ -186,14 +188,13 @@ struct ocelot_ace_stats { struct ocelot_ace_rule { struct list_head list; - struct ocelot_port *port; u16 prio; u32 id; enum ocelot_ace_action action; struct ocelot_ace_stats stats; - int chip_port; + u16 ingress_port_mask; enum ocelot_vcap_bit dmac_mc; enum ocelot_vcap_bit dmac_bc; @@ -209,24 +210,21 @@ struct ocelot_ace_rule { struct ocelot_ace_frame_ipv4 ipv4; struct ocelot_ace_frame_ipv6 ipv6; } frame; + struct ocelot_policer pol; + u32 pol_ix; }; -struct ocelot_acl_block { - struct list_head rules; - struct ocelot *ocelot; - int count; -}; - -int ocelot_ace_rule_offload_add(struct ocelot_ace_rule *rule); -int ocelot_ace_rule_offload_del(struct ocelot_ace_rule *rule); -int ocelot_ace_rule_stats_update(struct ocelot_ace_rule *rule); +int ocelot_ace_rule_offload_add(struct ocelot *ocelot, + struct ocelot_ace_rule *rule); +int ocelot_ace_rule_offload_del(struct ocelot *ocelot, + struct ocelot_ace_rule *rule); +int ocelot_ace_rule_stats_update(struct ocelot *ocelot, + struct ocelot_ace_rule *rule); int ocelot_ace_init(struct ocelot *ocelot); -void ocelot_ace_deinit(void); -int ocelot_setup_tc_block_flower_bind(struct ocelot_port_private *priv, - struct flow_block_offload *f); -void ocelot_setup_tc_block_flower_unbind(struct ocelot_port_private *priv, - struct flow_block_offload *f); +int ocelot_setup_tc_cls_flower(struct ocelot_port_private *priv, + struct flow_cls_offload *f, + bool ingress); #endif /* _MSCC_OCELOT_ACE_H_ */ diff --git a/drivers/net/ethernet/mscc/ocelot_board.c b/drivers/net/ethernet/mscc/ocelot_board.c index 1135a18019c7..0ac9fbf77a01 100644 --- a/drivers/net/ethernet/mscc/ocelot_board.c +++ b/drivers/net/ethernet/mscc/ocelot_board.c @@ -14,9 +14,14 @@ #include <linux/skbuff.h> #include <net/switchdev.h> +#include <soc/mscc/ocelot_vcap.h> #include "ocelot.h" #define IFH_EXTRACT_BITFIELD64(x, o, w) (((x) >> (o)) & GENMASK_ULL((w) - 1, 0)) +#define VSC7514_VCAP_IS2_CNT 64 +#define VSC7514_VCAP_IS2_ENTRY_WIDTH 376 +#define VSC7514_VCAP_IS2_ACTION_WIDTH 99 +#define VSC7514_VCAP_PORT_CNT 11 static int ocelot_parse_ifh(u32 *_ifh, struct frame_info *info) { @@ -211,29 +216,6 @@ static const struct of_device_id mscc_ocelot_match[] = { }; MODULE_DEVICE_TABLE(of, mscc_ocelot_match); -static void ocelot_port_pcs_init(struct ocelot *ocelot, int port) -{ - struct ocelot_port *ocelot_port = ocelot->ports[port]; - - /* Disable HDX fast control */ - ocelot_port_writel(ocelot_port, DEV_PORT_MISC_HDX_FAST_DIS, - DEV_PORT_MISC); - - /* SGMII only for now */ - ocelot_port_writel(ocelot_port, PCS1G_MODE_CFG_SGMII_MODE_ENA, - PCS1G_MODE_CFG); - ocelot_port_writel(ocelot_port, PCS1G_SD_CFG_SD_SEL, PCS1G_SD_CFG); - - /* Enable PCS */ - ocelot_port_writel(ocelot_port, PCS1G_CFG_PCS_ENA, PCS1G_CFG); - - /* No aneg on SGMII */ - ocelot_port_writel(ocelot_port, 0, PCS1G_ANEG_CFG); - - /* No loopback */ - ocelot_port_writel(ocelot_port, 0, PCS1G_LB_CFG); -} - static int ocelot_reset(struct ocelot *ocelot) { int retries = 100; @@ -258,10 +240,132 @@ static int ocelot_reset(struct ocelot *ocelot) } static const struct ocelot_ops ocelot_ops = { - .pcs_init = ocelot_port_pcs_init, .reset = ocelot_reset, }; +static const struct vcap_field vsc7514_vcap_is2_keys[] = { + /* Common: 46 bits */ + [VCAP_IS2_TYPE] = { 0, 4}, + [VCAP_IS2_HK_FIRST] = { 4, 1}, + [VCAP_IS2_HK_PAG] = { 5, 8}, + [VCAP_IS2_HK_IGR_PORT_MASK] = { 13, 12}, + [VCAP_IS2_HK_RSV2] = { 25, 1}, + [VCAP_IS2_HK_HOST_MATCH] = { 26, 1}, + [VCAP_IS2_HK_L2_MC] = { 27, 1}, + [VCAP_IS2_HK_L2_BC] = { 28, 1}, + [VCAP_IS2_HK_VLAN_TAGGED] = { 29, 1}, + [VCAP_IS2_HK_VID] = { 30, 12}, + [VCAP_IS2_HK_DEI] = { 42, 1}, + [VCAP_IS2_HK_PCP] = { 43, 3}, + /* MAC_ETYPE / MAC_LLC / MAC_SNAP / OAM common */ + [VCAP_IS2_HK_L2_DMAC] = { 46, 48}, + [VCAP_IS2_HK_L2_SMAC] = { 94, 48}, + /* MAC_ETYPE (TYPE=000) */ + [VCAP_IS2_HK_MAC_ETYPE_ETYPE] = {142, 16}, + [VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD0] = {158, 16}, + [VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD1] = {174, 8}, + [VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD2] = {182, 3}, + /* MAC_LLC (TYPE=001) */ + [VCAP_IS2_HK_MAC_LLC_L2_LLC] = {142, 40}, + /* MAC_SNAP (TYPE=010) */ + [VCAP_IS2_HK_MAC_SNAP_L2_SNAP] = {142, 40}, + /* MAC_ARP (TYPE=011) */ + [VCAP_IS2_HK_MAC_ARP_SMAC] = { 46, 48}, + [VCAP_IS2_HK_MAC_ARP_ADDR_SPACE_OK] = { 94, 1}, + [VCAP_IS2_HK_MAC_ARP_PROTO_SPACE_OK] = { 95, 1}, + [VCAP_IS2_HK_MAC_ARP_LEN_OK] = { 96, 1}, + [VCAP_IS2_HK_MAC_ARP_TARGET_MATCH] = { 97, 1}, + [VCAP_IS2_HK_MAC_ARP_SENDER_MATCH] = { 98, 1}, + [VCAP_IS2_HK_MAC_ARP_OPCODE_UNKNOWN] = { 99, 1}, + [VCAP_IS2_HK_MAC_ARP_OPCODE] = {100, 2}, + [VCAP_IS2_HK_MAC_ARP_L3_IP4_DIP] = {102, 32}, + [VCAP_IS2_HK_MAC_ARP_L3_IP4_SIP] = {134, 32}, + [VCAP_IS2_HK_MAC_ARP_DIP_EQ_SIP] = {166, 1}, + /* IP4_TCP_UDP / IP4_OTHER common */ + [VCAP_IS2_HK_IP4] = { 46, 1}, + [VCAP_IS2_HK_L3_FRAGMENT] = { 47, 1}, + [VCAP_IS2_HK_L3_FRAG_OFS_GT0] = { 48, 1}, + [VCAP_IS2_HK_L3_OPTIONS] = { 49, 1}, + [VCAP_IS2_HK_IP4_L3_TTL_GT0] = { 50, 1}, + [VCAP_IS2_HK_L3_TOS] = { 51, 8}, + [VCAP_IS2_HK_L3_IP4_DIP] = { 59, 32}, + [VCAP_IS2_HK_L3_IP4_SIP] = { 91, 32}, + [VCAP_IS2_HK_DIP_EQ_SIP] = {123, 1}, + /* IP4_TCP_UDP (TYPE=100) */ + [VCAP_IS2_HK_TCP] = {124, 1}, + [VCAP_IS2_HK_L4_SPORT] = {125, 16}, + [VCAP_IS2_HK_L4_DPORT] = {141, 16}, + [VCAP_IS2_HK_L4_RNG] = {157, 8}, + [VCAP_IS2_HK_L4_SPORT_EQ_DPORT] = {165, 1}, + [VCAP_IS2_HK_L4_SEQUENCE_EQ0] = {166, 1}, + [VCAP_IS2_HK_L4_URG] = {167, 1}, + [VCAP_IS2_HK_L4_ACK] = {168, 1}, + [VCAP_IS2_HK_L4_PSH] = {169, 1}, + [VCAP_IS2_HK_L4_RST] = {170, 1}, + [VCAP_IS2_HK_L4_SYN] = {171, 1}, + [VCAP_IS2_HK_L4_FIN] = {172, 1}, + [VCAP_IS2_HK_L4_1588_DOM] = {173, 8}, + [VCAP_IS2_HK_L4_1588_VER] = {181, 4}, + /* IP4_OTHER (TYPE=101) */ + [VCAP_IS2_HK_IP4_L3_PROTO] = {124, 8}, + [VCAP_IS2_HK_L3_PAYLOAD] = {132, 56}, + /* IP6_STD (TYPE=110) */ + [VCAP_IS2_HK_IP6_L3_TTL_GT0] = { 46, 1}, + [VCAP_IS2_HK_L3_IP6_SIP] = { 47, 128}, + [VCAP_IS2_HK_IP6_L3_PROTO] = {175, 8}, + /* OAM (TYPE=111) */ + [VCAP_IS2_HK_OAM_MEL_FLAGS] = {142, 7}, + [VCAP_IS2_HK_OAM_VER] = {149, 5}, + [VCAP_IS2_HK_OAM_OPCODE] = {154, 8}, + [VCAP_IS2_HK_OAM_FLAGS] = {162, 8}, + [VCAP_IS2_HK_OAM_MEPID] = {170, 16}, + [VCAP_IS2_HK_OAM_CCM_CNTS_EQ0] = {186, 1}, + [VCAP_IS2_HK_OAM_IS_Y1731] = {187, 1}, +}; + +static const struct vcap_field vsc7514_vcap_is2_actions[] = { + [VCAP_IS2_ACT_HIT_ME_ONCE] = { 0, 1}, + [VCAP_IS2_ACT_CPU_COPY_ENA] = { 1, 1}, + [VCAP_IS2_ACT_CPU_QU_NUM] = { 2, 3}, + [VCAP_IS2_ACT_MASK_MODE] = { 5, 2}, + [VCAP_IS2_ACT_MIRROR_ENA] = { 7, 1}, + [VCAP_IS2_ACT_LRN_DIS] = { 8, 1}, + [VCAP_IS2_ACT_POLICE_ENA] = { 9, 1}, + [VCAP_IS2_ACT_POLICE_IDX] = { 10, 9}, + [VCAP_IS2_ACT_POLICE_VCAP_ONLY] = { 19, 1}, + [VCAP_IS2_ACT_PORT_MASK] = { 20, 11}, + [VCAP_IS2_ACT_REW_OP] = { 31, 9}, + [VCAP_IS2_ACT_SMAC_REPLACE_ENA] = { 40, 1}, + [VCAP_IS2_ACT_RSV] = { 41, 2}, + [VCAP_IS2_ACT_ACL_ID] = { 43, 6}, + [VCAP_IS2_ACT_HIT_CNT] = { 49, 32}, +}; + +static const struct vcap_props vsc7514_vcap_props[] = { + [VCAP_IS2] = { + .tg_width = 2, + .sw_count = 4, + .entry_count = VSC7514_VCAP_IS2_CNT, + .entry_width = VSC7514_VCAP_IS2_ENTRY_WIDTH, + .action_count = VSC7514_VCAP_IS2_CNT + + VSC7514_VCAP_PORT_CNT + 2, + .action_width = 99, + .action_type_width = 1, + .action_table = { + [IS2_ACTION_TYPE_NORMAL] = { + .width = 49, + .count = 2 + }, + [IS2_ACTION_TYPE_SMAC_SIP] = { + .width = 6, + .count = 4 + }, + }, + .counter_words = 4, + .counter_width = 32, + }, +}; + static int mscc_ocelot_probe(struct platform_device *pdev) { struct device_node *np = pdev->dev.of_node; @@ -349,8 +453,6 @@ static int mscc_ocelot_probe(struct platform_device *pdev) ocelot->ptp = 1; } - ocelot->num_cpu_ports = 1; /* 1 port on the switch, two groups */ - ports = of_get_child_by_name(np, "ethernet-ports"); if (!ports) { dev_err(&pdev->dev, "no ethernet-ports child node found\n"); @@ -362,9 +464,14 @@ static int mscc_ocelot_probe(struct platform_device *pdev) ocelot->ports = devm_kcalloc(&pdev->dev, ocelot->num_phys_ports, sizeof(struct ocelot_port *), GFP_KERNEL); + ocelot->vcap_is2_keys = vsc7514_vcap_is2_keys; + ocelot->vcap_is2_actions = vsc7514_vcap_is2_actions; + ocelot->vcap = vsc7514_vcap_props; + ocelot_init(ocelot); - ocelot_set_cpu_port(ocelot, ocelot->num_phys_ports, - OCELOT_TAG_PREFIX_NONE, OCELOT_TAG_PREFIX_NONE); + /* No NPI port */ + ocelot_configure_cpu(ocelot, -1, OCELOT_TAG_PREFIX_NONE, + OCELOT_TAG_PREFIX_NONE); for_each_available_child_of_node(ports, portnp) { struct ocelot_port_private *priv; diff --git a/drivers/net/ethernet/mscc/ocelot_flower.c b/drivers/net/ethernet/mscc/ocelot_flower.c index 3d65b99b9734..341923311fec 100644 --- a/drivers/net/ethernet/mscc/ocelot_flower.c +++ b/drivers/net/ethernet/mscc/ocelot_flower.c @@ -8,27 +8,35 @@ #include "ocelot_ace.h" -struct ocelot_port_block { - struct ocelot_acl_block *block; - struct ocelot_port_private *priv; -}; - static int ocelot_flower_parse_action(struct flow_cls_offload *f, - struct ocelot_ace_rule *rule) + struct ocelot_ace_rule *ace) { const struct flow_action_entry *a; + s64 burst; + u64 rate; int i; - if (f->rule->action.num_entries != 1) + if (!flow_offload_has_one_action(&f->rule->action)) + return -EOPNOTSUPP; + + if (!flow_action_basic_hw_stats_check(&f->rule->action, + f->common.extack)) return -EOPNOTSUPP; flow_action_for_each(i, a, &f->rule->action) { switch (a->id) { case FLOW_ACTION_DROP: - rule->action = OCELOT_ACL_ACTION_DROP; + ace->action = OCELOT_ACL_ACTION_DROP; break; case FLOW_ACTION_TRAP: - rule->action = OCELOT_ACL_ACTION_TRAP; + ace->action = OCELOT_ACL_ACTION_TRAP; + break; + case FLOW_ACTION_POLICE: + ace->action = OCELOT_ACL_ACTION_POLICE; + rate = a->police.rate_bytes_ps; + ace->pol.rate = div_u64(rate, 1000) * 8; + burst = rate * PSCHED_NS2TICKS(a->police.burst); + ace->pol.burst = div_u64(burst, PSCHED_TICKS_PER_SEC); break; default: return -EOPNOTSUPP; @@ -39,7 +47,7 @@ static int ocelot_flower_parse_action(struct flow_cls_offload *f, } static int ocelot_flower_parse(struct flow_cls_offload *f, - struct ocelot_ace_rule *ocelot_rule) + struct ocelot_ace_rule *ace) { struct flow_rule *rule = flow_cls_offload_flow_rule(f); struct flow_dissector *dissector = rule->match.dissector; @@ -84,14 +92,14 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, return -EOPNOTSUPP; flow_rule_match_eth_addrs(rule, &match); - ocelot_rule->type = OCELOT_ACE_TYPE_ETYPE; - ether_addr_copy(ocelot_rule->frame.etype.dmac.value, + ace->type = OCELOT_ACE_TYPE_ETYPE; + ether_addr_copy(ace->frame.etype.dmac.value, match.key->dst); - ether_addr_copy(ocelot_rule->frame.etype.smac.value, + ether_addr_copy(ace->frame.etype.smac.value, match.key->src); - ether_addr_copy(ocelot_rule->frame.etype.dmac.mask, + ether_addr_copy(ace->frame.etype.dmac.mask, match.mask->dst); - ether_addr_copy(ocelot_rule->frame.etype.smac.mask, + ether_addr_copy(ace->frame.etype.smac.mask, match.mask->src); goto finished_key_parsing; } @@ -101,17 +109,17 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, flow_rule_match_basic(rule, &match); if (ntohs(match.key->n_proto) == ETH_P_IP) { - ocelot_rule->type = OCELOT_ACE_TYPE_IPV4; - ocelot_rule->frame.ipv4.proto.value[0] = + ace->type = OCELOT_ACE_TYPE_IPV4; + ace->frame.ipv4.proto.value[0] = match.key->ip_proto; - ocelot_rule->frame.ipv4.proto.mask[0] = + ace->frame.ipv4.proto.mask[0] = match.mask->ip_proto; } if (ntohs(match.key->n_proto) == ETH_P_IPV6) { - ocelot_rule->type = OCELOT_ACE_TYPE_IPV6; - ocelot_rule->frame.ipv6.proto.value[0] = + ace->type = OCELOT_ACE_TYPE_IPV6; + ace->frame.ipv6.proto.value[0] = match.key->ip_proto; - ocelot_rule->frame.ipv6.proto.mask[0] = + ace->frame.ipv6.proto.mask[0] = match.mask->ip_proto; } } @@ -122,16 +130,16 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, u8 *tmp; flow_rule_match_ipv4_addrs(rule, &match); - tmp = &ocelot_rule->frame.ipv4.sip.value.addr[0]; + tmp = &ace->frame.ipv4.sip.value.addr[0]; memcpy(tmp, &match.key->src, 4); - tmp = &ocelot_rule->frame.ipv4.sip.mask.addr[0]; + tmp = &ace->frame.ipv4.sip.mask.addr[0]; memcpy(tmp, &match.mask->src, 4); - tmp = &ocelot_rule->frame.ipv4.dip.value.addr[0]; + tmp = &ace->frame.ipv4.dip.value.addr[0]; memcpy(tmp, &match.key->dst, 4); - tmp = &ocelot_rule->frame.ipv4.dip.mask.addr[0]; + tmp = &ace->frame.ipv4.dip.mask.addr[0]; memcpy(tmp, &match.mask->dst, 4); } @@ -144,213 +152,111 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, struct flow_match_ports match; flow_rule_match_ports(rule, &match); - ocelot_rule->frame.ipv4.sport.value = ntohs(match.key->src); - ocelot_rule->frame.ipv4.sport.mask = ntohs(match.mask->src); - ocelot_rule->frame.ipv4.dport.value = ntohs(match.key->dst); - ocelot_rule->frame.ipv4.dport.mask = ntohs(match.mask->dst); + ace->frame.ipv4.sport.value = ntohs(match.key->src); + ace->frame.ipv4.sport.mask = ntohs(match.mask->src); + ace->frame.ipv4.dport.value = ntohs(match.key->dst); + ace->frame.ipv4.dport.mask = ntohs(match.mask->dst); } if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) { struct flow_match_vlan match; flow_rule_match_vlan(rule, &match); - ocelot_rule->type = OCELOT_ACE_TYPE_ANY; - ocelot_rule->vlan.vid.value = match.key->vlan_id; - ocelot_rule->vlan.vid.mask = match.mask->vlan_id; - ocelot_rule->vlan.pcp.value[0] = match.key->vlan_priority; - ocelot_rule->vlan.pcp.mask[0] = match.mask->vlan_priority; + ace->type = OCELOT_ACE_TYPE_ANY; + ace->vlan.vid.value = match.key->vlan_id; + ace->vlan.vid.mask = match.mask->vlan_id; + ace->vlan.pcp.value[0] = match.key->vlan_priority; + ace->vlan.pcp.mask[0] = match.mask->vlan_priority; } finished_key_parsing: - ocelot_rule->prio = f->common.prio; - ocelot_rule->id = f->cookie; - return ocelot_flower_parse_action(f, ocelot_rule); + ace->prio = f->common.prio; + ace->id = f->cookie; + return ocelot_flower_parse_action(f, ace); } static -struct ocelot_ace_rule *ocelot_ace_rule_create(struct flow_cls_offload *f, - struct ocelot_port_block *block) +struct ocelot_ace_rule *ocelot_ace_rule_create(struct ocelot *ocelot, int port, + struct flow_cls_offload *f) { - struct ocelot_ace_rule *rule; + struct ocelot_ace_rule *ace; - rule = kzalloc(sizeof(*rule), GFP_KERNEL); - if (!rule) + ace = kzalloc(sizeof(*ace), GFP_KERNEL); + if (!ace) return NULL; - rule->port = &block->priv->port; - rule->chip_port = block->priv->chip_port; - return rule; + ace->ingress_port_mask = BIT(port); + return ace; } -static int ocelot_flower_replace(struct flow_cls_offload *f, - struct ocelot_port_block *port_block) +int ocelot_cls_flower_replace(struct ocelot *ocelot, int port, + struct flow_cls_offload *f, bool ingress) { - struct ocelot_ace_rule *rule; + struct ocelot_ace_rule *ace; int ret; - rule = ocelot_ace_rule_create(f, port_block); - if (!rule) + ace = ocelot_ace_rule_create(ocelot, port, f); + if (!ace) return -ENOMEM; - ret = ocelot_flower_parse(f, rule); + ret = ocelot_flower_parse(f, ace); if (ret) { - kfree(rule); + kfree(ace); return ret; } - ret = ocelot_ace_rule_offload_add(rule); - if (ret) - return ret; - - port_block->priv->tc.offload_cnt++; - return 0; + return ocelot_ace_rule_offload_add(ocelot, ace); } +EXPORT_SYMBOL_GPL(ocelot_cls_flower_replace); -static int ocelot_flower_destroy(struct flow_cls_offload *f, - struct ocelot_port_block *port_block) +int ocelot_cls_flower_destroy(struct ocelot *ocelot, int port, + struct flow_cls_offload *f, bool ingress) { - struct ocelot_ace_rule rule; - int ret; - - rule.prio = f->common.prio; - rule.port = &port_block->priv->port; - rule.id = f->cookie; + struct ocelot_ace_rule ace; - ret = ocelot_ace_rule_offload_del(&rule); - if (ret) - return ret; + ace.prio = f->common.prio; + ace.id = f->cookie; - port_block->priv->tc.offload_cnt--; - return 0; + return ocelot_ace_rule_offload_del(ocelot, &ace); } +EXPORT_SYMBOL_GPL(ocelot_cls_flower_destroy); -static int ocelot_flower_stats_update(struct flow_cls_offload *f, - struct ocelot_port_block *port_block) +int ocelot_cls_flower_stats(struct ocelot *ocelot, int port, + struct flow_cls_offload *f, bool ingress) { - struct ocelot_ace_rule rule; + struct ocelot_ace_rule ace; int ret; - rule.prio = f->common.prio; - rule.port = &port_block->priv->port; - rule.id = f->cookie; - ret = ocelot_ace_rule_stats_update(&rule); + ace.prio = f->common.prio; + ace.id = f->cookie; + ret = ocelot_ace_rule_stats_update(ocelot, &ace); if (ret) return ret; - flow_stats_update(&f->stats, 0x0, rule.stats.pkts, 0x0); + flow_stats_update(&f->stats, 0x0, ace.stats.pkts, 0x0, + FLOW_ACTION_HW_STATS_IMMEDIATE); return 0; } +EXPORT_SYMBOL_GPL(ocelot_cls_flower_stats); -static int ocelot_setup_tc_cls_flower(struct flow_cls_offload *f, - struct ocelot_port_block *port_block) +int ocelot_setup_tc_cls_flower(struct ocelot_port_private *priv, + struct flow_cls_offload *f, + bool ingress) { + struct ocelot *ocelot = priv->port.ocelot; + int port = priv->chip_port; + + if (!ingress) + return -EOPNOTSUPP; + switch (f->command) { case FLOW_CLS_REPLACE: - return ocelot_flower_replace(f, port_block); + return ocelot_cls_flower_replace(ocelot, port, f, ingress); case FLOW_CLS_DESTROY: - return ocelot_flower_destroy(f, port_block); + return ocelot_cls_flower_destroy(ocelot, port, f, ingress); case FLOW_CLS_STATS: - return ocelot_flower_stats_update(f, port_block); - default: - return -EOPNOTSUPP; - } -} - -static int ocelot_setup_tc_block_cb_flower(enum tc_setup_type type, - void *type_data, void *cb_priv) -{ - struct ocelot_port_block *port_block = cb_priv; - - if (!tc_cls_can_offload_and_chain0(port_block->priv->dev, type_data)) - return -EOPNOTSUPP; - - switch (type) { - case TC_SETUP_CLSFLOWER: - return ocelot_setup_tc_cls_flower(type_data, cb_priv); - case TC_SETUP_CLSMATCHALL: - return 0; + return ocelot_cls_flower_stats(ocelot, port, f, ingress); default: return -EOPNOTSUPP; } } - -static struct ocelot_port_block* -ocelot_port_block_create(struct ocelot_port_private *priv) -{ - struct ocelot_port_block *port_block; - - port_block = kzalloc(sizeof(*port_block), GFP_KERNEL); - if (!port_block) - return NULL; - - port_block->priv = priv; - - return port_block; -} - -static void ocelot_port_block_destroy(struct ocelot_port_block *block) -{ - kfree(block); -} - -static void ocelot_tc_block_unbind(void *cb_priv) -{ - struct ocelot_port_block *port_block = cb_priv; - - ocelot_port_block_destroy(port_block); -} - -int ocelot_setup_tc_block_flower_bind(struct ocelot_port_private *priv, - struct flow_block_offload *f) -{ - struct ocelot_port_block *port_block; - struct flow_block_cb *block_cb; - int ret; - - if (f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS) - return -EOPNOTSUPP; - - block_cb = flow_block_cb_lookup(f->block, - ocelot_setup_tc_block_cb_flower, priv); - if (!block_cb) { - port_block = ocelot_port_block_create(priv); - if (!port_block) - return -ENOMEM; - - block_cb = flow_block_cb_alloc(ocelot_setup_tc_block_cb_flower, - priv, port_block, - ocelot_tc_block_unbind); - if (IS_ERR(block_cb)) { - ret = PTR_ERR(block_cb); - goto err_cb_register; - } - flow_block_cb_add(block_cb, f); - list_add_tail(&block_cb->driver_list, f->driver_block_list); - } else { - port_block = flow_block_cb_priv(block_cb); - } - - flow_block_cb_incref(block_cb); - return 0; - -err_cb_register: - ocelot_port_block_destroy(port_block); - - return ret; -} - -void ocelot_setup_tc_block_flower_unbind(struct ocelot_port_private *priv, - struct flow_block_offload *f) -{ - struct flow_block_cb *block_cb; - - block_cb = flow_block_cb_lookup(f->block, - ocelot_setup_tc_block_cb_flower, priv); - if (!block_cb) - return; - - if (!flow_block_cb_decref(block_cb)) { - flow_block_cb_remove(block_cb, f); - list_del(&block_cb->driver_list); - } -} diff --git a/drivers/net/ethernet/mscc/ocelot_police.c b/drivers/net/ethernet/mscc/ocelot_police.c index faddce43f2e3..2e1d8e187332 100644 --- a/drivers/net/ethernet/mscc/ocelot_police.c +++ b/drivers/net/ethernet/mscc/ocelot_police.c @@ -4,6 +4,7 @@ * Copyright (c) 2019 Microsemi Corporation */ +#include <soc/mscc/ocelot.h> #include "ocelot_police.h" enum mscc_qos_rate_mode { @@ -203,6 +204,7 @@ int ocelot_port_policer_add(struct ocelot *ocelot, int port, return 0; } +EXPORT_SYMBOL(ocelot_port_policer_add); int ocelot_port_policer_del(struct ocelot *ocelot, int port) { @@ -225,3 +227,28 @@ int ocelot_port_policer_del(struct ocelot *ocelot, int port) return 0; } +EXPORT_SYMBOL(ocelot_port_policer_del); + +int ocelot_ace_policer_add(struct ocelot *ocelot, u32 pol_ix, + struct ocelot_policer *pol) +{ + struct qos_policer_conf pp = { 0 }; + + if (!pol) + return -EINVAL; + + pp.mode = MSCC_QOS_RATE_MODE_DATA; + pp.pir = pol->rate; + pp.pbs = pol->burst; + + return qos_policer_conf_set(ocelot, 0, pol_ix, &pp); +} + +int ocelot_ace_policer_del(struct ocelot *ocelot, u32 pol_ix) +{ + struct qos_policer_conf pp = { 0 }; + + pp.mode = MSCC_QOS_RATE_MODE_DISABLED; + + return qos_policer_conf_set(ocelot, 0, pol_ix, &pp); +} diff --git a/drivers/net/ethernet/mscc/ocelot_police.h b/drivers/net/ethernet/mscc/ocelot_police.h index ae9509229463..792abd28010a 100644 --- a/drivers/net/ethernet/mscc/ocelot_police.h +++ b/drivers/net/ethernet/mscc/ocelot_police.h @@ -9,14 +9,9 @@ #include "ocelot.h" -struct ocelot_policer { - u32 rate; /* kilobit per second */ - u32 burst; /* bytes */ -}; +int ocelot_ace_policer_add(struct ocelot *ocelot, u32 pol_ix, + struct ocelot_policer *pol); -int ocelot_port_policer_add(struct ocelot *ocelot, int port, - struct ocelot_policer *pol); - -int ocelot_port_policer_del(struct ocelot *ocelot, int port); +int ocelot_ace_policer_del(struct ocelot *ocelot, u32 pol_ix); #endif /* _MSCC_OCELOT_POLICE_H_ */ diff --git a/drivers/net/ethernet/mscc/ocelot_tc.c b/drivers/net/ethernet/mscc/ocelot_tc.c index a4f7fbd76507..d326e231f0ad 100644 --- a/drivers/net/ethernet/mscc/ocelot_tc.c +++ b/drivers/net/ethernet/mscc/ocelot_tc.c @@ -4,8 +4,8 @@ * Copyright (c) 2019 Microsemi Corporation */ +#include <soc/mscc/ocelot.h> #include "ocelot_tc.h" -#include "ocelot_police.h" #include "ocelot_ace.h" #include <net/pkt_cls.h> @@ -20,9 +20,6 @@ static int ocelot_setup_tc_cls_matchall(struct ocelot_port_private *priv, int port = priv->chip_port; int err; - netdev_dbg(priv->dev, "%s: port %u command %d cookie %lu\n", - __func__, port, f->command, f->cookie); - if (!ingress) { NL_SET_ERR_MSG_MOD(extack, "Only ingress is supported"); return -EOPNOTSUPP; @@ -99,17 +96,10 @@ static int ocelot_setup_tc_block_cb(enum tc_setup_type type, switch (type) { case TC_SETUP_CLSMATCHALL: - netdev_dbg(priv->dev, "tc_block_cb: TC_SETUP_CLSMATCHALL %s\n", - ingress ? "ingress" : "egress"); - return ocelot_setup_tc_cls_matchall(priv, type_data, ingress); case TC_SETUP_CLSFLOWER: - return 0; + return ocelot_setup_tc_cls_flower(priv, type_data, ingress); default: - netdev_dbg(priv->dev, "tc_block_cb: type %d %s\n", - type, - ingress ? "ingress" : "egress"); - return -EOPNOTSUPP; } } @@ -137,10 +127,6 @@ static int ocelot_setup_tc_block(struct ocelot_port_private *priv, { struct flow_block_cb *block_cb; flow_setup_cb_t *cb; - int err; - - netdev_dbg(priv->dev, "tc_block command %d, binder_type %d\n", - f->command, f->binder_type); if (f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS) { cb = ocelot_setup_tc_block_cb_ig; @@ -162,11 +148,6 @@ static int ocelot_setup_tc_block(struct ocelot_port_private *priv, if (IS_ERR(block_cb)) return PTR_ERR(block_cb); - err = ocelot_setup_tc_block_flower_bind(priv, f); - if (err < 0) { - flow_block_cb_free(block_cb); - return err; - } flow_block_cb_add(block_cb, f); list_add_tail(&block_cb->driver_list, f->driver_block_list); return 0; @@ -175,7 +156,6 @@ static int ocelot_setup_tc_block(struct ocelot_port_private *priv, if (!block_cb) return -ENOENT; - ocelot_setup_tc_block_flower_unbind(priv, f); flow_block_cb_remove(block_cb, f); list_del(&block_cb->driver_list); return 0; diff --git a/drivers/net/ethernet/mscc/ocelot_vcap.h b/drivers/net/ethernet/mscc/ocelot_vcap.h deleted file mode 100644 index e22eac1da783..000000000000 --- a/drivers/net/ethernet/mscc/ocelot_vcap.h +++ /dev/null @@ -1,403 +0,0 @@ -/* SPDX-License-Identifier: (GPL-2.0 OR MIT) - * Microsemi Ocelot Switch driver - * Copyright (c) 2019 Microsemi Corporation - */ - -#ifndef _OCELOT_VCAP_H_ -#define _OCELOT_VCAP_H_ - -/* ================================================================= - * VCAP Common - * ================================================================= - */ - -/* VCAP Type-Group values */ -#define VCAP_TG_NONE 0 /* Entry is invalid */ -#define VCAP_TG_FULL 1 /* Full entry */ -#define VCAP_TG_HALF 2 /* Half entry */ -#define VCAP_TG_QUARTER 3 /* Quarter entry */ - -/* ================================================================= - * VCAP IS2 - * ================================================================= - */ - -#define VCAP_IS2_CNT 64 -#define VCAP_IS2_ENTRY_WIDTH 376 -#define VCAP_IS2_ACTION_WIDTH 99 -#define VCAP_PORT_CNT 11 - -/* IS2 half key types */ -#define IS2_TYPE_ETYPE 0 -#define IS2_TYPE_LLC 1 -#define IS2_TYPE_SNAP 2 -#define IS2_TYPE_ARP 3 -#define IS2_TYPE_IP_UDP_TCP 4 -#define IS2_TYPE_IP_OTHER 5 -#define IS2_TYPE_IPV6 6 -#define IS2_TYPE_OAM 7 -#define IS2_TYPE_SMAC_SIP6 8 -#define IS2_TYPE_ANY 100 /* Pseudo type */ - -/* IS2 half key type mask for matching any IP */ -#define IS2_TYPE_MASK_IP_ANY 0xe - -/* IS2 action types */ -#define IS2_ACTION_TYPE_NORMAL 0 -#define IS2_ACTION_TYPE_SMAC_SIP 1 - -/* IS2 MASK_MODE values */ -#define IS2_ACT_MASK_MODE_NONE 0 -#define IS2_ACT_MASK_MODE_FILTER 1 -#define IS2_ACT_MASK_MODE_POLICY 2 -#define IS2_ACT_MASK_MODE_REDIR 3 - -/* IS2 REW_OP values */ -#define IS2_ACT_REW_OP_NONE 0 -#define IS2_ACT_REW_OP_PTP_ONE 2 -#define IS2_ACT_REW_OP_PTP_TWO 3 -#define IS2_ACT_REW_OP_SPECIAL 8 -#define IS2_ACT_REW_OP_PTP_ORG 9 -#define IS2_ACT_REW_OP_PTP_ONE_SUB_DELAY_1 (IS2_ACT_REW_OP_PTP_ONE | (1 << 3)) -#define IS2_ACT_REW_OP_PTP_ONE_SUB_DELAY_2 (IS2_ACT_REW_OP_PTP_ONE | (2 << 3)) -#define IS2_ACT_REW_OP_PTP_ONE_ADD_DELAY (IS2_ACT_REW_OP_PTP_ONE | (1 << 5)) -#define IS2_ACT_REW_OP_PTP_ONE_ADD_SUB BIT(7) - -#define VCAP_PORT_WIDTH 4 - -/* IS2 quarter key - SMAC_SIP4 */ -#define IS2_QKO_IGR_PORT 0 -#define IS2_QKL_IGR_PORT VCAP_PORT_WIDTH -#define IS2_QKO_L2_SMAC (IS2_QKO_IGR_PORT + IS2_QKL_IGR_PORT) -#define IS2_QKL_L2_SMAC 48 -#define IS2_QKO_L3_IP4_SIP (IS2_QKO_L2_SMAC + IS2_QKL_L2_SMAC) -#define IS2_QKL_L3_IP4_SIP 32 - -/* IS2 half key - common */ -#define IS2_HKO_TYPE 0 -#define IS2_HKL_TYPE 4 -#define IS2_HKO_FIRST (IS2_HKO_TYPE + IS2_HKL_TYPE) -#define IS2_HKL_FIRST 1 -#define IS2_HKO_PAG (IS2_HKO_FIRST + IS2_HKL_FIRST) -#define IS2_HKL_PAG 8 -#define IS2_HKO_IGR_PORT_MASK (IS2_HKO_PAG + IS2_HKL_PAG) -#define IS2_HKL_IGR_PORT_MASK (VCAP_PORT_CNT + 1) -#define IS2_HKO_SERVICE_FRM (IS2_HKO_IGR_PORT_MASK + IS2_HKL_IGR_PORT_MASK) -#define IS2_HKL_SERVICE_FRM 1 -#define IS2_HKO_HOST_MATCH (IS2_HKO_SERVICE_FRM + IS2_HKL_SERVICE_FRM) -#define IS2_HKL_HOST_MATCH 1 -#define IS2_HKO_L2_MC (IS2_HKO_HOST_MATCH + IS2_HKL_HOST_MATCH) -#define IS2_HKL_L2_MC 1 -#define IS2_HKO_L2_BC (IS2_HKO_L2_MC + IS2_HKL_L2_MC) -#define IS2_HKL_L2_BC 1 -#define IS2_HKO_VLAN_TAGGED (IS2_HKO_L2_BC + IS2_HKL_L2_BC) -#define IS2_HKL_VLAN_TAGGED 1 -#define IS2_HKO_VID (IS2_HKO_VLAN_TAGGED + IS2_HKL_VLAN_TAGGED) -#define IS2_HKL_VID 12 -#define IS2_HKO_DEI (IS2_HKO_VID + IS2_HKL_VID) -#define IS2_HKL_DEI 1 -#define IS2_HKO_PCP (IS2_HKO_DEI + IS2_HKL_DEI) -#define IS2_HKL_PCP 3 - -/* IS2 half key - MAC_ETYPE/MAC_LLC/MAC_SNAP/OAM common */ -#define IS2_HKO_L2_DMAC (IS2_HKO_PCP + IS2_HKL_PCP) -#define IS2_HKL_L2_DMAC 48 -#define IS2_HKO_L2_SMAC (IS2_HKO_L2_DMAC + IS2_HKL_L2_DMAC) -#define IS2_HKL_L2_SMAC 48 - -/* IS2 half key - MAC_ETYPE */ -#define IS2_HKO_MAC_ETYPE_ETYPE (IS2_HKO_L2_SMAC + IS2_HKL_L2_SMAC) -#define IS2_HKL_MAC_ETYPE_ETYPE 16 -#define IS2_HKO_MAC_ETYPE_L2_PAYLOAD \ - (IS2_HKO_MAC_ETYPE_ETYPE + IS2_HKL_MAC_ETYPE_ETYPE) -#define IS2_HKL_MAC_ETYPE_L2_PAYLOAD 27 - -/* IS2 half key - MAC_LLC */ -#define IS2_HKO_MAC_LLC_L2_LLC IS2_HKO_MAC_ETYPE_ETYPE -#define IS2_HKL_MAC_LLC_L2_LLC 40 - -/* IS2 half key - MAC_SNAP */ -#define IS2_HKO_MAC_SNAP_L2_SNAP IS2_HKO_MAC_ETYPE_ETYPE -#define IS2_HKL_MAC_SNAP_L2_SNAP 40 - -/* IS2 half key - ARP */ -#define IS2_HKO_MAC_ARP_L2_SMAC IS2_HKO_L2_DMAC -#define IS2_HKL_MAC_ARP_L2_SMAC 48 -#define IS2_HKO_MAC_ARP_ARP_ADDR_SPACE_OK \ - (IS2_HKO_MAC_ARP_L2_SMAC + IS2_HKL_MAC_ARP_L2_SMAC) -#define IS2_HKL_MAC_ARP_ARP_ADDR_SPACE_OK 1 -#define IS2_HKO_MAC_ARP_ARP_PROTO_SPACE_OK \ - (IS2_HKO_MAC_ARP_ARP_ADDR_SPACE_OK + IS2_HKL_MAC_ARP_ARP_ADDR_SPACE_OK) -#define IS2_HKL_MAC_ARP_ARP_PROTO_SPACE_OK 1 -#define IS2_HKO_MAC_ARP_ARP_LEN_OK \ - (IS2_HKO_MAC_ARP_ARP_PROTO_SPACE_OK + \ - IS2_HKL_MAC_ARP_ARP_PROTO_SPACE_OK) -#define IS2_HKL_MAC_ARP_ARP_LEN_OK 1 -#define IS2_HKO_MAC_ARP_ARP_TGT_MATCH \ - (IS2_HKO_MAC_ARP_ARP_LEN_OK + IS2_HKL_MAC_ARP_ARP_LEN_OK) -#define IS2_HKL_MAC_ARP_ARP_TGT_MATCH 1 -#define IS2_HKO_MAC_ARP_ARP_SENDER_MATCH \ - (IS2_HKO_MAC_ARP_ARP_TGT_MATCH + IS2_HKL_MAC_ARP_ARP_TGT_MATCH) -#define IS2_HKL_MAC_ARP_ARP_SENDER_MATCH 1 -#define IS2_HKO_MAC_ARP_ARP_OPCODE_UNKNOWN \ - (IS2_HKO_MAC_ARP_ARP_SENDER_MATCH + IS2_HKL_MAC_ARP_ARP_SENDER_MATCH) -#define IS2_HKL_MAC_ARP_ARP_OPCODE_UNKNOWN 1 -#define IS2_HKO_MAC_ARP_ARP_OPCODE \ - (IS2_HKO_MAC_ARP_ARP_OPCODE_UNKNOWN + \ - IS2_HKL_MAC_ARP_ARP_OPCODE_UNKNOWN) -#define IS2_HKL_MAC_ARP_ARP_OPCODE 2 -#define IS2_HKO_MAC_ARP_L3_IP4_DIP \ - (IS2_HKO_MAC_ARP_ARP_OPCODE + IS2_HKL_MAC_ARP_ARP_OPCODE) -#define IS2_HKL_MAC_ARP_L3_IP4_DIP 32 -#define IS2_HKO_MAC_ARP_L3_IP4_SIP \ - (IS2_HKO_MAC_ARP_L3_IP4_DIP + IS2_HKL_MAC_ARP_L3_IP4_DIP) -#define IS2_HKL_MAC_ARP_L3_IP4_SIP 32 -#define IS2_HKO_MAC_ARP_DIP_EQ_SIP \ - (IS2_HKO_MAC_ARP_L3_IP4_SIP + IS2_HKL_MAC_ARP_L3_IP4_SIP) -#define IS2_HKL_MAC_ARP_DIP_EQ_SIP 1 - -/* IS2 half key - IP4_TCP_UDP/IP4_OTHER common */ -#define IS2_HKO_IP4 IS2_HKO_L2_DMAC -#define IS2_HKL_IP4 1 -#define IS2_HKO_L3_FRAGMENT (IS2_HKO_IP4 + IS2_HKL_IP4) -#define IS2_HKL_L3_FRAGMENT 1 -#define IS2_HKO_L3_FRAG_OFS_GT0 (IS2_HKO_L3_FRAGMENT + IS2_HKL_L3_FRAGMENT) -#define IS2_HKL_L3_FRAG_OFS_GT0 1 -#define IS2_HKO_L3_OPTIONS (IS2_HKO_L3_FRAG_OFS_GT0 + IS2_HKL_L3_FRAG_OFS_GT0) -#define IS2_HKL_L3_OPTIONS 1 -#define IS2_HKO_L3_TTL_GT0 (IS2_HKO_L3_OPTIONS + IS2_HKL_L3_OPTIONS) -#define IS2_HKL_L3_TTL_GT0 1 -#define IS2_HKO_L3_TOS (IS2_HKO_L3_TTL_GT0 + IS2_HKL_L3_TTL_GT0) -#define IS2_HKL_L3_TOS 8 -#define IS2_HKO_L3_IP4_DIP (IS2_HKO_L3_TOS + IS2_HKL_L3_TOS) -#define IS2_HKL_L3_IP4_DIP 32 -#define IS2_HKO_L3_IP4_SIP (IS2_HKO_L3_IP4_DIP + IS2_HKL_L3_IP4_DIP) -#define IS2_HKL_L3_IP4_SIP 32 -#define IS2_HKO_DIP_EQ_SIP (IS2_HKO_L3_IP4_SIP + IS2_HKL_L3_IP4_SIP) -#define IS2_HKL_DIP_EQ_SIP 1 - -/* IS2 half key - IP4_TCP_UDP */ -#define IS2_HKO_IP4_TCP_UDP_TCP (IS2_HKO_DIP_EQ_SIP + IS2_HKL_DIP_EQ_SIP) -#define IS2_HKL_IP4_TCP_UDP_TCP 1 -#define IS2_HKO_IP4_TCP_UDP_L4_DPORT \ - (IS2_HKO_IP4_TCP_UDP_TCP + IS2_HKL_IP4_TCP_UDP_TCP) -#define IS2_HKL_IP4_TCP_UDP_L4_DPORT 16 -#define IS2_HKO_IP4_TCP_UDP_L4_SPORT \ - (IS2_HKO_IP4_TCP_UDP_L4_DPORT + IS2_HKL_IP4_TCP_UDP_L4_DPORT) -#define IS2_HKL_IP4_TCP_UDP_L4_SPORT 16 -#define IS2_HKO_IP4_TCP_UDP_L4_RNG \ - (IS2_HKO_IP4_TCP_UDP_L4_SPORT + IS2_HKL_IP4_TCP_UDP_L4_SPORT) -#define IS2_HKL_IP4_TCP_UDP_L4_RNG 8 -#define IS2_HKO_IP4_TCP_UDP_SPORT_EQ_DPORT \ - (IS2_HKO_IP4_TCP_UDP_L4_RNG + IS2_HKL_IP4_TCP_UDP_L4_RNG) -#define IS2_HKL_IP4_TCP_UDP_SPORT_EQ_DPORT 1 -#define IS2_HKO_IP4_TCP_UDP_SEQUENCE_EQ0 \ - (IS2_HKO_IP4_TCP_UDP_SPORT_EQ_DPORT + \ - IS2_HKL_IP4_TCP_UDP_SPORT_EQ_DPORT) -#define IS2_HKL_IP4_TCP_UDP_SEQUENCE_EQ0 1 -#define IS2_HKO_IP4_TCP_UDP_L4_FIN \ - (IS2_HKO_IP4_TCP_UDP_SEQUENCE_EQ0 + IS2_HKL_IP4_TCP_UDP_SEQUENCE_EQ0) -#define IS2_HKL_IP4_TCP_UDP_L4_FIN 1 -#define IS2_HKO_IP4_TCP_UDP_L4_SYN \ - (IS2_HKO_IP4_TCP_UDP_L4_FIN + IS2_HKL_IP4_TCP_UDP_L4_FIN) -#define IS2_HKL_IP4_TCP_UDP_L4_SYN 1 -#define IS2_HKO_IP4_TCP_UDP_L4_RST \ - (IS2_HKO_IP4_TCP_UDP_L4_SYN + IS2_HKL_IP4_TCP_UDP_L4_SYN) -#define IS2_HKL_IP4_TCP_UDP_L4_RST 1 -#define IS2_HKO_IP4_TCP_UDP_L4_PSH \ - (IS2_HKO_IP4_TCP_UDP_L4_RST + IS2_HKL_IP4_TCP_UDP_L4_RST) -#define IS2_HKL_IP4_TCP_UDP_L4_PSH 1 -#define IS2_HKO_IP4_TCP_UDP_L4_ACK \ - (IS2_HKO_IP4_TCP_UDP_L4_PSH + IS2_HKL_IP4_TCP_UDP_L4_PSH) -#define IS2_HKL_IP4_TCP_UDP_L4_ACK 1 -#define IS2_HKO_IP4_TCP_UDP_L4_URG \ - (IS2_HKO_IP4_TCP_UDP_L4_ACK + IS2_HKL_IP4_TCP_UDP_L4_ACK) -#define IS2_HKL_IP4_TCP_UDP_L4_URG 1 -#define IS2_HKO_IP4_TCP_UDP_L4_1588_DOM \ - (IS2_HKO_IP4_TCP_UDP_L4_URG + IS2_HKL_IP4_TCP_UDP_L4_URG) -#define IS2_HKL_IP4_TCP_UDP_L4_1588_DOM 8 -#define IS2_HKO_IP4_TCP_UDP_L4_1588_VER \ - (IS2_HKO_IP4_TCP_UDP_L4_1588_DOM + IS2_HKL_IP4_TCP_UDP_L4_1588_DOM) -#define IS2_HKL_IP4_TCP_UDP_L4_1588_VER 4 - -/* IS2 half key - IP4_OTHER */ -#define IS2_HKO_IP4_OTHER_L3_PROTO IS2_HKO_IP4_TCP_UDP_TCP -#define IS2_HKL_IP4_OTHER_L3_PROTO 8 -#define IS2_HKO_IP4_OTHER_L3_PAYLOAD \ - (IS2_HKO_IP4_OTHER_L3_PROTO + IS2_HKL_IP4_OTHER_L3_PROTO) -#define IS2_HKL_IP4_OTHER_L3_PAYLOAD 56 - -/* IS2 half key - IP6_STD */ -#define IS2_HKO_IP6_STD_L3_TTL_GT0 IS2_HKO_L2_DMAC -#define IS2_HKL_IP6_STD_L3_TTL_GT0 1 -#define IS2_HKO_IP6_STD_L3_IP6_SIP \ - (IS2_HKO_IP6_STD_L3_TTL_GT0 + IS2_HKL_IP6_STD_L3_TTL_GT0) -#define IS2_HKL_IP6_STD_L3_IP6_SIP 128 -#define IS2_HKO_IP6_STD_L3_PROTO \ - (IS2_HKO_IP6_STD_L3_IP6_SIP + IS2_HKL_IP6_STD_L3_IP6_SIP) -#define IS2_HKL_IP6_STD_L3_PROTO 8 - -/* IS2 half key - OAM */ -#define IS2_HKO_OAM_OAM_MEL_FLAGS IS2_HKO_MAC_ETYPE_ETYPE -#define IS2_HKL_OAM_OAM_MEL_FLAGS 7 -#define IS2_HKO_OAM_OAM_VER \ - (IS2_HKO_OAM_OAM_MEL_FLAGS + IS2_HKL_OAM_OAM_MEL_FLAGS) -#define IS2_HKL_OAM_OAM_VER 5 -#define IS2_HKO_OAM_OAM_OPCODE (IS2_HKO_OAM_OAM_VER + IS2_HKL_OAM_OAM_VER) -#define IS2_HKL_OAM_OAM_OPCODE 8 -#define IS2_HKO_OAM_OAM_FLAGS (IS2_HKO_OAM_OAM_OPCODE + IS2_HKL_OAM_OAM_OPCODE) -#define IS2_HKL_OAM_OAM_FLAGS 8 -#define IS2_HKO_OAM_OAM_MEPID (IS2_HKO_OAM_OAM_FLAGS + IS2_HKL_OAM_OAM_FLAGS) -#define IS2_HKL_OAM_OAM_MEPID 16 -#define IS2_HKO_OAM_OAM_CCM_CNTS_EQ0 \ - (IS2_HKO_OAM_OAM_MEPID + IS2_HKL_OAM_OAM_MEPID) -#define IS2_HKL_OAM_OAM_CCM_CNTS_EQ0 1 - -/* IS2 half key - SMAC_SIP6 */ -#define IS2_HKO_SMAC_SIP6_IGR_PORT IS2_HKL_TYPE -#define IS2_HKL_SMAC_SIP6_IGR_PORT VCAP_PORT_WIDTH -#define IS2_HKO_SMAC_SIP6_L2_SMAC \ - (IS2_HKO_SMAC_SIP6_IGR_PORT + IS2_HKL_SMAC_SIP6_IGR_PORT) -#define IS2_HKL_SMAC_SIP6_L2_SMAC 48 -#define IS2_HKO_SMAC_SIP6_L3_IP6_SIP \ - (IS2_HKO_SMAC_SIP6_L2_SMAC + IS2_HKL_SMAC_SIP6_L2_SMAC) -#define IS2_HKL_SMAC_SIP6_L3_IP6_SIP 128 - -/* IS2 full key - common */ -#define IS2_FKO_TYPE 0 -#define IS2_FKL_TYPE 2 -#define IS2_FKO_FIRST (IS2_FKO_TYPE + IS2_FKL_TYPE) -#define IS2_FKL_FIRST 1 -#define IS2_FKO_PAG (IS2_FKO_FIRST + IS2_FKL_FIRST) -#define IS2_FKL_PAG 8 -#define IS2_FKO_IGR_PORT_MASK (IS2_FKO_PAG + IS2_FKL_PAG) -#define IS2_FKL_IGR_PORT_MASK (VCAP_PORT_CNT + 1) -#define IS2_FKO_SERVICE_FRM (IS2_FKO_IGR_PORT_MASK + IS2_FKL_IGR_PORT_MASK) -#define IS2_FKL_SERVICE_FRM 1 -#define IS2_FKO_HOST_MATCH (IS2_FKO_SERVICE_FRM + IS2_FKL_SERVICE_FRM) -#define IS2_FKL_HOST_MATCH 1 -#define IS2_FKO_L2_MC (IS2_FKO_HOST_MATCH + IS2_FKL_HOST_MATCH) -#define IS2_FKL_L2_MC 1 -#define IS2_FKO_L2_BC (IS2_FKO_L2_MC + IS2_FKL_L2_MC) -#define IS2_FKL_L2_BC 1 -#define IS2_FKO_VLAN_TAGGED (IS2_FKO_L2_BC + IS2_FKL_L2_BC) -#define IS2_FKL_VLAN_TAGGED 1 -#define IS2_FKO_VID (IS2_FKO_VLAN_TAGGED + IS2_FKL_VLAN_TAGGED) -#define IS2_FKL_VID 12 -#define IS2_FKO_DEI (IS2_FKO_VID + IS2_FKL_VID) -#define IS2_FKL_DEI 1 -#define IS2_FKO_PCP (IS2_FKO_DEI + IS2_FKL_DEI) -#define IS2_FKL_PCP 3 - -/* IS2 full key - IP6_TCP_UDP/IP6_OTHER common */ -#define IS2_FKO_L3_TTL_GT0 (IS2_FKO_PCP + IS2_FKL_PCP) -#define IS2_FKL_L3_TTL_GT0 1 -#define IS2_FKO_L3_TOS (IS2_FKO_L3_TTL_GT0 + IS2_FKL_L3_TTL_GT0) -#define IS2_FKL_L3_TOS 8 -#define IS2_FKO_L3_IP6_DIP (IS2_FKO_L3_TOS + IS2_FKL_L3_TOS) -#define IS2_FKL_L3_IP6_DIP 128 -#define IS2_FKO_L3_IP6_SIP (IS2_FKO_L3_IP6_DIP + IS2_FKL_L3_IP6_DIP) -#define IS2_FKL_L3_IP6_SIP 128 -#define IS2_FKO_DIP_EQ_SIP (IS2_FKO_L3_IP6_SIP + IS2_FKL_L3_IP6_SIP) -#define IS2_FKL_DIP_EQ_SIP 1 - -/* IS2 full key - IP6_TCP_UDP */ -#define IS2_FKO_IP6_TCP_UDP_TCP (IS2_FKO_DIP_EQ_SIP + IS2_FKL_DIP_EQ_SIP) -#define IS2_FKL_IP6_TCP_UDP_TCP 1 -#define IS2_FKO_IP6_TCP_UDP_L4_DPORT \ - (IS2_FKO_IP6_TCP_UDP_TCP + IS2_FKL_IP6_TCP_UDP_TCP) -#define IS2_FKL_IP6_TCP_UDP_L4_DPORT 16 -#define IS2_FKO_IP6_TCP_UDP_L4_SPORT \ - (IS2_FKO_IP6_TCP_UDP_L4_DPORT + IS2_FKL_IP6_TCP_UDP_L4_DPORT) -#define IS2_FKL_IP6_TCP_UDP_L4_SPORT 16 -#define IS2_FKO_IP6_TCP_UDP_L4_RNG \ - (IS2_FKO_IP6_TCP_UDP_L4_SPORT + IS2_FKL_IP6_TCP_UDP_L4_SPORT) -#define IS2_FKL_IP6_TCP_UDP_L4_RNG 8 -#define IS2_FKO_IP6_TCP_UDP_SPORT_EQ_DPORT \ - (IS2_FKO_IP6_TCP_UDP_L4_RNG + IS2_FKL_IP6_TCP_UDP_L4_RNG) -#define IS2_FKL_IP6_TCP_UDP_SPORT_EQ_DPORT 1 -#define IS2_FKO_IP6_TCP_UDP_SEQUENCE_EQ0 \ - (IS2_FKO_IP6_TCP_UDP_SPORT_EQ_DPORT + \ - IS2_FKL_IP6_TCP_UDP_SPORT_EQ_DPORT) -#define IS2_FKL_IP6_TCP_UDP_SEQUENCE_EQ0 1 -#define IS2_FKO_IP6_TCP_UDP_L4_FIN \ - (IS2_FKO_IP6_TCP_UDP_SEQUENCE_EQ0 + IS2_FKL_IP6_TCP_UDP_SEQUENCE_EQ0) -#define IS2_FKL_IP6_TCP_UDP_L4_FIN 1 -#define IS2_FKO_IP6_TCP_UDP_L4_SYN \ - (IS2_FKO_IP6_TCP_UDP_L4_FIN + IS2_FKL_IP6_TCP_UDP_L4_FIN) -#define IS2_FKL_IP6_TCP_UDP_L4_SYN 1 -#define IS2_FKO_IP6_TCP_UDP_L4_RST \ - (IS2_FKO_IP6_TCP_UDP_L4_SYN + IS2_FKL_IP6_TCP_UDP_L4_SYN) -#define IS2_FKL_IP6_TCP_UDP_L4_RST 1 -#define IS2_FKO_IP6_TCP_UDP_L4_PSH \ - (IS2_FKO_IP6_TCP_UDP_L4_RST + IS2_FKL_IP6_TCP_UDP_L4_RST) -#define IS2_FKL_IP6_TCP_UDP_L4_PSH 1 -#define IS2_FKO_IP6_TCP_UDP_L4_ACK \ - (IS2_FKO_IP6_TCP_UDP_L4_PSH + IS2_FKL_IP6_TCP_UDP_L4_PSH) -#define IS2_FKL_IP6_TCP_UDP_L4_ACK 1 -#define IS2_FKO_IP6_TCP_UDP_L4_URG \ - (IS2_FKO_IP6_TCP_UDP_L4_ACK + IS2_FKL_IP6_TCP_UDP_L4_ACK) -#define IS2_FKL_IP6_TCP_UDP_L4_URG 1 -#define IS2_FKO_IP6_TCP_UDP_L4_1588_DOM \ - (IS2_FKO_IP6_TCP_UDP_L4_URG + IS2_FKL_IP6_TCP_UDP_L4_URG) -#define IS2_FKL_IP6_TCP_UDP_L4_1588_DOM 8 -#define IS2_FKO_IP6_TCP_UDP_L4_1588_VER \ - (IS2_FKO_IP6_TCP_UDP_L4_1588_DOM + IS2_FKL_IP6_TCP_UDP_L4_1588_DOM) -#define IS2_FKL_IP6_TCP_UDP_L4_1588_VER 4 - -/* IS2 full key - IP6_OTHER */ -#define IS2_FKO_IP6_OTHER_L3_PROTO IS2_FKO_IP6_TCP_UDP_TCP -#define IS2_FKL_IP6_OTHER_L3_PROTO 8 -#define IS2_FKO_IP6_OTHER_L3_PAYLOAD \ - (IS2_FKO_IP6_OTHER_L3_PROTO + IS2_FKL_IP6_OTHER_L3_PROTO) -#define IS2_FKL_IP6_OTHER_L3_PAYLOAD 56 - -/* IS2 full key - CUSTOM */ -#define IS2_FKO_CUSTOM_CUSTOM_TYPE IS2_FKO_L3_TTL_GT0 -#define IS2_FKL_CUSTOM_CUSTOM_TYPE 1 -#define IS2_FKO_CUSTOM_CUSTOM \ - (IS2_FKO_CUSTOM_CUSTOM_TYPE + IS2_FKL_CUSTOM_CUSTOM_TYPE) -#define IS2_FKL_CUSTOM_CUSTOM 320 - -/* IS2 action - BASE_TYPE */ -#define IS2_AO_HIT_ME_ONCE 0 -#define IS2_AL_HIT_ME_ONCE 1 -#define IS2_AO_CPU_COPY_ENA (IS2_AO_HIT_ME_ONCE + IS2_AL_HIT_ME_ONCE) -#define IS2_AL_CPU_COPY_ENA 1 -#define IS2_AO_CPU_QU_NUM (IS2_AO_CPU_COPY_ENA + IS2_AL_CPU_COPY_ENA) -#define IS2_AL_CPU_QU_NUM 3 -#define IS2_AO_MASK_MODE (IS2_AO_CPU_QU_NUM + IS2_AL_CPU_QU_NUM) -#define IS2_AL_MASK_MODE 2 -#define IS2_AO_MIRROR_ENA (IS2_AO_MASK_MODE + IS2_AL_MASK_MODE) -#define IS2_AL_MIRROR_ENA 1 -#define IS2_AO_LRN_DIS (IS2_AO_MIRROR_ENA + IS2_AL_MIRROR_ENA) -#define IS2_AL_LRN_DIS 1 -#define IS2_AO_POLICE_ENA (IS2_AO_LRN_DIS + IS2_AL_LRN_DIS) -#define IS2_AL_POLICE_ENA 1 -#define IS2_AO_POLICE_IDX (IS2_AO_POLICE_ENA + IS2_AL_POLICE_ENA) -#define IS2_AL_POLICE_IDX 9 -#define IS2_AO_POLICE_VCAP_ONLY (IS2_AO_POLICE_IDX + IS2_AL_POLICE_IDX) -#define IS2_AL_POLICE_VCAP_ONLY 1 -#define IS2_AO_PORT_MASK (IS2_AO_POLICE_VCAP_ONLY + IS2_AL_POLICE_VCAP_ONLY) -#define IS2_AL_PORT_MASK VCAP_PORT_CNT -#define IS2_AO_REW_OP (IS2_AO_PORT_MASK + IS2_AL_PORT_MASK) -#define IS2_AL_REW_OP 9 -#define IS2_AO_LM_CNT_DIS (IS2_AO_REW_OP + IS2_AL_REW_OP) -#define IS2_AL_LM_CNT_DIS 1 -#define IS2_AO_ISDX_ENA \ - (IS2_AO_LM_CNT_DIS + IS2_AL_LM_CNT_DIS + 1) /* Reserved bit */ -#define IS2_AL_ISDX_ENA 1 -#define IS2_AO_ACL_ID (IS2_AO_ISDX_ENA + IS2_AL_ISDX_ENA) -#define IS2_AL_ACL_ID 6 - -/* IS2 action - SMAC_SIP */ -#define IS2_AO_SMAC_SIP_CPU_COPY_ENA 0 -#define IS2_AL_SMAC_SIP_CPU_COPY_ENA 1 -#define IS2_AO_SMAC_SIP_CPU_QU_NUM 1 -#define IS2_AL_SMAC_SIP_CPU_QU_NUM 3 -#define IS2_AO_SMAC_SIP_FWD_KILL_ENA 4 -#define IS2_AL_SMAC_SIP_FWD_KILL_ENA 1 -#define IS2_AO_SMAC_SIP_HOST_MATCH 5 -#define IS2_AL_SMAC_SIP_HOST_MATCH 1 - -#endif /* _OCELOT_VCAP_H_ */ diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c index 2ee0d0be113a..2616fd735aab 100644 --- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c +++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c @@ -1920,6 +1920,7 @@ myri10ge_phys_id(struct net_device *netdev, enum ethtool_phys_id_state state) } static const struct ethtool_ops myri10ge_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = myri10ge_get_drvinfo, .get_coalesce = myri10ge_get_coalesce, .set_coalesce = myri10ge_set_coalesce, diff --git a/drivers/net/ethernet/natsemi/jazzsonic.c b/drivers/net/ethernet/natsemi/jazzsonic.c index 51fa82b429a3..bfa0c0d39600 100644 --- a/drivers/net/ethernet/natsemi/jazzsonic.c +++ b/drivers/net/ethernet/natsemi/jazzsonic.c @@ -147,39 +147,12 @@ static int sonic_probe1(struct net_device *dev) dev->dev_addr[i*2+1] = val >> 8; } - err = -ENOMEM; - - /* Initialize the device structure. */ - lp->dma_bitmode = SONIC_BITMODE32; - /* Allocate the entire chunk of memory for the descriptors. - Note that this cannot cross a 64K boundary. */ - lp->descriptors = dma_alloc_coherent(lp->device, - SIZEOF_SONIC_DESC * - SONIC_BUS_SCALE(lp->dma_bitmode), - &lp->descriptors_laddr, - GFP_KERNEL); - if (lp->descriptors == NULL) + err = sonic_alloc_descriptors(dev); + if (err) goto out; - /* Now set up the pointers to point to the appropriate places */ - lp->cda = lp->descriptors; - lp->tda = lp->cda + (SIZEOF_SONIC_CDA - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rda = lp->tda + (SIZEOF_SONIC_TD * SONIC_NUM_TDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rra = lp->rda + (SIZEOF_SONIC_RD * SONIC_NUM_RDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - - lp->cda_laddr = lp->descriptors_laddr; - lp->tda_laddr = lp->cda_laddr + (SIZEOF_SONIC_CDA - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rda_laddr = lp->tda_laddr + (SIZEOF_SONIC_TD * SONIC_NUM_TDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rra_laddr = lp->rda_laddr + (SIZEOF_SONIC_RD * SONIC_NUM_RDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - dev->netdev_ops = &sonic_netdev_ops; dev->watchdog_timeo = TX_TIMEOUT; diff --git a/drivers/net/ethernet/natsemi/macsonic.c b/drivers/net/ethernet/natsemi/macsonic.c index 0937fc2a928e..1b5559aacb38 100644 --- a/drivers/net/ethernet/natsemi/macsonic.c +++ b/drivers/net/ethernet/natsemi/macsonic.c @@ -114,17 +114,6 @@ static inline void bit_reverse_addr(unsigned char addr[6]) addr[i] = bitrev8(addr[i]); } -static irqreturn_t macsonic_interrupt(int irq, void *dev_id) -{ - irqreturn_t result; - unsigned long flags; - - local_irq_save(flags); - result = sonic_interrupt(irq, dev_id); - local_irq_restore(flags); - return result; -} - static int macsonic_open(struct net_device* dev) { int retval; @@ -135,12 +124,12 @@ static int macsonic_open(struct net_device* dev) dev->name, dev->irq); goto err; } - /* Under the A/UX interrupt scheme, the onboard SONIC interrupt comes - * in at priority level 3. However, we sometimes get the level 2 inter- - * rupt as well, which must prevent re-entrance of the sonic handler. + /* Under the A/UX interrupt scheme, the onboard SONIC interrupt gets + * moved from level 2 to level 3. Unfortunately we still get some + * level 2 interrupts so register the handler for both. */ if (dev->irq == IRQ_AUTO_3) { - retval = request_irq(IRQ_NUBUS_9, macsonic_interrupt, 0, + retval = request_irq(IRQ_NUBUS_9, sonic_interrupt, 0, "sonic", dev); if (retval) { printk(KERN_ERR "%s: unable to get IRQ %d.\n", @@ -186,33 +175,10 @@ static const struct net_device_ops macsonic_netdev_ops = { static int macsonic_init(struct net_device *dev) { struct sonic_local* lp = netdev_priv(dev); + int err = sonic_alloc_descriptors(dev); - /* Allocate the entire chunk of memory for the descriptors. - Note that this cannot cross a 64K boundary. */ - lp->descriptors = dma_alloc_coherent(lp->device, - SIZEOF_SONIC_DESC * - SONIC_BUS_SCALE(lp->dma_bitmode), - &lp->descriptors_laddr, - GFP_KERNEL); - if (lp->descriptors == NULL) - return -ENOMEM; - - /* Now set up the pointers to point to the appropriate places */ - lp->cda = lp->descriptors; - lp->tda = lp->cda + (SIZEOF_SONIC_CDA - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rda = lp->tda + (SIZEOF_SONIC_TD * SONIC_NUM_TDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rra = lp->rda + (SIZEOF_SONIC_RD * SONIC_NUM_RDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - - lp->cda_laddr = lp->descriptors_laddr; - lp->tda_laddr = lp->cda_laddr + (SIZEOF_SONIC_CDA - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rda_laddr = lp->tda_laddr + (SIZEOF_SONIC_TD * SONIC_NUM_TDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rra_laddr = lp->rda_laddr + (SIZEOF_SONIC_RD * SONIC_NUM_RDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); + if (err) + return err; dev->netdev_ops = &macsonic_netdev_ops; dev->watchdog_timeo = TX_TIMEOUT; diff --git a/drivers/net/ethernet/natsemi/sonic.c b/drivers/net/ethernet/natsemi/sonic.c index 31be3ba66877..dd3605aa5f23 100644 --- a/drivers/net/ethernet/natsemi/sonic.c +++ b/drivers/net/ethernet/natsemi/sonic.c @@ -50,6 +50,42 @@ static void sonic_msg_init(struct net_device *dev) netif_dbg(lp, drv, dev, "%s", version); } +static int sonic_alloc_descriptors(struct net_device *dev) +{ + struct sonic_local *lp = netdev_priv(dev); + + /* Allocate a chunk of memory for the descriptors. Note that this + * must not cross a 64K boundary. It is smaller than one page which + * means that page alignment is a sufficient condition. + */ + lp->descriptors = + dma_alloc_coherent(lp->device, + SIZEOF_SONIC_DESC * + SONIC_BUS_SCALE(lp->dma_bitmode), + &lp->descriptors_laddr, GFP_KERNEL); + + if (!lp->descriptors) + return -ENOMEM; + + lp->cda = lp->descriptors; + lp->tda = lp->cda + SIZEOF_SONIC_CDA * + SONIC_BUS_SCALE(lp->dma_bitmode); + lp->rda = lp->tda + SIZEOF_SONIC_TD * SONIC_NUM_TDS * + SONIC_BUS_SCALE(lp->dma_bitmode); + lp->rra = lp->rda + SIZEOF_SONIC_RD * SONIC_NUM_RDS * + SONIC_BUS_SCALE(lp->dma_bitmode); + + lp->cda_laddr = lp->descriptors_laddr; + lp->tda_laddr = lp->cda_laddr + SIZEOF_SONIC_CDA * + SONIC_BUS_SCALE(lp->dma_bitmode); + lp->rda_laddr = lp->tda_laddr + SIZEOF_SONIC_TD * SONIC_NUM_TDS * + SONIC_BUS_SCALE(lp->dma_bitmode); + lp->rra_laddr = lp->rda_laddr + SIZEOF_SONIC_RD * SONIC_NUM_RDS * + SONIC_BUS_SCALE(lp->dma_bitmode); + + return 0; +} + /* * Open/initialize the SONIC controller. * @@ -264,7 +300,7 @@ static int sonic_send_packet(struct sk_buff *skb, struct net_device *dev) spin_lock_irqsave(&lp->lock, flags); - entry = lp->next_tx; + entry = (lp->eol_tx + 1) & SONIC_TDS_MASK; sonic_tda_put(dev, entry, SONIC_TD_STATUS, 0); /* clear status */ sonic_tda_put(dev, entry, SONIC_TD_FRAG_COUNT, 1); /* single fragment */ @@ -275,27 +311,26 @@ static int sonic_send_packet(struct sk_buff *skb, struct net_device *dev) sonic_tda_put(dev, entry, SONIC_TD_LINK, sonic_tda_get(dev, entry, SONIC_TD_LINK) | SONIC_EOL); - wmb(); + sonic_tda_put(dev, lp->eol_tx, SONIC_TD_LINK, ~SONIC_EOL & + sonic_tda_get(dev, lp->eol_tx, SONIC_TD_LINK)); + + netif_dbg(lp, tx_queued, dev, "%s: issuing Tx command\n", __func__); + + SONIC_WRITE(SONIC_CMD, SONIC_CR_TXP); + lp->tx_len[entry] = length; lp->tx_laddr[entry] = laddr; lp->tx_skb[entry] = skb; - wmb(); - sonic_tda_put(dev, lp->eol_tx, SONIC_TD_LINK, - sonic_tda_get(dev, lp->eol_tx, SONIC_TD_LINK) & ~SONIC_EOL); lp->eol_tx = entry; - lp->next_tx = (entry + 1) & SONIC_TDS_MASK; - if (lp->tx_skb[lp->next_tx] != NULL) { + entry = (entry + 1) & SONIC_TDS_MASK; + if (lp->tx_skb[entry]) { /* The ring is full, the ISR has yet to process the next TD. */ netif_dbg(lp, tx_queued, dev, "%s: stopping queue\n", __func__); netif_stop_queue(dev); /* after this packet, wait for ISR to free up some TDAs */ - } else netif_start_queue(dev); - - netif_dbg(lp, tx_queued, dev, "%s: issuing Tx command\n", __func__); - - SONIC_WRITE(SONIC_CMD, SONIC_CR_TXP); + } spin_unlock_irqrestore(&lp->lock, flags); @@ -594,11 +629,6 @@ static void sonic_rx(struct net_device *dev) if (rbe) SONIC_WRITE(SONIC_ISR, SONIC_INT_RBE); - /* - * If any worth-while packets have been received, netif_rx() - * has done a mark_bh(NET_BH) for us and will work on them - * when we get to the bottom-half routine. - */ } @@ -780,7 +810,7 @@ static int sonic_init(struct net_device *dev) SONIC_WRITE(SONIC_UTDA, lp->tda_laddr >> 16); SONIC_WRITE(SONIC_CTDA, lp->tda_laddr & 0xffff); - lp->cur_tx = lp->next_tx = 0; + lp->cur_tx = 0; lp->eol_tx = SONIC_NUM_TDS - 1; /* diff --git a/drivers/net/ethernet/natsemi/sonic.h b/drivers/net/ethernet/natsemi/sonic.h index e0e4cba6f6f6..3cbb62c860c8 100644 --- a/drivers/net/ethernet/natsemi/sonic.h +++ b/drivers/net/ethernet/natsemi/sonic.h @@ -321,7 +321,6 @@ struct sonic_local { unsigned int cur_tx; /* first unacked transmit packet */ unsigned int eol_rx; unsigned int eol_tx; /* last unacked transmit packet */ - unsigned int next_tx; /* next free TD */ int msg_enable; struct device *device; /* generic device */ struct net_device_stats stats; @@ -342,6 +341,7 @@ static void sonic_multicast_list(struct net_device *dev); static int sonic_init(struct net_device *dev); static void sonic_tx_timeout(struct net_device *dev, unsigned int txqueue); static void sonic_msg_init(struct net_device *dev); +static int sonic_alloc_descriptors(struct net_device *dev); /* Internal inlines for reading/writing DMA buffers. Note that bus size and endianness matter here, whereas they don't for registers, diff --git a/drivers/net/ethernet/natsemi/xtsonic.c b/drivers/net/ethernet/natsemi/xtsonic.c index e1b886e87a76..dda9ec7d9cee 100644 --- a/drivers/net/ethernet/natsemi/xtsonic.c +++ b/drivers/net/ethernet/natsemi/xtsonic.c @@ -167,47 +167,11 @@ static int __init sonic_probe1(struct net_device *dev) dev->dev_addr[i*2+1] = val >> 8; } - /* Initialize the device structure. */ - lp->dma_bitmode = SONIC_BITMODE32; - /* - * Allocate local private descriptor areas in uncached space. - * The entire structure must be located within the same 64kb segment. - * A simple way to ensure this is to allocate twice the - * size of the structure -- given that the structure is - * much less than 64 kB, at least one of the halves of - * the allocated area will be contained entirely in 64 kB. - * We also allocate extra space for a pointer to allow freeing - * this structure later on (in xtsonic_cleanup_module()). - */ - lp->descriptors = dma_alloc_coherent(lp->device, - SIZEOF_SONIC_DESC * - SONIC_BUS_SCALE(lp->dma_bitmode), - &lp->descriptors_laddr, - GFP_KERNEL); - if (lp->descriptors == NULL) { - err = -ENOMEM; + err = sonic_alloc_descriptors(dev); + if (err) goto out; - } - - lp->cda = lp->descriptors; - lp->tda = lp->cda + (SIZEOF_SONIC_CDA - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rda = lp->tda + (SIZEOF_SONIC_TD * SONIC_NUM_TDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rra = lp->rda + (SIZEOF_SONIC_RD * SONIC_NUM_RDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - - /* get the virtual dma address */ - - lp->cda_laddr = lp->descriptors_laddr; - lp->tda_laddr = lp->cda_laddr + (SIZEOF_SONIC_CDA - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rda_laddr = lp->tda_laddr + (SIZEOF_SONIC_TD * SONIC_NUM_TDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); - lp->rra_laddr = lp->rda_laddr + (SIZEOF_SONIC_RD * SONIC_NUM_RDS - * SONIC_BUS_SCALE(lp->dma_bitmode)); dev->netdev_ops = &xtsonic_netdev_ops; dev->watchdog_timeo = TX_TIMEOUT; diff --git a/drivers/net/ethernet/netronome/nfp/bpf/fw.h b/drivers/net/ethernet/netronome/nfp/bpf/fw.h index a83a0ad5e27d..4268a7e0f344 100644 --- a/drivers/net/ethernet/netronome/nfp/bpf/fw.h +++ b/drivers/net/ethernet/netronome/nfp/bpf/fw.h @@ -104,14 +104,14 @@ struct cmsg_req_map_op { __be32 tid; __be32 count; __be32 flags; - u8 data[0]; + u8 data[]; }; struct cmsg_reply_map_op { struct cmsg_reply_map_simple reply_hdr; __be32 count; __be32 resv; - u8 data[0]; + u8 data[]; }; struct cmsg_bpf_event { @@ -120,6 +120,6 @@ struct cmsg_bpf_event { __be64 map_ptr; __be32 data_size; __be32 pkt_size; - u8 data[0]; + u8 data[]; }; #endif diff --git a/drivers/net/ethernet/netronome/nfp/flower/action.c b/drivers/net/ethernet/netronome/nfp/flower/action.c index c06600fb47ff..1c76e1592ca2 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/action.c +++ b/drivers/net/ethernet/netronome/nfp/flower/action.c @@ -1207,6 +1207,10 @@ int nfp_flower_compile_action(struct nfp_app *app, bool pkt_host = false; u32 csum_updated = 0; + if (!flow_action_hw_stats_check(&flow->rule->action, extack, + FLOW_ACTION_HW_STATS_DELAYED_BIT)) + return -EOPNOTSUPP; + memset(nfp_flow->action_data, 0, NFP_FL_MAX_A_SIZ); nfp_flow->meta.act_len = 0; tun_type = NFP_FL_TUNNEL_NONE; diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h index 9b50d76bbc09..bf516285510f 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h +++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h @@ -587,7 +587,7 @@ struct nfp_flower_cmsg_mac_repr { u8 info; u8 nbi_port; u8 phys_port; - } ports[0]; + } ports[]; }; #define NFP_FLOWER_CMSG_MAC_REPR_NBI GENMASK(1, 0) @@ -619,7 +619,7 @@ struct nfp_flower_cmsg_merge_hint { struct { __be32 host_ctx; __be64 host_cookie; - } __packed flow[0]; + } __packed flow[]; }; enum nfp_flower_cmsg_port_type { diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c index 7ca5c1becfcf..c694dbc239d0 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/offload.c +++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c @@ -1490,7 +1490,8 @@ nfp_flower_get_stats(struct nfp_app *app, struct net_device *netdev, nfp_flower_update_merge_stats(app, nfp_flow); flow_stats_update(&flow->stats, priv->stats[ctx_id].bytes, - priv->stats[ctx_id].pkts, priv->stats[ctx_id].used); + priv->stats[ctx_id].pkts, priv->stats[ctx_id].used, + FLOW_ACTION_HW_STATS_DELAYED); priv->stats[ctx_id].pkts = 0; priv->stats[ctx_id].bytes = 0; diff --git a/drivers/net/ethernet/netronome/nfp/flower/qos_conf.c b/drivers/net/ethernet/netronome/nfp/flower/qos_conf.c index 124a43dc136a..d18a830e4264 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/qos_conf.c +++ b/drivers/net/ethernet/netronome/nfp/flower/qos_conf.c @@ -320,7 +320,8 @@ nfp_flower_stats_rate_limiter(struct nfp_app *app, struct net_device *netdev, spin_unlock_bh(&fl_priv->qos_stats_lock); flow_stats_update(&flow->stats, diff_bytes, diff_pkts, - repr_priv->qos_table.last_update); + repr_priv->qos_table.last_update, + FLOW_ACTION_HW_STATS_DELAYED); return 0; } diff --git a/drivers/net/ethernet/netronome/nfp/nfp_devlink.c b/drivers/net/ethernet/netronome/nfp/nfp_devlink.c index c50fce42f473..07dbf4d72227 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_devlink.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_devlink.c @@ -211,7 +211,7 @@ static const struct nfp_devlink_versions { enum nfp_nsp_versions id; const char *key; } nfp_devlink_versions_nsp[] = { - { NFP_VERSIONS_BUNDLE, "fw.bundle_id", }, + { NFP_VERSIONS_BUNDLE, DEVLINK_INFO_VERSION_GENERIC_FW_BUNDLE_ID, }, { NFP_VERSIONS_BSP, DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, }, { NFP_VERSIONS_CPLD, "fw.cpld", }, { NFP_VERSIONS_APP, DEVLINK_INFO_VERSION_GENERIC_FW_APP, }, diff --git a/drivers/net/ethernet/netronome/nfp/nfp_main.h b/drivers/net/ethernet/netronome/nfp/nfp_main.h index 5d5812fd9317..fa6b13a05941 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_main.h +++ b/drivers/net/ethernet/netronome/nfp/nfp_main.h @@ -42,7 +42,7 @@ struct nfp_shared_buf; */ struct nfp_dumpspec { u32 size; - u8 data[0]; + u8 data[]; }; /** diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_debugdump.c b/drivers/net/ethernet/netronome/nfp/nfp_net_debugdump.c index 769ceef09756..a614df095b08 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_debugdump.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_debugdump.c @@ -36,7 +36,7 @@ enum nfp_dumpspec_type { struct nfp_dump_tl { __be32 type; __be32 length; /* chunk length to follow, aligned to 8 bytes */ - char data[0]; + char data[]; }; /* NFP CPP parameters */ @@ -62,7 +62,7 @@ struct nfp_dumpspec_csr { struct nfp_dumpspec_rtsym { struct nfp_dump_tl tl; - char rtsym[0]; + char rtsym[]; }; /* header for register dumpable */ @@ -79,7 +79,7 @@ struct nfp_dump_rtsym { struct nfp_dump_common_cpp cpp; __be32 error; /* error code encountered while reading */ u8 padded_name_length; /* pad so data starts at 8 byte boundary */ - char rtsym[0]; + char rtsym[]; /* after padded_name_length, there is dump_length data */ }; @@ -92,7 +92,7 @@ struct nfp_dump_error { struct nfp_dump_tl tl; __be32 error; char padding[4]; - char spec[0]; + char spec[]; }; /* to track state through debug size calculation TLV traversal */ diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c index d648e32c0520..2779f1526d1e 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c @@ -1343,26 +1343,6 @@ static int nfp_net_set_coalesce(struct net_device *netdev, struct nfp_net *nn = netdev_priv(netdev); unsigned int factor; - if (ec->rx_coalesce_usecs_irq || - ec->rx_max_coalesced_frames_irq || - ec->tx_coalesce_usecs_irq || - ec->tx_max_coalesced_frames_irq || - ec->stats_block_coalesce_usecs || - ec->use_adaptive_rx_coalesce || - ec->use_adaptive_tx_coalesce || - ec->pkt_rate_low || - ec->rx_coalesce_usecs_low || - ec->rx_max_coalesced_frames_low || - ec->tx_coalesce_usecs_low || - ec->tx_max_coalesced_frames_low || - ec->pkt_rate_high || - ec->rx_coalesce_usecs_high || - ec->rx_max_coalesced_frames_high || - ec->tx_coalesce_usecs_high || - ec->tx_max_coalesced_frames_high || - ec->rate_sample_interval) - return -EOPNOTSUPP; - /* Compute factor used to convert coalesce '_usecs' parameters to * ME timestamp ticks. There are 16 ME clock cycles for each timestamp * count. @@ -1476,6 +1456,8 @@ static int nfp_net_set_channels(struct net_device *netdev, } static const struct ethtool_ops nfp_net_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = nfp_net_get_drvinfo, .get_link = ethtool_op_get_link, .get_ringparam = nfp_net_get_ringparam, diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.h b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.h index e0f13dfe1f39..48a74accbbd3 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.h +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.h @@ -18,7 +18,7 @@ struct nfp_port; */ struct nfp_reprs { unsigned int num_reprs; - struct net_device __rcu *reprs[0]; + struct net_device __rcu *reprs[]; }; /** diff --git a/drivers/net/ethernet/netronome/nfp/nfpcore/nfp6000_pcie.c b/drivers/net/ethernet/netronome/nfp/nfpcore/nfp6000_pcie.c index 684e4e036c55..a486008eb80a 100644 --- a/drivers/net/ethernet/netronome/nfp/nfpcore/nfp6000_pcie.c +++ b/drivers/net/ethernet/netronome/nfp/nfpcore/nfp6000_pcie.c @@ -1247,19 +1247,16 @@ static void nfp6000_free(struct nfp_cpp *cpp) static int nfp6000_read_serial(struct device *dev, u8 *serial) { struct pci_dev *pdev = to_pci_dev(dev); - int pos; - u32 reg; + u64 dsn; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_DSN); - if (!pos) { + dsn = pci_get_dsn(pdev); + if (!dsn) { dev_err(dev, "can't find PCIe Serial Number Capability\n"); return -EINVAL; } - pci_read_config_dword(pdev, pos + 4, ®); - put_unaligned_be16(reg >> 16, serial + 4); - pci_read_config_dword(pdev, pos + 8, ®); - put_unaligned_be32(reg, serial); + put_unaligned_be32((u32)(dsn >> 32), serial); + put_unaligned_be16((u16)(dsn >> 16), serial + 4); return 0; } @@ -1267,18 +1264,15 @@ static int nfp6000_read_serial(struct device *dev, u8 *serial) static int nfp6000_get_interface(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); - int pos; - u32 reg; + u64 dsn; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_DSN); - if (!pos) { + dsn = pci_get_dsn(pdev); + if (!dsn) { dev_err(dev, "can't find PCIe Serial Number Capability\n"); return -EINVAL; } - pci_read_config_dword(pdev, pos + 4, ®); - - return reg & 0xffff; + return dsn & 0xffff; } static const struct nfp_cpp_operations nfp6000_pcie_ops = { diff --git a/drivers/net/ethernet/netronome/nfp/nfpcore/nfp_nsp.h b/drivers/net/ethernet/netronome/nfp/nfpcore/nfp_nsp.h index 1531c1870020..f5360bae6f75 100644 --- a/drivers/net/ethernet/netronome/nfp/nfpcore/nfp_nsp.h +++ b/drivers/net/ethernet/netronome/nfp/nfpcore/nfp_nsp.h @@ -183,7 +183,7 @@ struct nfp_eth_table { bool is_split; unsigned int fec_modes_supported; - } ports[0]; + } ports[]; }; struct nfp_eth_table *nfp_eth_read_ports(struct nfp_cpp *cpp); diff --git a/drivers/net/ethernet/ni/nixge.c b/drivers/net/ethernet/ni/nixge.c index 49c7987c2abd..2fdd0753b3af 100644 --- a/drivers/net/ethernet/ni/nixge.c +++ b/drivers/net/ethernet/ni/nixge.c @@ -1019,27 +1019,6 @@ static int nixge_ethtools_set_coalesce(struct net_device *ndev, return -EBUSY; } - if (ecoalesce->rx_coalesce_usecs || - ecoalesce->rx_coalesce_usecs_irq || - ecoalesce->rx_max_coalesced_frames_irq || - ecoalesce->tx_coalesce_usecs || - ecoalesce->tx_coalesce_usecs_irq || - ecoalesce->tx_max_coalesced_frames_irq || - ecoalesce->stats_block_coalesce_usecs || - ecoalesce->use_adaptive_rx_coalesce || - ecoalesce->use_adaptive_tx_coalesce || - ecoalesce->pkt_rate_low || - ecoalesce->rx_coalesce_usecs_low || - ecoalesce->rx_max_coalesced_frames_low || - ecoalesce->tx_coalesce_usecs_low || - ecoalesce->tx_max_coalesced_frames_low || - ecoalesce->pkt_rate_high || - ecoalesce->rx_coalesce_usecs_high || - ecoalesce->rx_max_coalesced_frames_high || - ecoalesce->tx_coalesce_usecs_high || - ecoalesce->tx_max_coalesced_frames_high || - ecoalesce->rate_sample_interval) - return -EOPNOTSUPP; if (ecoalesce->rx_max_coalesced_frames) priv->coalesce_count_rx = ecoalesce->rx_max_coalesced_frames; if (ecoalesce->tx_max_coalesced_frames) @@ -1083,6 +1062,7 @@ static int nixge_ethtools_set_phys_id(struct net_device *ndev, } static const struct ethtool_ops nixge_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = nixge_ethtools_get_drvinfo, .get_coalesce = nixge_ethtools_get_coalesce, .set_coalesce = nixge_ethtools_set_coalesce, diff --git a/drivers/net/ethernet/pensando/ionic/ionic.h b/drivers/net/ethernet/pensando/ionic/ionic.h index bb106a32f416..23ccc0da2341 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic.h +++ b/drivers/net/ethernet/pensando/ionic/ionic.h @@ -12,12 +12,12 @@ struct ionic_lif; #define IONIC_DRV_NAME "ionic" #define IONIC_DRV_DESCRIPTION "Pensando Ethernet NIC Driver" -#define IONIC_DRV_VERSION "0.20.0-k" #define PCI_VENDOR_ID_PENSANDO 0x1dd8 #define PCI_DEVICE_ID_PENSANDO_IONIC_ETH_PF 0x1002 #define PCI_DEVICE_ID_PENSANDO_IONIC_ETH_VF 0x1003 +#define PCI_DEVICE_ID_PENSANDO_IONIC_ETH_MGMT 0x1004 #define DEVCMD_TIMEOUT 10 @@ -42,6 +42,7 @@ struct ionic { struct dentry *dentry; struct ionic_dev_bar bars[IONIC_BARS_MAX]; unsigned int num_bars; + bool is_mgmt_nic; struct ionic_identity ident; struct list_head lifs; struct ionic_lif *master_lif; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c index 448d7b23b2f7..60fc191a35e5 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c @@ -15,6 +15,7 @@ static const struct pci_device_id ionic_id_table[] = { { PCI_VDEVICE(PENSANDO, PCI_DEVICE_ID_PENSANDO_IONIC_ETH_PF) }, { PCI_VDEVICE(PENSANDO, PCI_DEVICE_ID_PENSANDO_IONIC_ETH_VF) }, + { PCI_VDEVICE(PENSANDO, PCI_DEVICE_ID_PENSANDO_IONIC_ETH_MGMT) }, { 0, } /* end of table */ }; MODULE_DEVICE_TABLE(pci, ionic_id_table); @@ -37,6 +38,9 @@ int ionic_bus_alloc_irq_vectors(struct ionic *ionic, unsigned int nintrs) void ionic_bus_free_irq_vectors(struct ionic *ionic) { + if (!ionic->nintrs) + return; + pci_free_irq_vectors(ionic->pdev); } @@ -221,6 +225,9 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_drvdata(pdev, ionic); mutex_init(&ionic->dev_cmd_lock); + ionic->is_mgmt_nic = + ent->device == PCI_DEVICE_ID_PENSANDO_IONIC_ETH_MGMT; + /* Query system for DMA addressing limitation for the device. */ err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(IONIC_ADDR_LEN)); if (err) { @@ -245,6 +252,8 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) } pci_set_master(pdev); + if (!ionic->is_mgmt_nic) + pcie_print_link_status(pdev); err = ionic_map_bars(ionic); if (err) @@ -346,6 +355,11 @@ err_out_reset: ionic_reset(ionic); err_out_teardown: ionic_dev_teardown(ionic); + /* Don't fail the probe for these errors, keep + * the hw interface around for inspection + */ + return 0; + err_out_unmap_bars: ionic_unmap_bars(ionic); pci_release_regions(pdev); @@ -369,11 +383,14 @@ static void ionic_remove(struct pci_dev *pdev) if (!ionic) return; - ionic_devlink_unregister(ionic); - ionic_lifs_unregister(ionic); - ionic_lifs_deinit(ionic); - ionic_lifs_free(ionic); - ionic_bus_free_irq_vectors(ionic); + if (ionic->master_lif) { + ionic_devlink_unregister(ionic); + ionic_lifs_unregister(ionic); + ionic_lifs_deinit(ionic); + ionic_lifs_free(ionic); + ionic_bus_free_irq_vectors(ionic); + } + ionic_port_reset(ionic); ionic_reset(ionic); ionic_dev_teardown(ionic); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c b/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c index bc03cecf80cc..5f8fc58d42b3 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c @@ -228,7 +228,13 @@ DEFINE_SHOW_ATTRIBUTE(netdev); void ionic_debugfs_add_lif(struct ionic_lif *lif) { - lif->dentry = debugfs_create_dir(lif->name, lif->ionic->dentry); + struct dentry *lif_dentry; + + lif_dentry = debugfs_create_dir(lif->name, lif->ionic->dentry); + if (IS_ERR_OR_NULL(lif_dentry)) + return; + lif->dentry = lif_dentry; + debugfs_create_file("netdev", 0400, lif->dentry, lif->netdev, &netdev_fops); } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c index 46107de5e6c3..f4ae40ae1e53 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c @@ -14,11 +14,15 @@ static void ionic_watchdog_cb(struct timer_list *t) { struct ionic *ionic = from_timer(ionic, t, watchdog_timer); + int hb; mod_timer(&ionic->watchdog_timer, round_jiffies(jiffies + ionic->watchdog_period)); - ionic_heartbeat_check(ionic); + hb = ionic_heartbeat_check(ionic); + + if (hb >= 0 && ionic->master_lif) + ionic_link_status_check_request(ionic->master_lif); } void ionic_init_devinfo(struct ionic *ionic) @@ -82,6 +86,7 @@ int ionic_dev_setup(struct ionic *ionic) return -EFAULT; } + idev->last_fw_status = 0xff; timer_setup(&ionic->watchdog_timer, ionic_watchdog_cb, 0); ionic->watchdog_period = IONIC_WATCHDOG_SECS * HZ; mod_timer(&ionic->watchdog_timer, @@ -115,8 +120,43 @@ int ionic_heartbeat_check(struct ionic *ionic) * fw_status != 0xff (bad PCI read) */ fw_status = ioread8(&idev->dev_info_regs->fw_status); - if (fw_status == 0xff || - !(fw_status & IONIC_FW_STS_F_RUNNING)) + if (fw_status != 0xff) + fw_status &= IONIC_FW_STS_F_RUNNING; /* use only the run bit */ + + /* is this a transition? */ + if (fw_status != idev->last_fw_status && + idev->last_fw_status != 0xff) { + struct ionic_lif *lif = ionic->master_lif; + bool trigger = false; + + if (!fw_status || fw_status == 0xff) { + dev_info(ionic->dev, "FW stopped %u\n", fw_status); + if (lif && !test_bit(IONIC_LIF_F_FW_RESET, lif->state)) + trigger = true; + } else { + dev_info(ionic->dev, "FW running %u\n", fw_status); + if (lif && test_bit(IONIC_LIF_F_FW_RESET, lif->state)) + trigger = true; + } + + if (trigger) { + struct ionic_deferred_work *work; + + work = kzalloc(sizeof(*work), GFP_ATOMIC); + if (!work) { + dev_err(ionic->dev, "%s OOM\n", __func__); + } else { + work->type = IONIC_DW_TYPE_LIF_RESET; + if (fw_status & IONIC_FW_STS_F_RUNNING && + fw_status != 0xff) + work->fw_status = 1; + ionic_lif_deferred_enqueue(&lif->deferred, work); + } + } + } + idev->last_fw_status = fw_status; + + if (!fw_status || fw_status == 0xff) return -ENXIO; /* early FW has no heartbeat, else FW will return non-zero */ diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.h b/drivers/net/ethernet/pensando/ionic/ionic_dev.h index 7838e342c4fd..587398b01997 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.h @@ -132,6 +132,7 @@ struct ionic_dev { unsigned long last_hb_time; u32 last_hb; + u8 last_fw_status; u64 __iomem *db_pages; dma_addr_t phy_db_pages; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_devlink.c b/drivers/net/ethernet/pensando/ionic/ionic_devlink.c index 6fb27dcc5787..273c889faaad 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_devlink.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_devlink.c @@ -77,6 +77,10 @@ int ionic_devlink_register(struct ionic *ionic) return err; } + /* don't register the mgmt_nic as a port */ + if (ionic->is_mgmt_nic) + return 0; + devlink_port_attrs_set(&ionic->dl_port, DEVLINK_PORT_FLAVOUR_PHYSICAL, 0, false, 0, NULL, 0); err = devlink_port_register(dl, &ionic->dl_port, 0); @@ -93,6 +97,7 @@ void ionic_devlink_unregister(struct ionic *ionic) { struct devlink *dl = priv_to_devlink(ionic); - devlink_port_unregister(&ionic->dl_port); + if (ionic->dl_port.registered) + devlink_port_unregister(&ionic->dl_port); devlink_unregister(dl); } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c index f778fff034f5..6996229facfd 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c @@ -3,6 +3,7 @@ #include <linux/module.h> #include <linux/netdevice.h> +#include <linux/sfp.h> #include "ionic.h" #include "ionic_bus.h" @@ -86,7 +87,6 @@ static void ionic_get_drvinfo(struct net_device *netdev, struct ionic *ionic = lif->ionic; strlcpy(drvinfo->driver, IONIC_DRV_NAME, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, IONIC_DRV_VERSION, sizeof(drvinfo->version)); strlcpy(drvinfo->fw_version, ionic->idev.dev_info.fw_version, sizeof(drvinfo->fw_version)); strlcpy(drvinfo->bus_info, ionic_bus_info(ionic), @@ -412,28 +412,6 @@ static int ionic_set_coalesce(struct net_device *netdev, unsigned int i; u32 coal; - if (coalesce->rx_max_coalesced_frames || - coalesce->rx_coalesce_usecs_irq || - coalesce->rx_max_coalesced_frames_irq || - coalesce->tx_max_coalesced_frames || - coalesce->tx_coalesce_usecs_irq || - coalesce->tx_max_coalesced_frames_irq || - coalesce->stats_block_coalesce_usecs || - coalesce->use_adaptive_rx_coalesce || - coalesce->use_adaptive_tx_coalesce || - coalesce->pkt_rate_low || - coalesce->rx_coalesce_usecs_low || - coalesce->rx_max_coalesced_frames_low || - coalesce->tx_coalesce_usecs_low || - coalesce->tx_max_coalesced_frames_low || - coalesce->pkt_rate_high || - coalesce->rx_coalesce_usecs_high || - coalesce->rx_max_coalesced_frames_high || - coalesce->tx_coalesce_usecs_high || - coalesce->tx_max_coalesced_frames_high || - coalesce->rate_sample_interval) - return -EINVAL; - ident = &lif->ionic->ident; if (ident->dev.intr_coal_div == 0) { netdev_warn(netdev, "bad HW value in dev.intr_coal_div = %d\n", @@ -462,7 +440,7 @@ static int ionic_set_coalesce(struct net_device *netdev, if (coal != lif->rx_coalesce_hw) { lif->rx_coalesce_hw = coal; - if (test_bit(IONIC_LIF_UP, lif->state)) { + if (test_bit(IONIC_LIF_F_UP, lif->state)) { for (i = 0; i < lif->nxqs; i++) { qcq = lif->rxqcqs[i].qcq; ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, @@ -509,11 +487,11 @@ static int ionic_set_ringparam(struct net_device *netdev, ring->rx_pending == lif->nrxq_descs) return 0; - err = ionic_wait_for_bit(lif, IONIC_LIF_QUEUE_RESET); + err = ionic_wait_for_bit(lif, IONIC_LIF_F_QUEUE_RESET); if (err) return err; - running = test_bit(IONIC_LIF_UP, lif->state); + running = test_bit(IONIC_LIF_F_UP, lif->state); if (running) ionic_stop(netdev); @@ -522,7 +500,7 @@ static int ionic_set_ringparam(struct net_device *netdev, if (running) ionic_open(netdev); - clear_bit(IONIC_LIF_QUEUE_RESET, lif->state); + clear_bit(IONIC_LIF_F_QUEUE_RESET, lif->state); return 0; } @@ -553,11 +531,11 @@ static int ionic_set_channels(struct net_device *netdev, if (ch->combined_count == lif->nxqs) return 0; - err = ionic_wait_for_bit(lif, IONIC_LIF_QUEUE_RESET); + err = ionic_wait_for_bit(lif, IONIC_LIF_F_QUEUE_RESET); if (err) return err; - running = test_bit(IONIC_LIF_UP, lif->state); + running = test_bit(IONIC_LIF_F_UP, lif->state); if (running) ionic_stop(netdev); @@ -565,7 +543,7 @@ static int ionic_set_channels(struct net_device *netdev, if (running) ionic_open(netdev); - clear_bit(IONIC_LIF_QUEUE_RESET, lif->state); + clear_bit(IONIC_LIF_F_QUEUE_RESET, lif->state); return 0; } @@ -575,7 +553,7 @@ static u32 ionic_get_priv_flags(struct net_device *netdev) struct ionic_lif *lif = netdev_priv(netdev); u32 priv_flags = 0; - if (test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) + if (test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) priv_flags |= PRIV_F_SW_DBG_STATS; return priv_flags; @@ -584,14 +562,10 @@ static u32 ionic_get_priv_flags(struct net_device *netdev) static int ionic_set_priv_flags(struct net_device *netdev, u32 priv_flags) { struct ionic_lif *lif = netdev_priv(netdev); - u32 flags = lif->flags; - clear_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state); + clear_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state); if (priv_flags & PRIV_F_SW_DBG_STATS) - set_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state); - - if (flags != lif->flags) - lif->flags = flags; + set_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state); return 0; } @@ -704,23 +678,27 @@ static int ionic_get_module_info(struct net_device *netdev, struct ionic_lif *lif = netdev_priv(netdev); struct ionic_dev *idev = &lif->ionic->idev; struct ionic_xcvr_status *xcvr; + struct sfp_eeprom_base *sfp; xcvr = &idev->port_info->status.xcvr; + sfp = (struct sfp_eeprom_base *) xcvr->sprom; /* report the module data type and length */ - switch (xcvr->sprom[0]) { - case 0x03: /* SFP */ + switch (sfp->phys_id) { + case SFF8024_ID_SFP: modinfo->type = ETH_MODULE_SFF_8079; modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN; break; - case 0x0D: /* QSFP */ - case 0x11: /* QSFP28 */ + case SFF8024_ID_QSFP_8436_8636: + case SFF8024_ID_QSFP28_8636: modinfo->type = ETH_MODULE_SFF_8436; modinfo->eeprom_len = ETH_MODULE_SFF_8436_LEN; break; default: netdev_info(netdev, "unknown xcvr type 0x%02x\n", xcvr->sprom[0]); + modinfo->type = 0; + modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN; break; } @@ -784,6 +762,7 @@ static int ionic_nway_reset(struct net_device *netdev) } static const struct ethtool_ops ionic_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_drvinfo = ionic_get_drvinfo, .get_regs_len = ionic_get_regs_len, .get_regs = ionic_get_regs, diff --git a/drivers/net/ethernet/pensando/ionic/ionic_if.h b/drivers/net/ethernet/pensando/ionic/ionic_if.h index 51adf5059834..ceeb7629e7a0 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_if.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_if.h @@ -4,8 +4,6 @@ #ifndef _IONIC_IF_H_ #define _IONIC_IF_H_ -#pragma pack(push, 1) - #define IONIC_DEV_INFO_SIGNATURE 0x44455649 /* 'DEVI' */ #define IONIC_DEV_INFO_VERSION 1 #define IONIC_IFNAMSIZ 16 @@ -366,7 +364,7 @@ union ionic_lif_config { u8 rsvd2[2]; __le64 features; __le32 queue_count[IONIC_QTYPE_MAX]; - }; + } __packed; __le32 words[64]; }; @@ -417,7 +415,7 @@ union ionic_lif_identity { __le32 max_frame_size; u8 rsvd2[106]; union ionic_lif_config config; - } eth; + } __packed eth; struct { u8 version; @@ -439,8 +437,8 @@ union ionic_lif_identity { struct ionic_lif_logical_qtype rq_qtype; struct ionic_lif_logical_qtype cq_qtype; struct ionic_lif_logical_qtype eq_qtype; - } rdma; - }; + } __packed rdma; + } __packed; __le32 words[512]; }; @@ -526,7 +524,7 @@ struct ionic_q_init_cmd { __le64 sg_ring_base; __le32 eq_index; u8 rsvd2[16]; -}; +} __packed; /** * struct ionic_q_init_comp - Queue init command completion @@ -1095,7 +1093,7 @@ struct ionic_port_status { u8 status; u8 rsvd[51]; struct ionic_xcvr_status xcvr; -}; +} __packed; /** * struct ionic_port_identify_cmd - Port identify command @@ -1251,7 +1249,7 @@ struct ionic_port_getattr_comp { u8 pause_type; u8 loopback_mode; u8 rsvd2[11]; - }; + } __packed; u8 color; }; @@ -1319,7 +1317,7 @@ struct ionic_dev_setattr_cmd { char name[IONIC_IFNAMSIZ]; __le64 features; u8 rsvd2[60]; - }; + } __packed; }; /** @@ -1334,7 +1332,7 @@ struct ionic_dev_setattr_comp { union { __le64 features; u8 rsvd2[11]; - }; + } __packed; u8 color; }; @@ -1361,7 +1359,7 @@ struct ionic_dev_getattr_comp { union { __le64 features; u8 rsvd2[11]; - }; + } __packed; u8 color; }; @@ -1426,7 +1424,7 @@ struct ionic_lif_setattr_cmd { } rss; u8 stats_ctl; u8 rsvd[60]; - }; + } __packed; }; /** @@ -1444,7 +1442,7 @@ struct ionic_lif_setattr_comp { union { __le64 features; u8 rsvd2[11]; - }; + } __packed; u8 color; }; @@ -1483,7 +1481,7 @@ struct ionic_lif_getattr_comp { u8 mac[6]; __le64 features; u8 rsvd2[11]; - }; + } __packed; u8 color; }; @@ -1688,7 +1686,7 @@ struct ionic_vf_setattr_cmd { u8 linkstate; __le64 stats_pa; u8 pad[60]; - }; + } __packed; }; struct ionic_vf_setattr_comp { @@ -1726,7 +1724,7 @@ struct ionic_vf_getattr_comp { u8 linkstate; __le64 stats_pa; u8 pad[11]; - }; + } __packed; u8 color; }; @@ -2472,7 +2470,7 @@ union ionic_dev_cmd_regs { union ionic_dev_cmd_comp comp; u8 rsvd[48]; u32 data[478]; - }; + } __packed; u32 words[512]; }; @@ -2485,7 +2483,7 @@ union ionic_dev_regs { struct { union ionic_dev_info_regs info; union ionic_dev_cmd_regs devcmd; - }; + } __packed; __le32 words[1024]; }; @@ -2575,6 +2573,4 @@ struct ionic_identity { union ionic_qos_identity qos; }; -#pragma pack(pop) - #endif /* _IONIC_IF_H_ */ diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index 938e19ee0bcd..4b8a76098ca3 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -21,6 +21,12 @@ static void ionic_lif_rx_mode(struct ionic_lif *lif, unsigned int rx_mode); static int ionic_lif_addr_add(struct ionic_lif *lif, const u8 *addr); static int ionic_lif_addr_del(struct ionic_lif *lif, const u8 *addr); static void ionic_link_status_check(struct ionic_lif *lif); +static void ionic_lif_handle_fw_down(struct ionic_lif *lif); +static void ionic_lif_handle_fw_up(struct ionic_lif *lif); +static void ionic_lif_set_netdev_info(struct ionic_lif *lif); + +static int ionic_start_queues(struct ionic_lif *lif); +static void ionic_stop_queues(struct ionic_lif *lif); static void ionic_lif_deferred_work(struct work_struct *work) { @@ -50,6 +56,12 @@ static void ionic_lif_deferred_work(struct work_struct *work) case IONIC_DW_TYPE_LINK_STATUS: ionic_link_status_check(lif); break; + case IONIC_DW_TYPE_LIF_RESET: + if (w->fw_status) + ionic_lif_handle_fw_up(lif); + else + ionic_lif_handle_fw_down(lif); + break; default: break; } @@ -58,8 +70,8 @@ static void ionic_lif_deferred_work(struct work_struct *work) } } -static void ionic_lif_deferred_enqueue(struct ionic_deferred *def, - struct ionic_deferred_work *work) +void ionic_lif_deferred_enqueue(struct ionic_deferred *def, + struct ionic_deferred_work *work) { spin_lock_bh(&def->lock); list_add_tail(&work->list, &def->list); @@ -73,40 +85,47 @@ static void ionic_link_status_check(struct ionic_lif *lif) u16 link_status; bool link_up; + if (!test_bit(IONIC_LIF_F_LINK_CHECK_REQUESTED, lif->state)) + return; + + if (lif->ionic->is_mgmt_nic) + return; + link_status = le16_to_cpu(lif->info->status.link_status); link_up = link_status == IONIC_PORT_OPER_STATUS_UP; - /* filter out the no-change cases */ - if (link_up == netif_carrier_ok(netdev)) - goto link_out; - if (link_up) { - netdev_info(netdev, "Link up - %d Gbps\n", - le32_to_cpu(lif->info->status.link_speed) / 1000); + if (!netif_carrier_ok(netdev)) { + u32 link_speed; - if (test_bit(IONIC_LIF_UP, lif->state)) { - netif_tx_wake_all_queues(lif->netdev); + ionic_port_identify(lif->ionic); + link_speed = le32_to_cpu(lif->info->status.link_speed); + netdev_info(netdev, "Link up - %d Gbps\n", + link_speed / 1000); netif_carrier_on(netdev); } + + if (netif_running(lif->netdev)) + ionic_start_queues(lif); } else { - netdev_info(netdev, "Link down\n"); + if (netif_carrier_ok(netdev)) { + netdev_info(netdev, "Link down\n"); + netif_carrier_off(netdev); + } - /* carrier off first to avoid watchdog timeout */ - netif_carrier_off(netdev); - if (test_bit(IONIC_LIF_UP, lif->state)) - netif_tx_stop_all_queues(netdev); + if (netif_running(lif->netdev)) + ionic_stop_queues(lif); } -link_out: - clear_bit(IONIC_LIF_LINK_CHECK_REQUESTED, lif->state); + clear_bit(IONIC_LIF_F_LINK_CHECK_REQUESTED, lif->state); } -static void ionic_link_status_check_request(struct ionic_lif *lif) +void ionic_link_status_check_request(struct ionic_lif *lif) { struct ionic_deferred_work *work; /* we only need one request outstanding at a time */ - if (test_and_set_bit(IONIC_LIF_LINK_CHECK_REQUESTED, lif->state)) + if (test_and_set_bit(IONIC_LIF_F_LINK_CHECK_REQUESTED, lif->state)) return; if (in_interrupt()) { @@ -244,38 +263,19 @@ static int ionic_qcq_disable(struct ionic_qcq *qcq) return ionic_adminq_post_wait(lif, &ctx); } -static void ionic_lif_quiesce(struct ionic_lif *lif) -{ - struct ionic_admin_ctx ctx = { - .work = COMPLETION_INITIALIZER_ONSTACK(ctx.work), - .cmd.lif_setattr = { - .opcode = IONIC_CMD_LIF_SETATTR, - .attr = IONIC_LIF_ATTR_STATE, - .index = lif->index, - .state = IONIC_LIF_DISABLE - }, - }; - - ionic_adminq_post_wait(lif, &ctx); -} - static void ionic_lif_qcq_deinit(struct ionic_lif *lif, struct ionic_qcq *qcq) { struct ionic_dev *idev = &lif->ionic->idev; - struct device *dev = lif->ionic->dev; if (!qcq) return; - ionic_debugfs_del_qcq(qcq); - if (!(qcq->flags & IONIC_QCQ_F_INITED)) return; if (qcq->flags & IONIC_QCQ_F_INTR) { ionic_intr_mask(idev->intr_ctrl, qcq->intr.index, IONIC_INTR_MASK_SET); - devm_free_irq(dev, qcq->intr.vector, &qcq->napi); netif_napi_del(&qcq->napi); } @@ -289,12 +289,18 @@ static void ionic_qcq_free(struct ionic_lif *lif, struct ionic_qcq *qcq) if (!qcq) return; + ionic_debugfs_del_qcq(qcq); + dma_free_coherent(dev, qcq->total_size, qcq->base, qcq->base_pa); qcq->base = NULL; qcq->base_pa = 0; - if (qcq->flags & IONIC_QCQ_F_INTR) + if (qcq->flags & IONIC_QCQ_F_INTR) { + irq_set_affinity_hint(qcq->intr.vector, NULL); + devm_free_irq(dev, qcq->intr.vector, &qcq->napi); + qcq->intr.vector = 0; ionic_intr_free(lif, qcq->intr.index); + } devm_kfree(dev, qcq->cq.info); qcq->cq.info = NULL; @@ -318,19 +324,21 @@ static void ionic_qcqs_free(struct ionic_lif *lif) lif->adminqcq = NULL; } - for (i = 0; i < lif->nxqs; i++) - if (lif->rxqcqs[i].stats) - devm_kfree(dev, lif->rxqcqs[i].stats); - - devm_kfree(dev, lif->rxqcqs); - lif->rxqcqs = NULL; - - for (i = 0; i < lif->nxqs; i++) - if (lif->txqcqs[i].stats) - devm_kfree(dev, lif->txqcqs[i].stats); + if (lif->rxqcqs) { + for (i = 0; i < lif->nxqs; i++) + if (lif->rxqcqs[i].stats) + devm_kfree(dev, lif->rxqcqs[i].stats); + devm_kfree(dev, lif->rxqcqs); + lif->rxqcqs = NULL; + } - devm_kfree(dev, lif->txqcqs); - lif->txqcqs = NULL; + if (lif->txqcqs) { + for (i = 0; i < lif->nxqs; i++) + if (lif->txqcqs[i].stats) + devm_kfree(dev, lif->txqcqs[i].stats); + devm_kfree(dev, lif->txqcqs); + lif->txqcqs = NULL; + } } static void ionic_link_qcq_interrupts(struct ionic_qcq *src_qcq, @@ -424,8 +432,15 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, ionic_intr_mask_assert(idev->intr_ctrl, new->intr.index, IONIC_INTR_MASK_SET); - new->intr.cpu = new->intr.index % num_online_cpus(); - if (cpu_online(new->intr.cpu)) + err = ionic_request_irq(lif, new); + if (err) { + netdev_warn(lif->netdev, "irq request failed %d\n", err); + goto err_out_free_intr; + } + + new->intr.cpu = cpumask_local_spread(new->intr.index, + dev_to_node(dev)); + if (new->intr.cpu != -1) cpumask_set_cpu(new->intr.cpu, &new->intr.affinity_mask); } else { @@ -437,13 +452,13 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, if (!new->cq.info) { netdev_err(lif->netdev, "Cannot allocate completion queue info\n"); err = -ENOMEM; - goto err_out_free_intr; + goto err_out_free_irq; } err = ionic_cq_init(lif, &new->cq, &new->intr, num_descs, cq_desc_size); if (err) { netdev_err(lif->netdev, "Cannot initialize completion queue\n"); - goto err_out_free_intr; + goto err_out_free_irq; } new->base = dma_alloc_coherent(dev, total_size, &new->base_pa, @@ -451,7 +466,7 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, if (!new->base) { netdev_err(lif->netdev, "Cannot allocate queue DMA memory\n"); err = -ENOMEM; - goto err_out_free_intr; + goto err_out_free_irq; } new->total_size = total_size; @@ -477,8 +492,12 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, return 0; +err_out_free_irq: + if (flags & IONIC_QCQ_F_INTR) + devm_free_irq(dev, new->intr.vector, &new->napi); err_out_free_intr: - ionic_intr_free(lif, new->intr.index); + if (flags & IONIC_QCQ_F_INTR) + ionic_intr_free(lif, new->intr.index); err_out: dev_err(dev, "qcq alloc of %s%d failed %d\n", name, index, err); return err; @@ -500,6 +519,7 @@ static int ionic_qcqs_alloc(struct ionic_lif *lif) 0, lif->kern_pid, &lif->adminqcq); if (err) return err; + ionic_debugfs_add_qcq(lif, lif->adminqcq); if (lif->ionic->nnqs_per_lif) { flags = IONIC_QCQ_F_NOTIFYQ; @@ -510,6 +530,7 @@ static int ionic_qcqs_alloc(struct ionic_lif *lif) 0, lif->kern_pid, &lif->notifyqcq); if (err) goto err_out_free_adminqcq; + ionic_debugfs_add_qcq(lif, lif->notifyqcq); /* Let the notifyq ride on the adminq interrupt */ ionic_link_qcq_interrupts(lif->adminqcq, lif->notifyqcq); @@ -594,6 +615,10 @@ static int ionic_lif_txq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) dev_dbg(dev, "txq_init.ring_base 0x%llx\n", ctx.cmd.q_init.ring_base); dev_dbg(dev, "txq_init.ring_size %d\n", ctx.cmd.q_init.ring_size); + q->tail = q->info; + q->head = q->tail; + cq->tail = cq->info; + err = ionic_adminq_post_wait(lif, &ctx); if (err) return err; @@ -607,8 +632,6 @@ static int ionic_lif_txq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) qcq->flags |= IONIC_QCQ_F_INITED; - ionic_debugfs_add_qcq(lif, qcq); - return 0; } @@ -641,6 +664,10 @@ static int ionic_lif_rxq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) dev_dbg(dev, "rxq_init.ring_base 0x%llx\n", ctx.cmd.q_init.ring_base); dev_dbg(dev, "rxq_init.ring_size %d\n", ctx.cmd.q_init.ring_size); + q->tail = q->info; + q->head = q->tail; + cq->tail = cq->info; + err = ionic_adminq_post_wait(lif, &ctx); if (err) return err; @@ -655,16 +682,8 @@ static int ionic_lif_rxq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) netif_napi_add(lif->netdev, &qcq->napi, ionic_rx_napi, NAPI_POLL_WEIGHT); - err = ionic_request_irq(lif, qcq); - if (err) { - netif_napi_del(&qcq->napi); - return err; - } - qcq->flags |= IONIC_QCQ_F_INITED; - ionic_debugfs_add_qcq(lif, qcq); - return 0; } @@ -672,6 +691,7 @@ static bool ionic_notifyq_service(struct ionic_cq *cq, struct ionic_cq_info *cq_info) { union ionic_notifyq_comp *comp = cq_info->cq_desc; + struct ionic_deferred_work *work; struct net_device *netdev; struct ionic_queue *q; struct ionic_lif *lif; @@ -697,11 +717,13 @@ static bool ionic_notifyq_service(struct ionic_cq *cq, ionic_link_status_check_request(lif); break; case IONIC_EVENT_RESET: - netdev_info(netdev, "Notifyq IONIC_EVENT_RESET eid=%lld\n", - eid); - netdev_info(netdev, " reset_code=%d state=%d\n", - comp->reset.reset_code, - comp->reset.state); + work = kzalloc(sizeof(*work), GFP_ATOMIC); + if (!work) { + netdev_err(lif->netdev, "%s OOM\n", __func__); + } else { + work->type = IONIC_DW_TYPE_LIF_RESET; + ionic_lif_deferred_enqueue(&lif->deferred, work); + } break; default: netdev_warn(netdev, "Notifyq unknown event ecode=%d eid=%lld\n", @@ -831,7 +853,7 @@ static int ionic_lif_addr_add(struct ionic_lif *lif, const u8 *addr) memcpy(ctx.cmd.rx_filter_add.mac.addr, addr, ETH_ALEN); err = ionic_adminq_post_wait(lif, &ctx); - if (err) + if (err && err != -EEXIST) return err; return ionic_rx_filter_save(lif, 0, IONIC_RXQ_INDEX_ANY, 0, &ctx); @@ -861,7 +883,7 @@ static int ionic_lif_addr_del(struct ionic_lif *lif, const u8 *addr) spin_unlock_bh(&lif->rx_filters.lock); err = ionic_adminq_post_wait(lif, &ctx); - if (err) + if (err && err != -EEXIST) return err; netdev_dbg(lif->netdev, "rx_filter del ADDR %pM (id %d)\n", addr, @@ -1093,6 +1115,7 @@ static int ionic_set_nic_features(struct ionic_lif *lif, u64 vlan_flags = IONIC_ETH_HW_VLAN_TX_TAG | IONIC_ETH_HW_VLAN_RX_STRIP | IONIC_ETH_HW_VLAN_RX_FILTER; + u64 old_hw_features; int err; ctx.cmd.lif_setattr.features = ionic_netdev_features_to_nic(features); @@ -1100,9 +1123,13 @@ static int ionic_set_nic_features(struct ionic_lif *lif, if (err) return err; + old_hw_features = lif->hw_features; lif->hw_features = le64_to_cpu(ctx.cmd.lif_setattr.features & ctx.comp.lif_setattr.features); + if ((old_hw_features ^ lif->hw_features) & IONIC_ETH_HW_RX_HASH) + ionic_lif_rss_config(lif, lif->rss_types, NULL, NULL); + if ((vlan_flags & features) && !(vlan_flags & le64_to_cpu(ctx.comp.lif_setattr.features))) dev_info_once(lif->ionic->dev, "NIC is not supporting vlan offload, likely in SmartNIC mode\n"); @@ -1149,6 +1176,10 @@ static int ionic_init_nic_features(struct ionic_lif *lif) netdev_features_t features; int err; + /* no netdev features on the management device */ + if (lif->ionic->is_mgmt_nic) + return 0; + /* set up what we expect to support by default */ features = NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX | @@ -1205,7 +1236,8 @@ static int ionic_init_nic_features(struct ionic_lif *lif) netdev->hw_features |= netdev->hw_enc_features; netdev->features |= netdev->hw_features; - netdev->priv_flags |= IFF_UNICAST_FLT; + netdev->priv_flags |= IFF_UNICAST_FLT | + IFF_LIVE_ADDR_CHANGE; return 0; } @@ -1356,13 +1388,15 @@ int ionic_lif_rss_config(struct ionic_lif *lif, const u16 types, .cmd.lif_setattr = { .opcode = IONIC_CMD_LIF_SETATTR, .attr = IONIC_LIF_ATTR_RSS, - .rss.types = cpu_to_le16(types), .rss.addr = cpu_to_le64(lif->rss_ind_tbl_pa), }, }; unsigned int i, tbl_sz; - lif->rss_types = types; + if (lif->hw_features & IONIC_ETH_HW_RX_HASH) { + lif->rss_types = types; + ctx.cmd.lif_setattr.rss.types = cpu_to_le16(types); + } if (key) memcpy(lif->rss_hash_key, key, IONIC_RSS_HASH_KEY_SIZE); @@ -1413,10 +1447,22 @@ static void ionic_lif_rss_deinit(struct ionic_lif *lif) static void ionic_txrx_disable(struct ionic_lif *lif) { unsigned int i; + int err; - for (i = 0; i < lif->nxqs; i++) { - ionic_qcq_disable(lif->txqcqs[i].qcq); - ionic_qcq_disable(lif->rxqcqs[i].qcq); + if (lif->txqcqs) { + for (i = 0; i < lif->nxqs; i++) { + err = ionic_qcq_disable(lif->txqcqs[i].qcq); + if (err == -ETIMEDOUT) + break; + } + } + + if (lif->rxqcqs) { + for (i = 0; i < lif->nxqs; i++) { + err = ionic_qcq_disable(lif->rxqcqs[i].qcq); + if (err == -ETIMEDOUT) + break; + } } } @@ -1424,26 +1470,40 @@ static void ionic_txrx_deinit(struct ionic_lif *lif) { unsigned int i; - for (i = 0; i < lif->nxqs; i++) { - ionic_lif_qcq_deinit(lif, lif->txqcqs[i].qcq); - ionic_tx_flush(&lif->txqcqs[i].qcq->cq); + if (lif->txqcqs) { + for (i = 0; i < lif->nxqs; i++) { + ionic_lif_qcq_deinit(lif, lif->txqcqs[i].qcq); + ionic_tx_flush(&lif->txqcqs[i].qcq->cq); + ionic_tx_empty(&lif->txqcqs[i].qcq->q); + } + } - ionic_lif_qcq_deinit(lif, lif->rxqcqs[i].qcq); - ionic_rx_flush(&lif->rxqcqs[i].qcq->cq); - ionic_rx_empty(&lif->rxqcqs[i].qcq->q); + if (lif->rxqcqs) { + for (i = 0; i < lif->nxqs; i++) { + ionic_lif_qcq_deinit(lif, lif->rxqcqs[i].qcq); + ionic_rx_flush(&lif->rxqcqs[i].qcq->cq); + ionic_rx_empty(&lif->rxqcqs[i].qcq->q); + } } + lif->rx_mode = 0; } static void ionic_txrx_free(struct ionic_lif *lif) { unsigned int i; - for (i = 0; i < lif->nxqs; i++) { - ionic_qcq_free(lif, lif->txqcqs[i].qcq); - lif->txqcqs[i].qcq = NULL; + if (lif->txqcqs) { + for (i = 0; i < lif->nxqs; i++) { + ionic_qcq_free(lif, lif->txqcqs[i].qcq); + lif->txqcqs[i].qcq = NULL; + } + } - ionic_qcq_free(lif, lif->rxqcqs[i].qcq); - lif->rxqcqs[i].qcq = NULL; + if (lif->rxqcqs) { + for (i = 0; i < lif->nxqs; i++) { + ionic_qcq_free(lif, lif->rxqcqs[i].qcq); + lif->rxqcqs[i].qcq = NULL; + } } } @@ -1465,6 +1525,7 @@ static int ionic_txrx_alloc(struct ionic_lif *lif) goto err_out; lif->txqcqs[i].qcq->stats = lif->txqcqs[i].stats; + ionic_debugfs_add_qcq(lif, lif->txqcqs[i].qcq); } flags = IONIC_QCQ_F_RX_STATS | IONIC_QCQ_F_SG | IONIC_QCQ_F_INTR; @@ -1485,6 +1546,7 @@ static int ionic_txrx_alloc(struct ionic_lif *lif) lif->rx_coalesce_hw); ionic_link_qcq_interrupts(lif->rxqcqs[i].qcq, lif->txqcqs[i].qcq); + ionic_debugfs_add_qcq(lif, lif->rxqcqs[i].qcq); } return 0; @@ -1533,14 +1595,15 @@ static int ionic_txrx_enable(struct ionic_lif *lif) int i, err; for (i = 0; i < lif->nxqs; i++) { - err = ionic_qcq_enable(lif->txqcqs[i].qcq); + ionic_rx_fill(&lif->rxqcqs[i].qcq->q); + err = ionic_qcq_enable(lif->rxqcqs[i].qcq); if (err) goto err_out; - ionic_rx_fill(&lif->rxqcqs[i].qcq->q); - err = ionic_qcq_enable(lif->rxqcqs[i].qcq); + err = ionic_qcq_enable(lif->txqcqs[i].qcq); if (err) { - ionic_qcq_disable(lif->txqcqs[i].qcq); + if (err != -ETIMEDOUT) + ionic_qcq_disable(lif->rxqcqs[i].qcq); goto err_out; } } @@ -1549,74 +1612,84 @@ static int ionic_txrx_enable(struct ionic_lif *lif) err_out: while (i--) { - ionic_qcq_disable(lif->rxqcqs[i].qcq); - ionic_qcq_disable(lif->txqcqs[i].qcq); + err = ionic_qcq_disable(lif->txqcqs[i].qcq); + if (err == -ETIMEDOUT) + break; + err = ionic_qcq_disable(lif->rxqcqs[i].qcq); + if (err == -ETIMEDOUT) + break; } return err; } +static int ionic_start_queues(struct ionic_lif *lif) +{ + int err; + + if (test_and_set_bit(IONIC_LIF_F_UP, lif->state)) + return 0; + + err = ionic_txrx_enable(lif); + if (err) { + clear_bit(IONIC_LIF_F_UP, lif->state); + return err; + } + netif_tx_wake_all_queues(lif->netdev); + + return 0; +} + int ionic_open(struct net_device *netdev) { struct ionic_lif *lif = netdev_priv(netdev); int err; - netif_carrier_off(netdev); - err = ionic_txrx_alloc(lif); if (err) return err; err = ionic_txrx_init(lif); if (err) - goto err_txrx_free; - - err = ionic_txrx_enable(lif); - if (err) - goto err_txrx_deinit; - - netif_set_real_num_tx_queues(netdev, lif->nxqs); - netif_set_real_num_rx_queues(netdev, lif->nxqs); - - set_bit(IONIC_LIF_UP, lif->state); + goto err_out; - ionic_link_status_check_request(lif); - if (netif_carrier_ok(netdev)) - netif_tx_wake_all_queues(netdev); + /* don't start the queues until we have link */ + if (netif_carrier_ok(netdev)) { + err = ionic_start_queues(lif); + if (err) + goto err_txrx_deinit; + } return 0; err_txrx_deinit: ionic_txrx_deinit(lif); -err_txrx_free: +err_out: ionic_txrx_free(lif); return err; } +static void ionic_stop_queues(struct ionic_lif *lif) +{ + if (!test_and_clear_bit(IONIC_LIF_F_UP, lif->state)) + return; + + ionic_txrx_disable(lif); + netif_tx_disable(lif->netdev); +} + int ionic_stop(struct net_device *netdev) { struct ionic_lif *lif = netdev_priv(netdev); - int err = 0; - if (!test_bit(IONIC_LIF_UP, lif->state)) { - dev_dbg(lif->ionic->dev, "%s: %s state=DOWN\n", - __func__, lif->name); + if (test_bit(IONIC_LIF_F_FW_RESET, lif->state)) return 0; - } - dev_dbg(lif->ionic->dev, "%s: %s state=UP\n", __func__, lif->name); - clear_bit(IONIC_LIF_UP, lif->state); - - /* carrier off before disabling queues to avoid watchdog timeout */ - netif_carrier_off(netdev); - netif_tx_stop_all_queues(netdev); - netif_tx_disable(netdev); - ionic_txrx_disable(lif); - ionic_lif_quiesce(lif); + ionic_stop_queues(lif); ionic_txrx_deinit(lif); ionic_txrx_free(lif); - return err; + return 0; } static int ionic_get_vf_config(struct net_device *netdev, @@ -1871,7 +1944,7 @@ int ionic_reset_queues(struct ionic_lif *lif) /* Put off the next watchdog timeout */ netif_trans_update(lif->netdev); - err = ionic_wait_for_bit(lif, IONIC_LIF_QUEUE_RESET); + err = ionic_wait_for_bit(lif, IONIC_LIF_F_QUEUE_RESET); if (err) return err; @@ -1881,7 +1954,7 @@ int ionic_reset_queues(struct ionic_lif *lif) if (!err && running) ionic_open(lif->netdev); - clear_bit(IONIC_LIF_QUEUE_RESET, lif->state); + clear_bit(IONIC_LIF_F_QUEUE_RESET, lif->state); return err; } @@ -1910,6 +1983,8 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index ionic_ethtool_set_ops(netdev); netdev->watchdog_timeo = 2 * HZ; + netif_carrier_off(netdev); + netdev->min_mtu = IONIC_MIN_MTU; netdev->max_mtu = IONIC_MAX_MTU; @@ -1944,6 +2019,8 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index goto err_out_free_netdev; } + ionic_debugfs_add_lif(lif); + /* allocate queues */ err = ionic_qcqs_alloc(lif); if (err) @@ -2003,6 +2080,80 @@ static void ionic_lif_reset(struct ionic_lif *lif) mutex_unlock(&lif->ionic->dev_cmd_lock); } +static void ionic_lif_handle_fw_down(struct ionic_lif *lif) +{ + struct ionic *ionic = lif->ionic; + + if (test_and_set_bit(IONIC_LIF_F_FW_RESET, lif->state)) + return; + + dev_info(ionic->dev, "FW Down: Stopping LIFs\n"); + + netif_device_detach(lif->netdev); + + if (test_bit(IONIC_LIF_F_UP, lif->state)) { + dev_info(ionic->dev, "Surprise FW stop, stopping queues\n"); + ionic_stop_queues(lif); + } + + if (netif_running(lif->netdev)) { + ionic_txrx_deinit(lif); + ionic_txrx_free(lif); + } + ionic_lifs_deinit(ionic); + ionic_qcqs_free(lif); + + dev_info(ionic->dev, "FW Down: LIFs stopped\n"); +} + +static void ionic_lif_handle_fw_up(struct ionic_lif *lif) +{ + struct ionic *ionic = lif->ionic; + int err; + + if (!test_bit(IONIC_LIF_F_FW_RESET, lif->state)) + return; + + dev_info(ionic->dev, "FW Up: restarting LIFs\n"); + + err = ionic_qcqs_alloc(lif); + if (err) + goto err_out; + + err = ionic_lifs_init(ionic); + if (err) + goto err_qcqs_free; + + if (lif->registered) + ionic_lif_set_netdev_info(lif); + + if (netif_running(lif->netdev)) { + err = ionic_txrx_alloc(lif); + if (err) + goto err_lifs_deinit; + + err = ionic_txrx_init(lif); + if (err) + goto err_txrx_free; + } + + clear_bit(IONIC_LIF_F_FW_RESET, lif->state); + ionic_link_status_check_request(lif); + netif_device_attach(lif->netdev); + dev_info(ionic->dev, "FW Up: LIFs restarted\n"); + + return; + +err_txrx_free: + ionic_txrx_free(lif); +err_lifs_deinit: + ionic_lifs_deinit(ionic); +err_qcqs_free: + ionic_qcqs_free(lif); +err_out: + dev_err(ionic->dev, "FW Up: LIFs restart failed - err %d\n", err); +} + static void ionic_lif_free(struct ionic_lif *lif) { struct device *dev = lif->ionic->dev; @@ -2015,7 +2166,8 @@ static void ionic_lif_free(struct ionic_lif *lif) /* free queues */ ionic_qcqs_free(lif); - ionic_lif_reset(lif); + if (!test_bit(IONIC_LIF_F_FW_RESET, lif->state)) + ionic_lif_reset(lif); /* free lif info */ dma_free_coherent(dev, lif->info_sz, lif->info, lif->info_pa); @@ -2048,13 +2200,17 @@ void ionic_lifs_free(struct ionic *ionic) static void ionic_lif_deinit(struct ionic_lif *lif) { - if (!test_bit(IONIC_LIF_INITED, lif->state)) + if (!test_and_clear_bit(IONIC_LIF_F_INITED, lif->state)) return; - clear_bit(IONIC_LIF_INITED, lif->state); + if (!test_bit(IONIC_LIF_F_FW_RESET, lif->state)) { + cancel_work_sync(&lif->deferred.work); + cancel_work_sync(&lif->tx_timeout_work); + } ionic_rx_filters_deinit(lif); - ionic_lif_rss_deinit(lif); + if (lif->netdev->features & NETIF_F_RXHASH) + ionic_lif_rss_deinit(lif); napi_disable(&lif->adminqcq->napi); ionic_lif_qcq_deinit(lif, lif->notifyqcq); @@ -2107,13 +2263,6 @@ static int ionic_lif_adminq_init(struct ionic_lif *lif) netif_napi_add(lif->netdev, &qcq->napi, ionic_adminq_napi, NAPI_POLL_WEIGHT); - err = ionic_request_irq(lif, qcq); - if (err) { - netdev_warn(lif->netdev, "adminq irq request failed %d\n", err); - netif_napi_del(&qcq->napi); - return err; - } - napi_enable(&qcq->napi); if (qcq->flags & IONIC_QCQ_F_INTR) @@ -2122,8 +2271,6 @@ static int ionic_lif_adminq_init(struct ionic_lif *lif) qcq->flags |= IONIC_QCQ_F_INITED; - ionic_debugfs_add_qcq(lif, qcq); - return 0; } @@ -2159,6 +2306,7 @@ static int ionic_lif_notifyq_init(struct ionic_lif *lif) if (err) return err; + lif->last_eid = 0; q->hw_type = ctx.comp.q_init.hw_type; q->hw_index = le32_to_cpu(ctx.comp.q_init.hw_index); q->dbval = IONIC_DBELL_QID(q->hw_index); @@ -2171,8 +2319,6 @@ static int ionic_lif_notifyq_init(struct ionic_lif *lif) qcq->flags |= IONIC_QCQ_F_INITED; - ionic_debugfs_add_qcq(lif, qcq); - return 0; } @@ -2201,8 +2347,8 @@ static int ionic_station_set(struct ionic_lif *lif) addr.sa_family = AF_INET; err = eth_prepare_mac_addr_change(netdev, &addr); if (err) { - netdev_warn(lif->netdev, "ignoring bad MAC addr from NIC %pM\n", - addr.sa_data); + netdev_warn(lif->netdev, "ignoring bad MAC addr from NIC %pM - err %d\n", + addr.sa_data, err); return 0; } @@ -2226,8 +2372,6 @@ static int ionic_lif_init(struct ionic_lif *lif) int dbpage_num; int err; - ionic_debugfs_add_lif(lif); - mutex_lock(&lif->ionic->dev_cmd_lock); ionic_dev_cmd_lif_init(idev, lif->index, lif->info_pa); err = ionic_dev_cmd_wait(lif->ionic, DEVCMD_TIMEOUT); @@ -2287,7 +2431,7 @@ static int ionic_lif_init(struct ionic_lif *lif) lif->rx_copybreak = IONIC_RX_COPYBREAK_DEFAULT; - set_bit(IONIC_LIF_INITED, lif->state); + set_bit(IONIC_LIF_F_INITED, lif->state); INIT_WORK(&lif->tx_timeout_work, ionic_tx_timeout_work); @@ -2375,6 +2519,12 @@ int ionic_lifs_register(struct ionic *ionic) { int err; + /* the netdev is not registered on the management device, it is + * only used as a vehicle for napi operations on the adminq + */ + if (ionic->is_mgmt_nic) + return 0; + INIT_WORK(&ionic->nb_work, ionic_lif_notify_work); ionic->nb.notifier_call = ionic_lif_notify; @@ -2408,9 +2558,8 @@ void ionic_lifs_unregister(struct ionic *ionic) * current model, so don't bother searching the * ionic->lif for candidates to unregister */ - cancel_work_sync(&ionic->master_lif->deferred.work); - cancel_work_sync(&ionic->master_lif->tx_timeout_work); - if (ionic->master_lif->netdev->reg_state == NETREG_REGISTERED) + if (ionic->master_lif && + ionic->master_lif->netdev->reg_state == NETREG_REGISTERED) unregister_netdev(ionic->master_lif->netdev); } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h index 9c5a7dd45f9d..5d4ffda5c05f 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h @@ -98,6 +98,7 @@ struct ionic_deferred_work { union { unsigned int rx_mode; u8 addr[ETH_ALEN]; + u8 fw_status; }; }; @@ -121,14 +122,15 @@ struct ionic_lif_sw_stats { }; enum ionic_lif_state_flags { - IONIC_LIF_INITED, - IONIC_LIF_SW_DEBUG_STATS, - IONIC_LIF_UP, - IONIC_LIF_LINK_CHECK_REQUESTED, - IONIC_LIF_QUEUE_RESET, + IONIC_LIF_F_INITED, + IONIC_LIF_F_SW_DEBUG_STATS, + IONIC_LIF_F_UP, + IONIC_LIF_F_LINK_CHECK_REQUESTED, + IONIC_LIF_F_QUEUE_RESET, + IONIC_LIF_F_FW_RESET, /* leave this as last */ - IONIC_LIF_STATE_SIZE + IONIC_LIF_F_STATE_SIZE }; #define IONIC_LIF_NAME_MAX_SZ 32 @@ -136,7 +138,7 @@ struct ionic_lif { char name[IONIC_LIF_NAME_MAX_SZ]; struct list_head list; struct net_device *netdev; - DECLARE_BITMAP(state, IONIC_LIF_STATE_SIZE); + DECLARE_BITMAP(state, IONIC_LIF_F_STATE_SIZE); struct ionic *ionic; bool registered; unsigned int index; @@ -179,7 +181,6 @@ struct ionic_lif { u32 rx_coalesce_usecs; /* what the user asked for */ u32 rx_coalesce_hw; /* what the hw is using */ - u32 flags; struct work_struct tx_timeout_work; }; @@ -225,6 +226,9 @@ static inline u32 ionic_coal_hw_to_usec(struct ionic *ionic, u32 units) return (units * div) / mult; } +void ionic_link_status_check_request(struct ionic_lif *lif); +void ionic_lif_deferred_enqueue(struct ionic_deferred *def, + struct ionic_deferred_work *work); int ionic_lifs_alloc(struct ionic *ionic); void ionic_lifs_free(struct ionic *ionic); void ionic_lifs_deinit(struct ionic *ionic); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_main.c b/drivers/net/ethernet/pensando/ionic/ionic_main.c index a8e3fb73b465..588c62e9add7 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_main.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_main.c @@ -6,6 +6,7 @@ #include <linux/module.h> #include <linux/netdevice.h> #include <linux/utsname.h> +#include <linux/vermagic.h> #include "ionic.h" #include "ionic_bus.h" @@ -15,7 +16,6 @@ MODULE_DESCRIPTION(IONIC_DRV_DESCRIPTION); MODULE_AUTHOR("Pensando Systems, Inc"); MODULE_LICENSE("GPL"); -MODULE_VERSION(IONIC_DRV_VERSION); static const char *ionic_error_to_str(enum ionic_status_code code) { @@ -58,6 +58,8 @@ static const char *ionic_error_to_str(enum ionic_status_code code) return "IONIC_RC_BAD_ADDR"; case IONIC_RC_DEV_CMD: return "IONIC_RC_DEV_CMD"; + case IONIC_RC_ENOSUPP: + return "IONIC_RC_ENOSUPP"; case IONIC_RC_ERROR: return "IONIC_RC_ERROR"; case IONIC_RC_ERDMA: @@ -76,6 +78,7 @@ static int ionic_error_to_errno(enum ionic_status_code code) case IONIC_RC_EQTYPE: case IONIC_RC_EQID: case IONIC_RC_EINVAL: + case IONIC_RC_ENOSUPP: return -EINVAL; case IONIC_RC_EPERM: return -EPERM; @@ -240,11 +243,16 @@ static void ionic_adminq_cb(struct ionic_queue *q, static int ionic_adminq_post(struct ionic_lif *lif, struct ionic_admin_ctx *ctx) { - struct ionic_queue *adminq = &lif->adminqcq->q; + struct ionic_queue *adminq; int err = 0; WARN_ON(in_interrupt()); + if (!lif->adminqcq) + return -EIO; + + adminq = &lif->adminqcq->q; + spin_lock(&lif->adminq_lock); if (!ionic_q_has_space(adminq, 1)) { err = -ENOSPC; @@ -278,9 +286,11 @@ int ionic_adminq_post_wait(struct ionic_lif *lif, struct ionic_admin_ctx *ctx) err = ionic_adminq_post(lif, ctx); if (err) { - name = ionic_opcode_to_str(ctx->cmd.cmd.opcode); - netdev_err(netdev, "Posting of %s (%d) failed: %d\n", - name, ctx->cmd.cmd.opcode, err); + if (!test_bit(IONIC_LIF_F_FW_RESET, lif->state)) { + name = ionic_opcode_to_str(ctx->cmd.cmd.opcode); + netdev_err(netdev, "Posting of %s (%d) failed: %d\n", + name, ctx->cmd.cmd.opcode, err); + } return err; } @@ -357,7 +367,10 @@ try_again: done, duration / HZ, duration); if (!done && hb) { - ionic_dev_cmd_clean(ionic); + /* It is possible (but unlikely) that FW was busy and missed a + * heartbeat check but is still alive and will process this + * request, so don't clean the dev_cmd in this case. + */ dev_warn(ionic->dev, "DEVCMD %s (%d) failed - FW halted\n", ionic_opcode_to_str(opcode), opcode); return -ENXIO; @@ -414,7 +427,7 @@ int ionic_identify(struct ionic *ionic) memset(ident, 0, sizeof(*ident)); ident->drv.os_type = cpu_to_le32(IONIC_OS_TYPE_LINUX); - strncpy(ident->drv.driver_ver_str, IONIC_DRV_VERSION, + strncpy(ident->drv.driver_ver_str, UTS_RELEASE, sizeof(ident->drv.driver_ver_str) - 1); mutex_lock(&ionic->dev_cmd_lock); @@ -558,8 +571,6 @@ int ionic_port_reset(struct ionic *ionic) static int __init ionic_init_module(void) { - pr_info("%s %s, ver %s\n", - IONIC_DRV_NAME, IONIC_DRV_DESCRIPTION, IONIC_DRV_VERSION); ionic_debugfs_create(); return ionic_bus_register_driver(); } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_stats.c b/drivers/net/ethernet/pensando/ionic/ionic_stats.c index a1e9796a660a..8f2a8fb029f1 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_stats.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_stats.c @@ -118,8 +118,8 @@ static u64 ionic_sw_stats_get_count(struct ionic_lif *lif) /* rx stats */ total += MAX_Q(lif) * IONIC_NUM_RX_STATS; - if (test_bit(IONIC_LIF_UP, lif->state) && - test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_F_UP, lif->state) && + test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) { /* tx debug stats */ total += MAX_Q(lif) * (IONIC_NUM_DBG_CQ_STATS + IONIC_NUM_TX_Q_STATS + @@ -151,8 +151,8 @@ static void ionic_sw_stats_get_strings(struct ionic_lif *lif, u8 **buf) *buf += ETH_GSTRING_LEN; } - if (test_bit(IONIC_LIF_UP, lif->state) && - test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_F_UP, lif->state) && + test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) { for (i = 0; i < IONIC_NUM_TX_Q_STATS; i++) { snprintf(*buf, ETH_GSTRING_LEN, "txq_%d_%s", @@ -190,8 +190,8 @@ static void ionic_sw_stats_get_strings(struct ionic_lif *lif, u8 **buf) *buf += ETH_GSTRING_LEN; } - if (test_bit(IONIC_LIF_UP, lif->state) && - test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_F_UP, lif->state) && + test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) { for (i = 0; i < IONIC_NUM_DBG_CQ_STATS; i++) { snprintf(*buf, ETH_GSTRING_LEN, "rxq_%d_cq_%s", @@ -247,8 +247,8 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) (*buf)++; } - if (test_bit(IONIC_LIF_UP, lif->state) && - test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_F_UP, lif->state) && + test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) { txqcq = lif_to_txqcq(lif, q_num); for (i = 0; i < IONIC_NUM_TX_Q_STATS; i++) { **buf = IONIC_READ_STAT64(&txqcq->q, @@ -281,8 +281,8 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) (*buf)++; } - if (test_bit(IONIC_LIF_UP, lif->state) && - test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_F_UP, lif->state) && + test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) { rxqcq = lif_to_rxqcq(lif, q_num); for (i = 0; i < IONIC_NUM_DBG_CQ_STATS; i++) { **buf = IONIC_READ_STAT64(&rxqcq->cq, diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c index e452f4242ba0..d233b6e77b1e 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c @@ -158,7 +158,7 @@ static void ionic_rx_clean(struct ionic_queue *q, struct ionic_desc_info *desc_i } /* no packet processing while resetting */ - if (unlikely(test_bit(IONIC_LIF_QUEUE_RESET, q->lif->state))) { + if (unlikely(test_bit(IONIC_LIF_F_QUEUE_RESET, q->lif->state))) { stats->dropped++; return; } @@ -593,6 +593,22 @@ void ionic_tx_flush(struct ionic_cq *cq) work_done, 0); } +void ionic_tx_empty(struct ionic_queue *q) +{ + struct ionic_desc_info *desc_info; + int done = 0; + + /* walk the not completed tx entries, if any */ + while (q->head != q->tail) { + desc_info = q->tail; + q->tail = desc_info->next; + ionic_tx_clean(q, desc_info, NULL, desc_info->cb_arg); + desc_info->cb = NULL; + desc_info->cb_arg = NULL; + done++; + } +} + static int ionic_tx_tcp_inner_pseudo_csum(struct sk_buff *skb) { int err; @@ -632,10 +648,7 @@ static int ionic_tx_tcp_pseudo_csum(struct sk_buff *skb) ip_hdr(skb)->daddr, 0, IPPROTO_TCP, 0); } else if (skb->protocol == cpu_to_be16(ETH_P_IPV6)) { - tcp_hdr(skb)->check = - ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - 0, IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); } return 0; @@ -1026,7 +1039,7 @@ netdev_tx_t ionic_start_xmit(struct sk_buff *skb, struct net_device *netdev) int ndescs; int err; - if (unlikely(!test_bit(IONIC_LIF_UP, lif->state))) { + if (unlikely(!test_bit(IONIC_LIF_F_UP, lif->state))) { dev_kfree_skb(skb); return NETDEV_TX_OK; } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.h b/drivers/net/ethernet/pensando/ionic/ionic_txrx.h index 53775c62c85a..71973e3c35a6 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.h @@ -9,6 +9,7 @@ void ionic_tx_flush(struct ionic_cq *cq); void ionic_rx_fill(struct ionic_queue *q); void ionic_rx_empty(struct ionic_queue *q); +void ionic_tx_empty(struct ionic_queue *q); int ionic_rx_napi(struct napi_struct *napi, int budget); netdev_tx_t ionic_start_xmit(struct sk_buff *skb, struct net_device *netdev); diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic.h b/drivers/net/ethernet/qlogic/netxen/netxen_nic.h index 3dce769d83a1..86153660d245 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic.h +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic.h @@ -1316,7 +1316,7 @@ struct netxen_minidump_template_hdr { u32 driver_info_word4; u32 saved_state_array[NX_DUMP_STATE_ARRAY_LEN]; u32 capture_size_array[NX_DUMP_CAP_SIZE_ARRAY_LEN]; - u32 rsvd[0]; + u32 rsvd[]; }; /* Common Entry Header: Common to All Entry Types */ diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c index 6a2d91d58968..66f45fce90fa 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c @@ -748,24 +748,7 @@ static int netxen_set_intr_coalesce(struct net_device *netdev, if (ethcoal->rx_coalesce_usecs > 0xffff || ethcoal->rx_max_coalesced_frames > 0xffff || ethcoal->tx_coalesce_usecs > 0xffff || - ethcoal->tx_max_coalesced_frames > 0xffff || - ethcoal->rx_coalesce_usecs_irq || - ethcoal->rx_max_coalesced_frames_irq || - ethcoal->tx_coalesce_usecs_irq || - ethcoal->tx_max_coalesced_frames_irq || - ethcoal->stats_block_coalesce_usecs || - ethcoal->use_adaptive_rx_coalesce || - ethcoal->use_adaptive_tx_coalesce || - ethcoal->pkt_rate_low || - ethcoal->rx_coalesce_usecs_low || - ethcoal->rx_max_coalesced_frames_low || - ethcoal->tx_coalesce_usecs_low || - ethcoal->tx_max_coalesced_frames_low || - ethcoal->pkt_rate_high || - ethcoal->rx_coalesce_usecs_high || - ethcoal->rx_max_coalesced_frames_high || - ethcoal->tx_coalesce_usecs_high || - ethcoal->tx_max_coalesced_frames_high) + ethcoal->tx_max_coalesced_frames > 0xffff) return -EINVAL; if (!ethcoal->rx_coalesce_usecs || @@ -923,6 +906,8 @@ netxen_get_dump_data(struct net_device *netdev, struct ethtool_dump *dump, } const struct ethtool_ops netxen_nic_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = netxen_nic_get_drvinfo, .get_regs_len = netxen_nic_get_regs_len, .get_regs = netxen_nic_get_regs, diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 03bdd2e26329..38a65b984e47 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -4691,26 +4691,20 @@ static void qed_chain_free_single(struct qed_dev *cdev, static void qed_chain_free_pbl(struct qed_dev *cdev, struct qed_chain *p_chain) { - void **pp_virt_addr_tbl = p_chain->pbl.pp_virt_addr_tbl; + struct addr_tbl_entry *pp_addr_tbl = p_chain->pbl.pp_addr_tbl; u32 page_cnt = p_chain->page_cnt, i, pbl_size; - u8 *p_pbl_virt = p_chain->pbl_sp.p_virt_table; - if (!pp_virt_addr_tbl) + if (!pp_addr_tbl) return; - if (!p_pbl_virt) - goto out; - for (i = 0; i < page_cnt; i++) { - if (!pp_virt_addr_tbl[i]) + if (!pp_addr_tbl[i].virt_addr || !pp_addr_tbl[i].dma_map) break; dma_free_coherent(&cdev->pdev->dev, QED_CHAIN_PAGE_SIZE, - pp_virt_addr_tbl[i], - *(dma_addr_t *)p_pbl_virt); - - p_pbl_virt += QED_CHAIN_PBL_ENTRY_SIZE; + pp_addr_tbl[i].virt_addr, + pp_addr_tbl[i].dma_map); } pbl_size = page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; @@ -4720,9 +4714,9 @@ static void qed_chain_free_pbl(struct qed_dev *cdev, struct qed_chain *p_chain) pbl_size, p_chain->pbl_sp.p_virt_table, p_chain->pbl_sp.p_phys_table); -out: - vfree(p_chain->pbl.pp_virt_addr_tbl); - p_chain->pbl.pp_virt_addr_tbl = NULL; + + vfree(p_chain->pbl.pp_addr_tbl); + p_chain->pbl.pp_addr_tbl = NULL; } void qed_chain_free(struct qed_dev *cdev, struct qed_chain *p_chain) @@ -4823,19 +4817,19 @@ qed_chain_alloc_pbl(struct qed_dev *cdev, { u32 page_cnt = p_chain->page_cnt, size, i; dma_addr_t p_phys = 0, p_pbl_phys = 0; - void **pp_virt_addr_tbl = NULL; + struct addr_tbl_entry *pp_addr_tbl; u8 *p_pbl_virt = NULL; void *p_virt = NULL; - size = page_cnt * sizeof(*pp_virt_addr_tbl); - pp_virt_addr_tbl = vzalloc(size); - if (!pp_virt_addr_tbl) + size = page_cnt * sizeof(*pp_addr_tbl); + pp_addr_tbl = vzalloc(size); + if (!pp_addr_tbl) return -ENOMEM; /* The allocation of the PBL table is done with its full size, since it * is expected to be successive. * qed_chain_init_pbl_mem() is called even in a case of an allocation - * failure, since pp_virt_addr_tbl was previously allocated, and it + * failure, since tbl was previously allocated, and it * should be saved to allow its freeing during the error flow. */ size = page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; @@ -4849,8 +4843,7 @@ qed_chain_alloc_pbl(struct qed_dev *cdev, p_chain->b_external_pbl = true; } - qed_chain_init_pbl_mem(p_chain, p_pbl_virt, p_pbl_phys, - pp_virt_addr_tbl); + qed_chain_init_pbl_mem(p_chain, p_pbl_virt, p_pbl_phys, pp_addr_tbl); if (!p_pbl_virt) return -ENOMEM; @@ -4869,7 +4862,8 @@ qed_chain_alloc_pbl(struct qed_dev *cdev, /* Fill the PBL table with the physical address of the page */ *(dma_addr_t *)p_pbl_virt = p_phys; /* Keep the virtual address of the page */ - p_chain->pbl.pp_virt_addr_tbl[i] = p_virt; + p_chain->pbl.pp_addr_tbl[i].virt_addr = p_virt; + p_chain->pbl.pp_addr_tbl[i].dma_map = p_phys; p_pbl_virt += QED_CHAIN_PBL_ENTRY_SIZE; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 2c189c637cca..96356e897c80 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -1087,9 +1087,6 @@ static void qed_update_pf_params(struct qed_dev *cdev, #define QED_PERIODIC_DB_REC_INTERVAL_MS 100 #define QED_PERIODIC_DB_REC_INTERVAL \ msecs_to_jiffies(QED_PERIODIC_DB_REC_INTERVAL_MS) -#define QED_PERIODIC_DB_REC_WAIT_COUNT 10 -#define QED_PERIODIC_DB_REC_WAIT_INTERVAL \ - (QED_PERIODIC_DB_REC_INTERVAL_MS / QED_PERIODIC_DB_REC_WAIT_COUNT) static int qed_slowpath_delayed_work(struct qed_hwfn *hwfn, enum qed_slowpath_wq_flag wq_flag, @@ -1123,7 +1120,7 @@ void qed_periodic_db_rec_start(struct qed_hwfn *p_hwfn) static void qed_slowpath_wq_stop(struct qed_dev *cdev) { - int i, sleep_count = QED_PERIODIC_DB_REC_WAIT_COUNT; + int i; if (IS_VF(cdev)) return; @@ -1135,13 +1132,7 @@ static void qed_slowpath_wq_stop(struct qed_dev *cdev) /* Stop queuing new delayed works */ cdev->hwfns[i].slowpath_wq_active = false; - /* Wait until the last periodic doorbell recovery is executed */ - while (test_bit(QED_SLOWPATH_PERIODIC_DB_REC, - &cdev->hwfns[i].slowpath_task_flags) && - sleep_count--) - msleep(QED_PERIODIC_DB_REC_WAIT_INTERVAL); - - flush_workqueue(cdev->hwfns[i].slowpath_wq); + cancel_delayed_work(&cdev->hwfns[i].slowpath_task); destroy_workqueue(cdev->hwfns[i].slowpath_wq); } } diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index 8a426afb6a55..812c7766e096 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -1566,7 +1566,7 @@ static int qede_selftest_transmit_traffic(struct qede_dev *edev, static int qede_selftest_receive_traffic(struct qede_dev *edev) { - u16 hw_comp_cons, sw_comp_cons, sw_rx_index, len; + u16 sw_rx_index, len; struct eth_fast_path_rx_reg_cqe *fp_cqe; struct qede_rx_queue *rxq = NULL; struct sw_rx_data *sw_rx_data; @@ -1596,17 +1596,6 @@ static int qede_selftest_receive_traffic(struct qede_dev *edev) continue; } - hw_comp_cons = le16_to_cpu(*rxq->hw_cons_ptr); - sw_comp_cons = qed_chain_get_cons_idx(&rxq->rx_comp_ring); - - /* Memory barrier to prevent the CPU from doing speculative - * reads of CQE/BD before reading hw_comp_cons. If the CQE is - * read before it is written by FW, then FW writes CQE and SB, - * and then the CPU reads the hw_comp_cons, it will use an old - * CQE. - */ - rmb(); - /* Get the CQE from the completion ring */ cqe = (union eth_rx_cqe *)qed_chain_consume(&rxq->rx_comp_ring); @@ -2087,6 +2076,7 @@ err: } static const struct ethtool_ops qede_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_link_ksettings = qede_get_link_ksettings, .set_link_ksettings = qede_set_link_ksettings, .get_drvinfo = qede_get_drvinfo, @@ -2133,6 +2123,7 @@ static const struct ethtool_ops qede_ethtool_ops = { }; static const struct ethtool_ops qede_vf_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS, .get_link_ksettings = qede_get_link_ksettings, .get_drvinfo = qede_get_drvinfo, .get_msglevel = qede_get_msglevel, diff --git a/drivers/net/ethernet/qlogic/qede/qede_filter.c b/drivers/net/ethernet/qlogic/qede/qede_filter.c index d1ce4531d01a..fe72bb6c9455 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_filter.c +++ b/drivers/net/ethernet/qlogic/qede/qede_filter.c @@ -1746,7 +1746,8 @@ unlock: } static int qede_parse_actions(struct qede_dev *edev, - struct flow_action *flow_action) + struct flow_action *flow_action, + struct netlink_ext_ack *extack) { const struct flow_action_entry *act; int i; @@ -1756,6 +1757,9 @@ static int qede_parse_actions(struct qede_dev *edev, return -EINVAL; } + if (!flow_action_basic_hw_stats_check(flow_action, extack)) + return -EOPNOTSUPP; + flow_action_for_each(i, act, flow_action) { switch (act->id) { case FLOW_ACTION_DROP: @@ -1970,7 +1974,7 @@ int qede_add_tc_flower_fltr(struct qede_dev *edev, __be16 proto, } /* parse tc actions and get the vf_id */ - if (qede_parse_actions(edev, &f->rule->action)) + if (qede_parse_actions(edev, &f->rule->action, f->common.extack)) goto unlock; if (qede_flow_find_fltr(edev, &t)) { @@ -2038,7 +2042,7 @@ static int qede_flow_spec_validate(struct qede_dev *edev, return -EINVAL; } - if (qede_parse_actions(edev, flow_action)) + if (qede_parse_actions(edev, flow_action, NULL)) return -EINVAL; return 0; diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h index 374a4d4371f9..134611aa2c9a 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h @@ -418,7 +418,7 @@ struct qlcnic_83xx_dump_template_hdr { u32 saved_state[16]; u32 cap_sizes[8]; u32 ocm_wnd_reg[16]; - u32 rsvd[0]; + u32 rsvd[]; }; struct qlcnic_82xx_dump_template_hdr { @@ -436,7 +436,7 @@ struct qlcnic_82xx_dump_template_hdr { u32 cap_sizes[8]; u32 rsvd[7]; u32 capabilities; - u32 rsvd1[0]; + u32 rsvd1[]; }; #define QLC_PEX_DMA_READ_SIZE (PAGE_SIZE * 16) @@ -740,7 +740,7 @@ struct qlcnic_hostrq_rx_ctx { The following is packed: - N hostrq_rds_rings - N hostrq_sds_rings */ - char data[0]; + char data[]; } __packed; struct qlcnic_cardrsp_rds_ring{ @@ -769,7 +769,7 @@ struct qlcnic_cardrsp_rx_ctx { The following is packed: - N cardrsp_rds_rings - N cardrs_sds_rings */ - char data[0]; + char data[]; } __packed; #define SIZEOF_HOSTRQ_RX(HOSTRQ_RX, rds_rings, sds_rings) \ diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c index 75d83c3cbf27..5c2a3acf1e89 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c @@ -1542,24 +1542,7 @@ static int qlcnic_set_intr_coalesce(struct net_device *netdev, if (ethcoal->rx_coalesce_usecs > 0xffff || ethcoal->rx_max_coalesced_frames > 0xffff || ethcoal->tx_coalesce_usecs > 0xffff || - ethcoal->tx_max_coalesced_frames > 0xffff || - ethcoal->rx_coalesce_usecs_irq || - ethcoal->rx_max_coalesced_frames_irq || - ethcoal->tx_coalesce_usecs_irq || - ethcoal->tx_max_coalesced_frames_irq || - ethcoal->stats_block_coalesce_usecs || - ethcoal->use_adaptive_rx_coalesce || - ethcoal->use_adaptive_tx_coalesce || - ethcoal->pkt_rate_low || - ethcoal->rx_coalesce_usecs_low || - ethcoal->rx_max_coalesced_frames_low || - ethcoal->tx_coalesce_usecs_low || - ethcoal->tx_max_coalesced_frames_low || - ethcoal->pkt_rate_high || - ethcoal->rx_coalesce_usecs_high || - ethcoal->rx_max_coalesced_frames_high || - ethcoal->tx_coalesce_usecs_high || - ethcoal->tx_max_coalesced_frames_high) + ethcoal->tx_max_coalesced_frames > 0xffff) return -EINVAL; err = qlcnic_config_intr_coalesce(adapter, ethcoal); @@ -1834,6 +1817,8 @@ qlcnic_set_dump(struct net_device *netdev, struct ethtool_dump *val) } const struct ethtool_ops qlcnic_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = qlcnic_get_drvinfo, .get_regs_len = qlcnic_get_regs_len, .get_regs = qlcnic_get_regs, @@ -1865,6 +1850,8 @@ const struct ethtool_ops qlcnic_ethtool_ops = { }; const struct ethtool_ops qlcnic_sriov_vf_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = qlcnic_get_drvinfo, .get_regs_len = qlcnic_get_regs_len, .get_regs = qlcnic_get_regs, diff --git a/drivers/net/ethernet/qualcomm/emac/emac-mac.c b/drivers/net/ethernet/qualcomm/emac/emac-mac.c index bebe38d74d66..251d4ac4af02 100644 --- a/drivers/net/ethernet/qualcomm/emac/emac-mac.c +++ b/drivers/net/ethernet/qualcomm/emac/emac-mac.c @@ -1288,11 +1288,8 @@ static int emac_tso_csum(struct emac_adapter *adpt, memset(tpd, 0, sizeof(*tpd)); memset(&extra_tpd, 0, sizeof(extra_tpd)); - ipv6_hdr(skb)->payload_len = 0; - tcp_hdr(skb)->check = - ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - 0, IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); + TPD_PKT_LEN_SET(&extra_tpd, skb->len); TPD_LSO_SET(&extra_tpd, 1); TPD_LSOV_SET(&extra_tpd, 1); diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c index fbf4cbcf1a65..1305522f72d6 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c @@ -57,7 +57,7 @@ static int rmnet_register_real_device(struct net_device *real_dev) if (rmnet_is_real_dev_registered(real_dev)) return 0; - port = kzalloc(sizeof(*port), GFP_ATOMIC); + port = kzalloc(sizeof(*port), GFP_KERNEL); if (!port) return -ENOMEM; @@ -122,13 +122,12 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev, } real_dev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK])); - if (!real_dev || !dev) + if (!real_dev) { + NL_SET_ERR_MSG_MOD(extack, "link does not exist"); return -ENODEV; + } - if (!data[IFLA_RMNET_MUX_ID]) - return -EINVAL; - - ep = kzalloc(sizeof(*ep), GFP_ATOMIC); + ep = kzalloc(sizeof(*ep), GFP_KERNEL); if (!ep) return -ENOMEM; @@ -139,7 +138,7 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev, goto err0; port = rmnet_get_port_rtnl(real_dev); - err = rmnet_vnd_newlink(mux_id, dev, port, real_dev, ep); + err = rmnet_vnd_newlink(mux_id, dev, port, real_dev, ep, extack); if (err) goto err1; @@ -263,12 +262,16 @@ static int rmnet_rtnl_validate(struct nlattr *tb[], struct nlattr *data[], { u16 mux_id; - if (!data || !data[IFLA_RMNET_MUX_ID]) + if (!data || !data[IFLA_RMNET_MUX_ID]) { + NL_SET_ERR_MSG_MOD(extack, "MUX ID not specified"); return -EINVAL; + } mux_id = nla_get_u16(data[IFLA_RMNET_MUX_ID]); - if (mux_id > (RMNET_MAX_LOGICAL_EP - 1)) + if (mux_id > (RMNET_MAX_LOGICAL_EP - 1)) { + NL_SET_ERR_MSG_MOD(extack, "invalid MUX ID"); return -ERANGE; + } return 0; } @@ -406,14 +409,22 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, /* If there is more than one rmnet dev attached, its probably being * used for muxing. Skip the briding in that case */ - if (port->nr_rmnet_devs > 1) + if (port->nr_rmnet_devs > 1) { + NL_SET_ERR_MSG_MOD(extack, "more than one rmnet dev attached"); return -EINVAL; + } - if (port->rmnet_mode != RMNET_EPMODE_VND) + if (port->rmnet_mode != RMNET_EPMODE_VND) { + NL_SET_ERR_MSG_MOD(extack, "bridge device already exists"); return -EINVAL; + } + + if (rmnet_is_real_dev_registered(slave_dev)) { + NL_SET_ERR_MSG_MOD(extack, + "slave cannot be another rmnet dev"); - if (rmnet_is_real_dev_registered(slave_dev)) return -EBUSY; + } err = rmnet_register_real_device(slave_dev); if (err) @@ -475,4 +486,5 @@ static void __exit rmnet_exit(void) module_init(rmnet_init) module_exit(rmnet_exit) +MODULE_ALIAS_RTNL_LINK("rmnet"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c index 26ad40f19c64..d58b51d277f1 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c @@ -212,6 +212,8 @@ void rmnet_vnd_setup(struct net_device *rmnet_dev) rmnet_dev->needs_free_netdev = true; rmnet_dev->ethtool_ops = &rmnet_ethtool_ops; + rmnet_dev->features |= NETIF_F_LLTX; + /* This perm addr will be used as interface identifier by IPv6 */ rmnet_dev->addr_assign_type = NET_ADDR_RANDOM; eth_random_addr(rmnet_dev->perm_addr); @@ -222,16 +224,17 @@ void rmnet_vnd_setup(struct net_device *rmnet_dev) int rmnet_vnd_newlink(u8 id, struct net_device *rmnet_dev, struct rmnet_port *port, struct net_device *real_dev, - struct rmnet_endpoint *ep) + struct rmnet_endpoint *ep, + struct netlink_ext_ack *extack) + { struct rmnet_priv *priv = netdev_priv(rmnet_dev); int rc; - if (ep->egress_dev) - return -EINVAL; - - if (rmnet_get_endpoint(port, id)) + if (rmnet_get_endpoint(port, id)) { + NL_SET_ERR_MSG_MOD(extack, "MUX ID already exists"); return -EBUSY; + } rmnet_dev->hw_features = NETIF_F_RXCSUM; rmnet_dev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM; diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h index 14d77c709d4a..4967f3461ed1 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h @@ -11,7 +11,8 @@ int rmnet_vnd_do_flow_control(struct net_device *dev, int enable); int rmnet_vnd_newlink(u8 id, struct net_device *rmnet_dev, struct rmnet_port *port, struct net_device *real_dev, - struct rmnet_endpoint *ep); + struct rmnet_endpoint *ep, + struct netlink_ext_ack *extack); int rmnet_vnd_dellink(u8 id, struct rmnet_port *port, struct rmnet_endpoint *ep); void rmnet_vnd_rx_fixup(struct sk_buff *skb, struct net_device *dev); diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 791d99b9e1cf..55cb5730beb6 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -212,7 +212,6 @@ enum rtl_registers { /* Unlimited maximum PCI burst. */ #define RX_DMA_BURST (7 << RXCFG_DMA_SHIFT) - RxMissed = 0x4c, Cfg9346 = 0x50, Config0 = 0x51, Config1 = 0x52, @@ -576,6 +575,7 @@ struct rtl8169_tc_offsets { __le64 tx_errors; __le32 tx_multi_collision; __le16 tx_aborted; + __le16 rx_missed; }; enum rtl_flag { @@ -689,6 +689,12 @@ static void rtl_unlock_config_regs(struct rtl8169_private *tp) RTL_W8(tp, Cfg9346, Cfg9346_Unlock); } +static void rtl_pci_commit(struct rtl8169_private *tp) +{ + /* Read an arbitrary register to commit a preceding PCI write */ + RTL_R8(tp, ChipCmd); +} + static bool rtl_is_8125(struct rtl8169_private *tp) { return tp->mac_version >= RTL_GIGA_MAC_VER_60; @@ -1302,10 +1308,6 @@ static void rtl_irq_disable(struct rtl8169_private *tp) tp->irq_enabled = 0; } -#define RTL_EVENT_NAPI_RX (RxOK | RxErr) -#define RTL_EVENT_NAPI_TX (TxOK | TxErr) -#define RTL_EVENT_NAPI (RTL_EVENT_NAPI_RX | RTL_EVENT_NAPI_TX) - static void rtl_irq_enable(struct rtl8169_private *tp) { tp->irq_enabled = 1; @@ -1319,18 +1321,13 @@ static void rtl8169_irq_mask_and_ack(struct rtl8169_private *tp) { rtl_irq_disable(tp); rtl_ack_events(tp, 0xffffffff); - /* PCI commit */ - RTL_R8(tp, ChipCmd); + rtl_pci_commit(tp); } static void rtl_link_chg_patch(struct rtl8169_private *tp) { - struct net_device *dev = tp->dev; struct phy_device *phydev = tp->phydev; - if (!netif_running(dev)) - return; - if (tp->mac_version == RTL_GIGA_MAC_VER_34 || tp->mac_version == RTL_GIGA_MAC_VER_38) { if (phydev->speed == SPEED_1000) { @@ -1536,7 +1533,7 @@ static int rtl8169_set_features(struct net_device *dev, } RTL_W16(tp, CPlusCmd, tp->cp_cmd); - RTL_R16(tp, CPlusCmd); + rtl_pci_commit(tp); rtl_unlock_work(tp); @@ -1622,7 +1619,7 @@ static bool rtl8169_do_counters(struct rtl8169_private *tp, u32 counter_cmd) u32 cmd; RTL_W32(tp, CounterAddrHigh, (u64)paddr >> 32); - RTL_R32(tp, CounterAddrHigh); + rtl_pci_commit(tp); cmd = (u64)paddr & DMA_BIT_MASK(32); RTL_W32(tp, CounterAddrLow, cmd); RTL_W32(tp, CounterAddrLow, cmd | counter_cmd); @@ -1689,6 +1686,7 @@ static bool rtl8169_init_counter_offsets(struct rtl8169_private *tp) tp->tc_offset.tx_errors = counters->tx_errors; tp->tc_offset.tx_multi_collision = counters->tx_multi_collision; tp->tc_offset.tx_aborted = counters->tx_aborted; + tp->tc_offset.rx_missed = counters->rx_missed; tp->tc_offset.inited = true; return ret; @@ -1946,7 +1944,7 @@ static int rtl_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec) tp->cp_cmd = (tp->cp_cmd & ~INTT_MASK) | cp01; RTL_W16(tp, CPlusCmd, tp->cp_cmd); - RTL_R16(tp, CPlusCmd); + rtl_pci_commit(tp); rtl_unlock_work(tp); @@ -2008,6 +2006,8 @@ out: } static const struct ethtool_ops rtl8169_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = rtl8169_get_drvinfo, .get_regs_len = rtl8169_get_regs_len, .get_link = ethtool_op_get_link, @@ -2044,7 +2044,7 @@ static void rtl_enable_eee(struct rtl8169_private *tp) phy_write_mmd(phydev, MDIO_MMD_AN, MDIO_AN_EEE_ADV, adv); } -static void rtl8169_get_mac_version(struct rtl8169_private *tp) +static enum mac_version rtl8169_get_mac_version(u16 xid, bool gmii) { /* * The driver currently handles the 8168Bf and the 8168Be identically @@ -2060,7 +2060,7 @@ static void rtl8169_get_mac_version(struct rtl8169_private *tp) static const struct rtl_mac_info { u16 mask; u16 val; - u16 mac_version; + enum mac_version ver; } mac_info[] = { /* 8125 family. */ { 0x7cf, 0x608, RTL_GIGA_MAC_VER_60 }, @@ -2147,22 +2147,22 @@ static void rtl8169_get_mac_version(struct rtl8169_private *tp) { 0x000, 0x000, RTL_GIGA_MAC_NONE } }; const struct rtl_mac_info *p = mac_info; - u16 reg = RTL_R32(tp, TxConfig) >> 20; + enum mac_version ver; - while ((reg & p->mask) != p->val) + while ((xid & p->mask) != p->val) p++; - tp->mac_version = p->mac_version; - - if (tp->mac_version == RTL_GIGA_MAC_NONE) { - dev_err(tp_to_dev(tp), "unknown chip XID %03x\n", reg & 0xfcf); - } else if (!tp->supports_gmii) { - if (tp->mac_version == RTL_GIGA_MAC_VER_42) - tp->mac_version = RTL_GIGA_MAC_VER_43; - else if (tp->mac_version == RTL_GIGA_MAC_VER_45) - tp->mac_version = RTL_GIGA_MAC_VER_47; - else if (tp->mac_version == RTL_GIGA_MAC_VER_46) - tp->mac_version = RTL_GIGA_MAC_VER_48; + ver = p->ver; + + if (ver != RTL_GIGA_MAC_NONE && !gmii) { + if (ver == RTL_GIGA_MAC_VER_42) + ver = RTL_GIGA_MAC_VER_43; + else if (ver == RTL_GIGA_MAC_VER_45) + ver = RTL_GIGA_MAC_VER_47; + else if (ver == RTL_GIGA_MAC_VER_46) + ver = RTL_GIGA_MAC_VER_48; } + + return ver; } static void rtl_release_firmware(struct rtl8169_private *tp) @@ -2228,8 +2228,8 @@ u16 rtl8168h_2_get_adc_bias_ioffset(struct rtl8169_private *tp) static void rtl_schedule_task(struct rtl8169_private *tp, enum rtl_flag flag) { - if (!test_and_set_bit(flag, tp->wk.flags)) - schedule_work(&tp->wk.work); + set_bit(flag, tp->wk.flags); + schedule_work(&tp->wk.work); } static void rtl8169_init_phy(struct rtl8169_private *tp) @@ -2264,10 +2264,10 @@ static void rtl_rar_set(struct rtl8169_private *tp, u8 *addr) rtl_unlock_config_regs(tp); RTL_W32(tp, MAC4, addr[4] | addr[5] << 8); - RTL_R32(tp, MAC4); + rtl_pci_commit(tp); RTL_W32(tp, MAC0, addr[0] | addr[1] << 8 | addr[2] << 16 | addr[3] << 24); - RTL_R32(tp, MAC0); + rtl_pci_commit(tp); if (tp->mac_version == RTL_GIGA_MAC_VER_34) rtl_rar_exgmac_set(tp, addr); @@ -2471,66 +2471,52 @@ static void r8168b_1_hw_jumbo_disable(struct rtl8169_private *tp) RTL_W8(tp, Config4, RTL_R8(tp, Config4) & ~(1 << 0)); } -static void rtl_hw_jumbo_enable(struct rtl8169_private *tp) +static void rtl_jumbo_config(struct rtl8169_private *tp) { - rtl_unlock_config_regs(tp); - switch (tp->mac_version) { - case RTL_GIGA_MAC_VER_12: - case RTL_GIGA_MAC_VER_17: - pcie_set_readrq(tp->pci_dev, 512); - r8168b_1_hw_jumbo_enable(tp); - break; - case RTL_GIGA_MAC_VER_18 ... RTL_GIGA_MAC_VER_26: - pcie_set_readrq(tp->pci_dev, 512); - r8168c_hw_jumbo_enable(tp); - break; - case RTL_GIGA_MAC_VER_27 ... RTL_GIGA_MAC_VER_28: - r8168dp_hw_jumbo_enable(tp); - break; - case RTL_GIGA_MAC_VER_31 ... RTL_GIGA_MAC_VER_33: - pcie_set_readrq(tp->pci_dev, 512); - r8168e_hw_jumbo_enable(tp); - break; - default: - break; - } - rtl_lock_config_regs(tp); -} + bool jumbo = tp->dev->mtu > ETH_DATA_LEN; -static void rtl_hw_jumbo_disable(struct rtl8169_private *tp) -{ rtl_unlock_config_regs(tp); switch (tp->mac_version) { case RTL_GIGA_MAC_VER_12: case RTL_GIGA_MAC_VER_17: - r8168b_1_hw_jumbo_disable(tp); + if (jumbo) { + pcie_set_readrq(tp->pci_dev, 512); + r8168b_1_hw_jumbo_enable(tp); + } else { + r8168b_1_hw_jumbo_disable(tp); + } break; case RTL_GIGA_MAC_VER_18 ... RTL_GIGA_MAC_VER_26: - r8168c_hw_jumbo_disable(tp); + if (jumbo) { + pcie_set_readrq(tp->pci_dev, 512); + r8168c_hw_jumbo_enable(tp); + } else { + r8168c_hw_jumbo_disable(tp); + } break; case RTL_GIGA_MAC_VER_27 ... RTL_GIGA_MAC_VER_28: - r8168dp_hw_jumbo_disable(tp); + if (jumbo) + r8168dp_hw_jumbo_enable(tp); + else + r8168dp_hw_jumbo_disable(tp); break; case RTL_GIGA_MAC_VER_31 ... RTL_GIGA_MAC_VER_33: - r8168e_hw_jumbo_disable(tp); + if (jumbo) { + pcie_set_readrq(tp->pci_dev, 512); + r8168e_hw_jumbo_enable(tp); + } else { + r8168e_hw_jumbo_disable(tp); + } break; default: break; } rtl_lock_config_regs(tp); - if (pci_is_pcie(tp->pci_dev) && tp->supports_gmii) + if (!jumbo && pci_is_pcie(tp->pci_dev) && tp->supports_gmii) pcie_set_readrq(tp->pci_dev, 4096); } -static void rtl_jumbo_config(struct rtl8169_private *tp, int mtu) -{ - if (mtu > ETH_DATA_LEN) - rtl_hw_jumbo_enable(tp); - else - rtl_hw_jumbo_disable(tp); -} - DECLARE_RTL_COND(rtl_chipcmd_cond) { return RTL_R8(tp, ChipCmd) & CmdReset; @@ -3838,9 +3824,6 @@ static void rtl_hw_start_8168(struct rtl8169_private *tp) static void rtl_hw_start_8169(struct rtl8169_private *tp) { - if (tp->mac_version == RTL_GIGA_MAC_VER_05) - pci_write_config_byte(tp->pci_dev, PCI_CACHE_LINE_SIZE, 0x08); - RTL_W8(tp, EarlyTxThres, NoEarlyTx); tp->cp_cmd |= PCIMulRW; @@ -3853,8 +3836,6 @@ static void rtl_hw_start_8169(struct rtl8169_private *tp) rtl8169_set_magic_reg(tp, tp->mac_version); - RTL_W32(tp, RxMissed, 0); - /* disable interrupt coalescing */ RTL_W16(tp, IntrMitigate, 0x0000); } @@ -3877,10 +3858,11 @@ static void rtl_hw_start(struct rtl8169_private *tp) rtl_set_rx_tx_desc_registers(tp); rtl_lock_config_regs(tp); - rtl_jumbo_config(tp, tp->dev->mtu); + rtl_jumbo_config(tp); /* Initially a 10 us delay. Turned it into a PCI commit. - FR */ - RTL_R16(tp, CPlusCmd); + rtl_pci_commit(tp); + RTL_W8(tp, ChipCmd, CmdTxEnb | CmdRxEnb); rtl_init_rxcfg(tp); rtl_set_tx_config_registers(tp); @@ -3892,10 +3874,9 @@ static int rtl8169_change_mtu(struct net_device *dev, int new_mtu) { struct rtl8169_private *tp = netdev_priv(dev); - rtl_jumbo_config(tp, new_mtu); - dev->mtu = new_mtu; netdev_update_features(dev); + rtl_jumbo_config(tp); return 0; } @@ -3910,6 +3891,7 @@ static inline void rtl8169_mark_to_asic(struct RxDesc *desc) { u32 eor = le32_to_cpu(desc->opts1) & RingEnd; + desc->opts2 = 0; /* Force memory writes to complete before releasing descriptor */ dma_wmb(); @@ -3991,17 +3973,15 @@ static int rtl8169_init_ring(struct rtl8169_private *tp) return rtl8169_rx_fill(tp); } -static void rtl8169_unmap_tx_skb(struct device *d, struct ring_info *tx_skb, - struct TxDesc *desc) +static void rtl8169_unmap_tx_skb(struct rtl8169_private *tp, unsigned int entry) { - unsigned int len = tx_skb->len; + struct ring_info *tx_skb = tp->tx_skb + entry; + struct TxDesc *desc = tp->TxDescArray + entry; - dma_unmap_single(d, le64_to_cpu(desc->addr), len, DMA_TO_DEVICE); - - desc->opts1 = 0x00; - desc->opts2 = 0x00; - desc->addr = 0x00; - tx_skb->len = 0; + dma_unmap_single(tp_to_dev(tp), le64_to_cpu(desc->addr), tx_skb->len, + DMA_TO_DEVICE); + memset(desc, 0, sizeof(*desc)); + memset(tx_skb, 0, sizeof(*tx_skb)); } static void rtl8169_tx_clear_range(struct rtl8169_private *tp, u32 start, @@ -4017,12 +3997,9 @@ static void rtl8169_tx_clear_range(struct rtl8169_private *tp, u32 start, if (len) { struct sk_buff *skb = tx_skb->skb; - rtl8169_unmap_tx_skb(tp_to_dev(tp), tx_skb, - tp->TxDescArray + entry); - if (skb) { + rtl8169_unmap_tx_skb(tp, entry); + if (skb) dev_consume_skb_any(skb); - tx_skb->skb = NULL; - } } } } @@ -4063,57 +4040,56 @@ static void rtl8169_tx_timeout(struct net_device *dev, unsigned int txqueue) rtl_schedule_task(tp, RTL_FLAG_TASK_RESET_PENDING); } -static __le32 rtl8169_get_txd_opts1(u32 opts0, u32 len, unsigned int entry) +static int rtl8169_tx_map(struct rtl8169_private *tp, const u32 *opts, u32 len, + void *addr, unsigned int entry, bool desc_own) { - u32 status = opts0 | len; + struct TxDesc *txd = tp->TxDescArray + entry; + struct device *d = tp_to_dev(tp); + dma_addr_t mapping; + u32 opts1; + int ret; + + mapping = dma_map_single(d, addr, len, DMA_TO_DEVICE); + ret = dma_mapping_error(d, mapping); + if (unlikely(ret)) { + if (net_ratelimit()) + netif_err(tp, drv, tp->dev, "Failed to map TX data!\n"); + return ret; + } + txd->addr = cpu_to_le64(mapping); + txd->opts2 = cpu_to_le32(opts[1]); + + opts1 = opts[0] | len; if (entry == NUM_TX_DESC - 1) - status |= RingEnd; + opts1 |= RingEnd; + if (desc_own) + opts1 |= DescOwn; + txd->opts1 = cpu_to_le32(opts1); + + tp->tx_skb[entry].len = len; - return cpu_to_le32(status); + return 0; } static int rtl8169_xmit_frags(struct rtl8169_private *tp, struct sk_buff *skb, - u32 *opts) + const u32 *opts, unsigned int entry) { struct skb_shared_info *info = skb_shinfo(skb); - unsigned int cur_frag, entry; - struct TxDesc *uninitialized_var(txd); - struct device *d = tp_to_dev(tp); + unsigned int cur_frag; - entry = tp->cur_tx; for (cur_frag = 0; cur_frag < info->nr_frags; cur_frag++) { const skb_frag_t *frag = info->frags + cur_frag; - dma_addr_t mapping; - u32 len; - void *addr; + void *addr = skb_frag_address(frag); + u32 len = skb_frag_size(frag); entry = (entry + 1) % NUM_TX_DESC; - txd = tp->TxDescArray + entry; - len = skb_frag_size(frag); - addr = skb_frag_address(frag); - mapping = dma_map_single(d, addr, len, DMA_TO_DEVICE); - if (unlikely(dma_mapping_error(d, mapping))) { - if (net_ratelimit()) - netif_err(tp, drv, tp->dev, - "Failed to map TX fragments DMA!\n"); + if (unlikely(rtl8169_tx_map(tp, opts, len, addr, entry, true))) goto err_out; - } - - txd->opts1 = rtl8169_get_txd_opts1(opts[0], len, entry); - txd->opts2 = cpu_to_le32(opts[1]); - txd->addr = cpu_to_le64(mapping); - - tp->tx_skb[entry].len = len; - } - - if (cur_frag) { - tp->tx_skb[entry].skb = skb; - txd->opts1 |= cpu_to_le32(LastFrag); } - return cur_frag; + return 0; err_out: rtl8169_tx_clear_range(tp, tp->cur_tx + 1, cur_frag); @@ -4125,36 +4101,13 @@ static bool rtl_test_hw_pad_bug(struct rtl8169_private *tp, struct sk_buff *skb) return skb->len < ETH_ZLEN && tp->mac_version == RTL_GIGA_MAC_VER_34; } -/* msdn_giant_send_check() - * According to the document of microsoft, the TCP Pseudo Header excludes the - * packet length for IPv6 TCP large packets. - */ -static int msdn_giant_send_check(struct sk_buff *skb) -{ - const struct ipv6hdr *ipv6h; - struct tcphdr *th; - int ret; - - ret = skb_cow_head(skb, 0); - if (ret) - return ret; - - ipv6h = ipv6_hdr(skb); - th = tcp_hdr(skb); - - th->check = 0; - th->check = ~tcp_v6_check(0, &ipv6h->saddr, &ipv6h->daddr, 0); - - return ret; -} - static void rtl8169_tso_csum_v1(struct sk_buff *skb, u32 *opts) { u32 mss = skb_shinfo(skb)->gso_size; if (mss) { opts[0] |= TD_LSO; - opts[0] |= min(mss, TD_MSS_MAX) << TD0_MSS_SHIFT; + opts[0] |= mss << TD0_MSS_SHIFT; } else if (skb->ip_summed == CHECKSUM_PARTIAL) { const struct iphdr *ip = ip_hdr(skb); @@ -4180,9 +4133,10 @@ static bool rtl8169_tso_csum_v2(struct rtl8169_private *tp, break; case htons(ETH_P_IPV6): - if (msdn_giant_send_check(skb)) + if (skb_cow_head(skb, 0)) return false; + tcp_v6_gso_csum_prep(skb); opts[0] |= TD1_GTSENV6; break; @@ -4192,7 +4146,7 @@ static bool rtl8169_tso_csum_v2(struct rtl8169_private *tp, } opts[0] |= transport_offset << GTTCPHO_SHIFT; - opts[1] |= min(mss, TD_MSS_MAX) << TD1_MSS_SHIFT; + opts[1] |= mss << TD1_MSS_SHIFT; } else if (skb->ip_summed == CHECKSUM_PARTIAL) { u8 ip_protocol; @@ -4260,56 +4214,44 @@ static void rtl8169_doorbell(struct rtl8169_private *tp) static netdev_tx_t rtl8169_start_xmit(struct sk_buff *skb, struct net_device *dev) { + unsigned int frags = skb_shinfo(skb)->nr_frags; struct rtl8169_private *tp = netdev_priv(dev); unsigned int entry = tp->cur_tx % NUM_TX_DESC; - struct TxDesc *txd = tp->TxDescArray + entry; - struct device *d = tp_to_dev(tp); - dma_addr_t mapping; - u32 opts[2], len; - bool stop_queue; - bool door_bell; - int frags; + struct TxDesc *txd_first, *txd_last; + bool stop_queue, door_bell; + u32 opts[2]; - if (unlikely(!rtl_tx_slots_avail(tp, skb_shinfo(skb)->nr_frags))) { + txd_first = tp->TxDescArray + entry; + + if (unlikely(!rtl_tx_slots_avail(tp, frags))) { netif_err(tp, drv, dev, "BUG! Tx Ring full when queue awake!\n"); goto err_stop_0; } - if (unlikely(le32_to_cpu(txd->opts1) & DescOwn)) + if (unlikely(le32_to_cpu(txd_first->opts1) & DescOwn)) goto err_stop_0; opts[1] = rtl8169_tx_vlan_tag(skb); - opts[0] = DescOwn; + opts[0] = 0; - if (rtl_chip_supports_csum_v2(tp)) { - if (!rtl8169_tso_csum_v2(tp, skb, opts)) - goto err_dma_0; - } else { + if (!rtl_chip_supports_csum_v2(tp)) rtl8169_tso_csum_v1(skb, opts); - } - - len = skb_headlen(skb); - mapping = dma_map_single(d, skb->data, len, DMA_TO_DEVICE); - if (unlikely(dma_mapping_error(d, mapping))) { - if (net_ratelimit()) - netif_err(tp, drv, dev, "Failed to map TX DMA!\n"); + else if (!rtl8169_tso_csum_v2(tp, skb, opts)) goto err_dma_0; - } - tp->tx_skb[entry].len = len; - txd->addr = cpu_to_le64(mapping); + if (unlikely(rtl8169_tx_map(tp, opts, skb_headlen(skb), skb->data, + entry, false))) + goto err_dma_0; - frags = rtl8169_xmit_frags(tp, skb, opts); - if (frags < 0) - goto err_dma_1; - else if (frags) - opts[0] |= FirstFrag; - else { - opts[0] |= FirstFrag | LastFrag; - tp->tx_skb[entry].skb = skb; + if (frags) { + if (rtl8169_xmit_frags(tp, skb, opts, entry)) + goto err_dma_1; + entry = (entry + frags) % NUM_TX_DESC; } - txd->opts2 = cpu_to_le32(opts[1]); + txd_last = tp->TxDescArray + entry; + txd_last->opts1 |= cpu_to_le32(LastFrag); + tp->tx_skb[entry].skb = skb; skb_tx_timestamp(skb); @@ -4318,7 +4260,7 @@ static netdev_tx_t rtl8169_start_xmit(struct sk_buff *skb, door_bell = __netdev_sent_queue(dev, skb->len, netdev_xmit_more()); - txd->opts1 = rtl8169_get_txd_opts1(opts[0], len, entry); + txd_first->opts1 |= cpu_to_le32(DescOwn | FirstFrag); /* Force all memory writes to complete before notifying device */ wmb(); @@ -4354,7 +4296,7 @@ static netdev_tx_t rtl8169_start_xmit(struct sk_buff *skb, return NETDEV_TX_OK; err_dma_1: - rtl8169_unmap_tx_skb(d, tp->tx_skb + entry, txd); + rtl8169_unmap_tx_skb(tp, entry); err_dma_0: dev_kfree_skb_any(skb); dev->stats.tx_dropped++; @@ -4403,13 +4345,15 @@ static void rtl8169_pcierr_interrupt(struct net_device *dev) { struct rtl8169_private *tp = netdev_priv(dev); struct pci_dev *pdev = tp->pci_dev; - u16 pci_status, pci_cmd; + int pci_status_errs; + u16 pci_cmd; pci_read_config_word(pdev, PCI_COMMAND, &pci_cmd); - pci_read_config_word(pdev, PCI_STATUS, &pci_status); - netif_err(tp, intr, dev, "PCI error (cmd = 0x%04x, status = 0x%04x)\n", - pci_cmd, pci_status); + pci_status_errs = pci_status_get_and_clear_errors(pdev); + + netif_err(tp, intr, dev, "PCI error (cmd = 0x%04x, status_errs = 0x%04x)\n", + pci_cmd, pci_status_errs); /* * The recovery sequence below admits a very elaborated explanation: @@ -4426,11 +4370,6 @@ static void rtl8169_pcierr_interrupt(struct net_device *dev) pci_write_config_word(pdev, PCI_COMMAND, pci_cmd); - pci_write_config_word(pdev, PCI_STATUS, - pci_status & (PCI_STATUS_DETECTED_PARITY | - PCI_STATUS_SIG_SYSTEM_ERROR | PCI_STATUS_REC_MASTER_ABORT | - PCI_STATUS_REC_TARGET_ABORT | PCI_STATUS_SIG_TARGET_ABORT)); - rtl_schedule_task(tp, RTL_FLAG_TASK_RESET_PENDING); } @@ -4441,33 +4380,24 @@ static void rtl_tx(struct net_device *dev, struct rtl8169_private *tp, dirty_tx = tp->dirty_tx; smp_rmb(); - tx_left = tp->cur_tx - dirty_tx; - while (tx_left > 0) { + for (tx_left = tp->cur_tx - dirty_tx; tx_left > 0; tx_left--) { unsigned int entry = dirty_tx % NUM_TX_DESC; - struct ring_info *tx_skb = tp->tx_skb + entry; + struct sk_buff *skb = tp->tx_skb[entry].skb; u32 status; status = le32_to_cpu(tp->TxDescArray[entry].opts1); if (status & DescOwn) break; - /* This barrier is needed to keep us from reading - * any other fields out of the Tx descriptor until - * we know the status of DescOwn - */ - dma_rmb(); + rtl8169_unmap_tx_skb(tp, entry); - rtl8169_unmap_tx_skb(tp_to_dev(tp), tx_skb, - tp->TxDescArray + entry); - if (tx_skb->skb) { + if (skb) { pkts_compl++; - bytes_compl += tx_skb->skb->len; - napi_consume_skb(tx_skb->skb, budget); - tx_skb->skb = NULL; + bytes_compl += skb->len; + napi_consume_skb(skb, budget); } dirty_tx++; - tx_left--; } if (tp->dirty_tx != dirty_tx) { @@ -4606,7 +4536,6 @@ process_pkt: u64_stats_update_end(&tp->rx_stats.syncp); } release_descriptor: - desc->opts2 = 0; rtl8169_mark_to_asic(desc); } @@ -4636,8 +4565,7 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance) if (unlikely(status & RxFIFOOver && tp->mac_version == RTL_GIGA_MAC_VER_11)) { netif_stop_queue(tp->dev); - /* XXX - Hack alert. See rtl_task(). */ - set_bit(RTL_FLAG_TASK_RESET_PENDING, tp->wk.flags); + rtl_schedule_task(tp, RTL_FLAG_TASK_RESET_PENDING); } rtl_irq_disable(tp); @@ -4650,31 +4578,17 @@ out: static void rtl_task(struct work_struct *work) { - static const struct { - int bitnr; - void (*action)(struct rtl8169_private *); - } rtl_work[] = { - { RTL_FLAG_TASK_RESET_PENDING, rtl_reset_work }, - }; struct rtl8169_private *tp = container_of(work, struct rtl8169_private, wk.work); - struct net_device *dev = tp->dev; - int i; rtl_lock_work(tp); - if (!netif_running(dev) || + if (!netif_running(tp->dev) || !test_bit(RTL_FLAG_TASK_ENABLED, tp->wk.flags)) goto out_unlock; - for (i = 0; i < ARRAY_SIZE(rtl_work); i++) { - bool pending; - - pending = test_and_clear_bit(rtl_work[i].bitnr, tp->wk.flags); - if (pending) - rtl_work[i].action(tp); - } - + if (test_and_clear_bit(RTL_FLAG_TASK_RESET_PENDING, tp->wk.flags)) + rtl_reset_work(tp); out_unlock: rtl_unlock_work(tp); } @@ -4697,17 +4611,6 @@ static int rtl8169_poll(struct napi_struct *napi, int budget) return work_done; } -static void rtl8169_rx_missed(struct net_device *dev) -{ - struct rtl8169_private *tp = netdev_priv(dev); - - if (tp->mac_version > RTL_GIGA_MAC_VER_06) - return; - - dev->stats.rx_missed_errors += RTL_R32(tp, RxMissed) & 0xffffff; - RTL_W32(tp, RxMissed, 0); -} - static void r8169_phylink_handler(struct net_device *ndev) { struct rtl8169_private *tp = netdev_priv(ndev); @@ -4757,12 +4660,6 @@ static void rtl8169_down(struct net_device *dev) netif_stop_queue(dev); rtl8169_hw_reset(tp); - /* - * At this point device interrupts can not be enabled in any function, - * as netif_running is not true (rtl8169_interrupt, rtl8169_reset_task) - * and napi is disabled (rtl8169_poll). - */ - rtl8169_rx_missed(dev); /* Give a racing hard_start_xmit a few cycles to complete. */ synchronize_rcu(); @@ -4907,8 +4804,7 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) pm_runtime_get_noresume(&pdev->dev); - if (netif_running(dev) && pm_runtime_active(&pdev->dev)) - rtl8169_rx_missed(dev); + netdev_stats_to_stats64(stats, &dev->stats); do { start = u64_stats_fetch_begin_irq(&tp->rx_stats.syncp); @@ -4922,15 +4818,6 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) stats->tx_bytes = tp->tx_stats.bytes; } while (u64_stats_fetch_retry_irq(&tp->tx_stats.syncp, start)); - stats->rx_dropped = dev->stats.rx_dropped; - stats->tx_dropped = dev->stats.tx_dropped; - stats->rx_length_errors = dev->stats.rx_length_errors; - stats->rx_errors = dev->stats.rx_errors; - stats->rx_crc_errors = dev->stats.rx_crc_errors; - stats->rx_fifo_errors = dev->stats.rx_fifo_errors; - stats->rx_missed_errors = dev->stats.rx_missed_errors; - stats->multicast = dev->stats.multicast; - /* * Fetch additional counter values missing in stats collected by driver * from tally counters. @@ -4948,6 +4835,8 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) le32_to_cpu(tp->tc_offset.tx_multi_collision); stats->tx_aborted_errors = le16_to_cpu(counters->tx_aborted) - le16_to_cpu(tp->tc_offset.tx_aborted); + stats->rx_missed_errors = le16_to_cpu(counters->rx_missed) - + le16_to_cpu(tp->tc_offset.rx_missed); pm_runtime_put_noidle(&pdev->dev); } @@ -5033,7 +4922,6 @@ static int rtl8169_runtime_suspend(struct device *device) rtl8169_net_suspend(dev); /* Update counters before going runtime suspend */ - rtl8169_rx_missed(dev); rtl8169_update_counters(tp); return 0; @@ -5098,8 +4986,7 @@ static void rtl_wol_shutdown_quirk(struct rtl8169_private *tp) pci_clear_master(tp->pci_dev); RTL_W8(tp, ChipCmd, CmdRxEnb); - /* PCI commit */ - RTL_R8(tp, ChipCmd); + rtl_pci_commit(tp); break; default: break; @@ -5173,7 +5060,7 @@ static const struct net_device_ops rtl_netdev_ops = { static void rtl_set_irq_mask(struct rtl8169_private *tp) { - tp->irq_mask = RTL_EVENT_NAPI | LinkChg; + tp->irq_mask = RxOK | RxErr | TxOK | TxErr | LinkChg; if (tp->mac_version <= RTL_GIGA_MAC_VER_06) tp->irq_mask |= SYSErr | RxOverflow | RxFIFOOver; @@ -5449,9 +5336,10 @@ done: static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { struct rtl8169_private *tp; + int jumbo_max, region, rc; + enum mac_version chipset; struct net_device *dev; - int chipset, region; - int jumbo_max, rc; + u16 xid; dev = devm_alloc_etherdev(&pdev->dev, sizeof (*tp)); if (!dev) @@ -5509,10 +5397,16 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) tp->mmio_addr = pcim_iomap_table(pdev)[region]; + xid = (RTL_R32(tp, TxConfig) >> 20) & 0xfcf; + /* Identify chip attached to board */ - rtl8169_get_mac_version(tp); - if (tp->mac_version == RTL_GIGA_MAC_NONE) + chipset = rtl8169_get_mac_version(xid, tp->supports_gmii); + if (chipset == RTL_GIGA_MAC_NONE) { + dev_err(&pdev->dev, "unknown chip XID %03x\n", xid); return -ENODEV; + } + + tp->mac_version = chipset; tp->cp_cmd = RTL_R16(tp, CPlusCmd); @@ -5530,8 +5424,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_master(pdev); - chipset = tp->mac_version; - rc = rtl_alloc_irq(tp); if (rc < 0) { dev_err(&pdev->dev, "Can't allocate interrupt\n"); @@ -5549,9 +5441,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) netif_napi_add(dev, &tp->napi, rtl8169_poll, NAPI_POLL_WEIGHT); - dev->features |= NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO | - NETIF_F_RXCSUM | NETIF_F_HW_VLAN_CTAG_TX | - NETIF_F_HW_VLAN_CTAG_RX; dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO | NETIF_F_RXCSUM | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX; @@ -5573,7 +5462,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (rtl_chip_supports_csum_v2(tp)) { dev->hw_features |= NETIF_F_IPV6_CSUM | NETIF_F_TSO6; - dev->features |= NETIF_F_IPV6_CSUM | NETIF_F_TSO6; dev->gso_max_size = RTL_GSO_MAX_SIZE_V2; dev->gso_max_segs = RTL_GSO_MAX_SEGS_V2; } else { @@ -5588,9 +5476,10 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) tp->mac_version == RTL_GIGA_MAC_VER_22) { dev->vlan_features &= ~(NETIF_F_ALL_TSO | NETIF_F_SG); dev->hw_features &= ~(NETIF_F_ALL_TSO | NETIF_F_SG); - dev->features &= ~(NETIF_F_ALL_TSO | NETIF_F_SG); } + dev->features |= dev->hw_features; + dev->hw_features |= NETIF_F_RXALL; dev->hw_features |= NETIF_F_RXFCS; @@ -5622,8 +5511,7 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_mdio_unregister; netif_info(tp, probe, dev, "%s, %pM, XID %03x, IRQ %d\n", - rtl_chip_infos[chipset].name, dev->dev_addr, - (RTL_R32(tp, TxConfig) >> 20) & 0xfcf, + rtl_chip_infos[chipset].name, dev->dev_addr, xid, pci_irq_vector(pdev, 0)); if (jumbo_max) diff --git a/drivers/net/ethernet/realtek/r8169_phy_config.c b/drivers/net/ethernet/realtek/r8169_phy_config.c index e367e77c773b..b73f7d023e99 100644 --- a/drivers/net/ethernet/realtek/r8169_phy_config.c +++ b/drivers/net/ethernet/realtek/r8169_phy_config.c @@ -796,6 +796,11 @@ static void rtl8168g_disable_aldps(struct phy_device *phydev) phy_modify_paged(phydev, 0x0a43, 0x10, BIT(2), 0); } +static void rtl8168g_enable_gphy_10m(struct phy_device *phydev) +{ + phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11)); +} + static void rtl8168g_phy_adjust_10m_aldps(struct phy_device *phydev) { phy_modify_paged(phydev, 0x0bcc, 0x14, BIT(8), 0); @@ -904,8 +909,7 @@ static void rtl8168h_1_hw_phy_config(struct rtl8169_private *tp, r8168g_phy_param(phydev, 0x0811, 0x0000, 0x0800); phy_modify_paged(phydev, 0x0a42, 0x16, 0x0000, 0x0002); - /* enable GPHY 10M */ - phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11)); + rtl8168g_enable_gphy_10m(phydev); /* SAR ADC performance */ phy_modify_paged(phydev, 0x0bca, 0x17, BIT(12) | BIT(13), BIT(14)); @@ -940,8 +944,7 @@ static void rtl8168h_2_hw_phy_config(struct rtl8169_private *tp, r8168g_phy_param(phydev, 0x0811, 0x0000, 0x0800); phy_modify_paged(phydev, 0x0a42, 0x16, 0x0000, 0x0002); - /* enable GPHY 10M */ - phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11)); + rtl8168g_enable_gphy_10m(phydev); ioffset = rtl8168h_2_get_adc_bias_ioffset(tp); if (ioffset != 0xffff) @@ -1063,8 +1066,7 @@ static void rtl8117_hw_phy_config(struct rtl8169_private *tp, r8168g_phy_param(phydev, 0x8011, 0x0000, 0x0800); - /* enable GPHY 10M */ - phy_modify_paged(phydev, 0x0a44, 0x11, 0, BIT(11)); + rtl8168g_enable_gphy_10m(phydev); r8168g_phy_param(phydev, 0x8016, 0x0000, 0x0400); @@ -1171,7 +1173,7 @@ static void rtl8125_1_hw_phy_config(struct rtl8169_private *tp, phy_write_paged(phydev, 0xbc3, 0x12, 0x5555); phy_modify_paged(phydev, 0xbf0, 0x15, 0x0e00, 0x0a00); phy_modify_paged(phydev, 0xa5c, 0x10, 0x0400, 0x0000); - phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800); + rtl8168g_enable_gphy_10m(phydev); rtl8125_config_eee_phy(phydev); } @@ -1236,7 +1238,7 @@ static void rtl8125_2_hw_phy_config(struct rtl8169_private *tp, phy_modify_paged(phydev, 0xa5d, 0x12, 0x0000, 0x0020); phy_modify_paged(phydev, 0xad4, 0x17, 0x0010, 0x0000); phy_modify_paged(phydev, 0xa86, 0x15, 0x0001, 0x0000); - phy_modify_paged(phydev, 0xa44, 0x11, 0x0000, 0x0800); + rtl8168g_enable_gphy_10m(phydev); rtl8125_config_eee_phy(phydev); } diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 58ca126518a2..8ed73f44405d 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -142,69 +142,6 @@ static const u16 sh_eth_offset_gigabit[SH_ETH_MAX_REGISTER_OFFSET] = { [FWALCR1] = 0x00b4, }; -static const u16 sh_eth_offset_fast_rz[SH_ETH_MAX_REGISTER_OFFSET] = { - SH_ETH_OFFSET_DEFAULTS, - - [EDSR] = 0x0000, - [EDMR] = 0x0400, - [EDTRR] = 0x0408, - [EDRRR] = 0x0410, - [EESR] = 0x0428, - [EESIPR] = 0x0430, - [TDLAR] = 0x0010, - [TDFAR] = 0x0014, - [TDFXR] = 0x0018, - [TDFFR] = 0x001c, - [RDLAR] = 0x0030, - [RDFAR] = 0x0034, - [RDFXR] = 0x0038, - [RDFFR] = 0x003c, - [TRSCER] = 0x0438, - [RMFCR] = 0x0440, - [TFTR] = 0x0448, - [FDR] = 0x0450, - [RMCR] = 0x0458, - [RPADIR] = 0x0460, - [FCFTR] = 0x0468, - [CSMR] = 0x04E4, - - [ECMR] = 0x0500, - [RFLR] = 0x0508, - [ECSR] = 0x0510, - [ECSIPR] = 0x0518, - [PIR] = 0x0520, - [APR] = 0x0554, - [MPR] = 0x0558, - [PFTCR] = 0x055c, - [PFRCR] = 0x0560, - [TPAUSER] = 0x0564, - [MAHR] = 0x05c0, - [MALR] = 0x05c8, - [CEFCR] = 0x0740, - [FRECR] = 0x0748, - [TSFRCR] = 0x0750, - [TLFRCR] = 0x0758, - [RFCR] = 0x0760, - [MAFCR] = 0x0778, - - [ARSTR] = 0x0000, - [TSU_CTRST] = 0x0004, - [TSU_FWSLC] = 0x0038, - [TSU_VTAG0] = 0x0058, - [TSU_ADSBSY] = 0x0060, - [TSU_TEN] = 0x0064, - [TSU_POST1] = 0x0070, - [TSU_POST2] = 0x0074, - [TSU_POST3] = 0x0078, - [TSU_POST4] = 0x007c, - [TSU_ADRH0] = 0x0100, - - [TXNLCR0] = 0x0080, - [TXALCR0] = 0x0084, - [RXNLCR0] = 0x0088, - [RXALCR0] = 0x008C, -}; - static const u16 sh_eth_offset_fast_rcar[SH_ETH_MAX_REGISTER_OFFSET] = { SH_ETH_OFFSET_DEFAULTS, @@ -569,6 +506,9 @@ static void sh_eth_set_rate_gether(struct net_device *ndev) { struct sh_eth_private *mdp = netdev_priv(ndev); + if (WARN_ON(!mdp->cd->gecmr)) + return; + switch (mdp->speed) { case 10: /* 10BASE */ sh_eth_write(ndev, GECMR_10, GECMR); @@ -590,7 +530,7 @@ static struct sh_eth_cpu_data r7s72100_data = { .chip_reset = sh_eth_chip_reset, .set_duplex = sh_eth_set_duplex, - .register_type = SH_ETH_REG_FAST_RZ, + .register_type = SH_ETH_REG_GIGABIT, .edtrr_trns = EDTRR_TRNS_GETHER, .ecsr_value = ECSR_ICD, @@ -663,6 +603,7 @@ static struct sh_eth_cpu_data r8a7740_data = { .apr = 1, .mpr = 1, .tpauser = 1, + .gecmr = 1, .bculr = 1, .hw_swap = 1, .rpadir = 1, @@ -788,6 +729,7 @@ static struct sh_eth_cpu_data r8a77980_data = { .apr = 1, .mpr = 1, .tpauser = 1, + .gecmr = 1, .bculr = 1, .hw_swap = 1, .nbst = 1, @@ -957,6 +899,9 @@ static void sh_eth_set_rate_giga(struct net_device *ndev) { struct sh_eth_private *mdp = netdev_priv(ndev); + if (WARN_ON(!mdp->cd->gecmr)) + return; + switch (mdp->speed) { case 10: /* 10BASE */ sh_eth_write(ndev, 0x00000000, GECMR); @@ -1002,6 +947,7 @@ static struct sh_eth_cpu_data sh7757_data_giga = { .apr = 1, .mpr = 1, .tpauser = 1, + .gecmr = 1, .bculr = 1, .hw_swap = 1, .rpadir = 1, @@ -1042,6 +988,7 @@ static struct sh_eth_cpu_data sh7734_data = { .apr = 1, .mpr = 1, .tpauser = 1, + .gecmr = 1, .bculr = 1, .hw_swap = 1, .no_trimd = 1, @@ -1083,6 +1030,7 @@ static struct sh_eth_cpu_data sh7763_data = { .apr = 1, .mpr = 1, .tpauser = 1, + .gecmr = 1, .bculr = 1, .hw_swap = 1, .no_trimd = 1, @@ -2140,11 +2088,13 @@ static size_t __sh_eth_get_regs(struct net_device *ndev, u32 *buf) add_reg(EESR); add_reg(EESIPR); add_reg(TDLAR); - add_reg(TDFAR); + if (!cd->no_xdfar) + add_reg(TDFAR); add_reg(TDFXR); add_reg(TDFFR); add_reg(RDLAR); - add_reg(RDFAR); + if (!cd->no_xdfar) + add_reg(RDFAR); add_reg(RDFXR); add_reg(RDFFR); add_reg(TRSCER); @@ -2179,21 +2129,26 @@ static size_t __sh_eth_get_regs(struct net_device *ndev, u32 *buf) if (cd->tpauser) add_reg(TPAUSER); add_reg(TPAUSECR); - add_reg(GECMR); + if (cd->gecmr) + add_reg(GECMR); if (cd->bculr) add_reg(BCULR); add_reg(MAHR); add_reg(MALR); - add_reg(TROCR); - add_reg(CDCR); - add_reg(LCCR); - add_reg(CNDCR); + if (!cd->no_tx_cntrs) { + add_reg(TROCR); + add_reg(CDCR); + add_reg(LCCR); + add_reg(CNDCR); + } add_reg(CEFCR); add_reg(FRECR); add_reg(TSFRCR); add_reg(TLFRCR); - add_reg(CERCR); - add_reg(CEECR); + if (cd->cexcr) { + add_reg(CERCR); + add_reg(CEECR); + } add_reg(MAFCR); if (cd->rtrate) add_reg(RTRATE); @@ -3121,9 +3076,6 @@ static const u16 *sh_eth_get_register_offset(int register_type) case SH_ETH_REG_GIGABIT: reg_offset = sh_eth_offset_gigabit; break; - case SH_ETH_REG_FAST_RZ: - reg_offset = sh_eth_offset_fast_rz; - break; case SH_ETH_REG_FAST_RCAR: reg_offset = sh_eth_offset_fast_rcar; break; diff --git a/drivers/net/ethernet/renesas/sh_eth.h b/drivers/net/ethernet/renesas/sh_eth.h index 850726301e1c..c1b3751b12c4 100644 --- a/drivers/net/ethernet/renesas/sh_eth.h +++ b/drivers/net/ethernet/renesas/sh_eth.h @@ -145,7 +145,6 @@ enum { enum { SH_ETH_REG_GIGABIT, - SH_ETH_REG_FAST_RZ, SH_ETH_REG_FAST_RCAR, SH_ETH_REG_FAST_SH4, SH_ETH_REG_FAST_SH3_SH2 @@ -490,6 +489,7 @@ struct sh_eth_cpu_data { unsigned apr:1; /* EtherC has APR */ unsigned mpr:1; /* EtherC has MPR */ unsigned tpauser:1; /* EtherC has TPAUSER */ + unsigned gecmr:1; /* EtherC has GECMR */ unsigned bculr:1; /* EtherC has BCULR */ unsigned tsu:1; /* EtherC has TSU */ unsigned hw_swap:1; /* E-DMAC has DE bit in EDMR */ diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c index 466483c4ac67..21465cb3d60a 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c @@ -476,6 +476,7 @@ static int sxgbe_get_regs_len(struct net_device *dev) } static const struct ethtool_ops sxgbe_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = sxgbe_getdrvinfo, .get_msglevel = sxgbe_getmsglevel, .set_msglevel = sxgbe_setmsglevel, diff --git a/drivers/net/ethernet/sfc/ef10_sriov.c b/drivers/net/ethernet/sfc/ef10_sriov.c index 14393767ef9f..4580b30caae1 100644 --- a/drivers/net/ethernet/sfc/ef10_sriov.c +++ b/drivers/net/ethernet/sfc/ef10_sriov.c @@ -685,10 +685,70 @@ reset_nic: return rc ? rc : rc2; } -int efx_ef10_sriov_set_vf_spoofchk(struct efx_nic *efx, int vf_i, - bool spoofchk) +static int efx_ef10_sriov_set_privilege_mask(struct efx_nic *efx, int vf_i, + u32 mask, u32 value) { - return spoofchk ? -EOPNOTSUPP : 0; + MCDI_DECLARE_BUF(pm_outbuf, MC_CMD_PRIVILEGE_MASK_OUT_LEN); + MCDI_DECLARE_BUF(pm_inbuf, MC_CMD_PRIVILEGE_MASK_IN_LEN); + struct efx_ef10_nic_data *nic_data = efx->nic_data; + u32 old_mask, new_mask; + size_t outlen; + int rc; + + EFX_WARN_ON_PARANOID((value & ~mask) != 0); + + /* Get privilege mask */ + MCDI_POPULATE_DWORD_2(pm_inbuf, PRIVILEGE_MASK_IN_FUNCTION, + PRIVILEGE_MASK_IN_FUNCTION_PF, nic_data->pf_index, + PRIVILEGE_MASK_IN_FUNCTION_VF, vf_i); + + rc = efx_mcdi_rpc(efx, MC_CMD_PRIVILEGE_MASK, + pm_inbuf, sizeof(pm_inbuf), + pm_outbuf, sizeof(pm_outbuf), &outlen); + + if (rc != 0) + return rc; + if (outlen != MC_CMD_PRIVILEGE_MASK_OUT_LEN) + return -EIO; + + old_mask = MCDI_DWORD(pm_outbuf, PRIVILEGE_MASK_OUT_OLD_MASK); + + new_mask = old_mask & ~mask; + new_mask |= value; + + if (new_mask == old_mask) + return 0; + + new_mask |= MC_CMD_PRIVILEGE_MASK_IN_DO_CHANGE; + + /* Set privilege mask */ + MCDI_SET_DWORD(pm_inbuf, PRIVILEGE_MASK_IN_NEW_MASK, new_mask); + + rc = efx_mcdi_rpc(efx, MC_CMD_PRIVILEGE_MASK, + pm_inbuf, sizeof(pm_inbuf), + pm_outbuf, sizeof(pm_outbuf), &outlen); + + if (rc != 0) + return rc; + if (outlen != MC_CMD_PRIVILEGE_MASK_OUT_LEN) + return -EIO; + + return 0; +} + +int efx_ef10_sriov_set_vf_spoofchk(struct efx_nic *efx, int vf_i, bool spoofchk) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + /* Can't enable spoofchk if firmware doesn't support it. */ + if (!(nic_data->datapath_caps & + BIT(MC_CMD_GET_CAPABILITIES_OUT_TX_MAC_SECURITY_FILTERING_LBN)) && + spoofchk) + return -EOPNOTSUPP; + + return efx_ef10_sriov_set_privilege_mask(efx, vf_i, + MC_CMD_PRIVILEGE_MASK_IN_GRP_MAC_SPOOFING_TX, + spoofchk ? 0 : MC_CMD_PRIVILEGE_MASK_IN_GRP_MAC_SPOOFING_TX); } int efx_ef10_sriov_set_vf_link_state(struct efx_nic *efx, int vf_i, diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 4481f21a1f43..256807c28ff7 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -113,7 +113,6 @@ MODULE_PARM_DESC(debug, "Bitmapped debugging message enable value"); * *************************************************************************/ -static const struct efx_channel_type efx_default_channel_type; static void efx_remove_port(struct efx_nic *efx); static int efx_xdp_setup_prog(struct efx_nic *efx, struct bpf_prog *prog); static int efx_xdp(struct net_device *dev, struct netdev_bpf *xdp); diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h index 95395d67ea2d..66dcab140449 100644 --- a/drivers/net/ethernet/sfc/efx.h +++ b/drivers/net/ethernet/sfc/efx.h @@ -151,24 +151,6 @@ static inline s32 efx_filter_get_rx_ids(struct efx_nic *efx, int efx_filter_rfs(struct net_device *net_dev, const struct sk_buff *skb, u16 rxq_index, u32 flow_id); bool __efx_filter_rfs_expire(struct efx_channel *channel, unsigned int quota); -static inline void efx_filter_rfs_expire(struct work_struct *data) -{ - struct delayed_work *dwork = to_delayed_work(data); - struct efx_channel *channel; - unsigned int time, quota; - - channel = container_of(dwork, struct efx_channel, filter_work); - time = jiffies - channel->rfs_last_expiry; - quota = channel->rfs_filter_count * time / (30 * HZ); - if (quota > 20 && __efx_filter_rfs_expire(channel, min(channel->rfs_filter_count, quota))) - channel->rfs_last_expiry += time; - /* Ensure we do more work eventually even if NAPI poll is not happening */ - schedule_delayed_work(dwork, 30 * HZ); -} -#define efx_filter_rfs_enabled() 1 -#else -static inline void efx_filter_rfs_expire(struct work_struct *data) {} -#define efx_filter_rfs_enabled() 0 #endif /* RSS contexts */ diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c index 73d4e39b5b16..c492523b986c 100644 --- a/drivers/net/ethernet/sfc/efx_channels.c +++ b/drivers/net/ethernet/sfc/efx_channels.c @@ -485,6 +485,23 @@ void efx_remove_eventq(struct efx_channel *channel) * *************************************************************************/ +#ifdef CONFIG_RFS_ACCEL +static void efx_filter_rfs_expire(struct work_struct *data) +{ + struct delayed_work *dwork = to_delayed_work(data); + struct efx_channel *channel; + unsigned int time, quota; + + channel = container_of(dwork, struct efx_channel, filter_work); + time = jiffies - channel->rfs_last_expiry; + quota = channel->rfs_filter_count * time / (30 * HZ); + if (quota >= 20 && __efx_filter_rfs_expire(channel, min(channel->rfs_filter_count, quota))) + channel->rfs_last_expiry += time; + /* Ensure we do more work eventually even if NAPI poll is not happening */ + schedule_delayed_work(dwork, 30 * HZ); +} +#endif + /* Allocate and initialise a channel structure. */ struct efx_channel * efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel) @@ -1167,6 +1184,9 @@ static int efx_poll(struct napi_struct *napi, int budget) struct efx_channel *channel = container_of(napi, struct efx_channel, napi_str); struct efx_nic *efx = channel->efx; +#ifdef CONFIG_RFS_ACCEL + unsigned int time; +#endif int spent; netif_vdbg(efx, intr, efx->net_dev, @@ -1186,7 +1206,10 @@ static int efx_poll(struct napi_struct *napi, int budget) #ifdef CONFIG_RFS_ACCEL /* Perhaps expire some ARFS filters */ - mod_delayed_work(system_wq, &channel->filter_work, 0); + time = jiffies - channel->rfs_last_expiry; + /* Would our quota be >= 20? */ + if (channel->rfs_filter_count * time >= 600 * HZ) + mod_delayed_work(system_wq, &channel->filter_work, 0); #endif /* There is no race here; although napi_disable() will diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c index b0d76bc19673..1799ff9a45d9 100644 --- a/drivers/net/ethernet/sfc/efx_common.c +++ b/drivers/net/ethernet/sfc/efx_common.c @@ -200,11 +200,11 @@ void efx_link_status_changed(struct efx_nic *efx) unsigned int efx_xdp_max_mtu(struct efx_nic *efx) { /* The maximum MTU that we can fit in a single page, allowing for - * framing, overhead and XDP headroom. + * framing, overhead and XDP headroom + tailroom. */ int overhead = EFX_MAX_FRAME_LEN(0) + sizeof(struct efx_rx_page_state) + efx->rx_prefix_size + efx->type->rx_buffer_padding + - efx->rx_ip_align + XDP_PACKET_HEADROOM; + efx->rx_ip_align + EFX_XDP_HEADROOM + EFX_XDP_TAILROOM; return PAGE_SIZE - overhead; } @@ -302,8 +302,9 @@ static void efx_start_datapath(struct efx_nic *efx) efx->rx_dma_len = (efx->rx_prefix_size + EFX_MAX_FRAME_LEN(efx->net_dev->mtu) + efx->type->rx_buffer_padding); - rx_buf_len = (sizeof(struct efx_rx_page_state) + XDP_PACKET_HEADROOM + - efx->rx_ip_align + efx->rx_dma_len); + rx_buf_len = (sizeof(struct efx_rx_page_state) + EFX_XDP_HEADROOM + + efx->rx_ip_align + efx->rx_dma_len + EFX_XDP_TAILROOM); + if (rx_buf_len <= PAGE_SIZE) { efx->rx_scatter = efx->type->always_rx_scatter; efx->rx_buffer_order = 0; diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c index 993b5769525b..04e88d05e8ff 100644 --- a/drivers/net/ethernet/sfc/ethtool.c +++ b/drivers/net/ethernet/sfc/ethtool.c @@ -232,9 +232,6 @@ static int efx_ethtool_set_coalesce(struct net_device *net_dev, bool adaptive, rx_may_override_tx; int rc; - if (coalesce->use_adaptive_tx_coalesce) - return -EINVAL; - efx_get_irq_moderation(efx, &tx_usecs, &rx_usecs, &adaptive); if (coalesce->rx_coalesce_usecs != rx_usecs) @@ -582,6 +579,7 @@ efx_ethtool_get_rxnfc(struct net_device *net_dev, case ETHTOOL_GRXFH: { struct efx_rss_context *ctx = &efx->rss_context; + __u64 data; mutex_lock(&efx->rss_lock); if (info->flow_type & FLOW_RSS && info->rss_context) { @@ -591,35 +589,38 @@ efx_ethtool_get_rxnfc(struct net_device *net_dev, goto out_unlock; } } - info->data = 0; + + data = 0; if (!efx_rss_active(ctx)) /* No RSS */ - goto out_unlock; + goto out_setdata_unlock; + switch (info->flow_type & ~FLOW_RSS) { case UDP_V4_FLOW: - if (ctx->rx_hash_udp_4tuple) - /* fall through */ - case TCP_V4_FLOW: - info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3; - /* fall through */ - case SCTP_V4_FLOW: - case AH_ESP_V4_FLOW: - case IPV4_FLOW: - info->data |= RXH_IP_SRC | RXH_IP_DST; - break; case UDP_V6_FLOW: if (ctx->rx_hash_udp_4tuple) - /* fall through */ + data = (RXH_L4_B_0_1 | RXH_L4_B_2_3 | + RXH_IP_SRC | RXH_IP_DST); + else + data = RXH_IP_SRC | RXH_IP_DST; + break; + case TCP_V4_FLOW: case TCP_V6_FLOW: - info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3; - /* fall through */ + data = (RXH_L4_B_0_1 | RXH_L4_B_2_3 | + RXH_IP_SRC | RXH_IP_DST); + break; + case SCTP_V4_FLOW: case SCTP_V6_FLOW: + case AH_ESP_V4_FLOW: case AH_ESP_V6_FLOW: + case IPV4_FLOW: case IPV6_FLOW: - info->data |= RXH_IP_SRC | RXH_IP_DST; + data = RXH_IP_SRC | RXH_IP_DST; break; default: break; } +out_setdata_unlock: + info->data = data; out_unlock: mutex_unlock(&efx->rss_lock); return rc; @@ -1134,6 +1135,9 @@ static int efx_ethtool_set_fecparam(struct net_device *net_dev, } const struct ethtool_ops efx_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_USECS_IRQ | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_drvinfo = efx_ethtool_get_drvinfo, .get_regs_len = efx_ethtool_get_regs_len, .get_regs = efx_ethtool_get_regs, diff --git a/drivers/net/ethernet/sfc/falcon/ethtool.c b/drivers/net/ethernet/sfc/falcon/ethtool.c index 08bd6a321918..db90d94e24c9 100644 --- a/drivers/net/ethernet/sfc/falcon/ethtool.c +++ b/drivers/net/ethernet/sfc/falcon/ethtool.c @@ -603,9 +603,6 @@ static int ef4_ethtool_set_coalesce(struct net_device *net_dev, bool adaptive, rx_may_override_tx; int rc; - if (coalesce->use_adaptive_tx_coalesce) - return -EINVAL; - ef4_get_irq_moderation(efx, &tx_usecs, &rx_usecs, &adaptive); if (coalesce->rx_coalesce_usecs != rx_usecs) @@ -1311,6 +1308,9 @@ static int ef4_ethtool_get_module_info(struct net_device *net_dev, } const struct ethtool_ops ef4_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_USECS_IRQ | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_drvinfo = ef4_ethtool_get_drvinfo, .get_regs_len = ef4_ethtool_get_regs_len, .get_regs = ef4_ethtool_get_regs, diff --git a/drivers/net/ethernet/sfc/falcon/falcon_boards.c b/drivers/net/ethernet/sfc/falcon/falcon_boards.c index 605f486fa675..729a05c1b0cf 100644 --- a/drivers/net/ethernet/sfc/falcon/falcon_boards.c +++ b/drivers/net/ethernet/sfc/falcon/falcon_boards.c @@ -88,11 +88,11 @@ static int ef4_init_lm87(struct ef4_nic *efx, const struct i2c_board_info *info, const u8 *reg_values) { struct falcon_board *board = falcon_board(efx); - struct i2c_client *client = i2c_new_device(&board->i2c_adap, info); + struct i2c_client *client = i2c_new_client_device(&board->i2c_adap, info); int rc; - if (!client) - return -EIO; + if (IS_ERR(client)) + return PTR_ERR(client); /* Read-to-clear alarm/interrupt status */ i2c_smbus_read_byte_data(client, LM87_REG_ALARMS1); diff --git a/drivers/net/ethernet/sfc/falcon/net_driver.h b/drivers/net/ethernet/sfc/falcon/net_driver.h index a49ea2e719b6..a529ff395ead 100644 --- a/drivers/net/ethernet/sfc/falcon/net_driver.h +++ b/drivers/net/ethernet/sfc/falcon/net_driver.h @@ -288,7 +288,7 @@ struct ef4_rx_buffer { struct ef4_rx_page_state { dma_addr_t dma_addr; - unsigned int __pad[0] ____cacheline_aligned; + unsigned int __pad[] ____cacheline_aligned; }; /** diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 8164f0edcbf0..b084e623b5f4 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -91,6 +91,12 @@ #define EFX_RX_BUF_ALIGNMENT 4 #endif +/* Non-standard XDP_PACKET_HEADROOM and tailroom to satisfy XDP_REDIRECT and + * still fit two standard MTU size packets into a single 4K page. + */ +#define EFX_XDP_HEADROOM 128 +#define EFX_XDP_TAILROOM SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) + /* Forward declare Precision Time Protocol (PTP) support structure. */ struct efx_ptp_data; struct hwtstamp_config; @@ -333,7 +339,7 @@ struct efx_rx_buffer { struct efx_rx_page_state { dma_addr_t dma_addr; - unsigned int __pad[0] ____cacheline_aligned; + unsigned int __pad[] ____cacheline_aligned; }; /** diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c index a2042f16babc..260352d97d9d 100644 --- a/drivers/net/ethernet/sfc/rx.c +++ b/drivers/net/ethernet/sfc/rx.c @@ -302,7 +302,7 @@ static bool efx_do_xdp(struct efx_nic *efx, struct efx_channel *channel, efx->rx_prefix_size); xdp.data = *ehp; - xdp.data_hard_start = xdp.data - XDP_PACKET_HEADROOM; + xdp.data_hard_start = xdp.data - EFX_XDP_HEADROOM; /* No support yet for XDP metadata */ xdp_set_data_meta_invalid(&xdp); diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c index ee8beb87bdc1..e10c23833515 100644 --- a/drivers/net/ethernet/sfc/rx_common.c +++ b/drivers/net/ethernet/sfc/rx_common.c @@ -412,10 +412,10 @@ static int efx_init_rx_buffers(struct efx_rx_queue *rx_queue, bool atomic) index = rx_queue->added_count & rx_queue->ptr_mask; rx_buf = efx_rx_buffer(rx_queue, index); rx_buf->dma_addr = dma_addr + efx->rx_ip_align + - XDP_PACKET_HEADROOM; + EFX_XDP_HEADROOM; rx_buf->page = page; rx_buf->page_offset = page_offset + efx->rx_ip_align + - XDP_PACKET_HEADROOM; + EFX_XDP_HEADROOM; rx_buf->len = efx->rx_dma_len; rx_buf->flags = 0; ++rx_queue->added_count; @@ -433,7 +433,7 @@ static int efx_init_rx_buffers(struct efx_rx_queue *rx_queue, bool atomic) void efx_rx_config_page_split(struct efx_nic *efx) { efx->rx_page_buf_step = ALIGN(efx->rx_dma_len + efx->rx_ip_align + - XDP_PACKET_HEADROOM, + EFX_XDP_HEADROOM + EFX_XDP_TAILROOM, EFX_RX_BUF_ALIGNMENT); efx->rx_bufs_per_page = efx->rx_buffer_order ? 1 : ((PAGE_SIZE - sizeof(struct efx_rx_page_state)) / diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c index 8aafc54a4684..19b58563cb78 100644 --- a/drivers/net/ethernet/sfc/tx.c +++ b/drivers/net/ethernet/sfc/tx.c @@ -287,9 +287,8 @@ static int efx_tx_tso_fallback(struct efx_tx_queue *tx_queue, return PTR_ERR(segments); dev_consume_skb_any(skb); - skb = segments; - skb_list_walk_safe(skb, skb, next) { + skb_list_walk_safe(segments, skb, next) { skb_mark_not_on_list(skb); efx_enqueue_skb(tx_queue, skb); } diff --git a/drivers/net/ethernet/sgi/ioc3-eth.c b/drivers/net/ethernet/sgi/ioc3-eth.c index db6b2988e632..7305e8e86c51 100644 --- a/drivers/net/ethernet/sgi/ioc3-eth.c +++ b/drivers/net/ethernet/sgi/ioc3-eth.c @@ -582,40 +582,23 @@ static void ioc3_timer(struct timer_list *t) /* Try to find a PHY. There is no apparent relation between the MII addresses * in the SGI documentation and what we find in reality, so we simply probe - * for the PHY. It seems IOC3 PHYs usually live on address 31. One of my - * onboard IOC3s has the special oddity that probing doesn't seem to find it - * yet the interface seems to work fine, so if probing fails we for now will - * simply default to PHY 31 instead of bailing out. + * for the PHY. */ static int ioc3_mii_init(struct ioc3_private *ip) { - int ioc3_phy_workaround = 1; - int i, found = 0, res = 0; u16 word; + int i; for (i = 0; i < 32; i++) { word = ioc3_mdio_read(ip->mii.dev, i, MII_PHYSID1); if (word != 0xffff && word != 0x0000) { - found = 1; - break; /* Found a PHY */ + ip->mii.phy_id = i; + return 0; } } - - if (!found) { - if (ioc3_phy_workaround) { - i = 31; - } else { - ip->mii.phy_id = -1; - res = -ENODEV; - goto out; - } - } - - ip->mii.phy_id = i; - -out: - return res; + ip->mii.phy_id = -1; + return -ENODEV; } static void ioc3_mii_start(struct ioc3_private *ip) diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c index e8224b543dfc..a5a0fb60193a 100644 --- a/drivers/net/ethernet/socionext/netsec.c +++ b/drivers/net/ethernet/socionext/netsec.c @@ -589,6 +589,8 @@ static void netsec_et_set_msglevel(struct net_device *dev, u32 datum) } static const struct ethtool_ops netsec_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = netsec_et_get_drvinfo, .get_link_ksettings = phy_ethtool_get_link_ksettings, .set_link_ksettings = phy_ethtool_set_link_ksettings, @@ -896,9 +898,9 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, case XDP_TX: ret = netsec_xdp_xmit_back(priv, xdp); if (ret != NETSEC_XDP_TX) - __page_pool_put_page(dring->page_pool, - virt_to_head_page(xdp->data), - len, true); + page_pool_put_page(dring->page_pool, + virt_to_head_page(xdp->data), len, + true); break; case XDP_REDIRECT: err = xdp_do_redirect(priv->ndev, xdp, prog); @@ -906,9 +908,9 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, ret = NETSEC_XDP_REDIR; } else { ret = NETSEC_XDP_CONSUMED; - __page_pool_put_page(dring->page_pool, - virt_to_head_page(xdp->data), - len, true); + page_pool_put_page(dring->page_pool, + virt_to_head_page(xdp->data), len, + true); } break; default: @@ -919,9 +921,8 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, /* fall through -- handle aborts by dropping packet */ case XDP_DROP: ret = NETSEC_XDP_CONSUMED; - __page_pool_put_page(dring->page_pool, - virt_to_head_page(xdp->data), - len, true); + page_pool_put_page(dring->page_pool, + virt_to_head_page(xdp->data), len, true); break; } @@ -1020,8 +1021,8 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) * cache state. Since we paid the allocation cost if * building an skb fails try to put the page into cache */ - __page_pool_put_page(dring->page_pool, page, - pkt_len, true); + page_pool_put_page(dring->page_pool, page, pkt_len, + true); netif_err(priv, drv, priv->ndev, "rx failed to build skb\n"); break; @@ -1148,11 +1149,7 @@ static netdev_tx_t netsec_netdev_start_xmit(struct sk_buff *skb, ~tcp_v4_check(0, ip_hdr(skb)->saddr, ip_hdr(skb)->daddr, 0); } else { - ipv6_hdr(skb)->payload_len = 0; - tcp_hdr(skb)->check = - ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - 0, IPPROTO_TCP, 0); + tcp_v6_gso_csum_prep(skb); } tx_ctrl.tcp_seg_offload_flag = true; @@ -1199,7 +1196,7 @@ static void netsec_uninit_pkt_dring(struct netsec_priv *priv, int id) if (id == NETSEC_RING_RX) { struct page *page = virt_to_page(desc->addr); - page_pool_put_page(dring->page_pool, page, false); + page_pool_put_full_page(dring->page_pool, page, false); } else if (id == NETSEC_RING_TX) { dma_unmap_single(priv->dev, desc->dma_addr, desc->len, DMA_TO_DEVICE); diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index 338e25a6374e..b46f8d2ae6d7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -3,6 +3,7 @@ config STMMAC_ETH tristate "STMicroelectronics Multi-Gigabit Ethernet driver" depends on HAS_IOMEM && HAS_DMA select MII + select MDIO_XPCS select PAGE_POOL select PHYLINK select CRC32 @@ -197,6 +198,15 @@ config DWMAC_SUN8I EMAC ethernet controller. endif +config DWMAC_INTEL + tristate "Intel GMAC support" + default X86 + depends on X86 && STMMAC_ETH && PCI + depends on COMMON_CLK + ---help--- + This selects the Intel platform specific bus support for the + stmmac driver. This driver is used for Intel Quark/EHL/TGL. + config STMMAC_PCI tristate "STMMAC PCI bus support" depends on STMMAC_ETH && PCI diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile index c59926d96bcc..5a6f265bc540 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Makefile +++ b/drivers/net/ethernet/stmicro/stmmac/Makefile @@ -30,5 +30,6 @@ obj-$(CONFIG_DWMAC_GENERIC) += dwmac-generic.o stmmac-platform-objs:= stmmac_platform.o dwmac-altr-socfpga-objs := altr_tse_pcs.o dwmac-socfpga.o +obj-$(CONFIG_DWMAC_INTEL) += dwmac-intel.o obj-$(CONFIG_STMMAC_PCI) += stmmac-pci.o stmmac-pci-objs:= stmmac_pci.o diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 487099092693..6208a68a331d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -15,6 +15,7 @@ #include <linux/netdevice.h> #include <linux/stmmac.h> #include <linux/phy.h> +#include <linux/mdio-xpcs.h> #include <linux/module.h> #if IS_ENABLED(CONFIG_VLAN_8021Q) #define STMMAC_VLAN_TAG_USED @@ -33,6 +34,11 @@ #define DWMAC_CORE_5_00 0x50 #define DWMAC_CORE_5_10 0x51 #define DWXGMAC_CORE_2_10 0x21 +#define DWXLGMAC_CORE_2_00 0x20 + +/* Device ID */ +#define DWXGMAC_ID 0x76 +#define DWXLGMAC_ID 0x27 #define STMMAC_CHAN0 0 /* Always supported and default for all chips */ @@ -425,6 +431,12 @@ struct mac_link { u32 speed5000; u32 speed10000; } xgmii; + struct { + u32 speed25000; + u32 speed40000; + u32 speed50000; + u32 speed100000; + } xlgmii; }; struct mii_regs { @@ -446,6 +458,8 @@ struct mac_device_info { const struct stmmac_hwtimestamp *ptp; const struct stmmac_tc_ops *tc; const struct stmmac_mmc_ops *mmc; + const struct mdio_xpcs_ops *xpcs; + struct mdio_xpcs_args xpcs_args; struct mii_regs mii; /* MII register Addresses */ struct mac_link link; void __iomem *pcsr; /* vpointer to device CSRs */ @@ -456,6 +470,9 @@ struct mac_device_info { unsigned int pcs; unsigned int pmt; unsigned int ps; + unsigned int xlgmac; + unsigned int num_vlan; + u32 vlan_filter[32]; }; struct stmmac_rx_routing { @@ -467,6 +484,7 @@ int dwmac100_setup(struct stmmac_priv *priv); int dwmac1000_setup(struct stmmac_priv *priv); int dwmac4_setup(struct stmmac_priv *priv); int dwxgmac2_setup(struct stmmac_priv *priv); +int dwxlgmac2_setup(struct stmmac_priv *priv); void stmmac_set_mac_addr(void __iomem *ioaddr, u8 addr[6], unsigned int high, unsigned int low); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c new file mode 100644 index 000000000000..5419d4e478c0 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -0,0 +1,592 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020, Intel Corporation + */ + +#include <linux/clk-provider.h> +#include <linux/pci.h> +#include <linux/dmi.h> +#include "stmmac.h" + +/* This struct is used to associate PCI Function of MAC controller on a board, + * discovered via DMI, with the address of PHY connected to the MAC. The + * negative value of the address means that MAC controller is not connected + * with PHY. + */ +struct stmmac_pci_func_data { + unsigned int func; + int phy_addr; +}; + +struct stmmac_pci_dmi_data { + const struct stmmac_pci_func_data *func; + size_t nfuncs; +}; + +struct stmmac_pci_info { + int (*setup)(struct pci_dev *pdev, struct plat_stmmacenet_data *plat); +}; + +static int stmmac_pci_find_phy_addr(struct pci_dev *pdev, + const struct dmi_system_id *dmi_list) +{ + const struct stmmac_pci_func_data *func_data; + const struct stmmac_pci_dmi_data *dmi_data; + const struct dmi_system_id *dmi_id; + int func = PCI_FUNC(pdev->devfn); + size_t n; + + dmi_id = dmi_first_match(dmi_list); + if (!dmi_id) + return -ENODEV; + + dmi_data = dmi_id->driver_data; + func_data = dmi_data->func; + + for (n = 0; n < dmi_data->nfuncs; n++, func_data++) + if (func_data->func == func) + return func_data->phy_addr; + + return -ENODEV; +} + +static void common_default_data(struct plat_stmmacenet_data *plat) +{ + plat->clk_csr = 2; /* clk_csr_i = 20-35MHz & MDC = clk_csr_i/16 */ + plat->has_gmac = 1; + plat->force_sf_dma_mode = 1; + + plat->mdio_bus_data->needs_reset = true; + + /* Set default value for multicast hash bins */ + plat->multicast_filter_bins = HASH_TABLE_SIZE; + + /* Set default value for unicast filter entries */ + plat->unicast_filter_entries = 1; + + /* Set the maxmtu to a default of JUMBO_LEN */ + plat->maxmtu = JUMBO_LEN; + + /* Set default number of RX and TX queues to use */ + plat->tx_queues_to_use = 1; + plat->rx_queues_to_use = 1; + + /* Disable Priority config by default */ + plat->tx_queues_cfg[0].use_prio = false; + plat->rx_queues_cfg[0].use_prio = false; + + /* Disable RX queues routing by default */ + plat->rx_queues_cfg[0].pkt_route = 0x0; +} + +static int intel_mgbe_common_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + int i; + + plat->clk_csr = 5; + plat->has_gmac = 0; + plat->has_gmac4 = 1; + plat->force_sf_dma_mode = 0; + plat->tso_en = 1; + + plat->rx_sched_algorithm = MTL_RX_ALGORITHM_SP; + + for (i = 0; i < plat->rx_queues_to_use; i++) { + plat->rx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; + plat->rx_queues_cfg[i].chan = i; + + /* Disable Priority config by default */ + plat->rx_queues_cfg[i].use_prio = false; + + /* Disable RX queues routing by default */ + plat->rx_queues_cfg[i].pkt_route = 0x0; + } + + for (i = 0; i < plat->tx_queues_to_use; i++) { + plat->tx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; + + /* Disable Priority config by default */ + plat->tx_queues_cfg[i].use_prio = false; + } + + /* FIFO size is 4096 bytes for 1 tx/rx queue */ + plat->tx_fifo_size = plat->tx_queues_to_use * 4096; + plat->rx_fifo_size = plat->rx_queues_to_use * 4096; + + plat->tx_sched_algorithm = MTL_TX_ALGORITHM_WRR; + plat->tx_queues_cfg[0].weight = 0x09; + plat->tx_queues_cfg[1].weight = 0x0A; + plat->tx_queues_cfg[2].weight = 0x0B; + plat->tx_queues_cfg[3].weight = 0x0C; + plat->tx_queues_cfg[4].weight = 0x0D; + plat->tx_queues_cfg[5].weight = 0x0E; + plat->tx_queues_cfg[6].weight = 0x0F; + plat->tx_queues_cfg[7].weight = 0x10; + + plat->dma_cfg->pbl = 32; + plat->dma_cfg->pblx8 = true; + plat->dma_cfg->fixed_burst = 0; + plat->dma_cfg->mixed_burst = 0; + plat->dma_cfg->aal = 0; + + plat->axi = devm_kzalloc(&pdev->dev, sizeof(*plat->axi), + GFP_KERNEL); + if (!plat->axi) + return -ENOMEM; + + plat->axi->axi_lpi_en = 0; + plat->axi->axi_xit_frm = 0; + plat->axi->axi_wr_osr_lmt = 1; + plat->axi->axi_rd_osr_lmt = 1; + plat->axi->axi_blen[0] = 4; + plat->axi->axi_blen[1] = 8; + plat->axi->axi_blen[2] = 16; + + plat->ptp_max_adj = plat->clk_ptp_rate; + + /* Set system clock */ + plat->stmmac_clk = clk_register_fixed_rate(&pdev->dev, + "stmmac-clk", NULL, 0, + plat->clk_ptp_rate); + + if (IS_ERR(plat->stmmac_clk)) { + dev_warn(&pdev->dev, "Fail to register stmmac-clk\n"); + plat->stmmac_clk = NULL; + } + clk_prepare_enable(plat->stmmac_clk); + + /* Set default value for multicast hash bins */ + plat->multicast_filter_bins = HASH_TABLE_SIZE; + + /* Set default value for unicast filter entries */ + plat->unicast_filter_entries = 1; + + /* Set the maxmtu to a default of JUMBO_LEN */ + plat->maxmtu = JUMBO_LEN; + + return 0; +} + +static int ehl_common_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + int ret; + + plat->rx_queues_to_use = 8; + plat->tx_queues_to_use = 8; + plat->clk_ptp_rate = 200000000; + ret = intel_mgbe_common_data(pdev, plat); + if (ret) + return ret; + + return 0; +} + +static int ehl_sgmii_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->bus_id = 1; + plat->phy_addr = 0; + plat->phy_interface = PHY_INTERFACE_MODE_SGMII; + + return ehl_common_data(pdev, plat); +} + +static struct stmmac_pci_info ehl_sgmii1g_pci_info = { + .setup = ehl_sgmii_data, +}; + +static int ehl_rgmii_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->bus_id = 1; + plat->phy_addr = 0; + plat->phy_interface = PHY_INTERFACE_MODE_RGMII; + + return ehl_common_data(pdev, plat); +} + +static struct stmmac_pci_info ehl_rgmii1g_pci_info = { + .setup = ehl_rgmii_data, +}; + +static int ehl_pse0_common_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->bus_id = 2; + plat->phy_addr = 1; + return ehl_common_data(pdev, plat); +} + +static int ehl_pse0_rgmii1g_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->phy_interface = PHY_INTERFACE_MODE_RGMII_ID; + return ehl_pse0_common_data(pdev, plat); +} + +static struct stmmac_pci_info ehl_pse0_rgmii1g_pci_info = { + .setup = ehl_pse0_rgmii1g_data, +}; + +static int ehl_pse0_sgmii1g_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->phy_interface = PHY_INTERFACE_MODE_SGMII; + return ehl_pse0_common_data(pdev, plat); +} + +static struct stmmac_pci_info ehl_pse0_sgmii1g_pci_info = { + .setup = ehl_pse0_sgmii1g_data, +}; + +static int ehl_pse1_common_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->bus_id = 3; + plat->phy_addr = 1; + return ehl_common_data(pdev, plat); +} + +static int ehl_pse1_rgmii1g_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->phy_interface = PHY_INTERFACE_MODE_RGMII_ID; + return ehl_pse1_common_data(pdev, plat); +} + +static struct stmmac_pci_info ehl_pse1_rgmii1g_pci_info = { + .setup = ehl_pse1_rgmii1g_data, +}; + +static int ehl_pse1_sgmii1g_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->phy_interface = PHY_INTERFACE_MODE_SGMII; + return ehl_pse1_common_data(pdev, plat); +} + +static struct stmmac_pci_info ehl_pse1_sgmii1g_pci_info = { + .setup = ehl_pse1_sgmii1g_data, +}; + +static int tgl_common_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + int ret; + + plat->rx_queues_to_use = 6; + plat->tx_queues_to_use = 4; + plat->clk_ptp_rate = 200000000; + ret = intel_mgbe_common_data(pdev, plat); + if (ret) + return ret; + + return 0; +} + +static int tgl_sgmii_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + plat->bus_id = 1; + plat->phy_addr = 0; + plat->phy_interface = PHY_INTERFACE_MODE_SGMII; + return tgl_common_data(pdev, plat); +} + +static struct stmmac_pci_info tgl_sgmii1g_pci_info = { + .setup = tgl_sgmii_data, +}; + +static const struct stmmac_pci_func_data galileo_stmmac_func_data[] = { + { + .func = 6, + .phy_addr = 1, + }, +}; + +static const struct stmmac_pci_dmi_data galileo_stmmac_dmi_data = { + .func = galileo_stmmac_func_data, + .nfuncs = ARRAY_SIZE(galileo_stmmac_func_data), +}; + +static const struct stmmac_pci_func_data iot2040_stmmac_func_data[] = { + { + .func = 6, + .phy_addr = 1, + }, + { + .func = 7, + .phy_addr = 1, + }, +}; + +static const struct stmmac_pci_dmi_data iot2040_stmmac_dmi_data = { + .func = iot2040_stmmac_func_data, + .nfuncs = ARRAY_SIZE(iot2040_stmmac_func_data), +}; + +static const struct dmi_system_id quark_pci_dmi[] = { + { + .matches = { + DMI_EXACT_MATCH(DMI_BOARD_NAME, "Galileo"), + }, + .driver_data = (void *)&galileo_stmmac_dmi_data, + }, + { + .matches = { + DMI_EXACT_MATCH(DMI_BOARD_NAME, "GalileoGen2"), + }, + .driver_data = (void *)&galileo_stmmac_dmi_data, + }, + /* There are 2 types of SIMATIC IOT2000: IOT2020 and IOT2040. + * The asset tag "6ES7647-0AA00-0YA2" is only for IOT2020 which + * has only one pci network device while other asset tags are + * for IOT2040 which has two. + */ + { + .matches = { + DMI_EXACT_MATCH(DMI_BOARD_NAME, "SIMATIC IOT2000"), + DMI_EXACT_MATCH(DMI_BOARD_ASSET_TAG, + "6ES7647-0AA00-0YA2"), + }, + .driver_data = (void *)&galileo_stmmac_dmi_data, + }, + { + .matches = { + DMI_EXACT_MATCH(DMI_BOARD_NAME, "SIMATIC IOT2000"), + }, + .driver_data = (void *)&iot2040_stmmac_dmi_data, + }, + {} +}; + +static int quark_default_data(struct pci_dev *pdev, + struct plat_stmmacenet_data *plat) +{ + int ret; + + /* Set common default data first */ + common_default_data(plat); + + /* Refuse to load the driver and register net device if MAC controller + * does not connect to any PHY interface. + */ + ret = stmmac_pci_find_phy_addr(pdev, quark_pci_dmi); + if (ret < 0) { + /* Return error to the caller on DMI enabled boards. */ + if (dmi_get_system_info(DMI_BOARD_NAME)) + return ret; + + /* Galileo boards with old firmware don't support DMI. We always + * use 1 here as PHY address, so at least the first found MAC + * controller would be probed. + */ + ret = 1; + } + + plat->bus_id = pci_dev_id(pdev); + plat->phy_addr = ret; + plat->phy_interface = PHY_INTERFACE_MODE_RMII; + + plat->dma_cfg->pbl = 16; + plat->dma_cfg->pblx8 = true; + plat->dma_cfg->fixed_burst = 1; + /* AXI (TODO) */ + + return 0; +} + +static const struct stmmac_pci_info quark_pci_info = { + .setup = quark_default_data, +}; + +/** + * intel_eth_pci_probe + * + * @pdev: pci device pointer + * @id: pointer to table of device id/id's. + * + * Description: This probing function gets called for all PCI devices which + * match the ID table and are not "owned" by other driver yet. This function + * gets passed a "struct pci_dev *" for each device whose entry in the ID table + * matches the device. The probe functions returns zero when the driver choose + * to take "ownership" of the device or an error code(-ve no) otherwise. + */ +static int intel_eth_pci_probe(struct pci_dev *pdev, + const struct pci_device_id *id) +{ + struct stmmac_pci_info *info = (struct stmmac_pci_info *)id->driver_data; + struct plat_stmmacenet_data *plat; + struct stmmac_resources res; + int i; + int ret; + + plat = devm_kzalloc(&pdev->dev, sizeof(*plat), GFP_KERNEL); + if (!plat) + return -ENOMEM; + + plat->mdio_bus_data = devm_kzalloc(&pdev->dev, + sizeof(*plat->mdio_bus_data), + GFP_KERNEL); + if (!plat->mdio_bus_data) + return -ENOMEM; + + plat->dma_cfg = devm_kzalloc(&pdev->dev, sizeof(*plat->dma_cfg), + GFP_KERNEL); + if (!plat->dma_cfg) + return -ENOMEM; + + /* Enable pci device */ + ret = pci_enable_device(pdev); + if (ret) { + dev_err(&pdev->dev, "%s: ERROR: failed to enable device\n", + __func__); + return ret; + } + + /* Get the base address of device */ + for (i = 0; i < PCI_STD_NUM_BARS; i++) { + if (pci_resource_len(pdev, i) == 0) + continue; + ret = pcim_iomap_regions(pdev, BIT(i), pci_name(pdev)); + if (ret) + return ret; + break; + } + + pci_set_master(pdev); + + ret = info->setup(pdev, plat); + if (ret) + return ret; + + pci_enable_msi(pdev); + + memset(&res, 0, sizeof(res)); + res.addr = pcim_iomap_table(pdev)[i]; + res.wol_irq = pdev->irq; + res.irq = pdev->irq; + + return stmmac_dvr_probe(&pdev->dev, plat, &res); +} + +/** + * intel_eth_pci_remove + * + * @pdev: platform device pointer + * Description: this function calls the main to free the net resources + * and releases the PCI resources. + */ +static void intel_eth_pci_remove(struct pci_dev *pdev) +{ + struct net_device *ndev = dev_get_drvdata(&pdev->dev); + struct stmmac_priv *priv = netdev_priv(ndev); + int i; + + stmmac_dvr_remove(&pdev->dev); + + if (priv->plat->stmmac_clk) + clk_unregister_fixed_rate(priv->plat->stmmac_clk); + + for (i = 0; i < PCI_STD_NUM_BARS; i++) { + if (pci_resource_len(pdev, i) == 0) + continue; + pcim_iounmap_regions(pdev, BIT(i)); + break; + } + + pci_disable_device(pdev); +} + +static int __maybe_unused intel_eth_pci_suspend(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + int ret; + + ret = stmmac_suspend(dev); + if (ret) + return ret; + + ret = pci_save_state(pdev); + if (ret) + return ret; + + pci_disable_device(pdev); + pci_wake_from_d3(pdev, true); + return 0; +} + +static int __maybe_unused intel_eth_pci_resume(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + int ret; + + pci_restore_state(pdev); + pci_set_power_state(pdev, PCI_D0); + + ret = pci_enable_device(pdev); + if (ret) + return ret; + + pci_set_master(pdev); + + return stmmac_resume(dev); +} + +static SIMPLE_DEV_PM_OPS(intel_eth_pm_ops, intel_eth_pci_suspend, + intel_eth_pci_resume); + +#define PCI_DEVICE_ID_INTEL_QUARK_ID 0x0937 +#define PCI_DEVICE_ID_INTEL_EHL_RGMII1G_ID 0x4b30 +#define PCI_DEVICE_ID_INTEL_EHL_SGMII1G_ID 0x4b31 +#define PCI_DEVICE_ID_INTEL_EHL_SGMII2G5_ID 0x4b32 +/* Intel(R) Programmable Services Engine (Intel(R) PSE) consist of 2 MAC + * which are named PSE0 and PSE1 + */ +#define PCI_DEVICE_ID_INTEL_EHL_PSE0_RGMII1G_ID 0x4ba0 +#define PCI_DEVICE_ID_INTEL_EHL_PSE0_SGMII1G_ID 0x4ba1 +#define PCI_DEVICE_ID_INTEL_EHL_PSE0_SGMII2G5_ID 0x4ba2 +#define PCI_DEVICE_ID_INTEL_EHL_PSE1_RGMII1G_ID 0x4bb0 +#define PCI_DEVICE_ID_INTEL_EHL_PSE1_SGMII1G_ID 0x4bb1 +#define PCI_DEVICE_ID_INTEL_EHL_PSE1_SGMII2G5_ID 0x4bb2 +#define PCI_DEVICE_ID_INTEL_TGL_SGMII1G_ID 0xa0ac + +static const struct pci_device_id intel_eth_pci_id_table[] = { + { PCI_DEVICE_DATA(INTEL, QUARK_ID, &quark_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_RGMII1G_ID, &ehl_rgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_SGMII1G_ID, &ehl_sgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_SGMII2G5_ID, &ehl_sgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE0_RGMII1G_ID, + &ehl_pse0_rgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE0_SGMII1G_ID, + &ehl_pse0_sgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE0_SGMII2G5_ID, + &ehl_pse0_sgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE1_RGMII1G_ID, + &ehl_pse1_rgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE1_SGMII1G_ID, + &ehl_pse1_sgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE1_SGMII2G5_ID, + &ehl_pse1_sgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, TGL_SGMII1G_ID, &tgl_sgmii1g_pci_info) }, + {} +}; + +MODULE_DEVICE_TABLE(pci, intel_eth_pci_id_table); + +static struct pci_driver intel_eth_pci_driver = { + .name = "intel-eth-pci", + .id_table = intel_eth_pci_id_table, + .probe = intel_eth_pci_probe, + .remove = intel_eth_pci_remove, + .driver = { + .pm = &intel_eth_pm_ops, + }, +}; + +module_pci_driver(intel_eth_pci_driver); + +MODULE_DESCRIPTION("INTEL 10/100/1000 Ethernet PCI driver"); +MODULE_AUTHOR("Voon Weifeng <weifeng.voon@intel.com>"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c index 9b7be996d07b..b2dc99289687 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c @@ -304,7 +304,7 @@ static int stm32mp1_parse_data(struct stm32_dwmac *dwmac, /* Get ETH_CLK clocks */ dwmac->clk_eth_ck = devm_clk_get(dev, "eth-ck"); if (IS_ERR(dwmac->clk_eth_ck)) { - dev_warn(dev, "No phy clock provided...\n"); + dev_info(dev, "No phy clock provided...\n"); dwmac->clk_eth_ck = NULL; } @@ -324,7 +324,7 @@ static int stm32mp1_parse_data(struct stm32_dwmac *dwmac, /* Get IRQ information early to have an ability to ask for deferred * probe if needed before we went too far with resource allocation. */ - dwmac->irq_pwr_wakeup = platform_get_irq_byname(pdev, + dwmac->irq_pwr_wakeup = platform_get_irq_byname_optional(pdev, "stm32_pwr_wakeup"); if (dwmac->irq_pwr_wakeup == -EPROBE_DEFER) return -EPROBE_DEFER; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index af50af27550b..28cac28253b8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -18,6 +18,7 @@ #define GMAC_PACKET_FILTER 0x00000008 #define GMAC_HASH_TAB(x) (0x10 + (x) * 4) #define GMAC_VLAN_TAG 0x00000050 +#define GMAC_VLAN_TAG_DATA 0x00000054 #define GMAC_VLAN_HASH_TABLE 0x00000058 #define GMAC_RX_FLOW_CTRL 0x00000090 #define GMAC_VLAN_INCL 0x00000060 @@ -90,6 +91,29 @@ #define GMAC_VLAN_VLC GENMASK(17, 16) #define GMAC_VLAN_VLC_SHIFT 16 +/* MAC VLAN Tag */ +#define GMAC_VLAN_TAG_VID GENMASK(15, 0) +#define GMAC_VLAN_TAG_ETV BIT(16) + +/* MAC VLAN Tag Control */ +#define GMAC_VLAN_TAG_CTRL_OB BIT(0) +#define GMAC_VLAN_TAG_CTRL_CT BIT(1) +#define GMAC_VLAN_TAG_CTRL_OFS_MASK GENMASK(6, 2) +#define GMAC_VLAN_TAG_CTRL_OFS_SHIFT 2 +#define GMAC_VLAN_TAG_CTRL_EVLS_MASK GENMASK(22, 21) +#define GMAC_VLAN_TAG_CTRL_EVLS_SHIFT 21 +#define GMAC_VLAN_TAG_CTRL_EVLRXS BIT(24) + +#define GMAC_VLAN_TAG_STRIP_NONE (0x0 << GMAC_VLAN_TAG_CTRL_EVLS_SHIFT) +#define GMAC_VLAN_TAG_STRIP_PASS (0x1 << GMAC_VLAN_TAG_CTRL_EVLS_SHIFT) +#define GMAC_VLAN_TAG_STRIP_FAIL (0x2 << GMAC_VLAN_TAG_CTRL_EVLS_SHIFT) +#define GMAC_VLAN_TAG_STRIP_ALL (0x3 << GMAC_VLAN_TAG_CTRL_EVLS_SHIFT) + +/* MAC VLAN Tag Data/Filter */ +#define GMAC_VLAN_TAG_DATA_VID GENMASK(15, 0) +#define GMAC_VLAN_TAG_DATA_VEN BIT(16) +#define GMAC_VLAN_TAG_DATA_ETV BIT(17) + /* MAC RX Queue Enable */ #define GMAC_RX_QUEUE_CLEAR(queue) ~(GENMASK(1, 0) << ((queue) * 2)) #define GMAC_RX_AV_QUEUE_ENABLE(queue) BIT((queue) * 2) @@ -248,6 +272,7 @@ enum power_event { #define GMAC_HW_FEAT_FRPBS GENMASK(12, 11) #define GMAC_HW_FEAT_FRPSEL BIT(10) #define GMAC_HW_FEAT_DVLAN BIT(5) +#define GMAC_HW_FEAT_NRVF GENMASK(2, 0) /* MAC HW ADDR regs */ #define GMAC_HI_DCS GENMASK(18, 16) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index dc09d2131e40..39692d15d80c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -394,6 +394,156 @@ static void dwmac4_set_eee_timer(struct mac_device_info *hw, int ls, int tw) writel(value, ioaddr + GMAC4_LPI_TIMER_CTRL); } +static void dwmac4_write_single_vlan(struct net_device *dev, u16 vid) +{ + void __iomem *ioaddr = (void __iomem *)dev->base_addr; + u32 val; + + val = readl(ioaddr + GMAC_VLAN_TAG); + val &= ~GMAC_VLAN_TAG_VID; + val |= GMAC_VLAN_TAG_ETV | vid; + + writel(val, ioaddr + GMAC_VLAN_TAG); +} + +static int dwmac4_write_vlan_filter(struct net_device *dev, + struct mac_device_info *hw, + u8 index, u32 data) +{ + void __iomem *ioaddr = (void __iomem *)dev->base_addr; + int i, timeout = 10; + u32 val; + + if (index >= hw->num_vlan) + return -EINVAL; + + writel(data, ioaddr + GMAC_VLAN_TAG_DATA); + + val = readl(ioaddr + GMAC_VLAN_TAG); + val &= ~(GMAC_VLAN_TAG_CTRL_OFS_MASK | + GMAC_VLAN_TAG_CTRL_CT | + GMAC_VLAN_TAG_CTRL_OB); + val |= (index << GMAC_VLAN_TAG_CTRL_OFS_SHIFT) | GMAC_VLAN_TAG_CTRL_OB; + + writel(val, ioaddr + GMAC_VLAN_TAG); + + for (i = 0; i < timeout; i++) { + val = readl(ioaddr + GMAC_VLAN_TAG); + if (!(val & GMAC_VLAN_TAG_CTRL_OB)) + return 0; + udelay(1); + } + + netdev_err(dev, "Timeout accessing MAC_VLAN_Tag_Filter\n"); + + return -EBUSY; +} + +static int dwmac4_add_hw_vlan_rx_fltr(struct net_device *dev, + struct mac_device_info *hw, + __be16 proto, u16 vid) +{ + int index = -1; + u32 val = 0; + int i, ret; + + if (vid > 4095) + return -EINVAL; + + /* Single Rx VLAN Filter */ + if (hw->num_vlan == 1) { + /* For single VLAN filter, VID 0 means VLAN promiscuous */ + if (vid == 0) { + netdev_warn(dev, "Adding VLAN ID 0 is not supported\n"); + return -EPERM; + } + + if (hw->vlan_filter[0] & GMAC_VLAN_TAG_VID) { + netdev_err(dev, "Only single VLAN ID supported\n"); + return -EPERM; + } + + hw->vlan_filter[0] = vid; + dwmac4_write_single_vlan(dev, vid); + + return 0; + } + + /* Extended Rx VLAN Filter Enable */ + val |= GMAC_VLAN_TAG_DATA_ETV | GMAC_VLAN_TAG_DATA_VEN | vid; + + for (i = 0; i < hw->num_vlan; i++) { + if (hw->vlan_filter[i] == val) + return 0; + else if (!(hw->vlan_filter[i] & GMAC_VLAN_TAG_DATA_VEN)) + index = i; + } + + if (index == -1) { + netdev_err(dev, "MAC_VLAN_Tag_Filter full (size: %0u)\n", + hw->num_vlan); + return -EPERM; + } + + ret = dwmac4_write_vlan_filter(dev, hw, index, val); + + if (!ret) + hw->vlan_filter[index] = val; + + return ret; +} + +static int dwmac4_del_hw_vlan_rx_fltr(struct net_device *dev, + struct mac_device_info *hw, + __be16 proto, u16 vid) +{ + int i, ret = 0; + + /* Single Rx VLAN Filter */ + if (hw->num_vlan == 1) { + if ((hw->vlan_filter[0] & GMAC_VLAN_TAG_VID) == vid) { + hw->vlan_filter[0] = 0; + dwmac4_write_single_vlan(dev, 0); + } + return 0; + } + + /* Extended Rx VLAN Filter Enable */ + for (i = 0; i < hw->num_vlan; i++) { + if ((hw->vlan_filter[i] & GMAC_VLAN_TAG_DATA_VID) == vid) { + ret = dwmac4_write_vlan_filter(dev, hw, i, 0); + + if (!ret) + hw->vlan_filter[i] = 0; + else + return ret; + } + } + + return ret; +} + +static void dwmac4_restore_hw_vlan_rx_fltr(struct net_device *dev, + struct mac_device_info *hw) +{ + u32 val; + int i; + + /* Single Rx VLAN Filter */ + if (hw->num_vlan == 1) { + dwmac4_write_single_vlan(dev, hw->vlan_filter[0]); + return; + } + + /* Extended Rx VLAN Filter Enable */ + for (i = 0; i < hw->num_vlan; i++) { + if (hw->vlan_filter[i] & GMAC_VLAN_TAG_DATA_VEN) { + val = hw->vlan_filter[i]; + dwmac4_write_vlan_filter(dev, hw, i, val); + } + } +} + static void dwmac4_set_filter(struct mac_device_info *hw, struct net_device *dev) { @@ -469,6 +619,10 @@ static void dwmac4_set_filter(struct mac_device_info *hw, } } + /* VLAN filtering */ + if (dev->features & NETIF_F_HW_VLAN_CTAG_FILTER) + value |= GMAC_PACKET_FILTER_VTFE; + writel(value, ioaddr + GMAC_PACKET_FILTER); } @@ -947,6 +1101,9 @@ const struct stmmac_ops dwmac4_ops = { .set_arp_offload = dwmac4_set_arp_offload, .config_l3_filter = dwmac4_config_l3_filter, .config_l4_filter = dwmac4_config_l4_filter, + .add_hw_vlan_rx_fltr = dwmac4_add_hw_vlan_rx_fltr, + .del_hw_vlan_rx_fltr = dwmac4_del_hw_vlan_rx_fltr, + .restore_hw_vlan_rx_fltr = dwmac4_restore_hw_vlan_rx_fltr, }; const struct stmmac_ops dwmac410_ops = { @@ -987,6 +1144,9 @@ const struct stmmac_ops dwmac410_ops = { .config_l4_filter = dwmac4_config_l4_filter, .est_configure = dwmac5_est_configure, .fpe_configure = dwmac5_fpe_configure, + .add_hw_vlan_rx_fltr = dwmac4_add_hw_vlan_rx_fltr, + .del_hw_vlan_rx_fltr = dwmac4_del_hw_vlan_rx_fltr, + .restore_hw_vlan_rx_fltr = dwmac4_restore_hw_vlan_rx_fltr, }; const struct stmmac_ops dwmac510_ops = { @@ -1032,8 +1192,42 @@ const struct stmmac_ops dwmac510_ops = { .config_l4_filter = dwmac4_config_l4_filter, .est_configure = dwmac5_est_configure, .fpe_configure = dwmac5_fpe_configure, + .add_hw_vlan_rx_fltr = dwmac4_add_hw_vlan_rx_fltr, + .del_hw_vlan_rx_fltr = dwmac4_del_hw_vlan_rx_fltr, + .restore_hw_vlan_rx_fltr = dwmac4_restore_hw_vlan_rx_fltr, }; +static u32 dwmac4_get_num_vlan(void __iomem *ioaddr) +{ + u32 val, num_vlan; + + val = readl(ioaddr + GMAC_HW_FEATURE3); + switch (val & GMAC_HW_FEAT_NRVF) { + case 0: + num_vlan = 1; + break; + case 1: + num_vlan = 4; + break; + case 2: + num_vlan = 8; + break; + case 3: + num_vlan = 16; + break; + case 4: + num_vlan = 24; + break; + case 5: + num_vlan = 32; + break; + default: + num_vlan = 1; + } + + return num_vlan; +} + int dwmac4_setup(struct stmmac_priv *priv) { struct mac_device_info *mac = priv->hw; @@ -1062,6 +1256,7 @@ int dwmac4_setup(struct stmmac_priv *priv) mac->mii.reg_mask = GENMASK(20, 16); mac->mii.clk_csr_shift = 8; mac->mii.clk_csr_mask = GENMASK(11, 8); + mac->num_vlan = dwmac4_get_num_vlan(priv->ioaddr); return 0; } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c index 9becca280074..6e30d7eb4983 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c @@ -6,6 +6,7 @@ */ #include <linux/io.h> +#include <linux/iopoll.h> #include <linux/delay.h> #include "common.h" #include "dwmac4_dma.h" @@ -14,22 +15,14 @@ int dwmac4_dma_reset(void __iomem *ioaddr) { u32 value = readl(ioaddr + DMA_BUS_MODE); - int limit; /* DMA SW reset */ value |= DMA_BUS_MODE_SFT_RESET; writel(value, ioaddr + DMA_BUS_MODE); - limit = 10; - while (limit--) { - if (!(readl(ioaddr + DMA_BUS_MODE) & DMA_BUS_MODE_SFT_RESET)) - break; - mdelay(10); - } - - if (limit < 0) - return -EBUSY; - return 0; + return readl_poll_timeout(ioaddr + DMA_BUS_MODE, value, + !(value & DMA_BUS_MODE_SFT_RESET), + 10000, 100000); } void dwmac4_set_rx_tail_ptr(void __iomem *ioaddr, u32 tail_ptr, u32 chan) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c index 688d36095333..cb87d31a99df 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c @@ -16,19 +16,14 @@ int dwmac_dma_reset(void __iomem *ioaddr) { u32 value = readl(ioaddr + DMA_BUS_MODE); - int err; /* DMA SW reset */ value |= DMA_BUS_MODE_SFT_RESET; writel(value, ioaddr + DMA_BUS_MODE); - err = readl_poll_timeout(ioaddr + DMA_BUS_MODE, value, + return readl_poll_timeout(ioaddr + DMA_BUS_MODE, value, !(value & DMA_BUS_MODE_SFT_RESET), 10000, 100000); - if (err) - return -EBUSY; - - return 0; } /* CSR1 enables the transmit DMA to check for new descriptor */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c index 67b754a56288..0e4575f7bedb 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c @@ -9,6 +9,7 @@ #include <linux/iopoll.h> #include "stmmac.h" #include "stmmac_ptp.h" +#include "dwxlgmac2.h" #include "dwxgmac2.h" static void dwxgmac2_core_init(struct mac_device_info *hw, @@ -1485,6 +1486,67 @@ const struct stmmac_ops dwxgmac210_ops = { .fpe_configure = dwxgmac3_fpe_configure, }; +static void dwxlgmac2_rx_queue_enable(struct mac_device_info *hw, u8 mode, + u32 queue) +{ + void __iomem *ioaddr = hw->pcsr; + u32 value; + + value = readl(ioaddr + XLGMAC_RXQ_ENABLE_CTRL0) & ~XGMAC_RXQEN(queue); + if (mode == MTL_QUEUE_AVB) + value |= 0x1 << XGMAC_RXQEN_SHIFT(queue); + else if (mode == MTL_QUEUE_DCB) + value |= 0x2 << XGMAC_RXQEN_SHIFT(queue); + writel(value, ioaddr + XLGMAC_RXQ_ENABLE_CTRL0); +} + +const struct stmmac_ops dwxlgmac2_ops = { + .core_init = dwxgmac2_core_init, + .set_mac = dwxgmac2_set_mac, + .rx_ipc = dwxgmac2_rx_ipc, + .rx_queue_enable = dwxlgmac2_rx_queue_enable, + .rx_queue_prio = dwxgmac2_rx_queue_prio, + .tx_queue_prio = dwxgmac2_tx_queue_prio, + .rx_queue_routing = NULL, + .prog_mtl_rx_algorithms = dwxgmac2_prog_mtl_rx_algorithms, + .prog_mtl_tx_algorithms = dwxgmac2_prog_mtl_tx_algorithms, + .set_mtl_tx_queue_weight = dwxgmac2_set_mtl_tx_queue_weight, + .map_mtl_to_dma = dwxgmac2_map_mtl_to_dma, + .config_cbs = dwxgmac2_config_cbs, + .dump_regs = dwxgmac2_dump_regs, + .host_irq_status = dwxgmac2_host_irq_status, + .host_mtl_irq_status = dwxgmac2_host_mtl_irq_status, + .flow_ctrl = dwxgmac2_flow_ctrl, + .pmt = dwxgmac2_pmt, + .set_umac_addr = dwxgmac2_set_umac_addr, + .get_umac_addr = dwxgmac2_get_umac_addr, + .set_eee_mode = dwxgmac2_set_eee_mode, + .reset_eee_mode = dwxgmac2_reset_eee_mode, + .set_eee_timer = dwxgmac2_set_eee_timer, + .set_eee_pls = dwxgmac2_set_eee_pls, + .pcs_ctrl_ane = NULL, + .pcs_rane = NULL, + .pcs_get_adv_lp = NULL, + .debug = NULL, + .set_filter = dwxgmac2_set_filter, + .safety_feat_config = dwxgmac3_safety_feat_config, + .safety_feat_irq_status = dwxgmac3_safety_feat_irq_status, + .safety_feat_dump = dwxgmac3_safety_feat_dump, + .set_mac_loopback = dwxgmac2_set_mac_loopback, + .rss_configure = dwxgmac2_rss_configure, + .update_vlan_hash = dwxgmac2_update_vlan_hash, + .rxp_config = dwxgmac3_rxp_config, + .get_mac_tx_timestamp = dwxgmac2_get_mac_tx_timestamp, + .flex_pps_config = dwxgmac2_flex_pps_config, + .sarc_configure = dwxgmac2_sarc_configure, + .enable_vlan = dwxgmac2_enable_vlan, + .config_l3_filter = dwxgmac2_config_l3_filter, + .config_l4_filter = dwxgmac2_config_l4_filter, + .set_arp_offload = dwxgmac2_set_arp_offload, + .est_configure = dwxgmac3_est_configure, + .fpe_configure = dwxgmac3_fpe_configure, +}; + int dwxgmac2_setup(struct stmmac_priv *priv) { struct mac_device_info *mac = priv->hw; @@ -1521,3 +1583,40 @@ int dwxgmac2_setup(struct stmmac_priv *priv) return 0; } + +int dwxlgmac2_setup(struct stmmac_priv *priv) +{ + struct mac_device_info *mac = priv->hw; + + dev_info(priv->device, "\tXLGMAC\n"); + + priv->dev->priv_flags |= IFF_UNICAST_FLT; + mac->pcsr = priv->ioaddr; + mac->multicast_filter_bins = priv->plat->multicast_filter_bins; + mac->unicast_filter_entries = priv->plat->unicast_filter_entries; + mac->mcast_bits_log2 = 0; + + if (mac->multicast_filter_bins) + mac->mcast_bits_log2 = ilog2(mac->multicast_filter_bins); + + mac->link.duplex = 0; + mac->link.speed1000 = XLGMAC_CONFIG_SS_1000; + mac->link.speed2500 = XLGMAC_CONFIG_SS_2500; + mac->link.xgmii.speed10000 = XLGMAC_CONFIG_SS_10G; + mac->link.xlgmii.speed25000 = XLGMAC_CONFIG_SS_25G; + mac->link.xlgmii.speed40000 = XLGMAC_CONFIG_SS_40G; + mac->link.xlgmii.speed50000 = XLGMAC_CONFIG_SS_50G; + mac->link.xlgmii.speed100000 = XLGMAC_CONFIG_SS_100G; + mac->link.speed_mask = XLGMAC_CONFIG_SS; + + mac->mii.addr = XGMAC_MDIO_ADDR; + mac->mii.data = XGMAC_MDIO_DATA; + mac->mii.addr_shift = 16; + mac->mii.addr_mask = GENMASK(20, 16); + mac->mii.reg_shift = 0; + mac->mii.reg_mask = GENMASK(15, 0); + mac->mii.clk_csr_shift = 19; + mac->mii.clk_csr_mask = GENMASK(21, 19); + + return 0; +} diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxlgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxlgmac2.h new file mode 100644 index 000000000000..726090d49221 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/dwxlgmac2.h @@ -0,0 +1,22 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2020 Synopsys, Inc. and/or its affiliates. + * Synopsys DesignWare XLGMAC definitions. + */ + +#ifndef __STMMAC_DWXLGMAC2_H__ +#define __STMMAC_DWXLGMAC2_H__ + +/* MAC Registers */ +#define XLGMAC_CONFIG_SS GENMASK(30, 28) +#define XLGMAC_CONFIG_SS_SHIFT 28 +#define XLGMAC_CONFIG_SS_40G (0x0 << XLGMAC_CONFIG_SS_SHIFT) +#define XLGMAC_CONFIG_SS_25G (0x1 << XLGMAC_CONFIG_SS_SHIFT) +#define XLGMAC_CONFIG_SS_50G (0x2 << XLGMAC_CONFIG_SS_SHIFT) +#define XLGMAC_CONFIG_SS_100G (0x3 << XLGMAC_CONFIG_SS_SHIFT) +#define XLGMAC_CONFIG_SS_10G (0x4 << XLGMAC_CONFIG_SS_SHIFT) +#define XLGMAC_CONFIG_SS_2500 (0x6 << XLGMAC_CONFIG_SS_SHIFT) +#define XLGMAC_CONFIG_SS_1000 (0x7 << XLGMAC_CONFIG_SS_SHIFT) +#define XLGMAC_RXQ_ENABLE_CTRL0 0x00000140 + +#endif /* __STMMAC_DWXLGMAC2_H__ */ diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index 3af2e5015245..bb7114f970f8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -23,6 +23,18 @@ static u32 stmmac_get_id(struct stmmac_priv *priv, u32 id_reg) return reg & GENMASK(7, 0); } +static u32 stmmac_get_dev_id(struct stmmac_priv *priv, u32 id_reg) +{ + u32 reg = readl(priv->ioaddr + id_reg); + + if (!reg) { + dev_info(priv->device, "Version ID not available\n"); + return 0x0; + } + + return (reg & GENMASK(15, 8)) >> 8; +} + static void stmmac_dwmac_mode_quirk(struct stmmac_priv *priv) { struct mac_device_info *mac = priv->hw; @@ -69,11 +81,18 @@ static int stmmac_dwmac4_quirks(struct stmmac_priv *priv) return 0; } +static int stmmac_dwxlgmac_quirks(struct stmmac_priv *priv) +{ + priv->hw->xlgmac = true; + return 0; +} + static const struct stmmac_hwif_entry { bool gmac; bool gmac4; bool xgmac; u32 min_id; + u32 dev_id; const struct stmmac_regs_off regs; const void *desc; const void *dma; @@ -199,6 +218,7 @@ static const struct stmmac_hwif_entry { .gmac4 = false, .xgmac = true, .min_id = DWXGMAC_CORE_2_10, + .dev_id = DWXGMAC_ID, .regs = { .ptp_off = PTP_XGMAC_OFFSET, .mmc_off = MMC_XGMAC_OFFSET, @@ -212,6 +232,25 @@ static const struct stmmac_hwif_entry { .mmc = &dwxgmac_mmc_ops, .setup = dwxgmac2_setup, .quirks = NULL, + }, { + .gmac = false, + .gmac4 = false, + .xgmac = true, + .min_id = DWXLGMAC_CORE_2_00, + .dev_id = DWXLGMAC_ID, + .regs = { + .ptp_off = PTP_XGMAC_OFFSET, + .mmc_off = MMC_XGMAC_OFFSET, + }, + .desc = &dwxgmac210_desc_ops, + .dma = &dwxgmac210_dma_ops, + .mac = &dwxlgmac2_ops, + .hwtimestamp = &stmmac_ptp, + .mode = NULL, + .tc = &dwmac510_tc_ops, + .mmc = &dwxgmac_mmc_ops, + .setup = dwxlgmac2_setup, + .quirks = stmmac_dwxlgmac_quirks, }, }; @@ -223,13 +262,15 @@ int stmmac_hwif_init(struct stmmac_priv *priv) const struct stmmac_hwif_entry *entry; struct mac_device_info *mac; bool needs_setup = true; + u32 id, dev_id = 0; int i, ret; - u32 id; if (needs_gmac) { id = stmmac_get_id(priv, GMAC_VERSION); } else if (needs_gmac4 || needs_xgmac) { id = stmmac_get_id(priv, GMAC4_VERSION); + if (needs_xgmac) + dev_id = stmmac_get_dev_id(priv, GMAC4_VERSION); } else { id = 0; } @@ -267,6 +308,8 @@ int stmmac_hwif_init(struct stmmac_priv *priv) /* Use synopsys_id var because some setups can override this */ if (priv->synopsys_id < entry->min_id) continue; + if (needs_xgmac && (dev_id ^ entry->dev_id)) + continue; /* Only use generic HW helpers if needed */ mac->desc = mac->desc ? : entry->desc; diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index df63b0367aff..ffe2d63389b8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -369,6 +369,14 @@ struct stmmac_ops { void (*update_vlan_hash)(struct mac_device_info *hw, u32 hash, __le16 perfect_match, bool is_double); void (*enable_vlan)(struct mac_device_info *hw, u32 type); + int (*add_hw_vlan_rx_fltr)(struct net_device *dev, + struct mac_device_info *hw, + __be16 proto, u16 vid); + int (*del_hw_vlan_rx_fltr)(struct net_device *dev, + struct mac_device_info *hw, + __be16 proto, u16 vid); + void (*restore_hw_vlan_rx_fltr)(struct net_device *dev, + struct mac_device_info *hw); /* TX Timestamp */ int (*get_mac_tx_timestamp)(struct mac_device_info *hw, u64 *ts); /* Source Address Insertion / Replacement */ @@ -461,6 +469,12 @@ struct stmmac_ops { stmmac_do_void_callback(__priv, mac, update_vlan_hash, __args) #define stmmac_enable_vlan(__priv, __args...) \ stmmac_do_void_callback(__priv, mac, enable_vlan, __args) +#define stmmac_add_hw_vlan_rx_fltr(__priv, __args...) \ + stmmac_do_callback(__priv, mac, add_hw_vlan_rx_fltr, __args) +#define stmmac_del_hw_vlan_rx_fltr(__priv, __args...) \ + stmmac_do_callback(__priv, mac, del_hw_vlan_rx_fltr, __args) +#define stmmac_restore_hw_vlan_rx_fltr(__priv, __args...) \ + stmmac_do_void_callback(__priv, mac, restore_hw_vlan_rx_fltr, __args) #define stmmac_get_mac_tx_timestamp(__priv, __args...) \ stmmac_do_callback(__priv, mac, get_mac_tx_timestamp, __args) #define stmmac_sarc_configure(__priv, __args...) \ @@ -577,6 +591,18 @@ struct stmmac_mmc_ops { #define stmmac_mmc_read(__priv, __args...) \ stmmac_do_void_callback(__priv, mmc, read, __args) +/* XPCS callbacks */ +#define stmmac_xpcs_validate(__priv, __args...) \ + stmmac_do_callback(__priv, xpcs, validate, __args) +#define stmmac_xpcs_config(__priv, __args...) \ + stmmac_do_callback(__priv, xpcs, config, __args) +#define stmmac_xpcs_get_state(__priv, __args...) \ + stmmac_do_callback(__priv, xpcs, get_state, __args) +#define stmmac_xpcs_link_up(__priv, __args...) \ + stmmac_do_callback(__priv, xpcs, link_up, __args) +#define stmmac_xpcs_probe(__priv, __args...) \ + stmmac_do_callback(__priv, xpcs, probe, __args) + struct stmmac_regs_off { u32 ptp_off; u32 mmc_off; @@ -593,6 +619,7 @@ extern const struct stmmac_dma_ops dwmac410_dma_ops; extern const struct stmmac_ops dwmac510_ops; extern const struct stmmac_tc_ops dwmac510_tc_ops; extern const struct stmmac_ops dwxgmac210_ops; +extern const struct stmmac_ops dwxlgmac2_ops; extern const struct stmmac_dma_ops dwxgmac210_dma_ops; extern const struct stmmac_desc_ops dwxgmac210_desc_ops; extern const struct stmmac_mmc_ops dwmac_mmc_ops; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index b29603ec744c..eae11c585025 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -732,20 +732,6 @@ static int stmmac_set_coalesce(struct net_device *dev, u32 rx_cnt = priv->plat->rx_queues_to_use; unsigned int rx_riwt; - /* Check not supported parameters */ - if ((ec->rx_coalesce_usecs_irq) || - (ec->rx_max_coalesced_frames_irq) || (ec->tx_coalesce_usecs_irq) || - (ec->use_adaptive_rx_coalesce) || (ec->use_adaptive_tx_coalesce) || - (ec->pkt_rate_low) || (ec->rx_coalesce_usecs_low) || - (ec->rx_max_coalesced_frames_low) || (ec->tx_coalesce_usecs_high) || - (ec->tx_max_coalesced_frames_low) || (ec->pkt_rate_high) || - (ec->tx_coalesce_usecs_low) || (ec->rx_coalesce_usecs_high) || - (ec->rx_max_coalesced_frames_high) || - (ec->tx_max_coalesced_frames_irq) || - (ec->stats_block_coalesce_usecs) || - (ec->tx_max_coalesced_frames_high) || (ec->rate_sample_interval)) - return -EOPNOTSUPP; - if (priv->use_riwt && (ec->rx_coalesce_usecs > 0)) { rx_riwt = stmmac_usec2riwt(ec->rx_coalesce_usecs, priv); @@ -914,6 +900,8 @@ static int stmmac_set_tunable(struct net_device *dev, } static const struct ethtool_ops stmmac_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .begin = stmmac_check_if_running, .get_drvinfo = stmmac_ethtool_getdrvinfo, .get_msglevel = stmmac_ethtool_getmsglevel, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c index 020159622559..fcf080243a0f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c @@ -10,6 +10,7 @@ *******************************************************************************/ #include <linux/io.h> +#include <linux/iopoll.h> #include <linux/delay.h> #include "common.h" #include "stmmac_ptp.h" @@ -53,7 +54,6 @@ static void config_sub_second_increment(void __iomem *ioaddr, static int init_systime(void __iomem *ioaddr, u32 sec, u32 nsec) { - int limit; u32 value; writel(sec, ioaddr + PTP_STSUR); @@ -64,16 +64,9 @@ static int init_systime(void __iomem *ioaddr, u32 sec, u32 nsec) writel(value, ioaddr + PTP_TCR); /* wait for present system time initialize to complete */ - limit = 10; - while (limit--) { - if (!(readl(ioaddr + PTP_TCR) & PTP_TCR_TSINIT)) - break; - mdelay(10); - } - if (limit < 0) - return -EBUSY; - - return 0; + return readl_poll_timeout(ioaddr + PTP_TCR, value, + !(value & PTP_TCR_TSINIT), + 10000, 100000); } static int config_addend(void __iomem *ioaddr, u32 addend) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 7da18c9afa01..2fb671e61ee8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -849,6 +849,38 @@ static void stmmac_validate(struct phylink_config *config, phylink_set(mac_supported, 10000baseKX4_Full); phylink_set(mac_supported, 10000baseKR_Full); } + if (!max_speed || (max_speed >= 25000)) { + phylink_set(mac_supported, 25000baseCR_Full); + phylink_set(mac_supported, 25000baseKR_Full); + phylink_set(mac_supported, 25000baseSR_Full); + } + if (!max_speed || (max_speed >= 40000)) { + phylink_set(mac_supported, 40000baseKR4_Full); + phylink_set(mac_supported, 40000baseCR4_Full); + phylink_set(mac_supported, 40000baseSR4_Full); + phylink_set(mac_supported, 40000baseLR4_Full); + } + if (!max_speed || (max_speed >= 50000)) { + phylink_set(mac_supported, 50000baseCR2_Full); + phylink_set(mac_supported, 50000baseKR2_Full); + phylink_set(mac_supported, 50000baseSR2_Full); + phylink_set(mac_supported, 50000baseKR_Full); + phylink_set(mac_supported, 50000baseSR_Full); + phylink_set(mac_supported, 50000baseCR_Full); + phylink_set(mac_supported, 50000baseLR_ER_FR_Full); + phylink_set(mac_supported, 50000baseDR_Full); + } + if (!max_speed || (max_speed >= 100000)) { + phylink_set(mac_supported, 100000baseKR4_Full); + phylink_set(mac_supported, 100000baseSR4_Full); + phylink_set(mac_supported, 100000baseCR4_Full); + phylink_set(mac_supported, 100000baseLR4_ER4_Full); + phylink_set(mac_supported, 100000baseKR2_Full); + phylink_set(mac_supported, 100000baseSR2_Full); + phylink_set(mac_supported, 100000baseCR2_Full); + phylink_set(mac_supported, 100000baseLR2_ER2_FR2_Full); + phylink_set(mac_supported, 100000baseDR2_Full); + } } /* Half-Duplex can only work with single queue */ @@ -858,33 +890,65 @@ static void stmmac_validate(struct phylink_config *config, phylink_set(mask, 1000baseT_Half); } - bitmap_and(supported, supported, mac_supported, - __ETHTOOL_LINK_MODE_MASK_NBITS); - bitmap_andnot(supported, supported, mask, - __ETHTOOL_LINK_MODE_MASK_NBITS); - bitmap_and(state->advertising, state->advertising, mac_supported, - __ETHTOOL_LINK_MODE_MASK_NBITS); - bitmap_andnot(state->advertising, state->advertising, mask, - __ETHTOOL_LINK_MODE_MASK_NBITS); + linkmode_and(supported, supported, mac_supported); + linkmode_andnot(supported, supported, mask); + + linkmode_and(state->advertising, state->advertising, mac_supported); + linkmode_andnot(state->advertising, state->advertising, mask); + + /* If PCS is supported, check which modes it supports. */ + stmmac_xpcs_validate(priv, &priv->hw->xpcs_args, supported, state); } static void stmmac_mac_pcs_get_state(struct phylink_config *config, struct phylink_link_state *state) { + struct stmmac_priv *priv = netdev_priv(to_net_dev(config->dev)); + state->link = 0; + stmmac_xpcs_get_state(priv, &priv->hw->xpcs_args, state); } static void stmmac_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { struct stmmac_priv *priv = netdev_priv(to_net_dev(config->dev)); + + stmmac_xpcs_config(priv, &priv->hw->xpcs_args, state); +} + +static void stmmac_mac_an_restart(struct phylink_config *config) +{ + /* Not Supported */ +} + +static void stmmac_mac_link_down(struct phylink_config *config, + unsigned int mode, phy_interface_t interface) +{ + struct stmmac_priv *priv = netdev_priv(to_net_dev(config->dev)); + + stmmac_mac_set(priv, priv->ioaddr, false); + priv->eee_active = false; + stmmac_eee_init(priv); + stmmac_set_eee_pls(priv, priv->hw, false); +} + +static void stmmac_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause) +{ + struct stmmac_priv *priv = netdev_priv(to_net_dev(config->dev)); u32 ctrl; + stmmac_xpcs_link_up(priv, &priv->hw->xpcs_args, speed, interface); + ctrl = readl(priv->ioaddr + MAC_CTRL_REG); ctrl &= ~priv->hw->link.speed_mask; - if (state->interface == PHY_INTERFACE_MODE_USXGMII) { - switch (state->speed) { + if (interface == PHY_INTERFACE_MODE_USXGMII) { + switch (speed) { case SPEED_10000: ctrl |= priv->hw->link.xgmii.speed10000; break; @@ -897,8 +961,34 @@ static void stmmac_mac_config(struct phylink_config *config, unsigned int mode, default: return; } + } else if (interface == PHY_INTERFACE_MODE_XLGMII) { + switch (speed) { + case SPEED_100000: + ctrl |= priv->hw->link.xlgmii.speed100000; + break; + case SPEED_50000: + ctrl |= priv->hw->link.xlgmii.speed50000; + break; + case SPEED_40000: + ctrl |= priv->hw->link.xlgmii.speed40000; + break; + case SPEED_25000: + ctrl |= priv->hw->link.xlgmii.speed25000; + break; + case SPEED_10000: + ctrl |= priv->hw->link.xgmii.speed10000; + break; + case SPEED_2500: + ctrl |= priv->hw->link.speed2500; + break; + case SPEED_1000: + ctrl |= priv->hw->link.speed1000; + break; + default: + return; + } } else { - switch (state->speed) { + switch (speed) { case SPEED_2500: ctrl |= priv->hw->link.speed2500; break; @@ -916,44 +1006,21 @@ static void stmmac_mac_config(struct phylink_config *config, unsigned int mode, } } - priv->speed = state->speed; + priv->speed = speed; if (priv->plat->fix_mac_speed) - priv->plat->fix_mac_speed(priv->plat->bsp_priv, state->speed); + priv->plat->fix_mac_speed(priv->plat->bsp_priv, speed); - if (!state->duplex) + if (!duplex) ctrl &= ~priv->hw->link.duplex; else ctrl |= priv->hw->link.duplex; /* Flow Control operation */ - if (state->pause) - stmmac_mac_flow_ctrl(priv, state->duplex); + if (tx_pause && rx_pause) + stmmac_mac_flow_ctrl(priv, duplex); writel(ctrl, priv->ioaddr + MAC_CTRL_REG); -} - -static void stmmac_mac_an_restart(struct phylink_config *config) -{ - /* Not Supported */ -} - -static void stmmac_mac_link_down(struct phylink_config *config, - unsigned int mode, phy_interface_t interface) -{ - struct stmmac_priv *priv = netdev_priv(to_net_dev(config->dev)); - - stmmac_mac_set(priv, priv->ioaddr, false); - priv->eee_active = false; - stmmac_eee_init(priv); - stmmac_set_eee_pls(priv, priv->hw, false); -} - -static void stmmac_mac_link_up(struct phylink_config *config, - unsigned int mode, phy_interface_t interface, - struct phy_device *phy) -{ - struct stmmac_priv *priv = netdev_priv(to_net_dev(config->dev)); stmmac_mac_set(priv, priv->ioaddr, true); if (phy && priv->dma_cap.eee) { @@ -1043,6 +1110,10 @@ static int stmmac_phy_setup(struct stmmac_priv *priv) priv->phylink_config.dev = &priv->dev->dev; priv->phylink_config.type = PHYLINK_NETDEV; + priv->phylink_config.pcs_poll = true; + + if (!fwnode) + fwnode = dev_fwnode(priv->device); phylink = phylink_create(&priv->phylink_config, fwnode, mode, &stmmac_phylink_mac_ops); @@ -1251,11 +1322,11 @@ static void stmmac_free_rx_buffer(struct stmmac_priv *priv, u32 queue, int i) struct stmmac_rx_buffer *buf = &rx_q->buf_pool[i]; if (buf->page) - page_pool_put_page(rx_q->page_pool, buf->page, false); + page_pool_put_full_page(rx_q->page_pool, buf->page, false); buf->page = NULL; if (buf->sec_page) - page_pool_put_page(rx_q->page_pool, buf->sec_page, false); + page_pool_put_full_page(rx_q->page_pool, buf->sec_page, false); buf->sec_page = NULL; } @@ -2687,7 +2758,8 @@ static int stmmac_open(struct net_device *dev) int ret; if (priv->hw->pcs != STMMAC_PCS_TBI && - priv->hw->pcs != STMMAC_PCS_RTBI) { + priv->hw->pcs != STMMAC_PCS_RTBI && + priv->hw->xpcs == NULL) { ret = stmmac_init_phy(dev); if (ret) { netdev_err(priv->dev, @@ -4494,6 +4566,8 @@ static int stmmac_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid return ret; } + ret = stmmac_add_hw_vlan_rx_fltr(priv, ndev, priv->hw, proto, vid); + return ret; } @@ -4501,11 +4575,16 @@ static int stmmac_vlan_rx_kill_vid(struct net_device *ndev, __be16 proto, u16 vi { struct stmmac_priv *priv = netdev_priv(ndev); bool is_double = false; + int ret; if (be16_to_cpu(proto) == ETH_P_8021AD) is_double = true; clear_bit(vid, priv->active_vlans); + ret = stmmac_del_hw_vlan_rx_fltr(priv, ndev, priv->hw, proto, vid); + if (ret) + return ret; + return stmmac_vlan_update(priv, is_double); } @@ -5096,6 +5175,8 @@ int stmmac_resume(struct device *dev) stmmac_init_coalesce(priv); stmmac_set_rx_mode(ndev); + stmmac_restore_hw_vlan_rx_fltr(priv, ndev, priv->hw); + stmmac_enable_all_queues(priv); stmmac_start_all_queues(priv); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index cfe5d8b73142..b2a707e2ef43 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -382,6 +382,14 @@ int stmmac_mdio_register(struct net_device *ndev) max_addr = PHY_MAX_ADDR; } + if (mdio_bus_data->has_xpcs) { + priv->hw->xpcs = mdio_xpcs_get_ops(); + if (!priv->hw->xpcs) { + err = -ENODEV; + goto bus_register_fail; + } + } + if (mdio_bus_data->needs_reset) new_bus->reset = &stmmac_mdio_reset; @@ -433,6 +441,25 @@ int stmmac_mdio_register(struct net_device *ndev) found = 1; } + /* Try to probe the XPCS by scanning all addresses. */ + if (priv->hw->xpcs) { + struct mdio_xpcs_args *xpcs = &priv->hw->xpcs_args; + int ret, mode = priv->plat->phy_interface; + max_addr = PHY_MAX_ADDR; + + xpcs->bus = new_bus; + + for (addr = 0; addr < max_addr; addr++) { + xpcs->addr = addr; + + ret = stmmac_xpcs_probe(priv, xpcs, mode); + if (!ret) { + found = 1; + break; + } + } + } + if (!found && !mdio_node) { dev_warn(dev, "No PHY found\n"); mdiobus_unregister(new_bus); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index fe2c9fa6a71c..3fb21f7ac9fb 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -15,49 +15,10 @@ #include "stmmac.h" -/* - * This struct is used to associate PCI Function of MAC controller on a board, - * discovered via DMI, with the address of PHY connected to the MAC. The - * negative value of the address means that MAC controller is not connected - * with PHY. - */ -struct stmmac_pci_func_data { - unsigned int func; - int phy_addr; -}; - -struct stmmac_pci_dmi_data { - const struct stmmac_pci_func_data *func; - size_t nfuncs; -}; - struct stmmac_pci_info { int (*setup)(struct pci_dev *pdev, struct plat_stmmacenet_data *plat); }; -static int stmmac_pci_find_phy_addr(struct pci_dev *pdev, - const struct dmi_system_id *dmi_list) -{ - const struct stmmac_pci_func_data *func_data; - const struct stmmac_pci_dmi_data *dmi_data; - const struct dmi_system_id *dmi_id; - int func = PCI_FUNC(pdev->devfn); - size_t n; - - dmi_id = dmi_first_match(dmi_list); - if (!dmi_id) - return -ENODEV; - - dmi_data = dmi_id->driver_data; - func_data = dmi_data->func; - - for (n = 0; n < dmi_data->nfuncs; n++, func_data++) - if (func_data->func == func) - return func_data->phy_addr; - - return -ENODEV; -} - static void common_default_data(struct plat_stmmacenet_data *plat) { plat->clk_csr = 2; /* clk_csr_i = 20-35MHz & MDC = clk_csr_i/16 */ @@ -108,272 +69,6 @@ static const struct stmmac_pci_info stmmac_pci_info = { .setup = stmmac_default_data, }; -static int intel_mgbe_common_data(struct pci_dev *pdev, - struct plat_stmmacenet_data *plat) -{ - int i; - - plat->clk_csr = 5; - plat->has_gmac = 0; - plat->has_gmac4 = 1; - plat->force_sf_dma_mode = 0; - plat->tso_en = 1; - - plat->rx_sched_algorithm = MTL_RX_ALGORITHM_SP; - - for (i = 0; i < plat->rx_queues_to_use; i++) { - plat->rx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; - plat->rx_queues_cfg[i].chan = i; - - /* Disable Priority config by default */ - plat->rx_queues_cfg[i].use_prio = false; - - /* Disable RX queues routing by default */ - plat->rx_queues_cfg[i].pkt_route = 0x0; - } - - for (i = 0; i < plat->tx_queues_to_use; i++) { - plat->tx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; - - /* Disable Priority config by default */ - plat->tx_queues_cfg[i].use_prio = false; - } - - /* FIFO size is 4096 bytes for 1 tx/rx queue */ - plat->tx_fifo_size = plat->tx_queues_to_use * 4096; - plat->rx_fifo_size = plat->rx_queues_to_use * 4096; - - plat->tx_sched_algorithm = MTL_TX_ALGORITHM_WRR; - plat->tx_queues_cfg[0].weight = 0x09; - plat->tx_queues_cfg[1].weight = 0x0A; - plat->tx_queues_cfg[2].weight = 0x0B; - plat->tx_queues_cfg[3].weight = 0x0C; - plat->tx_queues_cfg[4].weight = 0x0D; - plat->tx_queues_cfg[5].weight = 0x0E; - plat->tx_queues_cfg[6].weight = 0x0F; - plat->tx_queues_cfg[7].weight = 0x10; - - plat->dma_cfg->pbl = 32; - plat->dma_cfg->pblx8 = true; - plat->dma_cfg->fixed_burst = 0; - plat->dma_cfg->mixed_burst = 0; - plat->dma_cfg->aal = 0; - - plat->axi = devm_kzalloc(&pdev->dev, sizeof(*plat->axi), - GFP_KERNEL); - if (!plat->axi) - return -ENOMEM; - - plat->axi->axi_lpi_en = 0; - plat->axi->axi_xit_frm = 0; - plat->axi->axi_wr_osr_lmt = 1; - plat->axi->axi_rd_osr_lmt = 1; - plat->axi->axi_blen[0] = 4; - plat->axi->axi_blen[1] = 8; - plat->axi->axi_blen[2] = 16; - - plat->ptp_max_adj = plat->clk_ptp_rate; - - /* Set system clock */ - plat->stmmac_clk = clk_register_fixed_rate(&pdev->dev, - "stmmac-clk", NULL, 0, - plat->clk_ptp_rate); - - if (IS_ERR(plat->stmmac_clk)) { - dev_warn(&pdev->dev, "Fail to register stmmac-clk\n"); - plat->stmmac_clk = NULL; - } - clk_prepare_enable(plat->stmmac_clk); - - /* Set default value for multicast hash bins */ - plat->multicast_filter_bins = HASH_TABLE_SIZE; - - /* Set default value for unicast filter entries */ - plat->unicast_filter_entries = 1; - - /* Set the maxmtu to a default of JUMBO_LEN */ - plat->maxmtu = JUMBO_LEN; - - return 0; -} - -static int ehl_common_data(struct pci_dev *pdev, - struct plat_stmmacenet_data *plat) -{ - int ret; - - plat->rx_queues_to_use = 8; - plat->tx_queues_to_use = 8; - plat->clk_ptp_rate = 200000000; - ret = intel_mgbe_common_data(pdev, plat); - if (ret) - return ret; - - return 0; -} - -static int ehl_sgmii_data(struct pci_dev *pdev, - struct plat_stmmacenet_data *plat) -{ - plat->bus_id = 1; - plat->phy_addr = 0; - plat->phy_interface = PHY_INTERFACE_MODE_SGMII; - - return ehl_common_data(pdev, plat); -} - -static struct stmmac_pci_info ehl_sgmii1g_pci_info = { - .setup = ehl_sgmii_data, -}; - -static int ehl_rgmii_data(struct pci_dev *pdev, - struct plat_stmmacenet_data *plat) -{ - plat->bus_id = 1; - plat->phy_addr = 0; - plat->phy_interface = PHY_INTERFACE_MODE_RGMII; - - return ehl_common_data(pdev, plat); -} - -static struct stmmac_pci_info ehl_rgmii1g_pci_info = { - .setup = ehl_rgmii_data, -}; - -static int tgl_common_data(struct pci_dev *pdev, - struct plat_stmmacenet_data *plat) -{ - int ret; - - plat->rx_queues_to_use = 6; - plat->tx_queues_to_use = 4; - plat->clk_ptp_rate = 200000000; - ret = intel_mgbe_common_data(pdev, plat); - if (ret) - return ret; - - return 0; -} - -static int tgl_sgmii_data(struct pci_dev *pdev, - struct plat_stmmacenet_data *plat) -{ - plat->bus_id = 1; - plat->phy_addr = 0; - plat->phy_interface = PHY_INTERFACE_MODE_SGMII; - return tgl_common_data(pdev, plat); -} - -static struct stmmac_pci_info tgl_sgmii1g_pci_info = { - .setup = tgl_sgmii_data, -}; - -static const struct stmmac_pci_func_data galileo_stmmac_func_data[] = { - { - .func = 6, - .phy_addr = 1, - }, -}; - -static const struct stmmac_pci_dmi_data galileo_stmmac_dmi_data = { - .func = galileo_stmmac_func_data, - .nfuncs = ARRAY_SIZE(galileo_stmmac_func_data), -}; - -static const struct stmmac_pci_func_data iot2040_stmmac_func_data[] = { - { - .func = 6, - .phy_addr = 1, - }, - { - .func = 7, - .phy_addr = 1, - }, -}; - -static const struct stmmac_pci_dmi_data iot2040_stmmac_dmi_data = { - .func = iot2040_stmmac_func_data, - .nfuncs = ARRAY_SIZE(iot2040_stmmac_func_data), -}; - -static const struct dmi_system_id quark_pci_dmi[] = { - { - .matches = { - DMI_EXACT_MATCH(DMI_BOARD_NAME, "Galileo"), - }, - .driver_data = (void *)&galileo_stmmac_dmi_data, - }, - { - .matches = { - DMI_EXACT_MATCH(DMI_BOARD_NAME, "GalileoGen2"), - }, - .driver_data = (void *)&galileo_stmmac_dmi_data, - }, - /* - * There are 2 types of SIMATIC IOT2000: IOT2020 and IOT2040. - * The asset tag "6ES7647-0AA00-0YA2" is only for IOT2020 which - * has only one pci network device while other asset tags are - * for IOT2040 which has two. - */ - { - .matches = { - DMI_EXACT_MATCH(DMI_BOARD_NAME, "SIMATIC IOT2000"), - DMI_EXACT_MATCH(DMI_BOARD_ASSET_TAG, - "6ES7647-0AA00-0YA2"), - }, - .driver_data = (void *)&galileo_stmmac_dmi_data, - }, - { - .matches = { - DMI_EXACT_MATCH(DMI_BOARD_NAME, "SIMATIC IOT2000"), - }, - .driver_data = (void *)&iot2040_stmmac_dmi_data, - }, - {} -}; - -static int quark_default_data(struct pci_dev *pdev, - struct plat_stmmacenet_data *plat) -{ - int ret; - - /* Set common default data first */ - common_default_data(plat); - - /* - * Refuse to load the driver and register net device if MAC controller - * does not connect to any PHY interface. - */ - ret = stmmac_pci_find_phy_addr(pdev, quark_pci_dmi); - if (ret < 0) { - /* Return error to the caller on DMI enabled boards. */ - if (dmi_get_system_info(DMI_BOARD_NAME)) - return ret; - - /* - * Galileo boards with old firmware don't support DMI. We always - * use 1 here as PHY address, so at least the first found MAC - * controller would be probed. - */ - ret = 1; - } - - plat->bus_id = pci_dev_id(pdev); - plat->phy_addr = ret; - plat->phy_interface = PHY_INTERFACE_MODE_RMII; - - plat->dma_cfg->pbl = 16; - plat->dma_cfg->pblx8 = true; - plat->dma_cfg->fixed_burst = 1; - /* AXI (TODO) */ - - return 0; -} - -static const struct stmmac_pci_info quark_pci_info = { - .setup = quark_default_data, -}; - static int snps_gmac5_default_data(struct pci_dev *pdev, struct plat_stmmacenet_data *plat) { @@ -579,28 +274,15 @@ static int __maybe_unused stmmac_pci_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(stmmac_pm_ops, stmmac_pci_suspend, stmmac_pci_resume); /* synthetic ID, no official vendor */ -#define PCI_VENDOR_ID_STMMAC 0x700 - -#define STMMAC_QUARK_ID 0x0937 -#define STMMAC_DEVICE_ID 0x1108 -#define STMMAC_EHL_RGMII1G_ID 0x4b30 -#define STMMAC_EHL_SGMII1G_ID 0x4b31 -#define STMMAC_TGL_SGMII1G_ID 0xa0ac -#define STMMAC_GMAC5_ID 0x7102 - -#define STMMAC_DEVICE(vendor_id, dev_id, info) { \ - PCI_VDEVICE(vendor_id, dev_id), \ - .driver_data = (kernel_ulong_t)&info \ - } +#define PCI_VENDOR_ID_STMMAC 0x0700 + +#define PCI_DEVICE_ID_STMMAC_STMMAC 0x1108 +#define PCI_DEVICE_ID_SYNOPSYS_GMAC5_ID 0x7102 static const struct pci_device_id stmmac_id_table[] = { - STMMAC_DEVICE(STMMAC, STMMAC_DEVICE_ID, stmmac_pci_info), - STMMAC_DEVICE(STMICRO, PCI_DEVICE_ID_STMICRO_MAC, stmmac_pci_info), - STMMAC_DEVICE(INTEL, STMMAC_QUARK_ID, quark_pci_info), - STMMAC_DEVICE(INTEL, STMMAC_EHL_RGMII1G_ID, ehl_rgmii1g_pci_info), - STMMAC_DEVICE(INTEL, STMMAC_EHL_SGMII1G_ID, ehl_sgmii1g_pci_info), - STMMAC_DEVICE(INTEL, STMMAC_TGL_SGMII1G_ID, tgl_sgmii1g_pci_info), - STMMAC_DEVICE(SYNOPSYS, STMMAC_GMAC5_ID, snps_gmac5_pci_info), + { PCI_DEVICE_DATA(STMMAC, STMMAC, &stmmac_pci_info) }, + { PCI_DEVICE_DATA(STMICRO, MAC, &stmmac_pci_info) }, + { PCI_DEVICE_DATA(SYNOPSYS, GMAC5_ID, &snps_gmac5_pci_info) }, {} }; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 13fafd905db8..bcda49dcf619 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -588,7 +588,7 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac) if (IS_ERR(plat->clk_ptp_ref)) { plat->clk_ptp_rate = clk_get_rate(plat->stmmac_clk); plat->clk_ptp_ref = NULL; - dev_warn(&pdev->dev, "PTP uses main clock\n"); + dev_info(&pdev->dev, "PTP uses main clock\n"); } else { plat->clk_ptp_rate = clk_get_rate(plat->clk_ptp_ref); dev_dbg(&pdev->dev, "PTP rate %d\n", plat->clk_ptp_rate); @@ -645,8 +645,6 @@ EXPORT_SYMBOL_GPL(stmmac_remove_config_dt); int stmmac_get_platform_resources(struct platform_device *pdev, struct stmmac_resources *stmmac_res) { - struct resource *res; - memset(stmmac_res, 0, sizeof(*stmmac_res)); /* Get IRQ information early to have an ability to ask for deferred @@ -680,8 +678,7 @@ int stmmac_get_platform_resources(struct platform_device *pdev, dev_info(&pdev->dev, "IRQ eth_lpi not found\n"); } - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - stmmac_res->addr = devm_ioremap_resource(&pdev->dev, res); + stmmac_res->addr = devm_platform_ioremap_resource(pdev, 0); return PTR_ERR_OR_ZERO(stmmac_res->addr); } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c index 2aba2673d6c3..e6696495f126 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c @@ -380,7 +380,7 @@ static int stmmac_test_phy_loopback(struct stmmac_priv *priv) int ret; if (!priv->dev->phydev) - return -EBUSY; + return -EOPNOTSUPP; ret = phy_loopback(priv->dev->phydev, true); if (ret) @@ -1387,6 +1387,7 @@ static int __stmmac_test_l3filt(struct stmmac_priv *priv, u32 dst, u32 src, cls->rule = rule; rule->action.entries[0].id = FLOW_ACTION_DROP; + rule->action.entries[0].hw_stats = FLOW_ACTION_HW_STATS_ANY; rule->action.num_entries = 1; attr.dst = priv->dev->dev_addr; @@ -1515,6 +1516,7 @@ static int __stmmac_test_l4filt(struct stmmac_priv *priv, u32 dst, u32 src, cls->rule = rule; rule->action.entries[0].id = FLOW_ACTION_DROP; + rule->action.entries[0].hw_stats = FLOW_ACTION_HW_STATS_ANY; rule->action.num_entries = 1; attr.dst = priv->dev->dev_addr; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c index 7a01dee2f9a8..3d747846f482 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c @@ -367,7 +367,8 @@ static int tc_setup_cbs(struct stmmac_priv *priv, static int tc_parse_flow_actions(struct stmmac_priv *priv, struct flow_action *action, - struct stmmac_flow_entry *entry) + struct stmmac_flow_entry *entry, + struct netlink_ext_ack *extack) { struct flow_action_entry *act; int i; @@ -375,6 +376,9 @@ static int tc_parse_flow_actions(struct stmmac_priv *priv, if (!flow_action_has_entries(action)) return -EINVAL; + if (!flow_action_basic_hw_stats_check(action, extack)) + return -EOPNOTSUPP; + flow_action_for_each(i, act, action) { switch (act->id) { case FLOW_ACTION_DROP: @@ -530,7 +534,8 @@ static int tc_add_flow(struct stmmac_priv *priv, return -ENOENT; } - ret = tc_parse_flow_actions(priv, &rule->action, entry); + ret = tc_parse_flow_actions(priv, &rule->action, entry, + cls->common.extack); if (ret) return ret; diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index 6ec9163e232c..e6d1aa882fa5 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -1716,34 +1716,26 @@ static int cas_pci_interrupt(struct net_device *dev, struct cas *cp, pr_cont("\n"); if (stat & PCI_ERR_OTHER) { - u16 cfg; + int pci_errs; /* Interrogate PCI config space for the * true cause. */ - pci_read_config_word(cp->pdev, PCI_STATUS, &cfg); - netdev_err(dev, "Read PCI cfg space status [%04x]\n", cfg); - if (cfg & PCI_STATUS_PARITY) + pci_errs = pci_status_get_and_clear_errors(cp->pdev); + + netdev_err(dev, "PCI status errors[%04x]\n", pci_errs); + if (pci_errs & PCI_STATUS_PARITY) netdev_err(dev, "PCI parity error detected\n"); - if (cfg & PCI_STATUS_SIG_TARGET_ABORT) + if (pci_errs & PCI_STATUS_SIG_TARGET_ABORT) netdev_err(dev, "PCI target abort\n"); - if (cfg & PCI_STATUS_REC_TARGET_ABORT) + if (pci_errs & PCI_STATUS_REC_TARGET_ABORT) netdev_err(dev, "PCI master acks target abort\n"); - if (cfg & PCI_STATUS_REC_MASTER_ABORT) + if (pci_errs & PCI_STATUS_REC_MASTER_ABORT) netdev_err(dev, "PCI master abort\n"); - if (cfg & PCI_STATUS_SIG_SYSTEM_ERROR) + if (pci_errs & PCI_STATUS_SIG_SYSTEM_ERROR) netdev_err(dev, "PCI system error SERR#\n"); - if (cfg & PCI_STATUS_DETECTED_PARITY) + if (pci_errs & PCI_STATUS_DETECTED_PARITY) netdev_err(dev, "PCI parity error\n"); - - /* Write the error bits back to clear them. */ - cfg &= (PCI_STATUS_PARITY | - PCI_STATUS_SIG_TARGET_ABORT | - PCI_STATUS_REC_TARGET_ABORT | - PCI_STATUS_REC_MASTER_ABORT | - PCI_STATUS_SIG_SYSTEM_ERROR | - PCI_STATUS_DETECTED_PARITY); - pci_write_config_word(cp->pdev, PCI_STATUS, cfg); } /* For all PCI errors, we should reset the chip. */ diff --git a/drivers/net/ethernet/sun/sungem.c b/drivers/net/ethernet/sun/sungem.c index 8358064fbd48..2d392a7b179a 100644 --- a/drivers/net/ethernet/sun/sungem.c +++ b/drivers/net/ethernet/sun/sungem.c @@ -545,37 +545,25 @@ static int gem_pci_interrupt(struct net_device *dev, struct gem *gp, u32 gem_sta } if (pci_estat & GREG_PCIESTAT_OTHER) { - u16 pci_cfg_stat; + int pci_errs; /* Interrogate PCI config space for the * true cause. */ - pci_read_config_word(gp->pdev, PCI_STATUS, - &pci_cfg_stat); - netdev_err(dev, "Read PCI cfg space status [%04x]\n", - pci_cfg_stat); - if (pci_cfg_stat & PCI_STATUS_PARITY) + pci_errs = pci_status_get_and_clear_errors(gp->pdev); + netdev_err(dev, "PCI status errors[%04x]\n", pci_errs); + if (pci_errs & PCI_STATUS_PARITY) netdev_err(dev, "PCI parity error detected\n"); - if (pci_cfg_stat & PCI_STATUS_SIG_TARGET_ABORT) + if (pci_errs & PCI_STATUS_SIG_TARGET_ABORT) netdev_err(dev, "PCI target abort\n"); - if (pci_cfg_stat & PCI_STATUS_REC_TARGET_ABORT) + if (pci_errs & PCI_STATUS_REC_TARGET_ABORT) netdev_err(dev, "PCI master acks target abort\n"); - if (pci_cfg_stat & PCI_STATUS_REC_MASTER_ABORT) + if (pci_errs & PCI_STATUS_REC_MASTER_ABORT) netdev_err(dev, "PCI master abort\n"); - if (pci_cfg_stat & PCI_STATUS_SIG_SYSTEM_ERROR) + if (pci_errs & PCI_STATUS_SIG_SYSTEM_ERROR) netdev_err(dev, "PCI system error SERR#\n"); - if (pci_cfg_stat & PCI_STATUS_DETECTED_PARITY) + if (pci_errs & PCI_STATUS_DETECTED_PARITY) netdev_err(dev, "PCI parity error\n"); - - /* Write the error bits back to clear them. */ - pci_cfg_stat &= (PCI_STATUS_PARITY | - PCI_STATUS_SIG_TARGET_ABORT | - PCI_STATUS_REC_TARGET_ABORT | - PCI_STATUS_REC_MASTER_ABORT | - PCI_STATUS_SIG_SYSTEM_ERROR | - PCI_STATUS_DETECTED_PARITY); - pci_write_config_word(gp->pdev, - PCI_STATUS, pci_cfg_stat); } /* For all PCI errors, we should reset the chip. */ diff --git a/drivers/net/ethernet/synopsys/dwc-xlgmac-ethtool.c b/drivers/net/ethernet/synopsys/dwc-xlgmac-ethtool.c index fde722136869..bc198eadfcab 100644 --- a/drivers/net/ethernet/synopsys/dwc-xlgmac-ethtool.c +++ b/drivers/net/ethernet/synopsys/dwc-xlgmac-ethtool.c @@ -151,7 +151,6 @@ static int xlgmac_ethtool_get_coalesce(struct net_device *netdev, { struct xlgmac_pdata *pdata = netdev_priv(netdev); - memset(ec, 0, sizeof(struct ethtool_coalesce)); ec->rx_coalesce_usecs = pdata->rx_usecs; ec->rx_max_coalesced_frames = pdata->rx_frames; ec->tx_max_coalesced_frames = pdata->tx_frames; @@ -167,20 +166,6 @@ static int xlgmac_ethtool_set_coalesce(struct net_device *netdev, unsigned int rx_frames, rx_riwt, rx_usecs; unsigned int tx_frames; - /* Check for not supported parameters */ - if ((ec->rx_coalesce_usecs_irq) || (ec->rx_max_coalesced_frames_irq) || - (ec->tx_coalesce_usecs) || (ec->tx_coalesce_usecs_high) || - (ec->tx_max_coalesced_frames_irq) || (ec->tx_coalesce_usecs_irq) || - (ec->stats_block_coalesce_usecs) || (ec->pkt_rate_low) || - (ec->use_adaptive_rx_coalesce) || (ec->use_adaptive_tx_coalesce) || - (ec->rx_max_coalesced_frames_low) || (ec->rx_coalesce_usecs_low) || - (ec->tx_coalesce_usecs_low) || (ec->tx_max_coalesced_frames_low) || - (ec->pkt_rate_high) || (ec->rx_coalesce_usecs_high) || - (ec->rx_max_coalesced_frames_high) || - (ec->tx_max_coalesced_frames_high) || - (ec->rate_sample_interval)) - return -EOPNOTSUPP; - rx_usecs = ec->rx_coalesce_usecs; rx_riwt = hw_ops->usec_to_riwt(pdata, rx_usecs); rx_frames = ec->rx_max_coalesced_frames; @@ -257,6 +242,8 @@ static void xlgmac_ethtool_get_ethtool_stats(struct net_device *netdev, } static const struct ethtool_ops xlgmac_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = xlgmac_ethtool_get_drvinfo, .get_link = ethtool_op_get_link, .get_msglevel = xlgmac_ethtool_get_msglevel, diff --git a/drivers/net/ethernet/tehuti/tehuti.c b/drivers/net/ethernet/tehuti/tehuti.c index 0f8a924fc60c..40a2ce0ca808 100644 --- a/drivers/net/ethernet/tehuti/tehuti.c +++ b/drivers/net/ethernet/tehuti/tehuti.c @@ -2373,6 +2373,8 @@ static void bdx_get_ethtool_stats(struct net_device *netdev, static void bdx_set_ethtool_ops(struct net_device *netdev) { static const struct ethtool_ops bdx_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = bdx_get_drvinfo, .get_link = ethtool_op_get_link, .get_coalesce = bdx_get_coalesce, diff --git a/drivers/net/ethernet/tehuti/tehuti.h b/drivers/net/ethernet/tehuti/tehuti.h index 5fc03c8eba0c..909e7296cecf 100644 --- a/drivers/net/ethernet/tehuti/tehuti.h +++ b/drivers/net/ethernet/tehuti/tehuti.h @@ -330,7 +330,7 @@ struct txd_desc { u16 length; u32 va_lo; u32 va_hi; - struct pbl pbl[0]; /* Fragments */ + struct pbl pbl[]; /* Fragments */ } __packed; /* Register region size */ diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig index bf98e0fa7d8b..89cec778cf2d 100644 --- a/drivers/net/ethernet/ti/Kconfig +++ b/drivers/net/ethernet/ti/Kconfig @@ -6,7 +6,7 @@ config NET_VENDOR_TI bool "Texas Instruments (TI) devices" default y - depends on PCI || EISA || AR7 || ARCH_DAVINCI || ARCH_OMAP2PLUS || ARCH_KEYSTONE + depends on PCI || EISA || AR7 || ARCH_DAVINCI || ARCH_OMAP2PLUS || ARCH_KEYSTONE || ARCH_K3 ---help--- If you have a network (Ethernet) card belonging to this class, say Y. @@ -31,7 +31,7 @@ config TI_DAVINCI_EMAC config TI_DAVINCI_MDIO tristate "TI DaVinci MDIO Support" - depends on ARCH_DAVINCI || ARCH_OMAP2PLUS || ARCH_KEYSTONE || COMPILE_TEST + depends on ARCH_DAVINCI || ARCH_OMAP2PLUS || ARCH_KEYSTONE || ARCH_K3 || COMPILE_TEST select PHYLIB ---help--- This driver supports TI's DaVinci MDIO module. @@ -53,6 +53,7 @@ config TI_CPSW select MFD_SYSCON select PAGE_POOL select REGMAP + imply PHY_TI_GMII_SEL ---help--- This driver supports TI's CPSW Ethernet Switch. @@ -94,6 +95,21 @@ config TI_CPTS_MOD imply PTP_1588_CLOCK default m +config TI_K3_AM65_CPSW_NUSS + tristate "TI K3 AM654x/J721E CPSW Ethernet driver" + depends on ARCH_K3 && OF && TI_K3_UDMA_GLUE_LAYER + select TI_DAVINCI_MDIO + imply PHY_TI_GMII_SEL + help + This driver supports TI K3 AM654/J721E CPSW2G Ethernet SubSystem. + The two-port Gigabit Ethernet MAC (MCU_CPSW0) subsystem provides + Ethernet packet communication for the device: One Ethernet port + (port 1) with selectable RGMII and RMII interfaces and an internal + Communications Port Programming Interface (CPPI) port (port 0). + + To compile this driver as a module, choose M here: the module + will be called ti-am65-cpsw-nuss. + config TI_KEYSTONE_NETCP tristate "TI Keystone NETCP Core Support" select TI_DAVINCI_MDIO diff --git a/drivers/net/ethernet/ti/Makefile b/drivers/net/ethernet/ti/Makefile index ecf776ad8689..53792190e9c2 100644 --- a/drivers/net/ethernet/ti/Makefile +++ b/drivers/net/ethernet/ti/Makefile @@ -23,3 +23,6 @@ obj-$(CONFIG_TI_KEYSTONE_NETCP) += keystone_netcp.o keystone_netcp-y := netcp_core.o cpsw_ale.o obj-$(CONFIG_TI_KEYSTONE_NETCP_ETHSS) += keystone_netcp_ethss.o keystone_netcp_ethss-y := netcp_ethss.o netcp_sgmii.o netcp_xgbepcsr.o cpsw_ale.o + +obj-$(CONFIG_TI_K3_AM65_CPSW_NUSS) += ti-am65-cpsw-nuss.o +ti-am65-cpsw-nuss-y := am65-cpsw-nuss.o cpsw_sl.o am65-cpsw-ethtool.o cpsw_ale.o k3-cppi-desc-pool.o diff --git a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c new file mode 100644 index 000000000000..c3502aa15ea0 --- /dev/null +++ b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c @@ -0,0 +1,747 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Texas Instruments K3 AM65 Ethernet Switch SubSystem Driver ethtool ops + * + * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com/ + * + */ + +#include <linux/net_tstamp.h> +#include <linux/phy.h> +#include <linux/platform_device.h> +#include <linux/pm_runtime.h> + +#include "am65-cpsw-nuss.h" +#include "cpsw_ale.h" + +#define AM65_CPSW_REGDUMP_VER 0x1 + +enum { + AM65_CPSW_REGDUMP_MOD_NUSS = 1, + AM65_CPSW_REGDUMP_MOD_RGMII_STATUS = 2, + AM65_CPSW_REGDUMP_MOD_MDIO = 3, + AM65_CPSW_REGDUMP_MOD_CPSW = 4, + AM65_CPSW_REGDUMP_MOD_CPSW_P0 = 5, + AM65_CPSW_REGDUMP_MOD_CPSW_P1 = 6, + AM65_CPSW_REGDUMP_MOD_CPSW_CPTS = 7, + AM65_CPSW_REGDUMP_MOD_CPSW_ALE = 8, + AM65_CPSW_REGDUMP_MOD_CPSW_ALE_TBL = 9, + AM65_CPSW_REGDUMP_MOD_LAST, +}; + +/** + * struct am65_cpsw_regdump_hdr - regdump record header + * + * @module_id: CPSW module ID + * @len: CPSW module registers space length in u32 + */ + +struct am65_cpsw_regdump_hdr { + u32 module_id; + u32 len; +}; + +/** + * struct am65_cpsw_regdump_item - regdump module description + * + * @hdr: CPSW module header + * @start_ofs: CPSW module registers start addr + * @end_ofs: CPSW module registers end addr + * + * Registers dump provided in the format: + * u32 : module ID + * u32 : dump length + * u32[..len]: registers values + */ +struct am65_cpsw_regdump_item { + struct am65_cpsw_regdump_hdr hdr; + u32 start_ofs; + u32 end_ofs; +}; + +#define AM65_CPSW_REGDUMP_REC(mod, start, end) { \ + .hdr.module_id = (mod), \ + .hdr.len = (((u32 *)(end)) - ((u32 *)(start)) + 1) * sizeof(u32) * 2 + \ + sizeof(struct am65_cpsw_regdump_hdr), \ + .start_ofs = (start), \ + .end_ofs = end, \ +} + +static const struct am65_cpsw_regdump_item am65_cpsw_regdump[] = { + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_NUSS, 0x0, 0x1c), + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_RGMII_STATUS, 0x30, 0x4c), + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_MDIO, 0xf00, 0xffc), + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_CPSW, 0x20000, 0x2011c), + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_CPSW_P0, 0x21000, 0x21320), + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_CPSW_P1, 0x22000, 0x223a4), + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_CPSW_CPTS, + 0x3d000, 0x3d048), + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_CPSW_ALE, 0x3e000, 0x3e13c), + AM65_CPSW_REGDUMP_REC(AM65_CPSW_REGDUMP_MOD_CPSW_ALE_TBL, 0, 0), +}; + +struct am65_cpsw_stats_regs { + u32 rx_good_frames; + u32 rx_broadcast_frames; + u32 rx_multicast_frames; + u32 rx_pause_frames; /* slave */ + u32 rx_crc_errors; + u32 rx_align_code_errors; /* slave */ + u32 rx_oversized_frames; + u32 rx_jabber_frames; /* slave */ + u32 rx_undersized_frames; + u32 rx_fragments; /* slave */ + u32 ale_drop; + u32 ale_overrun_drop; + u32 rx_octets; + u32 tx_good_frames; + u32 tx_broadcast_frames; + u32 tx_multicast_frames; + u32 tx_pause_frames; /* slave */ + u32 tx_deferred_frames; /* slave */ + u32 tx_collision_frames; /* slave */ + u32 tx_single_coll_frames; /* slave */ + u32 tx_mult_coll_frames; /* slave */ + u32 tx_excessive_collisions; /* slave */ + u32 tx_late_collisions; /* slave */ + u32 rx_ipg_error; /* slave 10G only */ + u32 tx_carrier_sense_errors; /* slave */ + u32 tx_octets; + u32 tx_64B_frames; + u32 tx_65_to_127B_frames; + u32 tx_128_to_255B_frames; + u32 tx_256_to_511B_frames; + u32 tx_512_to_1023B_frames; + u32 tx_1024B_frames; + u32 net_octets; + u32 rx_bottom_fifo_drop; + u32 rx_port_mask_drop; + u32 rx_top_fifo_drop; + u32 ale_rate_limit_drop; + u32 ale_vid_ingress_drop; + u32 ale_da_eq_sa_drop; + u32 ale_block_drop; /* K3 */ + u32 ale_secure_drop; /* K3 */ + u32 ale_auth_drop; /* K3 */ + u32 ale_unknown_ucast; + u32 ale_unknown_ucast_bytes; + u32 ale_unknown_mcast; + u32 ale_unknown_mcast_bytes; + u32 ale_unknown_bcast; + u32 ale_unknown_bcast_bytes; + u32 ale_pol_match; + u32 ale_pol_match_red; + u32 ale_pol_match_yellow; + u32 ale_mcast_sa_drop; /* K3 */ + u32 ale_dual_vlan_drop; /* K3 */ + u32 ale_len_err_drop; /* K3 */ + u32 ale_ip_next_hdr_drop; /* K3 */ + u32 ale_ipv4_frag_drop; /* K3 */ + u32 __rsvd_1[24]; + u32 iet_rx_assembly_err; /* K3 slave */ + u32 iet_rx_assembly_ok; /* K3 slave */ + u32 iet_rx_smd_err; /* K3 slave */ + u32 iet_rx_frag; /* K3 slave */ + u32 iet_tx_hold; /* K3 slave */ + u32 iet_tx_frag; /* K3 slave */ + u32 __rsvd_2[9]; + u32 tx_mem_protect_err; + /* following NU only */ + u32 tx_pri0; + u32 tx_pri1; + u32 tx_pri2; + u32 tx_pri3; + u32 tx_pri4; + u32 tx_pri5; + u32 tx_pri6; + u32 tx_pri7; + u32 tx_pri0_bcnt; + u32 tx_pri1_bcnt; + u32 tx_pri2_bcnt; + u32 tx_pri3_bcnt; + u32 tx_pri4_bcnt; + u32 tx_pri5_bcnt; + u32 tx_pri6_bcnt; + u32 tx_pri7_bcnt; + u32 tx_pri0_drop; + u32 tx_pri1_drop; + u32 tx_pri2_drop; + u32 tx_pri3_drop; + u32 tx_pri4_drop; + u32 tx_pri5_drop; + u32 tx_pri6_drop; + u32 tx_pri7_drop; + u32 tx_pri0_drop_bcnt; + u32 tx_pri1_drop_bcnt; + u32 tx_pri2_drop_bcnt; + u32 tx_pri3_drop_bcnt; + u32 tx_pri4_drop_bcnt; + u32 tx_pri5_drop_bcnt; + u32 tx_pri6_drop_bcnt; + u32 tx_pri7_drop_bcnt; +}; + +struct am65_cpsw_ethtool_stat { + char desc[ETH_GSTRING_LEN]; + int offset; +}; + +#define AM65_CPSW_STATS(prefix, field) \ +{ \ + #prefix#field, \ + offsetof(struct am65_cpsw_stats_regs, field) \ +} + +static const struct am65_cpsw_ethtool_stat am65_host_stats[] = { + AM65_CPSW_STATS(p0_, rx_good_frames), + AM65_CPSW_STATS(p0_, rx_broadcast_frames), + AM65_CPSW_STATS(p0_, rx_multicast_frames), + AM65_CPSW_STATS(p0_, rx_crc_errors), + AM65_CPSW_STATS(p0_, rx_oversized_frames), + AM65_CPSW_STATS(p0_, rx_undersized_frames), + AM65_CPSW_STATS(p0_, ale_drop), + AM65_CPSW_STATS(p0_, ale_overrun_drop), + AM65_CPSW_STATS(p0_, rx_octets), + AM65_CPSW_STATS(p0_, tx_good_frames), + AM65_CPSW_STATS(p0_, tx_broadcast_frames), + AM65_CPSW_STATS(p0_, tx_multicast_frames), + AM65_CPSW_STATS(p0_, tx_octets), + AM65_CPSW_STATS(p0_, tx_64B_frames), + AM65_CPSW_STATS(p0_, tx_65_to_127B_frames), + AM65_CPSW_STATS(p0_, tx_128_to_255B_frames), + AM65_CPSW_STATS(p0_, tx_256_to_511B_frames), + AM65_CPSW_STATS(p0_, tx_512_to_1023B_frames), + AM65_CPSW_STATS(p0_, tx_1024B_frames), + AM65_CPSW_STATS(p0_, net_octets), + AM65_CPSW_STATS(p0_, rx_bottom_fifo_drop), + AM65_CPSW_STATS(p0_, rx_port_mask_drop), + AM65_CPSW_STATS(p0_, rx_top_fifo_drop), + AM65_CPSW_STATS(p0_, ale_rate_limit_drop), + AM65_CPSW_STATS(p0_, ale_vid_ingress_drop), + AM65_CPSW_STATS(p0_, ale_da_eq_sa_drop), + AM65_CPSW_STATS(p0_, ale_block_drop), + AM65_CPSW_STATS(p0_, ale_secure_drop), + AM65_CPSW_STATS(p0_, ale_auth_drop), + AM65_CPSW_STATS(p0_, ale_unknown_ucast), + AM65_CPSW_STATS(p0_, ale_unknown_ucast_bytes), + AM65_CPSW_STATS(p0_, ale_unknown_mcast), + AM65_CPSW_STATS(p0_, ale_unknown_mcast_bytes), + AM65_CPSW_STATS(p0_, ale_unknown_bcast), + AM65_CPSW_STATS(p0_, ale_unknown_bcast_bytes), + AM65_CPSW_STATS(p0_, ale_pol_match), + AM65_CPSW_STATS(p0_, ale_pol_match_red), + AM65_CPSW_STATS(p0_, ale_pol_match_yellow), + AM65_CPSW_STATS(p0_, ale_mcast_sa_drop), + AM65_CPSW_STATS(p0_, ale_dual_vlan_drop), + AM65_CPSW_STATS(p0_, ale_len_err_drop), + AM65_CPSW_STATS(p0_, ale_ip_next_hdr_drop), + AM65_CPSW_STATS(p0_, ale_ipv4_frag_drop), + AM65_CPSW_STATS(p0_, tx_mem_protect_err), + AM65_CPSW_STATS(p0_, tx_pri0), + AM65_CPSW_STATS(p0_, tx_pri1), + AM65_CPSW_STATS(p0_, tx_pri2), + AM65_CPSW_STATS(p0_, tx_pri3), + AM65_CPSW_STATS(p0_, tx_pri4), + AM65_CPSW_STATS(p0_, tx_pri5), + AM65_CPSW_STATS(p0_, tx_pri6), + AM65_CPSW_STATS(p0_, tx_pri7), + AM65_CPSW_STATS(p0_, tx_pri0_bcnt), + AM65_CPSW_STATS(p0_, tx_pri1_bcnt), + AM65_CPSW_STATS(p0_, tx_pri2_bcnt), + AM65_CPSW_STATS(p0_, tx_pri3_bcnt), + AM65_CPSW_STATS(p0_, tx_pri4_bcnt), + AM65_CPSW_STATS(p0_, tx_pri5_bcnt), + AM65_CPSW_STATS(p0_, tx_pri6_bcnt), + AM65_CPSW_STATS(p0_, tx_pri7_bcnt), + AM65_CPSW_STATS(p0_, tx_pri0_drop), + AM65_CPSW_STATS(p0_, tx_pri1_drop), + AM65_CPSW_STATS(p0_, tx_pri2_drop), + AM65_CPSW_STATS(p0_, tx_pri3_drop), + AM65_CPSW_STATS(p0_, tx_pri4_drop), + AM65_CPSW_STATS(p0_, tx_pri5_drop), + AM65_CPSW_STATS(p0_, tx_pri6_drop), + AM65_CPSW_STATS(p0_, tx_pri7_drop), + AM65_CPSW_STATS(p0_, tx_pri0_drop_bcnt), + AM65_CPSW_STATS(p0_, tx_pri1_drop_bcnt), + AM65_CPSW_STATS(p0_, tx_pri2_drop_bcnt), + AM65_CPSW_STATS(p0_, tx_pri3_drop_bcnt), + AM65_CPSW_STATS(p0_, tx_pri4_drop_bcnt), + AM65_CPSW_STATS(p0_, tx_pri5_drop_bcnt), + AM65_CPSW_STATS(p0_, tx_pri6_drop_bcnt), + AM65_CPSW_STATS(p0_, tx_pri7_drop_bcnt), +}; + +static const struct am65_cpsw_ethtool_stat am65_slave_stats[] = { + AM65_CPSW_STATS(, rx_good_frames), + AM65_CPSW_STATS(, rx_broadcast_frames), + AM65_CPSW_STATS(, rx_multicast_frames), + AM65_CPSW_STATS(, rx_pause_frames), + AM65_CPSW_STATS(, rx_crc_errors), + AM65_CPSW_STATS(, rx_align_code_errors), + AM65_CPSW_STATS(, rx_oversized_frames), + AM65_CPSW_STATS(, rx_jabber_frames), + AM65_CPSW_STATS(, rx_undersized_frames), + AM65_CPSW_STATS(, rx_fragments), + AM65_CPSW_STATS(, ale_drop), + AM65_CPSW_STATS(, ale_overrun_drop), + AM65_CPSW_STATS(, rx_octets), + AM65_CPSW_STATS(, tx_good_frames), + AM65_CPSW_STATS(, tx_broadcast_frames), + AM65_CPSW_STATS(, tx_multicast_frames), + AM65_CPSW_STATS(, tx_pause_frames), + AM65_CPSW_STATS(, tx_deferred_frames), + AM65_CPSW_STATS(, tx_collision_frames), + AM65_CPSW_STATS(, tx_single_coll_frames), + AM65_CPSW_STATS(, tx_mult_coll_frames), + AM65_CPSW_STATS(, tx_excessive_collisions), + AM65_CPSW_STATS(, tx_late_collisions), + AM65_CPSW_STATS(, rx_ipg_error), + AM65_CPSW_STATS(, tx_carrier_sense_errors), + AM65_CPSW_STATS(, tx_octets), + AM65_CPSW_STATS(, tx_64B_frames), + AM65_CPSW_STATS(, tx_65_to_127B_frames), + AM65_CPSW_STATS(, tx_128_to_255B_frames), + AM65_CPSW_STATS(, tx_256_to_511B_frames), + AM65_CPSW_STATS(, tx_512_to_1023B_frames), + AM65_CPSW_STATS(, tx_1024B_frames), + AM65_CPSW_STATS(, net_octets), + AM65_CPSW_STATS(, rx_bottom_fifo_drop), + AM65_CPSW_STATS(, rx_port_mask_drop), + AM65_CPSW_STATS(, rx_top_fifo_drop), + AM65_CPSW_STATS(, ale_rate_limit_drop), + AM65_CPSW_STATS(, ale_vid_ingress_drop), + AM65_CPSW_STATS(, ale_da_eq_sa_drop), + AM65_CPSW_STATS(, ale_block_drop), + AM65_CPSW_STATS(, ale_secure_drop), + AM65_CPSW_STATS(, ale_auth_drop), + AM65_CPSW_STATS(, ale_unknown_ucast), + AM65_CPSW_STATS(, ale_unknown_ucast_bytes), + AM65_CPSW_STATS(, ale_unknown_mcast), + AM65_CPSW_STATS(, ale_unknown_mcast_bytes), + AM65_CPSW_STATS(, ale_unknown_bcast), + AM65_CPSW_STATS(, ale_unknown_bcast_bytes), + AM65_CPSW_STATS(, ale_pol_match), + AM65_CPSW_STATS(, ale_pol_match_red), + AM65_CPSW_STATS(, ale_pol_match_yellow), + AM65_CPSW_STATS(, ale_mcast_sa_drop), + AM65_CPSW_STATS(, ale_dual_vlan_drop), + AM65_CPSW_STATS(, ale_len_err_drop), + AM65_CPSW_STATS(, ale_ip_next_hdr_drop), + AM65_CPSW_STATS(, ale_ipv4_frag_drop), + AM65_CPSW_STATS(, iet_rx_assembly_err), + AM65_CPSW_STATS(, iet_rx_assembly_ok), + AM65_CPSW_STATS(, iet_rx_smd_err), + AM65_CPSW_STATS(, iet_rx_frag), + AM65_CPSW_STATS(, iet_tx_hold), + AM65_CPSW_STATS(, iet_tx_frag), + AM65_CPSW_STATS(, tx_mem_protect_err), + AM65_CPSW_STATS(, tx_pri0), + AM65_CPSW_STATS(, tx_pri1), + AM65_CPSW_STATS(, tx_pri2), + AM65_CPSW_STATS(, tx_pri3), + AM65_CPSW_STATS(, tx_pri4), + AM65_CPSW_STATS(, tx_pri5), + AM65_CPSW_STATS(, tx_pri6), + AM65_CPSW_STATS(, tx_pri7), + AM65_CPSW_STATS(, tx_pri0_bcnt), + AM65_CPSW_STATS(, tx_pri1_bcnt), + AM65_CPSW_STATS(, tx_pri2_bcnt), + AM65_CPSW_STATS(, tx_pri3_bcnt), + AM65_CPSW_STATS(, tx_pri4_bcnt), + AM65_CPSW_STATS(, tx_pri5_bcnt), + AM65_CPSW_STATS(, tx_pri6_bcnt), + AM65_CPSW_STATS(, tx_pri7_bcnt), + AM65_CPSW_STATS(, tx_pri0_drop), + AM65_CPSW_STATS(, tx_pri1_drop), + AM65_CPSW_STATS(, tx_pri2_drop), + AM65_CPSW_STATS(, tx_pri3_drop), + AM65_CPSW_STATS(, tx_pri4_drop), + AM65_CPSW_STATS(, tx_pri5_drop), + AM65_CPSW_STATS(, tx_pri6_drop), + AM65_CPSW_STATS(, tx_pri7_drop), + AM65_CPSW_STATS(, tx_pri0_drop_bcnt), + AM65_CPSW_STATS(, tx_pri1_drop_bcnt), + AM65_CPSW_STATS(, tx_pri2_drop_bcnt), + AM65_CPSW_STATS(, tx_pri3_drop_bcnt), + AM65_CPSW_STATS(, tx_pri4_drop_bcnt), + AM65_CPSW_STATS(, tx_pri5_drop_bcnt), + AM65_CPSW_STATS(, tx_pri6_drop_bcnt), + AM65_CPSW_STATS(, tx_pri7_drop_bcnt), +}; + +/* Ethtool priv_flags */ +static const char am65_cpsw_ethtool_priv_flags[][ETH_GSTRING_LEN] = { +#define AM65_CPSW_PRIV_P0_RX_PTYPE_RROBIN BIT(0) + "p0-rx-ptype-rrobin", +}; + +static int am65_cpsw_ethtool_op_begin(struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + int ret; + + ret = pm_runtime_get_sync(common->dev); + if (ret < 0) { + dev_err(common->dev, "ethtool begin failed %d\n", ret); + pm_runtime_put_noidle(common->dev); + } + + return ret; +} + +static void am65_cpsw_ethtool_op_complete(struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + int ret; + + ret = pm_runtime_put(common->dev); + if (ret < 0 && ret != -EBUSY) + dev_err(common->dev, "ethtool complete failed %d\n", ret); +} + +static void am65_cpsw_get_drvinfo(struct net_device *ndev, + struct ethtool_drvinfo *info) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + + strlcpy(info->driver, dev_driver_string(common->dev), + sizeof(info->driver)); + strlcpy(info->bus_info, dev_name(common->dev), sizeof(info->bus_info)); +} + +static u32 am65_cpsw_get_msglevel(struct net_device *ndev) +{ + struct am65_cpsw_ndev_priv *priv = am65_ndev_to_priv(ndev); + + return priv->msg_enable; +} + +static void am65_cpsw_set_msglevel(struct net_device *ndev, u32 value) +{ + struct am65_cpsw_ndev_priv *priv = am65_ndev_to_priv(ndev); + + priv->msg_enable = value; +} + +static void am65_cpsw_get_channels(struct net_device *ndev, + struct ethtool_channels *ch) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + + ch->max_rx = AM65_CPSW_MAX_RX_QUEUES; + ch->max_tx = AM65_CPSW_MAX_TX_QUEUES; + ch->rx_count = AM65_CPSW_MAX_RX_QUEUES; + ch->tx_count = common->tx_ch_num; +} + +static int am65_cpsw_set_channels(struct net_device *ndev, + struct ethtool_channels *chs) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + + if (!chs->rx_count || !chs->tx_count) + return -EINVAL; + + /* Check if interface is up. Can change the num queues when + * the interface is down. + */ + if (netif_running(ndev)) + return -EBUSY; + + am65_cpsw_nuss_remove_tx_chns(common); + + return am65_cpsw_nuss_update_tx_chns(common, chs->tx_count); +} + +static void am65_cpsw_get_ringparam(struct net_device *ndev, + struct ethtool_ringparam *ering) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + + /* not supported */ + ering->tx_pending = common->tx_chns[0].descs_num; + ering->rx_pending = common->rx_chns.descs_num; +} + +static void am65_cpsw_get_pauseparam(struct net_device *ndev, + struct ethtool_pauseparam *pause) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + pause->autoneg = AUTONEG_DISABLE; + pause->rx_pause = salve->rx_pause ? true : false; + pause->tx_pause = salve->tx_pause ? true : false; +} + +static int am65_cpsw_set_pauseparam(struct net_device *ndev, + struct ethtool_pauseparam *pause) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + if (!salve->phy) + return -EINVAL; + + if (!phy_validate_pause(salve->phy, pause)) + return -EINVAL; + + salve->rx_pause = pause->rx_pause ? true : false; + salve->tx_pause = pause->tx_pause ? true : false; + + phy_set_asym_pause(salve->phy, salve->rx_pause, salve->tx_pause); + + return 0; +} + +static void am65_cpsw_get_wol(struct net_device *ndev, + struct ethtool_wolinfo *wol) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + wol->supported = 0; + wol->wolopts = 0; + + if (salve->phy) + phy_ethtool_get_wol(salve->phy, wol); +} + +static int am65_cpsw_set_wol(struct net_device *ndev, + struct ethtool_wolinfo *wol) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + if (!salve->phy) + return -EOPNOTSUPP; + + return phy_ethtool_set_wol(salve->phy, wol); +} + +static int am65_cpsw_get_link_ksettings(struct net_device *ndev, + struct ethtool_link_ksettings *ecmd) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + if (!salve->phy) + return -EOPNOTSUPP; + + phy_ethtool_ksettings_get(salve->phy, ecmd); + return 0; +} + +static int +am65_cpsw_set_link_ksettings(struct net_device *ndev, + const struct ethtool_link_ksettings *ecmd) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + if (!salve->phy || phy_is_pseudo_fixed_link(salve->phy)) + return -EOPNOTSUPP; + + return phy_ethtool_ksettings_set(salve->phy, ecmd); +} + +static int am65_cpsw_get_eee(struct net_device *ndev, struct ethtool_eee *edata) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + if (!salve->phy || phy_is_pseudo_fixed_link(salve->phy)) + return -EOPNOTSUPP; + + return phy_ethtool_get_eee(salve->phy, edata); +} + +static int am65_cpsw_set_eee(struct net_device *ndev, struct ethtool_eee *edata) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + if (!salve->phy || phy_is_pseudo_fixed_link(salve->phy)) + return -EOPNOTSUPP; + + return phy_ethtool_set_eee(salve->phy, edata); +} + +static int am65_cpsw_nway_reset(struct net_device *ndev) +{ + struct am65_cpsw_slave_data *salve = am65_ndev_to_slave(ndev); + + if (!salve->phy || phy_is_pseudo_fixed_link(salve->phy)) + return -EOPNOTSUPP; + + return phy_restart_aneg(salve->phy); +} + +static int am65_cpsw_get_regs_len(struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + u32 i, regdump_len = 0; + + for (i = 0; i < ARRAY_SIZE(am65_cpsw_regdump); i++) { + if (am65_cpsw_regdump[i].hdr.module_id == + AM65_CPSW_REGDUMP_MOD_CPSW_ALE_TBL) { + regdump_len += sizeof(struct am65_cpsw_regdump_hdr); + regdump_len += common->ale->params.ale_entries * + ALE_ENTRY_WORDS * sizeof(u32); + continue; + } + regdump_len += am65_cpsw_regdump[i].hdr.len; + } + + return regdump_len; +} + +static void am65_cpsw_get_regs(struct net_device *ndev, + struct ethtool_regs *regs, void *p) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + u32 i, j, pos, *reg = p; + + /* update CPSW IP version */ + regs->version = AM65_CPSW_REGDUMP_VER; + + pos = 0; + for (i = 0; i < ARRAY_SIZE(am65_cpsw_regdump); i++) { + reg[pos++] = am65_cpsw_regdump[i].hdr.module_id; + + if (am65_cpsw_regdump[i].hdr.module_id == + AM65_CPSW_REGDUMP_MOD_CPSW_ALE_TBL) { + u32 ale_tbl_len = common->ale->params.ale_entries * + ALE_ENTRY_WORDS * sizeof(u32) + + sizeof(struct am65_cpsw_regdump_hdr); + reg[pos++] = ale_tbl_len; + cpsw_ale_dump(common->ale, ®[pos]); + pos += ale_tbl_len; + continue; + } + + reg[pos++] = am65_cpsw_regdump[i].hdr.len; + + j = am65_cpsw_regdump[i].start_ofs; + do { + reg[pos++] = j; + reg[pos++] = readl_relaxed(common->ss_base + j); + j += sizeof(u32); + } while (j <= am65_cpsw_regdump[i].end_ofs); + } +} + +static int am65_cpsw_get_sset_count(struct net_device *ndev, int sset) +{ + switch (sset) { + case ETH_SS_STATS: + return ARRAY_SIZE(am65_host_stats) + + ARRAY_SIZE(am65_slave_stats); + case ETH_SS_PRIV_FLAGS: + return ARRAY_SIZE(am65_cpsw_ethtool_priv_flags); + default: + return -EOPNOTSUPP; + } +} + +static void am65_cpsw_get_strings(struct net_device *ndev, + u32 stringset, u8 *data) +{ + const struct am65_cpsw_ethtool_stat *hw_stats; + u32 i, num_stats; + u8 *p = data; + + switch (stringset) { + case ETH_SS_STATS: + num_stats = ARRAY_SIZE(am65_host_stats); + hw_stats = am65_host_stats; + for (i = 0; i < num_stats; i++) { + memcpy(p, hw_stats[i].desc, ETH_GSTRING_LEN); + p += ETH_GSTRING_LEN; + } + + num_stats = ARRAY_SIZE(am65_slave_stats); + hw_stats = am65_slave_stats; + for (i = 0; i < num_stats; i++) { + memcpy(p, hw_stats[i].desc, ETH_GSTRING_LEN); + p += ETH_GSTRING_LEN; + } + break; + case ETH_SS_PRIV_FLAGS: + num_stats = ARRAY_SIZE(am65_cpsw_ethtool_priv_flags); + + for (i = 0; i < num_stats; i++) { + memcpy(p, am65_cpsw_ethtool_priv_flags[i], + ETH_GSTRING_LEN); + p += ETH_GSTRING_LEN; + } + break; + } +} + +static void am65_cpsw_get_ethtool_stats(struct net_device *ndev, + struct ethtool_stats *stats, u64 *data) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + const struct am65_cpsw_ethtool_stat *hw_stats; + struct am65_cpsw_host *host_p; + struct am65_cpsw_port *port; + u32 i, num_stats; + + host_p = am65_common_get_host(common); + port = am65_ndev_to_port(ndev); + num_stats = ARRAY_SIZE(am65_host_stats); + hw_stats = am65_host_stats; + for (i = 0; i < num_stats; i++) + *data++ = readl_relaxed(host_p->stat_base + + hw_stats[i].offset); + + num_stats = ARRAY_SIZE(am65_slave_stats); + hw_stats = am65_slave_stats; + for (i = 0; i < num_stats; i++) + *data++ = readl_relaxed(port->stat_base + + hw_stats[i].offset); +} + +static u32 am65_cpsw_get_ethtool_priv_flags(struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + u32 priv_flags = 0; + + if (common->pf_p0_rx_ptype_rrobin) + priv_flags |= AM65_CPSW_PRIV_P0_RX_PTYPE_RROBIN; + + return priv_flags; +} + +static int am65_cpsw_set_ethtool_priv_flags(struct net_device *ndev, u32 flags) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + + common->pf_p0_rx_ptype_rrobin = + !!(flags & AM65_CPSW_PRIV_P0_RX_PTYPE_RROBIN); + am65_cpsw_nuss_set_p0_ptype(common); + + return 0; +} + +const struct ethtool_ops am65_cpsw_ethtool_ops_slave = { + .begin = am65_cpsw_ethtool_op_begin, + .complete = am65_cpsw_ethtool_op_complete, + .get_drvinfo = am65_cpsw_get_drvinfo, + .get_msglevel = am65_cpsw_get_msglevel, + .set_msglevel = am65_cpsw_set_msglevel, + .get_channels = am65_cpsw_get_channels, + .set_channels = am65_cpsw_set_channels, + .get_ringparam = am65_cpsw_get_ringparam, + .get_regs_len = am65_cpsw_get_regs_len, + .get_regs = am65_cpsw_get_regs, + .get_sset_count = am65_cpsw_get_sset_count, + .get_strings = am65_cpsw_get_strings, + .get_ethtool_stats = am65_cpsw_get_ethtool_stats, + .get_ts_info = ethtool_op_get_ts_info, + .get_priv_flags = am65_cpsw_get_ethtool_priv_flags, + .set_priv_flags = am65_cpsw_set_ethtool_priv_flags, + + .get_link = ethtool_op_get_link, + .get_link_ksettings = am65_cpsw_get_link_ksettings, + .set_link_ksettings = am65_cpsw_set_link_ksettings, + .get_pauseparam = am65_cpsw_get_pauseparam, + .set_pauseparam = am65_cpsw_set_pauseparam, + .get_wol = am65_cpsw_get_wol, + .set_wol = am65_cpsw_set_wol, + .get_eee = am65_cpsw_get_eee, + .set_eee = am65_cpsw_set_eee, + .nway_reset = am65_cpsw_nway_reset, +}; diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c new file mode 100644 index 000000000000..f71c15c39492 --- /dev/null +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -0,0 +1,1965 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Texas Instruments K3 AM65 Ethernet Switch SubSystem Driver + * + * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com/ + * + */ + +#include <linux/etherdevice.h> +#include <linux/if_vlan.h> +#include <linux/interrupt.h> +#include <linux/kernel.h> +#include <linux/kmemleak.h> +#include <linux/module.h> +#include <linux/netdevice.h> +#include <linux/net_tstamp.h> +#include <linux/of.h> +#include <linux/of_mdio.h> +#include <linux/of_net.h> +#include <linux/of_device.h> +#include <linux/phy.h> +#include <linux/phy/phy.h> +#include <linux/platform_device.h> +#include <linux/pm_runtime.h> +#include <linux/regmap.h> +#include <linux/mfd/syscon.h> +#include <linux/dma/ti-cppi5.h> +#include <linux/dma/k3-udma-glue.h> + +#include "cpsw_ale.h" +#include "cpsw_sl.h" +#include "am65-cpsw-nuss.h" +#include "k3-cppi-desc-pool.h" + +#define AM65_CPSW_SS_BASE 0x0 +#define AM65_CPSW_SGMII_BASE 0x100 +#define AM65_CPSW_XGMII_BASE 0x2100 +#define AM65_CPSW_CPSW_NU_BASE 0x20000 +#define AM65_CPSW_NU_PORTS_BASE 0x1000 +#define AM65_CPSW_NU_STATS_BASE 0x1a000 +#define AM65_CPSW_NU_ALE_BASE 0x1e000 +#define AM65_CPSW_NU_CPTS_BASE 0x1d000 + +#define AM65_CPSW_NU_PORTS_OFFSET 0x1000 +#define AM65_CPSW_NU_STATS_PORT_OFFSET 0x200 + +#define AM65_CPSW_MAX_PORTS 8 + +#define AM65_CPSW_MIN_PACKET_SIZE VLAN_ETH_ZLEN +#define AM65_CPSW_MAX_PACKET_SIZE (VLAN_ETH_FRAME_LEN + ETH_FCS_LEN) + +#define AM65_CPSW_REG_CTL 0x004 +#define AM65_CPSW_REG_STAT_PORT_EN 0x014 +#define AM65_CPSW_REG_PTYPE 0x018 + +#define AM65_CPSW_P0_REG_CTL 0x004 +#define AM65_CPSW_PORT0_REG_FLOW_ID_OFFSET 0x008 + +#define AM65_CPSW_PORT_REG_PRI_CTL 0x01c +#define AM65_CPSW_PORT_REG_RX_PRI_MAP 0x020 +#define AM65_CPSW_PORT_REG_RX_MAXLEN 0x024 + +#define AM65_CPSW_PORTN_REG_SA_L 0x308 +#define AM65_CPSW_PORTN_REG_SA_H 0x30c +#define AM65_CPSW_PORTN_REG_TS_CTL 0x310 +#define AM65_CPSW_PORTN_REG_TS_SEQ_LTYPE_REG 0x314 +#define AM65_CPSW_PORTN_REG_TS_VLAN_LTYPE_REG 0x318 +#define AM65_CPSW_PORTN_REG_TS_CTL_LTYPE2 0x31C + +#define AM65_CPSW_CTL_VLAN_AWARE BIT(1) +#define AM65_CPSW_CTL_P0_ENABLE BIT(2) +#define AM65_CPSW_CTL_P0_TX_CRC_REMOVE BIT(13) +#define AM65_CPSW_CTL_P0_RX_PAD BIT(14) + +/* AM65_CPSW_P0_REG_CTL */ +#define AM65_CPSW_P0_REG_CTL_RX_CHECKSUM_EN BIT(0) + +/* AM65_CPSW_PORT_REG_PRI_CTL */ +#define AM65_CPSW_PORT_REG_PRI_CTL_RX_PTYPE_RROBIN BIT(8) + +/* AM65_CPSW_PN_TS_CTL register fields */ +#define AM65_CPSW_PN_TS_CTL_TX_ANX_F_EN BIT(4) +#define AM65_CPSW_PN_TS_CTL_TX_VLAN_LT1_EN BIT(5) +#define AM65_CPSW_PN_TS_CTL_TX_VLAN_LT2_EN BIT(6) +#define AM65_CPSW_PN_TS_CTL_TX_ANX_D_EN BIT(7) +#define AM65_CPSW_PN_TS_CTL_TX_ANX_E_EN BIT(10) +#define AM65_CPSW_PN_TS_CTL_TX_HOST_TS_EN BIT(11) +#define AM65_CPSW_PN_TS_CTL_MSG_TYPE_EN_SHIFT 16 + +/* AM65_CPSW_PORTN_REG_TS_SEQ_LTYPE_REG register fields */ +#define AM65_CPSW_PN_TS_SEQ_ID_OFFSET_SHIFT 16 + +/* AM65_CPSW_PORTN_REG_TS_CTL_LTYPE2 */ +#define AM65_CPSW_PN_TS_CTL_LTYPE2_TS_107 BIT(16) +#define AM65_CPSW_PN_TS_CTL_LTYPE2_TS_129 BIT(17) +#define AM65_CPSW_PN_TS_CTL_LTYPE2_TS_130 BIT(18) +#define AM65_CPSW_PN_TS_CTL_LTYPE2_TS_131 BIT(19) +#define AM65_CPSW_PN_TS_CTL_LTYPE2_TS_132 BIT(20) +#define AM65_CPSW_PN_TS_CTL_LTYPE2_TS_319 BIT(21) +#define AM65_CPSW_PN_TS_CTL_LTYPE2_TS_320 BIT(22) +#define AM65_CPSW_PN_TS_CTL_LTYPE2_TS_TTL_NONZERO BIT(23) + +/* The PTP event messages - Sync, Delay_Req, Pdelay_Req, and Pdelay_Resp. */ +#define AM65_CPSW_TS_EVENT_MSG_TYPE_BITS (BIT(0) | BIT(1) | BIT(2) | BIT(3)) + +#define AM65_CPSW_TS_SEQ_ID_OFFSET (0x1e) + +#define AM65_CPSW_TS_TX_ANX_ALL_EN \ + (AM65_CPSW_PN_TS_CTL_TX_ANX_D_EN | \ + AM65_CPSW_PN_TS_CTL_TX_ANX_E_EN | \ + AM65_CPSW_PN_TS_CTL_TX_ANX_F_EN) + +#define AM65_CPSW_ALE_AGEOUT_DEFAULT 30 +/* Number of TX/RX descriptors */ +#define AM65_CPSW_MAX_TX_DESC 500 +#define AM65_CPSW_MAX_RX_DESC 500 + +#define AM65_CPSW_NAV_PS_DATA_SIZE 16 +#define AM65_CPSW_NAV_SW_DATA_SIZE 16 + +#define AM65_CPSW_DEBUG (NETIF_MSG_HW | NETIF_MSG_DRV | NETIF_MSG_LINK | \ + NETIF_MSG_IFUP | NETIF_MSG_PROBE | NETIF_MSG_IFDOWN | \ + NETIF_MSG_RX_ERR | NETIF_MSG_TX_ERR) + +static void am65_cpsw_port_set_sl_mac(struct am65_cpsw_port *slave, + const u8 *dev_addr) +{ + u32 mac_hi = (dev_addr[0] << 0) | (dev_addr[1] << 8) | + (dev_addr[2] << 16) | (dev_addr[3] << 24); + u32 mac_lo = (dev_addr[4] << 0) | (dev_addr[5] << 8); + + writel(mac_hi, slave->port_base + AM65_CPSW_PORTN_REG_SA_H); + writel(mac_lo, slave->port_base + AM65_CPSW_PORTN_REG_SA_L); +} + +static void am65_cpsw_sl_ctl_reset(struct am65_cpsw_port *port) +{ + cpsw_sl_reset(port->slave.mac_sl, 100); + /* Max length register has to be restored after MAC SL reset */ + writel(AM65_CPSW_MAX_PACKET_SIZE, + port->port_base + AM65_CPSW_PORT_REG_RX_MAXLEN); +} + +static void am65_cpsw_nuss_get_ver(struct am65_cpsw_common *common) +{ + common->nuss_ver = readl(common->ss_base); + common->cpsw_ver = readl(common->cpsw_base); + dev_info(common->dev, + "initializing am65 cpsw nuss version 0x%08X, cpsw version 0x%08X Ports: %u\n", + common->nuss_ver, + common->cpsw_ver, + common->port_num + 1); +} + +void am65_cpsw_nuss_adjust_link(struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct phy_device *phy = port->slave.phy; + u32 mac_control = 0; + + if (!phy) + return; + + if (phy->link) { + mac_control = CPSW_SL_CTL_GMII_EN; + + if (phy->speed == 1000) + mac_control |= CPSW_SL_CTL_GIG; + if (phy->speed == 10 && phy_interface_is_rgmii(phy)) + /* Can be used with in band mode only */ + mac_control |= CPSW_SL_CTL_EXT_EN; + if (phy->duplex) + mac_control |= CPSW_SL_CTL_FULLDUPLEX; + + /* RGMII speed is 100M if !CPSW_SL_CTL_GIG*/ + + /* rx_pause/tx_pause */ + if (port->slave.rx_pause) + mac_control |= CPSW_SL_CTL_RX_FLOW_EN; + + if (port->slave.tx_pause) + mac_control |= CPSW_SL_CTL_TX_FLOW_EN; + + cpsw_sl_ctl_set(port->slave.mac_sl, mac_control); + + /* enable forwarding */ + cpsw_ale_control_set(common->ale, port->port_id, + ALE_PORT_STATE, ALE_PORT_STATE_FORWARD); + + netif_tx_wake_all_queues(ndev); + } else { + int tmo; + /* disable forwarding */ + cpsw_ale_control_set(common->ale, port->port_id, + ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); + + cpsw_sl_ctl_set(port->slave.mac_sl, CPSW_SL_CTL_CMD_IDLE); + + tmo = cpsw_sl_wait_for_idle(port->slave.mac_sl, 100); + dev_dbg(common->dev, "donw msc_sl %08x tmo %d\n", + cpsw_sl_reg_read(port->slave.mac_sl, CPSW_SL_MACSTATUS), + tmo); + + cpsw_sl_ctl_reset(port->slave.mac_sl); + + netif_tx_stop_all_queues(ndev); + } + + phy_print_status(phy); +} + +static int am65_cpsw_nuss_ndo_slave_add_vid(struct net_device *ndev, + __be16 proto, u16 vid) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + u32 port_mask, unreg_mcast = 0; + int ret; + + ret = pm_runtime_get_sync(common->dev); + if (ret < 0) { + pm_runtime_put_noidle(common->dev); + return ret; + } + + port_mask = BIT(port->port_id) | ALE_PORT_HOST; + if (!vid) + unreg_mcast = port_mask; + dev_info(common->dev, "Adding vlan %d to vlan filter\n", vid); + ret = cpsw_ale_add_vlan(common->ale, vid, port_mask, + unreg_mcast, port_mask, 0); + + pm_runtime_put(common->dev); + return ret; +} + +static int am65_cpsw_nuss_ndo_slave_kill_vid(struct net_device *ndev, + __be16 proto, u16 vid) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + int ret; + + ret = pm_runtime_get_sync(common->dev); + if (ret < 0) { + pm_runtime_put_noidle(common->dev); + return ret; + } + + dev_info(common->dev, "Removing vlan %d from vlan filter\n", vid); + ret = cpsw_ale_del_vlan(common->ale, vid, 0); + + pm_runtime_put(common->dev); + return ret; +} + +static void am65_cpsw_slave_set_promisc_2g(struct am65_cpsw_port *port, + bool promisc) +{ + struct am65_cpsw_common *common = port->common; + + if (promisc) { + /* Enable promiscuous mode */ + cpsw_ale_control_set(common->ale, port->port_id, + ALE_PORT_MACONLY_CAF, 1); + dev_dbg(common->dev, "promisc enabled\n"); + } else { + /* Disable promiscuous mode */ + cpsw_ale_control_set(common->ale, port->port_id, + ALE_PORT_MACONLY_CAF, 0); + dev_dbg(common->dev, "promisc disabled\n"); + } +} + +static void am65_cpsw_nuss_ndo_slave_set_rx_mode(struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + u32 port_mask; + bool promisc; + + promisc = !!(ndev->flags & IFF_PROMISC); + am65_cpsw_slave_set_promisc_2g(port, promisc); + + if (promisc) + return; + + /* Restore allmulti on vlans if necessary */ + cpsw_ale_set_allmulti(common->ale, + ndev->flags & IFF_ALLMULTI, port->port_id); + + port_mask = ALE_PORT_HOST; + /* Clear all mcast from ALE */ + cpsw_ale_flush_multicast(common->ale, port_mask, -1); + + if (!netdev_mc_empty(ndev)) { + struct netdev_hw_addr *ha; + + /* program multicast address list into ALE register */ + netdev_for_each_mc_addr(ha, ndev) { + cpsw_ale_add_mcast(common->ale, ha->addr, + port_mask, 0, 0, 0); + } + } +} + +static void am65_cpsw_nuss_ndo_host_tx_timeout(struct net_device *ndev, + unsigned int txqueue) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_tx_chn *tx_chn; + struct netdev_queue *netif_txq; + unsigned long trans_start; + + netif_txq = netdev_get_tx_queue(ndev, txqueue); + tx_chn = &common->tx_chns[txqueue]; + trans_start = netif_txq->trans_start; + + netdev_err(ndev, "txq:%d DRV_XOFF:%d tmo:%u dql_avail:%d free_desc:%zu\n", + txqueue, + netif_tx_queue_stopped(netif_txq), + jiffies_to_msecs(jiffies - trans_start), + dql_avail(&netif_txq->dql), + k3_cppi_desc_pool_avail(tx_chn->desc_pool)); + + if (netif_tx_queue_stopped(netif_txq)) { + /* try recover if stopped by us */ + txq_trans_update(netif_txq); + netif_tx_wake_queue(netif_txq); + } +} + +static int am65_cpsw_nuss_rx_push(struct am65_cpsw_common *common, + struct sk_buff *skb) +{ + struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns; + struct cppi5_host_desc_t *desc_rx; + struct device *dev = common->dev; + u32 pkt_len = skb_tailroom(skb); + dma_addr_t desc_dma; + dma_addr_t buf_dma; + void *swdata; + + desc_rx = k3_cppi_desc_pool_alloc(rx_chn->desc_pool); + if (!desc_rx) { + dev_err(dev, "Failed to allocate RXFDQ descriptor\n"); + return -ENOMEM; + } + desc_dma = k3_cppi_desc_pool_virt2dma(rx_chn->desc_pool, desc_rx); + + buf_dma = dma_map_single(dev, skb->data, pkt_len, DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(dev, buf_dma))) { + k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); + dev_err(dev, "Failed to map rx skb buffer\n"); + return -EINVAL; + } + + cppi5_hdesc_init(desc_rx, CPPI5_INFO0_HDESC_EPIB_PRESENT, + AM65_CPSW_NAV_PS_DATA_SIZE); + cppi5_hdesc_attach_buf(desc_rx, 0, 0, buf_dma, skb_tailroom(skb)); + swdata = cppi5_hdesc_get_swdata(desc_rx); + *((void **)swdata) = skb; + + return k3_udma_glue_push_rx_chn(rx_chn->rx_chn, 0, desc_rx, desc_dma); +} + +void am65_cpsw_nuss_set_p0_ptype(struct am65_cpsw_common *common) +{ + struct am65_cpsw_host *host_p = am65_common_get_host(common); + u32 val, pri_map; + + /* P0 set Receive Priority Type */ + val = readl(host_p->port_base + AM65_CPSW_PORT_REG_PRI_CTL); + + if (common->pf_p0_rx_ptype_rrobin) { + val |= AM65_CPSW_PORT_REG_PRI_CTL_RX_PTYPE_RROBIN; + /* Enet Ports fifos works in fixed priority mode only, so + * reset P0_Rx_Pri_Map so all packet will go in Enet fifo 0 + */ + pri_map = 0x0; + } else { + val &= ~AM65_CPSW_PORT_REG_PRI_CTL_RX_PTYPE_RROBIN; + /* restore P0_Rx_Pri_Map */ + pri_map = 0x76543210; + } + + writel(pri_map, host_p->port_base + AM65_CPSW_PORT_REG_RX_PRI_MAP); + writel(val, host_p->port_base + AM65_CPSW_PORT_REG_PRI_CTL); +} + +static int am65_cpsw_nuss_common_open(struct am65_cpsw_common *common, + netdev_features_t features) +{ + struct am65_cpsw_host *host_p = am65_common_get_host(common); + int port_idx, i, ret; + struct sk_buff *skb; + u32 val, port_mask; + + if (common->usage_count) + return 0; + + /* Control register */ + writel(AM65_CPSW_CTL_P0_ENABLE | AM65_CPSW_CTL_P0_TX_CRC_REMOVE | + AM65_CPSW_CTL_VLAN_AWARE | AM65_CPSW_CTL_P0_RX_PAD, + common->cpsw_base + AM65_CPSW_REG_CTL); + /* Max length register */ + writel(AM65_CPSW_MAX_PACKET_SIZE, + host_p->port_base + AM65_CPSW_PORT_REG_RX_MAXLEN); + /* set base flow_id */ + writel(common->rx_flow_id_base, + host_p->port_base + AM65_CPSW_PORT0_REG_FLOW_ID_OFFSET); + /* en tx crc offload */ + if (features & NETIF_F_HW_CSUM) + writel(AM65_CPSW_P0_REG_CTL_RX_CHECKSUM_EN, + host_p->port_base + AM65_CPSW_P0_REG_CTL); + + am65_cpsw_nuss_set_p0_ptype(common); + + /* enable statistic */ + val = BIT(HOST_PORT_NUM); + for (port_idx = 0; port_idx < common->port_num; port_idx++) { + struct am65_cpsw_port *port = &common->ports[port_idx]; + + if (!port->disabled) + val |= BIT(port->port_id); + } + writel(val, common->cpsw_base + AM65_CPSW_REG_STAT_PORT_EN); + + /* disable priority elevation */ + writel(0, common->cpsw_base + AM65_CPSW_REG_PTYPE); + + cpsw_ale_start(common->ale); + + /* limit to one RX flow only */ + cpsw_ale_control_set(common->ale, HOST_PORT_NUM, + ALE_DEFAULT_THREAD_ID, 0); + cpsw_ale_control_set(common->ale, HOST_PORT_NUM, + ALE_DEFAULT_THREAD_ENABLE, 1); + if (AM65_CPSW_IS_CPSW2G(common)) + cpsw_ale_control_set(common->ale, HOST_PORT_NUM, + ALE_PORT_NOLEARN, 1); + /* switch to vlan unaware mode */ + cpsw_ale_control_set(common->ale, HOST_PORT_NUM, ALE_VLAN_AWARE, 1); + cpsw_ale_control_set(common->ale, HOST_PORT_NUM, + ALE_PORT_STATE, ALE_PORT_STATE_FORWARD); + + /* default vlan cfg: create mask based on enabled ports */ + port_mask = GENMASK(common->port_num, 0) & + ~common->disabled_ports_mask; + + cpsw_ale_add_vlan(common->ale, 0, port_mask, + port_mask, port_mask, + port_mask & ~ALE_PORT_HOST); + + for (i = 0; i < common->rx_chns.descs_num; i++) { + skb = __netdev_alloc_skb_ip_align(NULL, + AM65_CPSW_MAX_PACKET_SIZE, + GFP_KERNEL); + if (!skb) { + dev_err(common->dev, "cannot allocate skb\n"); + return -ENOMEM; + } + + ret = am65_cpsw_nuss_rx_push(common, skb); + if (ret < 0) { + dev_err(common->dev, + "cannot submit skb to channel rx, error %d\n", + ret); + kfree_skb(skb); + return ret; + } + kmemleak_not_leak(skb); + } + k3_udma_glue_enable_rx_chn(common->rx_chns.rx_chn); + + for (i = 0; i < common->tx_ch_num; i++) { + ret = k3_udma_glue_enable_tx_chn(common->tx_chns[i].tx_chn); + if (ret) + return ret; + napi_enable(&common->tx_chns[i].napi_tx); + } + + napi_enable(&common->napi_rx); + + dev_dbg(common->dev, "cpsw_nuss started\n"); + return 0; +} + +static void am65_cpsw_nuss_tx_cleanup(void *data, dma_addr_t desc_dma); +static void am65_cpsw_nuss_rx_cleanup(void *data, dma_addr_t desc_dma); + +static int am65_cpsw_nuss_common_stop(struct am65_cpsw_common *common) +{ + int i; + + if (common->usage_count != 1) + return 0; + + cpsw_ale_control_set(common->ale, HOST_PORT_NUM, + ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); + + /* shutdown tx channels */ + atomic_set(&common->tdown_cnt, common->tx_ch_num); + /* ensure new tdown_cnt value is visible */ + smp_mb__after_atomic(); + reinit_completion(&common->tdown_complete); + + for (i = 0; i < common->tx_ch_num; i++) + k3_udma_glue_tdown_tx_chn(common->tx_chns[i].tx_chn, false); + + i = wait_for_completion_timeout(&common->tdown_complete, + msecs_to_jiffies(1000)); + if (!i) + dev_err(common->dev, "tx timeout\n"); + for (i = 0; i < common->tx_ch_num; i++) + napi_disable(&common->tx_chns[i].napi_tx); + + for (i = 0; i < common->tx_ch_num; i++) { + k3_udma_glue_reset_tx_chn(common->tx_chns[i].tx_chn, + &common->tx_chns[i], + am65_cpsw_nuss_tx_cleanup); + k3_udma_glue_disable_tx_chn(common->tx_chns[i].tx_chn); + } + + k3_udma_glue_tdown_rx_chn(common->rx_chns.rx_chn, true); + napi_disable(&common->napi_rx); + + for (i = 0; i < AM65_CPSW_MAX_RX_FLOWS; i++) + k3_udma_glue_reset_rx_chn(common->rx_chns.rx_chn, i, + &common->rx_chns, + am65_cpsw_nuss_rx_cleanup, !!i); + + k3_udma_glue_disable_rx_chn(common->rx_chns.rx_chn); + + cpsw_ale_stop(common->ale); + + writel(0, common->cpsw_base + AM65_CPSW_REG_CTL); + writel(0, common->cpsw_base + AM65_CPSW_REG_STAT_PORT_EN); + + dev_dbg(common->dev, "cpsw_nuss stopped\n"); + return 0; +} + +static int am65_cpsw_nuss_ndo_slave_stop(struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + int ret; + + if (port->slave.phy) + phy_stop(port->slave.phy); + + netif_tx_stop_all_queues(ndev); + + if (port->slave.phy) { + phy_disconnect(port->slave.phy); + port->slave.phy = NULL; + } + + ret = am65_cpsw_nuss_common_stop(common); + if (ret) + return ret; + + common->usage_count--; + pm_runtime_put(common->dev); + return 0; +} + +static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + u32 port_mask; + int ret, i; + + ret = pm_runtime_get_sync(common->dev); + if (ret < 0) { + pm_runtime_put_noidle(common->dev); + return ret; + } + + /* Notify the stack of the actual queue counts. */ + ret = netif_set_real_num_tx_queues(ndev, common->tx_ch_num); + if (ret) { + dev_err(common->dev, "cannot set real number of tx queues\n"); + return ret; + } + + ret = netif_set_real_num_rx_queues(ndev, AM65_CPSW_MAX_RX_QUEUES); + if (ret) { + dev_err(common->dev, "cannot set real number of rx queues\n"); + return ret; + } + + for (i = 0; i < common->tx_ch_num; i++) + netdev_tx_reset_queue(netdev_get_tx_queue(ndev, i)); + + ret = am65_cpsw_nuss_common_open(common, ndev->features); + if (ret) + return ret; + + common->usage_count++; + + am65_cpsw_port_set_sl_mac(port, ndev->dev_addr); + + if (port->slave.mac_only) + /* enable mac-only mode on port */ + cpsw_ale_control_set(common->ale, port->port_id, + ALE_PORT_MACONLY, 1); + if (AM65_CPSW_IS_CPSW2G(common)) + cpsw_ale_control_set(common->ale, port->port_id, + ALE_PORT_NOLEARN, 1); + + port_mask = BIT(port->port_id) | ALE_PORT_HOST; + cpsw_ale_add_ucast(common->ale, ndev->dev_addr, + HOST_PORT_NUM, ALE_SECURE, 0); + cpsw_ale_add_mcast(common->ale, ndev->broadcast, + port_mask, 0, 0, ALE_MCAST_FWD_2); + + /* mac_sl should be configured via phy-link interface */ + am65_cpsw_sl_ctl_reset(port); + + ret = phy_set_mode_ext(port->slave.ifphy, PHY_MODE_ETHERNET, + port->slave.phy_if); + if (ret) + goto error_cleanup; + + if (port->slave.phy_node) { + port->slave.phy = of_phy_connect(ndev, + port->slave.phy_node, + &am65_cpsw_nuss_adjust_link, + 0, port->slave.phy_if); + if (!port->slave.phy) { + dev_err(common->dev, "phy %pOF not found on slave %d\n", + port->slave.phy_node, + port->port_id); + ret = -ENODEV; + goto error_cleanup; + } + } + + phy_attached_info(port->slave.phy); + phy_start(port->slave.phy); + + return 0; + +error_cleanup: + am65_cpsw_nuss_ndo_slave_stop(ndev); + return ret; +} + +static void am65_cpsw_nuss_rx_cleanup(void *data, dma_addr_t desc_dma) +{ + struct am65_cpsw_rx_chn *rx_chn = data; + struct cppi5_host_desc_t *desc_rx; + struct sk_buff *skb; + dma_addr_t buf_dma; + u32 buf_dma_len; + void **swdata; + + desc_rx = k3_cppi_desc_pool_dma2virt(rx_chn->desc_pool, desc_dma); + swdata = cppi5_hdesc_get_swdata(desc_rx); + skb = *swdata; + cppi5_hdesc_get_obuf(desc_rx, &buf_dma, &buf_dma_len); + + dma_unmap_single(rx_chn->dev, buf_dma, buf_dma_len, DMA_FROM_DEVICE); + k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); + + dev_kfree_skb_any(skb); +} + +/* RX psdata[2] word format - checksum information */ +#define AM65_CPSW_RX_PSD_CSUM_ADD GENMASK(15, 0) +#define AM65_CPSW_RX_PSD_CSUM_ERR BIT(16) +#define AM65_CPSW_RX_PSD_IS_FRAGMENT BIT(17) +#define AM65_CPSW_RX_PSD_IS_TCP BIT(18) +#define AM65_CPSW_RX_PSD_IPV6_VALID BIT(19) +#define AM65_CPSW_RX_PSD_IPV4_VALID BIT(20) + +static void am65_cpsw_nuss_rx_csum(struct sk_buff *skb, u32 csum_info) +{ + /* HW can verify IPv4/IPv6 TCP/UDP packets checksum + * csum information provides in psdata[2] word: + * AM65_CPSW_RX_PSD_CSUM_ERR bit - indicates csum error + * AM65_CPSW_RX_PSD_IPV6_VALID and AM65_CPSW_RX_PSD_IPV4_VALID + * bits - indicates IPv4/IPv6 packet + * AM65_CPSW_RX_PSD_IS_FRAGMENT bit - indicates fragmented packet + * AM65_CPSW_RX_PSD_CSUM_ADD has value 0xFFFF for non fragmented packets + * or csum value for fragmented packets if !AM65_CPSW_RX_PSD_CSUM_ERR + */ + skb_checksum_none_assert(skb); + + if (unlikely(!(skb->dev->features & NETIF_F_RXCSUM))) + return; + + if ((csum_info & (AM65_CPSW_RX_PSD_IPV6_VALID | + AM65_CPSW_RX_PSD_IPV4_VALID)) && + !(csum_info & AM65_CPSW_RX_PSD_CSUM_ERR)) { + /* csum for fragmented packets is unsupported */ + if (!(csum_info & AM65_CPSW_RX_PSD_IS_FRAGMENT)) + skb->ip_summed = CHECKSUM_UNNECESSARY; + } +} + +static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_common *common, + u32 flow_idx) +{ + struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns; + u32 buf_dma_len, pkt_len, port_id = 0, csum_info; + struct am65_cpsw_ndev_priv *ndev_priv; + struct am65_cpsw_ndev_stats *stats; + struct cppi5_host_desc_t *desc_rx; + struct device *dev = common->dev; + struct sk_buff *skb, *new_skb; + dma_addr_t desc_dma, buf_dma; + struct am65_cpsw_port *port; + struct net_device *ndev; + void **swdata; + u32 *psdata; + int ret = 0; + + ret = k3_udma_glue_pop_rx_chn(rx_chn->rx_chn, flow_idx, &desc_dma); + if (ret) { + if (ret != -ENODATA) + dev_err(dev, "RX: pop chn fail %d\n", ret); + return ret; + } + + if (desc_dma & 0x1) { + dev_dbg(dev, "%s RX tdown flow: %u\n", __func__, flow_idx); + return 0; + } + + desc_rx = k3_cppi_desc_pool_dma2virt(rx_chn->desc_pool, desc_dma); + dev_dbg(dev, "%s flow_idx: %u desc %pad\n", + __func__, flow_idx, &desc_dma); + + swdata = cppi5_hdesc_get_swdata(desc_rx); + skb = *swdata; + cppi5_hdesc_get_obuf(desc_rx, &buf_dma, &buf_dma_len); + pkt_len = cppi5_hdesc_get_pktlen(desc_rx); + cppi5_desc_get_tags_ids(&desc_rx->hdr, &port_id, NULL); + dev_dbg(dev, "%s rx port_id:%d\n", __func__, port_id); + port = am65_common_get_port(common, port_id); + ndev = port->ndev; + skb->dev = ndev; + + psdata = cppi5_hdesc_get_psdata(desc_rx); + csum_info = psdata[2]; + dev_dbg(dev, "%s rx csum_info:%#x\n", __func__, csum_info); + + dma_unmap_single(dev, buf_dma, buf_dma_len, DMA_FROM_DEVICE); + + k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); + + new_skb = netdev_alloc_skb_ip_align(ndev, AM65_CPSW_MAX_PACKET_SIZE); + if (new_skb) { + skb_put(skb, pkt_len); + skb->protocol = eth_type_trans(skb, ndev); + am65_cpsw_nuss_rx_csum(skb, csum_info); + napi_gro_receive(&common->napi_rx, skb); + + ndev_priv = netdev_priv(ndev); + stats = this_cpu_ptr(ndev_priv->stats); + + u64_stats_update_begin(&stats->syncp); + stats->rx_packets++; + stats->rx_bytes += pkt_len; + u64_stats_update_end(&stats->syncp); + kmemleak_not_leak(new_skb); + } else { + ndev->stats.rx_dropped++; + new_skb = skb; + } + + if (netif_dormant(ndev)) { + dev_kfree_skb_any(new_skb); + ndev->stats.rx_dropped++; + return 0; + } + + ret = am65_cpsw_nuss_rx_push(common, new_skb); + if (WARN_ON(ret < 0)) { + dev_kfree_skb_any(new_skb); + ndev->stats.rx_errors++; + ndev->stats.rx_dropped++; + } + + return ret; +} + +static int am65_cpsw_nuss_rx_poll(struct napi_struct *napi_rx, int budget) +{ + struct am65_cpsw_common *common = am65_cpsw_napi_to_common(napi_rx); + int flow = AM65_CPSW_MAX_RX_FLOWS; + int cur_budget, ret; + int num_rx = 0; + + /* process every flow */ + while (flow--) { + cur_budget = budget - num_rx; + + while (cur_budget--) { + ret = am65_cpsw_nuss_rx_packets(common, flow); + if (ret) + break; + num_rx++; + } + + if (num_rx >= budget) + break; + } + + dev_dbg(common->dev, "%s num_rx:%d %d\n", __func__, num_rx, budget); + + if (num_rx < budget && napi_complete_done(napi_rx, num_rx)) + enable_irq(common->rx_chns.irq); + + return num_rx; +} + +static void am65_cpsw_nuss_xmit_free(struct am65_cpsw_tx_chn *tx_chn, + struct device *dev, + struct cppi5_host_desc_t *desc) +{ + struct cppi5_host_desc_t *first_desc, *next_desc; + dma_addr_t buf_dma, next_desc_dma; + u32 buf_dma_len; + + first_desc = desc; + next_desc = first_desc; + + cppi5_hdesc_get_obuf(first_desc, &buf_dma, &buf_dma_len); + + dma_unmap_single(dev, buf_dma, buf_dma_len, + DMA_TO_DEVICE); + + next_desc_dma = cppi5_hdesc_get_next_hbdesc(first_desc); + while (next_desc_dma) { + next_desc = k3_cppi_desc_pool_dma2virt(tx_chn->desc_pool, + next_desc_dma); + cppi5_hdesc_get_obuf(next_desc, &buf_dma, &buf_dma_len); + + dma_unmap_page(dev, buf_dma, buf_dma_len, + DMA_TO_DEVICE); + + next_desc_dma = cppi5_hdesc_get_next_hbdesc(next_desc); + + k3_cppi_desc_pool_free(tx_chn->desc_pool, next_desc); + } + + k3_cppi_desc_pool_free(tx_chn->desc_pool, first_desc); +} + +static void am65_cpsw_nuss_tx_cleanup(void *data, dma_addr_t desc_dma) +{ + struct am65_cpsw_tx_chn *tx_chn = data; + struct cppi5_host_desc_t *desc_tx; + struct sk_buff *skb; + void **swdata; + + desc_tx = k3_cppi_desc_pool_dma2virt(tx_chn->desc_pool, desc_dma); + swdata = cppi5_hdesc_get_swdata(desc_tx); + skb = *(swdata); + am65_cpsw_nuss_xmit_free(tx_chn, tx_chn->common->dev, desc_tx); + + dev_kfree_skb_any(skb); +} + +static int am65_cpsw_nuss_tx_compl_packets(struct am65_cpsw_common *common, + int chn, unsigned int budget) +{ + struct cppi5_host_desc_t *desc_tx; + struct device *dev = common->dev; + struct am65_cpsw_tx_chn *tx_chn; + struct netdev_queue *netif_txq; + unsigned int total_bytes = 0; + struct net_device *ndev; + struct sk_buff *skb; + dma_addr_t desc_dma; + int res, num_tx = 0; + void **swdata; + + tx_chn = &common->tx_chns[chn]; + + while (true) { + struct am65_cpsw_ndev_priv *ndev_priv; + struct am65_cpsw_ndev_stats *stats; + + res = k3_udma_glue_pop_tx_chn(tx_chn->tx_chn, &desc_dma); + if (res == -ENODATA) + break; + + if (desc_dma & 0x1) { + if (atomic_dec_and_test(&common->tdown_cnt)) + complete(&common->tdown_complete); + break; + } + + desc_tx = k3_cppi_desc_pool_dma2virt(tx_chn->desc_pool, + desc_dma); + swdata = cppi5_hdesc_get_swdata(desc_tx); + skb = *(swdata); + am65_cpsw_nuss_xmit_free(tx_chn, dev, desc_tx); + + ndev = skb->dev; + + ndev_priv = netdev_priv(ndev); + stats = this_cpu_ptr(ndev_priv->stats); + u64_stats_update_begin(&stats->syncp); + stats->tx_packets++; + stats->tx_bytes += skb->len; + u64_stats_update_end(&stats->syncp); + + total_bytes += skb->len; + napi_consume_skb(skb, budget); + num_tx++; + } + + if (!num_tx) + return 0; + + netif_txq = netdev_get_tx_queue(ndev, chn); + + netdev_tx_completed_queue(netif_txq, num_tx, total_bytes); + + if (netif_tx_queue_stopped(netif_txq)) { + /* Check whether the queue is stopped due to stalled tx dma, + * if the queue is stopped then wake the queue as + * we have free desc for tx + */ + __netif_tx_lock(netif_txq, smp_processor_id()); + if (netif_running(ndev) && + (k3_cppi_desc_pool_avail(tx_chn->desc_pool) >= + MAX_SKB_FRAGS)) + netif_tx_wake_queue(netif_txq); + + __netif_tx_unlock(netif_txq); + } + dev_dbg(dev, "%s:%u pkt:%d\n", __func__, chn, num_tx); + + return num_tx; +} + +static int am65_cpsw_nuss_tx_poll(struct napi_struct *napi_tx, int budget) +{ + struct am65_cpsw_tx_chn *tx_chn = am65_cpsw_napi_to_tx_chn(napi_tx); + int num_tx; + + num_tx = am65_cpsw_nuss_tx_compl_packets(tx_chn->common, tx_chn->id, + budget); + num_tx = min(num_tx, budget); + if (num_tx < budget) { + napi_complete(napi_tx); + enable_irq(tx_chn->irq); + } + + return num_tx; +} + +static irqreturn_t am65_cpsw_nuss_rx_irq(int irq, void *dev_id) +{ + struct am65_cpsw_common *common = dev_id; + + disable_irq_nosync(irq); + napi_schedule(&common->napi_rx); + + return IRQ_HANDLED; +} + +static irqreturn_t am65_cpsw_nuss_tx_irq(int irq, void *dev_id) +{ + struct am65_cpsw_tx_chn *tx_chn = dev_id; + + disable_irq_nosync(irq); + napi_schedule(&tx_chn->napi_tx); + + return IRQ_HANDLED; +} + +static netdev_tx_t am65_cpsw_nuss_ndo_slave_xmit(struct sk_buff *skb, + struct net_device *ndev) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct cppi5_host_desc_t *first_desc, *next_desc, *cur_desc; + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct device *dev = common->dev; + struct am65_cpsw_tx_chn *tx_chn; + struct netdev_queue *netif_txq; + dma_addr_t desc_dma, buf_dma; + int ret, q_idx, i; + void **swdata; + u32 *psdata; + u32 pkt_len; + + /* padding enabled in hw */ + pkt_len = skb_headlen(skb); + + q_idx = skb_get_queue_mapping(skb); + dev_dbg(dev, "%s skb_queue:%d\n", __func__, q_idx); + + tx_chn = &common->tx_chns[q_idx]; + netif_txq = netdev_get_tx_queue(ndev, q_idx); + + /* Map the linear buffer */ + buf_dma = dma_map_single(dev, skb->data, pkt_len, + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, buf_dma))) { + dev_err(dev, "Failed to map tx skb buffer\n"); + ndev->stats.tx_errors++; + goto err_free_skb; + } + + first_desc = k3_cppi_desc_pool_alloc(tx_chn->desc_pool); + if (!first_desc) { + dev_dbg(dev, "Failed to allocate descriptor\n"); + dma_unmap_single(dev, buf_dma, pkt_len, DMA_TO_DEVICE); + goto busy_stop_q; + } + + cppi5_hdesc_init(first_desc, CPPI5_INFO0_HDESC_EPIB_PRESENT, + AM65_CPSW_NAV_PS_DATA_SIZE); + cppi5_desc_set_pktids(&first_desc->hdr, 0, 0x3FFF); + cppi5_hdesc_set_pkttype(first_desc, 0x7); + cppi5_desc_set_tags_ids(&first_desc->hdr, 0, port->port_id); + + cppi5_hdesc_attach_buf(first_desc, buf_dma, pkt_len, buf_dma, pkt_len); + swdata = cppi5_hdesc_get_swdata(first_desc); + *(swdata) = skb; + psdata = cppi5_hdesc_get_psdata(first_desc); + + /* HW csum offload if enabled */ + psdata[2] = 0; + if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) { + unsigned int cs_start, cs_offset; + + cs_start = skb_transport_offset(skb); + cs_offset = cs_start + skb->csum_offset; + /* HW numerates bytes starting from 1 */ + psdata[2] = ((cs_offset + 1) << 24) | + ((cs_start + 1) << 16) | (skb->len - cs_start); + dev_dbg(dev, "%s tx psdata:%#x\n", __func__, psdata[2]); + } + + if (!skb_is_nonlinear(skb)) + goto done_tx; + + dev_dbg(dev, "fragmented SKB\n"); + + /* Handle the case where skb is fragmented in pages */ + cur_desc = first_desc; + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + u32 frag_size = skb_frag_size(frag); + + next_desc = k3_cppi_desc_pool_alloc(tx_chn->desc_pool); + if (!next_desc) { + dev_err(dev, "Failed to allocate descriptor\n"); + goto busy_free_descs; + } + + buf_dma = skb_frag_dma_map(dev, frag, 0, frag_size, + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, buf_dma))) { + dev_err(dev, "Failed to map tx skb page\n"); + k3_cppi_desc_pool_free(tx_chn->desc_pool, next_desc); + ndev->stats.tx_errors++; + goto err_free_descs; + } + + cppi5_hdesc_reset_hbdesc(next_desc); + cppi5_hdesc_attach_buf(next_desc, + buf_dma, frag_size, buf_dma, frag_size); + + desc_dma = k3_cppi_desc_pool_virt2dma(tx_chn->desc_pool, + next_desc); + cppi5_hdesc_link_hbdesc(cur_desc, desc_dma); + + pkt_len += frag_size; + cur_desc = next_desc; + } + WARN_ON(pkt_len != skb->len); + +done_tx: + skb_tx_timestamp(skb); + + /* report bql before sending packet */ + netdev_tx_sent_queue(netif_txq, pkt_len); + + cppi5_hdesc_set_pktlen(first_desc, pkt_len); + desc_dma = k3_cppi_desc_pool_virt2dma(tx_chn->desc_pool, first_desc); + ret = k3_udma_glue_push_tx_chn(tx_chn->tx_chn, first_desc, desc_dma); + if (ret) { + dev_err(dev, "can't push desc %d\n", ret); + /* inform bql */ + netdev_tx_completed_queue(netif_txq, 1, pkt_len); + ndev->stats.tx_errors++; + goto err_free_descs; + } + + if (k3_cppi_desc_pool_avail(tx_chn->desc_pool) < MAX_SKB_FRAGS) { + netif_tx_stop_queue(netif_txq); + /* Barrier, so that stop_queue visible to other cpus */ + smp_mb__after_atomic(); + dev_dbg(dev, "netif_tx_stop_queue %d\n", q_idx); + + /* re-check for smp */ + if (k3_cppi_desc_pool_avail(tx_chn->desc_pool) >= + MAX_SKB_FRAGS) { + netif_tx_wake_queue(netif_txq); + dev_dbg(dev, "netif_tx_wake_queue %d\n", q_idx); + } + } + + return NETDEV_TX_OK; + +err_free_descs: + am65_cpsw_nuss_xmit_free(tx_chn, dev, first_desc); +err_free_skb: + ndev->stats.tx_dropped++; + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + +busy_free_descs: + am65_cpsw_nuss_xmit_free(tx_chn, dev, first_desc); +busy_stop_q: + netif_tx_stop_queue(netif_txq); + return NETDEV_TX_BUSY; +} + +static int am65_cpsw_nuss_ndo_slave_set_mac_address(struct net_device *ndev, + void *addr) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct sockaddr *sockaddr = (struct sockaddr *)addr; + int ret; + + ret = eth_prepare_mac_addr_change(ndev, addr); + if (ret < 0) + return ret; + + ret = pm_runtime_get_sync(common->dev); + if (ret < 0) { + pm_runtime_put_noidle(common->dev); + return ret; + } + + cpsw_ale_del_ucast(common->ale, ndev->dev_addr, + HOST_PORT_NUM, 0, 0); + cpsw_ale_add_ucast(common->ale, sockaddr->sa_data, + HOST_PORT_NUM, ALE_SECURE, 0); + + am65_cpsw_port_set_sl_mac(port, addr); + eth_commit_mac_addr_change(ndev, sockaddr); + + pm_runtime_put(common->dev); + + return 0; +} + +static int am65_cpsw_nuss_ndo_slave_ioctl(struct net_device *ndev, + struct ifreq *req, int cmd) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + + if (!netif_running(ndev)) + return -EINVAL; + + if (!port->slave.phy) + return -EOPNOTSUPP; + + return phy_mii_ioctl(port->slave.phy, req, cmd); +} + +static void am65_cpsw_nuss_ndo_get_stats(struct net_device *dev, + struct rtnl_link_stats64 *stats) +{ + struct am65_cpsw_ndev_priv *ndev_priv = netdev_priv(dev); + unsigned int start; + int cpu; + + for_each_possible_cpu(cpu) { + struct am65_cpsw_ndev_stats *cpu_stats; + u64 rx_packets; + u64 rx_bytes; + u64 tx_packets; + u64 tx_bytes; + + cpu_stats = per_cpu_ptr(ndev_priv->stats, cpu); + do { + start = u64_stats_fetch_begin_irq(&cpu_stats->syncp); + rx_packets = cpu_stats->rx_packets; + rx_bytes = cpu_stats->rx_bytes; + tx_packets = cpu_stats->tx_packets; + tx_bytes = cpu_stats->tx_bytes; + } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); + + stats->rx_packets += rx_packets; + stats->rx_bytes += rx_bytes; + stats->tx_packets += tx_packets; + stats->tx_bytes += tx_bytes; + } + + stats->rx_errors = dev->stats.rx_errors; + stats->rx_dropped = dev->stats.rx_dropped; + stats->tx_dropped = dev->stats.tx_dropped; +} + +static int am65_cpsw_nuss_ndo_slave_set_features(struct net_device *ndev, + netdev_features_t features) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + netdev_features_t changes = features ^ ndev->features; + struct am65_cpsw_host *host_p; + + host_p = am65_common_get_host(common); + + if (changes & NETIF_F_HW_CSUM) { + bool enable = !!(features & NETIF_F_HW_CSUM); + + dev_info(common->dev, "Turn %s tx-checksum-ip-generic\n", + enable ? "ON" : "OFF"); + if (enable) + writel(AM65_CPSW_P0_REG_CTL_RX_CHECKSUM_EN, + host_p->port_base + AM65_CPSW_P0_REG_CTL); + else + writel(0, + host_p->port_base + AM65_CPSW_P0_REG_CTL); + } + + return 0; +} + +static const struct net_device_ops am65_cpsw_nuss_netdev_ops_2g = { + .ndo_open = am65_cpsw_nuss_ndo_slave_open, + .ndo_stop = am65_cpsw_nuss_ndo_slave_stop, + .ndo_start_xmit = am65_cpsw_nuss_ndo_slave_xmit, + .ndo_set_rx_mode = am65_cpsw_nuss_ndo_slave_set_rx_mode, + .ndo_get_stats64 = am65_cpsw_nuss_ndo_get_stats, + .ndo_validate_addr = eth_validate_addr, + .ndo_set_mac_address = am65_cpsw_nuss_ndo_slave_set_mac_address, + .ndo_tx_timeout = am65_cpsw_nuss_ndo_host_tx_timeout, + .ndo_vlan_rx_add_vid = am65_cpsw_nuss_ndo_slave_add_vid, + .ndo_vlan_rx_kill_vid = am65_cpsw_nuss_ndo_slave_kill_vid, + .ndo_do_ioctl = am65_cpsw_nuss_ndo_slave_ioctl, + .ndo_set_features = am65_cpsw_nuss_ndo_slave_set_features, +}; + +static void am65_cpsw_nuss_slave_disable_unused(struct am65_cpsw_port *port) +{ + struct am65_cpsw_common *common = port->common; + + if (!port->disabled) + return; + + common->disabled_ports_mask |= BIT(port->port_id); + cpsw_ale_control_set(common->ale, port->port_id, + ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); + + cpsw_sl_reset(port->slave.mac_sl, 100); + cpsw_sl_ctl_reset(port->slave.mac_sl); +} + +static void am65_cpsw_nuss_free_tx_chns(void *data) +{ + struct am65_cpsw_common *common = data; + int i; + + for (i = 0; i < common->tx_ch_num; i++) { + struct am65_cpsw_tx_chn *tx_chn = &common->tx_chns[i]; + + if (!IS_ERR_OR_NULL(tx_chn->tx_chn)) + k3_udma_glue_release_tx_chn(tx_chn->tx_chn); + + if (!IS_ERR_OR_NULL(tx_chn->desc_pool)) + k3_cppi_desc_pool_destroy(tx_chn->desc_pool); + + memset(tx_chn, 0, sizeof(*tx_chn)); + } +} + +void am65_cpsw_nuss_remove_tx_chns(struct am65_cpsw_common *common) +{ + struct device *dev = common->dev; + int i; + + devm_remove_action(dev, am65_cpsw_nuss_free_tx_chns, common); + + for (i = 0; i < common->tx_ch_num; i++) { + struct am65_cpsw_tx_chn *tx_chn = &common->tx_chns[i]; + + if (tx_chn->irq) + devm_free_irq(dev, tx_chn->irq, tx_chn); + + netif_napi_del(&tx_chn->napi_tx); + + if (!IS_ERR_OR_NULL(tx_chn->tx_chn)) + k3_udma_glue_release_tx_chn(tx_chn->tx_chn); + + if (!IS_ERR_OR_NULL(tx_chn->desc_pool)) + k3_cppi_desc_pool_destroy(tx_chn->desc_pool); + + memset(tx_chn, 0, sizeof(*tx_chn)); + } +} + +static int am65_cpsw_nuss_init_tx_chns(struct am65_cpsw_common *common) +{ + u32 max_desc_num = ALIGN(AM65_CPSW_MAX_TX_DESC, MAX_SKB_FRAGS); + struct k3_udma_glue_tx_channel_cfg tx_cfg = { 0 }; + struct device *dev = common->dev; + struct k3_ring_cfg ring_cfg = { + .elm_size = K3_RINGACC_RING_ELSIZE_8, + .mode = K3_RINGACC_RING_MODE_RING, + .flags = 0 + }; + u32 hdesc_size; + int i, ret = 0; + + hdesc_size = cppi5_hdesc_calc_size(true, AM65_CPSW_NAV_PS_DATA_SIZE, + AM65_CPSW_NAV_SW_DATA_SIZE); + + tx_cfg.swdata_size = AM65_CPSW_NAV_SW_DATA_SIZE; + tx_cfg.tx_cfg = ring_cfg; + tx_cfg.txcq_cfg = ring_cfg; + tx_cfg.tx_cfg.size = max_desc_num; + tx_cfg.txcq_cfg.size = max_desc_num; + + for (i = 0; i < common->tx_ch_num; i++) { + struct am65_cpsw_tx_chn *tx_chn = &common->tx_chns[i]; + + snprintf(tx_chn->tx_chn_name, + sizeof(tx_chn->tx_chn_name), "tx%d", i); + + tx_chn->common = common; + tx_chn->id = i; + tx_chn->descs_num = max_desc_num; + tx_chn->desc_pool = + k3_cppi_desc_pool_create_name(dev, + tx_chn->descs_num, + hdesc_size, + tx_chn->tx_chn_name); + if (IS_ERR(tx_chn->desc_pool)) { + ret = PTR_ERR(tx_chn->desc_pool); + dev_err(dev, "Failed to create poll %d\n", ret); + goto err; + } + + tx_chn->tx_chn = + k3_udma_glue_request_tx_chn(dev, + tx_chn->tx_chn_name, + &tx_cfg); + if (IS_ERR(tx_chn->tx_chn)) { + ret = PTR_ERR(tx_chn->tx_chn); + dev_err(dev, "Failed to request tx dma channel %d\n", + ret); + goto err; + } + + tx_chn->irq = k3_udma_glue_tx_get_irq(tx_chn->tx_chn); + if (tx_chn->irq <= 0) { + dev_err(dev, "Failed to get tx dma irq %d\n", + tx_chn->irq); + goto err; + } + + snprintf(tx_chn->tx_chn_name, + sizeof(tx_chn->tx_chn_name), "%s-tx%d", + dev_name(dev), tx_chn->id); + } + +err: + i = devm_add_action(dev, am65_cpsw_nuss_free_tx_chns, common); + if (i) { + dev_err(dev, "failed to add free_tx_chns action %d", i); + return i; + } + + return ret; +} + +static void am65_cpsw_nuss_free_rx_chns(void *data) +{ + struct am65_cpsw_common *common = data; + struct am65_cpsw_rx_chn *rx_chn; + + rx_chn = &common->rx_chns; + + if (!IS_ERR_OR_NULL(rx_chn->rx_chn)) + k3_udma_glue_release_rx_chn(rx_chn->rx_chn); + + if (!IS_ERR_OR_NULL(rx_chn->desc_pool)) + k3_cppi_desc_pool_destroy(rx_chn->desc_pool); +} + +static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common) +{ + struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns; + struct k3_udma_glue_rx_channel_cfg rx_cfg = { 0 }; + u32 max_desc_num = AM65_CPSW_MAX_RX_DESC; + struct device *dev = common->dev; + u32 hdesc_size; + u32 fdqring_id; + int i, ret = 0; + + hdesc_size = cppi5_hdesc_calc_size(true, AM65_CPSW_NAV_PS_DATA_SIZE, + AM65_CPSW_NAV_SW_DATA_SIZE); + + rx_cfg.swdata_size = AM65_CPSW_NAV_SW_DATA_SIZE; + rx_cfg.flow_id_num = AM65_CPSW_MAX_RX_FLOWS; + rx_cfg.flow_id_base = common->rx_flow_id_base; + + /* init all flows */ + rx_chn->dev = dev; + rx_chn->descs_num = max_desc_num; + rx_chn->desc_pool = k3_cppi_desc_pool_create_name(dev, + rx_chn->descs_num, + hdesc_size, "rx"); + if (IS_ERR(rx_chn->desc_pool)) { + ret = PTR_ERR(rx_chn->desc_pool); + dev_err(dev, "Failed to create rx poll %d\n", ret); + goto err; + } + + rx_chn->rx_chn = k3_udma_glue_request_rx_chn(dev, "rx", &rx_cfg); + if (IS_ERR(rx_chn->rx_chn)) { + ret = PTR_ERR(rx_chn->rx_chn); + dev_err(dev, "Failed to request rx dma channel %d\n", ret); + goto err; + } + + common->rx_flow_id_base = + k3_udma_glue_rx_get_flow_id_base(rx_chn->rx_chn); + dev_info(dev, "set new flow-id-base %u\n", common->rx_flow_id_base); + + fdqring_id = K3_RINGACC_RING_ID_ANY; + for (i = 0; i < rx_cfg.flow_id_num; i++) { + struct k3_ring_cfg rxring_cfg = { + .elm_size = K3_RINGACC_RING_ELSIZE_8, + .mode = K3_RINGACC_RING_MODE_RING, + .flags = 0, + }; + struct k3_ring_cfg fdqring_cfg = { + .elm_size = K3_RINGACC_RING_ELSIZE_8, + .mode = K3_RINGACC_RING_MODE_MESSAGE, + .flags = K3_RINGACC_RING_SHARED, + }; + struct k3_udma_glue_rx_flow_cfg rx_flow_cfg = { + .rx_cfg = rxring_cfg, + .rxfdq_cfg = fdqring_cfg, + .ring_rxq_id = K3_RINGACC_RING_ID_ANY, + .src_tag_lo_sel = + K3_UDMA_GLUE_SRC_TAG_LO_USE_REMOTE_SRC_TAG, + }; + + rx_flow_cfg.ring_rxfdq0_id = fdqring_id; + rx_flow_cfg.rx_cfg.size = max_desc_num; + rx_flow_cfg.rxfdq_cfg.size = max_desc_num; + + ret = k3_udma_glue_rx_flow_init(rx_chn->rx_chn, + i, &rx_flow_cfg); + if (ret) { + dev_err(dev, "Failed to init rx flow%d %d\n", i, ret); + goto err; + } + if (!i) + fdqring_id = + k3_udma_glue_rx_flow_get_fdq_id(rx_chn->rx_chn, + i); + + rx_chn->irq = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i); + + if (rx_chn->irq <= 0) { + dev_err(dev, "Failed to get rx dma irq %d\n", + rx_chn->irq); + ret = -ENXIO; + goto err; + } + } + +err: + i = devm_add_action(dev, am65_cpsw_nuss_free_rx_chns, common); + if (i) { + dev_err(dev, "failed to add free_rx_chns action %d", i); + return i; + } + + return ret; +} + +static int am65_cpsw_nuss_init_host_p(struct am65_cpsw_common *common) +{ + struct am65_cpsw_host *host_p = am65_common_get_host(common); + + host_p->common = common; + host_p->port_base = common->cpsw_base + AM65_CPSW_NU_PORTS_BASE; + host_p->stat_base = common->cpsw_base + AM65_CPSW_NU_STATS_BASE; + + return 0; +} + +static int am65_cpsw_am654_get_efuse_macid(struct device_node *of_node, + int slave, u8 *mac_addr) +{ + u32 mac_lo, mac_hi, offset; + struct regmap *syscon; + int ret; + + syscon = syscon_regmap_lookup_by_phandle(of_node, "ti,syscon-efuse"); + if (IS_ERR(syscon)) { + if (PTR_ERR(syscon) == -ENODEV) + return 0; + return PTR_ERR(syscon); + } + + ret = of_property_read_u32_index(of_node, "ti,syscon-efuse", 1, + &offset); + if (ret) + return ret; + + regmap_read(syscon, offset, &mac_lo); + regmap_read(syscon, offset + 4, &mac_hi); + + mac_addr[0] = (mac_hi >> 8) & 0xff; + mac_addr[1] = mac_hi & 0xff; + mac_addr[2] = (mac_lo >> 24) & 0xff; + mac_addr[3] = (mac_lo >> 16) & 0xff; + mac_addr[4] = (mac_lo >> 8) & 0xff; + mac_addr[5] = mac_lo & 0xff; + + return 0; +} + +static int am65_cpsw_nuss_init_slave_ports(struct am65_cpsw_common *common) +{ + struct device_node *node, *port_np; + struct device *dev = common->dev; + int ret; + + node = of_get_child_by_name(dev->of_node, "ethernet-ports"); + if (!node) + return -ENOENT; + + for_each_child_of_node(node, port_np) { + struct am65_cpsw_port *port; + const void *mac_addr; + u32 port_id; + + /* it is not a slave port node, continue */ + if (strcmp(port_np->name, "port")) + continue; + + ret = of_property_read_u32(port_np, "reg", &port_id); + if (ret < 0) { + dev_err(dev, "%pOF error reading port_id %d\n", + port_np, ret); + return ret; + } + + if (!port_id || port_id > common->port_num) { + dev_err(dev, "%pOF has invalid port_id %u %s\n", + port_np, port_id, port_np->name); + return -EINVAL; + } + + port = am65_common_get_port(common, port_id); + port->port_id = port_id; + port->common = common; + port->port_base = common->cpsw_base + AM65_CPSW_NU_PORTS_BASE + + AM65_CPSW_NU_PORTS_OFFSET * (port_id); + port->stat_base = common->cpsw_base + AM65_CPSW_NU_STATS_BASE + + (AM65_CPSW_NU_STATS_PORT_OFFSET * port_id); + port->name = of_get_property(port_np, "label", NULL); + + port->disabled = !of_device_is_available(port_np); + if (port->disabled) + continue; + + port->slave.ifphy = devm_of_phy_get(dev, port_np, NULL); + if (IS_ERR(port->slave.ifphy)) { + ret = PTR_ERR(port->slave.ifphy); + dev_err(dev, "%pOF error retrieving port phy: %d\n", + port_np, ret); + return ret; + } + + port->slave.mac_only = + of_property_read_bool(port_np, "ti,mac-only"); + + /* get phy/link info */ + if (of_phy_is_fixed_link(port_np)) { + ret = of_phy_register_fixed_link(port_np); + if (ret) { + if (ret != -EPROBE_DEFER) + dev_err(dev, "%pOF failed to register fixed-link phy: %d\n", + port_np, ret); + return ret; + } + port->slave.phy_node = of_node_get(port_np); + } else { + port->slave.phy_node = + of_parse_phandle(port_np, "phy-handle", 0); + } + + if (!port->slave.phy_node) { + dev_err(dev, + "slave[%d] no phy found\n", port_id); + return -ENODEV; + } + + ret = of_get_phy_mode(port_np, &port->slave.phy_if); + if (ret) { + dev_err(dev, "%pOF read phy-mode err %d\n", + port_np, ret); + return ret; + } + + port->slave.mac_sl = cpsw_sl_get("am65", dev, port->port_base); + if (IS_ERR(port->slave.mac_sl)) + return PTR_ERR(port->slave.mac_sl); + + mac_addr = of_get_mac_address(port_np); + if (!IS_ERR(mac_addr)) { + ether_addr_copy(port->slave.mac_addr, mac_addr); + } else if (am65_cpsw_am654_get_efuse_macid(port_np, + port->port_id, + port->slave.mac_addr) || + !is_valid_ether_addr(port->slave.mac_addr)) { + random_ether_addr(port->slave.mac_addr); + dev_err(dev, "Use random MAC address\n"); + } + } + of_node_put(node); + + return 0; +} + +static void am65_cpsw_pcpu_stats_free(void *data) +{ + struct am65_cpsw_ndev_stats __percpu *stats = data; + + free_percpu(stats); +} + +static int am65_cpsw_nuss_init_ndev_2g(struct am65_cpsw_common *common) +{ + struct am65_cpsw_ndev_priv *ndev_priv; + struct device *dev = common->dev; + struct am65_cpsw_port *port; + int ret; + + port = am65_common_get_port(common, 1); + + /* alloc netdev */ + port->ndev = devm_alloc_etherdev_mqs(common->dev, + sizeof(struct am65_cpsw_ndev_priv), + AM65_CPSW_MAX_TX_QUEUES, + AM65_CPSW_MAX_RX_QUEUES); + if (!port->ndev) { + dev_err(dev, "error allocating slave net_device %u\n", + port->port_id); + return -ENOMEM; + } + + ndev_priv = netdev_priv(port->ndev); + ndev_priv->port = port; + ndev_priv->msg_enable = AM65_CPSW_DEBUG; + SET_NETDEV_DEV(port->ndev, dev); + + ether_addr_copy(port->ndev->dev_addr, port->slave.mac_addr); + + port->ndev->min_mtu = AM65_CPSW_MIN_PACKET_SIZE; + port->ndev->max_mtu = AM65_CPSW_MAX_PACKET_SIZE; + port->ndev->hw_features = NETIF_F_SG | + NETIF_F_RXCSUM | + NETIF_F_HW_CSUM; + port->ndev->features = port->ndev->hw_features | + NETIF_F_HW_VLAN_CTAG_FILTER; + port->ndev->vlan_features |= NETIF_F_SG; + port->ndev->netdev_ops = &am65_cpsw_nuss_netdev_ops_2g; + port->ndev->ethtool_ops = &am65_cpsw_ethtool_ops_slave; + + /* Disable TX checksum offload by default due to HW bug */ + if (common->pdata->quirks & AM65_CPSW_QUIRK_I2027_NO_TX_CSUM) + port->ndev->features &= ~NETIF_F_HW_CSUM; + + ndev_priv->stats = netdev_alloc_pcpu_stats(struct am65_cpsw_ndev_stats); + if (!ndev_priv->stats) + return -ENOMEM; + + ret = devm_add_action_or_reset(dev, am65_cpsw_pcpu_stats_free, + ndev_priv->stats); + if (ret) { + dev_err(dev, "failed to add percpu stat free action %d", ret); + return ret; + } + + netif_napi_add(port->ndev, &common->napi_rx, + am65_cpsw_nuss_rx_poll, NAPI_POLL_WEIGHT); + + common->pf_p0_rx_ptype_rrobin = false; + + return ret; +} + +static int am65_cpsw_nuss_ndev_add_napi_2g(struct am65_cpsw_common *common) +{ + struct device *dev = common->dev; + struct am65_cpsw_port *port; + int i, ret = 0; + + port = am65_common_get_port(common, 1); + + for (i = 0; i < common->tx_ch_num; i++) { + struct am65_cpsw_tx_chn *tx_chn = &common->tx_chns[i]; + + netif_tx_napi_add(port->ndev, &tx_chn->napi_tx, + am65_cpsw_nuss_tx_poll, NAPI_POLL_WEIGHT); + + ret = devm_request_irq(dev, tx_chn->irq, + am65_cpsw_nuss_tx_irq, + 0, tx_chn->tx_chn_name, tx_chn); + if (ret) { + dev_err(dev, "failure requesting tx%u irq %u, %d\n", + tx_chn->id, tx_chn->irq, ret); + goto err; + } + } + +err: + return ret; +} + +static int am65_cpsw_nuss_ndev_reg_2g(struct am65_cpsw_common *common) +{ + struct device *dev = common->dev; + struct am65_cpsw_port *port; + int ret = 0; + + port = am65_common_get_port(common, 1); + ret = am65_cpsw_nuss_ndev_add_napi_2g(common); + if (ret) + goto err; + + ret = devm_request_irq(dev, common->rx_chns.irq, + am65_cpsw_nuss_rx_irq, + 0, dev_name(dev), common); + if (ret) { + dev_err(dev, "failure requesting rx irq %u, %d\n", + common->rx_chns.irq, ret); + goto err; + } + + ret = register_netdev(port->ndev); + if (ret) + dev_err(dev, "error registering slave net device %d\n", ret); + + /* can't auto unregister ndev using devm_add_action() due to + * devres release sequence in DD core for DMA + */ +err: + return ret; +} + +int am65_cpsw_nuss_update_tx_chns(struct am65_cpsw_common *common, int num_tx) +{ + int ret; + + common->tx_ch_num = num_tx; + ret = am65_cpsw_nuss_init_tx_chns(common); + if (ret) + return ret; + + return am65_cpsw_nuss_ndev_add_napi_2g(common); +} + +static void am65_cpsw_nuss_cleanup_ndev(struct am65_cpsw_common *common) +{ + struct am65_cpsw_port *port; + int i; + + for (i = 0; i < common->port_num; i++) { + port = &common->ports[i]; + if (port->ndev) + unregister_netdev(port->ndev); + } +} + +static const struct am65_cpsw_pdata am65x_sr1_0 = { + .quirks = AM65_CPSW_QUIRK_I2027_NO_TX_CSUM, +}; + +static const struct am65_cpsw_pdata j721e_sr1_0 = { + .quirks = 0, +}; + +static const struct of_device_id am65_cpsw_nuss_of_mtable[] = { + { .compatible = "ti,am654-cpsw-nuss", .data = &am65x_sr1_0 }, + { .compatible = "ti,j721e-cpsw-nuss", .data = &j721e_sr1_0 }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, am65_cpsw_nuss_of_mtable); + +static int am65_cpsw_nuss_probe(struct platform_device *pdev) +{ + struct cpsw_ale_params ale_params; + const struct of_device_id *of_id; + struct device *dev = &pdev->dev; + struct am65_cpsw_common *common; + struct device_node *node; + struct resource *res; + int ret, i; + + common = devm_kzalloc(dev, sizeof(struct am65_cpsw_common), GFP_KERNEL); + if (!common) + return -ENOMEM; + common->dev = dev; + + of_id = of_match_device(am65_cpsw_nuss_of_mtable, dev); + if (!of_id) + return -EINVAL; + common->pdata = of_id->data; + + res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "cpsw_nuss"); + common->ss_base = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(common->ss_base)) + return PTR_ERR(common->ss_base); + common->cpsw_base = common->ss_base + AM65_CPSW_CPSW_NU_BASE; + + node = of_get_child_by_name(dev->of_node, "ethernet-ports"); + if (!node) + return -ENOENT; + common->port_num = of_get_child_count(node); + if (common->port_num < 1 || common->port_num > AM65_CPSW_MAX_PORTS) + return -ENOENT; + of_node_put(node); + + if (common->port_num != 1) + return -EOPNOTSUPP; + + common->rx_flow_id_base = -1; + init_completion(&common->tdown_complete); + common->tx_ch_num = 1; + + ret = dma_coerce_mask_and_coherent(dev, DMA_BIT_MASK(48)); + if (ret) { + dev_err(dev, "error setting dma mask: %d\n", ret); + return ret; + } + + common->ports = devm_kcalloc(dev, common->port_num, + sizeof(*common->ports), + GFP_KERNEL); + if (!common->ports) + return -ENOMEM; + + pm_runtime_enable(dev); + ret = pm_runtime_get_sync(dev); + if (ret < 0) { + pm_runtime_put_noidle(dev); + pm_runtime_disable(dev); + return ret; + } + + ret = of_platform_populate(dev->of_node, NULL, NULL, dev); + /* We do not want to force this, as in some cases may not have child */ + if (ret) + dev_warn(dev, "populating child nodes err:%d\n", ret); + + am65_cpsw_nuss_get_ver(common); + + /* init tx channels */ + ret = am65_cpsw_nuss_init_tx_chns(common); + if (ret) + goto err_of_clear; + ret = am65_cpsw_nuss_init_rx_chns(common); + if (ret) + goto err_of_clear; + + ret = am65_cpsw_nuss_init_host_p(common); + if (ret) + goto err_of_clear; + + ret = am65_cpsw_nuss_init_slave_ports(common); + if (ret) + goto err_of_clear; + + /* init common data */ + ale_params.dev = dev; + ale_params.ale_ageout = AM65_CPSW_ALE_AGEOUT_DEFAULT; + ale_params.ale_entries = 0; + ale_params.ale_ports = common->port_num + 1; + ale_params.ale_regs = common->cpsw_base + AM65_CPSW_NU_ALE_BASE; + ale_params.nu_switch_ale = true; + + common->ale = cpsw_ale_create(&ale_params); + if (!common->ale) { + dev_err(dev, "error initializing ale engine\n"); + goto err_of_clear; + } + + /* init ports */ + for (i = 0; i < common->port_num; i++) + am65_cpsw_nuss_slave_disable_unused(&common->ports[i]); + + dev_set_drvdata(dev, common); + + ret = am65_cpsw_nuss_init_ndev_2g(common); + if (ret) + goto err_of_clear; + + ret = am65_cpsw_nuss_ndev_reg_2g(common); + if (ret) + goto err_of_clear; + + pm_runtime_put(dev); + return 0; + +err_of_clear: + of_platform_depopulate(dev); + pm_runtime_put_sync(dev); + pm_runtime_disable(dev); + return ret; +} + +static int am65_cpsw_nuss_remove(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct am65_cpsw_common *common; + int ret; + + common = dev_get_drvdata(dev); + + ret = pm_runtime_get_sync(&pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&pdev->dev); + return ret; + } + + /* must unregister ndevs here because DD release_driver routine calls + * dma_deconfigure(dev) before devres_release_all(dev) + */ + am65_cpsw_nuss_cleanup_ndev(common); + + of_platform_depopulate(dev); + + pm_runtime_put_sync(&pdev->dev); + pm_runtime_disable(&pdev->dev); + return 0; +} + +static struct platform_driver am65_cpsw_nuss_driver = { + .driver = { + .name = AM65_CPSW_DRV_NAME, + .of_match_table = am65_cpsw_nuss_of_mtable, + }, + .probe = am65_cpsw_nuss_probe, + .remove = am65_cpsw_nuss_remove, +}; + +module_platform_driver(am65_cpsw_nuss_driver); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Grygorii Strashko <grygorii.strashko@ti.com>"); +MODULE_DESCRIPTION("TI AM65 CPSW Ethernet driver"); diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.h b/drivers/net/ethernet/ti/am65-cpsw-nuss.h new file mode 100644 index 000000000000..41ae5b4c7931 --- /dev/null +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.h @@ -0,0 +1,142 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com/ + * + */ + +#ifndef AM65_CPSW_NUSS_H_ +#define AM65_CPSW_NUSS_H_ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/netdevice.h> + +#define HOST_PORT_NUM 0 + +#define AM65_CPSW_MAX_TX_QUEUES 8 +#define AM65_CPSW_MAX_RX_QUEUES 1 +#define AM65_CPSW_MAX_RX_FLOWS 1 + +struct am65_cpsw_slave_data { + bool mac_only; + struct cpsw_sl *mac_sl; + struct device_node *phy_node; + struct phy_device *phy; + phy_interface_t phy_if; + struct phy *ifphy; + bool rx_pause; + bool tx_pause; + u8 mac_addr[ETH_ALEN]; +}; + +struct am65_cpsw_port { + struct am65_cpsw_common *common; + struct net_device *ndev; + const char *name; + u32 port_id; + void __iomem *port_base; + void __iomem *stat_base; + bool disabled; + struct am65_cpsw_slave_data slave; +}; + +struct am65_cpsw_host { + struct am65_cpsw_common *common; + void __iomem *port_base; + void __iomem *stat_base; +}; + +struct am65_cpsw_tx_chn { + struct napi_struct napi_tx; + struct am65_cpsw_common *common; + struct k3_cppi_desc_pool *desc_pool; + struct k3_udma_glue_tx_channel *tx_chn; + int irq; + u32 id; + u32 descs_num; + char tx_chn_name[128]; +}; + +struct am65_cpsw_rx_chn { + struct device *dev; + struct k3_cppi_desc_pool *desc_pool; + struct k3_udma_glue_rx_channel *rx_chn; + u32 descs_num; + int irq; +}; + +#define AM65_CPSW_QUIRK_I2027_NO_TX_CSUM BIT(0) + +struct am65_cpsw_pdata { + u32 quirks; +}; + +struct am65_cpsw_common { + struct device *dev; + const struct am65_cpsw_pdata *pdata; + + void __iomem *ss_base; + void __iomem *cpsw_base; + + u32 port_num; + struct am65_cpsw_host host; + struct am65_cpsw_port *ports; + u32 disabled_ports_mask; + + int usage_count; /* number of opened ports */ + struct cpsw_ale *ale; + int tx_ch_num; + u32 rx_flow_id_base; + + struct am65_cpsw_tx_chn tx_chns[AM65_CPSW_MAX_TX_QUEUES]; + struct completion tdown_complete; + atomic_t tdown_cnt; + + struct am65_cpsw_rx_chn rx_chns; + struct napi_struct napi_rx; + + u32 nuss_ver; + u32 cpsw_ver; + + bool pf_p0_rx_ptype_rrobin; +}; + +struct am65_cpsw_ndev_stats { + u64 tx_packets; + u64 tx_bytes; + u64 rx_packets; + u64 rx_bytes; + struct u64_stats_sync syncp; +}; + +struct am65_cpsw_ndev_priv { + u32 msg_enable; + struct am65_cpsw_port *port; + struct am65_cpsw_ndev_stats __percpu *stats; +}; + +#define am65_ndev_to_priv(ndev) \ + ((struct am65_cpsw_ndev_priv *)netdev_priv(ndev)) +#define am65_ndev_to_port(ndev) (am65_ndev_to_priv(ndev)->port) +#define am65_ndev_to_common(ndev) (am65_ndev_to_port(ndev)->common) +#define am65_ndev_to_slave(ndev) (&am65_ndev_to_port(ndev)->slave) + +#define am65_common_get_host(common) (&(common)->host) +#define am65_common_get_port(common, id) (&(common)->ports[(id) - 1]) + +#define am65_cpsw_napi_to_common(pnapi) \ + container_of(pnapi, struct am65_cpsw_common, napi_rx) +#define am65_cpsw_napi_to_tx_chn(pnapi) \ + container_of(pnapi, struct am65_cpsw_tx_chn, napi_tx) + +#define AM65_CPSW_DRV_NAME "am65-cpsw-nuss" + +#define AM65_CPSW_IS_CPSW2G(common) ((common)->port_num == 1) + +extern const struct ethtool_ops am65_cpsw_ethtool_ops_slave; + +void am65_cpsw_nuss_adjust_link(struct net_device *ndev); +void am65_cpsw_nuss_set_p0_ptype(struct am65_cpsw_common *common); +void am65_cpsw_nuss_remove_tx_chns(struct am65_cpsw_common *common); +int am65_cpsw_nuss_update_tx_chns(struct am65_cpsw_common *common, int num_tx); + +#endif /* AM65_CPSW_NUSS_H_ */ diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 6ae4a72e6f43..c2c5bf87da01 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1211,6 +1211,7 @@ static int cpsw_set_channels(struct net_device *ndev, } static const struct ethtool_ops cpsw_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = cpsw_get_drvinfo, .get_msglevel = cpsw_get_msglevel, .set_msglevel = cpsw_set_msglevel, diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index ecdbde539eb7..0374e6936091 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -44,6 +44,8 @@ #define ALE_UNKNOWNVLAN_FORCE_UNTAG_EGRESS 0x9C #define ALE_VLAN_MASK_MUX(reg) (0xc0 + (0x4 * (reg))) +#define AM65_CPSW_ALE_THREAD_DEF_REG 0x134 + #define ALE_TABLE_WRITE BIT(31) #define ALE_TYPE_FREE 0 @@ -122,6 +124,8 @@ DEFINE_ALE_FIELD(mcast, 40, 1) DEFINE_ALE_FIELD(vlan_unreg_mcast_idx, 20, 3) DEFINE_ALE_FIELD(vlan_reg_mcast_idx, 44, 3) +#define NU_VLAN_UNREG_MCAST_IDX 1 + /* The MAC address field in the ALE entry cannot be macroized as above */ static inline void cpsw_ale_get_addr(u32 *ale_entry, u8 *addr) { @@ -455,6 +459,8 @@ int cpsw_ale_add_vlan(struct cpsw_ale *ale, u16 vid, int port_mask, int untag, cpsw_ale_set_vlan_unreg_mcast(ale_entry, unreg_mcast, ale->vlan_field_bits); } else { + cpsw_ale_set_vlan_unreg_mcast_idx(ale_entry, + NU_VLAN_UNREG_MCAST_IDX); cpsw_ale_set_vlan_mcast(ale, ale_entry, reg_mcast, unreg_mcast); } cpsw_ale_set_vlan_member_list(ale_entry, port_mask, @@ -775,6 +781,22 @@ static struct ale_control_info ale_controls[ALE_NUM_CONTROLS] = { .port_shift = 0, .bits = 1, }, + [ALE_PORT_MACONLY] = { + .name = "mac_only_port_mode", + .offset = ALE_PORTCTL, + .port_offset = 4, + .shift = 11, + .port_shift = 0, + .bits = 1, + }, + [ALE_PORT_MACONLY_CAF] = { + .name = "mac_only_port_caf", + .offset = ALE_PORTCTL, + .port_offset = 4, + .shift = 13, + .port_shift = 0, + .bits = 1, + }, [ALE_PORT_MCAST_LIMIT] = { .name = "mcast_limit", .offset = ALE_PORTCTL, @@ -823,6 +845,22 @@ static struct ale_control_info ale_controls[ALE_NUM_CONTROLS] = { .port_shift = 0, .bits = 6, }, + [ALE_DEFAULT_THREAD_ID] = { + .name = "default_thread_id", + .offset = AM65_CPSW_ALE_THREAD_DEF_REG, + .port_offset = 0, + .shift = 0, + .port_shift = 0, + .bits = 6, + }, + [ALE_DEFAULT_THREAD_ENABLE] = { + .name = "default_thread_id_enable", + .offset = AM65_CPSW_ALE_THREAD_DEF_REG, + .port_offset = 0, + .shift = 15, + .port_shift = 0, + .bits = 1, + }, }; int cpsw_ale_control_set(struct cpsw_ale *ale, int port, int control, diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h index 70d0955c2652..6a3cb6898728 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.h +++ b/drivers/net/ethernet/ti/cpsw_ale.h @@ -62,8 +62,12 @@ enum cpsw_ale_control { ALE_PORT_UNKNOWN_MCAST_FLOOD, ALE_PORT_UNKNOWN_REG_MCAST_FLOOD, ALE_PORT_UNTAGGED_EGRESS, + ALE_PORT_MACONLY, + ALE_PORT_MACONLY_CAF, ALE_PORT_BCAST_LIMIT, ALE_PORT_MCAST_LIMIT, + ALE_DEFAULT_THREAD_ID, + ALE_DEFAULT_THREAD_ENABLE, ALE_NUM_CONTROLS, }; diff --git a/drivers/net/ethernet/ti/cpsw_new.c b/drivers/net/ethernet/ti/cpsw_new.c index 71215db7934b..9209e613257d 100644 --- a/drivers/net/ethernet/ti/cpsw_new.c +++ b/drivers/net/ethernet/ti/cpsw_new.c @@ -1175,6 +1175,7 @@ static int cpsw_set_channels(struct net_device *ndev, } static const struct ethtool_ops cpsw_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = cpsw_get_drvinfo, .get_msglevel = cpsw_get_msglevel, .set_msglevel = cpsw_set_msglevel, diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index 75d4e16c692b..de282531f68b 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -481,6 +481,7 @@ static int emac_set_coalesce(struct net_device *ndev, * Ethtool support for EMAC adapter */ static const struct ethtool_ops ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, .get_drvinfo = emac_get_drvinfo, .get_link = ethtool_op_get_link, .get_coalesce = emac_get_coalesce, diff --git a/drivers/net/ethernet/ti/k3-cppi-desc-pool.c b/drivers/net/ethernet/ti/k3-cppi-desc-pool.c new file mode 100644 index 000000000000..ad7cfc1316ce --- /dev/null +++ b/drivers/net/ethernet/ti/k3-cppi-desc-pool.c @@ -0,0 +1,126 @@ +// SPDX-License-Identifier: GPL-2.0 +/* TI K3 CPPI5 descriptors pool API + * + * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/device.h> +#include <linux/dma-mapping.h> +#include <linux/err.h> +#include <linux/genalloc.h> +#include <linux/kernel.h> + +#include "k3-cppi-desc-pool.h" + +struct k3_cppi_desc_pool { + struct device *dev; + dma_addr_t dma_addr; + void *cpumem; /* dma_alloc map */ + size_t desc_size; + size_t mem_size; + size_t num_desc; + struct gen_pool *gen_pool; +}; + +void k3_cppi_desc_pool_destroy(struct k3_cppi_desc_pool *pool) +{ + if (!pool) + return; + + WARN(gen_pool_size(pool->gen_pool) != gen_pool_avail(pool->gen_pool), + "k3_knav_desc_pool size %zu != avail %zu", + gen_pool_size(pool->gen_pool), + gen_pool_avail(pool->gen_pool)); + if (pool->cpumem) + dma_free_coherent(pool->dev, pool->mem_size, pool->cpumem, + pool->dma_addr); + + gen_pool_destroy(pool->gen_pool); /* frees pool->name */ +} + +struct k3_cppi_desc_pool * +k3_cppi_desc_pool_create_name(struct device *dev, size_t size, + size_t desc_size, + const char *name) +{ + struct k3_cppi_desc_pool *pool; + const char *pool_name = NULL; + int ret = -ENOMEM; + + pool = devm_kzalloc(dev, sizeof(*pool), GFP_KERNEL); + if (!pool) + return ERR_PTR(ret); + + pool->dev = dev; + pool->desc_size = roundup_pow_of_two(desc_size); + pool->num_desc = size; + pool->mem_size = pool->num_desc * pool->desc_size; + + pool_name = kstrdup_const(name ? name : dev_name(pool->dev), + GFP_KERNEL); + if (!pool_name) + return ERR_PTR(-ENOMEM); + + pool->gen_pool = gen_pool_create(ilog2(pool->desc_size), -1); + if (IS_ERR(pool->gen_pool)) { + ret = PTR_ERR(pool->gen_pool); + dev_err(pool->dev, "pool create failed %d\n", ret); + kfree_const(pool_name); + goto gen_pool_create_fail; + } + + pool->gen_pool->name = pool_name; + + pool->cpumem = dma_alloc_coherent(pool->dev, pool->mem_size, + &pool->dma_addr, GFP_KERNEL); + + if (!pool->cpumem) + goto dma_alloc_fail; + + ret = gen_pool_add_virt(pool->gen_pool, (unsigned long)pool->cpumem, + (phys_addr_t)pool->dma_addr, pool->mem_size, + -1); + if (ret < 0) { + dev_err(pool->dev, "pool add failed %d\n", ret); + goto gen_pool_add_virt_fail; + } + + return pool; + +gen_pool_add_virt_fail: + dma_free_coherent(pool->dev, pool->mem_size, pool->cpumem, + pool->dma_addr); +dma_alloc_fail: + gen_pool_destroy(pool->gen_pool); /* frees pool->name */ +gen_pool_create_fail: + devm_kfree(pool->dev, pool); + return ERR_PTR(ret); +} + +dma_addr_t k3_cppi_desc_pool_virt2dma(struct k3_cppi_desc_pool *pool, + void *addr) +{ + return addr ? pool->dma_addr + (addr - pool->cpumem) : 0; +} + +void *k3_cppi_desc_pool_dma2virt(struct k3_cppi_desc_pool *pool, dma_addr_t dma) +{ + return dma ? pool->cpumem + (dma - pool->dma_addr) : NULL; +} + +void *k3_cppi_desc_pool_alloc(struct k3_cppi_desc_pool *pool) +{ + return (void *)gen_pool_alloc(pool->gen_pool, pool->desc_size); +} + +void k3_cppi_desc_pool_free(struct k3_cppi_desc_pool *pool, void *addr) +{ + gen_pool_free(pool->gen_pool, (unsigned long)addr, pool->desc_size); +} + +size_t k3_cppi_desc_pool_avail(struct k3_cppi_desc_pool *pool) +{ + return gen_pool_avail(pool->gen_pool) / pool->desc_size; +} diff --git a/drivers/net/ethernet/ti/k3-cppi-desc-pool.h b/drivers/net/ethernet/ti/k3-cppi-desc-pool.h new file mode 100644 index 000000000000..a7e3fa5e7b62 --- /dev/null +++ b/drivers/net/ethernet/ti/k3-cppi-desc-pool.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* TI K3 CPPI5 descriptors pool + * + * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com + */ + +#ifndef K3_CPPI_DESC_POOL_H_ +#define K3_CPPI_DESC_POOL_H_ + +#include <linux/device.h> +#include <linux/types.h> + +struct k3_cppi_desc_pool; + +void k3_cppi_desc_pool_destroy(struct k3_cppi_desc_pool *pool); +struct k3_cppi_desc_pool * +k3_cppi_desc_pool_create_name(struct device *dev, size_t size, + size_t desc_size, + const char *name); +#define k3_cppi_desc_pool_create(dev, size, desc_size) \ + k3_cppi_desc_pool_create_name(dev, size, desc_size, NULL) +dma_addr_t +k3_cppi_desc_pool_virt2dma(struct k3_cppi_desc_pool *pool, void *addr); +void * +k3_cppi_desc_pool_dma2virt(struct k3_cppi_desc_pool *pool, dma_addr_t dma); +void *k3_cppi_desc_pool_alloc(struct k3_cppi_desc_pool *pool); +void k3_cppi_desc_pool_free(struct k3_cppi_desc_pool *pool, void *addr); +size_t k3_cppi_desc_pool_avail(struct k3_cppi_desc_pool *pool); + +#endif /* K3_CPPI_DESC_POOL_H_ */ diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.h b/drivers/net/ethernet/toshiba/ps3_gelic_net.h index 805903dbddcc..68f324ed4eaf 100644 --- a/drivers/net/ethernet/toshiba/ps3_gelic_net.h +++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.h @@ -308,7 +308,7 @@ struct gelic_port { struct gelic_card *card; struct net_device *netdev; enum gelic_port_type type; - long priv[0]; /* long for alignment */ + long priv[]; /* long for alignment */ }; static inline struct gelic_card *port_to_card(struct gelic_port *p) diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_wireless.h b/drivers/net/ethernet/toshiba/ps3_gelic_wireless.h index 4041d946b649..1f203d1ae8db 100644 --- a/drivers/net/ethernet/toshiba/ps3_gelic_wireless.h +++ b/drivers/net/ethernet/toshiba/ps3_gelic_wireless.h @@ -158,7 +158,7 @@ struct gelic_eurus_scan_info { __be32 reserved2; __be32 reserved3; __be32 reserved4; - u8 elements[0]; /* ie */ + u8 elements[]; /* ie */ } __packed; /* the hypervisor returns bbs up to 16 */ diff --git a/drivers/net/ethernet/toshiba/spider_net.h b/drivers/net/ethernet/toshiba/spider_net.h index c0c68cbc898c..05b1a0736835 100644 --- a/drivers/net/ethernet/toshiba/spider_net.h +++ b/drivers/net/ethernet/toshiba/spider_net.h @@ -470,7 +470,7 @@ struct spider_net_card { struct spider_net_extra_stats spider_stats; /* Must be last item in struct */ - struct spider_net_descr darray[0]; + struct spider_net_descr darray[]; }; #endif diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index 3fd43d30b20d..b50c3ec3495b 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -367,7 +367,7 @@ struct TxFD { struct RxFD { struct FDesc fd; - struct BDesc bd[0]; /* variable length */ + struct BDesc bd[]; /* variable length */ }; struct FrFD { diff --git a/drivers/net/ethernet/via/via-velocity.c b/drivers/net/ethernet/via/via-velocity.c index 4b556b74541a..713dbc04b25b 100644 --- a/drivers/net/ethernet/via/via-velocity.c +++ b/drivers/net/ethernet/via/via-velocity.c @@ -3648,6 +3648,8 @@ static void velocity_get_ethtool_stats(struct net_device *dev, } static const struct ethtool_ops velocity_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = velocity_get_drvinfo, .get_wol = velocity_ethtool_get_wol, .set_wol = velocity_ethtool_set_wol, diff --git a/drivers/net/ethernet/xilinx/Kconfig b/drivers/net/ethernet/xilinx/Kconfig index 6304ebd8b5c6..0810af8193cb 100644 --- a/drivers/net/ethernet/xilinx/Kconfig +++ b/drivers/net/ethernet/xilinx/Kconfig @@ -32,7 +32,6 @@ config XILINX_AXI_EMAC config XILINX_LL_TEMAC tristate "Xilinx LL TEMAC (LocalLink Tri-mode Ethernet MAC) driver" - depends on PPC || MICROBLAZE || X86 || COMPILE_TEST select PHYLIB ---help--- This driver supports the Xilinx 10/100/1000 LocalLink TEMAC diff --git a/drivers/net/ethernet/xilinx/ll_temac.h b/drivers/net/ethernet/xilinx/ll_temac.h index 53fb8141f1a6..4a73127e10a6 100644 --- a/drivers/net/ethernet/xilinx/ll_temac.h +++ b/drivers/net/ethernet/xilinx/ll_temac.h @@ -369,18 +369,20 @@ struct temac_local { /* Buffer descriptors */ struct cdmac_bd *tx_bd_v; dma_addr_t tx_bd_p; + u32 tx_bd_num; struct cdmac_bd *rx_bd_v; dma_addr_t rx_bd_p; + u32 rx_bd_num; int tx_bd_ci; - int tx_bd_next; int tx_bd_tail; int rx_bd_ci; int rx_bd_tail; /* DMA channel control setup */ - u32 tx_chnl_ctrl; - u32 rx_chnl_ctrl; + u8 coalesce_count_tx; + u8 coalesce_delay_tx; u8 coalesce_count_rx; + u8 coalesce_delay_rx; struct delayed_work restart_work; }; diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 9461acec6f70..3e313e71ae36 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -58,8 +58,11 @@ #include "ll_temac.h" -#define TX_BD_NUM 64 -#define RX_BD_NUM 128 +/* Descriptors defines for Tx and Rx DMA */ +#define TX_BD_NUM_DEFAULT 64 +#define RX_BD_NUM_DEFAULT 1024 +#define TX_BD_NUM_MAX 4096 +#define RX_BD_NUM_MAX 4096 /* --------------------------------------------------------------------- * Low level register access functions @@ -301,7 +304,7 @@ static void temac_dma_bd_release(struct net_device *ndev) /* Reset Local Link (DMA) */ lp->dma_out(lp, DMA_CONTROL_REG, DMA_CONTROL_RST); - for (i = 0; i < RX_BD_NUM; i++) { + for (i = 0; i < lp->rx_bd_num; i++) { if (!lp->rx_skb[i]) break; else { @@ -312,12 +315,12 @@ static void temac_dma_bd_release(struct net_device *ndev) } if (lp->rx_bd_v) dma_free_coherent(ndev->dev.parent, - sizeof(*lp->rx_bd_v) * RX_BD_NUM, - lp->rx_bd_v, lp->rx_bd_p); + sizeof(*lp->rx_bd_v) * lp->rx_bd_num, + lp->rx_bd_v, lp->rx_bd_p); if (lp->tx_bd_v) dma_free_coherent(ndev->dev.parent, - sizeof(*lp->tx_bd_v) * TX_BD_NUM, - lp->tx_bd_v, lp->tx_bd_p); + sizeof(*lp->tx_bd_v) * lp->tx_bd_num, + lp->tx_bd_v, lp->tx_bd_p); } /** @@ -330,33 +333,33 @@ static int temac_dma_bd_init(struct net_device *ndev) dma_addr_t skb_dma_addr; int i; - lp->rx_skb = devm_kcalloc(&ndev->dev, RX_BD_NUM, sizeof(*lp->rx_skb), - GFP_KERNEL); + lp->rx_skb = devm_kcalloc(&ndev->dev, lp->rx_bd_num, + sizeof(*lp->rx_skb), GFP_KERNEL); if (!lp->rx_skb) goto out; /* allocate the tx and rx ring buffer descriptors. */ /* returns a virtual address and a physical address. */ lp->tx_bd_v = dma_alloc_coherent(ndev->dev.parent, - sizeof(*lp->tx_bd_v) * TX_BD_NUM, + sizeof(*lp->tx_bd_v) * lp->tx_bd_num, &lp->tx_bd_p, GFP_KERNEL); if (!lp->tx_bd_v) goto out; lp->rx_bd_v = dma_alloc_coherent(ndev->dev.parent, - sizeof(*lp->rx_bd_v) * RX_BD_NUM, + sizeof(*lp->rx_bd_v) * lp->rx_bd_num, &lp->rx_bd_p, GFP_KERNEL); if (!lp->rx_bd_v) goto out; - for (i = 0; i < TX_BD_NUM; i++) { + for (i = 0; i < lp->tx_bd_num; i++) { lp->tx_bd_v[i].next = cpu_to_be32(lp->tx_bd_p - + sizeof(*lp->tx_bd_v) * ((i + 1) % TX_BD_NUM)); + + sizeof(*lp->tx_bd_v) * ((i + 1) % lp->tx_bd_num)); } - for (i = 0; i < RX_BD_NUM; i++) { + for (i = 0; i < lp->rx_bd_num; i++) { lp->rx_bd_v[i].next = cpu_to_be32(lp->rx_bd_p - + sizeof(*lp->rx_bd_v) * ((i + 1) % RX_BD_NUM)); + + sizeof(*lp->rx_bd_v) * ((i + 1) % lp->rx_bd_num)); skb = netdev_alloc_skb_ip_align(ndev, XTE_MAX_JUMBO_FRAME_SIZE); @@ -376,21 +379,22 @@ static int temac_dma_bd_init(struct net_device *ndev) } /* Configure DMA channel (irq setup) */ - lp->dma_out(lp, TX_CHNL_CTRL, lp->tx_chnl_ctrl | + lp->dma_out(lp, TX_CHNL_CTRL, + lp->coalesce_delay_tx << 24 | lp->coalesce_count_tx << 16 | 0x00000400 | // Use 1 Bit Wide Counters. Currently Not Used! CHNL_CTRL_IRQ_EN | CHNL_CTRL_IRQ_ERR_EN | CHNL_CTRL_IRQ_DLY_EN | CHNL_CTRL_IRQ_COAL_EN); - lp->dma_out(lp, RX_CHNL_CTRL, lp->rx_chnl_ctrl | + lp->dma_out(lp, RX_CHNL_CTRL, + lp->coalesce_delay_rx << 24 | lp->coalesce_count_rx << 16 | CHNL_CTRL_IRQ_IOE | CHNL_CTRL_IRQ_EN | CHNL_CTRL_IRQ_ERR_EN | CHNL_CTRL_IRQ_DLY_EN | CHNL_CTRL_IRQ_COAL_EN); /* Init descriptor indexes */ lp->tx_bd_ci = 0; - lp->tx_bd_next = 0; lp->tx_bd_tail = 0; lp->rx_bd_ci = 0; - lp->rx_bd_tail = RX_BD_NUM - 1; + lp->rx_bd_tail = lp->rx_bd_num - 1; /* Enable RX DMA transfers */ wmb(); @@ -785,7 +789,7 @@ static void temac_start_xmit_done(struct net_device *ndev) ndev->stats.tx_bytes += be32_to_cpu(cur_p->len); lp->tx_bd_ci++; - if (lp->tx_bd_ci >= TX_BD_NUM) + if (lp->tx_bd_ci >= lp->tx_bd_num) lp->tx_bd_ci = 0; cur_p = &lp->tx_bd_v[lp->tx_bd_ci]; @@ -811,7 +815,7 @@ static inline int temac_check_tx_bd_space(struct temac_local *lp, int num_frag) return NETDEV_TX_BUSY; tail++; - if (tail >= TX_BD_NUM) + if (tail >= lp->tx_bd_num) tail = 0; cur_p = &lp->tx_bd_v[tail]; @@ -826,14 +830,13 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) { struct temac_local *lp = netdev_priv(ndev); struct cdmac_bd *cur_p; - dma_addr_t start_p, tail_p, skb_dma_addr; + dma_addr_t tail_p, skb_dma_addr; int ii; unsigned long num_frag; skb_frag_t *frag; num_frag = skb_shinfo(skb)->nr_frags; frag = &skb_shinfo(skb)->frags[0]; - start_p = lp->tx_bd_p + sizeof(*lp->tx_bd_v) * lp->tx_bd_tail; cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; if (temac_check_tx_bd_space(lp, num_frag + 1)) { @@ -876,7 +879,7 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) ptr_to_txbd((void *)skb, cur_p); for (ii = 0; ii < num_frag; ii++) { - if (++lp->tx_bd_tail >= TX_BD_NUM) + if (++lp->tx_bd_tail >= lp->tx_bd_num) lp->tx_bd_tail = 0; cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; @@ -886,7 +889,7 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) DMA_TO_DEVICE); if (dma_mapping_error(ndev->dev.parent, skb_dma_addr)) { if (--lp->tx_bd_tail < 0) - lp->tx_bd_tail = TX_BD_NUM - 1; + lp->tx_bd_tail = lp->tx_bd_num - 1; cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; while (--ii >= 0) { --frag; @@ -895,7 +898,7 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) skb_frag_size(frag), DMA_TO_DEVICE); if (--lp->tx_bd_tail < 0) - lp->tx_bd_tail = TX_BD_NUM - 1; + lp->tx_bd_tail = lp->tx_bd_num - 1; cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; } dma_unmap_single(ndev->dev.parent, @@ -914,7 +917,7 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) tail_p = lp->tx_bd_p + sizeof(*lp->tx_bd_v) * lp->tx_bd_tail; lp->tx_bd_tail++; - if (lp->tx_bd_tail >= TX_BD_NUM) + if (lp->tx_bd_tail >= lp->tx_bd_num) lp->tx_bd_tail = 0; skb_tx_timestamp(skb); @@ -934,7 +937,7 @@ static int ll_temac_recv_buffers_available(struct temac_local *lp) return 0; available = 1 + lp->rx_bd_tail - lp->rx_bd_ci; if (available <= 0) - available += RX_BD_NUM; + available += lp->rx_bd_num; return available; } @@ -1003,7 +1006,7 @@ static void ll_temac_recv(struct net_device *ndev) ndev->stats.rx_bytes += length; rx_bd = lp->rx_bd_ci; - if (++lp->rx_bd_ci >= RX_BD_NUM) + if (++lp->rx_bd_ci >= lp->rx_bd_num) lp->rx_bd_ci = 0; } while (rx_bd != lp->rx_bd_tail); @@ -1034,7 +1037,7 @@ static void ll_temac_recv(struct net_device *ndev) dma_addr_t skb_dma_addr; rx_bd = lp->rx_bd_tail + 1; - if (rx_bd >= RX_BD_NUM) + if (rx_bd >= lp->rx_bd_num) rx_bd = 0; bd = &lp->rx_bd_v[rx_bd]; @@ -1250,13 +1253,96 @@ static const struct attribute_group temac_attr_group = { .attrs = temac_device_attrs, }; -/* ethtool support */ +/* --------------------------------------------------------------------- + * ethtool support + */ + +static void ll_temac_ethtools_get_ringparam(struct net_device *ndev, + struct ethtool_ringparam *ering) +{ + struct temac_local *lp = netdev_priv(ndev); + + ering->rx_max_pending = RX_BD_NUM_MAX; + ering->rx_mini_max_pending = 0; + ering->rx_jumbo_max_pending = 0; + ering->tx_max_pending = TX_BD_NUM_MAX; + ering->rx_pending = lp->rx_bd_num; + ering->rx_mini_pending = 0; + ering->rx_jumbo_pending = 0; + ering->tx_pending = lp->tx_bd_num; +} + +static int ll_temac_ethtools_set_ringparam(struct net_device *ndev, + struct ethtool_ringparam *ering) +{ + struct temac_local *lp = netdev_priv(ndev); + + if (ering->rx_pending > RX_BD_NUM_MAX || + ering->rx_mini_pending || + ering->rx_jumbo_pending || + ering->rx_pending > TX_BD_NUM_MAX) + return -EINVAL; + + if (netif_running(ndev)) + return -EBUSY; + + lp->rx_bd_num = ering->rx_pending; + lp->tx_bd_num = ering->tx_pending; + return 0; +} + +static int ll_temac_ethtools_get_coalesce(struct net_device *ndev, + struct ethtool_coalesce *ec) +{ + struct temac_local *lp = netdev_priv(ndev); + + ec->rx_max_coalesced_frames = lp->coalesce_count_rx; + ec->tx_max_coalesced_frames = lp->coalesce_count_tx; + ec->rx_coalesce_usecs = (lp->coalesce_delay_rx * 512) / 100; + ec->tx_coalesce_usecs = (lp->coalesce_delay_tx * 512) / 100; + return 0; +} + +static int ll_temac_ethtools_set_coalesce(struct net_device *ndev, + struct ethtool_coalesce *ec) +{ + struct temac_local *lp = netdev_priv(ndev); + + if (netif_running(ndev)) { + netdev_err(ndev, + "Please stop netif before applying configuration\n"); + return -EFAULT; + } + + if (ec->rx_max_coalesced_frames) + lp->coalesce_count_rx = ec->rx_max_coalesced_frames; + if (ec->tx_max_coalesced_frames) + lp->coalesce_count_tx = ec->tx_max_coalesced_frames; + /* With typical LocalLink clock speed of 200 MHz and + * C_PRESCALAR=1023, each delay count corresponds to 5.12 us. + */ + if (ec->rx_coalesce_usecs) + lp->coalesce_delay_rx = + min(255U, (ec->rx_coalesce_usecs * 100) / 512); + if (ec->tx_coalesce_usecs) + lp->coalesce_delay_tx = + min(255U, (ec->tx_coalesce_usecs * 100) / 512); + + return 0; +} + static const struct ethtool_ops temac_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES, .nway_reset = phy_ethtool_nway_reset, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, .get_link_ksettings = phy_ethtool_get_link_ksettings, .set_link_ksettings = phy_ethtool_set_link_ksettings, + .get_ringparam = ll_temac_ethtools_get_ringparam, + .set_ringparam = ll_temac_ethtools_set_ringparam, + .get_coalesce = ll_temac_ethtools_get_coalesce, + .set_coalesce = ll_temac_ethtools_set_coalesce, }; static int temac_probe(struct platform_device *pdev) @@ -1300,6 +1386,8 @@ static int temac_probe(struct platform_device *pdev) lp->ndev = ndev; lp->dev = &pdev->dev; lp->options = XTE_OPTION_DEFAULTS; + lp->rx_bd_num = RX_BD_NUM_DEFAULT; + lp->tx_bd_num = TX_BD_NUM_DEFAULT; spin_lock_init(&lp->rx_lock); INIT_DELAYED_WORK(&lp->restart_work, ll_temac_restart_work_func); @@ -1364,6 +1452,14 @@ static int temac_probe(struct platform_device *pdev) /* Can checksum TCP/UDP over IPv4. */ ndev->features |= NETIF_F_IP_CSUM; + /* Defaults for IRQ delay/coalescing setup. These are + * configuration values, so does not belong in device-tree. + */ + lp->coalesce_delay_tx = 0x10; + lp->coalesce_count_tx = 0x22; + lp->coalesce_delay_rx = 0xff; + lp->coalesce_count_rx = 0x07; + /* Setup LocalLink DMA */ if (temac_np) { /* Find the DMA node, map the DMA registers, and @@ -1402,14 +1498,6 @@ static int temac_probe(struct platform_device *pdev) lp->rx_irq = irq_of_parse_and_map(dma_np, 0); lp->tx_irq = irq_of_parse_and_map(dma_np, 1); - /* Use defaults for IRQ delay/coalescing setup. These - * are configuration values, so does not belong in - * device-tree. - */ - lp->tx_chnl_ctrl = 0x10220000; - lp->rx_chnl_ctrl = 0xff070000; - lp->coalesce_count_rx = 0x07; - /* Finished with the DMA node; drop the reference */ of_node_put(dma_np); } else if (pdata) { @@ -1435,18 +1523,13 @@ static int temac_probe(struct platform_device *pdev) lp->tx_irq = platform_get_irq(pdev, 1); /* IRQ delay/coalescing setup */ - if (pdata->tx_irq_timeout || pdata->tx_irq_count) - lp->tx_chnl_ctrl = (pdata->tx_irq_timeout << 24) | - (pdata->tx_irq_count << 16); - else - lp->tx_chnl_ctrl = 0x10220000; + if (pdata->tx_irq_timeout || pdata->tx_irq_count) { + lp->coalesce_delay_tx = pdata->tx_irq_timeout; + lp->coalesce_count_tx = pdata->tx_irq_count; + } if (pdata->rx_irq_timeout || pdata->rx_irq_count) { - lp->rx_chnl_ctrl = (pdata->rx_irq_timeout << 24) | - (pdata->rx_irq_count << 16); + lp->coalesce_delay_rx = pdata->rx_irq_timeout; lp->coalesce_count_rx = pdata->rx_irq_count; - } else { - lp->rx_chnl_ctrl = 0xff070000; - lp->coalesce_count_rx = 0x07; } } diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h index 2dacfc85b3ba..fbaf3c987d9c 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet.h +++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h @@ -161,17 +161,11 @@ #define XAE_FCC_OFFSET 0x0000040C /* Flow Control Configuration */ #define XAE_EMMC_OFFSET 0x00000410 /* EMAC mode configuration */ #define XAE_PHYC_OFFSET 0x00000414 /* RGMII/SGMII configuration */ +#define XAE_ID_OFFSET 0x000004F8 /* Identification register */ #define XAE_MDIO_MC_OFFSET 0x00000500 /* MII Management Config */ #define XAE_MDIO_MCR_OFFSET 0x00000504 /* MII Management Control */ #define XAE_MDIO_MWD_OFFSET 0x00000508 /* MII Management Write Data */ #define XAE_MDIO_MRD_OFFSET 0x0000050C /* MII Management Read Data */ -#define XAE_MDIO_MIS_OFFSET 0x00000600 /* MII Management Interrupt Status */ -/* MII Mgmt Interrupt Pending register offset */ -#define XAE_MDIO_MIP_OFFSET 0x00000620 -/* MII Management Interrupt Enable register offset */ -#define XAE_MDIO_MIE_OFFSET 0x00000640 -/* MII Management Interrupt Clear register offset. */ -#define XAE_MDIO_MIC_OFFSET 0x00000660 #define XAE_UAW0_OFFSET 0x00000700 /* Unicast address word 0 */ #define XAE_UAW1_OFFSET 0x00000704 /* Unicast address word 1 */ #define XAE_FMI_OFFSET 0x00000708 /* Filter Mask Index */ @@ -335,6 +329,7 @@ #define XAE_FEATURE_PARTIAL_TX_CSUM (1 << 1) #define XAE_FEATURE_FULL_RX_CSUM (1 << 2) #define XAE_FEATURE_FULL_TX_CSUM (1 << 3) +#define XAE_FEATURE_DMA_64BIT (1 << 4) #define XAE_NO_CSUM_OFFLOAD 0 @@ -347,9 +342,9 @@ /** * struct axidma_bd - Axi Dma buffer descriptor layout * @next: MM2S/S2MM Next Descriptor Pointer - * @reserved1: Reserved and not used + * @next_msb: MM2S/S2MM Next Descriptor Pointer (high 32 bits) * @phys: MM2S/S2MM Buffer Address - * @reserved2: Reserved and not used + * @phys_msb: MM2S/S2MM Buffer Address (high 32 bits) * @reserved3: Reserved and not used * @reserved4: Reserved and not used * @cntrl: MM2S/S2MM Control value @@ -362,9 +357,9 @@ */ struct axidma_bd { u32 next; /* Physical address of next buffer descriptor */ - u32 reserved1; + u32 next_msb; /* high 32 bits for IP >= v7.1, reserved on older IP */ u32 phys; - u32 reserved2; + u32 phys_msb; /* for IP >= v7.1, reserved for older IP */ u32 reserved3; u32 reserved4; u32 cntrl; @@ -435,7 +430,7 @@ struct axienet_local { void __iomem *regs; void __iomem *dma_regs; - struct tasklet_struct dma_err_tasklet; + struct work_struct dma_err_task; int tx_irq; int rx_irq; diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index 20746b801959..fa5dc2993520 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -147,6 +147,34 @@ static inline void axienet_dma_out32(struct axienet_local *lp, iowrite32(value, lp->dma_regs + reg); } +static void axienet_dma_out_addr(struct axienet_local *lp, off_t reg, + dma_addr_t addr) +{ + axienet_dma_out32(lp, reg, lower_32_bits(addr)); + + if (lp->features & XAE_FEATURE_DMA_64BIT) + axienet_dma_out32(lp, reg + 4, upper_32_bits(addr)); +} + +static void desc_set_phys_addr(struct axienet_local *lp, dma_addr_t addr, + struct axidma_bd *desc) +{ + desc->phys = lower_32_bits(addr); + if (lp->features & XAE_FEATURE_DMA_64BIT) + desc->phys_msb = upper_32_bits(addr); +} + +static dma_addr_t desc_get_phys_addr(struct axienet_local *lp, + struct axidma_bd *desc) +{ + dma_addr_t ret = desc->phys; + + if (lp->features & XAE_FEATURE_DMA_64BIT) + ret |= ((dma_addr_t)desc->phys_msb << 16) << 16; + + return ret; +} + /** * axienet_dma_bd_release - Release buffer descriptor rings * @ndev: Pointer to the net_device structure @@ -160,24 +188,41 @@ static void axienet_dma_bd_release(struct net_device *ndev) int i; struct axienet_local *lp = netdev_priv(ndev); + /* If we end up here, tx_bd_v must have been DMA allocated. */ + dma_free_coherent(ndev->dev.parent, + sizeof(*lp->tx_bd_v) * lp->tx_bd_num, + lp->tx_bd_v, + lp->tx_bd_p); + + if (!lp->rx_bd_v) + return; + for (i = 0; i < lp->rx_bd_num; i++) { - dma_unmap_single(ndev->dev.parent, lp->rx_bd_v[i].phys, - lp->max_frm_size, DMA_FROM_DEVICE); + dma_addr_t phys; + + /* A NULL skb means this descriptor has not been initialised + * at all. + */ + if (!lp->rx_bd_v[i].skb) + break; + dev_kfree_skb(lp->rx_bd_v[i].skb); - } - if (lp->rx_bd_v) { - dma_free_coherent(ndev->dev.parent, - sizeof(*lp->rx_bd_v) * lp->rx_bd_num, - lp->rx_bd_v, - lp->rx_bd_p); - } - if (lp->tx_bd_v) { - dma_free_coherent(ndev->dev.parent, - sizeof(*lp->tx_bd_v) * lp->tx_bd_num, - lp->tx_bd_v, - lp->tx_bd_p); + /* For each descriptor, we programmed cntrl with the (non-zero) + * descriptor size, after it had been successfully allocated. + * So a non-zero value in there means we need to unmap it. + */ + if (lp->rx_bd_v[i].cntrl) { + phys = desc_get_phys_addr(lp, &lp->rx_bd_v[i]); + dma_unmap_single(ndev->dev.parent, phys, + lp->max_frm_size, DMA_FROM_DEVICE); + } } + + dma_free_coherent(ndev->dev.parent, + sizeof(*lp->rx_bd_v) * lp->rx_bd_num, + lp->rx_bd_v, + lp->rx_bd_p); } /** @@ -207,7 +252,7 @@ static int axienet_dma_bd_init(struct net_device *ndev) sizeof(*lp->tx_bd_v) * lp->tx_bd_num, &lp->tx_bd_p, GFP_KERNEL); if (!lp->tx_bd_v) - goto out; + return -ENOMEM; lp->rx_bd_v = dma_alloc_coherent(ndev->dev.parent, sizeof(*lp->rx_bd_v) * lp->rx_bd_num, @@ -216,25 +261,37 @@ static int axienet_dma_bd_init(struct net_device *ndev) goto out; for (i = 0; i < lp->tx_bd_num; i++) { - lp->tx_bd_v[i].next = lp->tx_bd_p + - sizeof(*lp->tx_bd_v) * - ((i + 1) % lp->tx_bd_num); + dma_addr_t addr = lp->tx_bd_p + + sizeof(*lp->tx_bd_v) * + ((i + 1) % lp->tx_bd_num); + + lp->tx_bd_v[i].next = lower_32_bits(addr); + if (lp->features & XAE_FEATURE_DMA_64BIT) + lp->tx_bd_v[i].next_msb = upper_32_bits(addr); } for (i = 0; i < lp->rx_bd_num; i++) { - lp->rx_bd_v[i].next = lp->rx_bd_p + - sizeof(*lp->rx_bd_v) * - ((i + 1) % lp->rx_bd_num); + dma_addr_t addr; + + addr = lp->rx_bd_p + sizeof(*lp->rx_bd_v) * + ((i + 1) % lp->rx_bd_num); + lp->rx_bd_v[i].next = lower_32_bits(addr); + if (lp->features & XAE_FEATURE_DMA_64BIT) + lp->rx_bd_v[i].next_msb = upper_32_bits(addr); skb = netdev_alloc_skb_ip_align(ndev, lp->max_frm_size); if (!skb) goto out; lp->rx_bd_v[i].skb = skb; - lp->rx_bd_v[i].phys = dma_map_single(ndev->dev.parent, - skb->data, - lp->max_frm_size, - DMA_FROM_DEVICE); + addr = dma_map_single(ndev->dev.parent, skb->data, + lp->max_frm_size, DMA_FROM_DEVICE); + if (dma_mapping_error(ndev->dev.parent, addr)) { + netdev_err(ndev, "DMA mapping error\n"); + goto out; + } + desc_set_phys_addr(lp, addr, &lp->rx_bd_v[i]); + lp->rx_bd_v[i].cntrl = lp->max_frm_size; } @@ -267,18 +324,18 @@ static int axienet_dma_bd_init(struct net_device *ndev) /* Populate the tail pointer and bring the Rx Axi DMA engine out of * halted state. This will make the Rx side ready for reception. */ - axienet_dma_out32(lp, XAXIDMA_RX_CDESC_OFFSET, lp->rx_bd_p); + axienet_dma_out_addr(lp, XAXIDMA_RX_CDESC_OFFSET, lp->rx_bd_p); cr = axienet_dma_in32(lp, XAXIDMA_RX_CR_OFFSET); axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, cr | XAXIDMA_CR_RUNSTOP_MASK); - axienet_dma_out32(lp, XAXIDMA_RX_TDESC_OFFSET, lp->rx_bd_p + - (sizeof(*lp->rx_bd_v) * (lp->rx_bd_num - 1))); + axienet_dma_out_addr(lp, XAXIDMA_RX_TDESC_OFFSET, lp->rx_bd_p + + (sizeof(*lp->rx_bd_v) * (lp->rx_bd_num - 1))); /* Write to the RS (Run-stop) bit in the Tx channel control register. * Tx channel is now ready to run. But only after we write to the * tail pointer register that the Tx channel will start transmitting. */ - axienet_dma_out32(lp, XAXIDMA_TX_CDESC_OFFSET, lp->tx_bd_p); + axienet_dma_out_addr(lp, XAXIDMA_TX_CDESC_OFFSET, lp->tx_bd_p); cr = axienet_dma_in32(lp, XAXIDMA_TX_CR_OFFSET); axienet_dma_out32(lp, XAXIDMA_TX_CR_OFFSET, cr | XAXIDMA_CR_RUNSTOP_MASK); @@ -437,9 +494,10 @@ static void axienet_setoptions(struct net_device *ndev, u32 options) lp->options |= options; } -static void __axienet_device_reset(struct axienet_local *lp) +static int __axienet_device_reset(struct axienet_local *lp) { u32 timeout; + /* Reset Axi DMA. This would reset Axi Ethernet core as well. The reset * process of Axi DMA takes a while to complete as all pending * commands/transfers will be flushed or completed during this @@ -455,9 +513,11 @@ static void __axienet_device_reset(struct axienet_local *lp) if (--timeout == 0) { netdev_err(lp->ndev, "%s: DMA reset timeout!\n", __func__); - break; + return -ETIMEDOUT; } } + + return 0; } /** @@ -470,13 +530,17 @@ static void __axienet_device_reset(struct axienet_local *lp) * areconnected to Axi Ethernet reset lines, this in turn resets the Axi * Ethernet core. No separate hardware reset is done for the Axi Ethernet * core. + * Returns 0 on success or a negative error number otherwise. */ -static void axienet_device_reset(struct net_device *ndev) +static int axienet_device_reset(struct net_device *ndev) { u32 axienet_status; struct axienet_local *lp = netdev_priv(ndev); + int ret; - __axienet_device_reset(lp); + ret = __axienet_device_reset(lp); + if (ret) + return ret; lp->max_frm_size = XAE_MAX_VLAN_FRAME_SIZE; lp->options |= XAE_OPTION_VLAN; @@ -491,9 +555,11 @@ static void axienet_device_reset(struct net_device *ndev) lp->options |= XAE_OPTION_JUMBO; } - if (axienet_dma_bd_init(ndev)) { + ret = axienet_dma_bd_init(ndev); + if (ret) { netdev_err(ndev, "%s: descriptor allocation failed\n", __func__); + return ret; } axienet_status = axienet_ior(lp, XAE_RCW1_OFFSET); @@ -518,36 +584,54 @@ static void axienet_device_reset(struct net_device *ndev) axienet_setoptions(ndev, lp->options); netif_trans_update(ndev); + + return 0; } /** - * axienet_start_xmit_done - Invoked once a transmit is completed by the - * Axi DMA Tx channel. + * axienet_free_tx_chain - Clean up a series of linked TX descriptors. * @ndev: Pointer to the net_device structure + * @first_bd: Index of first descriptor to clean up + * @nr_bds: Number of descriptors to clean up, can be -1 if unknown. + * @sizep: Pointer to a u32 filled with the total sum of all bytes + * in all cleaned-up descriptors. Ignored if NULL. * - * This function is invoked from the Axi DMA Tx isr to notify the completion - * of transmit operation. It clears fields in the corresponding Tx BDs and - * unmaps the corresponding buffer so that CPU can regain ownership of the - * buffer. It finally invokes "netif_wake_queue" to restart transmission if - * required. + * Would either be called after a successful transmit operation, or after + * there was an error when setting up the chain. + * Returns the number of descriptors handled. */ -static void axienet_start_xmit_done(struct net_device *ndev) +static int axienet_free_tx_chain(struct net_device *ndev, u32 first_bd, + int nr_bds, u32 *sizep) { - u32 size = 0; - u32 packets = 0; struct axienet_local *lp = netdev_priv(ndev); struct axidma_bd *cur_p; - unsigned int status = 0; - - cur_p = &lp->tx_bd_v[lp->tx_bd_ci]; - status = cur_p->status; - while (status & XAXIDMA_BD_STS_COMPLETE_MASK) { - dma_unmap_single(ndev->dev.parent, cur_p->phys, - (cur_p->cntrl & XAXIDMA_BD_CTRL_LENGTH_MASK), - DMA_TO_DEVICE); - if (cur_p->skb) + int max_bds = nr_bds; + unsigned int status; + dma_addr_t phys; + int i; + + if (max_bds == -1) + max_bds = lp->tx_bd_num; + + for (i = 0; i < max_bds; i++) { + cur_p = &lp->tx_bd_v[(first_bd + i) % lp->tx_bd_num]; + status = cur_p->status; + + /* If no number is given, clean up *all* descriptors that have + * been completed by the MAC. + */ + if (nr_bds == -1 && !(status & XAXIDMA_BD_STS_COMPLETE_MASK)) + break; + + phys = desc_get_phys_addr(lp, cur_p); + dma_unmap_single(ndev->dev.parent, phys, + (cur_p->cntrl & XAXIDMA_BD_CTRL_LENGTH_MASK), + DMA_TO_DEVICE); + + if (cur_p->skb && (status & XAXIDMA_BD_STS_COMPLETE_MASK)) dev_consume_skb_irq(cur_p->skb); - /*cur_p->phys = 0;*/ + + cur_p->cntrl = 0; cur_p->app0 = 0; cur_p->app1 = 0; cur_p->app2 = 0; @@ -555,15 +639,36 @@ static void axienet_start_xmit_done(struct net_device *ndev) cur_p->status = 0; cur_p->skb = NULL; - size += status & XAXIDMA_BD_STS_ACTUAL_LEN_MASK; - packets++; - - if (++lp->tx_bd_ci >= lp->tx_bd_num) - lp->tx_bd_ci = 0; - cur_p = &lp->tx_bd_v[lp->tx_bd_ci]; - status = cur_p->status; + if (sizep) + *sizep += status & XAXIDMA_BD_STS_ACTUAL_LEN_MASK; } + return i; +} + +/** + * axienet_start_xmit_done - Invoked once a transmit is completed by the + * Axi DMA Tx channel. + * @ndev: Pointer to the net_device structure + * + * This function is invoked from the Axi DMA Tx isr to notify the completion + * of transmit operation. It clears fields in the corresponding Tx BDs and + * unmaps the corresponding buffer so that CPU can regain ownership of the + * buffer. It finally invokes "netif_wake_queue" to restart transmission if + * required. + */ +static void axienet_start_xmit_done(struct net_device *ndev) +{ + struct axienet_local *lp = netdev_priv(ndev); + u32 packets = 0; + u32 size = 0; + + packets = axienet_free_tx_chain(ndev, lp->tx_bd_ci, -1, &size); + + lp->tx_bd_ci += packets; + if (lp->tx_bd_ci >= lp->tx_bd_num) + lp->tx_bd_ci -= lp->tx_bd_num; + ndev->stats.tx_packets += packets; ndev->stats.tx_bytes += size; @@ -617,9 +722,10 @@ axienet_start_xmit(struct sk_buff *skb, struct net_device *ndev) u32 csum_start_off; u32 csum_index_off; skb_frag_t *frag; - dma_addr_t tail_p; + dma_addr_t tail_p, phys; struct axienet_local *lp = netdev_priv(ndev); struct axidma_bd *cur_p; + u32 orig_tail_ptr = lp->tx_bd_tail; num_frag = skb_shinfo(skb)->nr_frags; cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; @@ -655,19 +761,37 @@ axienet_start_xmit(struct sk_buff *skb, struct net_device *ndev) cur_p->app0 |= 2; /* Tx Full Checksum Offload Enabled */ } + phys = dma_map_single(ndev->dev.parent, skb->data, + skb_headlen(skb), DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(ndev->dev.parent, phys))) { + if (net_ratelimit()) + netdev_err(ndev, "TX DMA mapping error\n"); + ndev->stats.tx_dropped++; + return NETDEV_TX_OK; + } + desc_set_phys_addr(lp, phys, cur_p); cur_p->cntrl = skb_headlen(skb) | XAXIDMA_BD_CTRL_TXSOF_MASK; - cur_p->phys = dma_map_single(ndev->dev.parent, skb->data, - skb_headlen(skb), DMA_TO_DEVICE); for (ii = 0; ii < num_frag; ii++) { if (++lp->tx_bd_tail >= lp->tx_bd_num) lp->tx_bd_tail = 0; cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; frag = &skb_shinfo(skb)->frags[ii]; - cur_p->phys = dma_map_single(ndev->dev.parent, - skb_frag_address(frag), - skb_frag_size(frag), - DMA_TO_DEVICE); + phys = dma_map_single(ndev->dev.parent, + skb_frag_address(frag), + skb_frag_size(frag), + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(ndev->dev.parent, phys))) { + if (net_ratelimit()) + netdev_err(ndev, "TX DMA mapping error\n"); + ndev->stats.tx_dropped++; + axienet_free_tx_chain(ndev, orig_tail_ptr, ii + 1, + NULL); + lp->tx_bd_tail = orig_tail_ptr; + + return NETDEV_TX_OK; + } + desc_set_phys_addr(lp, phys, cur_p); cur_p->cntrl = skb_frag_size(frag); } @@ -676,7 +800,7 @@ axienet_start_xmit(struct sk_buff *skb, struct net_device *ndev) tail_p = lp->tx_bd_p + sizeof(*lp->tx_bd_v) * lp->tx_bd_tail; /* Start the transfer */ - axienet_dma_out32(lp, XAXIDMA_TX_TDESC_OFFSET, tail_p); + axienet_dma_out_addr(lp, XAXIDMA_TX_TDESC_OFFSET, tail_p); if (++lp->tx_bd_tail >= lp->tx_bd_num) lp->tx_bd_tail = 0; @@ -706,10 +830,12 @@ static void axienet_recv(struct net_device *ndev) cur_p = &lp->rx_bd_v[lp->rx_bd_ci]; while ((cur_p->status & XAXIDMA_BD_STS_COMPLETE_MASK)) { + dma_addr_t phys; + tail_p = lp->rx_bd_p + sizeof(*lp->rx_bd_v) * lp->rx_bd_ci; - dma_unmap_single(ndev->dev.parent, cur_p->phys, - lp->max_frm_size, + phys = desc_get_phys_addr(lp, cur_p); + dma_unmap_single(ndev->dev.parent, phys, lp->max_frm_size, DMA_FROM_DEVICE); skb = cur_p->skb; @@ -745,9 +871,17 @@ static void axienet_recv(struct net_device *ndev) if (!new_skb) return; - cur_p->phys = dma_map_single(ndev->dev.parent, new_skb->data, - lp->max_frm_size, - DMA_FROM_DEVICE); + phys = dma_map_single(ndev->dev.parent, new_skb->data, + lp->max_frm_size, + DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(ndev->dev.parent, phys))) { + if (net_ratelimit()) + netdev_err(ndev, "RX DMA mapping error\n"); + dev_kfree_skb(new_skb); + return; + } + desc_set_phys_addr(lp, phys, cur_p); + cur_p->cntrl = lp->max_frm_size; cur_p->status = 0; cur_p->skb = new_skb; @@ -761,7 +895,7 @@ static void axienet_recv(struct net_device *ndev) ndev->stats.rx_bytes += size; if (tail_p) - axienet_dma_out32(lp, XAXIDMA_RX_TDESC_OFFSET, tail_p); + axienet_dma_out_addr(lp, XAXIDMA_RX_TDESC_OFFSET, tail_p); } /** @@ -791,7 +925,8 @@ static irqreturn_t axienet_tx_irq(int irq, void *_ndev) return IRQ_NONE; if (status & XAXIDMA_IRQ_ERROR_MASK) { dev_err(&ndev->dev, "DMA Tx error 0x%x\n", status); - dev_err(&ndev->dev, "Current BD is at: 0x%x\n", + dev_err(&ndev->dev, "Current BD is at: 0x%x%08x\n", + (lp->tx_bd_v[lp->tx_bd_ci]).phys_msb, (lp->tx_bd_v[lp->tx_bd_ci]).phys); cr = axienet_dma_in32(lp, XAXIDMA_TX_CR_OFFSET); @@ -806,7 +941,7 @@ static irqreturn_t axienet_tx_irq(int irq, void *_ndev) /* Write to the Rx channel control register */ axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, cr); - tasklet_schedule(&lp->dma_err_tasklet); + schedule_work(&lp->dma_err_task); axienet_dma_out32(lp, XAXIDMA_TX_SR_OFFSET, status); } out: @@ -840,7 +975,8 @@ static irqreturn_t axienet_rx_irq(int irq, void *_ndev) return IRQ_NONE; if (status & XAXIDMA_IRQ_ERROR_MASK) { dev_err(&ndev->dev, "DMA Rx error 0x%x\n", status); - dev_err(&ndev->dev, "Current BD is at: 0x%x\n", + dev_err(&ndev->dev, "Current BD is at: 0x%x%08x\n", + (lp->rx_bd_v[lp->rx_bd_ci]).phys_msb, (lp->rx_bd_v[lp->rx_bd_ci]).phys); cr = axienet_dma_in32(lp, XAXIDMA_TX_CR_OFFSET); @@ -855,7 +991,7 @@ static irqreturn_t axienet_rx_irq(int irq, void *_ndev) /* write to the Rx channel control register */ axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, cr); - tasklet_schedule(&lp->dma_err_tasklet); + schedule_work(&lp->dma_err_task); axienet_dma_out32(lp, XAXIDMA_RX_SR_OFFSET, status); } out: @@ -891,7 +1027,7 @@ static irqreturn_t axienet_eth_irq(int irq, void *_ndev) return IRQ_HANDLED; } -static void axienet_dma_err_handler(unsigned long data); +static void axienet_dma_err_handler(struct work_struct *work); /** * axienet_open - Driver open routine. @@ -921,8 +1057,9 @@ static int axienet_open(struct net_device *ndev) */ mutex_lock(&lp->mii_bus->mdio_lock); axienet_mdio_disable(lp); - axienet_device_reset(ndev); - ret = axienet_mdio_enable(lp); + ret = axienet_device_reset(ndev); + if (ret == 0) + ret = axienet_mdio_enable(lp); mutex_unlock(&lp->mii_bus->mdio_lock); if (ret < 0) return ret; @@ -935,9 +1072,8 @@ static int axienet_open(struct net_device *ndev) phylink_start(lp->phylink); - /* Enable tasklets for Axi DMA error handling */ - tasklet_init(&lp->dma_err_tasklet, axienet_dma_err_handler, - (unsigned long) lp); + /* Enable worker thread for Axi DMA error handling */ + INIT_WORK(&lp->dma_err_task, axienet_dma_err_handler); /* Enable interrupts for Axi DMA Tx */ ret = request_irq(lp->tx_irq, axienet_tx_irq, IRQF_SHARED, @@ -966,7 +1102,7 @@ err_rx_irq: err_tx_irq: phylink_stop(lp->phylink); phylink_disconnect_phy(lp->phylink); - tasklet_kill(&lp->dma_err_tasklet); + cancel_work_sync(&lp->dma_err_task); dev_err(lp->dev, "request_irq() failed\n"); return ret; } @@ -1025,7 +1161,7 @@ static int axienet_stop(struct net_device *ndev) axienet_mdio_enable(lp); mutex_unlock(&lp->mii_bus->mdio_lock); - tasklet_kill(&lp->dma_err_tasklet); + cancel_work_sync(&lp->dma_err_task); if (lp->eth_irq > 0) free_irq(lp->eth_irq, ndev); @@ -1083,6 +1219,16 @@ static void axienet_poll_controller(struct net_device *ndev) } #endif +static int axienet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) +{ + struct axienet_local *lp = netdev_priv(dev); + + if (!netif_running(dev)) + return -EINVAL; + + return phylink_mii_ioctl(lp->phylink, rq, cmd); +} + static const struct net_device_ops axienet_netdev_ops = { .ndo_open = axienet_open, .ndo_stop = axienet_stop, @@ -1090,6 +1236,7 @@ static const struct net_device_ops axienet_netdev_ops = { .ndo_change_mtu = axienet_change_mtu, .ndo_set_mac_address = netdev_set_mac_address, .ndo_validate_addr = eth_validate_addr, + .ndo_do_ioctl = axienet_ioctl, .ndo_set_rx_mode = axienet_set_multicast_list, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = axienet_poll_controller, @@ -1170,10 +1317,6 @@ static void axienet_ethtools_get_regs(struct net_device *ndev, data[20] = axienet_ior(lp, XAE_MDIO_MCR_OFFSET); data[21] = axienet_ior(lp, XAE_MDIO_MWD_OFFSET); data[22] = axienet_ior(lp, XAE_MDIO_MRD_OFFSET); - data[23] = axienet_ior(lp, XAE_MDIO_MIS_OFFSET); - data[24] = axienet_ior(lp, XAE_MDIO_MIP_OFFSET); - data[25] = axienet_ior(lp, XAE_MDIO_MIE_OFFSET); - data[26] = axienet_ior(lp, XAE_MDIO_MIC_OFFSET); data[27] = axienet_ior(lp, XAE_UAW0_OFFSET); data[28] = axienet_ior(lp, XAE_UAW1_OFFSET); data[29] = axienet_ior(lp, XAE_FMI_OFFSET); @@ -1309,27 +1452,6 @@ static int axienet_ethtools_set_coalesce(struct net_device *ndev, return -EFAULT; } - if ((ecoalesce->rx_coalesce_usecs) || - (ecoalesce->rx_coalesce_usecs_irq) || - (ecoalesce->rx_max_coalesced_frames_irq) || - (ecoalesce->tx_coalesce_usecs) || - (ecoalesce->tx_coalesce_usecs_irq) || - (ecoalesce->tx_max_coalesced_frames_irq) || - (ecoalesce->stats_block_coalesce_usecs) || - (ecoalesce->use_adaptive_rx_coalesce) || - (ecoalesce->use_adaptive_tx_coalesce) || - (ecoalesce->pkt_rate_low) || - (ecoalesce->rx_coalesce_usecs_low) || - (ecoalesce->rx_max_coalesced_frames_low) || - (ecoalesce->tx_coalesce_usecs_low) || - (ecoalesce->tx_max_coalesced_frames_low) || - (ecoalesce->pkt_rate_high) || - (ecoalesce->rx_coalesce_usecs_high) || - (ecoalesce->rx_max_coalesced_frames_high) || - (ecoalesce->tx_coalesce_usecs_high) || - (ecoalesce->tx_max_coalesced_frames_high) || - (ecoalesce->rate_sample_interval)) - return -EOPNOTSUPP; if (ecoalesce->rx_max_coalesced_frames) lp->coalesce_count_rx = ecoalesce->rx_max_coalesced_frames; if (ecoalesce->tx_max_coalesced_frames) @@ -1357,6 +1479,7 @@ axienet_ethtools_set_link_ksettings(struct net_device *ndev, } static const struct ethtool_ops axienet_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_MAX_FRAMES, .get_drvinfo = axienet_ethtools_get_drvinfo, .get_regs_len = axienet_ethtools_get_regs_len, .get_regs = axienet_ethtools_get_regs, @@ -1441,6 +1564,22 @@ static void axienet_mac_an_restart(struct phylink_config *config) static void axienet_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { + /* nothing meaningful to do */ +} + +static void axienet_mac_link_down(struct phylink_config *config, + unsigned int mode, + phy_interface_t interface) +{ + /* nothing meaningful to do */ +} + +static void axienet_mac_link_up(struct phylink_config *config, + struct phy_device *phy, + unsigned int mode, phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause) +{ struct net_device *ndev = to_net_dev(config->dev); struct axienet_local *lp = netdev_priv(ndev); u32 emmc_reg, fcc_reg; @@ -1448,7 +1587,7 @@ static void axienet_mac_config(struct phylink_config *config, unsigned int mode, emmc_reg = axienet_ior(lp, XAE_EMMC_OFFSET); emmc_reg &= ~XAE_EMMC_LINKSPEED_MASK; - switch (state->speed) { + switch (speed) { case SPEED_1000: emmc_reg |= XAE_EMMC_LINKSPD_1000; break; @@ -1467,32 +1606,17 @@ static void axienet_mac_config(struct phylink_config *config, unsigned int mode, axienet_iow(lp, XAE_EMMC_OFFSET, emmc_reg); fcc_reg = axienet_ior(lp, XAE_FCC_OFFSET); - if (state->pause & MLO_PAUSE_TX) + if (tx_pause) fcc_reg |= XAE_FCC_FCTX_MASK; else fcc_reg &= ~XAE_FCC_FCTX_MASK; - if (state->pause & MLO_PAUSE_RX) + if (rx_pause) fcc_reg |= XAE_FCC_FCRX_MASK; else fcc_reg &= ~XAE_FCC_FCRX_MASK; axienet_iow(lp, XAE_FCC_OFFSET, fcc_reg); } -static void axienet_mac_link_down(struct phylink_config *config, - unsigned int mode, - phy_interface_t interface) -{ - /* nothing meaningful to do */ -} - -static void axienet_mac_link_up(struct phylink_config *config, - unsigned int mode, - phy_interface_t interface, - struct phy_device *phy) -{ - /* nothing meaningful to do */ -} - static const struct phylink_mac_ops axienet_phylink_ops = { .validate = axienet_validate, .mac_pcs_get_state = axienet_mac_pcs_get_state, @@ -1503,17 +1627,18 @@ static const struct phylink_mac_ops axienet_phylink_ops = { }; /** - * axienet_dma_err_handler - Tasklet handler for Axi DMA Error - * @data: Data passed + * axienet_dma_err_handler - Work queue task for Axi DMA Error + * @work: pointer to work_struct * * Resets the Axi DMA and Axi Ethernet devices, and reconfigures the * Tx/Rx BDs. */ -static void axienet_dma_err_handler(unsigned long data) +static void axienet_dma_err_handler(struct work_struct *work) { u32 axienet_status; u32 cr, i; - struct axienet_local *lp = (struct axienet_local *) data; + struct axienet_local *lp = container_of(work, struct axienet_local, + dma_err_task); struct net_device *ndev = lp->ndev; struct axidma_bd *cur_p; @@ -1533,14 +1658,18 @@ static void axienet_dma_err_handler(unsigned long data) for (i = 0; i < lp->tx_bd_num; i++) { cur_p = &lp->tx_bd_v[i]; - if (cur_p->phys) - dma_unmap_single(ndev->dev.parent, cur_p->phys, + if (cur_p->cntrl) { + dma_addr_t addr = desc_get_phys_addr(lp, cur_p); + + dma_unmap_single(ndev->dev.parent, addr, (cur_p->cntrl & XAXIDMA_BD_CTRL_LENGTH_MASK), DMA_TO_DEVICE); + } if (cur_p->skb) dev_kfree_skb_irq(cur_p->skb); cur_p->phys = 0; + cur_p->phys_msb = 0; cur_p->cntrl = 0; cur_p->status = 0; cur_p->app0 = 0; @@ -1594,18 +1723,18 @@ static void axienet_dma_err_handler(unsigned long data) /* Populate the tail pointer and bring the Rx Axi DMA engine out of * halted state. This will make the Rx side ready for reception. */ - axienet_dma_out32(lp, XAXIDMA_RX_CDESC_OFFSET, lp->rx_bd_p); + axienet_dma_out_addr(lp, XAXIDMA_RX_CDESC_OFFSET, lp->rx_bd_p); cr = axienet_dma_in32(lp, XAXIDMA_RX_CR_OFFSET); axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, cr | XAXIDMA_CR_RUNSTOP_MASK); - axienet_dma_out32(lp, XAXIDMA_RX_TDESC_OFFSET, lp->rx_bd_p + - (sizeof(*lp->rx_bd_v) * (lp->rx_bd_num - 1))); + axienet_dma_out_addr(lp, XAXIDMA_RX_TDESC_OFFSET, lp->rx_bd_p + + (sizeof(*lp->rx_bd_v) * (lp->rx_bd_num - 1))); /* Write to the RS (Run-stop) bit in the Tx channel control register. * Tx channel is now ready to run. But only after we write to the * tail pointer register that the Tx channel will start transmitting */ - axienet_dma_out32(lp, XAXIDMA_TX_CDESC_OFFSET, lp->tx_bd_p); + axienet_dma_out_addr(lp, XAXIDMA_TX_CDESC_OFFSET, lp->tx_bd_p); cr = axienet_dma_in32(lp, XAXIDMA_TX_CR_OFFSET); axienet_dma_out32(lp, XAXIDMA_TX_CR_OFFSET, cr | XAXIDMA_CR_RUNSTOP_MASK); @@ -1651,6 +1780,7 @@ static int axienet_probe(struct platform_device *pdev) struct net_device *ndev; const void *mac_addr; struct resource *ethres; + int addr_width = 32; u32 value; ndev = alloc_etherdev(sizeof(*lp)); @@ -1781,7 +1911,7 @@ static int axienet_probe(struct platform_device *pdev) lp->rx_irq = irq_of_parse_and_map(np, 1); lp->tx_irq = irq_of_parse_and_map(np, 0); of_node_put(np); - lp->eth_irq = platform_get_irq(pdev, 0); + lp->eth_irq = platform_get_irq_optional(pdev, 0); } else { /* Check for these resources directly on the Ethernet node. */ struct resource *res = platform_get_resource(pdev, @@ -1789,7 +1919,7 @@ static int axienet_probe(struct platform_device *pdev) lp->dma_regs = devm_ioremap_resource(&pdev->dev, res); lp->rx_irq = platform_get_irq(pdev, 1); lp->tx_irq = platform_get_irq(pdev, 0); - lp->eth_irq = platform_get_irq(pdev, 2); + lp->eth_irq = platform_get_irq_optional(pdev, 2); } if (IS_ERR(lp->dma_regs)) { dev_err(&pdev->dev, "could not map DMA regs\n"); @@ -1802,6 +1932,36 @@ static int axienet_probe(struct platform_device *pdev) goto free_netdev; } + /* Autodetect the need for 64-bit DMA pointers. + * When the IP is configured for a bus width bigger than 32 bits, + * writing the MSB registers is mandatory, even if they are all 0. + * We can detect this case by writing all 1's to one such register + * and see if that sticks: when the IP is configured for 32 bits + * only, those registers are RES0. + * Those MSB registers were introduced in IP v7.1, which we check first. + */ + if ((axienet_ior(lp, XAE_ID_OFFSET) >> 24) >= 0x9) { + void __iomem *desc = lp->dma_regs + XAXIDMA_TX_CDESC_OFFSET + 4; + + iowrite32(0x0, desc); + if (ioread32(desc) == 0) { /* sanity check */ + iowrite32(0xffffffff, desc); + if (ioread32(desc) > 0) { + lp->features |= XAE_FEATURE_DMA_64BIT; + addr_width = 64; + dev_info(&pdev->dev, + "autodetected 64-bit DMA range\n"); + } + iowrite32(0x0, desc); + } + } + + ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(addr_width)); + if (ret) { + dev_err(&pdev->dev, "No suitable DMA available\n"); + goto free_netdev; + } + /* Check for Ethernet core IRQ (optional) */ if (lp->eth_irq <= 0) dev_info(&pdev->dev, "Ethernet core IRQ not defined\n"); |