diff options
Diffstat (limited to 'drivers')
441 files changed, 21700 insertions, 13546 deletions
diff --git a/drivers/atm/solos-pci.c b/drivers/atm/solos-pci.c index c909b7b7d5f1..d70abe77f737 100644 --- a/drivers/atm/solos-pci.c +++ b/drivers/atm/solos-pci.c @@ -42,7 +42,8 @@ #include <linux/swab.h> #include <linux/slab.h> -#define VERSION "0.07" +#define VERSION "1.04" +#define DRIVER_VERSION 0x01 #define PTAG "solos-pci" #define CONFIG_RAM_SIZE 128 @@ -56,16 +57,21 @@ #define FLASH_BUSY 0x60 #define FPGA_MODE 0x5C #define FLASH_MODE 0x58 +#define GPIO_STATUS 0x54 +#define DRIVER_VER 0x50 #define TX_DMA_ADDR(port) (0x40 + (4 * (port))) #define RX_DMA_ADDR(port) (0x30 + (4 * (port))) #define DATA_RAM_SIZE 32768 #define BUF_SIZE 2048 #define OLD_BUF_SIZE 4096 /* For FPGA versions <= 2*/ -#define FPGA_PAGE 528 /* FPGA flash page size*/ -#define SOLOS_PAGE 512 /* Solos flash page size*/ -#define FPGA_BLOCK (FPGA_PAGE * 8) /* FPGA flash block size*/ -#define SOLOS_BLOCK (SOLOS_PAGE * 8) /* Solos flash block size*/ +/* Old boards use ATMEL AD45DB161D flash */ +#define ATMEL_FPGA_PAGE 528 /* FPGA flash page size*/ +#define ATMEL_SOLOS_PAGE 512 /* Solos flash page size*/ +#define ATMEL_FPGA_BLOCK (ATMEL_FPGA_PAGE * 8) /* FPGA block size*/ +#define ATMEL_SOLOS_BLOCK (ATMEL_SOLOS_PAGE * 8) /* Solos block size*/ +/* Current boards use M25P/M25PE SPI flash */ +#define SPI_FLASH_BLOCK (256 * 64) #define RX_BUF(card, nr) ((card->buffers) + (nr)*(card->buffer_size)*2) #define TX_BUF(card, nr) ((card->buffers) + (nr)*(card->buffer_size)*2 + (card->buffer_size)) @@ -122,11 +128,14 @@ struct solos_card { struct sk_buff_head cli_queue[4]; struct sk_buff *tx_skb[4]; struct sk_buff *rx_skb[4]; + unsigned char *dma_bounce; wait_queue_head_t param_wq; wait_queue_head_t fw_wq; int using_dma; + int dma_alignment; int fpga_version; int buffer_size; + int atmel_flash; }; @@ -451,7 +460,6 @@ static ssize_t console_show(struct device *dev, struct device_attribute *attr, len = skb->len; memcpy(buf, skb->data, len); - dev_dbg(&card->dev->dev, "len: %d\n", len); kfree_skb(skb); return len; @@ -498,6 +506,78 @@ static ssize_t console_store(struct device *dev, struct device_attribute *attr, return err?:count; } +struct geos_gpio_attr { + struct device_attribute attr; + int offset; +}; + +#define SOLOS_GPIO_ATTR(_name, _mode, _show, _store, _offset) \ + struct geos_gpio_attr gpio_attr_##_name = { \ + .attr = __ATTR(_name, _mode, _show, _store), \ + .offset = _offset } + +static ssize_t geos_gpio_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct pci_dev *pdev = container_of(dev, struct pci_dev, dev); + struct geos_gpio_attr *gattr = container_of(attr, struct geos_gpio_attr, attr); + struct solos_card *card = pci_get_drvdata(pdev); + uint32_t data32; + + if (count != 1 && (count != 2 || buf[1] != '\n')) + return -EINVAL; + + spin_lock_irq(&card->param_queue_lock); + data32 = ioread32(card->config_regs + GPIO_STATUS); + if (buf[0] == '1') { + data32 |= 1 << gattr->offset; + iowrite32(data32, card->config_regs + GPIO_STATUS); + } else if (buf[0] == '0') { + data32 &= ~(1 << gattr->offset); + iowrite32(data32, card->config_regs + GPIO_STATUS); + } else { + count = -EINVAL; + } + spin_lock_irq(&card->param_queue_lock); + return count; +} + +static ssize_t geos_gpio_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct pci_dev *pdev = container_of(dev, struct pci_dev, dev); + struct geos_gpio_attr *gattr = container_of(attr, struct geos_gpio_attr, attr); + struct solos_card *card = pci_get_drvdata(pdev); + uint32_t data32; + + data32 = ioread32(card->config_regs + GPIO_STATUS); + data32 = (data32 >> gattr->offset) & 1; + + return sprintf(buf, "%d\n", data32); +} + +static ssize_t hardware_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct pci_dev *pdev = container_of(dev, struct pci_dev, dev); + struct geos_gpio_attr *gattr = container_of(attr, struct geos_gpio_attr, attr); + struct solos_card *card = pci_get_drvdata(pdev); + uint32_t data32; + + data32 = ioread32(card->config_regs + GPIO_STATUS); + switch (gattr->offset) { + case 0: + /* HardwareVersion */ + data32 = data32 & 0x1F; + break; + case 1: + /* HardwareVariant */ + data32 = (data32 >> 5) & 0x0F; + break; + } + return sprintf(buf, "%d\n", data32); +} + static DEVICE_ATTR(console, 0644, console_show, console_store); @@ -506,6 +586,14 @@ static DEVICE_ATTR(console, 0644, console_show, console_store); #include "solos-attrlist.c" +static SOLOS_GPIO_ATTR(GPIO1, 0644, geos_gpio_show, geos_gpio_store, 9); +static SOLOS_GPIO_ATTR(GPIO2, 0644, geos_gpio_show, geos_gpio_store, 10); +static SOLOS_GPIO_ATTR(GPIO3, 0644, geos_gpio_show, geos_gpio_store, 11); +static SOLOS_GPIO_ATTR(GPIO4, 0644, geos_gpio_show, geos_gpio_store, 12); +static SOLOS_GPIO_ATTR(GPIO5, 0644, geos_gpio_show, geos_gpio_store, 13); +static SOLOS_GPIO_ATTR(PushButton, 0444, geos_gpio_show, NULL, 14); +static SOLOS_GPIO_ATTR(HardwareVersion, 0444, hardware_show, NULL, 0); +static SOLOS_GPIO_ATTR(HardwareVariant, 0444, hardware_show, NULL, 1); #undef SOLOS_ATTR_RO #undef SOLOS_ATTR_RW @@ -522,6 +610,23 @@ static struct attribute_group solos_attr_group = { .name = "parameters", }; +static struct attribute *gpio_attrs[] = { + &gpio_attr_GPIO1.attr.attr, + &gpio_attr_GPIO2.attr.attr, + &gpio_attr_GPIO3.attr.attr, + &gpio_attr_GPIO4.attr.attr, + &gpio_attr_GPIO5.attr.attr, + &gpio_attr_PushButton.attr.attr, + &gpio_attr_HardwareVersion.attr.attr, + &gpio_attr_HardwareVariant.attr.attr, + NULL +}; + +static struct attribute_group gpio_attr_group = { + .attrs = gpio_attrs, + .name = "gpio", +}; + static int flash_upgrade(struct solos_card *card, int chip) { const struct firmware *fw; @@ -533,16 +638,25 @@ static int flash_upgrade(struct solos_card *card, int chip) switch (chip) { case 0: fw_name = "solos-FPGA.bin"; - blocksize = FPGA_BLOCK; + if (card->atmel_flash) + blocksize = ATMEL_FPGA_BLOCK; + else + blocksize = SPI_FLASH_BLOCK; break; case 1: fw_name = "solos-Firmware.bin"; - blocksize = SOLOS_BLOCK; + if (card->atmel_flash) + blocksize = ATMEL_SOLOS_BLOCK; + else + blocksize = SPI_FLASH_BLOCK; break; case 2: if (card->fpga_version > LEGACY_BUFFERS){ fw_name = "solos-db-FPGA.bin"; - blocksize = FPGA_BLOCK; + if (card->atmel_flash) + blocksize = ATMEL_FPGA_BLOCK; + else + blocksize = SPI_FLASH_BLOCK; } else { dev_info(&card->dev->dev, "FPGA version doesn't support" " daughter board upgrades\n"); @@ -552,7 +666,10 @@ static int flash_upgrade(struct solos_card *card, int chip) case 3: if (card->fpga_version > LEGACY_BUFFERS){ fw_name = "solos-Firmware.bin"; - blocksize = SOLOS_BLOCK; + if (card->atmel_flash) + blocksize = ATMEL_SOLOS_BLOCK; + else + blocksize = SPI_FLASH_BLOCK; } else { dev_info(&card->dev->dev, "FPGA version doesn't support" " daughter board upgrades\n"); @@ -568,6 +685,9 @@ static int flash_upgrade(struct solos_card *card, int chip) dev_info(&card->dev->dev, "Flash upgrade starting\n"); + /* New FPGAs require driver version before permitting flash upgrades */ + iowrite32(DRIVER_VERSION, card->config_regs + DRIVER_VER); + numblocks = fw->size / blocksize; dev_info(&card->dev->dev, "Firmware size: %zd\n", fw->size); dev_info(&card->dev->dev, "Number of blocks: %d\n", numblocks); @@ -597,9 +717,13 @@ static int flash_upgrade(struct solos_card *card, int chip) /* dev_info(&card->dev->dev, "Set FPGA Flash mode to Block Write\n"); */ iowrite32(((chip * 2) + 1), card->config_regs + FLASH_MODE); - /* Copy block to buffer, swapping each 16 bits */ + /* Copy block to buffer, swapping each 16 bits for Atmel flash */ for(i = 0; i < blocksize; i += 4) { - uint32_t word = swahb32p((uint32_t *)(fw->data + offset + i)); + uint32_t word; + if (card->atmel_flash) + word = swahb32p((uint32_t *)(fw->data + offset + i)); + else + word = *(uint32_t *)(fw->data + offset + i); if(card->fpga_version > LEGACY_BUFFERS) iowrite32(word, FLASH_BUF + i); else @@ -961,7 +1085,12 @@ static uint32_t fpga_tx(struct solos_card *card) tx_started |= 1 << port; oldskb = skb; /* We're done with this skb already */ } else if (skb && card->using_dma) { - SKB_CB(skb)->dma_addr = pci_map_single(card->dev, skb->data, + unsigned char *data = skb->data; + if ((unsigned long)data & card->dma_alignment) { + data = card->dma_bounce + (BUF_SIZE * port); + memcpy(data, skb->data, skb->len); + } + SKB_CB(skb)->dma_addr = pci_map_single(card->dev, data, skb->len, PCI_DMA_TODEVICE); card->tx_skb[port] = skb; iowrite32(SKB_CB(skb)->dma_addr, @@ -1133,18 +1262,33 @@ static int fpga_probe(struct pci_dev *dev, const struct pci_device_id *id) db_fpga_upgrade = db_firmware_upgrade = 0; } + /* Stopped using Atmel flash after 0.03-38 */ + if (fpga_ver < 39) + card->atmel_flash = 1; + else + card->atmel_flash = 0; + + data32 = ioread32(card->config_regs + PORTS); + card->nr_ports = (data32 & 0x000000FF); + if (card->fpga_version >= DMA_SUPPORTED) { pci_set_master(dev); card->using_dma = 1; + if (1) { /* All known FPGA versions so far */ + card->dma_alignment = 3; + card->dma_bounce = kmalloc(card->nr_ports * BUF_SIZE, GFP_KERNEL); + if (!card->dma_bounce) { + dev_warn(&card->dev->dev, "Failed to allocate DMA bounce buffers\n"); + /* Fallback to MMIO doesn't work */ + goto out_unmap_both; + } + } } else { card->using_dma = 0; /* Set RX empty flag for all ports */ iowrite32(0xF0, card->config_regs + FLAGS_ADDR); } - data32 = ioread32(card->config_regs + PORTS); - card->nr_ports = (data32 & 0x000000FF); - pci_set_drvdata(dev, card); tasklet_init(&card->tlet, solos_bh, (unsigned long)card); @@ -1179,6 +1323,10 @@ static int fpga_probe(struct pci_dev *dev, const struct pci_device_id *id) if (err) goto out_free_irq; + if (card->fpga_version >= DMA_SUPPORTED && + sysfs_create_group(&card->dev->dev.kobj, &gpio_attr_group)) + dev_err(&card->dev->dev, "Could not register parameter group for GPIOs\n"); + return 0; out_free_irq: @@ -1187,6 +1335,7 @@ static int fpga_probe(struct pci_dev *dev, const struct pci_device_id *id) tasklet_kill(&card->tlet); out_unmap_both: + kfree(card->dma_bounce); pci_set_drvdata(dev, NULL); pci_iounmap(dev, card->buffers); out_unmap_config: @@ -1289,11 +1438,16 @@ static void fpga_remove(struct pci_dev *dev) iowrite32(1, card->config_regs + FPGA_MODE); (void)ioread32(card->config_regs + FPGA_MODE); + if (card->fpga_version >= DMA_SUPPORTED) + sysfs_remove_group(&card->dev->dev.kobj, &gpio_attr_group); + atm_remove(card); free_irq(dev->irq, card); tasklet_kill(&card->tlet); + kfree(card->dma_bounce); + /* Release device from reset */ iowrite32(0, card->config_regs + FPGA_MODE); (void)ioread32(card->config_regs + FPGA_MODE); diff --git a/drivers/base/dma-buf.c b/drivers/base/dma-buf.c index 460e22dee36d..a3f79c495a41 100644 --- a/drivers/base/dma-buf.c +++ b/drivers/base/dma-buf.c @@ -298,6 +298,8 @@ void dma_buf_unmap_attachment(struct dma_buf_attachment *attach, struct sg_table *sg_table, enum dma_data_direction direction) { + might_sleep(); + if (WARN_ON(!attach || !attach->dmabuf || !sg_table)) return; diff --git a/drivers/bcma/driver_chipcommon_pmu.c b/drivers/bcma/driver_chipcommon_pmu.c index e162999bf916..c62c788b3289 100644 --- a/drivers/bcma/driver_chipcommon_pmu.c +++ b/drivers/bcma/driver_chipcommon_pmu.c @@ -13,12 +13,13 @@ #include <linux/export.h> #include <linux/bcma/bcma.h> -static u32 bcma_chipco_pll_read(struct bcma_drv_cc *cc, u32 offset) +u32 bcma_chipco_pll_read(struct bcma_drv_cc *cc, u32 offset) { bcma_cc_write32(cc, BCMA_CC_PLLCTL_ADDR, offset); bcma_cc_read32(cc, BCMA_CC_PLLCTL_ADDR); return bcma_cc_read32(cc, BCMA_CC_PLLCTL_DATA); } +EXPORT_SYMBOL_GPL(bcma_chipco_pll_read); void bcma_chipco_pll_write(struct bcma_drv_cc *cc, u32 offset, u32 value) { diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index d2ed7f18d1ac..175649468c95 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -1,5 +1,5 @@ /* Copyright (c) 2012 Coraid, Inc. See COPYING for GPL terms. */ -#define VERSION "50" +#define VERSION "81" #define AOE_MAJOR 152 #define DEVICE_NAME "aoe" @@ -10,7 +10,7 @@ #define AOE_PARTITIONS (16) #endif -#define WHITESPACE " \t\v\f\n" +#define WHITESPACE " \t\v\f\n," enum { AOECMD_ATA, @@ -73,21 +73,29 @@ enum { DEVFL_TKILL = (1<<1), /* flag for timer to know when to kill self */ DEVFL_EXT = (1<<2), /* device accepts lba48 commands */ DEVFL_GDALLOC = (1<<3), /* need to alloc gendisk */ - DEVFL_KICKME = (1<<4), /* slow polling network card catch */ - DEVFL_NEWSIZE = (1<<5), /* need to update dev size in block layer */ + DEVFL_GD_NOW = (1<<4), /* allocating gendisk */ + DEVFL_KICKME = (1<<5), /* slow polling network card catch */ + DEVFL_NEWSIZE = (1<<6), /* need to update dev size in block layer */ + DEVFL_FREEING = (1<<7), /* set when device is being cleaned up */ + DEVFL_FREED = (1<<8), /* device has been cleaned up */ }; enum { DEFAULTBCNT = 2 * 512, /* 2 sectors */ MIN_BUFS = 16, - NTARGETS = 8, + NTARGETS = 4, NAOEIFS = 8, NSKBPOOLMAX = 256, NFACTIVE = 61, TIMERTICK = HZ / 10, - MINTIMER = HZ >> 2, - MAXTIMER = HZ << 1, + RTTSCALE = 8, + RTTDSCALE = 3, + RTTAVG_INIT = USEC_PER_SEC / 4 << RTTSCALE, + RTTDEV_INIT = RTTAVG_INIT / 4, + + HARD_SCORN_SECS = 10, /* try another remote port after this */ + MAX_TAINT = 1000, /* cap on aoetgt taint */ }; struct buf { @@ -100,10 +108,17 @@ struct buf { struct request *rq; }; +enum frame_flags { + FFL_PROBE = 1, +}; + struct frame { struct list_head head; u32 tag; + struct timeval sent; /* high-res time packet was sent */ + u32 sent_jiffs; /* low-res jiffies-based sent time */ ulong waited; + ulong waited_total; struct aoetgt *t; /* parent target I belong to */ sector_t lba; struct sk_buff *skb; /* command skb freed on module exit */ @@ -112,6 +127,7 @@ struct frame { struct bio_vec *bv; ulong bcnt; ulong bv_off; + char flags; }; struct aoeif { @@ -122,28 +138,31 @@ struct aoeif { struct aoetgt { unsigned char addr[6]; - ushort nframes; + ushort nframes; /* cap on frames to use */ struct aoedev *d; /* parent device I belong to */ struct list_head ffree; /* list of free frames */ struct aoeif ifs[NAOEIFS]; struct aoeif *ifp; /* current aoeif in use */ - ushort nout; - ushort maxout; - ulong falloc; - ulong lastwadj; /* last window adjustment */ + ushort nout; /* number of AoE commands outstanding */ + ushort maxout; /* current value for max outstanding */ + ushort next_cwnd; /* incr maxout after decrementing to zero */ + ushort ssthresh; /* slow start threshold */ + ulong falloc; /* number of allocated frames */ + int taint; /* how much we want to avoid this aoetgt */ int minbcnt; int wpkts, rpkts; + char nout_probes; }; struct aoedev { struct aoedev *next; ulong sysminor; ulong aoemajor; + u32 rttavg; /* scaled AoE round trip time average */ + u32 rttdev; /* scaled round trip time mean deviation */ u16 aoeminor; u16 flags; u16 nopen; /* (bd_openers isn't available without sleeping) */ - u16 rttavg; /* round trip average of requests/responses */ - u16 mintimer; u16 fw_ver; /* version of blade's firmware */ u16 lasttag; /* last tag sent */ u16 useme; @@ -151,7 +170,7 @@ struct aoedev { struct work_struct work;/* disk create work struct */ struct gendisk *gd; struct request_queue *blkq; - struct hd_geometry geo; + struct hd_geometry geo; sector_t ssize; struct timer_list timer; spinlock_t lock; @@ -164,11 +183,12 @@ struct aoedev { } ip; ulong maxbcnt; struct list_head factive[NFACTIVE]; /* hash of active frames */ - struct aoetgt *targets[NTARGETS]; + struct list_head rexmitq; /* deferred retransmissions */ + struct aoetgt **targets; + ulong ntargets; /* number of allocated aoetgt pointers */ struct aoetgt **tgt; /* target in use when working */ - struct aoetgt *htgt; /* target needing rexmit assistance */ - ulong ntargets; ulong kicked; + char ident[512]; }; /* kthread tracking */ @@ -195,6 +215,7 @@ void aoecmd_cfg(ushort aoemajor, unsigned char aoeminor); struct sk_buff *aoecmd_ata_rsp(struct sk_buff *); void aoecmd_cfg_rsp(struct sk_buff *); void aoecmd_sleepwork(struct work_struct *); +void aoecmd_wreset(struct aoetgt *t); void aoecmd_cleanslate(struct aoedev *); void aoecmd_exit(void); int aoecmd_init(void); diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index 00dfc5008ad4..a129f8c8073d 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -16,11 +16,19 @@ #include <linux/netdevice.h> #include <linux/mutex.h> #include <linux/export.h> +#include <linux/moduleparam.h> +#include <scsi/sg.h> #include "aoe.h" static DEFINE_MUTEX(aoeblk_mutex); static struct kmem_cache *buf_pool_cache; +/* GPFS needs a larger value than the default. */ +static int aoe_maxsectors; +module_param(aoe_maxsectors, int, 0644); +MODULE_PARM_DESC(aoe_maxsectors, + "When nonzero, set the maximum number of sectors per I/O request"); + static ssize_t aoedisk_show_state(struct device *dev, struct device_attribute *attr, char *page) { @@ -59,7 +67,7 @@ static ssize_t aoedisk_show_netif(struct device *dev, nd = nds; ne = nd + ARRAY_SIZE(nds); t = d->targets; - te = t + NTARGETS; + te = t + d->ntargets; for (; t < te && *t; t++) { ifp = (*t)->ifs; e = ifp + NAOEIFS; @@ -91,6 +99,14 @@ static ssize_t aoedisk_show_fwver(struct device *dev, return snprintf(page, PAGE_SIZE, "0x%04x\n", (unsigned int) d->fw_ver); } +static ssize_t aoedisk_show_payload(struct device *dev, + struct device_attribute *attr, char *page) +{ + struct gendisk *disk = dev_to_disk(dev); + struct aoedev *d = disk->private_data; + + return snprintf(page, PAGE_SIZE, "%lu\n", d->maxbcnt); +} static DEVICE_ATTR(state, S_IRUGO, aoedisk_show_state, NULL); static DEVICE_ATTR(mac, S_IRUGO, aoedisk_show_mac, NULL); @@ -99,12 +115,14 @@ static struct device_attribute dev_attr_firmware_version = { .attr = { .name = "firmware-version", .mode = S_IRUGO }, .show = aoedisk_show_fwver, }; +static DEVICE_ATTR(payload, S_IRUGO, aoedisk_show_payload, NULL); static struct attribute *aoe_attrs[] = { &dev_attr_state.attr, &dev_attr_mac.attr, &dev_attr_netif.attr, &dev_attr_firmware_version.attr, + &dev_attr_payload.attr, NULL, }; @@ -129,9 +147,18 @@ aoeblk_open(struct block_device *bdev, fmode_t mode) struct aoedev *d = bdev->bd_disk->private_data; ulong flags; + if (!virt_addr_valid(d)) { + pr_crit("aoe: invalid device pointer in %s\n", + __func__); + WARN_ON(1); + return -ENODEV; + } + if (!(d->flags & DEVFL_UP) || d->flags & DEVFL_TKILL) + return -ENODEV; + mutex_lock(&aoeblk_mutex); spin_lock_irqsave(&d->lock, flags); - if (d->flags & DEVFL_UP) { + if (d->flags & DEVFL_UP && !(d->flags & DEVFL_TKILL)) { d->nopen++; spin_unlock_irqrestore(&d->lock, flags); mutex_unlock(&aoeblk_mutex); @@ -195,9 +222,38 @@ aoeblk_getgeo(struct block_device *bdev, struct hd_geometry *geo) return 0; } +static int +aoeblk_ioctl(struct block_device *bdev, fmode_t mode, uint cmd, ulong arg) +{ + struct aoedev *d; + + if (!arg) + return -EINVAL; + + d = bdev->bd_disk->private_data; + if ((d->flags & DEVFL_UP) == 0) { + pr_err("aoe: disk not up\n"); + return -ENODEV; + } + + if (cmd == HDIO_GET_IDENTITY) { + if (!copy_to_user((void __user *) arg, &d->ident, + sizeof(d->ident))) + return 0; + return -EFAULT; + } + + /* udev calls scsi_id, which uses SG_IO, resulting in noise */ + if (cmd != SG_IO) + pr_info("aoe: unknown ioctl 0x%x\n", cmd); + + return -ENOTTY; +} + static const struct block_device_operations aoe_bdops = { .open = aoeblk_open, .release = aoeblk_release, + .ioctl = aoeblk_ioctl, .getgeo = aoeblk_getgeo, .owner = THIS_MODULE, }; @@ -212,6 +268,18 @@ aoeblk_gdalloc(void *vp) struct request_queue *q; enum { KB = 1024, MB = KB * KB, READ_AHEAD = 2 * MB, }; ulong flags; + int late = 0; + + spin_lock_irqsave(&d->lock, flags); + if (d->flags & DEVFL_GDALLOC + && !(d->flags & DEVFL_TKILL) + && !(d->flags & DEVFL_GD_NOW)) + d->flags |= DEVFL_GD_NOW; + else + late = 1; + spin_unlock_irqrestore(&d->lock, flags); + if (late) + return; gd = alloc_disk(AOE_PARTITIONS); if (gd == NULL) { @@ -231,23 +299,24 @@ aoeblk_gdalloc(void *vp) if (q == NULL) { pr_err("aoe: cannot allocate block queue for %ld.%d\n", d->aoemajor, d->aoeminor); - mempool_destroy(mp); - goto err_disk; + goto err_mempool; } - d->blkq = blk_alloc_queue(GFP_KERNEL); - if (!d->blkq) - goto err_mempool; - d->blkq->backing_dev_info.name = "aoe"; - if (bdi_init(&d->blkq->backing_dev_info)) - goto err_blkq; spin_lock_irqsave(&d->lock, flags); - blk_queue_max_hw_sectors(d->blkq, BLK_DEF_MAX_SECTORS); + WARN_ON(!(d->flags & DEVFL_GD_NOW)); + WARN_ON(!(d->flags & DEVFL_GDALLOC)); + WARN_ON(d->flags & DEVFL_TKILL); + WARN_ON(d->gd); + WARN_ON(d->flags & DEVFL_UP); + blk_queue_max_hw_sectors(q, BLK_DEF_MAX_SECTORS); + q->backing_dev_info.name = "aoe"; q->backing_dev_info.ra_pages = READ_AHEAD / PAGE_CACHE_SIZE; d->bufpool = mp; d->blkq = gd->queue = q; q->queuedata = d; d->gd = gd; + if (aoe_maxsectors) + blk_queue_max_hw_sectors(q, aoe_maxsectors); gd->major = AOE_MAJOR; gd->first_minor = d->sysminor; gd->fops = &aoe_bdops; @@ -263,18 +332,21 @@ aoeblk_gdalloc(void *vp) add_disk(gd); aoedisk_add_sysfs(d); + + spin_lock_irqsave(&d->lock, flags); + WARN_ON(!(d->flags & DEVFL_GD_NOW)); + d->flags &= ~DEVFL_GD_NOW; + spin_unlock_irqrestore(&d->lock, flags); return; -err_blkq: - blk_cleanup_queue(d->blkq); - d->blkq = NULL; err_mempool: - mempool_destroy(d->bufpool); + mempool_destroy(mp); err_disk: put_disk(gd); err: spin_lock_irqsave(&d->lock, flags); - d->flags &= ~DEVFL_GDALLOC; + d->flags &= ~DEVFL_GD_NOW; + schedule_work(&d->work); spin_unlock_irqrestore(&d->lock, flags); } diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c index ed57a890c643..42e67ad6bd20 100644 --- a/drivers/block/aoe/aoechr.c +++ b/drivers/block/aoe/aoechr.c @@ -39,6 +39,11 @@ struct ErrMsg { }; static DEFINE_MUTEX(aoechr_mutex); + +/* A ring buffer of error messages, to be read through + * "/dev/etherd/err". When no messages are present, + * readers will block waiting for messages to appear. + */ static struct ErrMsg emsgs[NMSG]; static int emsgs_head_idx, emsgs_tail_idx; static struct completion emsgs_comp; @@ -282,7 +287,7 @@ aoechr_init(void) int n, i; n = register_chrdev(AOE_MAJOR, "aoechr", &aoe_fops); - if (n < 0) { + if (n < 0) { printk(KERN_ERR "aoe: can't register char device\n"); return n; } diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 9fe4f1865558..25ef5c014fca 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -22,6 +22,7 @@ #define MAXIOC (8192) /* default meant to avoid most soft lockups */ static void ktcomplete(struct frame *, struct sk_buff *); +static int count_targets(struct aoedev *d, int *untainted); static struct buf *nextbuf(struct aoedev *); @@ -29,7 +30,7 @@ static int aoe_deadsecs = 60 * 3; module_param(aoe_deadsecs, int, 0644); MODULE_PARM_DESC(aoe_deadsecs, "After aoe_deadsecs seconds, give up and fail dev."); -static int aoe_maxout = 16; +static int aoe_maxout = 64; module_param(aoe_maxout, int, 0644); MODULE_PARM_DESC(aoe_maxout, "Only aoe_maxout outstanding packets for every MAC on eX.Y."); @@ -43,6 +44,8 @@ static struct { spinlock_t lock; } iocq; +static struct page *empty_page; + static struct sk_buff * new_skb(ulong len) { @@ -59,6 +62,23 @@ new_skb(ulong len) } static struct frame * +getframe_deferred(struct aoedev *d, u32 tag) +{ + struct list_head *head, *pos, *nx; + struct frame *f; + + head = &d->rexmitq; + list_for_each_safe(pos, nx, head) { + f = list_entry(pos, struct frame, head); + if (f->tag == tag) { + list_del(pos); + return f; + } + } + return NULL; +} + +static struct frame * getframe(struct aoedev *d, u32 tag) { struct frame *f; @@ -162,8 +182,10 @@ aoe_freetframe(struct frame *f) t = f->t; f->buf = NULL; + f->lba = 0; f->bv = NULL; f->r_skb = NULL; + f->flags = 0; list_add(&f->head, &t->ffree); } @@ -217,20 +239,25 @@ newframe(struct aoedev *d) struct frame *f; struct aoetgt *t, **tt; int totout = 0; + int use_tainted; + int has_untainted; - if (d->targets[0] == NULL) { /* shouldn't happen, but I'm paranoid */ + if (!d->targets || !d->targets[0]) { printk(KERN_ERR "aoe: NULL TARGETS!\n"); return NULL; } tt = d->tgt; /* last used target */ - for (;;) { + for (use_tainted = 0, has_untainted = 0;;) { tt++; - if (tt >= &d->targets[NTARGETS] || !*tt) + if (tt >= &d->targets[d->ntargets] || !*tt) tt = d->targets; t = *tt; - totout += t->nout; + if (!t->taint) { + has_untainted = 1; + totout += t->nout; + } if (t->nout < t->maxout - && t != d->htgt + && (use_tainted || !t->taint) && t->ifp->nd) { f = newtframe(d, t); if (f) { @@ -239,8 +266,12 @@ newframe(struct aoedev *d) return f; } } - if (tt == d->tgt) /* we've looped and found nada */ - break; + if (tt == d->tgt) { /* we've looped and found nada */ + if (!use_tainted && !has_untainted) + use_tainted = 1; + else + break; + } } if (totout == 0) { d->kicked++; @@ -277,21 +308,68 @@ fhash(struct frame *f) list_add_tail(&f->head, &d->factive[n]); } +static void +ata_rw_frameinit(struct frame *f) +{ + struct aoetgt *t; + struct aoe_hdr *h; + struct aoe_atahdr *ah; + struct sk_buff *skb; + char writebit, extbit; + + skb = f->skb; + h = (struct aoe_hdr *) skb_mac_header(skb); + ah = (struct aoe_atahdr *) (h + 1); + skb_put(skb, sizeof(*h) + sizeof(*ah)); + memset(h, 0, skb->len); + + writebit = 0x10; + extbit = 0x4; + + t = f->t; + f->tag = aoehdr_atainit(t->d, t, h); + fhash(f); + t->nout++; + f->waited = 0; + f->waited_total = 0; + if (f->buf) + f->lba = f->buf->sector; + + /* set up ata header */ + ah->scnt = f->bcnt >> 9; + put_lba(ah, f->lba); + if (t->d->flags & DEVFL_EXT) { + ah->aflags |= AOEAFL_EXT; + } else { + extbit = 0; + ah->lba3 &= 0x0f; + ah->lba3 |= 0xe0; /* LBA bit + obsolete 0xa0 */ + } + if (f->buf && bio_data_dir(f->buf->bio) == WRITE) { + skb_fillup(skb, f->bv, f->bv_off, f->bcnt); + ah->aflags |= AOEAFL_WRITE; + skb->len += f->bcnt; + skb->data_len = f->bcnt; + skb->truesize += f->bcnt; + t->wpkts++; + } else { + t->rpkts++; + writebit = 0; + } + + ah->cmdstat = ATA_CMD_PIO_READ | writebit | extbit; + skb->dev = t->ifp->nd; +} + static int aoecmd_ata_rw(struct aoedev *d) { struct frame *f; - struct aoe_hdr *h; - struct aoe_atahdr *ah; struct buf *buf; struct aoetgt *t; struct sk_buff *skb; struct sk_buff_head queue; ulong bcnt, fbcnt; - char writebit, extbit; - - writebit = 0x10; - extbit = 0x4; buf = nextbuf(d); if (buf == NULL) @@ -326,50 +404,18 @@ aoecmd_ata_rw(struct aoedev *d) } while (fbcnt); /* initialize the headers & frame */ - skb = f->skb; - h = (struct aoe_hdr *) skb_mac_header(skb); - ah = (struct aoe_atahdr *) (h+1); - skb_put(skb, sizeof *h + sizeof *ah); - memset(h, 0, skb->len); - f->tag = aoehdr_atainit(d, t, h); - fhash(f); - t->nout++; - f->waited = 0; f->buf = buf; f->bcnt = bcnt; - f->lba = buf->sector; - - /* set up ata header */ - ah->scnt = bcnt >> 9; - put_lba(ah, buf->sector); - if (d->flags & DEVFL_EXT) { - ah->aflags |= AOEAFL_EXT; - } else { - extbit = 0; - ah->lba3 &= 0x0f; - ah->lba3 |= 0xe0; /* LBA bit + obsolete 0xa0 */ - } - if (bio_data_dir(buf->bio) == WRITE) { - skb_fillup(skb, f->bv, f->bv_off, bcnt); - ah->aflags |= AOEAFL_WRITE; - skb->len += bcnt; - skb->data_len = bcnt; - skb->truesize += bcnt; - t->wpkts++; - } else { - t->rpkts++; - writebit = 0; - } - - ah->cmdstat = ATA_CMD_PIO_READ | writebit | extbit; + ata_rw_frameinit(f); /* mark all tracking fields and load out */ buf->nframesout += 1; buf->sector += bcnt >> 9; - skb->dev = t->ifp->nd; - skb = skb_clone(skb, GFP_ATOMIC); + skb = skb_clone(f->skb, GFP_ATOMIC); if (skb) { + do_gettimeofday(&f->sent); + f->sent_jiffs = (u32) jiffies; __skb_queue_head_init(&queue); __skb_queue_tail(&queue, skb); aoenet_xmit(&queue); @@ -442,11 +488,14 @@ resend(struct aoedev *d, struct frame *f) h = (struct aoe_hdr *) skb_mac_header(skb); ah = (struct aoe_atahdr *) (h+1); - snprintf(buf, sizeof buf, - "%15s e%ld.%d oldtag=%08x@%08lx newtag=%08x s=%pm d=%pm nout=%d\n", - "retransmit", d->aoemajor, d->aoeminor, f->tag, jiffies, n, - h->src, h->dst, t->nout); - aoechr_error(buf); + if (!(f->flags & FFL_PROBE)) { + snprintf(buf, sizeof(buf), + "%15s e%ld.%d oldtag=%08x@%08lx newtag=%08x s=%pm d=%pm nout=%d\n", + "retransmit", d->aoemajor, d->aoeminor, + f->tag, jiffies, n, + h->src, h->dst, t->nout); + aoechr_error(buf); + } f->tag = n; fhash(f); @@ -458,12 +507,46 @@ resend(struct aoedev *d, struct frame *f) skb = skb_clone(skb, GFP_ATOMIC); if (skb == NULL) return; + do_gettimeofday(&f->sent); + f->sent_jiffs = (u32) jiffies; __skb_queue_head_init(&queue); __skb_queue_tail(&queue, skb); aoenet_xmit(&queue); } static int +tsince_hr(struct frame *f) +{ + struct timeval now; + int n; + + do_gettimeofday(&now); + n = now.tv_usec - f->sent.tv_usec; + n += (now.tv_sec - f->sent.tv_sec) * USEC_PER_SEC; + + if (n < 0) + n = -n; + + /* For relatively long periods, use jiffies to avoid + * discrepancies caused by updates to the system time. + * + * On system with HZ of 1000, 32-bits is over 49 days + * worth of jiffies, or over 71 minutes worth of usecs. + * + * Jiffies overflow is handled by subtraction of unsigned ints: + * (gdb) print (unsigned) 2 - (unsigned) 0xfffffffe + * $3 = 4 + * (gdb) + */ + if (n > USEC_PER_SEC / 4) { + n = ((u32) jiffies) - f->sent_jiffs; + n *= USEC_PER_SEC / HZ; + } + + return n; +} + +static int tsince(u32 tag) { int n; @@ -472,7 +555,7 @@ tsince(u32 tag) n -= tag & 0xffff; if (n < 0) n += 1<<16; - return n; + return jiffies_to_usecs(n + 1); } static struct aoeif * @@ -503,70 +586,189 @@ ejectif(struct aoetgt *t, struct aoeif *ifp) dev_put(nd); } -static int -sthtith(struct aoedev *d) +static struct frame * +reassign_frame(struct frame *f) { - struct frame *f, *nf; - struct list_head *nx, *pos, *head; + struct frame *nf; struct sk_buff *skb; - struct aoetgt *ht = d->htgt; - int i; - for (i = 0; i < NFACTIVE; i++) { - head = &d->factive[i]; - list_for_each_safe(pos, nx, head) { - f = list_entry(pos, struct frame, head); - if (f->t != ht) - continue; + nf = newframe(f->t->d); + if (!nf) + return NULL; + if (nf->t == f->t) { + aoe_freetframe(nf); + return NULL; + } - nf = newframe(d); - if (!nf) - return 0; + skb = nf->skb; + nf->skb = f->skb; + nf->buf = f->buf; + nf->bcnt = f->bcnt; + nf->lba = f->lba; + nf->bv = f->bv; + nf->bv_off = f->bv_off; + nf->waited = 0; + nf->waited_total = f->waited_total; + nf->sent = f->sent; + nf->sent_jiffs = f->sent_jiffs; + f->skb = skb; + + return nf; +} - /* remove frame from active list */ - list_del(pos); +static void +probe(struct aoetgt *t) +{ + struct aoedev *d; + struct frame *f; + struct sk_buff *skb; + struct sk_buff_head queue; + size_t n, m; + int frag; - /* reassign all pertinent bits to new outbound frame */ - skb = nf->skb; - nf->skb = f->skb; - nf->buf = f->buf; - nf->bcnt = f->bcnt; - nf->lba = f->lba; - nf->bv = f->bv; - nf->bv_off = f->bv_off; - nf->waited = 0; - f->skb = skb; + d = t->d; + f = newtframe(d, t); + if (!f) { + pr_err("%s %pm for e%ld.%d: %s\n", + "aoe: cannot probe remote address", + t->addr, + (long) d->aoemajor, d->aoeminor, + "no frame available"); + return; + } + f->flags |= FFL_PROBE; + ifrotate(t); + f->bcnt = t->d->maxbcnt ? t->d->maxbcnt : DEFAULTBCNT; + ata_rw_frameinit(f); + skb = f->skb; + for (frag = 0, n = f->bcnt; n > 0; ++frag, n -= m) { + if (n < PAGE_SIZE) + m = n; + else + m = PAGE_SIZE; + skb_fill_page_desc(skb, frag, empty_page, 0, m); + } + skb->len += f->bcnt; + skb->data_len = f->bcnt; + skb->truesize += f->bcnt; + + skb = skb_clone(f->skb, GFP_ATOMIC); + if (skb) { + do_gettimeofday(&f->sent); + f->sent_jiffs = (u32) jiffies; + __skb_queue_head_init(&queue); + __skb_queue_tail(&queue, skb); + aoenet_xmit(&queue); + } +} + +static long +rto(struct aoedev *d) +{ + long t; + + t = 2 * d->rttavg >> RTTSCALE; + t += 8 * d->rttdev >> RTTDSCALE; + if (t == 0) + t = 1; + + return t; +} + +static void +rexmit_deferred(struct aoedev *d) +{ + struct aoetgt *t; + struct frame *f; + struct frame *nf; + struct list_head *pos, *nx, *head; + int since; + int untainted; + + count_targets(d, &untainted); + + head = &d->rexmitq; + list_for_each_safe(pos, nx, head) { + f = list_entry(pos, struct frame, head); + t = f->t; + if (t->taint) { + if (!(f->flags & FFL_PROBE)) { + nf = reassign_frame(f); + if (nf) { + if (t->nout_probes == 0 + && untainted > 0) { + probe(t); + t->nout_probes++; + } + list_replace(&f->head, &nf->head); + pos = &nf->head; + aoe_freetframe(f); + f = nf; + t = f->t; + } + } else if (untainted < 1) { + /* don't probe w/o other untainted aoetgts */ + goto stop_probe; + } else if (tsince_hr(f) < t->taint * rto(d)) { + /* reprobe slowly when taint is high */ + continue; + } + } else if (f->flags & FFL_PROBE) { +stop_probe: /* don't probe untainted aoetgts */ + list_del(pos); aoe_freetframe(f); - ht->nout--; - nf->t->nout++; - resend(d, nf); + /* leaving d->kicked, because this is routine */ + f->t->d->flags |= DEVFL_KICKME; + continue; } + if (t->nout >= t->maxout) + continue; + list_del(pos); + t->nout++; + if (f->flags & FFL_PROBE) + t->nout_probes++; + since = tsince_hr(f); + f->waited += since; + f->waited_total += since; + resend(d, f); } - /* We've cleaned up the outstanding so take away his - * interfaces so he won't be used. We should remove him from - * the target array here, but cleaning up a target is - * involved. PUNT! - */ - memset(ht->ifs, 0, sizeof ht->ifs); - d->htgt = NULL; - return 1; } -static inline unsigned char -ata_scnt(unsigned char *packet) { - struct aoe_hdr *h; - struct aoe_atahdr *ah; +/* An aoetgt accumulates demerits quickly, and successful + * probing redeems the aoetgt slowly. + */ +static void +scorn(struct aoetgt *t) +{ + int n; - h = (struct aoe_hdr *) packet; - ah = (struct aoe_atahdr *) (h+1); - return ah->scnt; + n = t->taint++; + t->taint += t->taint * 2; + if (n > t->taint) + t->taint = n; + if (t->taint > MAX_TAINT) + t->taint = MAX_TAINT; +} + +static int +count_targets(struct aoedev *d, int *untainted) +{ + int i, good; + + for (i = good = 0; i < d->ntargets && d->targets[i]; ++i) + if (d->targets[i]->taint == 0) + good++; + + if (untainted) + *untainted = good; + return i; } static void rexmit_timer(ulong vp) { struct aoedev *d; - struct aoetgt *t, **tt, **te; + struct aoetgt *t; struct aoeif *ifp; struct frame *f; struct list_head *head, *pos, *nx; @@ -574,15 +776,18 @@ rexmit_timer(ulong vp) register long timeout; ulong flags, n; int i; + int utgts; /* number of aoetgt descriptors (not slots) */ + int since; d = (struct aoedev *) vp; - /* timeout is always ~150% of the moving average */ - timeout = d->rttavg; - timeout += timeout >> 1; - spin_lock_irqsave(&d->lock, flags); + /* timeout based on observed timings and variations */ + timeout = rto(d); + + utgts = count_targets(d, NULL); + if (d->flags & DEVFL_TKILL) { spin_unlock_irqrestore(&d->lock, flags); return; @@ -593,67 +798,61 @@ rexmit_timer(ulong vp) head = &d->factive[i]; list_for_each_safe(pos, nx, head) { f = list_entry(pos, struct frame, head); - if (tsince(f->tag) < timeout) + if (tsince_hr(f) < timeout) break; /* end of expired frames */ /* move to flist for later processing */ list_move_tail(pos, &flist); } } - /* window check */ - tt = d->targets; - te = tt + d->ntargets; - for (; tt < te && (t = *tt); tt++) { - if (t->nout == t->maxout - && t->maxout < t->nframes - && (jiffies - t->lastwadj)/HZ > 10) { - t->maxout++; - t->lastwadj = jiffies; - } - } - - if (!list_empty(&flist)) { /* retransmissions necessary */ - n = d->rttavg <<= 1; - if (n > MAXTIMER) - d->rttavg = MAXTIMER; - } /* process expired frames */ while (!list_empty(&flist)) { pos = flist.next; f = list_entry(pos, struct frame, head); - n = f->waited += timeout; - n /= HZ; - if (n > aoe_deadsecs) { + since = tsince_hr(f); + n = f->waited_total + since; + n /= USEC_PER_SEC; + if (aoe_deadsecs + && n > aoe_deadsecs + && !(f->flags & FFL_PROBE)) { /* Waited too long. Device failure. * Hang all frames on first hash bucket for downdev * to clean up. */ list_splice(&flist, &d->factive[0]); aoedev_downdev(d); - break; + goto out; } - list_del(pos); t = f->t; - if (n > aoe_deadsecs/2) - d->htgt = t; /* see if another target can help */ - - if (t->nout == t->maxout) { - if (t->maxout > 1) - t->maxout--; - t->lastwadj = jiffies; + n = f->waited + since; + n /= USEC_PER_SEC; + if (aoe_deadsecs && utgts > 0 + && (n > aoe_deadsecs / utgts || n > HARD_SCORN_SECS)) + scorn(t); /* avoid this target */ + + if (t->maxout != 1) { + t->ssthresh = t->maxout / 2; + t->maxout = 1; } - ifp = getif(t, f->skb->dev); - if (ifp && ++ifp->lost > (t->nframes << 1) - && (ifp != t->ifs || t->ifs[1].nd)) { - ejectif(t, ifp); - ifp = NULL; + if (f->flags & FFL_PROBE) { + t->nout_probes--; + } else { + ifp = getif(t, f->skb->dev); + if (ifp && ++ifp->lost > (t->nframes << 1) + && (ifp != t->ifs || t->ifs[1].nd)) { + ejectif(t, ifp); + ifp = NULL; + } } - resend(d, f); + list_move_tail(pos, &d->rexmitq); + t->nout--; } + rexmit_deferred(d); - if ((d->flags & DEVFL_KICKME || d->htgt) && d->blkq) { +out: + if ((d->flags & DEVFL_KICKME) && d->blkq) { d->flags &= ~DEVFL_KICKME; d->blkq->request_fn(d->blkq); } @@ -774,8 +973,7 @@ nextbuf(struct aoedev *d) void aoecmd_work(struct aoedev *d) { - if (d->htgt && !sthtith(d)) - return; + rexmit_deferred(d); while (aoecmd_ata_rw(d)) ; } @@ -809,6 +1007,17 @@ aoecmd_sleepwork(struct work_struct *work) } static void +ata_ident_fixstring(u16 *id, int ns) +{ + u16 s; + + while (ns-- > 0) { + s = *id; + *id++ = s >> 8 | s << 8; + } +} + +static void ataid_complete(struct aoedev *d, struct aoetgt *t, unsigned char *id) { u64 ssize; @@ -843,6 +1052,11 @@ ataid_complete(struct aoedev *d, struct aoetgt *t, unsigned char *id) d->geo.sectors = get_unaligned_le16(&id[56 << 1]); } + ata_ident_fixstring((u16 *) &id[10<<1], 10); /* serial */ + ata_ident_fixstring((u16 *) &id[23<<1], 4); /* firmware */ + ata_ident_fixstring((u16 *) &id[27<<1], 20); /* model */ + memcpy(d->ident, id, sizeof(d->ident)); + if (d->ssize != ssize) printk(KERN_INFO "aoe: %pm e%ld.%d v%04x has %llu sectors\n", @@ -862,26 +1076,28 @@ ataid_complete(struct aoedev *d, struct aoetgt *t, unsigned char *id) } static void -calc_rttavg(struct aoedev *d, int rtt) +calc_rttavg(struct aoedev *d, struct aoetgt *t, int rtt) { register long n; n = rtt; - if (n < 0) { - n = -rtt; - if (n < MINTIMER) - n = MINTIMER; - else if (n > MAXTIMER) - n = MAXTIMER; - d->mintimer += (n - d->mintimer) >> 1; - } else if (n < d->mintimer) - n = d->mintimer; - else if (n > MAXTIMER) - n = MAXTIMER; - - /* g == .25; cf. Congestion Avoidance and Control, Jacobson & Karels; 1988 */ - n -= d->rttavg; - d->rttavg += n >> 2; + + /* cf. Congestion Avoidance and Control, Jacobson & Karels, 1988 */ + n -= d->rttavg >> RTTSCALE; + d->rttavg += n; + if (n < 0) + n = -n; + n -= d->rttdev >> RTTDSCALE; + d->rttdev += n; + + if (!t || t->maxout >= t->nframes) + return; + if (t->maxout < t->ssthresh) + t->maxout += 1; + else if (t->nout == t->maxout && t->next_cwnd-- == 0) { + t->maxout += 1; + t->next_cwnd = t->maxout; + } } static struct aoetgt * @@ -890,7 +1106,7 @@ gettgt(struct aoedev *d, char *addr) struct aoetgt **t, **e; t = d->targets; - e = t + NTARGETS; + e = t + d->ntargets; for (; t < e && *t; t++) if (memcmp((*t)->addr, addr, sizeof((*t)->addr)) == 0) return *t; @@ -966,19 +1182,22 @@ ktiocomplete(struct frame *f) struct aoeif *ifp; struct aoedev *d; long n; + int untainted; if (f == NULL) return; t = f->t; d = t->d; + skb = f->r_skb; + buf = f->buf; + if (f->flags & FFL_PROBE) + goto out; + if (!skb) /* just fail the buf. */ + goto noskb; hout = (struct aoe_hdr *) skb_mac_header(f->skb); ahout = (struct aoe_atahdr *) (hout+1); - buf = f->buf; - skb = f->r_skb; - if (skb == NULL) - goto noskb; /* just fail the buf. */ hin = (struct aoe_hdr *) skb->data; skb_pull(skb, sizeof(*hin)); @@ -988,9 +1207,9 @@ ktiocomplete(struct frame *f) pr_err("aoe: ata error cmd=%2.2Xh stat=%2.2Xh from e%ld.%d\n", ahout->cmdstat, ahin->cmdstat, d->aoemajor, d->aoeminor); -noskb: if (buf) +noskb: if (buf) clear_bit(BIO_UPTODATE, &buf->bio->bi_flags); - goto badrsp; + goto out; } n = ahout->scnt << 9; @@ -998,8 +1217,10 @@ noskb: if (buf) case ATA_CMD_PIO_READ: case ATA_CMD_PIO_READ_EXT: if (skb->len < n) { - pr_err("aoe: runt data size in read. skb->len=%d need=%ld\n", - skb->len, n); + pr_err("%s e%ld.%d. skb->len=%d need=%ld\n", + "aoe: runt data size in read from", + (long) d->aoemajor, d->aoeminor, + skb->len, n); clear_bit(BIO_UPTODATE, &buf->bio->bi_flags); break; } @@ -1010,13 +1231,13 @@ noskb: if (buf) ifp = getif(t, skb->dev); if (ifp) ifp->lost = 0; - if (d->htgt == t) /* I'll help myself, thank you. */ - d->htgt = NULL; spin_unlock_irq(&d->lock); break; case ATA_CMD_ID_ATA: if (skb->len < 512) { - pr_info("aoe: runt data size in ataid. skb->len=%d\n", + pr_info("%s e%ld.%d. skb->len=%d need=512\n", + "aoe: runt data size in ataid from", + (long) d->aoemajor, d->aoeminor, skb->len); break; } @@ -1032,16 +1253,23 @@ noskb: if (buf) be16_to_cpu(get_unaligned(&hin->major)), hin->minor); } -badrsp: +out: spin_lock_irq(&d->lock); + if (t->taint > 0 + && --t->taint > 0 + && t->nout_probes == 0) { + count_targets(d, &untainted); + if (untainted > 0) { + probe(t); + t->nout_probes++; + } + } aoe_freetframe(f); if (buf && --buf->nframesout == 0 && buf->resid == 0) aoe_end_buf(d, buf); - aoecmd_work(d); - spin_unlock_irq(&d->lock); aoedev_put(d); dev_kfree_skb(skb); @@ -1141,7 +1369,6 @@ aoecmd_ata_rsp(struct sk_buff *skb) struct aoedev *d; struct aoe_hdr *h; struct frame *f; - struct aoetgt *t; u32 n; ulong flags; char ebuf[128]; @@ -1162,23 +1389,32 @@ aoecmd_ata_rsp(struct sk_buff *skb) n = be32_to_cpu(get_unaligned(&h->tag)); f = getframe(d, n); - if (f == NULL) { - calc_rttavg(d, -tsince(n)); - spin_unlock_irqrestore(&d->lock, flags); - aoedev_put(d); - snprintf(ebuf, sizeof ebuf, - "%15s e%d.%d tag=%08x@%08lx\n", - "unexpected rsp", - get_unaligned_be16(&h->major), - h->minor, - get_unaligned_be32(&h->tag), - jiffies); - aoechr_error(ebuf); - return skb; + if (f) { + calc_rttavg(d, f->t, tsince_hr(f)); + f->t->nout--; + if (f->flags & FFL_PROBE) + f->t->nout_probes--; + } else { + f = getframe_deferred(d, n); + if (f) { + calc_rttavg(d, NULL, tsince_hr(f)); + } else { + calc_rttavg(d, NULL, tsince(n)); + spin_unlock_irqrestore(&d->lock, flags); + aoedev_put(d); + snprintf(ebuf, sizeof(ebuf), + "%15s e%d.%d tag=%08x@%08lx s=%pm d=%pm\n", + "unexpected rsp", + get_unaligned_be16(&h->major), + h->minor, + get_unaligned_be32(&h->tag), + jiffies, + h->src, + h->dst); + aoechr_error(ebuf); + return skb; + } } - t = f->t; - calc_rttavg(d, tsince(f->tag)); - t->nout--; aoecmd_work(d); spin_unlock_irqrestore(&d->lock, flags); @@ -1201,7 +1437,7 @@ aoecmd_cfg(ushort aoemajor, unsigned char aoeminor) aoecmd_cfg_pkts(aoemajor, aoeminor, &queue); aoenet_xmit(&queue); } - + struct sk_buff * aoecmd_ata_id(struct aoedev *d) { @@ -1227,6 +1463,7 @@ aoecmd_ata_id(struct aoedev *d) fhash(f); t->nout++; f->waited = 0; + f->waited_total = 0; /* set up ata header */ ah->scnt = 1; @@ -1235,41 +1472,69 @@ aoecmd_ata_id(struct aoedev *d) skb->dev = t->ifp->nd; - d->rttavg = MAXTIMER; + d->rttavg = RTTAVG_INIT; + d->rttdev = RTTDEV_INIT; d->timer.function = rexmit_timer; - return skb_clone(skb, GFP_ATOMIC); + skb = skb_clone(skb, GFP_ATOMIC); + if (skb) { + do_gettimeofday(&f->sent); + f->sent_jiffs = (u32) jiffies; + } + + return skb; } - + +static struct aoetgt ** +grow_targets(struct aoedev *d) +{ + ulong oldn, newn; + struct aoetgt **tt; + + oldn = d->ntargets; + newn = oldn * 2; + tt = kcalloc(newn, sizeof(*d->targets), GFP_ATOMIC); + if (!tt) + return NULL; + memmove(tt, d->targets, sizeof(*d->targets) * oldn); + d->tgt = tt + (d->tgt - d->targets); + kfree(d->targets); + d->targets = tt; + d->ntargets = newn; + + return &d->targets[oldn]; +} + static struct aoetgt * addtgt(struct aoedev *d, char *addr, ulong nframes) { struct aoetgt *t, **tt, **te; tt = d->targets; - te = tt + NTARGETS; + te = tt + d->ntargets; for (; tt < te && *tt; tt++) ; if (tt == te) { - printk(KERN_INFO - "aoe: device addtgt failure; too many targets\n"); - return NULL; + tt = grow_targets(d); + if (!tt) + goto nomem; } t = kzalloc(sizeof(*t), GFP_ATOMIC); - if (!t) { - printk(KERN_INFO "aoe: cannot allocate memory to add target\n"); - return NULL; - } - - d->ntargets++; + if (!t) + goto nomem; t->nframes = nframes; t->d = d; memcpy(t->addr, addr, sizeof t->addr); t->ifp = t->ifs; - t->maxout = t->nframes; + aoecmd_wreset(t); + t->maxout = t->nframes / 2; INIT_LIST_HEAD(&t->ffree); return *tt = t; + + nomem: + pr_info("aoe: cannot allocate memory to add target\n"); + return NULL; } static void @@ -1279,7 +1544,7 @@ setdbcnt(struct aoedev *d) int bcnt = 0; t = d->targets; - e = t + NTARGETS; + e = t + d->ntargets; for (; t < e && *t; t++) if (bcnt == 0 || bcnt > (*t)->minbcnt) bcnt = (*t)->minbcnt; @@ -1373,7 +1638,11 @@ aoecmd_cfg_rsp(struct sk_buff *skb) spin_lock_irqsave(&d->lock, flags); t = gettgt(d, h->src); - if (!t) { + if (t) { + t->nframes = n; + if (n < t->maxout) + aoecmd_wreset(t); + } else { t = addtgt(d, h->src, n); if (!t) goto bail; @@ -1402,17 +1671,26 @@ bail: } void +aoecmd_wreset(struct aoetgt *t) +{ + t->maxout = 1; + t->ssthresh = t->nframes / 2; + t->next_cwnd = t->nframes; +} + +void aoecmd_cleanslate(struct aoedev *d) { struct aoetgt **t, **te; - d->mintimer = MINTIMER; + d->rttavg = RTTAVG_INIT; + d->rttdev = RTTDEV_INIT; d->maxbcnt = 0; t = d->targets; - te = t + NTARGETS; + te = t + d->ntargets; for (; t < te && *t; t++) - (*t)->maxout = (*t)->nframes; + aoecmd_wreset(*t); } void @@ -1460,6 +1738,14 @@ aoe_flush_iocq(void) int __init aoecmd_init(void) { + void *p; + + /* get_zeroed_page returns page with ref count 1 */ + p = (void *) get_zeroed_page(GFP_KERNEL | __GFP_REPEAT); + if (!p) + return -ENOMEM; + empty_page = virt_to_page(p); + INIT_LIST_HEAD(&iocq.head); spin_lock_init(&iocq.lock); init_waitqueue_head(&ktiowq); @@ -1475,4 +1761,7 @@ aoecmd_exit(void) { aoe_ktstop(&kts); aoe_flush_iocq(); + + free_page((unsigned long) page_address(empty_page)); + empty_page = NULL; } diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index 90e5b537f94b..98f2965778b9 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -15,7 +15,6 @@ #include "aoe.h" static void dummy_timer(ulong); -static void aoedev_freedev(struct aoedev *); static void freetgt(struct aoedev *d, struct aoetgt *t); static void skbpoolfree(struct aoedev *d); @@ -69,25 +68,34 @@ minor_get_static(ulong *sysminor, ulong aoemaj, int aoemin) NPERSHELF = 16, }; + if (aoemin >= NPERSHELF) { + pr_err("aoe: %s %d slots per shelf\n", + "static minor device numbers support only", + NPERSHELF); + error = -1; + goto out; + } + n = aoemaj * NPERSHELF + aoemin; - if (aoemin >= NPERSHELF || n >= N_DEVS) { + if (n >= N_DEVS) { pr_err("aoe: %s with e%ld.%d\n", "cannot use static minor device numbers", aoemaj, aoemin); error = -1; - } else { - spin_lock_irqsave(&used_minors_lock, flags); - if (test_bit(n, used_minors)) { - pr_err("aoe: %s %lu\n", - "existing device already has static minor number", - n); - error = -1; - } else - set_bit(n, used_minors); - spin_unlock_irqrestore(&used_minors_lock, flags); + goto out; } - *sysminor = n; + spin_lock_irqsave(&used_minors_lock, flags); + if (test_bit(n, used_minors)) { + pr_err("aoe: %s %lu\n", + "existing device already has static minor number", + n); + error = -1; + } else + set_bit(n, used_minors); + spin_unlock_irqrestore(&used_minors_lock, flags); + *sysminor = n * AOE_PARTITIONS; +out: return error; } @@ -170,41 +178,50 @@ aoe_failip(struct aoedev *d) aoe_end_request(d, rq, 0); } +static void +downdev_frame(struct list_head *pos) +{ + struct frame *f; + + f = list_entry(pos, struct frame, head); + list_del(pos); + if (f->buf) { + f->buf->nframesout--; + aoe_failbuf(f->t->d, f->buf); + } + aoe_freetframe(f); +} + void aoedev_downdev(struct aoedev *d) { struct aoetgt *t, **tt, **te; - struct frame *f; struct list_head *head, *pos, *nx; struct request *rq; int i; d->flags &= ~DEVFL_UP; - /* clean out active buffers */ + /* clean out active and to-be-retransmitted buffers */ for (i = 0; i < NFACTIVE; i++) { head = &d->factive[i]; - list_for_each_safe(pos, nx, head) { - f = list_entry(pos, struct frame, head); - list_del(pos); - if (f->buf) { - f->buf->nframesout--; - aoe_failbuf(d, f->buf); - } - aoe_freetframe(f); - } + list_for_each_safe(pos, nx, head) + downdev_frame(pos); } + head = &d->rexmitq; + list_for_each_safe(pos, nx, head) + downdev_frame(pos); + /* reset window dressings */ tt = d->targets; - te = tt + NTARGETS; + te = tt + d->ntargets; for (; tt < te && (t = *tt); tt++) { - t->maxout = t->nframes; + aoecmd_wreset(t); t->nout = 0; } /* clean out the in-process request (if any) */ aoe_failip(d); - d->htgt = NULL; /* fast fail all pending I/O */ if (d->blkq) { @@ -218,12 +235,48 @@ aoedev_downdev(struct aoedev *d) set_capacity(d->gd, 0); } +/* return whether the user asked for this particular + * device to be flushed + */ +static int +user_req(char *s, size_t slen, struct aoedev *d) +{ + char *p; + size_t lim; + + if (!d->gd) + return 0; + p = strrchr(d->gd->disk_name, '/'); + if (!p) + p = d->gd->disk_name; + else + p += 1; + lim = sizeof(d->gd->disk_name); + lim -= p - d->gd->disk_name; + if (slen < lim) + lim = slen; + + return !strncmp(s, p, lim); +} + static void -aoedev_freedev(struct aoedev *d) +freedev(struct aoedev *d) { struct aoetgt **t, **e; + int freeing = 0; + unsigned long flags; + + spin_lock_irqsave(&d->lock, flags); + if (d->flags & DEVFL_TKILL + && !(d->flags & DEVFL_FREEING)) { + d->flags |= DEVFL_FREEING; + freeing = 1; + } + spin_unlock_irqrestore(&d->lock, flags); + if (!freeing) + return; - cancel_work_sync(&d->work); + del_timer_sync(&d->timer); if (d->gd) { aoedisk_rm_sysfs(d); del_gendisk(d->gd); @@ -231,61 +284,113 @@ aoedev_freedev(struct aoedev *d) blk_cleanup_queue(d->blkq); } t = d->targets; - e = t + NTARGETS; + e = t + d->ntargets; for (; t < e && *t; t++) freetgt(d, *t); if (d->bufpool) mempool_destroy(d->bufpool); skbpoolfree(d); minor_free(d->sysminor); - kfree(d); + + spin_lock_irqsave(&d->lock, flags); + d->flags |= DEVFL_FREED; + spin_unlock_irqrestore(&d->lock, flags); } -int -aoedev_flush(const char __user *str, size_t cnt) +enum flush_parms { + NOT_EXITING = 0, + EXITING = 1, +}; + +static int +flush(const char __user *str, size_t cnt, int exiting) { ulong flags; struct aoedev *d, **dd; - struct aoedev *rmd = NULL; char buf[16]; int all = 0; + int specified = 0; /* flush a specific device */ + unsigned int skipflags; + + skipflags = DEVFL_GDALLOC | DEVFL_NEWSIZE | DEVFL_TKILL; - if (cnt >= 3) { + if (!exiting && cnt >= 3) { if (cnt > sizeof buf) cnt = sizeof buf; if (copy_from_user(buf, str, cnt)) return -EFAULT; all = !strncmp(buf, "all", 3); + if (!all) + specified = 1; } + flush_scheduled_work(); + /* pass one: without sleeping, do aoedev_downdev */ spin_lock_irqsave(&devlist_lock, flags); - dd = &devlist; - while ((d = *dd)) { + for (d = devlist; d; d = d->next) { spin_lock(&d->lock); - if ((!all && (d->flags & DEVFL_UP)) - || (d->flags & (DEVFL_GDALLOC|DEVFL_NEWSIZE)) + if (exiting) { + /* unconditionally take each device down */ + } else if (specified) { + if (!user_req(buf, cnt, d)) + goto cont; + } else if ((!all && (d->flags & DEVFL_UP)) + || d->flags & skipflags || d->nopen - || d->ref) { - spin_unlock(&d->lock); - dd = &d->next; - continue; - } - *dd = d->next; + || d->ref) + goto cont; + aoedev_downdev(d); d->flags |= DEVFL_TKILL; +cont: spin_unlock(&d->lock); - d->next = rmd; - rmd = d; } spin_unlock_irqrestore(&devlist_lock, flags); - while ((d = rmd)) { - rmd = d->next; - del_timer_sync(&d->timer); - aoedev_freedev(d); /* must be able to sleep */ + + /* pass two: call freedev, which might sleep, + * for aoedevs marked with DEVFL_TKILL + */ +restart: + spin_lock_irqsave(&devlist_lock, flags); + for (d = devlist; d; d = d->next) { + spin_lock(&d->lock); + if (d->flags & DEVFL_TKILL + && !(d->flags & DEVFL_FREEING)) { + spin_unlock(&d->lock); + spin_unlock_irqrestore(&devlist_lock, flags); + freedev(d); + goto restart; + } + spin_unlock(&d->lock); } + + /* pass three: remove aoedevs marked with DEVFL_FREED */ + for (dd = &devlist, d = *dd; d; d = *dd) { + struct aoedev *doomed = NULL; + + spin_lock(&d->lock); + if (d->flags & DEVFL_FREED) { + *dd = d->next; + doomed = d; + } else { + dd = &d->next; + } + spin_unlock(&d->lock); + if (doomed) + kfree(doomed->targets); + kfree(doomed); + } + spin_unlock_irqrestore(&devlist_lock, flags); + return 0; } +int +aoedev_flush(const char __user *str, size_t cnt) +{ + return flush(str, cnt, NOT_EXITING); +} + /* This has been confirmed to occur once with Tms=3*1000 due to the * driver changing link and not processing its transmit ring. The * problem is hard enough to solve by returning an error that I'm @@ -332,13 +437,20 @@ aoedev_by_aoeaddr(ulong maj, int min, int do_alloc) struct aoedev *d; int i; ulong flags; - ulong sysminor; + ulong sysminor = 0; spin_lock_irqsave(&devlist_lock, flags); for (d=devlist; d; d=d->next) if (d->aoemajor == maj && d->aoeminor == min) { + spin_lock(&d->lock); + if (d->flags & DEVFL_TKILL) { + spin_unlock(&d->lock); + d = NULL; + goto out; + } d->ref++; + spin_unlock(&d->lock); break; } if (d || !do_alloc || minor_get(&sysminor, maj, min) < 0) @@ -346,6 +458,13 @@ aoedev_by_aoeaddr(ulong maj, int min, int do_alloc) d = kcalloc(1, sizeof *d, GFP_ATOMIC); if (!d) goto out; + d->targets = kcalloc(NTARGETS, sizeof(*d->targets), GFP_ATOMIC); + if (!d->targets) { + kfree(d); + d = NULL; + goto out; + } + d->ntargets = NTARGETS; INIT_WORK(&d->work, aoecmd_sleepwork); spin_lock_init(&d->lock); skb_queue_head_init(&d->skbpool); @@ -359,10 +478,12 @@ aoedev_by_aoeaddr(ulong maj, int min, int do_alloc) d->ref = 1; for (i = 0; i < NFACTIVE; i++) INIT_LIST_HEAD(&d->factive[i]); + INIT_LIST_HEAD(&d->rexmitq); d->sysminor = sysminor; d->aoemajor = maj; d->aoeminor = min; - d->mintimer = MINTIMER; + d->rttavg = RTTAVG_INIT; + d->rttdev = RTTDEV_INIT; d->next = devlist; devlist = d; out: @@ -396,21 +517,9 @@ freetgt(struct aoedev *d, struct aoetgt *t) void aoedev_exit(void) { - struct aoedev *d; - ulong flags; - + flush_scheduled_work(); aoe_flush_iocq(); - while ((d = devlist)) { - devlist = d->next; - - spin_lock_irqsave(&d->lock, flags); - aoedev_downdev(d); - d->flags |= DEVFL_TKILL; - spin_unlock_irqrestore(&d->lock, flags); - - del_timer_sync(&d->timer); - aoedev_freedev(d); - } + flush(NULL, 0, EXITING); } int __init diff --git a/drivers/block/aoe/aoemain.c b/drivers/block/aoe/aoemain.c index 04793c2c701b..4b987c2fefbe 100644 --- a/drivers/block/aoe/aoemain.c +++ b/drivers/block/aoe/aoemain.c @@ -105,7 +105,7 @@ aoe_init(void) aoechr_exit(); chr_fail: aoedev_exit(); - + printk(KERN_INFO "aoe: initialisation failure.\n"); return ret; } diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index 162c6471275c..71d3ea8d3006 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -31,7 +31,7 @@ enum { static char aoe_iflist[IFLISTSZ]; module_param_string(aoe_iflist, aoe_iflist, IFLISTSZ, 0600); -MODULE_PARM_DESC(aoe_iflist, "aoe_iflist=\"dev1 [dev2 ...]\""); +MODULE_PARM_DESC(aoe_iflist, "aoe_iflist=dev1[,dev2...]"); static wait_queue_head_t txwq; static struct ktstate kts; @@ -52,13 +52,18 @@ static struct sk_buff_head skbtxq; /* enters with txlock held */ static int -tx(void) +tx(void) __must_hold(&txlock) { struct sk_buff *skb; + struct net_device *ifp; while ((skb = skb_dequeue(&skbtxq))) { spin_unlock_irq(&txlock); - dev_queue_xmit(skb); + ifp = skb->dev; + if (dev_queue_xmit(skb) == NET_XMIT_DROP && net_ratelimit()) + pr_warn("aoe: packet could not be sent on %s. %s\n", + ifp ? ifp->name : "netif", + "consider increasing tx_queue_len"); spin_lock_irq(&txlock); } return 0; @@ -119,8 +124,8 @@ aoenet_xmit(struct sk_buff_head *queue) } } -/* - * (1) len doesn't include the header by default. I want this. +/* + * (1) len doesn't include the header by default. I want this. */ static int aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt, struct net_device *orig_dev) diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index ca83f96756ad..6526157edafc 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -41,8 +41,9 @@ #include <linux/spinlock.h> #include <linux/compat.h> #include <linux/mutex.h> +#include <linux/bitmap.h> +#include <linux/io.h> #include <asm/uaccess.h> -#include <asm/io.h> #include <linux/dma-mapping.h> #include <linux/blkdev.h> @@ -978,8 +979,7 @@ static CommandList_struct *cmd_alloc(ctlr_info_t *h) i = find_first_zero_bit(h->cmd_pool_bits, h->nr_cmds); if (i == h->nr_cmds) return NULL; - } while (test_and_set_bit(i & (BITS_PER_LONG - 1), - h->cmd_pool_bits + (i / BITS_PER_LONG)) != 0); + } while (test_and_set_bit(i, h->cmd_pool_bits) != 0); c = h->cmd_pool + i; memset(c, 0, sizeof(CommandList_struct)); cmd_dma_handle = h->cmd_pool_dhandle + i * sizeof(CommandList_struct); @@ -1046,8 +1046,7 @@ static void cmd_free(ctlr_info_t *h, CommandList_struct *c) int i; i = c - h->cmd_pool; - clear_bit(i & (BITS_PER_LONG - 1), - h->cmd_pool_bits + (i / BITS_PER_LONG)); + clear_bit(i, h->cmd_pool_bits); h->nr_frees++; } @@ -4268,10 +4267,7 @@ static void __devinit cciss_find_board_params(ctlr_info_t *h) static inline bool CISS_signature_present(ctlr_info_t *h) { - if ((readb(&h->cfgtable->Signature[0]) != 'C') || - (readb(&h->cfgtable->Signature[1]) != 'I') || - (readb(&h->cfgtable->Signature[2]) != 'S') || - (readb(&h->cfgtable->Signature[3]) != 'S')) { + if (!check_signature(h->cfgtable->Signature, "CISS", 4)) { dev_warn(&h->pdev->dev, "not a valid CISS config table\n"); return false; } @@ -4812,8 +4808,7 @@ static __devinit int cciss_init_reset_devices(struct pci_dev *pdev) static __devinit int cciss_allocate_cmd_pool(ctlr_info_t *h) { - h->cmd_pool_bits = kmalloc( - DIV_ROUND_UP(h->nr_cmds, BITS_PER_LONG) * + h->cmd_pool_bits = kmalloc(BITS_TO_LONGS(h->nr_cmds) * sizeof(unsigned long), GFP_KERNEL); h->cmd_pool = pci_alloc_consistent(h->pdev, h->nr_cmds * sizeof(CommandList_struct), @@ -5068,9 +5063,7 @@ reinit_after_soft_reset: pci_set_drvdata(pdev, h); /* command and error info recs zeroed out before they are used */ - memset(h->cmd_pool_bits, 0, - DIV_ROUND_UP(h->nr_cmds, BITS_PER_LONG) - * sizeof(unsigned long)); + bitmap_zero(h->cmd_pool_bits, h->nr_cmds); h->num_luns = 0; h->highest_lun = -1; diff --git a/drivers/block/drbd/Kconfig b/drivers/block/drbd/Kconfig index df0983787390..7845bd6ee414 100644 --- a/drivers/block/drbd/Kconfig +++ b/drivers/block/drbd/Kconfig @@ -2,13 +2,14 @@ # DRBD device driver configuration # -comment "DRBD disabled because PROC_FS, INET or CONNECTOR not selected" - depends on PROC_FS='n' || INET='n' || CONNECTOR='n' +comment "DRBD disabled because PROC_FS or INET not selected" + depends on PROC_FS='n' || INET='n' config BLK_DEV_DRBD tristate "DRBD Distributed Replicated Block Device support" - depends on PROC_FS && INET && CONNECTOR + depends on PROC_FS && INET select LRU_CACHE + select LIBCRC32C default n help @@ -58,7 +59,8 @@ config DRBD_FAULT_INJECTION 32 data read 64 read ahead 128 kmalloc of bitmap - 256 allocation of EE (epoch_entries) + 256 allocation of peer_requests + 512 insert data corruption on receiving side fault_devs: bitmask of minor numbers fault_rate: frequency in percent diff --git a/drivers/block/drbd/Makefile b/drivers/block/drbd/Makefile index 0d3f337ff5ff..8b450338075e 100644 --- a/drivers/block/drbd/Makefile +++ b/drivers/block/drbd/Makefile @@ -1,5 +1,7 @@ drbd-y := drbd_bitmap.o drbd_proc.o drbd-y += drbd_worker.o drbd_receiver.o drbd_req.o drbd_actlog.o drbd-y += drbd_main.o drbd_strings.o drbd_nl.o +drbd-y += drbd_interval.o drbd_state.o +drbd-y += drbd_nla.o obj-$(CONFIG_BLK_DEV_DRBD) += drbd.o diff --git a/drivers/block/drbd/drbd_actlog.c b/drivers/block/drbd/drbd_actlog.c index 3fbef018ce55..92510f8ad013 100644 --- a/drivers/block/drbd/drbd_actlog.c +++ b/drivers/block/drbd/drbd_actlog.c @@ -24,21 +24,73 @@ */ #include <linux/slab.h> +#include <linux/crc32c.h> #include <linux/drbd.h> +#include <linux/drbd_limits.h> +#include <linux/dynamic_debug.h> #include "drbd_int.h" #include "drbd_wrappers.h" -/* We maintain a trivial checksum in our on disk activity log. - * With that we can ensure correct operation even when the storage - * device might do a partial (last) sector write while losing power. - */ -struct __packed al_transaction { - u32 magic; - u32 tr_number; - struct __packed { - u32 pos; - u32 extent; } updates[1 + AL_EXTENTS_PT]; - u32 xor_sum; + +enum al_transaction_types { + AL_TR_UPDATE = 0, + AL_TR_INITIALIZED = 0xffff +}; +/* all fields on disc in big endian */ +struct __packed al_transaction_on_disk { + /* don't we all like magic */ + __be32 magic; + + /* to identify the most recent transaction block + * in the on disk ring buffer */ + __be32 tr_number; + + /* checksum on the full 4k block, with this field set to 0. */ + __be32 crc32c; + + /* type of transaction, special transaction types like: + * purge-all, set-all-idle, set-all-active, ... to-be-defined + * see also enum al_transaction_types */ + __be16 transaction_type; + + /* we currently allow only a few thousand extents, + * so 16bit will be enough for the slot number. */ + + /* how many updates in this transaction */ + __be16 n_updates; + + /* maximum slot number, "al-extents" in drbd.conf speak. + * Having this in each transaction should make reconfiguration + * of that parameter easier. */ + __be16 context_size; + + /* slot number the context starts with */ + __be16 context_start_slot_nr; + + /* Some reserved bytes. Expected usage is a 64bit counter of + * sectors-written since device creation, and other data generation tag + * supporting usage */ + __be32 __reserved[4]; + + /* --- 36 byte used --- */ + + /* Reserve space for up to AL_UPDATES_PER_TRANSACTION changes + * in one transaction, then use the remaining byte in the 4k block for + * context information. "Flexible" number of updates per transaction + * does not help, as we have to account for the case when all update + * slots are used anyways, so it would only complicate code without + * additional benefit. + */ + __be16 update_slot_nr[AL_UPDATES_PER_TRANSACTION]; + + /* but the extent number is 32bit, which at an extent size of 4 MiB + * allows to cover device sizes of up to 2**54 Byte (16 PiB) */ + __be32 update_extent_nr[AL_UPDATES_PER_TRANSACTION]; + + /* --- 420 bytes used (36 + 64*6) --- */ + + /* 4096 - 420 = 3676 = 919 * 4 */ + __be32 context[AL_CONTEXT_PER_TRANSACTION]; }; struct update_odbm_work { @@ -48,22 +100,11 @@ struct update_odbm_work { struct update_al_work { struct drbd_work w; - struct lc_element *al_ext; struct completion event; - unsigned int enr; - /* if old_enr != LC_FREE, write corresponding bitmap sector, too */ - unsigned int old_enr; -}; - -struct drbd_atodb_wait { - atomic_t count; - struct completion io_done; - struct drbd_conf *mdev; - int error; + int err; }; - -int w_al_write_transaction(struct drbd_conf *, struct drbd_work *, int); +static int al_write_transaction(struct drbd_conf *mdev); void *drbd_md_get_buffer(struct drbd_conf *mdev) { @@ -82,22 +123,24 @@ void drbd_md_put_buffer(struct drbd_conf *mdev) wake_up(&mdev->misc_wait); } -static bool md_io_allowed(struct drbd_conf *mdev) -{ - enum drbd_disk_state ds = mdev->state.disk; - return ds >= D_NEGOTIATING || ds == D_ATTACHING; -} - -void wait_until_done_or_disk_failure(struct drbd_conf *mdev, struct drbd_backing_dev *bdev, +void wait_until_done_or_force_detached(struct drbd_conf *mdev, struct drbd_backing_dev *bdev, unsigned int *done) { - long dt = bdev->dc.disk_timeout * HZ / 10; + long dt; + + rcu_read_lock(); + dt = rcu_dereference(bdev->disk_conf)->disk_timeout; + rcu_read_unlock(); + dt = dt * HZ / 10; if (dt == 0) dt = MAX_SCHEDULE_TIMEOUT; - dt = wait_event_timeout(mdev->misc_wait, *done || !md_io_allowed(mdev), dt); - if (dt == 0) + dt = wait_event_timeout(mdev->misc_wait, + *done || test_bit(FORCE_DETACH, &mdev->flags), dt); + if (dt == 0) { dev_err(DEV, "meta-data IO operation timed out\n"); + drbd_chk_io_error(mdev, 1, DRBD_FORCE_DETACH); + } } static int _drbd_md_sync_page_io(struct drbd_conf *mdev, @@ -106,7 +149,7 @@ static int _drbd_md_sync_page_io(struct drbd_conf *mdev, int rw, int size) { struct bio *bio; - int ok; + int err; mdev->md_io.done = 0; mdev->md_io.error = -ENODEV; @@ -118,8 +161,8 @@ static int _drbd_md_sync_page_io(struct drbd_conf *mdev, bio = bio_alloc_drbd(GFP_NOIO); bio->bi_bdev = bdev->md_bdev; bio->bi_sector = sector; - ok = (bio_add_page(bio, page, size, 0) == size); - if (!ok) + err = -EIO; + if (bio_add_page(bio, page, size, 0) != size) goto out; bio->bi_private = &mdev->md_io; bio->bi_end_io = drbd_md_io_complete; @@ -127,7 +170,7 @@ static int _drbd_md_sync_page_io(struct drbd_conf *mdev, if (!get_ldev_if_state(mdev, D_ATTACHING)) { /* Corresponding put_ldev in drbd_md_io_complete() */ dev_err(DEV, "ASSERT FAILED: get_ldev_if_state() == 1 in _drbd_md_sync_page_io()\n"); - ok = 0; + err = -ENODEV; goto out; } @@ -137,86 +180,47 @@ static int _drbd_md_sync_page_io(struct drbd_conf *mdev, bio_endio(bio, -EIO); else submit_bio(rw, bio); - wait_until_done_or_disk_failure(mdev, bdev, &mdev->md_io.done); - ok = bio_flagged(bio, BIO_UPTODATE) && mdev->md_io.error == 0; + wait_until_done_or_force_detached(mdev, bdev, &mdev->md_io.done); + if (bio_flagged(bio, BIO_UPTODATE)) + err = mdev->md_io.error; out: bio_put(bio); - return ok; + return err; } int drbd_md_sync_page_io(struct drbd_conf *mdev, struct drbd_backing_dev *bdev, sector_t sector, int rw) { - int logical_block_size, mask, ok; - int offset = 0; + int err; struct page *iop = mdev->md_io_page; D_ASSERT(atomic_read(&mdev->md_io_in_use) == 1); BUG_ON(!bdev->md_bdev); - logical_block_size = bdev_logical_block_size(bdev->md_bdev); - if (logical_block_size == 0) - logical_block_size = MD_SECTOR_SIZE; - - /* in case logical_block_size != 512 [ s390 only? ] */ - if (logical_block_size != MD_SECTOR_SIZE) { - mask = (logical_block_size / MD_SECTOR_SIZE) - 1; - D_ASSERT(mask == 1 || mask == 3 || mask == 7); - D_ASSERT(logical_block_size == (mask+1) * MD_SECTOR_SIZE); - offset = sector & mask; - sector = sector & ~mask; - iop = mdev->md_io_tmpp; - - if (rw & WRITE) { - /* these are GFP_KERNEL pages, pre-allocated - * on device initialization */ - void *p = page_address(mdev->md_io_page); - void *hp = page_address(mdev->md_io_tmpp); - - ok = _drbd_md_sync_page_io(mdev, bdev, iop, sector, - READ, logical_block_size); - - if (unlikely(!ok)) { - dev_err(DEV, "drbd_md_sync_page_io(,%llus," - "READ [logical_block_size!=512]) failed!\n", - (unsigned long long)sector); - return 0; - } - - memcpy(hp + offset*MD_SECTOR_SIZE, p, MD_SECTOR_SIZE); - } - } + dev_dbg(DEV, "meta_data io: %s [%d]:%s(,%llus,%s)\n", + current->comm, current->pid, __func__, + (unsigned long long)sector, (rw & WRITE) ? "WRITE" : "READ"); if (sector < drbd_md_first_sector(bdev) || - sector > drbd_md_last_sector(bdev)) + sector + 7 > drbd_md_last_sector(bdev)) dev_alert(DEV, "%s [%d]:%s(,%llus,%s) out of range md access!\n", current->comm, current->pid, __func__, (unsigned long long)sector, (rw & WRITE) ? "WRITE" : "READ"); - ok = _drbd_md_sync_page_io(mdev, bdev, iop, sector, rw, logical_block_size); - if (unlikely(!ok)) { - dev_err(DEV, "drbd_md_sync_page_io(,%llus,%s) failed!\n", - (unsigned long long)sector, (rw & WRITE) ? "WRITE" : "READ"); - return 0; - } - - if (logical_block_size != MD_SECTOR_SIZE && !(rw & WRITE)) { - void *p = page_address(mdev->md_io_page); - void *hp = page_address(mdev->md_io_tmpp); - - memcpy(p, hp + offset*MD_SECTOR_SIZE, MD_SECTOR_SIZE); + err = _drbd_md_sync_page_io(mdev, bdev, iop, sector, rw, MD_BLOCK_SIZE); + if (err) { + dev_err(DEV, "drbd_md_sync_page_io(,%llus,%s) failed with error %d\n", + (unsigned long long)sector, (rw & WRITE) ? "WRITE" : "READ", err); } - - return ok; + return err; } static struct lc_element *_al_get(struct drbd_conf *mdev, unsigned int enr) { struct lc_element *al_ext; struct lc_element *tmp; - unsigned long al_flags = 0; int wake; spin_lock_irq(&mdev->al_lock); @@ -231,76 +235,92 @@ static struct lc_element *_al_get(struct drbd_conf *mdev, unsigned int enr) return NULL; } } - al_ext = lc_get(mdev->act_log, enr); - al_flags = mdev->act_log->flags; + al_ext = lc_get(mdev->act_log, enr); spin_unlock_irq(&mdev->al_lock); - - /* - if (!al_ext) { - if (al_flags & LC_STARVING) - dev_warn(DEV, "Have to wait for LRU element (AL too small?)\n"); - if (al_flags & LC_DIRTY) - dev_warn(DEV, "Ongoing AL update (AL device too slow?)\n"); - } - */ - return al_ext; } -void drbd_al_begin_io(struct drbd_conf *mdev, sector_t sector) +void drbd_al_begin_io(struct drbd_conf *mdev, struct drbd_interval *i) { - unsigned int enr = (sector >> (AL_EXTENT_SHIFT-9)); - struct lc_element *al_ext; - struct update_al_work al_work; + /* for bios crossing activity log extent boundaries, + * we may need to activate two extents in one go */ + unsigned first = i->sector >> (AL_EXTENT_SHIFT-9); + unsigned last = i->size == 0 ? first : (i->sector + (i->size >> 9) - 1) >> (AL_EXTENT_SHIFT-9); + unsigned enr; + bool locked = false; + + D_ASSERT(first <= last); D_ASSERT(atomic_read(&mdev->local_cnt) > 0); - wait_event(mdev->al_wait, (al_ext = _al_get(mdev, enr))); + for (enr = first; enr <= last; enr++) + wait_event(mdev->al_wait, _al_get(mdev, enr) != NULL); + + /* Serialize multiple transactions. + * This uses test_and_set_bit, memory barrier is implicit. + */ + wait_event(mdev->al_wait, + mdev->act_log->pending_changes == 0 || + (locked = lc_try_lock_for_transaction(mdev->act_log))); - if (al_ext->lc_number != enr) { + if (locked) { /* drbd_al_write_transaction(mdev,al_ext,enr); * recurses into generic_make_request(), which * disallows recursion, bios being serialized on the * current->bio_tail list now. * we have to delegate updates to the activity log * to the worker thread. */ - init_completion(&al_work.event); - al_work.al_ext = al_ext; - al_work.enr = enr; - al_work.old_enr = al_ext->lc_number; - al_work.w.cb = w_al_write_transaction; - drbd_queue_work_front(&mdev->data.work, &al_work.w); - wait_for_completion(&al_work.event); - - mdev->al_writ_cnt++; - - spin_lock_irq(&mdev->al_lock); - lc_changed(mdev->act_log, al_ext); - spin_unlock_irq(&mdev->al_lock); + + /* Double check: it may have been committed by someone else, + * while we have been waiting for the lock. */ + if (mdev->act_log->pending_changes) { + bool write_al_updates; + + rcu_read_lock(); + write_al_updates = rcu_dereference(mdev->ldev->disk_conf)->al_updates; + rcu_read_unlock(); + + if (write_al_updates) { + al_write_transaction(mdev); + mdev->al_writ_cnt++; + } + + spin_lock_irq(&mdev->al_lock); + /* FIXME + if (err) + we need an "lc_cancel" here; + */ + lc_committed(mdev->act_log); + spin_unlock_irq(&mdev->al_lock); + } + lc_unlock(mdev->act_log); wake_up(&mdev->al_wait); } } -void drbd_al_complete_io(struct drbd_conf *mdev, sector_t sector) +void drbd_al_complete_io(struct drbd_conf *mdev, struct drbd_interval *i) { - unsigned int enr = (sector >> (AL_EXTENT_SHIFT-9)); + /* for bios crossing activity log extent boundaries, + * we may need to activate two extents in one go */ + unsigned first = i->sector >> (AL_EXTENT_SHIFT-9); + unsigned last = i->size == 0 ? first : (i->sector + (i->size >> 9) - 1) >> (AL_EXTENT_SHIFT-9); + unsigned enr; struct lc_element *extent; unsigned long flags; + D_ASSERT(first <= last); spin_lock_irqsave(&mdev->al_lock, flags); - extent = lc_find(mdev->act_log, enr); - - if (!extent) { - spin_unlock_irqrestore(&mdev->al_lock, flags); - dev_err(DEV, "al_complete_io() called on inactive extent %u\n", enr); - return; + for (enr = first; enr <= last; enr++) { + extent = lc_find(mdev->act_log, enr); + if (!extent) { + dev_err(DEV, "al_complete_io() called on inactive extent %u\n", enr); + continue; + } + lc_put(mdev->act_log, extent); } - - if (lc_put(mdev->act_log, extent) == 0) - wake_up(&mdev->al_wait); - spin_unlock_irqrestore(&mdev->al_lock, flags); + wake_up(&mdev->al_wait); } #if (PAGE_SHIFT + 3) < (AL_EXTENT_SHIFT - BM_BLOCK_SHIFT) @@ -326,296 +346,148 @@ static unsigned int rs_extent_to_bm_page(unsigned int rs_enr) return rs_enr >> /* bit to page */ ((PAGE_SHIFT + 3) - - /* al extent number to bit */ + /* resync extent number to bit */ (BM_EXT_SHIFT - BM_BLOCK_SHIFT)); } -int -w_al_write_transaction(struct drbd_conf *mdev, struct drbd_work *w, int unused) +static int +_al_write_transaction(struct drbd_conf *mdev) { - struct update_al_work *aw = container_of(w, struct update_al_work, w); - struct lc_element *updated = aw->al_ext; - const unsigned int new_enr = aw->enr; - const unsigned int evicted = aw->old_enr; - struct al_transaction *buffer; + struct al_transaction_on_disk *buffer; + struct lc_element *e; sector_t sector; - int i, n, mx; - unsigned int extent_nr; - u32 xor_sum = 0; + int i, mx; + unsigned extent_nr; + unsigned crc = 0; + int err = 0; if (!get_ldev(mdev)) { - dev_err(DEV, - "disk is %s, cannot start al transaction (-%d +%d)\n", - drbd_disk_str(mdev->state.disk), evicted, new_enr); - complete(&((struct update_al_work *)w)->event); - return 1; + dev_err(DEV, "disk is %s, cannot start al transaction\n", + drbd_disk_str(mdev->state.disk)); + return -EIO; } - /* do we have to do a bitmap write, first? - * TODO reduce maximum latency: - * submit both bios, then wait for both, - * instead of doing two synchronous sector writes. - * For now, we must not write the transaction, - * if we cannot write out the bitmap of the evicted extent. */ - if (mdev->state.conn < C_CONNECTED && evicted != LC_FREE) - drbd_bm_write_page(mdev, al_extent_to_bm_page(evicted)); /* The bitmap write may have failed, causing a state change. */ if (mdev->state.disk < D_INCONSISTENT) { dev_err(DEV, - "disk is %s, cannot write al transaction (-%d +%d)\n", - drbd_disk_str(mdev->state.disk), evicted, new_enr); - complete(&((struct update_al_work *)w)->event); + "disk is %s, cannot write al transaction\n", + drbd_disk_str(mdev->state.disk)); put_ldev(mdev); - return 1; + return -EIO; } buffer = drbd_md_get_buffer(mdev); /* protects md_io_buffer, al_tr_cycle, ... */ if (!buffer) { dev_err(DEV, "disk failed while waiting for md_io buffer\n"); - complete(&((struct update_al_work *)w)->event); put_ldev(mdev); - return 1; + return -ENODEV; } - buffer->magic = __constant_cpu_to_be32(DRBD_MAGIC); + memset(buffer, 0, sizeof(*buffer)); + buffer->magic = cpu_to_be32(DRBD_AL_MAGIC); buffer->tr_number = cpu_to_be32(mdev->al_tr_number); - n = lc_index_of(mdev->act_log, updated); + i = 0; + + /* Even though no one can start to change this list + * once we set the LC_LOCKED -- from drbd_al_begin_io(), + * lc_try_lock_for_transaction() --, someone may still + * be in the process of changing it. */ + spin_lock_irq(&mdev->al_lock); + list_for_each_entry(e, &mdev->act_log->to_be_changed, list) { + if (i == AL_UPDATES_PER_TRANSACTION) { + i++; + break; + } + buffer->update_slot_nr[i] = cpu_to_be16(e->lc_index); + buffer->update_extent_nr[i] = cpu_to_be32(e->lc_new_number); + if (e->lc_number != LC_FREE) + drbd_bm_mark_for_writeout(mdev, + al_extent_to_bm_page(e->lc_number)); + i++; + } + spin_unlock_irq(&mdev->al_lock); + BUG_ON(i > AL_UPDATES_PER_TRANSACTION); - buffer->updates[0].pos = cpu_to_be32(n); - buffer->updates[0].extent = cpu_to_be32(new_enr); + buffer->n_updates = cpu_to_be16(i); + for ( ; i < AL_UPDATES_PER_TRANSACTION; i++) { + buffer->update_slot_nr[i] = cpu_to_be16(-1); + buffer->update_extent_nr[i] = cpu_to_be32(LC_FREE); + } - xor_sum ^= new_enr; + buffer->context_size = cpu_to_be16(mdev->act_log->nr_elements); + buffer->context_start_slot_nr = cpu_to_be16(mdev->al_tr_cycle); - mx = min_t(int, AL_EXTENTS_PT, + mx = min_t(int, AL_CONTEXT_PER_TRANSACTION, mdev->act_log->nr_elements - mdev->al_tr_cycle); for (i = 0; i < mx; i++) { unsigned idx = mdev->al_tr_cycle + i; extent_nr = lc_element_by_index(mdev->act_log, idx)->lc_number; - buffer->updates[i+1].pos = cpu_to_be32(idx); - buffer->updates[i+1].extent = cpu_to_be32(extent_nr); - xor_sum ^= extent_nr; - } - for (; i < AL_EXTENTS_PT; i++) { - buffer->updates[i+1].pos = __constant_cpu_to_be32(-1); - buffer->updates[i+1].extent = __constant_cpu_to_be32(LC_FREE); - xor_sum ^= LC_FREE; + buffer->context[i] = cpu_to_be32(extent_nr); } - mdev->al_tr_cycle += AL_EXTENTS_PT; + for (; i < AL_CONTEXT_PER_TRANSACTION; i++) + buffer->context[i] = cpu_to_be32(LC_FREE); + + mdev->al_tr_cycle += AL_CONTEXT_PER_TRANSACTION; if (mdev->al_tr_cycle >= mdev->act_log->nr_elements) mdev->al_tr_cycle = 0; - buffer->xor_sum = cpu_to_be32(xor_sum); - sector = mdev->ldev->md.md_offset - + mdev->ldev->md.al_offset + mdev->al_tr_pos; - - if (!drbd_md_sync_page_io(mdev, mdev->ldev, sector, WRITE)) - drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR); + + mdev->ldev->md.al_offset + + mdev->al_tr_pos * (MD_BLOCK_SIZE>>9); - if (++mdev->al_tr_pos > - div_ceil(mdev->act_log->nr_elements, AL_EXTENTS_PT)) - mdev->al_tr_pos = 0; + crc = crc32c(0, buffer, 4096); + buffer->crc32c = cpu_to_be32(crc); - D_ASSERT(mdev->al_tr_pos < MD_AL_MAX_SIZE); - mdev->al_tr_number++; + if (drbd_bm_write_hinted(mdev)) + err = -EIO; + /* drbd_chk_io_error done already */ + else if (drbd_md_sync_page_io(mdev, mdev->ldev, sector, WRITE)) { + err = -EIO; + drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR); + } else { + /* advance ringbuffer position and transaction counter */ + mdev->al_tr_pos = (mdev->al_tr_pos + 1) % (MD_AL_SECTORS*512/MD_BLOCK_SIZE); + mdev->al_tr_number++; + } drbd_md_put_buffer(mdev); - - complete(&((struct update_al_work *)w)->event); put_ldev(mdev); - return 1; + return err; } -/** - * drbd_al_read_tr() - Read a single transaction from the on disk activity log - * @mdev: DRBD device. - * @bdev: Block device to read form. - * @b: pointer to an al_transaction. - * @index: On disk slot of the transaction to read. - * - * Returns -1 on IO error, 0 on checksum error and 1 upon success. - */ -static int drbd_al_read_tr(struct drbd_conf *mdev, - struct drbd_backing_dev *bdev, - struct al_transaction *b, - int index) -{ - sector_t sector; - int rv, i; - u32 xor_sum = 0; - - sector = bdev->md.md_offset + bdev->md.al_offset + index; - - /* Dont process error normally, - * as this is done before disk is attached! */ - if (!drbd_md_sync_page_io(mdev, bdev, sector, READ)) - return -1; - - rv = (be32_to_cpu(b->magic) == DRBD_MAGIC); - - for (i = 0; i < AL_EXTENTS_PT + 1; i++) - xor_sum ^= be32_to_cpu(b->updates[i].extent); - rv &= (xor_sum == be32_to_cpu(b->xor_sum)); - return rv; -} - -/** - * drbd_al_read_log() - Restores the activity log from its on disk representation. - * @mdev: DRBD device. - * @bdev: Block device to read form. - * - * Returns 1 on success, returns 0 when reading the log failed due to IO errors. - */ -int drbd_al_read_log(struct drbd_conf *mdev, struct drbd_backing_dev *bdev) +static int w_al_write_transaction(struct drbd_work *w, int unused) { - struct al_transaction *buffer; - int i; - int rv; - int mx; - int active_extents = 0; - int transactions = 0; - int found_valid = 0; - int from = 0; - int to = 0; - u32 from_tnr = 0; - u32 to_tnr = 0; - u32 cnr; - - mx = div_ceil(mdev->act_log->nr_elements, AL_EXTENTS_PT); - - /* lock out all other meta data io for now, - * and make sure the page is mapped. - */ - buffer = drbd_md_get_buffer(mdev); - if (!buffer) - return 0; - - /* Find the valid transaction in the log */ - for (i = 0; i <= mx; i++) { - rv = drbd_al_read_tr(mdev, bdev, buffer, i); - if (rv == 0) - continue; - if (rv == -1) { - drbd_md_put_buffer(mdev); - return 0; - } - cnr = be32_to_cpu(buffer->tr_number); - - if (++found_valid == 1) { - from = i; - to = i; - from_tnr = cnr; - to_tnr = cnr; - continue; - } - if ((int)cnr - (int)from_tnr < 0) { - D_ASSERT(from_tnr - cnr + i - from == mx+1); - from = i; - from_tnr = cnr; - } - if ((int)cnr - (int)to_tnr > 0) { - D_ASSERT(cnr - to_tnr == i - to); - to = i; - to_tnr = cnr; - } - } - - if (!found_valid) { - dev_warn(DEV, "No usable activity log found.\n"); - drbd_md_put_buffer(mdev); - return 1; - } - - /* Read the valid transactions. - * dev_info(DEV, "Reading from %d to %d.\n",from,to); */ - i = from; - while (1) { - int j, pos; - unsigned int extent_nr; - unsigned int trn; - - rv = drbd_al_read_tr(mdev, bdev, buffer, i); - ERR_IF(rv == 0) goto cancel; - if (rv == -1) { - drbd_md_put_buffer(mdev); - return 0; - } - - trn = be32_to_cpu(buffer->tr_number); - - spin_lock_irq(&mdev->al_lock); - - /* This loop runs backwards because in the cyclic - elements there might be an old version of the - updated element (in slot 0). So the element in slot 0 - can overwrite old versions. */ - for (j = AL_EXTENTS_PT; j >= 0; j--) { - pos = be32_to_cpu(buffer->updates[j].pos); - extent_nr = be32_to_cpu(buffer->updates[j].extent); - - if (extent_nr == LC_FREE) - continue; - - lc_set(mdev->act_log, extent_nr, pos); - active_extents++; - } - spin_unlock_irq(&mdev->al_lock); - - transactions++; - -cancel: - if (i == to) - break; - i++; - if (i > mx) - i = 0; - } - - mdev->al_tr_number = to_tnr+1; - mdev->al_tr_pos = to; - if (++mdev->al_tr_pos > - div_ceil(mdev->act_log->nr_elements, AL_EXTENTS_PT)) - mdev->al_tr_pos = 0; - - /* ok, we are done with it */ - drbd_md_put_buffer(mdev); + struct update_al_work *aw = container_of(w, struct update_al_work, w); + struct drbd_conf *mdev = w->mdev; + int err; - dev_info(DEV, "Found %d transactions (%d active extents) in activity log.\n", - transactions, active_extents); + err = _al_write_transaction(mdev); + aw->err = err; + complete(&aw->event); - return 1; + return err != -EIO ? err : 0; } -/** - * drbd_al_apply_to_bm() - Sets the bitmap to diry(1) where covered ba active AL extents - * @mdev: DRBD device. - */ -void drbd_al_apply_to_bm(struct drbd_conf *mdev) +/* Calls from worker context (see w_restart_disk_io()) need to write the + transaction directly. Others came through generic_make_request(), + those need to delegate it to the worker. */ +static int al_write_transaction(struct drbd_conf *mdev) { - unsigned int enr; - unsigned long add = 0; - char ppb[10]; - int i, tmp; - - wait_event(mdev->al_wait, lc_try_lock(mdev->act_log)); + struct update_al_work al_work; - for (i = 0; i < mdev->act_log->nr_elements; i++) { - enr = lc_element_by_index(mdev->act_log, i)->lc_number; - if (enr == LC_FREE) - continue; - tmp = drbd_bm_ALe_set_all(mdev, enr); - dynamic_dev_dbg(DEV, "AL: set %d bits in extent %u\n", tmp, enr); - add += tmp; - } + if (current == mdev->tconn->worker.task) + return _al_write_transaction(mdev); - lc_unlock(mdev->act_log); - wake_up(&mdev->al_wait); + init_completion(&al_work.event); + al_work.w.cb = w_al_write_transaction; + al_work.w.mdev = mdev; + drbd_queue_work_front(&mdev->tconn->sender_work, &al_work.w); + wait_for_completion(&al_work.event); - dev_info(DEV, "Marked additional %s as out-of-sync based on AL.\n", - ppsize(ppb, Bit2KB(add))); + return al_work.err; } static int _try_lc_del(struct drbd_conf *mdev, struct lc_element *al_ext) @@ -645,7 +517,7 @@ void drbd_al_shrink(struct drbd_conf *mdev) struct lc_element *al_ext; int i; - D_ASSERT(test_bit(__LC_DIRTY, &mdev->act_log->flags)); + D_ASSERT(test_bit(__LC_LOCKED, &mdev->act_log->flags)); for (i = 0; i < mdev->act_log->nr_elements; i++) { al_ext = lc_element_by_index(mdev->act_log, i); @@ -657,15 +529,17 @@ void drbd_al_shrink(struct drbd_conf *mdev) wake_up(&mdev->al_wait); } -static int w_update_odbm(struct drbd_conf *mdev, struct drbd_work *w, int unused) +static int w_update_odbm(struct drbd_work *w, int unused) { struct update_odbm_work *udw = container_of(w, struct update_odbm_work, w); + struct drbd_conf *mdev = w->mdev; + struct sib_info sib = { .sib_reason = SIB_SYNC_PROGRESS, }; if (!get_ldev(mdev)) { if (__ratelimit(&drbd_ratelimit_state)) dev_warn(DEV, "Can not update on disk bitmap, local IO disabled.\n"); kfree(udw); - return 1; + return 0; } drbd_bm_write_page(mdev, rs_extent_to_bm_page(udw->enr)); @@ -683,9 +557,9 @@ static int w_update_odbm(struct drbd_conf *mdev, struct drbd_work *w, int unused break; } } - drbd_bcast_sync_progress(mdev); + drbd_bcast_event(mdev, &sib); - return 1; + return 0; } @@ -755,7 +629,9 @@ static void drbd_try_clear_on_disk_bm(struct drbd_conf *mdev, sector_t sector, } ext->rs_left = rs_left; ext->rs_failed = success ? 0 : count; - lc_changed(mdev->resync, &ext->lce); + /* we don't keep a persistent log of the resync lru, + * we can commit any change right away. */ + lc_committed(mdev->resync); } lc_put(mdev->resync, &ext->lce); /* no race, we are within the al_lock! */ @@ -767,7 +643,8 @@ static void drbd_try_clear_on_disk_bm(struct drbd_conf *mdev, sector_t sector, if (udw) { udw->enr = ext->lce.lc_number; udw->w.cb = w_update_odbm; - drbd_queue_work_front(&mdev->data.work, &udw->w); + udw->w.mdev = mdev; + drbd_queue_work_front(&mdev->tconn->sender_work, &udw->w); } else { dev_warn(DEV, "Could not kmalloc an udw\n"); } @@ -813,16 +690,22 @@ void __drbd_set_in_sync(struct drbd_conf *mdev, sector_t sector, int size, int wake_up = 0; unsigned long flags; - if (size <= 0 || (size & 0x1ff) != 0 || size > DRBD_MAX_BIO_SIZE) { + if (size <= 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_BIO_SIZE) { dev_err(DEV, "drbd_set_in_sync: sector=%llus size=%d nonsense!\n", (unsigned long long)sector, size); return; } + + if (!get_ldev(mdev)) + return; /* no disk, no metadata, no bitmap to clear bits in */ + nr_sectors = drbd_get_capacity(mdev->this_bdev); esector = sector + (size >> 9) - 1; - ERR_IF(sector >= nr_sectors) return; - ERR_IF(esector >= nr_sectors) esector = (nr_sectors-1); + if (!expect(sector < nr_sectors)) + goto out; + if (!expect(esector < nr_sectors)) + esector = nr_sectors - 1; lbnr = BM_SECT_TO_BIT(nr_sectors-1); @@ -830,7 +713,7 @@ void __drbd_set_in_sync(struct drbd_conf *mdev, sector_t sector, int size, * round up start sector, round down end sector. we make sure we only * clear full, aligned, BM_BLOCK_SIZE (4K) blocks */ if (unlikely(esector < BM_SECT_PER_BIT-1)) - return; + goto out; if (unlikely(esector == (nr_sectors-1))) ebnr = lbnr; else @@ -838,14 +721,14 @@ void __drbd_set_in_sync(struct drbd_conf *mdev, sector_t sector, int size, sbnr = BM_SECT_TO_BIT(sector + BM_SECT_PER_BIT-1); if (sbnr > ebnr) - return; + goto out; /* * ok, (capacity & 7) != 0 sometimes, but who cares... * we count rs_{total,left} in bits, not sectors. */ count = drbd_bm_clear_bits(mdev, sbnr, ebnr); - if (count && get_ldev(mdev)) { + if (count) { drbd_advance_rs_marks(mdev, drbd_bm_total_weight(mdev)); spin_lock_irqsave(&mdev->al_lock, flags); drbd_try_clear_on_disk_bm(mdev, sector, count, true); @@ -854,8 +737,9 @@ void __drbd_set_in_sync(struct drbd_conf *mdev, sector_t sector, int size, /* just wake_up unconditional now, various lc_chaged(), * lc_put() in drbd_try_clear_on_disk_bm(). */ wake_up = 1; - put_ldev(mdev); } +out: + put_ldev(mdev); if (wake_up) wake_up(&mdev->al_wait); } @@ -871,7 +755,7 @@ void __drbd_set_in_sync(struct drbd_conf *mdev, sector_t sector, int size, int __drbd_set_out_of_sync(struct drbd_conf *mdev, sector_t sector, int size, const char *file, const unsigned int line) { - unsigned long sbnr, ebnr, lbnr, flags; + unsigned long sbnr, ebnr, flags; sector_t esector, nr_sectors; unsigned int enr, count = 0; struct lc_element *e; @@ -880,7 +764,7 @@ int __drbd_set_out_of_sync(struct drbd_conf *mdev, sector_t sector, int size, if (size == 0) return 0; - if (size < 0 || (size & 0x1ff) != 0 || size > DRBD_MAX_BIO_SIZE) { + if (size < 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_BIO_SIZE) { dev_err(DEV, "sector: %llus, size: %d\n", (unsigned long long)sector, size); return 0; @@ -892,12 +776,10 @@ int __drbd_set_out_of_sync(struct drbd_conf *mdev, sector_t sector, int size, nr_sectors = drbd_get_capacity(mdev->this_bdev); esector = sector + (size >> 9) - 1; - ERR_IF(sector >= nr_sectors) + if (!expect(sector < nr_sectors)) goto out; - ERR_IF(esector >= nr_sectors) - esector = (nr_sectors-1); - - lbnr = BM_SECT_TO_BIT(nr_sectors-1); + if (!expect(esector < nr_sectors)) + esector = nr_sectors - 1; /* we set it out of sync, * we do not need to round anything here */ @@ -940,7 +822,7 @@ struct bm_extent *_bme_get(struct drbd_conf *mdev, unsigned int enr) if (bm_ext->lce.lc_number != enr) { bm_ext->rs_left = drbd_bm_e_weight(mdev, enr); bm_ext->rs_failed = 0; - lc_changed(mdev->resync, &bm_ext->lce); + lc_committed(mdev->resync); wakeup = 1; } if (bm_ext->lce.refcnt == 1) @@ -956,7 +838,7 @@ struct bm_extent *_bme_get(struct drbd_conf *mdev, unsigned int enr) if (rs_flags & LC_STARVING) dev_warn(DEV, "Have to wait for element" " (resync LRU too small?)\n"); - BUG_ON(rs_flags & LC_DIRTY); + BUG_ON(rs_flags & LC_LOCKED); } return bm_ext; @@ -964,26 +846,12 @@ struct bm_extent *_bme_get(struct drbd_conf *mdev, unsigned int enr) static int _is_in_al(struct drbd_conf *mdev, unsigned int enr) { - struct lc_element *al_ext; - int rv = 0; + int rv; spin_lock_irq(&mdev->al_lock); - if (unlikely(enr == mdev->act_log->new_number)) - rv = 1; - else { - al_ext = lc_find(mdev->act_log, enr); - if (al_ext) { - if (al_ext->refcnt) - rv = 1; - } - } + rv = lc_is_used(mdev->act_log, enr); spin_unlock_irq(&mdev->al_lock); - /* - if (unlikely(rv)) { - dev_info(DEV, "Delaying sync read until app's write is done\n"); - } - */ return rv; } @@ -1113,13 +981,13 @@ int drbd_try_rs_begin_io(struct drbd_conf *mdev, sector_t sector) if (rs_flags & LC_STARVING) dev_warn(DEV, "Have to wait for element" " (resync LRU too small?)\n"); - BUG_ON(rs_flags & LC_DIRTY); + BUG_ON(rs_flags & LC_LOCKED); goto try_again; } if (bm_ext->lce.lc_number != enr) { bm_ext->rs_left = drbd_bm_e_weight(mdev, enr); bm_ext->rs_failed = 0; - lc_changed(mdev->resync, &bm_ext->lce); + lc_committed(mdev->resync); wake_up(&mdev->al_wait); D_ASSERT(test_bit(BME_LOCKED, &bm_ext->flags) == 0); } @@ -1130,8 +998,6 @@ int drbd_try_rs_begin_io(struct drbd_conf *mdev, sector_t sector) } check_al: for (i = 0; i < AL_EXT_PER_BM_SECT; i++) { - if (unlikely(al_enr+i == mdev->act_log->new_number)) - goto try_again; if (lc_is_used(mdev->act_log, al_enr+i)) goto try_again; } @@ -1266,7 +1132,7 @@ void drbd_rs_failed_io(struct drbd_conf *mdev, sector_t sector, int size) sector_t esector, nr_sectors; int wake_up = 0; - if (size <= 0 || (size & 0x1ff) != 0 || size > DRBD_MAX_BIO_SIZE) { + if (size <= 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_BIO_SIZE) { dev_err(DEV, "drbd_rs_failed_io: sector=%llus size=%d nonsense!\n", (unsigned long long)sector, size); return; @@ -1274,8 +1140,10 @@ void drbd_rs_failed_io(struct drbd_conf *mdev, sector_t sector, int size) nr_sectors = drbd_get_capacity(mdev->this_bdev); esector = sector + (size >> 9) - 1; - ERR_IF(sector >= nr_sectors) return; - ERR_IF(esector >= nr_sectors) esector = (nr_sectors-1); + if (!expect(sector < nr_sectors)) + return; + if (!expect(esector < nr_sectors)) + esector = nr_sectors - 1; lbnr = BM_SECT_TO_BIT(nr_sectors-1); diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c index d84566496746..8dc29502dc08 100644 --- a/drivers/block/drbd/drbd_bitmap.c +++ b/drivers/block/drbd/drbd_bitmap.c @@ -119,13 +119,9 @@ static void __bm_print_lock_info(struct drbd_conf *mdev, const char *func) if (!__ratelimit(&drbd_ratelimit_state)) return; dev_err(DEV, "FIXME %s in %s, bitmap locked for '%s' by %s\n", - current == mdev->receiver.task ? "receiver" : - current == mdev->asender.task ? "asender" : - current == mdev->worker.task ? "worker" : current->comm, - func, b->bm_why ?: "?", - b->bm_task == mdev->receiver.task ? "receiver" : - b->bm_task == mdev->asender.task ? "asender" : - b->bm_task == mdev->worker.task ? "worker" : "?"); + drbd_task_to_thread_name(mdev->tconn, current), + func, b->bm_why ?: "?", + drbd_task_to_thread_name(mdev->tconn, b->bm_task)); } void drbd_bm_lock(struct drbd_conf *mdev, char *why, enum bm_flag flags) @@ -142,13 +138,9 @@ void drbd_bm_lock(struct drbd_conf *mdev, char *why, enum bm_flag flags) if (trylock_failed) { dev_warn(DEV, "%s going to '%s' but bitmap already locked for '%s' by %s\n", - current == mdev->receiver.task ? "receiver" : - current == mdev->asender.task ? "asender" : - current == mdev->worker.task ? "worker" : current->comm, - why, b->bm_why ?: "?", - b->bm_task == mdev->receiver.task ? "receiver" : - b->bm_task == mdev->asender.task ? "asender" : - b->bm_task == mdev->worker.task ? "worker" : "?"); + drbd_task_to_thread_name(mdev->tconn, current), + why, b->bm_why ?: "?", + drbd_task_to_thread_name(mdev->tconn, b->bm_task)); mutex_lock(&b->bm_change); } if (BM_LOCKED_MASK & b->bm_flags) @@ -196,6 +188,9 @@ void drbd_bm_unlock(struct drbd_conf *mdev) /* to mark for lazy writeout once syncer cleared all clearable bits, * we if bits have been cleared since last IO. */ #define BM_PAGE_LAZY_WRITEOUT 28 +/* pages marked with this "HINT" will be considered for writeout + * on activity log transactions */ +#define BM_PAGE_HINT_WRITEOUT 27 /* store_page_idx uses non-atomic assignment. It is only used directly after * allocating the page. All other bm_set_page_* and bm_clear_page_* need to @@ -227,8 +222,7 @@ static void bm_page_unlock_io(struct drbd_conf *mdev, int page_nr) { struct drbd_bitmap *b = mdev->bitmap; void *addr = &page_private(b->bm_pages[page_nr]); - clear_bit(BM_PAGE_IO_LOCK, addr); - smp_mb__after_clear_bit(); + clear_bit_unlock(BM_PAGE_IO_LOCK, addr); wake_up(&mdev->bitmap->bm_io_wait); } @@ -246,6 +240,27 @@ static void bm_set_page_need_writeout(struct page *page) set_bit(BM_PAGE_NEED_WRITEOUT, &page_private(page)); } +/** + * drbd_bm_mark_for_writeout() - mark a page with a "hint" to be considered for writeout + * @mdev: DRBD device. + * @page_nr: the bitmap page to mark with the "hint" flag + * + * From within an activity log transaction, we mark a few pages with these + * hints, then call drbd_bm_write_hinted(), which will only write out changed + * pages which are flagged with this mark. + */ +void drbd_bm_mark_for_writeout(struct drbd_conf *mdev, int page_nr) +{ + struct page *page; + if (page_nr >= mdev->bitmap->bm_number_of_pages) { + dev_warn(DEV, "BAD: page_nr: %u, number_of_pages: %u\n", + page_nr, (int)mdev->bitmap->bm_number_of_pages); + return; + } + page = mdev->bitmap->bm_pages[page_nr]; + set_bit(BM_PAGE_HINT_WRITEOUT, &page_private(page)); +} + static int bm_test_page_unchanged(struct page *page) { volatile const unsigned long *addr = &page_private(page); @@ -373,14 +388,16 @@ static struct page **bm_realloc_pages(struct drbd_bitmap *b, unsigned long want) return old_pages; /* Trying kmalloc first, falling back to vmalloc. - * GFP_KERNEL is ok, as this is done when a lower level disk is - * "attached" to the drbd. Context is receiver thread or cqueue - * thread. As we have no disk yet, we are not in the IO path, - * not even the IO path of the peer. */ + * GFP_NOIO, as this is called while drbd IO is "suspended", + * and during resize or attach on diskless Primary, + * we must not block on IO to ourselves. + * Context is receiver thread or dmsetup. */ bytes = sizeof(struct page *)*want; - new_pages = kzalloc(bytes, GFP_KERNEL); + new_pages = kzalloc(bytes, GFP_NOIO); if (!new_pages) { - new_pages = vzalloc(bytes); + new_pages = __vmalloc(bytes, + GFP_NOIO | __GFP_HIGHMEM | __GFP_ZERO, + PAGE_KERNEL); if (!new_pages) return NULL; vmalloced = 1; @@ -390,7 +407,7 @@ static struct page **bm_realloc_pages(struct drbd_bitmap *b, unsigned long want) for (i = 0; i < have; i++) new_pages[i] = old_pages[i]; for (; i < want; i++) { - page = alloc_page(GFP_HIGHUSER); + page = alloc_page(GFP_NOIO | __GFP_HIGHMEM); if (!page) { bm_free_pages(new_pages + have, i - have); bm_vk_free(new_pages, vmalloced); @@ -439,7 +456,8 @@ int drbd_bm_init(struct drbd_conf *mdev) sector_t drbd_bm_capacity(struct drbd_conf *mdev) { - ERR_IF(!mdev->bitmap) return 0; + if (!expect(mdev->bitmap)) + return 0; return mdev->bitmap->bm_dev_capacity; } @@ -447,7 +465,8 @@ sector_t drbd_bm_capacity(struct drbd_conf *mdev) */ void drbd_bm_cleanup(struct drbd_conf *mdev) { - ERR_IF (!mdev->bitmap) return; + if (!expect(mdev->bitmap)) + return; bm_free_pages(mdev->bitmap->bm_pages, mdev->bitmap->bm_number_of_pages); bm_vk_free(mdev->bitmap->bm_pages, (BM_P_VMALLOCED & mdev->bitmap->bm_flags)); kfree(mdev->bitmap); @@ -610,7 +629,8 @@ int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity, int set_new_bits) int err = 0, growing; int opages_vmalloced; - ERR_IF(!b) return -ENOMEM; + if (!expect(b)) + return -ENOMEM; drbd_bm_lock(mdev, "resize", BM_LOCKED_MASK); @@ -732,8 +752,10 @@ unsigned long _drbd_bm_total_weight(struct drbd_conf *mdev) unsigned long s; unsigned long flags; - ERR_IF(!b) return 0; - ERR_IF(!b->bm_pages) return 0; + if (!expect(b)) + return 0; + if (!expect(b->bm_pages)) + return 0; spin_lock_irqsave(&b->bm_lock, flags); s = b->bm_set; @@ -756,8 +778,10 @@ unsigned long drbd_bm_total_weight(struct drbd_conf *mdev) size_t drbd_bm_words(struct drbd_conf *mdev) { struct drbd_bitmap *b = mdev->bitmap; - ERR_IF(!b) return 0; - ERR_IF(!b->bm_pages) return 0; + if (!expect(b)) + return 0; + if (!expect(b->bm_pages)) + return 0; return b->bm_words; } @@ -765,7 +789,8 @@ size_t drbd_bm_words(struct drbd_conf *mdev) unsigned long drbd_bm_bits(struct drbd_conf *mdev) { struct drbd_bitmap *b = mdev->bitmap; - ERR_IF(!b) return 0; + if (!expect(b)) + return 0; return b->bm_bits; } @@ -786,8 +811,10 @@ void drbd_bm_merge_lel(struct drbd_conf *mdev, size_t offset, size_t number, end = offset + number; - ERR_IF(!b) return; - ERR_IF(!b->bm_pages) return; + if (!expect(b)) + return; + if (!expect(b->bm_pages)) + return; if (number == 0) return; WARN_ON(offset >= b->bm_words); @@ -831,8 +858,10 @@ void drbd_bm_get_lel(struct drbd_conf *mdev, size_t offset, size_t number, end = offset + number; - ERR_IF(!b) return; - ERR_IF(!b->bm_pages) return; + if (!expect(b)) + return; + if (!expect(b->bm_pages)) + return; spin_lock_irq(&b->bm_lock); if ((offset >= b->bm_words) || @@ -860,8 +889,10 @@ void drbd_bm_get_lel(struct drbd_conf *mdev, size_t offset, size_t number, void drbd_bm_set_all(struct drbd_conf *mdev) { struct drbd_bitmap *b = mdev->bitmap; - ERR_IF(!b) return; - ERR_IF(!b->bm_pages) return; + if (!expect(b)) + return; + if (!expect(b->bm_pages)) + return; spin_lock_irq(&b->bm_lock); bm_memset(b, 0, 0xff, b->bm_words); @@ -874,8 +905,10 @@ void drbd_bm_set_all(struct drbd_conf *mdev) void drbd_bm_clear_all(struct drbd_conf *mdev) { struct drbd_bitmap *b = mdev->bitmap; - ERR_IF(!b) return; - ERR_IF(!b->bm_pages) return; + if (!expect(b)) + return; + if (!expect(b->bm_pages)) + return; spin_lock_irq(&b->bm_lock); bm_memset(b, 0, 0, b->bm_words); @@ -889,7 +922,8 @@ struct bm_aio_ctx { unsigned int done; unsigned flags; #define BM_AIO_COPY_PAGES 1 -#define BM_WRITE_ALL_PAGES 2 +#define BM_AIO_WRITE_HINTED 2 +#define BM_WRITE_ALL_PAGES 4 int error; struct kref kref; }; @@ -977,17 +1011,11 @@ static void bm_page_io_async(struct bm_aio_ctx *ctx, int page_nr, int rw) __must bm_set_page_unchanged(b->bm_pages[page_nr]); if (ctx->flags & BM_AIO_COPY_PAGES) { - void *src, *dest; page = mempool_alloc(drbd_md_io_page_pool, __GFP_HIGHMEM|__GFP_WAIT); - dest = kmap_atomic(page); - src = kmap_atomic(b->bm_pages[page_nr]); - memcpy(dest, src, PAGE_SIZE); - kunmap_atomic(src); - kunmap_atomic(dest); + copy_highpage(page, b->bm_pages[page_nr]); bm_store_page_idx(page, page_nr); } else page = b->bm_pages[page_nr]; - bio->bi_bdev = mdev->ldev->md_bdev; bio->bi_sector = on_disk_sector; /* bio_add_page of a single page to an empty bio will always succeed, @@ -1060,6 +1088,11 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w if (lazy_writeout_upper_idx && i == lazy_writeout_upper_idx) break; if (rw & WRITE) { + if ((flags & BM_AIO_WRITE_HINTED) && + !test_and_clear_bit(BM_PAGE_HINT_WRITEOUT, + &page_private(b->bm_pages[i]))) + continue; + if (!(flags & BM_WRITE_ALL_PAGES) && bm_test_page_unchanged(b->bm_pages[i])) { dynamic_dev_dbg(DEV, "skipped bm write for idx %u\n", i); @@ -1088,13 +1121,15 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w * "in_flight reached zero, all done" event. */ if (!atomic_dec_and_test(&ctx->in_flight)) - wait_until_done_or_disk_failure(mdev, mdev->ldev, &ctx->done); + wait_until_done_or_force_detached(mdev, mdev->ldev, &ctx->done); else kref_put(&ctx->kref, &bm_aio_ctx_destroy); - dev_info(DEV, "bitmap %s of %u pages took %lu jiffies\n", - rw == WRITE ? "WRITE" : "READ", - count, jiffies - now); + /* summary for global bitmap IO */ + if (flags == 0) + dev_info(DEV, "bitmap %s of %u pages took %lu jiffies\n", + rw == WRITE ? "WRITE" : "READ", + count, jiffies - now); if (ctx->error) { dev_alert(DEV, "we had at least one MD IO ERROR during bitmap IO\n"); @@ -1103,7 +1138,7 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w } if (atomic_read(&ctx->in_flight)) - err = -EIO; /* Disk failed during IO... */ + err = -EIO; /* Disk timeout/force-detach during IO... */ now = jiffies; if (rw == WRITE) { @@ -1115,8 +1150,9 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w } now = b->bm_set; - dev_info(DEV, "%s (%lu bits) marked out-of-sync by on disk bit-map.\n", - ppsize(ppb, now << (BM_BLOCK_SHIFT-10)), now); + if (flags == 0) + dev_info(DEV, "%s (%lu bits) marked out-of-sync by on disk bit-map.\n", + ppsize(ppb, now << (BM_BLOCK_SHIFT-10)), now); kref_put(&ctx->kref, &bm_aio_ctx_destroy); return err; @@ -1179,9 +1215,17 @@ int drbd_bm_write_copy_pages(struct drbd_conf *mdev) __must_hold(local) return bm_rw(mdev, WRITE, BM_AIO_COPY_PAGES, 0); } +/** + * drbd_bm_write_hinted() - Write bitmap pages with "hint" marks, if they have changed. + * @mdev: DRBD device. + */ +int drbd_bm_write_hinted(struct drbd_conf *mdev) __must_hold(local) +{ + return bm_rw(mdev, WRITE, BM_AIO_WRITE_HINTED | BM_AIO_COPY_PAGES, 0); +} /** - * drbd_bm_write_page: Writes a PAGE_SIZE aligned piece of bitmap + * drbd_bm_write_page() - Writes a PAGE_SIZE aligned piece of bitmap * @mdev: DRBD device. * @idx: bitmap page index * @@ -1222,11 +1266,11 @@ int drbd_bm_write_page(struct drbd_conf *mdev, unsigned int idx) __must_hold(loc } bm_page_io_async(ctx, idx, WRITE_SYNC); - wait_until_done_or_disk_failure(mdev, mdev->ldev, &ctx->done); + wait_until_done_or_force_detached(mdev, mdev->ldev, &ctx->done); if (ctx->error) drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR); - /* that should force detach, so the in memory bitmap will be + /* that causes us to detach, so the in memory bitmap will be * gone in a moment as well. */ mdev->bm_writ_cnt++; @@ -1289,8 +1333,10 @@ static unsigned long bm_find_next(struct drbd_conf *mdev, struct drbd_bitmap *b = mdev->bitmap; unsigned long i = DRBD_END_OF_BITMAP; - ERR_IF(!b) return i; - ERR_IF(!b->bm_pages) return i; + if (!expect(b)) + return i; + if (!expect(b->bm_pages)) + return i; spin_lock_irq(&b->bm_lock); if (BM_DONT_TEST & b->bm_flags) @@ -1391,8 +1437,10 @@ static int bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s, struct drbd_bitmap *b = mdev->bitmap; int c = 0; - ERR_IF(!b) return 1; - ERR_IF(!b->bm_pages) return 0; + if (!expect(b)) + return 1; + if (!expect(b->bm_pages)) + return 0; spin_lock_irqsave(&b->bm_lock, flags); if ((val ? BM_DONT_SET : BM_DONT_CLEAR) & b->bm_flags) @@ -1423,13 +1471,21 @@ static inline void bm_set_full_words_within_one_page(struct drbd_bitmap *b, { int i; int bits; + int changed = 0; unsigned long *paddr = kmap_atomic(b->bm_pages[page_nr]); for (i = first_word; i < last_word; i++) { bits = hweight_long(paddr[i]); paddr[i] = ~0UL; - b->bm_set += BITS_PER_LONG - bits; + changed += BITS_PER_LONG - bits; } kunmap_atomic(paddr); + if (changed) { + /* We only need lazy writeout, the information is still in the + * remote bitmap as well, and is reconstructed during the next + * bitmap exchange, if lost locally due to a crash. */ + bm_set_page_lazy_writeout(b->bm_pages[page_nr]); + b->bm_set += changed; + } } /* Same thing as drbd_bm_set_bits, @@ -1524,8 +1580,10 @@ int drbd_bm_test_bit(struct drbd_conf *mdev, const unsigned long bitnr) unsigned long *p_addr; int i; - ERR_IF(!b) return 0; - ERR_IF(!b->bm_pages) return 0; + if (!expect(b)) + return 0; + if (!expect(b->bm_pages)) + return 0; spin_lock_irqsave(&b->bm_lock, flags); if (BM_DONT_TEST & b->bm_flags) @@ -1559,8 +1617,10 @@ int drbd_bm_count_bits(struct drbd_conf *mdev, const unsigned long s, const unsi * robust in case we screwed up elsewhere, in that case pretend there * was one dirty bit in the requested area, so we won't try to do a * local read there (no bitmap probably implies no disk) */ - ERR_IF(!b) return 1; - ERR_IF(!b->bm_pages) return 1; + if (!expect(b)) + return 1; + if (!expect(b->bm_pages)) + return 1; spin_lock_irqsave(&b->bm_lock, flags); if (BM_DONT_TEST & b->bm_flags) @@ -1573,11 +1633,10 @@ int drbd_bm_count_bits(struct drbd_conf *mdev, const unsigned long s, const unsi bm_unmap(p_addr); p_addr = bm_map_pidx(b, idx); } - ERR_IF (bitnr >= b->bm_bits) { - dev_err(DEV, "bitnr=%lu bm_bits=%lu\n", bitnr, b->bm_bits); - } else { + if (expect(bitnr < b->bm_bits)) c += (0 != test_bit_le(bitnr - (page_nr << (PAGE_SHIFT+3)), p_addr)); - } + else + dev_err(DEV, "bitnr=%lu bm_bits=%lu\n", bitnr, b->bm_bits); } if (p_addr) bm_unmap(p_addr); @@ -1607,8 +1666,10 @@ int drbd_bm_e_weight(struct drbd_conf *mdev, unsigned long enr) unsigned long flags; unsigned long *p_addr, *bm; - ERR_IF(!b) return 0; - ERR_IF(!b->bm_pages) return 0; + if (!expect(b)) + return 0; + if (!expect(b->bm_pages)) + return 0; spin_lock_irqsave(&b->bm_lock, flags); if (BM_DONT_TEST & b->bm_flags) @@ -1630,47 +1691,3 @@ int drbd_bm_e_weight(struct drbd_conf *mdev, unsigned long enr) spin_unlock_irqrestore(&b->bm_lock, flags); return count; } - -/* Set all bits covered by the AL-extent al_enr. - * Returns number of bits changed. */ -unsigned long drbd_bm_ALe_set_all(struct drbd_conf *mdev, unsigned long al_enr) -{ - struct drbd_bitmap *b = mdev->bitmap; - unsigned long *p_addr, *bm; - unsigned long weight; - unsigned long s, e; - int count, i, do_now; - ERR_IF(!b) return 0; - ERR_IF(!b->bm_pages) return 0; - - spin_lock_irq(&b->bm_lock); - if (BM_DONT_SET & b->bm_flags) - bm_print_lock_info(mdev); - weight = b->bm_set; - - s = al_enr * BM_WORDS_PER_AL_EXT; - e = min_t(size_t, s + BM_WORDS_PER_AL_EXT, b->bm_words); - /* assert that s and e are on the same page */ - D_ASSERT((e-1) >> (PAGE_SHIFT - LN2_BPL + 3) - == s >> (PAGE_SHIFT - LN2_BPL + 3)); - count = 0; - if (s < b->bm_words) { - i = do_now = e-s; - p_addr = bm_map_pidx(b, bm_word_to_page_idx(b, s)); - bm = p_addr + MLPP(s); - while (i--) { - count += hweight_long(*bm); - *bm = -1UL; - bm++; - } - bm_unmap(p_addr); - b->bm_set += do_now*BITS_PER_LONG - count; - if (e == b->bm_words) - b->bm_set -= bm_clear_surplus(b); - } else { - dev_err(DEV, "start offset (%lu) too large in drbd_bm_ALe_set_all\n", s); - } - weight = b->bm_set - weight; - spin_unlock_irq(&b->bm_lock); - return weight; -} diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h index b953cc7c9c00..6b51afa1aae1 100644 --- a/drivers/block/drbd/drbd_int.h +++ b/drivers/block/drbd/drbd_int.h @@ -39,9 +39,13 @@ #include <linux/major.h> #include <linux/blkdev.h> #include <linux/genhd.h> +#include <linux/idr.h> #include <net/tcp.h> #include <linux/lru_cache.h> #include <linux/prefetch.h> +#include <linux/drbd_genl_api.h> +#include <linux/drbd.h> +#include "drbd_state.h" #ifdef __CHECKER__ # define __protected_by(x) __attribute__((require_context(x,1,999,"rdwr"))) @@ -61,7 +65,6 @@ extern unsigned int minor_count; extern bool disable_sendpage; extern bool allow_oos; -extern unsigned int cn_idx; #ifdef CONFIG_DRBD_FAULT_INJECTION extern int enable_faults; @@ -86,34 +89,44 @@ extern char usermode_helper[]; */ #define DRBD_SIGKILL SIGHUP -/* All EEs on the free list should have ID_VACANT (== 0) - * freshly allocated EEs get !ID_VACANT (== 1) - * so if it says "cannot dereference null pointer at address 0x00000001", - * it is most likely one of these :( */ - #define ID_IN_SYNC (4711ULL) #define ID_OUT_OF_SYNC (4712ULL) - #define ID_SYNCER (-1ULL) -#define ID_VACANT 0 -#define is_syncer_block_id(id) ((id) == ID_SYNCER) + #define UUID_NEW_BM_OFFSET ((u64)0x0001000000000000ULL) struct drbd_conf; +struct drbd_tconn; /* to shorten dev_warn(DEV, "msg"); and relatives statements */ #define DEV (disk_to_dev(mdev->vdisk)) +#define conn_printk(LEVEL, TCONN, FMT, ARGS...) \ + printk(LEVEL "d-con %s: " FMT, TCONN->name , ## ARGS) +#define conn_alert(TCONN, FMT, ARGS...) conn_printk(KERN_ALERT, TCONN, FMT, ## ARGS) +#define conn_crit(TCONN, FMT, ARGS...) conn_printk(KERN_CRIT, TCONN, FMT, ## ARGS) +#define conn_err(TCONN, FMT, ARGS...) conn_printk(KERN_ERR, TCONN, FMT, ## ARGS) +#define conn_warn(TCONN, FMT, ARGS...) conn_printk(KERN_WARNING, TCONN, FMT, ## ARGS) +#define conn_notice(TCONN, FMT, ARGS...) conn_printk(KERN_NOTICE, TCONN, FMT, ## ARGS) +#define conn_info(TCONN, FMT, ARGS...) conn_printk(KERN_INFO, TCONN, FMT, ## ARGS) +#define conn_dbg(TCONN, FMT, ARGS...) conn_printk(KERN_DEBUG, TCONN, FMT, ## ARGS) + #define D_ASSERT(exp) if (!(exp)) \ dev_err(DEV, "ASSERT( " #exp " ) in %s:%d\n", __FILE__, __LINE__) -#define ERR_IF(exp) if (({ \ - int _b = (exp) != 0; \ - if (_b) dev_err(DEV, "ASSERT FAILED: %s: (%s) in %s:%d\n", \ - __func__, #exp, __FILE__, __LINE__); \ - _b; \ - })) +/** + * expect - Make an assertion + * + * Unlike the assert macro, this macro returns a boolean result. + */ +#define expect(exp) ({ \ + bool _bool = (exp); \ + if (!_bool) \ + dev_err(DEV, "ASSERTION %s FAILED in %s\n", \ + #exp, __func__); \ + _bool; \ + }) /* Defines to control fault insertion */ enum { @@ -150,15 +163,12 @@ drbd_insert_fault(struct drbd_conf *mdev, unsigned int type) { /* usual integer division */ #define div_floor(A, B) ((A)/(B)) -/* drbd_meta-data.c (still in drbd_main.c) */ -/* 4th incarnation of the disk layout. */ -#define DRBD_MD_MAGIC (DRBD_MAGIC+4) - -extern struct drbd_conf **minor_table; extern struct ratelimit_state drbd_ratelimit_state; +extern struct idr minors; /* RCU, updates: genl_lock() */ +extern struct list_head drbd_tconns; /* RCU, updates: genl_lock() */ /* on the wire */ -enum drbd_packets { +enum drbd_packet { /* receiver (data socket) */ P_DATA = 0x00, P_DATA_REPLY = 0x01, /* Response to P_DATA_REQUEST */ @@ -186,7 +196,7 @@ enum drbd_packets { P_RECV_ACK = 0x15, /* Used in protocol B */ P_WRITE_ACK = 0x16, /* Used in protocol C */ P_RS_WRITE_ACK = 0x17, /* Is a P_WRITE_ACK, additionally call set_in_sync(). */ - P_DISCARD_ACK = 0x18, /* Used in proto C, two-primaries conflict detection */ + P_SUPERSEDED = 0x18, /* Used in proto C, two-primaries conflict detection */ P_NEG_ACK = 0x19, /* Sent if local disk is unusable */ P_NEG_DREPLY = 0x1a, /* Local disk is broken... */ P_NEG_RS_DREPLY = 0x1b, /* Local disk is broken... */ @@ -207,77 +217,23 @@ enum drbd_packets { P_DELAY_PROBE = 0x27, /* is used on BOTH sockets */ P_OUT_OF_SYNC = 0x28, /* Mark as out of sync (Outrunning), data socket */ P_RS_CANCEL = 0x29, /* meta: Used to cancel RS_DATA_REQUEST packet by SyncSource */ + P_CONN_ST_CHG_REQ = 0x2a, /* data sock: Connection wide state request */ + P_CONN_ST_CHG_REPLY = 0x2b, /* meta sock: Connection side state req reply */ + P_RETRY_WRITE = 0x2c, /* Protocol C: retry conflicting write request */ + P_PROTOCOL_UPDATE = 0x2d, /* data sock: is used in established connections */ - P_MAX_CMD = 0x2A, P_MAY_IGNORE = 0x100, /* Flag to test if (cmd > P_MAY_IGNORE) ... */ P_MAX_OPT_CMD = 0x101, /* special command ids for handshake */ - P_HAND_SHAKE_M = 0xfff1, /* First Packet on the MetaSock */ - P_HAND_SHAKE_S = 0xfff2, /* First Packet on the Socket */ + P_INITIAL_META = 0xfff1, /* First Packet on the MetaSock */ + P_INITIAL_DATA = 0xfff2, /* First Packet on the Socket */ - P_HAND_SHAKE = 0xfffe /* FIXED for the next century! */ + P_CONNECTION_FEATURES = 0xfffe /* FIXED for the next century! */ }; -static inline const char *cmdname(enum drbd_packets cmd) -{ - /* THINK may need to become several global tables - * when we want to support more than - * one PRO_VERSION */ - static const char *cmdnames[] = { - [P_DATA] = "Data", - [P_DATA_REPLY] = "DataReply", - [P_RS_DATA_REPLY] = "RSDataReply", - [P_BARRIER] = "Barrier", - [P_BITMAP] = "ReportBitMap", - [P_BECOME_SYNC_TARGET] = "BecomeSyncTarget", - [P_BECOME_SYNC_SOURCE] = "BecomeSyncSource", - [P_UNPLUG_REMOTE] = "UnplugRemote", - [P_DATA_REQUEST] = "DataRequest", - [P_RS_DATA_REQUEST] = "RSDataRequest", - [P_SYNC_PARAM] = "SyncParam", - [P_SYNC_PARAM89] = "SyncParam89", - [P_PROTOCOL] = "ReportProtocol", - [P_UUIDS] = "ReportUUIDs", - [P_SIZES] = "ReportSizes", - [P_STATE] = "ReportState", - [P_SYNC_UUID] = "ReportSyncUUID", - [P_AUTH_CHALLENGE] = "AuthChallenge", - [P_AUTH_RESPONSE] = "AuthResponse", - [P_PING] = "Ping", - [P_PING_ACK] = "PingAck", - [P_RECV_ACK] = "RecvAck", - [P_WRITE_ACK] = "WriteAck", - [P_RS_WRITE_ACK] = "RSWriteAck", - [P_DISCARD_ACK] = "DiscardAck", - [P_NEG_ACK] = "NegAck", - [P_NEG_DREPLY] = "NegDReply", - [P_NEG_RS_DREPLY] = "NegRSDReply", - [P_BARRIER_ACK] = "BarrierAck", - [P_STATE_CHG_REQ] = "StateChgRequest", - [P_STATE_CHG_REPLY] = "StateChgReply", - [P_OV_REQUEST] = "OVRequest", - [P_OV_REPLY] = "OVReply", - [P_OV_RESULT] = "OVResult", - [P_CSUM_RS_REQUEST] = "CsumRSRequest", - [P_RS_IS_IN_SYNC] = "CsumRSIsInSync", - [P_COMPRESSED_BITMAP] = "CBitmap", - [P_DELAY_PROBE] = "DelayProbe", - [P_OUT_OF_SYNC] = "OutOfSync", - [P_MAX_CMD] = NULL, - }; - - if (cmd == P_HAND_SHAKE_M) - return "HandShakeM"; - if (cmd == P_HAND_SHAKE_S) - return "HandShakeS"; - if (cmd == P_HAND_SHAKE) - return "HandShake"; - if (cmd >= P_MAX_CMD) - return "Unknown"; - return cmdnames[cmd]; -} +extern const char *cmdname(enum drbd_packet cmd); /* for sending/receiving the bitmap, * possibly in some encoding scheme */ @@ -337,37 +293,24 @@ struct p_header80 { u32 magic; u16 command; u16 length; /* bytes of data after this header */ - u8 payload[0]; } __packed; /* Header for big packets, Used for data packets exceeding 64kB */ struct p_header95 { u16 magic; /* use DRBD_MAGIC_BIG here */ u16 command; - u32 length; /* Use only 24 bits of that. Ignore the highest 8 bit. */ - u8 payload[0]; + u32 length; } __packed; -union p_header { - struct p_header80 h80; - struct p_header95 h95; -}; - -/* - * short commands, packets without payload, plain p_header: - * P_PING - * P_PING_ACK - * P_BECOME_SYNC_TARGET - * P_BECOME_SYNC_SOURCE - * P_UNPLUG_REMOTE - */ +struct p_header100 { + u32 magic; + u16 volume; + u16 command; + u32 length; + u32 pad; +} __packed; -/* - * commands with out-of-struct payload: - * P_BITMAP (no additional fields) - * P_DATA, P_DATA_REPLY (see p_data) - * P_COMPRESSED_BITMAP (see receive_compressed_bitmap) - */ +extern unsigned int drbd_header_size(struct drbd_tconn *tconn); /* these defines must not be changed without changing the protocol version */ #define DP_HARDBARRIER 1 /* depricated */ @@ -377,9 +320,10 @@ union p_header { #define DP_FUA 16 /* equals REQ_FUA */ #define DP_FLUSH 32 /* equals REQ_FLUSH */ #define DP_DISCARD 64 /* equals REQ_DISCARD */ +#define DP_SEND_RECEIVE_ACK 128 /* This is a proto B write request */ +#define DP_SEND_WRITE_ACK 256 /* This is a proto C write request */ struct p_data { - union p_header head; u64 sector; /* 64 bits sector number */ u64 block_id; /* to identify the request in protocol B&C */ u32 seq_num; @@ -390,21 +334,18 @@ struct p_data { * commands which share a struct: * p_block_ack: * P_RECV_ACK (proto B), P_WRITE_ACK (proto C), - * P_DISCARD_ACK (proto C, two-primaries conflict detection) + * P_SUPERSEDED (proto C, two-primaries conflict detection) * p_block_req: * P_DATA_REQUEST, P_RS_DATA_REQUEST */ struct p_block_ack { - struct p_header80 head; u64 sector; u64 block_id; u32 blksize; u32 seq_num; } __packed; - struct p_block_req { - struct p_header80 head; u64 sector; u64 block_id; u32 blksize; @@ -413,59 +354,52 @@ struct p_block_req { /* * commands with their own struct for additional fields: - * P_HAND_SHAKE + * P_CONNECTION_FEATURES * P_BARRIER * P_BARRIER_ACK * P_SYNC_PARAM * ReportParams */ -struct p_handshake { - struct p_header80 head; /* 8 bytes */ +struct p_connection_features { u32 protocol_min; u32 feature_flags; u32 protocol_max; /* should be more than enough for future enhancements - * for now, feature_flags and the reserverd array shall be zero. + * for now, feature_flags and the reserved array shall be zero. */ u32 _pad; - u64 reserverd[7]; + u64 reserved[7]; } __packed; -/* 80 bytes, FIXED for the next century */ struct p_barrier { - struct p_header80 head; u32 barrier; /* barrier number _handle_ only */ u32 pad; /* to multiple of 8 Byte */ } __packed; struct p_barrier_ack { - struct p_header80 head; u32 barrier; u32 set_size; } __packed; struct p_rs_param { - struct p_header80 head; - u32 rate; + u32 resync_rate; /* Since protocol version 88 and higher. */ char verify_alg[0]; } __packed; struct p_rs_param_89 { - struct p_header80 head; - u32 rate; + u32 resync_rate; /* protocol version 89: */ char verify_alg[SHARED_SECRET_MAX]; char csums_alg[SHARED_SECRET_MAX]; } __packed; struct p_rs_param_95 { - struct p_header80 head; - u32 rate; + u32 resync_rate; char verify_alg[SHARED_SECRET_MAX]; char csums_alg[SHARED_SECRET_MAX]; u32 c_plan_ahead; @@ -475,12 +409,11 @@ struct p_rs_param_95 { } __packed; enum drbd_conn_flags { - CF_WANT_LOSE = 1, + CF_DISCARD_MY_DATA = 1, CF_DRY_RUN = 2, }; struct p_protocol { - struct p_header80 head; u32 protocol; u32 after_sb_0p; u32 after_sb_1p; @@ -494,17 +427,14 @@ struct p_protocol { } __packed; struct p_uuids { - struct p_header80 head; u64 uuid[UI_EXTENDED_SIZE]; } __packed; struct p_rs_uuid { - struct p_header80 head; u64 uuid; } __packed; struct p_sizes { - struct p_header80 head; u64 d_size; /* size of disk */ u64 u_size; /* user requested size */ u64 c_size; /* current exported size */ @@ -514,18 +444,15 @@ struct p_sizes { } __packed; struct p_state { - struct p_header80 head; u32 state; } __packed; struct p_req_state { - struct p_header80 head; u32 mask; u32 val; } __packed; struct p_req_state_reply { - struct p_header80 head; u32 retcode; } __packed; @@ -539,15 +466,7 @@ struct p_drbd06_param { u32 bit_map_gen[5]; } __packed; -struct p_discard { - struct p_header80 head; - u64 block_id; - u32 seq_num; - u32 pad; -} __packed; - struct p_block_desc { - struct p_header80 head; u64 sector; u32 blksize; u32 pad; /* to multiple of 8 Byte */ @@ -563,7 +482,6 @@ enum drbd_bitmap_code { }; struct p_compressed_bm { - struct p_header80 head; /* (encoding & 0x0f): actual encoding, see enum drbd_bitmap_code * (encoding & 0x80): polarity (set/unset) of first runlength * ((encoding >> 4) & 0x07): pad_bits, number of trailing zero bits @@ -575,90 +493,22 @@ struct p_compressed_bm { } __packed; struct p_delay_probe93 { - struct p_header80 head; u32 seq_num; /* sequence number to match the two probe packets */ u32 offset; /* usecs the probe got sent after the reference time point */ } __packed; -/* DCBP: Drbd Compressed Bitmap Packet ... */ -static inline enum drbd_bitmap_code -DCBP_get_code(struct p_compressed_bm *p) -{ - return (enum drbd_bitmap_code)(p->encoding & 0x0f); -} - -static inline void -DCBP_set_code(struct p_compressed_bm *p, enum drbd_bitmap_code code) -{ - BUG_ON(code & ~0xf); - p->encoding = (p->encoding & ~0xf) | code; -} - -static inline int -DCBP_get_start(struct p_compressed_bm *p) -{ - return (p->encoding & 0x80) != 0; -} - -static inline void -DCBP_set_start(struct p_compressed_bm *p, int set) -{ - p->encoding = (p->encoding & ~0x80) | (set ? 0x80 : 0); -} - -static inline int -DCBP_get_pad_bits(struct p_compressed_bm *p) -{ - return (p->encoding >> 4) & 0x7; -} - -static inline void -DCBP_set_pad_bits(struct p_compressed_bm *p, int n) -{ - BUG_ON(n & ~0x7); - p->encoding = (p->encoding & (~0x7 << 4)) | (n << 4); -} - -/* one bitmap packet, including the p_header, - * should fit within one _architecture independend_ page. - * so we need to use the fixed size 4KiB page size - * most architectures have used for a long time. +/* + * Bitmap packets need to fit within a single page on the sender and receiver, + * so we are limited to 4 KiB (and not to PAGE_SIZE, which can be bigger). */ -#define BM_PACKET_PAYLOAD_BYTES (4096 - sizeof(struct p_header80)) -#define BM_PACKET_WORDS (BM_PACKET_PAYLOAD_BYTES/sizeof(long)) -#define BM_PACKET_VLI_BYTES_MAX (4096 - sizeof(struct p_compressed_bm)) -#if (PAGE_SIZE < 4096) -/* drbd_send_bitmap / receive_bitmap would break horribly */ -#error "PAGE_SIZE too small" -#endif - -union p_polymorph { - union p_header header; - struct p_handshake handshake; - struct p_data data; - struct p_block_ack block_ack; - struct p_barrier barrier; - struct p_barrier_ack barrier_ack; - struct p_rs_param_89 rs_param_89; - struct p_rs_param_95 rs_param_95; - struct p_protocol protocol; - struct p_sizes sizes; - struct p_uuids uuids; - struct p_state state; - struct p_req_state req_state; - struct p_req_state_reply req_state_reply; - struct p_block_req block_req; - struct p_delay_probe93 delay_probe93; - struct p_rs_uuid rs_uuid; - struct p_block_desc block_desc; -} __packed; +#define DRBD_SOCKET_BUFFER_SIZE 4096 /**********************************************************************/ enum drbd_thread_state { - None, - Running, - Exiting, - Restarting + NONE, + RUNNING, + EXITING, + RESTARTING }; struct drbd_thread { @@ -667,8 +517,9 @@ struct drbd_thread { struct completion stop; enum drbd_thread_state t_state; int (*function) (struct drbd_thread *); - struct drbd_conf *mdev; + struct drbd_tconn *tconn; int reset_cpu_mask; + char name[9]; }; static inline enum drbd_thread_state get_t_state(struct drbd_thread *thi) @@ -681,58 +532,54 @@ static inline enum drbd_thread_state get_t_state(struct drbd_thread *thi) return thi->t_state; } -struct drbd_work; -typedef int (*drbd_work_cb)(struct drbd_conf *, struct drbd_work *, int cancel); struct drbd_work { struct list_head list; - drbd_work_cb cb; + int (*cb)(struct drbd_work *, int cancel); + union { + struct drbd_conf *mdev; + struct drbd_tconn *tconn; + }; }; -struct drbd_tl_epoch; +#include "drbd_interval.h" + +extern int drbd_wait_misc(struct drbd_conf *, struct drbd_interval *); + struct drbd_request { struct drbd_work w; - struct drbd_conf *mdev; /* if local IO is not allowed, will be NULL. * if local IO _is_ allowed, holds the locally submitted bio clone, * or, after local IO completion, the ERR_PTR(error). - * see drbd_endio_pri(). */ + * see drbd_request_endio(). */ struct bio *private_bio; - struct hlist_node collision; - sector_t sector; - unsigned int size; - unsigned int epoch; /* barrier_nr */ + struct drbd_interval i; - /* barrier_nr: used to check on "completion" whether this req was in + /* epoch: used to check on "completion" whether this req was in * the current epoch, and we therefore have to close it, - * starting a new epoch... + * causing a p_barrier packet to be send, starting a new epoch. + * + * This corresponds to "barrier" in struct p_barrier[_ack], + * and to "barrier_nr" in struct drbd_epoch (and various + * comments/function parameters/local variable names). */ + unsigned int epoch; struct list_head tl_requests; /* ring list in the transfer log */ struct bio *master_bio; /* master bio pointer */ - unsigned long rq_state; /* see comments above _req_mod() */ unsigned long start_time; -}; - -struct drbd_tl_epoch { - struct drbd_work w; - struct list_head requests; /* requests before */ - struct drbd_tl_epoch *next; /* pointer to the next barrier */ - unsigned int br_number; /* the barriers identifier. */ - int n_writes; /* number of requests attached before this barrier */ -}; -struct drbd_request; + /* once it hits 0, we may complete the master_bio */ + atomic_t completion_ref; + /* once it hits 0, we may destroy this drbd_request object */ + struct kref kref; -/* These Tl_epoch_entries may be in one of 6 lists: - active_ee .. data packet being written - sync_ee .. syncer block being written - done_ee .. block written, need to send P_WRITE_ACK - read_ee .. [RS]P_DATA_REQUEST being read -*/ + unsigned rq_state; /* see comments above _req_mod() */ +}; struct drbd_epoch { + struct drbd_tconn *tconn; struct list_head list; unsigned int barrier_nr; atomic_t epoch_size; /* increased on every request added. */ @@ -762,17 +609,14 @@ struct digest_info { void *digest; }; -struct drbd_epoch_entry { +struct drbd_peer_request { struct drbd_work w; - struct hlist_node collision; struct drbd_epoch *epoch; /* for writes */ - struct drbd_conf *mdev; struct page *pages; atomic_t pending_bios; - unsigned int size; + struct drbd_interval i; /* see comments on ee flag bits below */ unsigned long flags; - sector_t sector; union { u64 block_id; struct digest_info *digest; @@ -793,31 +637,37 @@ enum { * we need to resubmit without the barrier flag. */ __EE_RESUBMITTED, - /* we may have several bios per epoch entry. + /* we may have several bios per peer request. * if any of those fail, we set this flag atomically * from the endio callback */ __EE_WAS_ERROR, /* This ee has a pointer to a digest instead of a block id */ __EE_HAS_DIGEST, + + /* Conflicting local requests need to be restarted after this request */ + __EE_RESTART_REQUESTS, + + /* The peer wants a write ACK for this (wire proto C) */ + __EE_SEND_WRITE_ACK, + + /* Is set when net_conf had two_primaries set while creating this peer_req */ + __EE_IN_INTERVAL_TREE, }; #define EE_CALL_AL_COMPLETE_IO (1<<__EE_CALL_AL_COMPLETE_IO) #define EE_MAY_SET_IN_SYNC (1<<__EE_MAY_SET_IN_SYNC) #define EE_RESUBMITTED (1<<__EE_RESUBMITTED) #define EE_WAS_ERROR (1<<__EE_WAS_ERROR) #define EE_HAS_DIGEST (1<<__EE_HAS_DIGEST) +#define EE_RESTART_REQUESTS (1<<__EE_RESTART_REQUESTS) +#define EE_SEND_WRITE_ACK (1<<__EE_SEND_WRITE_ACK) +#define EE_IN_INTERVAL_TREE (1<<__EE_IN_INTERVAL_TREE) -/* global flag bits */ +/* flag bits per mdev */ enum { - CREATE_BARRIER, /* next P_DATA is preceded by a P_BARRIER */ - SIGNAL_ASENDER, /* whether asender wants to be interrupted */ - SEND_PING, /* whether asender should send a ping asap */ - UNPLUG_REMOTE, /* sending a "UnplugRemote" could help */ MD_DIRTY, /* current uuids and flags not yet on disk */ - DISCARD_CONCURRENT, /* Set on one node, cleared on the peer! */ USE_DEGR_WFC_T, /* degr-wfc-timeout instead of wfc-timeout. */ - CLUSTER_ST_CHANGE, /* Cluster wide state change going on... */ CL_ST_CHG_SUCCESS, CL_ST_CHG_FAIL, CRASHED_PRIMARY, /* This node was a crashed primary. @@ -831,32 +681,18 @@ enum { once no more io in flight, start bitmap io */ BITMAP_IO_QUEUED, /* Started bitmap IO */ GO_DISKLESS, /* Disk is being detached, on io-error or admin request. */ - WAS_IO_ERROR, /* Local disk failed returned IO error */ + WAS_IO_ERROR, /* Local disk failed, returned IO error */ + WAS_READ_ERROR, /* Local disk READ failed (set additionally to the above) */ FORCE_DETACH, /* Force-detach from local disk, aborting any pending local IO */ RESYNC_AFTER_NEG, /* Resync after online grow after the attach&negotiate finished. */ - NET_CONGESTED, /* The data socket is congested */ - - CONFIG_PENDING, /* serialization of (re)configuration requests. - * if set, also prevents the device from dying */ - DEVICE_DYING, /* device became unconfigured, - * but worker thread is still handling the cleanup. - * reconfiguring (nl_disk_conf, nl_net_conf) is dissalowed, - * while this is set. */ RESIZE_PENDING, /* Size change detected locally, waiting for the response from * the peer, if it changed there as well. */ - CONN_DRY_RUN, /* Expect disconnect after resync handshake. */ - GOT_PING_ACK, /* set when we receive a ping_ack packet, misc wait gets woken */ NEW_CUR_UUID, /* Create new current UUID when thawing IO */ AL_SUSPENDED, /* Activity logging is currently suspended. */ AHEAD_TO_SYNC_SOURCE, /* Ahead -> SyncSource queued */ - STATE_SENT, /* Do not change state/UUIDs while this is set */ - - CALLBACK_PENDING, /* Whether we have a call_usermodehelper(, UMH_WAIT_PROC) - * pending, from drbd worker context. - * If set, bdi_write_congested() returns true, - * so shrink_page_list() would not recurse into, - * and potentially deadlock on, this drbd worker. - */ + B_RS_H_DONE, /* Before resync handler done (already executed) */ + DISCARD_MY_DATA, /* discard_my_data flag per volume */ + READ_BALANCE_RR, }; struct drbd_bitmap; /* opaque for drbd_conf */ @@ -894,24 +730,24 @@ enum bm_flag { struct drbd_work_queue { struct list_head q; - struct semaphore s; /* producers up it, worker down()s it */ spinlock_t q_lock; /* to protect the list. */ + wait_queue_head_t q_wait; }; struct drbd_socket { - struct drbd_work_queue work; struct mutex mutex; struct socket *socket; /* this way we get our * send/receive buffers off the stack */ - union p_polymorph sbuf; - union p_polymorph rbuf; + void *sbuf; + void *rbuf; }; struct drbd_md { u64 md_offset; /* sector offset to 'super' block */ u64 la_size_sect; /* last agreed size, unit sectors */ + spinlock_t uuid_lock; u64 uuid[UI_SIZE]; u64 device_uuid; u32 flags; @@ -921,24 +757,16 @@ struct drbd_md { s32 bm_offset; /* signed relative sector offset to bitmap */ /* u32 al_nr_extents; important for restoring the AL - * is stored into sync_conf.al_extents, which in turn + * is stored into ldev->dc.al_extents, which in turn * gets applied to act_log->nr_elements */ }; -/* for sync_conf and other types... */ -#define NL_PACKET(name, number, fields) struct name { fields }; -#define NL_INTEGER(pn,pr,member) int member; -#define NL_INT64(pn,pr,member) __u64 member; -#define NL_BIT(pn,pr,member) unsigned member:1; -#define NL_STRING(pn,pr,member,len) unsigned char member[len]; int member ## _len; -#include <linux/drbd_nl.h> - struct drbd_backing_dev { struct block_device *backing_bdev; struct block_device *md_bdev; struct drbd_md md; - struct disk_conf dc; /* The user provided config... */ + struct disk_conf *disk_conf; /* RCU, for updates: mdev->tconn->conf_update */ sector_t known_size; /* last known size of that backing device */ }; @@ -962,18 +790,116 @@ enum write_ordering_e { }; struct fifo_buffer { - int *values; unsigned int head_index; unsigned int size; + int total; /* sum of all values */ + int values[0]; +}; +extern struct fifo_buffer *fifo_alloc(int fifo_size); + +/* flag bits per tconn */ +enum { + NET_CONGESTED, /* The data socket is congested */ + RESOLVE_CONFLICTS, /* Set on one node, cleared on the peer! */ + SEND_PING, /* whether asender should send a ping asap */ + SIGNAL_ASENDER, /* whether asender wants to be interrupted */ + GOT_PING_ACK, /* set when we receive a ping_ack packet, ping_wait gets woken */ + CONN_WD_ST_CHG_REQ, /* A cluster wide state change on the connection is active */ + CONN_WD_ST_CHG_OKAY, + CONN_WD_ST_CHG_FAIL, + CONN_DRY_RUN, /* Expect disconnect after resync handshake. */ + CREATE_BARRIER, /* next P_DATA is preceded by a P_BARRIER */ + STATE_SENT, /* Do not change state/UUIDs while this is set */ + CALLBACK_PENDING, /* Whether we have a call_usermodehelper(, UMH_WAIT_PROC) + * pending, from drbd worker context. + * If set, bdi_write_congested() returns true, + * so shrink_page_list() would not recurse into, + * and potentially deadlock on, this drbd worker. + */ + DISCONNECT_SENT, +}; + +struct drbd_tconn { /* is a resource from the config file */ + char *name; /* Resource name */ + struct list_head all_tconn; /* linked on global drbd_tconns */ + struct kref kref; + struct idr volumes; /* <tconn, vnr> to mdev mapping */ + enum drbd_conns cstate; /* Only C_STANDALONE to C_WF_REPORT_PARAMS */ + unsigned susp:1; /* IO suspended by user */ + unsigned susp_nod:1; /* IO suspended because no data */ + unsigned susp_fen:1; /* IO suspended because fence peer handler runs */ + struct mutex cstate_mutex; /* Protects graceful disconnects */ + + unsigned long flags; + struct net_conf *net_conf; /* content protected by rcu */ + struct mutex conf_update; /* mutex for ready-copy-update of net_conf and disk_conf */ + wait_queue_head_t ping_wait; /* Woken upon reception of a ping, and a state change */ + struct res_opts res_opts; + + struct sockaddr_storage my_addr; + int my_addr_len; + struct sockaddr_storage peer_addr; + int peer_addr_len; + + struct drbd_socket data; /* data/barrier/cstate/parameter packets */ + struct drbd_socket meta; /* ping/ack (metadata) packets */ + int agreed_pro_version; /* actually used protocol version */ + unsigned long last_received; /* in jiffies, either socket */ + unsigned int ko_count; + + spinlock_t req_lock; + + struct list_head transfer_log; /* all requests not yet fully processed */ + + struct crypto_hash *cram_hmac_tfm; + struct crypto_hash *integrity_tfm; /* checksums we compute, updates protected by tconn->data->mutex */ + struct crypto_hash *peer_integrity_tfm; /* checksums we verify, only accessed from receiver thread */ + struct crypto_hash *csums_tfm; + struct crypto_hash *verify_tfm; + void *int_dig_in; + void *int_dig_vv; + + /* receiver side */ + struct drbd_epoch *current_epoch; + spinlock_t epoch_lock; + unsigned int epochs; + enum write_ordering_e write_ordering; + atomic_t current_tle_nr; /* transfer log epoch number */ + unsigned current_tle_writes; /* writes seen within this tl epoch */ + + unsigned long last_reconnect_jif; + struct drbd_thread receiver; + struct drbd_thread worker; + struct drbd_thread asender; + cpumask_var_t cpu_mask; + + /* sender side */ + struct drbd_work_queue sender_work; + + struct { + /* whether this sender thread + * has processed a single write yet. */ + bool seen_any_write_yet; + + /* Which barrier number to send with the next P_BARRIER */ + int current_epoch_nr; + + /* how many write requests have been sent + * with req->epoch == current_epoch_nr. + * If none, no P_BARRIER will be sent. */ + unsigned current_epoch_writes; + } send; }; struct drbd_conf { + struct drbd_tconn *tconn; + int vnr; /* volume number within the connection */ + struct kref kref; + /* things that are stored as / read from meta data on disk */ unsigned long flags; /* configured by drbdsetup */ - struct net_conf *net_conf; /* protected by get_net_conf() and put_net_conf() */ - struct syncer_conf sync_conf; struct drbd_backing_dev *ldev __protected_by(local); sector_t p_size; /* partner's disk size */ @@ -981,11 +907,7 @@ struct drbd_conf { struct block_device *this_bdev; struct gendisk *vdisk; - struct drbd_socket data; /* data/barrier/cstate/parameter packets */ - struct drbd_socket meta; /* ping/ack (metadata) packets */ - int agreed_pro_version; /* actually used protocol version */ - unsigned long last_received; /* in jiffies, either socket */ - unsigned int ko_count; + unsigned long last_reattach_jif; struct drbd_work resync_work, unplug_work, go_diskless, @@ -1005,10 +927,9 @@ struct drbd_conf { /* Used after attach while negotiating new disk state. */ union drbd_state new_state_tmp; - union drbd_state state; + union drbd_dev_state state; wait_queue_head_t misc_wait; wait_queue_head_t state_wait; /* upon each state change. */ - wait_queue_head_t net_cnt_wait; unsigned int send_cnt; unsigned int recv_cnt; unsigned int read_cnt; @@ -1018,17 +939,12 @@ struct drbd_conf { atomic_t ap_bio_cnt; /* Requests we need to complete */ atomic_t ap_pending_cnt; /* AP data packets on the wire, ack expected */ atomic_t rs_pending_cnt; /* RS request/data packets on the wire */ - atomic_t unacked_cnt; /* Need to send replys for */ + atomic_t unacked_cnt; /* Need to send replies for */ atomic_t local_cnt; /* Waiting for local completion */ - atomic_t net_cnt; /* Users of net_conf */ - spinlock_t req_lock; - struct drbd_tl_epoch *unused_spare_tle; /* for pre-allocation */ - struct drbd_tl_epoch *newest_tle; - struct drbd_tl_epoch *oldest_tle; - struct list_head out_of_sequence_requests; - struct list_head barrier_acked_requests; - struct hlist_head *tl_hash; - unsigned int tl_hash_s; + + /* Interval tree of pending local requests */ + struct rb_root read_requests; + struct rb_root write_requests; /* blocks to resync in this run [unit BM_BLOCK_SIZE] */ unsigned long rs_total; @@ -1048,9 +964,11 @@ struct drbd_conf { unsigned long rs_mark_time[DRBD_SYNC_MARKS]; /* current index into rs_mark_{left,time} */ int rs_last_mark; + unsigned long rs_last_bcast; /* [unit jiffies] */ /* where does the admin want us to start? (sector) */ sector_t ov_start_sector; + sector_t ov_stop_sector; /* where are we now? (sector) */ sector_t ov_position; /* Start sector of out of sync range (to merge printk reporting). */ @@ -1058,14 +976,7 @@ struct drbd_conf { /* size of out-of-sync range in sectors. */ sector_t ov_last_oos_size; unsigned long ov_left; /* in bits */ - struct crypto_hash *csums_tfm; - struct crypto_hash *verify_tfm; - unsigned long last_reattach_jif; - unsigned long last_reconnect_jif; - struct drbd_thread receiver; - struct drbd_thread worker; - struct drbd_thread asender; struct drbd_bitmap *bitmap; unsigned long bm_resync_fo; /* bit offset for drbd_bm_find_next */ @@ -1078,29 +989,19 @@ struct drbd_conf { int open_cnt; u64 *p_uuid; - struct drbd_epoch *current_epoch; - spinlock_t epoch_lock; - unsigned int epochs; - enum write_ordering_e write_ordering; + struct list_head active_ee; /* IO in progress (P_DATA gets written to disk) */ struct list_head sync_ee; /* IO in progress (P_RS_DATA_REPLY gets written to disk) */ - struct list_head done_ee; /* send ack */ - struct list_head read_ee; /* IO in progress (any read) */ + struct list_head done_ee; /* need to send P_WRITE_ACK */ + struct list_head read_ee; /* [RS]P_DATA_REQUEST being read */ struct list_head net_ee; /* zero-copy network send in progress */ - struct hlist_head *ee_hash; /* is proteced by req_lock! */ - unsigned int ee_hash_s; - - /* this one is protected by ee_lock, single thread */ - struct drbd_epoch_entry *last_write_w_barrier; int next_barrier_nr; - struct hlist_head *app_reads_hash; /* is proteced by req_lock */ struct list_head resync_reads; atomic_t pp_in_use; /* allocated from page pool */ atomic_t pp_in_use_by_net; /* sendpage()d, still referenced by tcp */ wait_queue_head_t ee_wait; struct page *md_io_page; /* one page buffer for md_io */ - struct page *md_io_tmpp; /* for logical_block_size != 512 */ struct drbd_md_io md_io; atomic_t md_io_in_use; /* protects the md_io, md_io_page and md_io_tmpp */ spinlock_t al_lock; @@ -1109,22 +1010,16 @@ struct drbd_conf { unsigned int al_tr_number; int al_tr_cycle; int al_tr_pos; /* position of the next transaction in the journal */ - struct crypto_hash *cram_hmac_tfm; - struct crypto_hash *integrity_w_tfm; /* to be used by the worker thread */ - struct crypto_hash *integrity_r_tfm; /* to be used by the receiver thread */ - void *int_dig_out; - void *int_dig_in; - void *int_dig_vv; wait_queue_head_t seq_wait; atomic_t packet_seq; unsigned int peer_seq; spinlock_t peer_seq_lock; unsigned int minor; unsigned long comm_bm_set; /* communicated number of set bits. */ - cpumask_var_t cpu_mask; struct bm_io_work bm_io_work; u64 ed_uuid; /* UUID of the exposed data */ - struct mutex state_mutex; + struct mutex own_state_mutex; + struct mutex *state_mutex; /* either own_state_mutex or mdev->tconn->cstate_mutex */ char congestion_reason; /* Why we where congested... */ atomic_t rs_sect_in; /* for incoming resync data rate, SyncTarget */ atomic_t rs_sect_ev; /* for submitted resync data rate, both */ @@ -1132,9 +1027,8 @@ struct drbd_conf { int rs_last_events; /* counter of read or write "events" (unit sectors) * on the lower level device when we last looked. */ int c_sync_rate; /* current resync rate after syncer throttle magic */ - struct fifo_buffer rs_plan_s; /* correction values of resync planer */ + struct fifo_buffer *rs_plan_s; /* correction values of resync planer (RCU, tconn->conn_update) */ int rs_in_flight; /* resync sectors in flight (to proxy, in proxy and from proxy) */ - int rs_planed; /* resync sectors already planned */ atomic_t ap_in_flight; /* App sectors in flight (waiting for ack) */ unsigned int peer_max_bio_size; unsigned int local_max_bio_size; @@ -1142,11 +1036,7 @@ struct drbd_conf { static inline struct drbd_conf *minor_to_mdev(unsigned int minor) { - struct drbd_conf *mdev; - - mdev = minor < minor_count ? minor_table[minor] : NULL; - - return mdev; + return (struct drbd_conf *)idr_find(&minors, minor); } static inline unsigned int mdev_to_minor(struct drbd_conf *mdev) @@ -1154,29 +1044,9 @@ static inline unsigned int mdev_to_minor(struct drbd_conf *mdev) return mdev->minor; } -/* returns 1 if it was successful, - * returns 0 if there was no data socket. - * so wherever you are going to use the data.socket, e.g. do - * if (!drbd_get_data_sock(mdev)) - * return 0; - * CODE(); - * drbd_put_data_sock(mdev); - */ -static inline int drbd_get_data_sock(struct drbd_conf *mdev) -{ - mutex_lock(&mdev->data.mutex); - /* drbd_disconnect() could have called drbd_free_sock() - * while we were waiting in down()... */ - if (unlikely(mdev->data.socket == NULL)) { - mutex_unlock(&mdev->data.mutex); - return 0; - } - return 1; -} - -static inline void drbd_put_data_sock(struct drbd_conf *mdev) +static inline struct drbd_conf *vnr_to_mdev(struct drbd_tconn *tconn, int vnr) { - mutex_unlock(&mdev->data.mutex); + return (struct drbd_conf *)idr_find(&tconn->volumes, vnr); } /* @@ -1185,106 +1055,77 @@ static inline void drbd_put_data_sock(struct drbd_conf *mdev) /* drbd_main.c */ -enum chg_state_flags { - CS_HARD = 1, - CS_VERBOSE = 2, - CS_WAIT_COMPLETE = 4, - CS_SERIALIZE = 8, - CS_ORDERED = CS_WAIT_COMPLETE + CS_SERIALIZE, -}; - enum dds_flags { DDSF_FORCED = 1, DDSF_NO_RESYNC = 2, /* Do not run a resync for the new space */ }; extern void drbd_init_set_defaults(struct drbd_conf *mdev); -extern enum drbd_state_rv drbd_change_state(struct drbd_conf *mdev, - enum chg_state_flags f, - union drbd_state mask, - union drbd_state val); -extern void drbd_force_state(struct drbd_conf *, union drbd_state, - union drbd_state); -extern enum drbd_state_rv _drbd_request_state(struct drbd_conf *, - union drbd_state, - union drbd_state, - enum chg_state_flags); -extern enum drbd_state_rv __drbd_set_state(struct drbd_conf *, union drbd_state, - enum chg_state_flags, - struct completion *done); -extern void print_st_err(struct drbd_conf *, union drbd_state, - union drbd_state, int); extern int drbd_thread_start(struct drbd_thread *thi); extern void _drbd_thread_stop(struct drbd_thread *thi, int restart, int wait); +extern char *drbd_task_to_thread_name(struct drbd_tconn *tconn, struct task_struct *task); #ifdef CONFIG_SMP -extern void drbd_thread_current_set_cpu(struct drbd_conf *mdev); -extern void drbd_calc_cpu_mask(struct drbd_conf *mdev); +extern void drbd_thread_current_set_cpu(struct drbd_thread *thi); +extern void drbd_calc_cpu_mask(struct drbd_tconn *tconn); #else #define drbd_thread_current_set_cpu(A) ({}) #define drbd_calc_cpu_mask(A) ({}) #endif -extern void drbd_free_resources(struct drbd_conf *mdev); -extern void tl_release(struct drbd_conf *mdev, unsigned int barrier_nr, +extern void tl_release(struct drbd_tconn *, unsigned int barrier_nr, unsigned int set_size); -extern void tl_clear(struct drbd_conf *mdev); -extern void _tl_add_barrier(struct drbd_conf *, struct drbd_tl_epoch *); -extern void drbd_free_sock(struct drbd_conf *mdev); -extern int drbd_send(struct drbd_conf *mdev, struct socket *sock, - void *buf, size_t size, unsigned msg_flags); -extern int drbd_send_protocol(struct drbd_conf *mdev); +extern void tl_clear(struct drbd_tconn *); +extern void drbd_free_sock(struct drbd_tconn *tconn); +extern int drbd_send(struct drbd_tconn *tconn, struct socket *sock, + void *buf, size_t size, unsigned msg_flags); +extern int drbd_send_all(struct drbd_tconn *, struct socket *, void *, size_t, + unsigned); + +extern int __drbd_send_protocol(struct drbd_tconn *tconn, enum drbd_packet cmd); +extern int drbd_send_protocol(struct drbd_tconn *tconn); extern int drbd_send_uuids(struct drbd_conf *mdev); extern int drbd_send_uuids_skip_initial_sync(struct drbd_conf *mdev); -extern int drbd_gen_and_send_sync_uuid(struct drbd_conf *mdev); +extern void drbd_gen_and_send_sync_uuid(struct drbd_conf *mdev); extern int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags flags); extern int drbd_send_state(struct drbd_conf *mdev, union drbd_state s); extern int drbd_send_current_state(struct drbd_conf *mdev); -extern int _drbd_send_cmd(struct drbd_conf *mdev, struct socket *sock, - enum drbd_packets cmd, struct p_header80 *h, - size_t size, unsigned msg_flags); -#define USE_DATA_SOCKET 1 -#define USE_META_SOCKET 0 -extern int drbd_send_cmd(struct drbd_conf *mdev, int use_data_socket, - enum drbd_packets cmd, struct p_header80 *h, - size_t size); -extern int drbd_send_cmd2(struct drbd_conf *mdev, enum drbd_packets cmd, - char *data, size_t size); -extern int drbd_send_sync_param(struct drbd_conf *mdev, struct syncer_conf *sc); -extern int drbd_send_b_ack(struct drbd_conf *mdev, u32 barrier_nr, - u32 set_size); -extern int drbd_send_ack(struct drbd_conf *mdev, enum drbd_packets cmd, - struct drbd_epoch_entry *e); -extern int drbd_send_ack_rp(struct drbd_conf *mdev, enum drbd_packets cmd, - struct p_block_req *rp); -extern int drbd_send_ack_dp(struct drbd_conf *mdev, enum drbd_packets cmd, - struct p_data *dp, int data_size); -extern int drbd_send_ack_ex(struct drbd_conf *mdev, enum drbd_packets cmd, +extern int drbd_send_sync_param(struct drbd_conf *mdev); +extern void drbd_send_b_ack(struct drbd_tconn *tconn, u32 barrier_nr, + u32 set_size); +extern int drbd_send_ack(struct drbd_conf *, enum drbd_packet, + struct drbd_peer_request *); +extern void drbd_send_ack_rp(struct drbd_conf *mdev, enum drbd_packet cmd, + struct p_block_req *rp); +extern void drbd_send_ack_dp(struct drbd_conf *mdev, enum drbd_packet cmd, + struct p_data *dp, int data_size); +extern int drbd_send_ack_ex(struct drbd_conf *mdev, enum drbd_packet cmd, sector_t sector, int blksize, u64 block_id); -extern int drbd_send_oos(struct drbd_conf *mdev, struct drbd_request *req); -extern int drbd_send_block(struct drbd_conf *mdev, enum drbd_packets cmd, - struct drbd_epoch_entry *e); +extern int drbd_send_out_of_sync(struct drbd_conf *, struct drbd_request *); +extern int drbd_send_block(struct drbd_conf *, enum drbd_packet, + struct drbd_peer_request *); extern int drbd_send_dblock(struct drbd_conf *mdev, struct drbd_request *req); extern int drbd_send_drequest(struct drbd_conf *mdev, int cmd, sector_t sector, int size, u64 block_id); -extern int drbd_send_drequest_csum(struct drbd_conf *mdev, - sector_t sector,int size, - void *digest, int digest_size, - enum drbd_packets cmd); +extern int drbd_send_drequest_csum(struct drbd_conf *mdev, sector_t sector, + int size, void *digest, int digest_size, + enum drbd_packet cmd); extern int drbd_send_ov_request(struct drbd_conf *mdev,sector_t sector,int size); extern int drbd_send_bitmap(struct drbd_conf *mdev); -extern int _drbd_send_bitmap(struct drbd_conf *mdev); -extern int drbd_send_sr_reply(struct drbd_conf *mdev, enum drbd_state_rv retcode); +extern void drbd_send_sr_reply(struct drbd_conf *mdev, enum drbd_state_rv retcode); +extern void conn_send_sr_reply(struct drbd_tconn *tconn, enum drbd_state_rv retcode); extern void drbd_free_bc(struct drbd_backing_dev *ldev); extern void drbd_mdev_cleanup(struct drbd_conf *mdev); void drbd_print_uuids(struct drbd_conf *mdev, const char *text); +extern void conn_md_sync(struct drbd_tconn *tconn); extern void drbd_md_sync(struct drbd_conf *mdev); extern int drbd_md_read(struct drbd_conf *mdev, struct drbd_backing_dev *bdev); extern void drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local); extern void _drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local); extern void drbd_uuid_new_current(struct drbd_conf *mdev) __must_hold(local); -extern void _drbd_uuid_new_current(struct drbd_conf *mdev) __must_hold(local); extern void drbd_uuid_set_bm(struct drbd_conf *mdev, u64 val) __must_hold(local); +extern void drbd_uuid_move_history(struct drbd_conf *mdev) __must_hold(local); +extern void __drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local); extern void drbd_md_set_flag(struct drbd_conf *mdev, int flags) __must_hold(local); extern void drbd_md_clear_flag(struct drbd_conf *mdev, int flags)__must_hold(local); extern int drbd_md_test_flag(struct drbd_backing_dev *, int); @@ -1302,33 +1143,52 @@ extern void drbd_queue_bitmap_io(struct drbd_conf *mdev, extern int drbd_bitmap_io(struct drbd_conf *mdev, int (*io_fn)(struct drbd_conf *), char *why, enum bm_flag flags); +extern int drbd_bitmap_io_from_worker(struct drbd_conf *mdev, + int (*io_fn)(struct drbd_conf *), + char *why, enum bm_flag flags); extern int drbd_bmio_set_n_write(struct drbd_conf *mdev); extern int drbd_bmio_clear_n_write(struct drbd_conf *mdev); extern void drbd_go_diskless(struct drbd_conf *mdev); extern void drbd_ldev_destroy(struct drbd_conf *mdev); - /* Meta data layout We reserve a 128MB Block (4k aligned) * either at the end of the backing device * or on a separate meta data device. */ -#define MD_RESERVED_SECT (128LU << 11) /* 128 MB, unit sectors */ /* The following numbers are sectors */ -#define MD_AL_OFFSET 8 /* 8 Sectors after start of meta area */ -#define MD_AL_MAX_SIZE 64 /* = 32 kb LOG ~ 3776 extents ~ 14 GB Storage */ -/* Allows up to about 3.8TB */ -#define MD_BM_OFFSET (MD_AL_OFFSET + MD_AL_MAX_SIZE) - -/* Since the smalles IO unit is usually 512 byte */ -#define MD_SECTOR_SHIFT 9 -#define MD_SECTOR_SIZE (1<<MD_SECTOR_SHIFT) - -/* activity log */ -#define AL_EXTENTS_PT ((MD_SECTOR_SIZE-12)/8-1) /* 61 ; Extents per 512B sector */ -#define AL_EXTENT_SHIFT 22 /* One extent represents 4M Storage */ +/* Allows up to about 3.8TB, so if you want more, + * you need to use the "flexible" meta data format. */ +#define MD_RESERVED_SECT (128LU << 11) /* 128 MB, unit sectors */ +#define MD_AL_OFFSET 8 /* 8 Sectors after start of meta area */ +#define MD_AL_SECTORS 64 /* = 32 kB on disk activity log ring buffer */ +#define MD_BM_OFFSET (MD_AL_OFFSET + MD_AL_SECTORS) + +/* we do all meta data IO in 4k blocks */ +#define MD_BLOCK_SHIFT 12 +#define MD_BLOCK_SIZE (1<<MD_BLOCK_SHIFT) + +/* One activity log extent represents 4M of storage */ +#define AL_EXTENT_SHIFT 22 #define AL_EXTENT_SIZE (1<<AL_EXTENT_SHIFT) +/* We could make these currently hardcoded constants configurable + * variables at create-md time (or even re-configurable at runtime?). + * Which will require some more changes to the DRBD "super block" + * and attach code. + * + * updates per transaction: + * This many changes to the active set can be logged with one transaction. + * This number is arbitrary. + * context per transaction: + * This many context extent numbers are logged with each transaction. + * This number is resulting from the transaction block size (4k), the layout + * of the transaction header, and the number of updates per transaction. + * See drbd_actlog.c:struct al_transaction_on_disk + * */ +#define AL_UPDATES_PER_TRANSACTION 64 // arbitrary +#define AL_CONTEXT_PER_TRANSACTION 919 // (4096 - 36 - 6*64)/4 + #if BITS_PER_LONG == 32 #define LN2_BPL 5 #define cpu_to_lel(A) cpu_to_le32(A) @@ -1364,11 +1224,14 @@ struct bm_extent { #define SLEEP_TIME (HZ/10) -#define BM_BLOCK_SHIFT 12 /* 4k per bit */ +/* We do bitmap IO in units of 4k blocks. + * We also still have a hardcoded 4k per bit relation. */ +#define BM_BLOCK_SHIFT 12 /* 4k per bit */ #define BM_BLOCK_SIZE (1<<BM_BLOCK_SHIFT) -/* (9+3) : 512 bytes @ 8 bits; representing 16M storage - * per sector of on disk bitmap */ -#define BM_EXT_SHIFT (BM_BLOCK_SHIFT + MD_SECTOR_SHIFT + 3) /* = 24 */ +/* mostly arbitrarily set the represented size of one bitmap extent, + * aka resync extent, to 16 MiB (which is also 512 Byte worth of bitmap + * at 4k per bit resolution) */ +#define BM_EXT_SHIFT 24 /* 16 MiB per resync extent */ #define BM_EXT_SIZE (1<<BM_EXT_SHIFT) #if (BM_EXT_SHIFT != 24) || (BM_BLOCK_SHIFT != 12) @@ -1436,17 +1299,20 @@ struct bm_extent { #endif #endif -/* Sector shift value for the "hash" functions of tl_hash and ee_hash tables. - * With a value of 8 all IO in one 128K block make it to the same slot of the - * hash table. */ -#define HT_SHIFT 8 -#define DRBD_MAX_BIO_SIZE (1U<<(9+HT_SHIFT)) +/* BIO_MAX_SIZE is 256 * PAGE_CACHE_SIZE, + * so for typical PAGE_CACHE_SIZE of 4k, that is (1<<20) Byte. + * Since we may live in a mixed-platform cluster, + * we limit us to a platform agnostic constant here for now. + * A followup commit may allow even bigger BIO sizes, + * once we thought that through. */ +#define DRBD_MAX_BIO_SIZE (1U << 20) +#if DRBD_MAX_BIO_SIZE > BIO_MAX_SIZE +#error Architecture not supported: DRBD_MAX_BIO_SIZE > BIO_MAX_SIZE +#endif #define DRBD_MAX_BIO_SIZE_SAFE (1U << 12) /* Works always = 4k */ -#define DRBD_MAX_SIZE_H80_PACKET (1U << 15) /* The old header only allows packets up to 32Kib data */ - -/* Number of elements in the app_reads_hash */ -#define APP_R_HSIZE 15 +#define DRBD_MAX_SIZE_H80_PACKET (1U << 15) /* Header 80 only allows packets up to 32KiB data */ +#define DRBD_MAX_BIO_SIZE_P95 (1U << 17) /* Protocol 95 to 99 allows bios up to 128KiB */ extern int drbd_bm_init(struct drbd_conf *mdev); extern int drbd_bm_resize(struct drbd_conf *mdev, sector_t sectors, int set_new_bits); @@ -1468,11 +1334,11 @@ extern int drbd_bm_test_bit(struct drbd_conf *mdev, unsigned long bitnr); extern int drbd_bm_e_weight(struct drbd_conf *mdev, unsigned long enr); extern int drbd_bm_write_page(struct drbd_conf *mdev, unsigned int idx) __must_hold(local); extern int drbd_bm_read(struct drbd_conf *mdev) __must_hold(local); +extern void drbd_bm_mark_for_writeout(struct drbd_conf *mdev, int page_nr); extern int drbd_bm_write(struct drbd_conf *mdev) __must_hold(local); +extern int drbd_bm_write_hinted(struct drbd_conf *mdev) __must_hold(local); extern int drbd_bm_write_all(struct drbd_conf *mdev) __must_hold(local); extern int drbd_bm_write_copy_pages(struct drbd_conf *mdev) __must_hold(local); -extern unsigned long drbd_bm_ALe_set_all(struct drbd_conf *mdev, - unsigned long al_enr); extern size_t drbd_bm_words(struct drbd_conf *mdev); extern unsigned long drbd_bm_bits(struct drbd_conf *mdev); extern sector_t drbd_bm_capacity(struct drbd_conf *mdev); @@ -1497,7 +1363,7 @@ extern void drbd_bm_unlock(struct drbd_conf *mdev); /* drbd_main.c */ extern struct kmem_cache *drbd_request_cache; -extern struct kmem_cache *drbd_ee_cache; /* epoch entries */ +extern struct kmem_cache *drbd_ee_cache; /* peer requests */ extern struct kmem_cache *drbd_bm_ext_cache; /* bitmap extents */ extern struct kmem_cache *drbd_al_ext_cache; /* activity log extents */ extern mempool_t *drbd_request_mempool; @@ -1537,12 +1403,22 @@ extern struct bio *bio_alloc_drbd(gfp_t gfp_mask); extern rwlock_t global_state_lock; -extern struct drbd_conf *drbd_new_device(unsigned int minor); -extern void drbd_free_mdev(struct drbd_conf *mdev); +extern int conn_lowest_minor(struct drbd_tconn *tconn); +enum drbd_ret_code conn_new_minor(struct drbd_tconn *tconn, unsigned int minor, int vnr); +extern void drbd_minor_destroy(struct kref *kref); + +extern int set_resource_options(struct drbd_tconn *tconn, struct res_opts *res_opts); +extern struct drbd_tconn *conn_create(const char *name, struct res_opts *res_opts); +extern void conn_destroy(struct kref *kref); +struct drbd_tconn *conn_get_by_name(const char *name); +extern struct drbd_tconn *conn_get_by_addrs(void *my_addr, int my_addr_len, + void *peer_addr, int peer_addr_len); +extern void conn_free_crypto(struct drbd_tconn *tconn); extern int proc_details; /* drbd_req */ +extern void __drbd_make_request(struct drbd_conf *, struct bio *, unsigned long); extern void drbd_make_request(struct request_queue *q, struct bio *bio); extern int drbd_read_remote(struct drbd_conf *mdev, struct drbd_request *req); extern int drbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bvm, struct bio_vec *bvec); @@ -1550,10 +1426,11 @@ extern int is_valid_ar_handle(struct drbd_request *, sector_t); /* drbd_nl.c */ +extern int drbd_msg_put_info(const char *info); extern void drbd_suspend_io(struct drbd_conf *mdev); extern void drbd_resume_io(struct drbd_conf *mdev); extern char *ppsize(char *buf, unsigned long long size); -extern sector_t drbd_new_dev_size(struct drbd_conf *, struct drbd_backing_dev *, int); +extern sector_t drbd_new_dev_size(struct drbd_conf *, struct drbd_backing_dev *, sector_t, int); enum determine_dev_size { dev_size_error = -1, unchanged = 0, shrunk = 1, grew = 2 }; extern enum determine_dev_size drbd_determine_dev_size(struct drbd_conf *, enum dds_flags) __must_hold(local); extern void resync_after_online_grow(struct drbd_conf *); @@ -1561,13 +1438,14 @@ extern void drbd_reconsider_max_bio_size(struct drbd_conf *mdev); extern enum drbd_state_rv drbd_set_role(struct drbd_conf *mdev, enum drbd_role new_role, int force); -extern enum drbd_disk_state drbd_try_outdate_peer(struct drbd_conf *mdev); -extern void drbd_try_outdate_peer_async(struct drbd_conf *mdev); +extern bool conn_try_outdate_peer(struct drbd_tconn *tconn); +extern void conn_try_outdate_peer_async(struct drbd_tconn *tconn); extern int drbd_khelper(struct drbd_conf *mdev, char *cmd); /* drbd_worker.c */ extern int drbd_worker(struct drbd_thread *thi); -extern int drbd_alter_sa(struct drbd_conf *mdev, int na); +enum drbd_ret_code drbd_resync_after_valid(struct drbd_conf *mdev, int o_minor); +void drbd_resync_after_changed(struct drbd_conf *mdev); extern void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side); extern void resume_next_sg(struct drbd_conf *mdev); extern void suspend_other_sg(struct drbd_conf *mdev); @@ -1576,13 +1454,13 @@ extern int drbd_resync_finished(struct drbd_conf *mdev); extern void *drbd_md_get_buffer(struct drbd_conf *mdev); extern void drbd_md_put_buffer(struct drbd_conf *mdev); extern int drbd_md_sync_page_io(struct drbd_conf *mdev, - struct drbd_backing_dev *bdev, sector_t sector, int rw); -extern void wait_until_done_or_disk_failure(struct drbd_conf *mdev, struct drbd_backing_dev *bdev, - unsigned int *done); -extern void drbd_ov_oos_found(struct drbd_conf*, sector_t, int); + struct drbd_backing_dev *bdev, sector_t sector, int rw); +extern void drbd_ov_out_of_sync_found(struct drbd_conf *, sector_t, int); +extern void wait_until_done_or_force_detached(struct drbd_conf *mdev, + struct drbd_backing_dev *bdev, unsigned int *done); extern void drbd_rs_controller_reset(struct drbd_conf *mdev); -static inline void ov_oos_print(struct drbd_conf *mdev) +static inline void ov_out_of_sync_print(struct drbd_conf *mdev) { if (mdev->ov_last_oos_size) { dev_err(DEV, "Out of sync: start=%llu, size=%lu (sectors)\n", @@ -1594,97 +1472,102 @@ static inline void ov_oos_print(struct drbd_conf *mdev) extern void drbd_csum_bio(struct drbd_conf *, struct crypto_hash *, struct bio *, void *); -extern void drbd_csum_ee(struct drbd_conf *, struct crypto_hash *, struct drbd_epoch_entry *, void *); +extern void drbd_csum_ee(struct drbd_conf *, struct crypto_hash *, + struct drbd_peer_request *, void *); /* worker callbacks */ -extern int w_req_cancel_conflict(struct drbd_conf *, struct drbd_work *, int); -extern int w_read_retry_remote(struct drbd_conf *, struct drbd_work *, int); -extern int w_e_end_data_req(struct drbd_conf *, struct drbd_work *, int); -extern int w_e_end_rsdata_req(struct drbd_conf *, struct drbd_work *, int); -extern int w_e_end_csum_rs_req(struct drbd_conf *, struct drbd_work *, int); -extern int w_e_end_ov_reply(struct drbd_conf *, struct drbd_work *, int); -extern int w_e_end_ov_req(struct drbd_conf *, struct drbd_work *, int); -extern int w_ov_finished(struct drbd_conf *, struct drbd_work *, int); -extern int w_resync_timer(struct drbd_conf *, struct drbd_work *, int); -extern int w_resume_next_sg(struct drbd_conf *, struct drbd_work *, int); -extern int w_send_write_hint(struct drbd_conf *, struct drbd_work *, int); -extern int w_send_dblock(struct drbd_conf *, struct drbd_work *, int); -extern int w_send_barrier(struct drbd_conf *, struct drbd_work *, int); -extern int w_send_read_req(struct drbd_conf *, struct drbd_work *, int); -extern int w_prev_work_done(struct drbd_conf *, struct drbd_work *, int); -extern int w_e_reissue(struct drbd_conf *, struct drbd_work *, int); -extern int w_restart_disk_io(struct drbd_conf *, struct drbd_work *, int); -extern int w_send_oos(struct drbd_conf *, struct drbd_work *, int); -extern int w_start_resync(struct drbd_conf *, struct drbd_work *, int); +extern int w_e_end_data_req(struct drbd_work *, int); +extern int w_e_end_rsdata_req(struct drbd_work *, int); +extern int w_e_end_csum_rs_req(struct drbd_work *, int); +extern int w_e_end_ov_reply(struct drbd_work *, int); +extern int w_e_end_ov_req(struct drbd_work *, int); +extern int w_ov_finished(struct drbd_work *, int); +extern int w_resync_timer(struct drbd_work *, int); +extern int w_send_write_hint(struct drbd_work *, int); +extern int w_make_resync_request(struct drbd_work *, int); +extern int w_send_dblock(struct drbd_work *, int); +extern int w_send_read_req(struct drbd_work *, int); +extern int w_prev_work_done(struct drbd_work *, int); +extern int w_e_reissue(struct drbd_work *, int); +extern int w_restart_disk_io(struct drbd_work *, int); +extern int w_send_out_of_sync(struct drbd_work *, int); +extern int w_start_resync(struct drbd_work *, int); extern void resync_timer_fn(unsigned long data); extern void start_resync_timer_fn(unsigned long data); /* drbd_receiver.c */ extern int drbd_rs_should_slow_down(struct drbd_conf *mdev, sector_t sector); -extern int drbd_submit_ee(struct drbd_conf *mdev, struct drbd_epoch_entry *e, - const unsigned rw, const int fault_type); -extern int drbd_release_ee(struct drbd_conf *mdev, struct list_head *list); -extern struct drbd_epoch_entry *drbd_alloc_ee(struct drbd_conf *mdev, - u64 id, - sector_t sector, - unsigned int data_size, - gfp_t gfp_mask) __must_hold(local); -extern void drbd_free_some_ee(struct drbd_conf *mdev, struct drbd_epoch_entry *e, - int is_net); -#define drbd_free_ee(m,e) drbd_free_some_ee(m, e, 0) -#define drbd_free_net_ee(m,e) drbd_free_some_ee(m, e, 1) -extern void drbd_wait_ee_list_empty(struct drbd_conf *mdev, - struct list_head *head); -extern void _drbd_wait_ee_list_empty(struct drbd_conf *mdev, - struct list_head *head); +extern int drbd_submit_peer_request(struct drbd_conf *, + struct drbd_peer_request *, const unsigned, + const int); +extern int drbd_free_peer_reqs(struct drbd_conf *, struct list_head *); +extern struct drbd_peer_request *drbd_alloc_peer_req(struct drbd_conf *, u64, + sector_t, unsigned int, + gfp_t) __must_hold(local); +extern void __drbd_free_peer_req(struct drbd_conf *, struct drbd_peer_request *, + int); +#define drbd_free_peer_req(m,e) __drbd_free_peer_req(m, e, 0) +#define drbd_free_net_peer_req(m,e) __drbd_free_peer_req(m, e, 1) +extern struct page *drbd_alloc_pages(struct drbd_conf *, unsigned int, bool); extern void drbd_set_recv_tcq(struct drbd_conf *mdev, int tcq_enabled); extern void _drbd_clear_done_ee(struct drbd_conf *mdev, struct list_head *to_be_freed); -extern void drbd_flush_workqueue(struct drbd_conf *mdev); -extern void drbd_free_tl_hash(struct drbd_conf *mdev); +extern void conn_flush_workqueue(struct drbd_tconn *tconn); +extern int drbd_connected(struct drbd_conf *mdev); +static inline void drbd_flush_workqueue(struct drbd_conf *mdev) +{ + conn_flush_workqueue(mdev->tconn); +} -/* yes, there is kernel_setsockopt, but only since 2.6.18. we don't need to - * mess with get_fs/set_fs, we know we are KERNEL_DS always. */ +/* Yes, there is kernel_setsockopt, but only since 2.6.18. + * So we have our own copy of it here. */ static inline int drbd_setsockopt(struct socket *sock, int level, int optname, - char __user *optval, int optlen) + char *optval, int optlen) { + mm_segment_t oldfs = get_fs(); + char __user *uoptval; int err; + + uoptval = (char __user __force *)optval; + + set_fs(KERNEL_DS); if (level == SOL_SOCKET) - err = sock_setsockopt(sock, level, optname, optval, optlen); + err = sock_setsockopt(sock, level, optname, uoptval, optlen); else - err = sock->ops->setsockopt(sock, level, optname, optval, + err = sock->ops->setsockopt(sock, level, optname, uoptval, optlen); + set_fs(oldfs); return err; } static inline void drbd_tcp_cork(struct socket *sock) { - int __user val = 1; + int val = 1; (void) drbd_setsockopt(sock, SOL_TCP, TCP_CORK, - (char __user *)&val, sizeof(val)); + (char*)&val, sizeof(val)); } static inline void drbd_tcp_uncork(struct socket *sock) { - int __user val = 0; + int val = 0; (void) drbd_setsockopt(sock, SOL_TCP, TCP_CORK, - (char __user *)&val, sizeof(val)); + (char*)&val, sizeof(val)); } static inline void drbd_tcp_nodelay(struct socket *sock) { - int __user val = 1; + int val = 1; (void) drbd_setsockopt(sock, SOL_TCP, TCP_NODELAY, - (char __user *)&val, sizeof(val)); + (char*)&val, sizeof(val)); } static inline void drbd_tcp_quickack(struct socket *sock) { - int __user val = 2; + int val = 2; (void) drbd_setsockopt(sock, SOL_TCP, TCP_QUICKACK, - (char __user *)&val, sizeof(val)); + (char*)&val, sizeof(val)); } -void drbd_bump_write_ordering(struct drbd_conf *mdev, enum write_ordering_e wo); +void drbd_bump_write_ordering(struct drbd_tconn *tconn, enum write_ordering_e wo); /* drbd_proc.c */ extern struct proc_dir_entry *drbd_proc; @@ -1693,8 +1576,8 @@ extern const char *drbd_conn_str(enum drbd_conns s); extern const char *drbd_role_str(enum drbd_role s); /* drbd_actlog.c */ -extern void drbd_al_begin_io(struct drbd_conf *mdev, sector_t sector); -extern void drbd_al_complete_io(struct drbd_conf *mdev, sector_t sector); +extern void drbd_al_begin_io(struct drbd_conf *mdev, struct drbd_interval *i); +extern void drbd_al_complete_io(struct drbd_conf *mdev, struct drbd_interval *i); extern void drbd_rs_complete_io(struct drbd_conf *mdev, sector_t sector); extern int drbd_rs_begin_io(struct drbd_conf *mdev, sector_t sector); extern int drbd_try_rs_begin_io(struct drbd_conf *mdev, sector_t sector); @@ -1702,7 +1585,6 @@ extern void drbd_rs_cancel_all(struct drbd_conf *mdev); extern int drbd_rs_del_all(struct drbd_conf *mdev); extern void drbd_rs_failed_io(struct drbd_conf *mdev, sector_t sector, int size); -extern int drbd_al_read_log(struct drbd_conf *mdev, struct drbd_backing_dev *); extern void drbd_advance_rs_marks(struct drbd_conf *mdev, unsigned long still_to_go); extern void __drbd_set_in_sync(struct drbd_conf *mdev, sector_t sector, int size, const char *file, const unsigned int line); @@ -1712,73 +1594,24 @@ extern int __drbd_set_out_of_sync(struct drbd_conf *mdev, sector_t sector, int size, const char *file, const unsigned int line); #define drbd_set_out_of_sync(mdev, sector, size) \ __drbd_set_out_of_sync(mdev, sector, size, __FILE__, __LINE__) -extern void drbd_al_apply_to_bm(struct drbd_conf *mdev); extern void drbd_al_shrink(struct drbd_conf *mdev); - /* drbd_nl.c */ - -void drbd_nl_cleanup(void); -int __init drbd_nl_init(void); -void drbd_bcast_state(struct drbd_conf *mdev, union drbd_state); -void drbd_bcast_sync_progress(struct drbd_conf *mdev); -void drbd_bcast_ee(struct drbd_conf *mdev, - const char *reason, const int dgs, - const char* seen_hash, const char* calc_hash, - const struct drbd_epoch_entry* e); - - -/** - * DOC: DRBD State macros - * - * These macros are used to express state changes in easily readable form. - * - * The NS macros expand to a mask and a value, that can be bit ored onto the - * current state as soon as the spinlock (req_lock) was taken. - * - * The _NS macros are used for state functions that get called with the - * spinlock. These macros expand directly to the new state value. - * - * Besides the basic forms NS() and _NS() additional _?NS[23] are defined - * to express state changes that affect more than one aspect of the state. - * - * E.g. NS2(conn, C_CONNECTED, peer, R_SECONDARY) - * Means that the network connection was established and that the peer - * is in secondary role. - */ -#define role_MASK R_MASK -#define peer_MASK R_MASK -#define disk_MASK D_MASK -#define pdsk_MASK D_MASK -#define conn_MASK C_MASK -#define susp_MASK 1 -#define user_isp_MASK 1 -#define aftr_isp_MASK 1 -#define susp_nod_MASK 1 -#define susp_fen_MASK 1 - -#define NS(T, S) \ - ({ union drbd_state mask; mask.i = 0; mask.T = T##_MASK; mask; }), \ - ({ union drbd_state val; val.i = 0; val.T = (S); val; }) -#define NS2(T1, S1, T2, S2) \ - ({ union drbd_state mask; mask.i = 0; mask.T1 = T1##_MASK; \ - mask.T2 = T2##_MASK; mask; }), \ - ({ union drbd_state val; val.i = 0; val.T1 = (S1); \ - val.T2 = (S2); val; }) -#define NS3(T1, S1, T2, S2, T3, S3) \ - ({ union drbd_state mask; mask.i = 0; mask.T1 = T1##_MASK; \ - mask.T2 = T2##_MASK; mask.T3 = T3##_MASK; mask; }), \ - ({ union drbd_state val; val.i = 0; val.T1 = (S1); \ - val.T2 = (S2); val.T3 = (S3); val; }) - -#define _NS(D, T, S) \ - D, ({ union drbd_state __ns; __ns.i = D->state.i; __ns.T = (S); __ns; }) -#define _NS2(D, T1, S1, T2, S2) \ - D, ({ union drbd_state __ns; __ns.i = D->state.i; __ns.T1 = (S1); \ - __ns.T2 = (S2); __ns; }) -#define _NS3(D, T1, S1, T2, S2, T3, S3) \ - D, ({ union drbd_state __ns; __ns.i = D->state.i; __ns.T1 = (S1); \ - __ns.T2 = (S2); __ns.T3 = (S3); __ns; }) +/* state info broadcast */ +struct sib_info { + enum drbd_state_info_bcast_reason sib_reason; + union { + struct { + char *helper_name; + unsigned helper_exit_code; + }; + struct { + union drbd_state os; + union drbd_state ns; + }; + }; +}; +void drbd_bcast_event(struct drbd_conf *mdev, const struct sib_info *sib); /* * inline helper functions @@ -1795,9 +1628,10 @@ static inline struct page *page_chain_next(struct page *page) #define page_chain_for_each_safe(page, n) \ for (; page && ({ n = page_chain_next(page); 1; }); page = n) -static inline int drbd_ee_has_active_page(struct drbd_epoch_entry *e) + +static inline int drbd_peer_req_has_active_page(struct drbd_peer_request *peer_req) { - struct page *page = e->pages; + struct page *page = peer_req->pages; page_chain_for_each(page) { if (page_count(page) > 1) return 1; @@ -1805,18 +1639,6 @@ static inline int drbd_ee_has_active_page(struct drbd_epoch_entry *e) return 0; } -static inline void drbd_state_lock(struct drbd_conf *mdev) -{ - wait_event(mdev->misc_wait, - !test_and_set_bit(CLUSTER_ST_CHANGE, &mdev->flags)); -} - -static inline void drbd_state_unlock(struct drbd_conf *mdev) -{ - clear_bit(CLUSTER_ST_CHANGE, &mdev->flags); - wake_up(&mdev->misc_wait); -} - static inline enum drbd_state_rv _drbd_set_state(struct drbd_conf *mdev, union drbd_state ns, enum chg_state_flags flags, struct completion *done) @@ -1830,48 +1652,71 @@ _drbd_set_state(struct drbd_conf *mdev, union drbd_state ns, return rv; } -/** - * drbd_request_state() - Reqest a state change - * @mdev: DRBD device. - * @mask: mask of state bits to change. - * @val: value of new state bits. - * - * This is the most graceful way of requesting a state change. It is verbose - * quite verbose in case the state change is not possible, and all those - * state changes are globally serialized. - */ -static inline int drbd_request_state(struct drbd_conf *mdev, - union drbd_state mask, - union drbd_state val) +static inline union drbd_state drbd_read_state(struct drbd_conf *mdev) { - return _drbd_request_state(mdev, mask, val, CS_VERBOSE + CS_ORDERED); + union drbd_state rv; + + rv.i = mdev->state.i; + rv.susp = mdev->tconn->susp; + rv.susp_nod = mdev->tconn->susp_nod; + rv.susp_fen = mdev->tconn->susp_fen; + + return rv; } enum drbd_force_detach_flags { - DRBD_IO_ERROR, + DRBD_READ_ERROR, + DRBD_WRITE_ERROR, DRBD_META_IO_ERROR, DRBD_FORCE_DETACH, }; #define __drbd_chk_io_error(m,f) __drbd_chk_io_error_(m,f, __func__) static inline void __drbd_chk_io_error_(struct drbd_conf *mdev, - enum drbd_force_detach_flags forcedetach, + enum drbd_force_detach_flags df, const char *where) { - switch (mdev->ldev->dc.on_io_error) { - case EP_PASS_ON: - if (forcedetach == DRBD_IO_ERROR) { + enum drbd_io_error_p ep; + + rcu_read_lock(); + ep = rcu_dereference(mdev->ldev->disk_conf)->on_io_error; + rcu_read_unlock(); + switch (ep) { + case EP_PASS_ON: /* FIXME would this be better named "Ignore"? */ + if (df == DRBD_READ_ERROR || df == DRBD_WRITE_ERROR) { if (__ratelimit(&drbd_ratelimit_state)) dev_err(DEV, "Local IO failed in %s.\n", where); if (mdev->state.disk > D_INCONSISTENT) _drbd_set_state(_NS(mdev, disk, D_INCONSISTENT), CS_HARD, NULL); break; } - /* NOTE fall through to detach case if forcedetach set */ + /* NOTE fall through for DRBD_META_IO_ERROR or DRBD_FORCE_DETACH */ case EP_DETACH: case EP_CALL_HELPER: + /* Remember whether we saw a READ or WRITE error. + * + * Recovery of the affected area for WRITE failure is covered + * by the activity log. + * READ errors may fall outside that area though. Certain READ + * errors can be "healed" by writing good data to the affected + * blocks, which triggers block re-allocation in lower layers. + * + * If we can not write the bitmap after a READ error, + * we may need to trigger a full sync (see w_go_diskless()). + * + * Force-detach is not really an IO error, but rather a + * desperate measure to try to deal with a completely + * unresponsive lower level IO stack. + * Still it should be treated as a WRITE error. + * + * Meta IO error is always WRITE error: + * we read meta data only once during attach, + * which will fail in case of errors. + */ set_bit(WAS_IO_ERROR, &mdev->flags); - if (forcedetach == DRBD_FORCE_DETACH) + if (df == DRBD_READ_ERROR) + set_bit(WAS_READ_ERROR, &mdev->flags); + if (df == DRBD_FORCE_DETACH) set_bit(FORCE_DETACH, &mdev->flags); if (mdev->state.disk > D_FAILED) { _drbd_set_state(_NS(mdev, disk, D_FAILED), CS_HARD, NULL); @@ -1896,9 +1741,9 @@ static inline void drbd_chk_io_error_(struct drbd_conf *mdev, { if (error) { unsigned long flags; - spin_lock_irqsave(&mdev->req_lock, flags); + spin_lock_irqsave(&mdev->tconn->req_lock, flags); __drbd_chk_io_error_(mdev, forcedetach, where); - spin_unlock_irqrestore(&mdev->req_lock, flags); + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); } } @@ -1910,9 +1755,9 @@ static inline void drbd_chk_io_error_(struct drbd_conf *mdev, * BTW, for internal meta data, this happens to be the maximum capacity * we could agree upon with our peer node. */ -static inline sector_t drbd_md_first_sector(struct drbd_backing_dev *bdev) +static inline sector_t _drbd_md_first_sector(int meta_dev_idx, struct drbd_backing_dev *bdev) { - switch (bdev->dc.meta_dev_idx) { + switch (meta_dev_idx) { case DRBD_MD_INDEX_INTERNAL: case DRBD_MD_INDEX_FLEX_INT: return bdev->md.md_offset + bdev->md.bm_offset; @@ -1922,13 +1767,30 @@ static inline sector_t drbd_md_first_sector(struct drbd_backing_dev *bdev) } } +static inline sector_t drbd_md_first_sector(struct drbd_backing_dev *bdev) +{ + int meta_dev_idx; + + rcu_read_lock(); + meta_dev_idx = rcu_dereference(bdev->disk_conf)->meta_dev_idx; + rcu_read_unlock(); + + return _drbd_md_first_sector(meta_dev_idx, bdev); +} + /** * drbd_md_last_sector() - Return the last sector number of the meta data area * @bdev: Meta data block device. */ static inline sector_t drbd_md_last_sector(struct drbd_backing_dev *bdev) { - switch (bdev->dc.meta_dev_idx) { + int meta_dev_idx; + + rcu_read_lock(); + meta_dev_idx = rcu_dereference(bdev->disk_conf)->meta_dev_idx; + rcu_read_unlock(); + + switch (meta_dev_idx) { case DRBD_MD_INDEX_INTERNAL: case DRBD_MD_INDEX_FLEX_INT: return bdev->md.md_offset + MD_AL_OFFSET - 1; @@ -1956,12 +1818,18 @@ static inline sector_t drbd_get_capacity(struct block_device *bdev) static inline sector_t drbd_get_max_capacity(struct drbd_backing_dev *bdev) { sector_t s; - switch (bdev->dc.meta_dev_idx) { + int meta_dev_idx; + + rcu_read_lock(); + meta_dev_idx = rcu_dereference(bdev->disk_conf)->meta_dev_idx; + rcu_read_unlock(); + + switch (meta_dev_idx) { case DRBD_MD_INDEX_INTERNAL: case DRBD_MD_INDEX_FLEX_INT: s = drbd_get_capacity(bdev->backing_bdev) ? min_t(sector_t, DRBD_MAX_SECTORS_FLEX, - drbd_md_first_sector(bdev)) + _drbd_md_first_sector(meta_dev_idx, bdev)) : 0; break; case DRBD_MD_INDEX_FLEX_EXT: @@ -1987,9 +1855,15 @@ static inline sector_t drbd_get_max_capacity(struct drbd_backing_dev *bdev) static inline sector_t drbd_md_ss__(struct drbd_conf *mdev, struct drbd_backing_dev *bdev) { - switch (bdev->dc.meta_dev_idx) { + int meta_dev_idx; + + rcu_read_lock(); + meta_dev_idx = rcu_dereference(bdev->disk_conf)->meta_dev_idx; + rcu_read_unlock(); + + switch (meta_dev_idx) { default: /* external, some index */ - return MD_RESERVED_SECT * bdev->dc.meta_dev_idx; + return MD_RESERVED_SECT * meta_dev_idx; case DRBD_MD_INDEX_INTERNAL: /* with drbd08, internal meta data is always "flexible" */ case DRBD_MD_INDEX_FLEX_INT: @@ -2015,9 +1889,8 @@ drbd_queue_work_front(struct drbd_work_queue *q, struct drbd_work *w) unsigned long flags; spin_lock_irqsave(&q->q_lock, flags); list_add(&w->list, &q->q); - up(&q->s); /* within the spinlock, - see comment near end of drbd_worker() */ spin_unlock_irqrestore(&q->q_lock, flags); + wake_up(&q->q_wait); } static inline void @@ -2026,41 +1899,35 @@ drbd_queue_work(struct drbd_work_queue *q, struct drbd_work *w) unsigned long flags; spin_lock_irqsave(&q->q_lock, flags); list_add_tail(&w->list, &q->q); - up(&q->s); /* within the spinlock, - see comment near end of drbd_worker() */ spin_unlock_irqrestore(&q->q_lock, flags); + wake_up(&q->q_wait); } -static inline void wake_asender(struct drbd_conf *mdev) -{ - if (test_bit(SIGNAL_ASENDER, &mdev->flags)) - force_sig(DRBD_SIG, mdev->asender.task); -} - -static inline void request_ping(struct drbd_conf *mdev) +static inline void wake_asender(struct drbd_tconn *tconn) { - set_bit(SEND_PING, &mdev->flags); - wake_asender(mdev); + if (test_bit(SIGNAL_ASENDER, &tconn->flags)) + force_sig(DRBD_SIG, tconn->asender.task); } -static inline int drbd_send_short_cmd(struct drbd_conf *mdev, - enum drbd_packets cmd) +static inline void request_ping(struct drbd_tconn *tconn) { - struct p_header80 h; - return drbd_send_cmd(mdev, USE_DATA_SOCKET, cmd, &h, sizeof(h)); + set_bit(SEND_PING, &tconn->flags); + wake_asender(tconn); } -static inline int drbd_send_ping(struct drbd_conf *mdev) -{ - struct p_header80 h; - return drbd_send_cmd(mdev, USE_META_SOCKET, P_PING, &h, sizeof(h)); -} +extern void *conn_prepare_command(struct drbd_tconn *, struct drbd_socket *); +extern void *drbd_prepare_command(struct drbd_conf *, struct drbd_socket *); +extern int conn_send_command(struct drbd_tconn *, struct drbd_socket *, + enum drbd_packet, unsigned int, void *, + unsigned int); +extern int drbd_send_command(struct drbd_conf *, struct drbd_socket *, + enum drbd_packet, unsigned int, void *, + unsigned int); -static inline int drbd_send_ping_ack(struct drbd_conf *mdev) -{ - struct p_header80 h; - return drbd_send_cmd(mdev, USE_META_SOCKET, P_PING_ACK, &h, sizeof(h)); -} +extern int drbd_send_ping(struct drbd_tconn *tconn); +extern int drbd_send_ping_ack(struct drbd_tconn *tconn); +extern int drbd_send_state_req(struct drbd_conf *, union drbd_state, union drbd_state); +extern int conn_send_state_req(struct drbd_tconn *, union drbd_state, union drbd_state); static inline void drbd_thread_stop(struct drbd_thread *thi) { @@ -2082,21 +1949,21 @@ static inline void drbd_thread_restart_nowait(struct drbd_thread *thi) * or implicit barrier packets as necessary. * increased: * w_send_barrier - * _req_mod(req, queue_for_net_write or queue_for_net_read); + * _req_mod(req, QUEUE_FOR_NET_WRITE or QUEUE_FOR_NET_READ); * it is much easier and equally valid to count what we queue for the * worker, even before it actually was queued or send. * (drbd_make_request_common; recovery path on read io-error) * decreased: * got_BarrierAck (respective tl_clear, tl_clear_barrier) - * _req_mod(req, data_received) + * _req_mod(req, DATA_RECEIVED) * [from receive_DataReply] - * _req_mod(req, write_acked_by_peer or recv_acked_by_peer or neg_acked) + * _req_mod(req, WRITE_ACKED_BY_PEER or RECV_ACKED_BY_PEER or NEG_ACKED) * [from got_BlockAck (P_WRITE_ACK, P_RECV_ACK)] * for some reason it is NOT decreased in got_NegAck, * but in the resulting cleanup code from report_params. * we should try to remember the reason for that... - * _req_mod(req, send_failed or send_canceled) - * _req_mod(req, connection_lost_while_pending) + * _req_mod(req, SEND_FAILED or SEND_CANCELED) + * _req_mod(req, CONNECTION_LOST_WHILE_PENDING) * [from tl_clear_barrier] */ static inline void inc_ap_pending(struct drbd_conf *mdev) @@ -2104,17 +1971,19 @@ static inline void inc_ap_pending(struct drbd_conf *mdev) atomic_inc(&mdev->ap_pending_cnt); } -#define ERR_IF_CNT_IS_NEGATIVE(which) \ - if (atomic_read(&mdev->which) < 0) \ +#define ERR_IF_CNT_IS_NEGATIVE(which, func, line) \ + if (atomic_read(&mdev->which) < 0) \ dev_err(DEV, "in %s:%d: " #which " = %d < 0 !\n", \ - __func__ , __LINE__ , \ - atomic_read(&mdev->which)) + func, line, \ + atomic_read(&mdev->which)) -#define dec_ap_pending(mdev) do { \ - typecheck(struct drbd_conf *, mdev); \ - if (atomic_dec_and_test(&mdev->ap_pending_cnt)) \ - wake_up(&mdev->misc_wait); \ - ERR_IF_CNT_IS_NEGATIVE(ap_pending_cnt); } while (0) +#define dec_ap_pending(mdev) _dec_ap_pending(mdev, __FUNCTION__, __LINE__) +static inline void _dec_ap_pending(struct drbd_conf *mdev, const char *func, int line) +{ + if (atomic_dec_and_test(&mdev->ap_pending_cnt)) + wake_up(&mdev->misc_wait); + ERR_IF_CNT_IS_NEGATIVE(ap_pending_cnt, func, line); +} /* counts how many resync-related answers we still expect from the peer * increase decrease @@ -2127,10 +1996,12 @@ static inline void inc_rs_pending(struct drbd_conf *mdev) atomic_inc(&mdev->rs_pending_cnt); } -#define dec_rs_pending(mdev) do { \ - typecheck(struct drbd_conf *, mdev); \ - atomic_dec(&mdev->rs_pending_cnt); \ - ERR_IF_CNT_IS_NEGATIVE(rs_pending_cnt); } while (0) +#define dec_rs_pending(mdev) _dec_rs_pending(mdev, __FUNCTION__, __LINE__) +static inline void _dec_rs_pending(struct drbd_conf *mdev, const char *func, int line) +{ + atomic_dec(&mdev->rs_pending_cnt); + ERR_IF_CNT_IS_NEGATIVE(rs_pending_cnt, func, line); +} /* counts how many answers we still need to send to the peer. * increased on @@ -2146,38 +2017,18 @@ static inline void inc_unacked(struct drbd_conf *mdev) atomic_inc(&mdev->unacked_cnt); } -#define dec_unacked(mdev) do { \ - typecheck(struct drbd_conf *, mdev); \ - atomic_dec(&mdev->unacked_cnt); \ - ERR_IF_CNT_IS_NEGATIVE(unacked_cnt); } while (0) - -#define sub_unacked(mdev, n) do { \ - typecheck(struct drbd_conf *, mdev); \ - atomic_sub(n, &mdev->unacked_cnt); \ - ERR_IF_CNT_IS_NEGATIVE(unacked_cnt); } while (0) - - -static inline void put_net_conf(struct drbd_conf *mdev) +#define dec_unacked(mdev) _dec_unacked(mdev, __FUNCTION__, __LINE__) +static inline void _dec_unacked(struct drbd_conf *mdev, const char *func, int line) { - if (atomic_dec_and_test(&mdev->net_cnt)) - wake_up(&mdev->net_cnt_wait); + atomic_dec(&mdev->unacked_cnt); + ERR_IF_CNT_IS_NEGATIVE(unacked_cnt, func, line); } -/** - * get_net_conf() - Increase ref count on mdev->net_conf; Returns 0 if nothing there - * @mdev: DRBD device. - * - * You have to call put_net_conf() when finished working with mdev->net_conf. - */ -static inline int get_net_conf(struct drbd_conf *mdev) +#define sub_unacked(mdev, n) _sub_unacked(mdev, n, __FUNCTION__, __LINE__) +static inline void _sub_unacked(struct drbd_conf *mdev, int n, const char *func, int line) { - int have_net_conf; - - atomic_inc(&mdev->net_cnt); - have_net_conf = mdev->state.conn >= C_UNCONNECTED; - if (!have_net_conf) - put_net_conf(mdev); - return have_net_conf; + atomic_sub(n, &mdev->unacked_cnt); + ERR_IF_CNT_IS_NEGATIVE(unacked_cnt, func, line); } /** @@ -2281,17 +2132,20 @@ static inline void drbd_get_syncer_progress(struct drbd_conf *mdev, * maybe re-implement using semaphores? */ static inline int drbd_get_max_buffers(struct drbd_conf *mdev) { - int mxb = 1000000; /* arbitrary limit on open requests */ - if (get_net_conf(mdev)) { - mxb = mdev->net_conf->max_buffers; - put_net_conf(mdev); - } + struct net_conf *nc; + int mxb; + + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + mxb = nc ? nc->max_buffers : 1000000; /* arbitrary limit on open requests */ + rcu_read_unlock(); + return mxb; } static inline int drbd_state_is_stable(struct drbd_conf *mdev) { - union drbd_state s = mdev->state; + union drbd_dev_state s = mdev->state; /* DO NOT add a default clause, we want the compiler to warn us * for any newly introduced state we may have forgotten to add here */ @@ -2325,7 +2179,7 @@ static inline int drbd_state_is_stable(struct drbd_conf *mdev) /* Allow IO in BM exchange states with new protocols */ case C_WF_BITMAP_S: - if (mdev->agreed_pro_version < 96) + if (mdev->tconn->agreed_pro_version < 96) return 0; break; @@ -2347,7 +2201,7 @@ static inline int drbd_state_is_stable(struct drbd_conf *mdev) /* disk state is stable as well. */ break; - /* no new io accepted during tansitional states */ + /* no new io accepted during transitional states */ case D_ATTACHING: case D_NEGOTIATING: case D_UNKNOWN: @@ -2359,16 +2213,18 @@ static inline int drbd_state_is_stable(struct drbd_conf *mdev) return 1; } -static inline int is_susp(union drbd_state s) +static inline int drbd_suspended(struct drbd_conf *mdev) { - return s.susp || s.susp_nod || s.susp_fen; + struct drbd_tconn *tconn = mdev->tconn; + + return tconn->susp || tconn->susp_fen || tconn->susp_nod; } static inline bool may_inc_ap_bio(struct drbd_conf *mdev) { int mxb = drbd_get_max_buffers(mdev); - if (is_susp(mdev->state)) + if (drbd_suspended(mdev)) return false; if (test_bit(SUSPEND_IO, &mdev->flags)) return false; @@ -2390,30 +2246,30 @@ static inline bool may_inc_ap_bio(struct drbd_conf *mdev) return true; } -static inline bool inc_ap_bio_cond(struct drbd_conf *mdev, int count) +static inline bool inc_ap_bio_cond(struct drbd_conf *mdev) { bool rv = false; - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); rv = may_inc_ap_bio(mdev); if (rv) - atomic_add(count, &mdev->ap_bio_cnt); - spin_unlock_irq(&mdev->req_lock); + atomic_inc(&mdev->ap_bio_cnt); + spin_unlock_irq(&mdev->tconn->req_lock); return rv; } -static inline void inc_ap_bio(struct drbd_conf *mdev, int count) +static inline void inc_ap_bio(struct drbd_conf *mdev) { /* we wait here * as long as the device is suspended * until the bitmap is no longer on the fly during connection - * handshake as long as we would exeed the max_buffer limit. + * handshake as long as we would exceed the max_buffer limit. * * to avoid races with the reconnect code, * we need to atomic_inc within the spinlock. */ - wait_event(mdev->misc_wait, inc_ap_bio_cond(mdev, count)); + wait_event(mdev->misc_wait, inc_ap_bio_cond(mdev)); } static inline void dec_ap_bio(struct drbd_conf *mdev) @@ -2425,7 +2281,7 @@ static inline void dec_ap_bio(struct drbd_conf *mdev) if (ap_bio == 0 && test_bit(BITMAP_IO, &mdev->flags)) { if (!test_and_set_bit(BITMAP_IO_QUEUED, &mdev->flags)) - drbd_queue_work(&mdev->data.work, &mdev->bm_io_work.w); + drbd_queue_work(&mdev->tconn->sender_work, &mdev->bm_io_work.w); } /* this currently does wake_up for every dec_ap_bio! @@ -2435,6 +2291,12 @@ static inline void dec_ap_bio(struct drbd_conf *mdev) wake_up(&mdev->misc_wait); } +static inline bool verify_can_do_stop_sector(struct drbd_conf *mdev) +{ + return mdev->tconn->agreed_pro_version >= 97 && + mdev->tconn->agreed_pro_version != 100; +} + static inline int drbd_set_ed_uuid(struct drbd_conf *mdev, u64 val) { int changed = mdev->ed_uuid != val; @@ -2442,40 +2304,6 @@ static inline int drbd_set_ed_uuid(struct drbd_conf *mdev, u64 val) return changed; } -static inline int seq_cmp(u32 a, u32 b) -{ - /* we assume wrap around at 32bit. - * for wrap around at 24bit (old atomic_t), - * we'd have to - * a <<= 8; b <<= 8; - */ - return (s32)(a) - (s32)(b); -} -#define seq_lt(a, b) (seq_cmp((a), (b)) < 0) -#define seq_gt(a, b) (seq_cmp((a), (b)) > 0) -#define seq_ge(a, b) (seq_cmp((a), (b)) >= 0) -#define seq_le(a, b) (seq_cmp((a), (b)) <= 0) -/* CAUTION: please no side effects in arguments! */ -#define seq_max(a, b) ((u32)(seq_gt((a), (b)) ? (a) : (b))) - -static inline void update_peer_seq(struct drbd_conf *mdev, unsigned int new_seq) -{ - unsigned int m; - spin_lock(&mdev->peer_seq_lock); - m = seq_max(mdev->peer_seq, new_seq); - mdev->peer_seq = m; - spin_unlock(&mdev->peer_seq_lock); - if (m == new_seq) - wake_up(&mdev->seq_wait); -} - -static inline void drbd_update_congested(struct drbd_conf *mdev) -{ - struct sock *sk = mdev->data.socket->sk; - if (sk->sk_wmem_queued > sk->sk_sndbuf * 4 / 5) - set_bit(NET_CONGESTED, &mdev->flags); -} - static inline int drbd_queue_order_type(struct drbd_conf *mdev) { /* sorry, we currently have no working implementation @@ -2490,10 +2318,15 @@ static inline void drbd_md_flush(struct drbd_conf *mdev) { int r; + if (mdev->ldev == NULL) { + dev_warn(DEV, "mdev->ldev == NULL in drbd_md_flush\n"); + return; + } + if (test_bit(MD_NO_FUA, &mdev->flags)) return; - r = blkdev_issue_flush(mdev->ldev->md_bdev, GFP_KERNEL, NULL); + r = blkdev_issue_flush(mdev->ldev->md_bdev, GFP_NOIO, NULL); if (r) { set_bit(MD_NO_FUA, &mdev->flags); dev_err(DEV, "meta data flush failed with status %d, disabling md-flushes\n", r); diff --git a/drivers/block/drbd/drbd_interval.c b/drivers/block/drbd/drbd_interval.c new file mode 100644 index 000000000000..89c497c630b4 --- /dev/null +++ b/drivers/block/drbd/drbd_interval.c @@ -0,0 +1,207 @@ +#include <asm/bug.h> +#include <linux/rbtree_augmented.h> +#include "drbd_interval.h" + +/** + * interval_end - return end of @node + */ +static inline +sector_t interval_end(struct rb_node *node) +{ + struct drbd_interval *this = rb_entry(node, struct drbd_interval, rb); + return this->end; +} + +/** + * compute_subtree_last - compute end of @node + * + * The end of an interval is the highest (start + (size >> 9)) value of this + * node and of its children. Called for @node and its parents whenever the end + * may have changed. + */ +static inline sector_t +compute_subtree_last(struct drbd_interval *node) +{ + sector_t max = node->sector + (node->size >> 9); + + if (node->rb.rb_left) { + sector_t left = interval_end(node->rb.rb_left); + if (left > max) + max = left; + } + if (node->rb.rb_right) { + sector_t right = interval_end(node->rb.rb_right); + if (right > max) + max = right; + } + return max; +} + +static void augment_propagate(struct rb_node *rb, struct rb_node *stop) +{ + while (rb != stop) { + struct drbd_interval *node = rb_entry(rb, struct drbd_interval, rb); + sector_t subtree_last = compute_subtree_last(node); + if (node->end == subtree_last) + break; + node->end = subtree_last; + rb = rb_parent(&node->rb); + } +} + +static void augment_copy(struct rb_node *rb_old, struct rb_node *rb_new) +{ + struct drbd_interval *old = rb_entry(rb_old, struct drbd_interval, rb); + struct drbd_interval *new = rb_entry(rb_new, struct drbd_interval, rb); + + new->end = old->end; +} + +static void augment_rotate(struct rb_node *rb_old, struct rb_node *rb_new) +{ + struct drbd_interval *old = rb_entry(rb_old, struct drbd_interval, rb); + struct drbd_interval *new = rb_entry(rb_new, struct drbd_interval, rb); + + new->end = old->end; + old->end = compute_subtree_last(old); +} + +static const struct rb_augment_callbacks augment_callbacks = { + augment_propagate, + augment_copy, + augment_rotate, +}; + +/** + * drbd_insert_interval - insert a new interval into a tree + */ +bool +drbd_insert_interval(struct rb_root *root, struct drbd_interval *this) +{ + struct rb_node **new = &root->rb_node, *parent = NULL; + + BUG_ON(!IS_ALIGNED(this->size, 512)); + + while (*new) { + struct drbd_interval *here = + rb_entry(*new, struct drbd_interval, rb); + + parent = *new; + if (this->sector < here->sector) + new = &(*new)->rb_left; + else if (this->sector > here->sector) + new = &(*new)->rb_right; + else if (this < here) + new = &(*new)->rb_left; + else if (this > here) + new = &(*new)->rb_right; + else + return false; + } + + rb_link_node(&this->rb, parent, new); + rb_insert_augmented(&this->rb, root, &augment_callbacks); + return true; +} + +/** + * drbd_contains_interval - check if a tree contains a given interval + * @sector: start sector of @interval + * @interval: may not be a valid pointer + * + * Returns if the tree contains the node @interval with start sector @start. + * Does not dereference @interval until @interval is known to be a valid object + * in @tree. Returns %false if @interval is in the tree but with a different + * sector number. + */ +bool +drbd_contains_interval(struct rb_root *root, sector_t sector, + struct drbd_interval *interval) +{ + struct rb_node *node = root->rb_node; + + while (node) { + struct drbd_interval *here = + rb_entry(node, struct drbd_interval, rb); + + if (sector < here->sector) + node = node->rb_left; + else if (sector > here->sector) + node = node->rb_right; + else if (interval < here) + node = node->rb_left; + else if (interval > here) + node = node->rb_right; + else + return true; + } + return false; +} + +/** + * drbd_remove_interval - remove an interval from a tree + */ +void +drbd_remove_interval(struct rb_root *root, struct drbd_interval *this) +{ + rb_erase_augmented(&this->rb, root, &augment_callbacks); +} + +/** + * drbd_find_overlap - search for an interval overlapping with [sector, sector + size) + * @sector: start sector + * @size: size, aligned to 512 bytes + * + * Returns an interval overlapping with [sector, sector + size), or NULL if + * there is none. When there is more than one overlapping interval in the + * tree, the interval with the lowest start sector is returned, and all other + * overlapping intervals will be on the right side of the tree, reachable with + * rb_next(). + */ +struct drbd_interval * +drbd_find_overlap(struct rb_root *root, sector_t sector, unsigned int size) +{ + struct rb_node *node = root->rb_node; + struct drbd_interval *overlap = NULL; + sector_t end = sector + (size >> 9); + + BUG_ON(!IS_ALIGNED(size, 512)); + + while (node) { + struct drbd_interval *here = + rb_entry(node, struct drbd_interval, rb); + + if (node->rb_left && + sector < interval_end(node->rb_left)) { + /* Overlap if any must be on left side */ + node = node->rb_left; + } else if (here->sector < end && + sector < here->sector + (here->size >> 9)) { + overlap = here; + break; + } else if (sector >= here->sector) { + /* Overlap if any must be on right side */ + node = node->rb_right; + } else + break; + } + return overlap; +} + +struct drbd_interval * +drbd_next_overlap(struct drbd_interval *i, sector_t sector, unsigned int size) +{ + sector_t end = sector + (size >> 9); + struct rb_node *node; + + for (;;) { + node = rb_next(&i->rb); + if (!node) + return NULL; + i = rb_entry(node, struct drbd_interval, rb); + if (i->sector >= end) + return NULL; + if (sector < i->sector + (i->size >> 9)) + return i; + } +} diff --git a/drivers/block/drbd/drbd_interval.h b/drivers/block/drbd/drbd_interval.h new file mode 100644 index 000000000000..f38fcb00c10d --- /dev/null +++ b/drivers/block/drbd/drbd_interval.h @@ -0,0 +1,40 @@ +#ifndef __DRBD_INTERVAL_H +#define __DRBD_INTERVAL_H + +#include <linux/types.h> +#include <linux/rbtree.h> + +struct drbd_interval { + struct rb_node rb; + sector_t sector; /* start sector of the interval */ + unsigned int size; /* size in bytes */ + sector_t end; /* highest interval end in subtree */ + int local:1 /* local or remote request? */; + int waiting:1; +}; + +static inline void drbd_clear_interval(struct drbd_interval *i) +{ + RB_CLEAR_NODE(&i->rb); +} + +static inline bool drbd_interval_empty(struct drbd_interval *i) +{ + return RB_EMPTY_NODE(&i->rb); +} + +extern bool drbd_insert_interval(struct rb_root *, struct drbd_interval *); +extern bool drbd_contains_interval(struct rb_root *, sector_t, + struct drbd_interval *); +extern void drbd_remove_interval(struct rb_root *, struct drbd_interval *); +extern struct drbd_interval *drbd_find_overlap(struct rb_root *, sector_t, + unsigned int); +extern struct drbd_interval *drbd_next_overlap(struct drbd_interval *, sector_t, + unsigned int); + +#define drbd_for_each_overlap(i, root, sector, size) \ + for (i = drbd_find_overlap(root, sector, size); \ + i; \ + i = drbd_next_overlap(i, sector, size)) + +#endif /* __DRBD_INTERVAL_H */ diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c index f55683ad4ffa..8c13eeb83c53 100644 --- a/drivers/block/drbd/drbd_main.c +++ b/drivers/block/drbd/drbd_main.c @@ -56,14 +56,6 @@ #include "drbd_vli.h" -struct after_state_chg_work { - struct drbd_work w; - union drbd_state os; - union drbd_state ns; - enum chg_state_flags flags; - struct completion *done; -}; - static DEFINE_MUTEX(drbd_main_mutex); int drbdd_init(struct drbd_thread *); int drbd_worker(struct drbd_thread *); @@ -72,21 +64,17 @@ int drbd_asender(struct drbd_thread *); int drbd_init(void); static int drbd_open(struct block_device *bdev, fmode_t mode); static int drbd_release(struct gendisk *gd, fmode_t mode); -static int w_after_state_ch(struct drbd_conf *mdev, struct drbd_work *w, int unused); -static void after_state_ch(struct drbd_conf *mdev, union drbd_state os, - union drbd_state ns, enum chg_state_flags flags); -static int w_md_sync(struct drbd_conf *mdev, struct drbd_work *w, int unused); +static int w_md_sync(struct drbd_work *w, int unused); static void md_sync_timer_fn(unsigned long data); -static int w_bitmap_io(struct drbd_conf *mdev, struct drbd_work *w, int unused); -static int w_go_diskless(struct drbd_conf *mdev, struct drbd_work *w, int unused); -static void _tl_clear(struct drbd_conf *mdev); +static int w_bitmap_io(struct drbd_work *w, int unused); +static int w_go_diskless(struct drbd_work *w, int unused); MODULE_AUTHOR("Philipp Reisner <phil@linbit.com>, " "Lars Ellenberg <lars@linbit.com>"); MODULE_DESCRIPTION("drbd - Distributed Replicated Block Device v" REL_VERSION); MODULE_VERSION(REL_VERSION); MODULE_LICENSE("GPL"); -MODULE_PARM_DESC(minor_count, "Maximum number of drbd devices (" +MODULE_PARM_DESC(minor_count, "Approximate number of drbd devices (" __stringify(DRBD_MINOR_COUNT_MIN) "-" __stringify(DRBD_MINOR_COUNT_MAX) ")"); MODULE_ALIAS_BLOCKDEV_MAJOR(DRBD_MAJOR); @@ -98,7 +86,6 @@ MODULE_PARM_DESC(allow_oos, "DONT USE!"); module_param(minor_count, uint, 0444); module_param(disable_sendpage, bool, 0644); module_param(allow_oos, bool, 0); -module_param(cn_idx, uint, 0444); module_param(proc_details, int, 0644); #ifdef CONFIG_DRBD_FAULT_INJECTION @@ -120,7 +107,6 @@ module_param(fault_devs, int, 0644); unsigned int minor_count = DRBD_MINOR_COUNT_DEF; bool disable_sendpage; bool allow_oos; -unsigned int cn_idx = CN_IDX_DRBD; int proc_details; /* Detail level in proc drbd*/ /* Module parameter for setting the user mode helper program @@ -132,10 +118,11 @@ module_param_string(usermode_helper, usermode_helper, sizeof(usermode_helper), 0 /* in 2.6.x, our device mapping and config info contains our virtual gendisks * as member "struct gendisk *vdisk;" */ -struct drbd_conf **minor_table; +struct idr minors; +struct list_head drbd_tconns; /* list of struct drbd_tconn */ struct kmem_cache *drbd_request_cache; -struct kmem_cache *drbd_ee_cache; /* epoch entries */ +struct kmem_cache *drbd_ee_cache; /* peer requests */ struct kmem_cache *drbd_bm_ext_cache; /* bitmap extents */ struct kmem_cache *drbd_al_ext_cache; /* activity log extents */ mempool_t *drbd_request_mempool; @@ -164,10 +151,15 @@ static const struct block_device_operations drbd_ops = { struct bio *bio_alloc_drbd(gfp_t gfp_mask) { + struct bio *bio; + if (!drbd_md_io_bio_set) return bio_alloc(gfp_mask, 1); - return bio_alloc_bioset(gfp_mask, 1, drbd_md_io_bio_set); + bio = bio_alloc_bioset(gfp_mask, 1, drbd_md_io_bio_set); + if (!bio) + return NULL; + return bio; } #ifdef __CHECKER__ @@ -190,158 +182,87 @@ int _get_ldev_if_state(struct drbd_conf *mdev, enum drbd_disk_state mins) #endif /** - * DOC: The transfer log - * - * The transfer log is a single linked list of &struct drbd_tl_epoch objects. - * mdev->newest_tle points to the head, mdev->oldest_tle points to the tail - * of the list. There is always at least one &struct drbd_tl_epoch object. - * - * Each &struct drbd_tl_epoch has a circular double linked list of requests - * attached. - */ -static int tl_init(struct drbd_conf *mdev) -{ - struct drbd_tl_epoch *b; - - /* during device minor initialization, we may well use GFP_KERNEL */ - b = kmalloc(sizeof(struct drbd_tl_epoch), GFP_KERNEL); - if (!b) - return 0; - INIT_LIST_HEAD(&b->requests); - INIT_LIST_HEAD(&b->w.list); - b->next = NULL; - b->br_number = 4711; - b->n_writes = 0; - b->w.cb = NULL; /* if this is != NULL, we need to dec_ap_pending in tl_clear */ - - mdev->oldest_tle = b; - mdev->newest_tle = b; - INIT_LIST_HEAD(&mdev->out_of_sequence_requests); - INIT_LIST_HEAD(&mdev->barrier_acked_requests); - - mdev->tl_hash = NULL; - mdev->tl_hash_s = 0; - - return 1; -} - -static void tl_cleanup(struct drbd_conf *mdev) -{ - D_ASSERT(mdev->oldest_tle == mdev->newest_tle); - D_ASSERT(list_empty(&mdev->out_of_sequence_requests)); - kfree(mdev->oldest_tle); - mdev->oldest_tle = NULL; - kfree(mdev->unused_spare_tle); - mdev->unused_spare_tle = NULL; - kfree(mdev->tl_hash); - mdev->tl_hash = NULL; - mdev->tl_hash_s = 0; -} - -/** - * _tl_add_barrier() - Adds a barrier to the transfer log - * @mdev: DRBD device. - * @new: Barrier to be added before the current head of the TL. - * - * The caller must hold the req_lock. - */ -void _tl_add_barrier(struct drbd_conf *mdev, struct drbd_tl_epoch *new) -{ - struct drbd_tl_epoch *newest_before; - - INIT_LIST_HEAD(&new->requests); - INIT_LIST_HEAD(&new->w.list); - new->w.cb = NULL; /* if this is != NULL, we need to dec_ap_pending in tl_clear */ - new->next = NULL; - new->n_writes = 0; - - newest_before = mdev->newest_tle; - new->br_number = newest_before->br_number+1; - if (mdev->newest_tle != new) { - mdev->newest_tle->next = new; - mdev->newest_tle = new; - } -} - -/** - * tl_release() - Free or recycle the oldest &struct drbd_tl_epoch object of the TL - * @mdev: DRBD device. + * tl_release() - mark as BARRIER_ACKED all requests in the corresponding transfer log epoch + * @tconn: DRBD connection. * @barrier_nr: Expected identifier of the DRBD write barrier packet. * @set_size: Expected number of requests before that barrier. * * In case the passed barrier_nr or set_size does not match the oldest - * &struct drbd_tl_epoch objects this function will cause a termination - * of the connection. + * epoch of not yet barrier-acked requests, this function will cause a + * termination of the connection. */ -void tl_release(struct drbd_conf *mdev, unsigned int barrier_nr, - unsigned int set_size) +void tl_release(struct drbd_tconn *tconn, unsigned int barrier_nr, + unsigned int set_size) { - struct drbd_tl_epoch *b, *nob; /* next old barrier */ - struct list_head *le, *tle; struct drbd_request *r; - - spin_lock_irq(&mdev->req_lock); - - b = mdev->oldest_tle; + struct drbd_request *req = NULL; + int expect_epoch = 0; + int expect_size = 0; + + spin_lock_irq(&tconn->req_lock); + + /* find oldest not yet barrier-acked write request, + * count writes in its epoch. */ + list_for_each_entry(r, &tconn->transfer_log, tl_requests) { + const unsigned s = r->rq_state; + if (!req) { + if (!(s & RQ_WRITE)) + continue; + if (!(s & RQ_NET_MASK)) + continue; + if (s & RQ_NET_DONE) + continue; + req = r; + expect_epoch = req->epoch; + expect_size ++; + } else { + if (r->epoch != expect_epoch) + break; + if (!(s & RQ_WRITE)) + continue; + /* if (s & RQ_DONE): not expected */ + /* if (!(s & RQ_NET_MASK)): not expected */ + expect_size++; + } + } /* first some paranoia code */ - if (b == NULL) { - dev_err(DEV, "BAD! BarrierAck #%u received, but no epoch in tl!?\n", - barrier_nr); + if (req == NULL) { + conn_err(tconn, "BAD! BarrierAck #%u received, but no epoch in tl!?\n", + barrier_nr); goto bail; } - if (b->br_number != barrier_nr) { - dev_err(DEV, "BAD! BarrierAck #%u received, expected #%u!\n", - barrier_nr, b->br_number); + if (expect_epoch != barrier_nr) { + conn_err(tconn, "BAD! BarrierAck #%u received, expected #%u!\n", + barrier_nr, expect_epoch); goto bail; } - if (b->n_writes != set_size) { - dev_err(DEV, "BAD! BarrierAck #%u received with n_writes=%u, expected n_writes=%u!\n", - barrier_nr, set_size, b->n_writes); + + if (expect_size != set_size) { + conn_err(tconn, "BAD! BarrierAck #%u received with n_writes=%u, expected n_writes=%u!\n", + barrier_nr, set_size, expect_size); goto bail; } - /* Clean up list of requests processed during current epoch */ - list_for_each_safe(le, tle, &b->requests) { - r = list_entry(le, struct drbd_request, tl_requests); - _req_mod(r, barrier_acked); - } - /* There could be requests on the list waiting for completion - of the write to the local disk. To avoid corruptions of - slab's data structures we have to remove the lists head. - - Also there could have been a barrier ack out of sequence, overtaking - the write acks - which would be a bug and violating write ordering. - To not deadlock in case we lose connection while such requests are - still pending, we need some way to find them for the - _req_mode(connection_lost_while_pending). - - These have been list_move'd to the out_of_sequence_requests list in - _req_mod(, barrier_acked) above. - */ - list_splice_init(&b->requests, &mdev->barrier_acked_requests); - - nob = b->next; - if (test_and_clear_bit(CREATE_BARRIER, &mdev->flags)) { - _tl_add_barrier(mdev, b); - if (nob) - mdev->oldest_tle = nob; - /* if nob == NULL b was the only barrier, and becomes the new - barrier. Therefore mdev->oldest_tle points already to b */ - } else { - D_ASSERT(nob != NULL); - mdev->oldest_tle = nob; - kfree(b); + /* Clean up list of requests processed during current epoch. */ + /* this extra list walk restart is paranoia, + * to catch requests being barrier-acked "unexpectedly". + * It usually should find the same req again, or some READ preceding it. */ + list_for_each_entry(req, &tconn->transfer_log, tl_requests) + if (req->epoch == expect_epoch) + break; + list_for_each_entry_safe_from(req, r, &tconn->transfer_log, tl_requests) { + if (req->epoch != expect_epoch) + break; + _req_mod(req, BARRIER_ACKED); } - - spin_unlock_irq(&mdev->req_lock); - dec_ap_pending(mdev); + spin_unlock_irq(&tconn->req_lock); return; bail: - spin_unlock_irq(&mdev->req_lock); - drbd_force_state(mdev, NS(conn, C_PROTOCOL_ERROR)); + spin_unlock_irq(&tconn->req_lock); + conn_request_state(tconn, NS(conn, C_PROTOCOL_ERROR), CS_HARD); } @@ -350,85 +271,24 @@ bail: * @mdev: DRBD device. * @what: The action/event to perform with all request objects * - * @what might be one of connection_lost_while_pending, resend, fail_frozen_disk_io, - * restart_frozen_disk_io. + * @what might be one of CONNECTION_LOST_WHILE_PENDING, RESEND, FAIL_FROZEN_DISK_IO, + * RESTART_FROZEN_DISK_IO. */ -static void _tl_restart(struct drbd_conf *mdev, enum drbd_req_event what) -{ - struct drbd_tl_epoch *b, *tmp, **pn; - struct list_head *le, *tle, carry_reads; - struct drbd_request *req; - int rv, n_writes, n_reads; - - b = mdev->oldest_tle; - pn = &mdev->oldest_tle; - while (b) { - n_writes = 0; - n_reads = 0; - INIT_LIST_HEAD(&carry_reads); - list_for_each_safe(le, tle, &b->requests) { - req = list_entry(le, struct drbd_request, tl_requests); - rv = _req_mod(req, what); - - n_writes += (rv & MR_WRITE) >> MR_WRITE_SHIFT; - n_reads += (rv & MR_READ) >> MR_READ_SHIFT; - } - tmp = b->next; - - if (n_writes) { - if (what == resend) { - b->n_writes = n_writes; - if (b->w.cb == NULL) { - b->w.cb = w_send_barrier; - inc_ap_pending(mdev); - set_bit(CREATE_BARRIER, &mdev->flags); - } - - drbd_queue_work(&mdev->data.work, &b->w); - } - pn = &b->next; - } else { - if (n_reads) - list_add(&carry_reads, &b->requests); - /* there could still be requests on that ring list, - * in case local io is still pending */ - list_del(&b->requests); - - /* dec_ap_pending corresponding to queue_barrier. - * the newest barrier may not have been queued yet, - * in which case w.cb is still NULL. */ - if (b->w.cb != NULL) - dec_ap_pending(mdev); - - if (b == mdev->newest_tle) { - /* recycle, but reinit! */ - D_ASSERT(tmp == NULL); - INIT_LIST_HEAD(&b->requests); - list_splice(&carry_reads, &b->requests); - INIT_LIST_HEAD(&b->w.list); - b->w.cb = NULL; - b->br_number = net_random(); - b->n_writes = 0; - - *pn = b; - break; - } - *pn = tmp; - kfree(b); - } - b = tmp; - list_splice(&carry_reads, &b->requests); - } - - /* Actions operating on the disk state, also want to work on - requests that got barrier acked. */ +/* must hold resource->req_lock */ +void _tl_restart(struct drbd_tconn *tconn, enum drbd_req_event what) +{ + struct drbd_request *req, *r; - list_for_each_safe(le, tle, &mdev->barrier_acked_requests) { - req = list_entry(le, struct drbd_request, tl_requests); + list_for_each_entry_safe(req, r, &tconn->transfer_log, tl_requests) _req_mod(req, what); - } } +void tl_restart(struct drbd_tconn *tconn, enum drbd_req_event what) +{ + spin_lock_irq(&tconn->req_lock); + _tl_restart(tconn, what); + spin_unlock_irq(&tconn->req_lock); +} /** * tl_clear() - Clears all requests and &struct drbd_tl_epoch objects out of the TL @@ -438,43 +298,9 @@ static void _tl_restart(struct drbd_conf *mdev, enum drbd_req_event what) * by the requests on the transfer gets marked as our of sync. Called from the * receiver thread and the worker thread. */ -void tl_clear(struct drbd_conf *mdev) +void tl_clear(struct drbd_tconn *tconn) { - spin_lock_irq(&mdev->req_lock); - _tl_clear(mdev); - spin_unlock_irq(&mdev->req_lock); -} - -static void _tl_clear(struct drbd_conf *mdev) -{ - struct list_head *le, *tle; - struct drbd_request *r; - - _tl_restart(mdev, connection_lost_while_pending); - - /* we expect this list to be empty. */ - D_ASSERT(list_empty(&mdev->out_of_sequence_requests)); - - /* but just in case, clean it up anyways! */ - list_for_each_safe(le, tle, &mdev->out_of_sequence_requests) { - r = list_entry(le, struct drbd_request, tl_requests); - /* It would be nice to complete outside of spinlock. - * But this is easier for now. */ - _req_mod(r, connection_lost_while_pending); - } - - /* ensure bit indicating barrier is required is clear */ - clear_bit(CREATE_BARRIER, &mdev->flags); - - memset(mdev->app_reads_hash, 0, APP_R_HSIZE*sizeof(void *)); - -} - -void tl_restart(struct drbd_conf *mdev, enum drbd_req_event what) -{ - spin_lock_irq(&mdev->req_lock); - _tl_restart(mdev, what); - spin_unlock_irq(&mdev->req_lock); + tl_restart(tconn, CONNECTION_LOST_WHILE_PENDING); } /** @@ -483,1377 +309,131 @@ void tl_restart(struct drbd_conf *mdev, enum drbd_req_event what) */ void tl_abort_disk_io(struct drbd_conf *mdev) { - struct drbd_tl_epoch *b; - struct list_head *le, *tle; - struct drbd_request *req; - - spin_lock_irq(&mdev->req_lock); - b = mdev->oldest_tle; - while (b) { - list_for_each_safe(le, tle, &b->requests) { - req = list_entry(le, struct drbd_request, tl_requests); - if (!(req->rq_state & RQ_LOCAL_PENDING)) - continue; - _req_mod(req, abort_disk_io); - } - b = b->next; - } + struct drbd_tconn *tconn = mdev->tconn; + struct drbd_request *req, *r; - list_for_each_safe(le, tle, &mdev->barrier_acked_requests) { - req = list_entry(le, struct drbd_request, tl_requests); + spin_lock_irq(&tconn->req_lock); + list_for_each_entry_safe(req, r, &tconn->transfer_log, tl_requests) { if (!(req->rq_state & RQ_LOCAL_PENDING)) continue; - _req_mod(req, abort_disk_io); - } - - spin_unlock_irq(&mdev->req_lock); -} - -/** - * cl_wide_st_chg() - true if the state change is a cluster wide one - * @mdev: DRBD device. - * @os: old (current) state. - * @ns: new (wanted) state. - */ -static int cl_wide_st_chg(struct drbd_conf *mdev, - union drbd_state os, union drbd_state ns) -{ - return (os.conn >= C_CONNECTED && ns.conn >= C_CONNECTED && - ((os.role != R_PRIMARY && ns.role == R_PRIMARY) || - (os.conn != C_STARTING_SYNC_T && ns.conn == C_STARTING_SYNC_T) || - (os.conn != C_STARTING_SYNC_S && ns.conn == C_STARTING_SYNC_S) || - (os.disk != D_FAILED && ns.disk == D_FAILED))) || - (os.conn >= C_CONNECTED && ns.conn == C_DISCONNECTING) || - (os.conn == C_CONNECTED && ns.conn == C_VERIFY_S); -} - -enum drbd_state_rv -drbd_change_state(struct drbd_conf *mdev, enum chg_state_flags f, - union drbd_state mask, union drbd_state val) -{ - unsigned long flags; - union drbd_state os, ns; - enum drbd_state_rv rv; - - spin_lock_irqsave(&mdev->req_lock, flags); - os = mdev->state; - ns.i = (os.i & ~mask.i) | val.i; - rv = _drbd_set_state(mdev, ns, f, NULL); - ns = mdev->state; - spin_unlock_irqrestore(&mdev->req_lock, flags); - - return rv; -} - -/** - * drbd_force_state() - Impose a change which happens outside our control on our state - * @mdev: DRBD device. - * @mask: mask of state bits to change. - * @val: value of new state bits. - */ -void drbd_force_state(struct drbd_conf *mdev, - union drbd_state mask, union drbd_state val) -{ - drbd_change_state(mdev, CS_HARD, mask, val); -} - -static enum drbd_state_rv is_valid_state(struct drbd_conf *, union drbd_state); -static enum drbd_state_rv is_valid_state_transition(struct drbd_conf *, - union drbd_state, - union drbd_state); -enum sanitize_state_warnings { - NO_WARNING, - ABORTED_ONLINE_VERIFY, - ABORTED_RESYNC, - CONNECTION_LOST_NEGOTIATING, - IMPLICITLY_UPGRADED_DISK, - IMPLICITLY_UPGRADED_PDSK, -}; -static union drbd_state sanitize_state(struct drbd_conf *mdev, union drbd_state os, - union drbd_state ns, enum sanitize_state_warnings *warn); -int drbd_send_state_req(struct drbd_conf *, - union drbd_state, union drbd_state); - -static enum drbd_state_rv -_req_st_cond(struct drbd_conf *mdev, union drbd_state mask, - union drbd_state val) -{ - union drbd_state os, ns; - unsigned long flags; - enum drbd_state_rv rv; - - if (test_and_clear_bit(CL_ST_CHG_SUCCESS, &mdev->flags)) - return SS_CW_SUCCESS; - - if (test_and_clear_bit(CL_ST_CHG_FAIL, &mdev->flags)) - return SS_CW_FAILED_BY_PEER; - - rv = 0; - spin_lock_irqsave(&mdev->req_lock, flags); - os = mdev->state; - ns.i = (os.i & ~mask.i) | val.i; - ns = sanitize_state(mdev, os, ns, NULL); - - if (!cl_wide_st_chg(mdev, os, ns)) - rv = SS_CW_NO_NEED; - if (!rv) { - rv = is_valid_state(mdev, ns); - if (rv == SS_SUCCESS) { - rv = is_valid_state_transition(mdev, ns, os); - if (rv == SS_SUCCESS) - rv = SS_UNKNOWN_ERROR; /* cont waiting, otherwise fail. */ - } - } - spin_unlock_irqrestore(&mdev->req_lock, flags); - - return rv; -} - -/** - * drbd_req_state() - Perform an eventually cluster wide state change - * @mdev: DRBD device. - * @mask: mask of state bits to change. - * @val: value of new state bits. - * @f: flags - * - * Should not be called directly, use drbd_request_state() or - * _drbd_request_state(). - */ -static enum drbd_state_rv -drbd_req_state(struct drbd_conf *mdev, union drbd_state mask, - union drbd_state val, enum chg_state_flags f) -{ - struct completion done; - unsigned long flags; - union drbd_state os, ns; - enum drbd_state_rv rv; - - init_completion(&done); - - if (f & CS_SERIALIZE) - mutex_lock(&mdev->state_mutex); - - spin_lock_irqsave(&mdev->req_lock, flags); - os = mdev->state; - ns.i = (os.i & ~mask.i) | val.i; - ns = sanitize_state(mdev, os, ns, NULL); - - if (cl_wide_st_chg(mdev, os, ns)) { - rv = is_valid_state(mdev, ns); - if (rv == SS_SUCCESS) - rv = is_valid_state_transition(mdev, ns, os); - spin_unlock_irqrestore(&mdev->req_lock, flags); - - if (rv < SS_SUCCESS) { - if (f & CS_VERBOSE) - print_st_err(mdev, os, ns, rv); - goto abort; - } - - drbd_state_lock(mdev); - if (!drbd_send_state_req(mdev, mask, val)) { - drbd_state_unlock(mdev); - rv = SS_CW_FAILED_BY_PEER; - if (f & CS_VERBOSE) - print_st_err(mdev, os, ns, rv); - goto abort; - } - - wait_event(mdev->state_wait, - (rv = _req_st_cond(mdev, mask, val))); - - if (rv < SS_SUCCESS) { - drbd_state_unlock(mdev); - if (f & CS_VERBOSE) - print_st_err(mdev, os, ns, rv); - goto abort; - } - spin_lock_irqsave(&mdev->req_lock, flags); - os = mdev->state; - ns.i = (os.i & ~mask.i) | val.i; - rv = _drbd_set_state(mdev, ns, f, &done); - drbd_state_unlock(mdev); - } else { - rv = _drbd_set_state(mdev, ns, f, &done); - } - - spin_unlock_irqrestore(&mdev->req_lock, flags); - - if (f & CS_WAIT_COMPLETE && rv == SS_SUCCESS) { - D_ASSERT(current != mdev->worker.task); - wait_for_completion(&done); - } - -abort: - if (f & CS_SERIALIZE) - mutex_unlock(&mdev->state_mutex); - - return rv; -} - -/** - * _drbd_request_state() - Request a state change (with flags) - * @mdev: DRBD device. - * @mask: mask of state bits to change. - * @val: value of new state bits. - * @f: flags - * - * Cousin of drbd_request_state(), useful with the CS_WAIT_COMPLETE - * flag, or when logging of failed state change requests is not desired. - */ -enum drbd_state_rv -_drbd_request_state(struct drbd_conf *mdev, union drbd_state mask, - union drbd_state val, enum chg_state_flags f) -{ - enum drbd_state_rv rv; - - wait_event(mdev->state_wait, - (rv = drbd_req_state(mdev, mask, val, f)) != SS_IN_TRANSIENT_STATE); - - return rv; -} - -static void print_st(struct drbd_conf *mdev, char *name, union drbd_state ns) -{ - dev_err(DEV, " %s = { cs:%s ro:%s/%s ds:%s/%s %c%c%c%c }\n", - name, - drbd_conn_str(ns.conn), - drbd_role_str(ns.role), - drbd_role_str(ns.peer), - drbd_disk_str(ns.disk), - drbd_disk_str(ns.pdsk), - is_susp(ns) ? 's' : 'r', - ns.aftr_isp ? 'a' : '-', - ns.peer_isp ? 'p' : '-', - ns.user_isp ? 'u' : '-' - ); -} - -void print_st_err(struct drbd_conf *mdev, union drbd_state os, - union drbd_state ns, enum drbd_state_rv err) -{ - if (err == SS_IN_TRANSIENT_STATE) - return; - dev_err(DEV, "State change failed: %s\n", drbd_set_st_err_str(err)); - print_st(mdev, " state", os); - print_st(mdev, "wanted", ns); -} - - -/** - * is_valid_state() - Returns an SS_ error code if ns is not valid - * @mdev: DRBD device. - * @ns: State to consider. - */ -static enum drbd_state_rv -is_valid_state(struct drbd_conf *mdev, union drbd_state ns) -{ - /* See drbd_state_sw_errors in drbd_strings.c */ - - enum drbd_fencing_p fp; - enum drbd_state_rv rv = SS_SUCCESS; - - fp = FP_DONT_CARE; - if (get_ldev(mdev)) { - fp = mdev->ldev->dc.fencing; - put_ldev(mdev); - } - - if (get_net_conf(mdev)) { - if (!mdev->net_conf->two_primaries && - ns.role == R_PRIMARY && ns.peer == R_PRIMARY) - rv = SS_TWO_PRIMARIES; - put_net_conf(mdev); - } - - if (rv <= 0) - /* already found a reason to abort */; - else if (ns.role == R_SECONDARY && mdev->open_cnt) - rv = SS_DEVICE_IN_USE; - - else if (ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.disk < D_UP_TO_DATE) - rv = SS_NO_UP_TO_DATE_DISK; - - else if (fp >= FP_RESOURCE && - ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.pdsk >= D_UNKNOWN) - rv = SS_PRIMARY_NOP; - - else if (ns.role == R_PRIMARY && ns.disk <= D_INCONSISTENT && ns.pdsk <= D_INCONSISTENT) - rv = SS_NO_UP_TO_DATE_DISK; - - else if (ns.conn > C_CONNECTED && ns.disk < D_INCONSISTENT) - rv = SS_NO_LOCAL_DISK; - - else if (ns.conn > C_CONNECTED && ns.pdsk < D_INCONSISTENT) - rv = SS_NO_REMOTE_DISK; - - else if (ns.conn > C_CONNECTED && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE) - rv = SS_NO_UP_TO_DATE_DISK; - - else if ((ns.conn == C_CONNECTED || - ns.conn == C_WF_BITMAP_S || - ns.conn == C_SYNC_SOURCE || - ns.conn == C_PAUSED_SYNC_S) && - ns.disk == D_OUTDATED) - rv = SS_CONNECTED_OUTDATES; - - else if ((ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T) && - (mdev->sync_conf.verify_alg[0] == 0)) - rv = SS_NO_VERIFY_ALG; - - else if ((ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T) && - mdev->agreed_pro_version < 88) - rv = SS_NOT_SUPPORTED; - - else if (ns.conn >= C_CONNECTED && ns.pdsk == D_UNKNOWN) - rv = SS_CONNECTED_OUTDATES; - - return rv; -} - -/** - * is_valid_state_transition() - Returns an SS_ error code if the state transition is not possible - * @mdev: DRBD device. - * @ns: new state. - * @os: old state. - */ -static enum drbd_state_rv -is_valid_state_transition(struct drbd_conf *mdev, union drbd_state ns, - union drbd_state os) -{ - enum drbd_state_rv rv = SS_SUCCESS; - - if ((ns.conn == C_STARTING_SYNC_T || ns.conn == C_STARTING_SYNC_S) && - os.conn > C_CONNECTED) - rv = SS_RESYNC_RUNNING; - - if (ns.conn == C_DISCONNECTING && os.conn == C_STANDALONE) - rv = SS_ALREADY_STANDALONE; - - if (ns.disk > D_ATTACHING && os.disk == D_DISKLESS) - rv = SS_IS_DISKLESS; - - if (ns.conn == C_WF_CONNECTION && os.conn < C_UNCONNECTED) - rv = SS_NO_NET_CONFIG; - - if (ns.disk == D_OUTDATED && os.disk < D_OUTDATED && os.disk != D_ATTACHING) - rv = SS_LOWER_THAN_OUTDATED; - - if (ns.conn == C_DISCONNECTING && os.conn == C_UNCONNECTED) - rv = SS_IN_TRANSIENT_STATE; - - if (ns.conn == os.conn && ns.conn == C_WF_REPORT_PARAMS) - rv = SS_IN_TRANSIENT_STATE; - - /* While establishing a connection only allow cstate to change. - Delay/refuse role changes, detach attach etc... */ - if (test_bit(STATE_SENT, &mdev->flags) && - !(os.conn == C_WF_REPORT_PARAMS || - (ns.conn == C_WF_REPORT_PARAMS && os.conn == C_WF_CONNECTION))) - rv = SS_IN_TRANSIENT_STATE; - - if ((ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T) && os.conn < C_CONNECTED) - rv = SS_NEED_CONNECTION; - - if ((ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T) && - ns.conn != os.conn && os.conn > C_CONNECTED) - rv = SS_RESYNC_RUNNING; - - if ((ns.conn == C_STARTING_SYNC_S || ns.conn == C_STARTING_SYNC_T) && - os.conn < C_CONNECTED) - rv = SS_NEED_CONNECTION; - - if ((ns.conn == C_SYNC_TARGET || ns.conn == C_SYNC_SOURCE) - && os.conn < C_WF_REPORT_PARAMS) - rv = SS_NEED_CONNECTION; /* No NetworkFailure -> SyncTarget etc... */ - - return rv; -} - -static void print_sanitize_warnings(struct drbd_conf *mdev, enum sanitize_state_warnings warn) -{ - static const char *msg_table[] = { - [NO_WARNING] = "", - [ABORTED_ONLINE_VERIFY] = "Online-verify aborted.", - [ABORTED_RESYNC] = "Resync aborted.", - [CONNECTION_LOST_NEGOTIATING] = "Connection lost while negotiating, no data!", - [IMPLICITLY_UPGRADED_DISK] = "Implicitly upgraded disk", - [IMPLICITLY_UPGRADED_PDSK] = "Implicitly upgraded pdsk", - }; - - if (warn != NO_WARNING) - dev_warn(DEV, "%s\n", msg_table[warn]); -} - -/** - * sanitize_state() - Resolves implicitly necessary additional changes to a state transition - * @mdev: DRBD device. - * @os: old state. - * @ns: new state. - * @warn_sync_abort: - * - * When we loose connection, we have to set the state of the peers disk (pdsk) - * to D_UNKNOWN. This rule and many more along those lines are in this function. - */ -static union drbd_state sanitize_state(struct drbd_conf *mdev, union drbd_state os, - union drbd_state ns, enum sanitize_state_warnings *warn) -{ - enum drbd_fencing_p fp; - enum drbd_disk_state disk_min, disk_max, pdsk_min, pdsk_max; - - if (warn) - *warn = NO_WARNING; - - fp = FP_DONT_CARE; - if (get_ldev(mdev)) { - fp = mdev->ldev->dc.fencing; - put_ldev(mdev); - } - - /* Disallow Network errors to configure a device's network part */ - if ((ns.conn >= C_TIMEOUT && ns.conn <= C_TEAR_DOWN) && - os.conn <= C_DISCONNECTING) - ns.conn = os.conn; - - /* After a network error (+C_TEAR_DOWN) only C_UNCONNECTED or C_DISCONNECTING can follow. - * If you try to go into some Sync* state, that shall fail (elsewhere). */ - if (os.conn >= C_TIMEOUT && os.conn <= C_TEAR_DOWN && - ns.conn != C_UNCONNECTED && ns.conn != C_DISCONNECTING && ns.conn <= C_CONNECTED) - ns.conn = os.conn; - - /* we cannot fail (again) if we already detached */ - if (ns.disk == D_FAILED && os.disk == D_DISKLESS) - ns.disk = D_DISKLESS; - - /* After C_DISCONNECTING only C_STANDALONE may follow */ - if (os.conn == C_DISCONNECTING && ns.conn != C_STANDALONE) - ns.conn = os.conn; - - if (ns.conn < C_CONNECTED) { - ns.peer_isp = 0; - ns.peer = R_UNKNOWN; - if (ns.pdsk > D_UNKNOWN || ns.pdsk < D_INCONSISTENT) - ns.pdsk = D_UNKNOWN; - } - - /* Clear the aftr_isp when becoming unconfigured */ - if (ns.conn == C_STANDALONE && ns.disk == D_DISKLESS && ns.role == R_SECONDARY) - ns.aftr_isp = 0; - - /* Abort resync if a disk fails/detaches */ - if (os.conn > C_CONNECTED && ns.conn > C_CONNECTED && - (ns.disk <= D_FAILED || ns.pdsk <= D_FAILED)) { - if (warn) - *warn = os.conn == C_VERIFY_S || os.conn == C_VERIFY_T ? - ABORTED_ONLINE_VERIFY : ABORTED_RESYNC; - ns.conn = C_CONNECTED; - } - - /* Connection breaks down before we finished "Negotiating" */ - if (ns.conn < C_CONNECTED && ns.disk == D_NEGOTIATING && - get_ldev_if_state(mdev, D_NEGOTIATING)) { - if (mdev->ed_uuid == mdev->ldev->md.uuid[UI_CURRENT]) { - ns.disk = mdev->new_state_tmp.disk; - ns.pdsk = mdev->new_state_tmp.pdsk; - } else { - if (warn) - *warn = CONNECTION_LOST_NEGOTIATING; - ns.disk = D_DISKLESS; - ns.pdsk = D_UNKNOWN; - } - put_ldev(mdev); - } - - /* D_CONSISTENT and D_OUTDATED vanish when we get connected */ - if (ns.conn >= C_CONNECTED && ns.conn < C_AHEAD) { - if (ns.disk == D_CONSISTENT || ns.disk == D_OUTDATED) - ns.disk = D_UP_TO_DATE; - if (ns.pdsk == D_CONSISTENT || ns.pdsk == D_OUTDATED) - ns.pdsk = D_UP_TO_DATE; - } - - /* Implications of the connection stat on the disk states */ - disk_min = D_DISKLESS; - disk_max = D_UP_TO_DATE; - pdsk_min = D_INCONSISTENT; - pdsk_max = D_UNKNOWN; - switch ((enum drbd_conns)ns.conn) { - case C_WF_BITMAP_T: - case C_PAUSED_SYNC_T: - case C_STARTING_SYNC_T: - case C_WF_SYNC_UUID: - case C_BEHIND: - disk_min = D_INCONSISTENT; - disk_max = D_OUTDATED; - pdsk_min = D_UP_TO_DATE; - pdsk_max = D_UP_TO_DATE; - break; - case C_VERIFY_S: - case C_VERIFY_T: - disk_min = D_UP_TO_DATE; - disk_max = D_UP_TO_DATE; - pdsk_min = D_UP_TO_DATE; - pdsk_max = D_UP_TO_DATE; - break; - case C_CONNECTED: - disk_min = D_DISKLESS; - disk_max = D_UP_TO_DATE; - pdsk_min = D_DISKLESS; - pdsk_max = D_UP_TO_DATE; - break; - case C_WF_BITMAP_S: - case C_PAUSED_SYNC_S: - case C_STARTING_SYNC_S: - case C_AHEAD: - disk_min = D_UP_TO_DATE; - disk_max = D_UP_TO_DATE; - pdsk_min = D_INCONSISTENT; - pdsk_max = D_CONSISTENT; /* D_OUTDATED would be nice. But explicit outdate necessary*/ - break; - case C_SYNC_TARGET: - disk_min = D_INCONSISTENT; - disk_max = D_INCONSISTENT; - pdsk_min = D_UP_TO_DATE; - pdsk_max = D_UP_TO_DATE; - break; - case C_SYNC_SOURCE: - disk_min = D_UP_TO_DATE; - disk_max = D_UP_TO_DATE; - pdsk_min = D_INCONSISTENT; - pdsk_max = D_INCONSISTENT; - break; - case C_STANDALONE: - case C_DISCONNECTING: - case C_UNCONNECTED: - case C_TIMEOUT: - case C_BROKEN_PIPE: - case C_NETWORK_FAILURE: - case C_PROTOCOL_ERROR: - case C_TEAR_DOWN: - case C_WF_CONNECTION: - case C_WF_REPORT_PARAMS: - case C_MASK: - break; - } - if (ns.disk > disk_max) - ns.disk = disk_max; - - if (ns.disk < disk_min) { - if (warn) - *warn = IMPLICITLY_UPGRADED_DISK; - ns.disk = disk_min; - } - if (ns.pdsk > pdsk_max) - ns.pdsk = pdsk_max; - - if (ns.pdsk < pdsk_min) { - if (warn) - *warn = IMPLICITLY_UPGRADED_PDSK; - ns.pdsk = pdsk_min; - } - - if (fp == FP_STONITH && - (ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.pdsk > D_OUTDATED) && - !(os.role == R_PRIMARY && os.conn < C_CONNECTED && os.pdsk > D_OUTDATED)) - ns.susp_fen = 1; /* Suspend IO while fence-peer handler runs (peer lost) */ - - if (mdev->sync_conf.on_no_data == OND_SUSPEND_IO && - (ns.role == R_PRIMARY && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE) && - !(os.role == R_PRIMARY && os.disk < D_UP_TO_DATE && os.pdsk < D_UP_TO_DATE)) - ns.susp_nod = 1; /* Suspend IO while no data available (no accessible data available) */ - - if (ns.aftr_isp || ns.peer_isp || ns.user_isp) { - if (ns.conn == C_SYNC_SOURCE) - ns.conn = C_PAUSED_SYNC_S; - if (ns.conn == C_SYNC_TARGET) - ns.conn = C_PAUSED_SYNC_T; - } else { - if (ns.conn == C_PAUSED_SYNC_S) - ns.conn = C_SYNC_SOURCE; - if (ns.conn == C_PAUSED_SYNC_T) - ns.conn = C_SYNC_TARGET; - } - - return ns; -} - -/* helper for __drbd_set_state */ -static void set_ov_position(struct drbd_conf *mdev, enum drbd_conns cs) -{ - if (mdev->agreed_pro_version < 90) - mdev->ov_start_sector = 0; - mdev->rs_total = drbd_bm_bits(mdev); - mdev->ov_position = 0; - if (cs == C_VERIFY_T) { - /* starting online verify from an arbitrary position - * does not fit well into the existing protocol. - * on C_VERIFY_T, we initialize ov_left and friends - * implicitly in receive_DataRequest once the - * first P_OV_REQUEST is received */ - mdev->ov_start_sector = ~(sector_t)0; - } else { - unsigned long bit = BM_SECT_TO_BIT(mdev->ov_start_sector); - if (bit >= mdev->rs_total) { - mdev->ov_start_sector = - BM_BIT_TO_SECT(mdev->rs_total - 1); - mdev->rs_total = 1; - } else - mdev->rs_total -= bit; - mdev->ov_position = mdev->ov_start_sector; - } - mdev->ov_left = mdev->rs_total; -} - -static void drbd_resume_al(struct drbd_conf *mdev) -{ - if (test_and_clear_bit(AL_SUSPENDED, &mdev->flags)) - dev_info(DEV, "Resumed AL updates\n"); -} - -/** - * __drbd_set_state() - Set a new DRBD state - * @mdev: DRBD device. - * @ns: new state. - * @flags: Flags - * @done: Optional completion, that will get completed after the after_state_ch() finished - * - * Caller needs to hold req_lock, and global_state_lock. Do not call directly. - */ -enum drbd_state_rv -__drbd_set_state(struct drbd_conf *mdev, union drbd_state ns, - enum chg_state_flags flags, struct completion *done) -{ - union drbd_state os; - enum drbd_state_rv rv = SS_SUCCESS; - enum sanitize_state_warnings ssw; - struct after_state_chg_work *ascw; - - os = mdev->state; - - ns = sanitize_state(mdev, os, ns, &ssw); - - if (ns.i == os.i) - return SS_NOTHING_TO_DO; - - if (!(flags & CS_HARD)) { - /* pre-state-change checks ; only look at ns */ - /* See drbd_state_sw_errors in drbd_strings.c */ - - rv = is_valid_state(mdev, ns); - if (rv < SS_SUCCESS) { - /* If the old state was illegal as well, then let - this happen...*/ - - if (is_valid_state(mdev, os) == rv) - rv = is_valid_state_transition(mdev, ns, os); - } else - rv = is_valid_state_transition(mdev, ns, os); - } - - if (rv < SS_SUCCESS) { - if (flags & CS_VERBOSE) - print_st_err(mdev, os, ns, rv); - return rv; - } - - print_sanitize_warnings(mdev, ssw); - - { - char *pbp, pb[300]; - pbp = pb; - *pbp = 0; - if (ns.role != os.role) - pbp += sprintf(pbp, "role( %s -> %s ) ", - drbd_role_str(os.role), - drbd_role_str(ns.role)); - if (ns.peer != os.peer) - pbp += sprintf(pbp, "peer( %s -> %s ) ", - drbd_role_str(os.peer), - drbd_role_str(ns.peer)); - if (ns.conn != os.conn) - pbp += sprintf(pbp, "conn( %s -> %s ) ", - drbd_conn_str(os.conn), - drbd_conn_str(ns.conn)); - if (ns.disk != os.disk) - pbp += sprintf(pbp, "disk( %s -> %s ) ", - drbd_disk_str(os.disk), - drbd_disk_str(ns.disk)); - if (ns.pdsk != os.pdsk) - pbp += sprintf(pbp, "pdsk( %s -> %s ) ", - drbd_disk_str(os.pdsk), - drbd_disk_str(ns.pdsk)); - if (is_susp(ns) != is_susp(os)) - pbp += sprintf(pbp, "susp( %d -> %d ) ", - is_susp(os), - is_susp(ns)); - if (ns.aftr_isp != os.aftr_isp) - pbp += sprintf(pbp, "aftr_isp( %d -> %d ) ", - os.aftr_isp, - ns.aftr_isp); - if (ns.peer_isp != os.peer_isp) - pbp += sprintf(pbp, "peer_isp( %d -> %d ) ", - os.peer_isp, - ns.peer_isp); - if (ns.user_isp != os.user_isp) - pbp += sprintf(pbp, "user_isp( %d -> %d ) ", - os.user_isp, - ns.user_isp); - dev_info(DEV, "%s\n", pb); - } - - /* solve the race between becoming unconfigured, - * worker doing the cleanup, and - * admin reconfiguring us: - * on (re)configure, first set CONFIG_PENDING, - * then wait for a potentially exiting worker, - * start the worker, and schedule one no_op. - * then proceed with configuration. - */ - if (ns.disk == D_DISKLESS && - ns.conn == C_STANDALONE && - ns.role == R_SECONDARY && - !test_and_set_bit(CONFIG_PENDING, &mdev->flags)) - set_bit(DEVICE_DYING, &mdev->flags); - - /* if we are going -> D_FAILED or D_DISKLESS, grab one extra reference - * on the ldev here, to be sure the transition -> D_DISKLESS resp. - * drbd_ldev_destroy() won't happen before our corresponding - * after_state_ch works run, where we put_ldev again. */ - if ((os.disk != D_FAILED && ns.disk == D_FAILED) || - (os.disk != D_DISKLESS && ns.disk == D_DISKLESS)) - atomic_inc(&mdev->local_cnt); - - mdev->state = ns; - - if (os.disk == D_ATTACHING && ns.disk >= D_NEGOTIATING) - drbd_print_uuids(mdev, "attached to UUIDs"); - - wake_up(&mdev->misc_wait); - wake_up(&mdev->state_wait); - - /* aborted verify run. log the last position */ - if ((os.conn == C_VERIFY_S || os.conn == C_VERIFY_T) && - ns.conn < C_CONNECTED) { - mdev->ov_start_sector = - BM_BIT_TO_SECT(drbd_bm_bits(mdev) - mdev->ov_left); - dev_info(DEV, "Online Verify reached sector %llu\n", - (unsigned long long)mdev->ov_start_sector); - } - - if ((os.conn == C_PAUSED_SYNC_T || os.conn == C_PAUSED_SYNC_S) && - (ns.conn == C_SYNC_TARGET || ns.conn == C_SYNC_SOURCE)) { - dev_info(DEV, "Syncer continues.\n"); - mdev->rs_paused += (long)jiffies - -(long)mdev->rs_mark_time[mdev->rs_last_mark]; - if (ns.conn == C_SYNC_TARGET) - mod_timer(&mdev->resync_timer, jiffies); - } - - if ((os.conn == C_SYNC_TARGET || os.conn == C_SYNC_SOURCE) && - (ns.conn == C_PAUSED_SYNC_T || ns.conn == C_PAUSED_SYNC_S)) { - dev_info(DEV, "Resync suspended\n"); - mdev->rs_mark_time[mdev->rs_last_mark] = jiffies; - } - - if (os.conn == C_CONNECTED && - (ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T)) { - unsigned long now = jiffies; - int i; - - set_ov_position(mdev, ns.conn); - mdev->rs_start = now; - mdev->rs_last_events = 0; - mdev->rs_last_sect_ev = 0; - mdev->ov_last_oos_size = 0; - mdev->ov_last_oos_start = 0; - - for (i = 0; i < DRBD_SYNC_MARKS; i++) { - mdev->rs_mark_left[i] = mdev->ov_left; - mdev->rs_mark_time[i] = now; - } - - drbd_rs_controller_reset(mdev); - - if (ns.conn == C_VERIFY_S) { - dev_info(DEV, "Starting Online Verify from sector %llu\n", - (unsigned long long)mdev->ov_position); - mod_timer(&mdev->resync_timer, jiffies); - } - } - - if (get_ldev(mdev)) { - u32 mdf = mdev->ldev->md.flags & ~(MDF_CONSISTENT|MDF_PRIMARY_IND| - MDF_CONNECTED_IND|MDF_WAS_UP_TO_DATE| - MDF_PEER_OUT_DATED|MDF_CRASHED_PRIMARY); - - if (test_bit(CRASHED_PRIMARY, &mdev->flags)) - mdf |= MDF_CRASHED_PRIMARY; - if (mdev->state.role == R_PRIMARY || - (mdev->state.pdsk < D_INCONSISTENT && mdev->state.peer == R_PRIMARY)) - mdf |= MDF_PRIMARY_IND; - if (mdev->state.conn > C_WF_REPORT_PARAMS) - mdf |= MDF_CONNECTED_IND; - if (mdev->state.disk > D_INCONSISTENT) - mdf |= MDF_CONSISTENT; - if (mdev->state.disk > D_OUTDATED) - mdf |= MDF_WAS_UP_TO_DATE; - if (mdev->state.pdsk <= D_OUTDATED && mdev->state.pdsk >= D_INCONSISTENT) - mdf |= MDF_PEER_OUT_DATED; - if (mdf != mdev->ldev->md.flags) { - mdev->ldev->md.flags = mdf; - drbd_md_mark_dirty(mdev); - } - if (os.disk < D_CONSISTENT && ns.disk >= D_CONSISTENT) - drbd_set_ed_uuid(mdev, mdev->ldev->md.uuid[UI_CURRENT]); - put_ldev(mdev); - } - - /* Peer was forced D_UP_TO_DATE & R_PRIMARY, consider to resync */ - if (os.disk == D_INCONSISTENT && os.pdsk == D_INCONSISTENT && - os.peer == R_SECONDARY && ns.peer == R_PRIMARY) - set_bit(CONSIDER_RESYNC, &mdev->flags); - - /* Receiver should clean up itself */ - if (os.conn != C_DISCONNECTING && ns.conn == C_DISCONNECTING) - drbd_thread_stop_nowait(&mdev->receiver); - - /* Now the receiver finished cleaning up itself, it should die */ - if (os.conn != C_STANDALONE && ns.conn == C_STANDALONE) - drbd_thread_stop_nowait(&mdev->receiver); - - /* Upon network failure, we need to restart the receiver. */ - if (os.conn > C_WF_CONNECTION && - ns.conn <= C_TEAR_DOWN && ns.conn >= C_TIMEOUT) - drbd_thread_restart_nowait(&mdev->receiver); - - /* Resume AL writing if we get a connection */ - if (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED) - drbd_resume_al(mdev); - - /* remember last connect and attach times so request_timer_fn() won't - * kill newly established sessions while we are still trying to thaw - * previously frozen IO */ - if (os.conn != C_WF_REPORT_PARAMS && ns.conn == C_WF_REPORT_PARAMS) - mdev->last_reconnect_jif = jiffies; - if ((os.disk == D_ATTACHING || os.disk == D_NEGOTIATING) && - ns.disk > D_NEGOTIATING) - mdev->last_reattach_jif = jiffies; - - ascw = kmalloc(sizeof(*ascw), GFP_ATOMIC); - if (ascw) { - ascw->os = os; - ascw->ns = ns; - ascw->flags = flags; - ascw->w.cb = w_after_state_ch; - ascw->done = done; - drbd_queue_work(&mdev->data.work, &ascw->w); - } else { - dev_warn(DEV, "Could not kmalloc an ascw\n"); - } - - return rv; -} - -static int w_after_state_ch(struct drbd_conf *mdev, struct drbd_work *w, int unused) -{ - struct after_state_chg_work *ascw = - container_of(w, struct after_state_chg_work, w); - after_state_ch(mdev, ascw->os, ascw->ns, ascw->flags); - if (ascw->flags & CS_WAIT_COMPLETE) { - D_ASSERT(ascw->done != NULL); - complete(ascw->done); - } - kfree(ascw); - - return 1; -} - -static void abw_start_sync(struct drbd_conf *mdev, int rv) -{ - if (rv) { - dev_err(DEV, "Writing the bitmap failed not starting resync.\n"); - _drbd_request_state(mdev, NS(conn, C_CONNECTED), CS_VERBOSE); - return; - } - - switch (mdev->state.conn) { - case C_STARTING_SYNC_T: - _drbd_request_state(mdev, NS(conn, C_WF_SYNC_UUID), CS_VERBOSE); - break; - case C_STARTING_SYNC_S: - drbd_start_resync(mdev, C_SYNC_SOURCE); - break; - } -} - -int drbd_bitmap_io_from_worker(struct drbd_conf *mdev, - int (*io_fn)(struct drbd_conf *), - char *why, enum bm_flag flags) -{ - int rv; - - D_ASSERT(current == mdev->worker.task); - - /* open coded non-blocking drbd_suspend_io(mdev); */ - set_bit(SUSPEND_IO, &mdev->flags); - - drbd_bm_lock(mdev, why, flags); - rv = io_fn(mdev); - drbd_bm_unlock(mdev); - - drbd_resume_io(mdev); - - return rv; -} - -/** - * after_state_ch() - Perform after state change actions that may sleep - * @mdev: DRBD device. - * @os: old state. - * @ns: new state. - * @flags: Flags - */ -static void after_state_ch(struct drbd_conf *mdev, union drbd_state os, - union drbd_state ns, enum chg_state_flags flags) -{ - enum drbd_fencing_p fp; - enum drbd_req_event what = nothing; - union drbd_state nsm = (union drbd_state){ .i = -1 }; - - if (os.conn != C_CONNECTED && ns.conn == C_CONNECTED) { - clear_bit(CRASHED_PRIMARY, &mdev->flags); - if (mdev->p_uuid) - mdev->p_uuid[UI_FLAGS] &= ~((u64)2); - } - - fp = FP_DONT_CARE; - if (get_ldev(mdev)) { - fp = mdev->ldev->dc.fencing; - put_ldev(mdev); - } - - /* Inform userspace about the change... */ - drbd_bcast_state(mdev, ns); - - if (!(os.role == R_PRIMARY && os.disk < D_UP_TO_DATE && os.pdsk < D_UP_TO_DATE) && - (ns.role == R_PRIMARY && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE)) - drbd_khelper(mdev, "pri-on-incon-degr"); - - /* Here we have the actions that are performed after a - state change. This function might sleep */ - - if (os.disk <= D_NEGOTIATING && ns.disk > D_NEGOTIATING) - mod_timer(&mdev->request_timer, jiffies + HZ); - - nsm.i = -1; - if (ns.susp_nod) { - if (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED) - what = resend; - - if ((os.disk == D_ATTACHING || os.disk == D_NEGOTIATING) && - ns.disk > D_NEGOTIATING) - what = restart_frozen_disk_io; - - if (what != nothing) - nsm.susp_nod = 0; - } - - if (ns.susp_fen) { - /* case1: The outdate peer handler is successful: */ - if (os.pdsk > D_OUTDATED && ns.pdsk <= D_OUTDATED) { - if (test_bit(NEW_CUR_UUID, &mdev->flags)) { - drbd_uuid_new_current(mdev); - clear_bit(NEW_CUR_UUID, &mdev->flags); - } - spin_lock_irq(&mdev->req_lock); - _tl_clear(mdev); - _drbd_set_state(_NS(mdev, susp_fen, 0), CS_VERBOSE, NULL); - spin_unlock_irq(&mdev->req_lock); - } - /* case2: The connection was established again: */ - if (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED) { - clear_bit(NEW_CUR_UUID, &mdev->flags); - what = resend; - nsm.susp_fen = 0; - } - } - - if (what != nothing) { - spin_lock_irq(&mdev->req_lock); - _tl_restart(mdev, what); - nsm.i &= mdev->state.i; - _drbd_set_state(mdev, nsm, CS_VERBOSE, NULL); - spin_unlock_irq(&mdev->req_lock); - } - - /* Became sync source. With protocol >= 96, we still need to send out - * the sync uuid now. Need to do that before any drbd_send_state, or - * the other side may go "paused sync" before receiving the sync uuids, - * which is unexpected. */ - if ((os.conn != C_SYNC_SOURCE && os.conn != C_PAUSED_SYNC_S) && - (ns.conn == C_SYNC_SOURCE || ns.conn == C_PAUSED_SYNC_S) && - mdev->agreed_pro_version >= 96 && get_ldev(mdev)) { - drbd_gen_and_send_sync_uuid(mdev); - put_ldev(mdev); - } - - /* Do not change the order of the if above and the two below... */ - if (os.pdsk == D_DISKLESS && ns.pdsk > D_DISKLESS) { /* attach on the peer */ - /* we probably will start a resync soon. - * make sure those things are properly reset. */ - mdev->rs_total = 0; - mdev->rs_failed = 0; - atomic_set(&mdev->rs_pending_cnt, 0); - drbd_rs_cancel_all(mdev); - - drbd_send_uuids(mdev); - drbd_send_state(mdev, ns); - } - /* No point in queuing send_bitmap if we don't have a connection - * anymore, so check also the _current_ state, not only the new state - * at the time this work was queued. */ - if (os.conn != C_WF_BITMAP_S && ns.conn == C_WF_BITMAP_S && - mdev->state.conn == C_WF_BITMAP_S) - drbd_queue_bitmap_io(mdev, &drbd_send_bitmap, NULL, - "send_bitmap (WFBitMapS)", - BM_LOCKED_TEST_ALLOWED); - - /* Lost contact to peer's copy of the data */ - if ((os.pdsk >= D_INCONSISTENT && - os.pdsk != D_UNKNOWN && - os.pdsk != D_OUTDATED) - && (ns.pdsk < D_INCONSISTENT || - ns.pdsk == D_UNKNOWN || - ns.pdsk == D_OUTDATED)) { - if (get_ldev(mdev)) { - if ((ns.role == R_PRIMARY || ns.peer == R_PRIMARY) && - mdev->ldev->md.uuid[UI_BITMAP] == 0 && ns.disk >= D_UP_TO_DATE) { - if (is_susp(mdev->state)) { - set_bit(NEW_CUR_UUID, &mdev->flags); - } else { - drbd_uuid_new_current(mdev); - drbd_send_uuids(mdev); - } - } - put_ldev(mdev); - } - } - - if (ns.pdsk < D_INCONSISTENT && get_ldev(mdev)) { - if (os.peer == R_SECONDARY && ns.peer == R_PRIMARY && - mdev->ldev->md.uuid[UI_BITMAP] == 0 && ns.disk >= D_UP_TO_DATE) { - drbd_uuid_new_current(mdev); - drbd_send_uuids(mdev); - } - /* D_DISKLESS Peer becomes secondary */ - if (os.peer == R_PRIMARY && ns.peer == R_SECONDARY) - /* We may still be Primary ourselves. - * No harm done if the bitmap still changes, - * redirtied pages will follow later. */ - drbd_bitmap_io_from_worker(mdev, &drbd_bm_write, - "demote diskless peer", BM_LOCKED_SET_ALLOWED); - put_ldev(mdev); - } - - /* Write out all changed bits on demote. - * Though, no need to da that just yet - * if there is a resync going on still */ - if (os.role == R_PRIMARY && ns.role == R_SECONDARY && - mdev->state.conn <= C_CONNECTED && get_ldev(mdev)) { - /* No changes to the bitmap expected this time, so assert that, - * even though no harm was done if it did change. */ - drbd_bitmap_io_from_worker(mdev, &drbd_bm_write, - "demote", BM_LOCKED_TEST_ALLOWED); - put_ldev(mdev); - } - - /* Last part of the attaching process ... */ - if (ns.conn >= C_CONNECTED && - os.disk == D_ATTACHING && ns.disk == D_NEGOTIATING) { - drbd_send_sizes(mdev, 0, 0); /* to start sync... */ - drbd_send_uuids(mdev); - drbd_send_state(mdev, ns); - } - - /* We want to pause/continue resync, tell peer. */ - if (ns.conn >= C_CONNECTED && - ((os.aftr_isp != ns.aftr_isp) || - (os.user_isp != ns.user_isp))) - drbd_send_state(mdev, ns); - - /* In case one of the isp bits got set, suspend other devices. */ - if ((!os.aftr_isp && !os.peer_isp && !os.user_isp) && - (ns.aftr_isp || ns.peer_isp || ns.user_isp)) - suspend_other_sg(mdev); - - /* Make sure the peer gets informed about eventual state - changes (ISP bits) while we were in WFReportParams. */ - if (os.conn == C_WF_REPORT_PARAMS && ns.conn >= C_CONNECTED) - drbd_send_state(mdev, ns); - - if (os.conn != C_AHEAD && ns.conn == C_AHEAD) - drbd_send_state(mdev, ns); - - /* We are in the progress to start a full sync... */ - if ((os.conn != C_STARTING_SYNC_T && ns.conn == C_STARTING_SYNC_T) || - (os.conn != C_STARTING_SYNC_S && ns.conn == C_STARTING_SYNC_S)) - /* no other bitmap changes expected during this phase */ - drbd_queue_bitmap_io(mdev, - &drbd_bmio_set_n_write, &abw_start_sync, - "set_n_write from StartingSync", BM_LOCKED_TEST_ALLOWED); - - /* We are invalidating our self... */ - if (os.conn < C_CONNECTED && ns.conn < C_CONNECTED && - os.disk > D_INCONSISTENT && ns.disk == D_INCONSISTENT) - /* other bitmap operation expected during this phase */ - drbd_queue_bitmap_io(mdev, &drbd_bmio_set_n_write, NULL, - "set_n_write from invalidate", BM_LOCKED_MASK); - - /* first half of local IO error, failure to attach, - * or administrative detach */ - if (os.disk != D_FAILED && ns.disk == D_FAILED) { - enum drbd_io_error_p eh = EP_PASS_ON; - int was_io_error = 0; - /* corresponding get_ldev was in __drbd_set_state, to serialize - * our cleanup here with the transition to D_DISKLESS. - * But is is still not save to dreference ldev here, since - * we might come from an failed Attach before ldev was set. */ - if (mdev->ldev) { - eh = mdev->ldev->dc.on_io_error; - was_io_error = test_and_clear_bit(WAS_IO_ERROR, &mdev->flags); - - if (was_io_error && eh == EP_CALL_HELPER) - drbd_khelper(mdev, "local-io-error"); - - /* Immediately allow completion of all application IO, - * that waits for completion from the local disk, - * if this was a force-detach due to disk_timeout - * or administrator request (drbdsetup detach --force). - * Do NOT abort otherwise. - * Aborting local requests may cause serious problems, - * if requests are completed to upper layers already, - * and then later the already submitted local bio completes. - * This can cause DMA into former bio pages that meanwhile - * have been re-used for other things. - * So aborting local requests may cause crashes, - * or even worse, silent data corruption. - */ - if (test_and_clear_bit(FORCE_DETACH, &mdev->flags)) - tl_abort_disk_io(mdev); - - /* current state still has to be D_FAILED, - * there is only one way out: to D_DISKLESS, - * and that may only happen after our put_ldev below. */ - if (mdev->state.disk != D_FAILED) - dev_err(DEV, - "ASSERT FAILED: disk is %s during detach\n", - drbd_disk_str(mdev->state.disk)); - - if (ns.conn >= C_CONNECTED) - drbd_send_state(mdev, ns); - - drbd_rs_cancel_all(mdev); - - /* In case we want to get something to stable storage still, - * this may be the last chance. - * Following put_ldev may transition to D_DISKLESS. */ - drbd_md_sync(mdev); - } - put_ldev(mdev); - } - - /* second half of local IO error, failure to attach, - * or administrative detach, - * after local_cnt references have reached zero again */ - if (os.disk != D_DISKLESS && ns.disk == D_DISKLESS) { - /* We must still be diskless, - * re-attach has to be serialized with this! */ - if (mdev->state.disk != D_DISKLESS) - dev_err(DEV, - "ASSERT FAILED: disk is %s while going diskless\n", - drbd_disk_str(mdev->state.disk)); - - if (ns.conn >= C_CONNECTED) - drbd_send_state(mdev, ns); - - /* corresponding get_ldev in __drbd_set_state - * this may finally trigger drbd_ldev_destroy. */ - put_ldev(mdev); - } - - /* Notify peer that I had a local IO error, and did not detached.. */ - if (os.disk == D_UP_TO_DATE && ns.disk == D_INCONSISTENT && ns.conn >= C_CONNECTED) - drbd_send_state(mdev, ns); - - /* Disks got bigger while they were detached */ - if (ns.disk > D_NEGOTIATING && ns.pdsk > D_NEGOTIATING && - test_and_clear_bit(RESYNC_AFTER_NEG, &mdev->flags)) { - if (ns.conn == C_CONNECTED) - resync_after_online_grow(mdev); - } - - /* A resync finished or aborted, wake paused devices... */ - if ((os.conn > C_CONNECTED && ns.conn <= C_CONNECTED) || - (os.peer_isp && !ns.peer_isp) || - (os.user_isp && !ns.user_isp)) - resume_next_sg(mdev); - - /* sync target done with resync. Explicitly notify peer, even though - * it should (at least for non-empty resyncs) already know itself. */ - if (os.disk < D_UP_TO_DATE && os.conn >= C_SYNC_SOURCE && ns.conn == C_CONNECTED) - drbd_send_state(mdev, ns); - - /* Wake up role changes, that were delayed because of connection establishing */ - if (os.conn == C_WF_REPORT_PARAMS && ns.conn != C_WF_REPORT_PARAMS) { - clear_bit(STATE_SENT, &mdev->flags); - wake_up(&mdev->state_wait); - } - - /* This triggers bitmap writeout of potentially still unwritten pages - * if the resync finished cleanly, or aborted because of peer disk - * failure, or because of connection loss. - * For resync aborted because of local disk failure, we cannot do - * any bitmap writeout anymore. - * No harm done if some bits change during this phase. - */ - if (os.conn > C_CONNECTED && ns.conn <= C_CONNECTED && get_ldev(mdev)) { - drbd_queue_bitmap_io(mdev, &drbd_bm_write_copy_pages, NULL, - "write from resync_finished", BM_LOCKED_CHANGE_ALLOWED); - put_ldev(mdev); - } - - /* free tl_hash if we Got thawed and are C_STANDALONE */ - if (ns.conn == C_STANDALONE && !is_susp(ns) && mdev->tl_hash) - drbd_free_tl_hash(mdev); - - /* Upon network connection, we need to start the receiver */ - if (os.conn == C_STANDALONE && ns.conn == C_UNCONNECTED) - drbd_thread_start(&mdev->receiver); - - /* Terminate worker thread if we are unconfigured - it will be - restarted as needed... */ - if (ns.disk == D_DISKLESS && - ns.conn == C_STANDALONE && - ns.role == R_SECONDARY) { - if (os.aftr_isp != ns.aftr_isp) - resume_next_sg(mdev); - /* set in __drbd_set_state, unless CONFIG_PENDING was set */ - if (test_bit(DEVICE_DYING, &mdev->flags)) - drbd_thread_stop_nowait(&mdev->worker); + if (req->w.mdev != mdev) + continue; + _req_mod(req, ABORT_DISK_IO); } - - drbd_md_sync(mdev); + spin_unlock_irq(&tconn->req_lock); } - static int drbd_thread_setup(void *arg) { struct drbd_thread *thi = (struct drbd_thread *) arg; - struct drbd_conf *mdev = thi->mdev; + struct drbd_tconn *tconn = thi->tconn; unsigned long flags; int retval; + snprintf(current->comm, sizeof(current->comm), "drbd_%c_%s", + thi->name[0], thi->tconn->name); + restart: retval = thi->function(thi); spin_lock_irqsave(&thi->t_lock, flags); - /* if the receiver has been "Exiting", the last thing it did + /* if the receiver has been "EXITING", the last thing it did * was set the conn state to "StandAlone", * if now a re-connect request comes in, conn state goes C_UNCONNECTED, * and receiver thread will be "started". - * drbd_thread_start needs to set "Restarting" in that case. + * drbd_thread_start needs to set "RESTARTING" in that case. * t_state check and assignment needs to be within the same spinlock, - * so either thread_start sees Exiting, and can remap to Restarting, - * or thread_start see None, and can proceed as normal. + * so either thread_start sees EXITING, and can remap to RESTARTING, + * or thread_start see NONE, and can proceed as normal. */ - if (thi->t_state == Restarting) { - dev_info(DEV, "Restarting %s\n", current->comm); - thi->t_state = Running; + if (thi->t_state == RESTARTING) { + conn_info(tconn, "Restarting %s thread\n", thi->name); + thi->t_state = RUNNING; spin_unlock_irqrestore(&thi->t_lock, flags); goto restart; } thi->task = NULL; - thi->t_state = None; + thi->t_state = NONE; smp_mb(); - complete(&thi->stop); + complete_all(&thi->stop); spin_unlock_irqrestore(&thi->t_lock, flags); - dev_info(DEV, "Terminating %s\n", current->comm); + conn_info(tconn, "Terminating %s\n", current->comm); /* Release mod reference taken when thread was started */ + + kref_put(&tconn->kref, &conn_destroy); module_put(THIS_MODULE); return retval; } -static void drbd_thread_init(struct drbd_conf *mdev, struct drbd_thread *thi, - int (*func) (struct drbd_thread *)) +static void drbd_thread_init(struct drbd_tconn *tconn, struct drbd_thread *thi, + int (*func) (struct drbd_thread *), char *name) { spin_lock_init(&thi->t_lock); thi->task = NULL; - thi->t_state = None; + thi->t_state = NONE; thi->function = func; - thi->mdev = mdev; + thi->tconn = tconn; + strncpy(thi->name, name, ARRAY_SIZE(thi->name)); } int drbd_thread_start(struct drbd_thread *thi) { - struct drbd_conf *mdev = thi->mdev; + struct drbd_tconn *tconn = thi->tconn; struct task_struct *nt; unsigned long flags; - const char *me = - thi == &mdev->receiver ? "receiver" : - thi == &mdev->asender ? "asender" : - thi == &mdev->worker ? "worker" : "NONSENSE"; - /* is used from state engine doing drbd_thread_stop_nowait, * while holding the req lock irqsave */ spin_lock_irqsave(&thi->t_lock, flags); switch (thi->t_state) { - case None: - dev_info(DEV, "Starting %s thread (from %s [%d])\n", - me, current->comm, current->pid); + case NONE: + conn_info(tconn, "Starting %s thread (from %s [%d])\n", + thi->name, current->comm, current->pid); /* Get ref on module for thread - this is released when thread exits */ if (!try_module_get(THIS_MODULE)) { - dev_err(DEV, "Failed to get module reference in drbd_thread_start\n"); + conn_err(tconn, "Failed to get module reference in drbd_thread_start\n"); spin_unlock_irqrestore(&thi->t_lock, flags); return false; } + kref_get(&thi->tconn->kref); + init_completion(&thi->stop); - D_ASSERT(thi->task == NULL); thi->reset_cpu_mask = 1; - thi->t_state = Running; + thi->t_state = RUNNING; spin_unlock_irqrestore(&thi->t_lock, flags); flush_signals(current); /* otherw. may get -ERESTARTNOINTR */ nt = kthread_create(drbd_thread_setup, (void *) thi, - "drbd%d_%s", mdev_to_minor(mdev), me); + "drbd_%c_%s", thi->name[0], thi->tconn->name); if (IS_ERR(nt)) { - dev_err(DEV, "Couldn't start thread\n"); + conn_err(tconn, "Couldn't start thread\n"); + kref_put(&tconn->kref, &conn_destroy); module_put(THIS_MODULE); return false; } spin_lock_irqsave(&thi->t_lock, flags); thi->task = nt; - thi->t_state = Running; + thi->t_state = RUNNING; spin_unlock_irqrestore(&thi->t_lock, flags); wake_up_process(nt); break; - case Exiting: - thi->t_state = Restarting; - dev_info(DEV, "Restarting %s thread (from %s [%d])\n", - me, current->comm, current->pid); + case EXITING: + thi->t_state = RESTARTING; + conn_info(tconn, "Restarting %s thread (from %s [%d])\n", + thi->name, current->comm, current->pid); /* fall through */ - case Running: - case Restarting: + case RUNNING: + case RESTARTING: default: spin_unlock_irqrestore(&thi->t_lock, flags); break; @@ -1867,12 +447,12 @@ void _drbd_thread_stop(struct drbd_thread *thi, int restart, int wait) { unsigned long flags; - enum drbd_thread_state ns = restart ? Restarting : Exiting; + enum drbd_thread_state ns = restart ? RESTARTING : EXITING; /* may be called from state engine, holding the req lock irqsave */ spin_lock_irqsave(&thi->t_lock, flags); - if (thi->t_state == None) { + if (thi->t_state == NONE) { spin_unlock_irqrestore(&thi->t_lock, flags); if (restart) drbd_thread_start(thi); @@ -1890,7 +470,6 @@ void _drbd_thread_stop(struct drbd_thread *thi, int restart, int wait) init_completion(&thi->stop); if (thi->task != current) force_sig(DRBD_SIGKILL, thi->task); - } spin_unlock_irqrestore(&thi->t_lock, flags); @@ -1899,6 +478,35 @@ void _drbd_thread_stop(struct drbd_thread *thi, int restart, int wait) wait_for_completion(&thi->stop); } +static struct drbd_thread *drbd_task_to_thread(struct drbd_tconn *tconn, struct task_struct *task) +{ + struct drbd_thread *thi = + task == tconn->receiver.task ? &tconn->receiver : + task == tconn->asender.task ? &tconn->asender : + task == tconn->worker.task ? &tconn->worker : NULL; + + return thi; +} + +char *drbd_task_to_thread_name(struct drbd_tconn *tconn, struct task_struct *task) +{ + struct drbd_thread *thi = drbd_task_to_thread(tconn, task); + return thi ? thi->name : task->comm; +} + +int conn_lowest_minor(struct drbd_tconn *tconn) +{ + struct drbd_conf *mdev; + int vnr = 0, m; + + rcu_read_lock(); + mdev = idr_get_next(&tconn->volumes, &vnr); + m = mdev ? mdev_to_minor(mdev) : -1; + rcu_read_unlock(); + + return m; +} + #ifdef CONFIG_SMP /** * drbd_calc_cpu_mask() - Generate CPU masks, spread over all CPUs @@ -1907,238 +515,345 @@ void _drbd_thread_stop(struct drbd_thread *thi, int restart, int wait) * Forces all threads of a device onto the same CPU. This is beneficial for * DRBD's performance. May be overwritten by user's configuration. */ -void drbd_calc_cpu_mask(struct drbd_conf *mdev) +void drbd_calc_cpu_mask(struct drbd_tconn *tconn) { int ord, cpu; /* user override. */ - if (cpumask_weight(mdev->cpu_mask)) + if (cpumask_weight(tconn->cpu_mask)) return; - ord = mdev_to_minor(mdev) % cpumask_weight(cpu_online_mask); + ord = conn_lowest_minor(tconn) % cpumask_weight(cpu_online_mask); for_each_online_cpu(cpu) { if (ord-- == 0) { - cpumask_set_cpu(cpu, mdev->cpu_mask); + cpumask_set_cpu(cpu, tconn->cpu_mask); return; } } /* should not be reached */ - cpumask_setall(mdev->cpu_mask); + cpumask_setall(tconn->cpu_mask); } /** * drbd_thread_current_set_cpu() - modifies the cpu mask of the _current_ thread * @mdev: DRBD device. + * @thi: drbd_thread object * * call in the "main loop" of _all_ threads, no need for any mutex, current won't die * prematurely. */ -void drbd_thread_current_set_cpu(struct drbd_conf *mdev) +void drbd_thread_current_set_cpu(struct drbd_thread *thi) { struct task_struct *p = current; - struct drbd_thread *thi = - p == mdev->asender.task ? &mdev->asender : - p == mdev->receiver.task ? &mdev->receiver : - p == mdev->worker.task ? &mdev->worker : - NULL; - ERR_IF(thi == NULL) - return; + if (!thi->reset_cpu_mask) return; thi->reset_cpu_mask = 0; - set_cpus_allowed_ptr(p, mdev->cpu_mask); + set_cpus_allowed_ptr(p, thi->tconn->cpu_mask); } #endif -/* the appropriate socket mutex must be held already */ -int _drbd_send_cmd(struct drbd_conf *mdev, struct socket *sock, - enum drbd_packets cmd, struct p_header80 *h, - size_t size, unsigned msg_flags) +/** + * drbd_header_size - size of a packet header + * + * The header size is a multiple of 8, so any payload following the header is + * word aligned on 64-bit architectures. (The bitmap send and receive code + * relies on this.) + */ +unsigned int drbd_header_size(struct drbd_tconn *tconn) { - int sent, ok; + if (tconn->agreed_pro_version >= 100) { + BUILD_BUG_ON(!IS_ALIGNED(sizeof(struct p_header100), 8)); + return sizeof(struct p_header100); + } else { + BUILD_BUG_ON(sizeof(struct p_header80) != + sizeof(struct p_header95)); + BUILD_BUG_ON(!IS_ALIGNED(sizeof(struct p_header80), 8)); + return sizeof(struct p_header80); + } +} - ERR_IF(!h) return false; - ERR_IF(!size) return false; +static unsigned int prepare_header80(struct p_header80 *h, enum drbd_packet cmd, int size) +{ + h->magic = cpu_to_be32(DRBD_MAGIC); + h->command = cpu_to_be16(cmd); + h->length = cpu_to_be16(size); + return sizeof(struct p_header80); +} - h->magic = BE_DRBD_MAGIC; +static unsigned int prepare_header95(struct p_header95 *h, enum drbd_packet cmd, int size) +{ + h->magic = cpu_to_be16(DRBD_MAGIC_BIG); h->command = cpu_to_be16(cmd); - h->length = cpu_to_be16(size-sizeof(struct p_header80)); + h->length = cpu_to_be32(size); + return sizeof(struct p_header95); +} - sent = drbd_send(mdev, sock, h, size, msg_flags); +static unsigned int prepare_header100(struct p_header100 *h, enum drbd_packet cmd, + int size, int vnr) +{ + h->magic = cpu_to_be32(DRBD_MAGIC_100); + h->volume = cpu_to_be16(vnr); + h->command = cpu_to_be16(cmd); + h->length = cpu_to_be32(size); + h->pad = 0; + return sizeof(struct p_header100); +} - ok = (sent == size); - if (!ok && !signal_pending(current)) - dev_warn(DEV, "short sent %s size=%d sent=%d\n", - cmdname(cmd), (int)size, sent); - return ok; +static unsigned int prepare_header(struct drbd_tconn *tconn, int vnr, + void *buffer, enum drbd_packet cmd, int size) +{ + if (tconn->agreed_pro_version >= 100) + return prepare_header100(buffer, cmd, size, vnr); + else if (tconn->agreed_pro_version >= 95 && + size > DRBD_MAX_SIZE_H80_PACKET) + return prepare_header95(buffer, cmd, size); + else + return prepare_header80(buffer, cmd, size); } -/* don't pass the socket. we may only look at it - * when we hold the appropriate socket mutex. - */ -int drbd_send_cmd(struct drbd_conf *mdev, int use_data_socket, - enum drbd_packets cmd, struct p_header80 *h, size_t size) +static void *__conn_prepare_command(struct drbd_tconn *tconn, + struct drbd_socket *sock) { - int ok = 0; - struct socket *sock; + if (!sock->socket) + return NULL; + return sock->sbuf + drbd_header_size(tconn); +} - if (use_data_socket) { - mutex_lock(&mdev->data.mutex); - sock = mdev->data.socket; - } else { - mutex_lock(&mdev->meta.mutex); - sock = mdev->meta.socket; - } +void *conn_prepare_command(struct drbd_tconn *tconn, struct drbd_socket *sock) +{ + void *p; - /* drbd_disconnect() could have called drbd_free_sock() - * while we were waiting in down()... */ - if (likely(sock != NULL)) - ok = _drbd_send_cmd(mdev, sock, cmd, h, size, 0); + mutex_lock(&sock->mutex); + p = __conn_prepare_command(tconn, sock); + if (!p) + mutex_unlock(&sock->mutex); - if (use_data_socket) - mutex_unlock(&mdev->data.mutex); - else - mutex_unlock(&mdev->meta.mutex); - return ok; + return p; } -int drbd_send_cmd2(struct drbd_conf *mdev, enum drbd_packets cmd, char *data, - size_t size) +void *drbd_prepare_command(struct drbd_conf *mdev, struct drbd_socket *sock) { - struct p_header80 h; - int ok; + return conn_prepare_command(mdev->tconn, sock); +} - h.magic = BE_DRBD_MAGIC; - h.command = cpu_to_be16(cmd); - h.length = cpu_to_be16(size); +static int __send_command(struct drbd_tconn *tconn, int vnr, + struct drbd_socket *sock, enum drbd_packet cmd, + unsigned int header_size, void *data, + unsigned int size) +{ + int msg_flags; + int err; - if (!drbd_get_data_sock(mdev)) - return 0; + /* + * Called with @data == NULL and the size of the data blocks in @size + * for commands that send data blocks. For those commands, omit the + * MSG_MORE flag: this will increase the likelihood that data blocks + * which are page aligned on the sender will end up page aligned on the + * receiver. + */ + msg_flags = data ? MSG_MORE : 0; + + header_size += prepare_header(tconn, vnr, sock->sbuf, cmd, + header_size + size); + err = drbd_send_all(tconn, sock->socket, sock->sbuf, header_size, + msg_flags); + if (data && !err) + err = drbd_send_all(tconn, sock->socket, data, size, 0); + return err; +} - ok = (sizeof(h) == - drbd_send(mdev, mdev->data.socket, &h, sizeof(h), 0)); - ok = ok && (size == - drbd_send(mdev, mdev->data.socket, data, size, 0)); +static int __conn_send_command(struct drbd_tconn *tconn, struct drbd_socket *sock, + enum drbd_packet cmd, unsigned int header_size, + void *data, unsigned int size) +{ + return __send_command(tconn, 0, sock, cmd, header_size, data, size); +} - drbd_put_data_sock(mdev); +int conn_send_command(struct drbd_tconn *tconn, struct drbd_socket *sock, + enum drbd_packet cmd, unsigned int header_size, + void *data, unsigned int size) +{ + int err; - return ok; + err = __conn_send_command(tconn, sock, cmd, header_size, data, size); + mutex_unlock(&sock->mutex); + return err; } -int drbd_send_sync_param(struct drbd_conf *mdev, struct syncer_conf *sc) +int drbd_send_command(struct drbd_conf *mdev, struct drbd_socket *sock, + enum drbd_packet cmd, unsigned int header_size, + void *data, unsigned int size) { + int err; + + err = __send_command(mdev->tconn, mdev->vnr, sock, cmd, header_size, + data, size); + mutex_unlock(&sock->mutex); + return err; +} + +int drbd_send_ping(struct drbd_tconn *tconn) +{ + struct drbd_socket *sock; + + sock = &tconn->meta; + if (!conn_prepare_command(tconn, sock)) + return -EIO; + return conn_send_command(tconn, sock, P_PING, 0, NULL, 0); +} + +int drbd_send_ping_ack(struct drbd_tconn *tconn) +{ + struct drbd_socket *sock; + + sock = &tconn->meta; + if (!conn_prepare_command(tconn, sock)) + return -EIO; + return conn_send_command(tconn, sock, P_PING_ACK, 0, NULL, 0); +} + +int drbd_send_sync_param(struct drbd_conf *mdev) +{ + struct drbd_socket *sock; struct p_rs_param_95 *p; - struct socket *sock; - int size, rv; - const int apv = mdev->agreed_pro_version; + int size; + const int apv = mdev->tconn->agreed_pro_version; + enum drbd_packet cmd; + struct net_conf *nc; + struct disk_conf *dc; + + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); size = apv <= 87 ? sizeof(struct p_rs_param) : apv == 88 ? sizeof(struct p_rs_param) - + strlen(mdev->sync_conf.verify_alg) + 1 + + strlen(nc->verify_alg) + 1 : apv <= 94 ? sizeof(struct p_rs_param_89) : /* apv >= 95 */ sizeof(struct p_rs_param_95); - /* used from admin command context and receiver/worker context. - * to avoid kmalloc, grab the socket right here, - * then use the pre-allocated sbuf there */ - mutex_lock(&mdev->data.mutex); - sock = mdev->data.socket; - - if (likely(sock != NULL)) { - enum drbd_packets cmd = apv >= 89 ? P_SYNC_PARAM89 : P_SYNC_PARAM; - - p = &mdev->data.sbuf.rs_param_95; + cmd = apv >= 89 ? P_SYNC_PARAM89 : P_SYNC_PARAM; - /* initialize verify_alg and csums_alg */ - memset(p->verify_alg, 0, 2 * SHARED_SECRET_MAX); + /* initialize verify_alg and csums_alg */ + memset(p->verify_alg, 0, 2 * SHARED_SECRET_MAX); - p->rate = cpu_to_be32(sc->rate); - p->c_plan_ahead = cpu_to_be32(sc->c_plan_ahead); - p->c_delay_target = cpu_to_be32(sc->c_delay_target); - p->c_fill_target = cpu_to_be32(sc->c_fill_target); - p->c_max_rate = cpu_to_be32(sc->c_max_rate); - - if (apv >= 88) - strcpy(p->verify_alg, mdev->sync_conf.verify_alg); - if (apv >= 89) - strcpy(p->csums_alg, mdev->sync_conf.csums_alg); - - rv = _drbd_send_cmd(mdev, sock, cmd, &p->head, size, 0); - } else - rv = 0; /* not ok */ + if (get_ldev(mdev)) { + dc = rcu_dereference(mdev->ldev->disk_conf); + p->resync_rate = cpu_to_be32(dc->resync_rate); + p->c_plan_ahead = cpu_to_be32(dc->c_plan_ahead); + p->c_delay_target = cpu_to_be32(dc->c_delay_target); + p->c_fill_target = cpu_to_be32(dc->c_fill_target); + p->c_max_rate = cpu_to_be32(dc->c_max_rate); + put_ldev(mdev); + } else { + p->resync_rate = cpu_to_be32(DRBD_RESYNC_RATE_DEF); + p->c_plan_ahead = cpu_to_be32(DRBD_C_PLAN_AHEAD_DEF); + p->c_delay_target = cpu_to_be32(DRBD_C_DELAY_TARGET_DEF); + p->c_fill_target = cpu_to_be32(DRBD_C_FILL_TARGET_DEF); + p->c_max_rate = cpu_to_be32(DRBD_C_MAX_RATE_DEF); + } - mutex_unlock(&mdev->data.mutex); + if (apv >= 88) + strcpy(p->verify_alg, nc->verify_alg); + if (apv >= 89) + strcpy(p->csums_alg, nc->csums_alg); + rcu_read_unlock(); - return rv; + return drbd_send_command(mdev, sock, cmd, size, NULL, 0); } -int drbd_send_protocol(struct drbd_conf *mdev) +int __drbd_send_protocol(struct drbd_tconn *tconn, enum drbd_packet cmd) { + struct drbd_socket *sock; struct p_protocol *p; - int size, cf, rv; + struct net_conf *nc; + int size, cf; - size = sizeof(struct p_protocol); + sock = &tconn->data; + p = __conn_prepare_command(tconn, sock); + if (!p) + return -EIO; - if (mdev->agreed_pro_version >= 87) - size += strlen(mdev->net_conf->integrity_alg) + 1; + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); - /* we must not recurse into our own queue, - * as that is blocked during handshake */ - p = kmalloc(size, GFP_NOIO); - if (p == NULL) - return 0; + if (nc->tentative && tconn->agreed_pro_version < 92) { + rcu_read_unlock(); + mutex_unlock(&sock->mutex); + conn_err(tconn, "--dry-run is not supported by peer"); + return -EOPNOTSUPP; + } - p->protocol = cpu_to_be32(mdev->net_conf->wire_protocol); - p->after_sb_0p = cpu_to_be32(mdev->net_conf->after_sb_0p); - p->after_sb_1p = cpu_to_be32(mdev->net_conf->after_sb_1p); - p->after_sb_2p = cpu_to_be32(mdev->net_conf->after_sb_2p); - p->two_primaries = cpu_to_be32(mdev->net_conf->two_primaries); + size = sizeof(*p); + if (tconn->agreed_pro_version >= 87) + size += strlen(nc->integrity_alg) + 1; + p->protocol = cpu_to_be32(nc->wire_protocol); + p->after_sb_0p = cpu_to_be32(nc->after_sb_0p); + p->after_sb_1p = cpu_to_be32(nc->after_sb_1p); + p->after_sb_2p = cpu_to_be32(nc->after_sb_2p); + p->two_primaries = cpu_to_be32(nc->two_primaries); cf = 0; - if (mdev->net_conf->want_lose) - cf |= CF_WANT_LOSE; - if (mdev->net_conf->dry_run) { - if (mdev->agreed_pro_version >= 92) - cf |= CF_DRY_RUN; - else { - dev_err(DEV, "--dry-run is not supported by peer"); - kfree(p); - return -1; - } - } + if (nc->discard_my_data) + cf |= CF_DISCARD_MY_DATA; + if (nc->tentative) + cf |= CF_DRY_RUN; p->conn_flags = cpu_to_be32(cf); - if (mdev->agreed_pro_version >= 87) - strcpy(p->integrity_alg, mdev->net_conf->integrity_alg); + if (tconn->agreed_pro_version >= 87) + strcpy(p->integrity_alg, nc->integrity_alg); + rcu_read_unlock(); - rv = drbd_send_cmd(mdev, USE_DATA_SOCKET, P_PROTOCOL, - (struct p_header80 *)p, size); - kfree(p); - return rv; + return __conn_send_command(tconn, sock, cmd, size, NULL, 0); +} + +int drbd_send_protocol(struct drbd_tconn *tconn) +{ + int err; + + mutex_lock(&tconn->data.mutex); + err = __drbd_send_protocol(tconn, P_PROTOCOL); + mutex_unlock(&tconn->data.mutex); + + return err; } int _drbd_send_uuids(struct drbd_conf *mdev, u64 uuid_flags) { - struct p_uuids p; + struct drbd_socket *sock; + struct p_uuids *p; int i; if (!get_ldev_if_state(mdev, D_NEGOTIATING)) - return 1; + return 0; + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) { + put_ldev(mdev); + return -EIO; + } + spin_lock_irq(&mdev->ldev->md.uuid_lock); for (i = UI_CURRENT; i < UI_SIZE; i++) - p.uuid[i] = mdev->ldev ? cpu_to_be64(mdev->ldev->md.uuid[i]) : 0; + p->uuid[i] = cpu_to_be64(mdev->ldev->md.uuid[i]); + spin_unlock_irq(&mdev->ldev->md.uuid_lock); mdev->comm_bm_set = drbd_bm_total_weight(mdev); - p.uuid[UI_SIZE] = cpu_to_be64(mdev->comm_bm_set); - uuid_flags |= mdev->net_conf->want_lose ? 1 : 0; + p->uuid[UI_SIZE] = cpu_to_be64(mdev->comm_bm_set); + rcu_read_lock(); + uuid_flags |= rcu_dereference(mdev->tconn->net_conf)->discard_my_data ? 1 : 0; + rcu_read_unlock(); uuid_flags |= test_bit(CRASHED_PRIMARY, &mdev->flags) ? 2 : 0; uuid_flags |= mdev->new_state_tmp.disk == D_INCONSISTENT ? 4 : 0; - p.uuid[UI_FLAGS] = cpu_to_be64(uuid_flags); + p->uuid[UI_FLAGS] = cpu_to_be64(uuid_flags); put_ldev(mdev); - - return drbd_send_cmd(mdev, USE_DATA_SOCKET, P_UUIDS, - (struct p_header80 *)&p, sizeof(p)); + return drbd_send_command(mdev, sock, P_UUIDS, sizeof(*p), NULL, 0); } int drbd_send_uuids(struct drbd_conf *mdev) @@ -2169,9 +884,10 @@ void drbd_print_uuids(struct drbd_conf *mdev, const char *text) } } -int drbd_gen_and_send_sync_uuid(struct drbd_conf *mdev) +void drbd_gen_and_send_sync_uuid(struct drbd_conf *mdev) { - struct p_rs_uuid p; + struct drbd_socket *sock; + struct p_rs_uuid *p; u64 uuid; D_ASSERT(mdev->state.disk == D_UP_TO_DATE); @@ -2184,24 +900,29 @@ int drbd_gen_and_send_sync_uuid(struct drbd_conf *mdev) drbd_uuid_set(mdev, UI_BITMAP, uuid); drbd_print_uuids(mdev, "updated sync UUID"); drbd_md_sync(mdev); - p.uuid = cpu_to_be64(uuid); - return drbd_send_cmd(mdev, USE_DATA_SOCKET, P_SYNC_UUID, - (struct p_header80 *)&p, sizeof(p)); + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (p) { + p->uuid = cpu_to_be64(uuid); + drbd_send_command(mdev, sock, P_SYNC_UUID, sizeof(*p), NULL, 0); + } } int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags flags) { - struct p_sizes p; + struct drbd_socket *sock; + struct p_sizes *p; sector_t d_size, u_size; int q_order_type; unsigned int max_bio_size; - int ok; if (get_ldev_if_state(mdev, D_NEGOTIATING)) { D_ASSERT(mdev->ldev->backing_bdev); d_size = drbd_get_max_capacity(mdev->ldev); - u_size = mdev->ldev->dc.disk_size; + rcu_read_lock(); + u_size = rcu_dereference(mdev->ldev->disk_conf)->disk_size; + rcu_read_unlock(); q_order_type = drbd_queue_order_type(mdev); max_bio_size = queue_max_hw_sectors(mdev->ldev->backing_bdev->bd_disk->queue) << 9; max_bio_size = min(max_bio_size, DRBD_MAX_BIO_SIZE); @@ -2213,20 +934,23 @@ int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags fl max_bio_size = DRBD_MAX_BIO_SIZE; /* ... multiple BIOs per peer_request */ } - /* Never allow old drbd (up to 8.3.7) to see more than 32KiB */ - if (mdev->agreed_pro_version <= 94) - max_bio_size = min(max_bio_size, DRBD_MAX_SIZE_H80_PACKET); + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; - p.d_size = cpu_to_be64(d_size); - p.u_size = cpu_to_be64(u_size); - p.c_size = cpu_to_be64(trigger_reply ? 0 : drbd_get_capacity(mdev->this_bdev)); - p.max_bio_size = cpu_to_be32(max_bio_size); - p.queue_order_type = cpu_to_be16(q_order_type); - p.dds_flags = cpu_to_be16(flags); + if (mdev->tconn->agreed_pro_version <= 94) + max_bio_size = min(max_bio_size, DRBD_MAX_SIZE_H80_PACKET); + else if (mdev->tconn->agreed_pro_version < 100) + max_bio_size = min(max_bio_size, DRBD_MAX_BIO_SIZE_P95); - ok = drbd_send_cmd(mdev, USE_DATA_SOCKET, P_SIZES, - (struct p_header80 *)&p, sizeof(p)); - return ok; + p->d_size = cpu_to_be64(d_size); + p->u_size = cpu_to_be64(u_size); + p->c_size = cpu_to_be64(trigger_reply ? 0 : drbd_get_capacity(mdev->this_bdev)); + p->max_bio_size = cpu_to_be32(max_bio_size); + p->queue_order_type = cpu_to_be16(q_order_type); + p->dds_flags = cpu_to_be16(flags); + return drbd_send_command(mdev, sock, P_SIZES, sizeof(*p), NULL, 0); } /** @@ -2235,34 +959,21 @@ int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags fl */ int drbd_send_current_state(struct drbd_conf *mdev) { - struct socket *sock; - struct p_state p; - int ok = 0; - - /* Grab state lock so we wont send state if we're in the middle - * of a cluster wide state change on another thread */ - drbd_state_lock(mdev); - - mutex_lock(&mdev->data.mutex); - - p.state = cpu_to_be32(mdev->state.i); /* Within the send mutex */ - sock = mdev->data.socket; + struct drbd_socket *sock; + struct p_state *p; - if (likely(sock != NULL)) { - ok = _drbd_send_cmd(mdev, sock, P_STATE, - (struct p_header80 *)&p, sizeof(p), 0); - } - - mutex_unlock(&mdev->data.mutex); - - drbd_state_unlock(mdev); - return ok; + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + p->state = cpu_to_be32(mdev->state.i); /* Within the send mutex */ + return drbd_send_command(mdev, sock, P_STATE, sizeof(*p), NULL, 0); } /** * drbd_send_state() - After a state change, sends the new state to the peer - * @mdev: DRBD device. - * @state: the state to send, not necessarily the current state. + * @mdev: DRBD device. + * @state: the state to send, not necessarily the current state. * * Each state change queues an "after_state_ch" work, which will eventually * send the resulting new state to the peer. If more state changes happen @@ -2271,50 +982,95 @@ int drbd_send_current_state(struct drbd_conf *mdev) */ int drbd_send_state(struct drbd_conf *mdev, union drbd_state state) { - struct socket *sock; - struct p_state p; - int ok = 0; + struct drbd_socket *sock; + struct p_state *p; - mutex_lock(&mdev->data.mutex); + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + p->state = cpu_to_be32(state.i); /* Within the send mutex */ + return drbd_send_command(mdev, sock, P_STATE, sizeof(*p), NULL, 0); +} - p.state = cpu_to_be32(state.i); - sock = mdev->data.socket; +int drbd_send_state_req(struct drbd_conf *mdev, union drbd_state mask, union drbd_state val) +{ + struct drbd_socket *sock; + struct p_req_state *p; - if (likely(sock != NULL)) { - ok = _drbd_send_cmd(mdev, sock, P_STATE, - (struct p_header80 *)&p, sizeof(p), 0); - } + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + p->mask = cpu_to_be32(mask.i); + p->val = cpu_to_be32(val.i); + return drbd_send_command(mdev, sock, P_STATE_CHG_REQ, sizeof(*p), NULL, 0); +} - mutex_unlock(&mdev->data.mutex); +int conn_send_state_req(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val) +{ + enum drbd_packet cmd; + struct drbd_socket *sock; + struct p_req_state *p; - return ok; + cmd = tconn->agreed_pro_version < 100 ? P_STATE_CHG_REQ : P_CONN_ST_CHG_REQ; + sock = &tconn->data; + p = conn_prepare_command(tconn, sock); + if (!p) + return -EIO; + p->mask = cpu_to_be32(mask.i); + p->val = cpu_to_be32(val.i); + return conn_send_command(tconn, sock, cmd, sizeof(*p), NULL, 0); } -int drbd_send_state_req(struct drbd_conf *mdev, - union drbd_state mask, union drbd_state val) +void drbd_send_sr_reply(struct drbd_conf *mdev, enum drbd_state_rv retcode) { - struct p_req_state p; + struct drbd_socket *sock; + struct p_req_state_reply *p; - p.mask = cpu_to_be32(mask.i); - p.val = cpu_to_be32(val.i); + sock = &mdev->tconn->meta; + p = drbd_prepare_command(mdev, sock); + if (p) { + p->retcode = cpu_to_be32(retcode); + drbd_send_command(mdev, sock, P_STATE_CHG_REPLY, sizeof(*p), NULL, 0); + } +} + +void conn_send_sr_reply(struct drbd_tconn *tconn, enum drbd_state_rv retcode) +{ + struct drbd_socket *sock; + struct p_req_state_reply *p; + enum drbd_packet cmd = tconn->agreed_pro_version < 100 ? P_STATE_CHG_REPLY : P_CONN_ST_CHG_REPLY; - return drbd_send_cmd(mdev, USE_DATA_SOCKET, P_STATE_CHG_REQ, - (struct p_header80 *)&p, sizeof(p)); + sock = &tconn->meta; + p = conn_prepare_command(tconn, sock); + if (p) { + p->retcode = cpu_to_be32(retcode); + conn_send_command(tconn, sock, cmd, sizeof(*p), NULL, 0); + } } -int drbd_send_sr_reply(struct drbd_conf *mdev, enum drbd_state_rv retcode) +static void dcbp_set_code(struct p_compressed_bm *p, enum drbd_bitmap_code code) { - struct p_req_state_reply p; + BUG_ON(code & ~0xf); + p->encoding = (p->encoding & ~0xf) | code; +} - p.retcode = cpu_to_be32(retcode); +static void dcbp_set_start(struct p_compressed_bm *p, int set) +{ + p->encoding = (p->encoding & ~0x80) | (set ? 0x80 : 0); +} - return drbd_send_cmd(mdev, USE_META_SOCKET, P_STATE_CHG_REPLY, - (struct p_header80 *)&p, sizeof(p)); +static void dcbp_set_pad_bits(struct p_compressed_bm *p, int n) +{ + BUG_ON(n & ~0x7); + p->encoding = (p->encoding & (~0x7 << 4)) | (n << 4); } int fill_bitmap_rle_bits(struct drbd_conf *mdev, - struct p_compressed_bm *p, - struct bm_xfer_ctx *c) + struct p_compressed_bm *p, + unsigned int size, + struct bm_xfer_ctx *c) { struct bitstream bs; unsigned long plain_bits; @@ -2322,19 +1078,21 @@ int fill_bitmap_rle_bits(struct drbd_conf *mdev, unsigned long rl; unsigned len; unsigned toggle; - int bits; + int bits, use_rle; /* may we use this feature? */ - if ((mdev->sync_conf.use_rle == 0) || - (mdev->agreed_pro_version < 90)) - return 0; + rcu_read_lock(); + use_rle = rcu_dereference(mdev->tconn->net_conf)->use_rle; + rcu_read_unlock(); + if (!use_rle || mdev->tconn->agreed_pro_version < 90) + return 0; if (c->bit_offset >= c->bm_bits) return 0; /* nothing to do. */ /* use at most thus many bytes */ - bitstream_init(&bs, p->code, BM_PACKET_VLI_BYTES_MAX, 0); - memset(p->code, 0, BM_PACKET_VLI_BYTES_MAX); + bitstream_init(&bs, p->code, size, 0); + memset(p->code, 0, size); /* plain bits covered in this code string */ plain_bits = 0; @@ -2356,12 +1114,12 @@ int fill_bitmap_rle_bits(struct drbd_conf *mdev, if (rl == 0) { /* the first checked bit was set, * store start value, */ - DCBP_set_start(p, 1); + dcbp_set_start(p, 1); /* but skip encoding of zero run length */ toggle = !toggle; continue; } - DCBP_set_start(p, 0); + dcbp_set_start(p, 0); } /* paranoia: catch zero runlength. @@ -2401,7 +1159,7 @@ int fill_bitmap_rle_bits(struct drbd_conf *mdev, bm_xfer_ctx_bit_to_word_offset(c); /* store pad_bits */ - DCBP_set_pad_bits(p, (8 - bs.cur.bit) & 0x7); + dcbp_set_pad_bits(p, (8 - bs.cur.bit) & 0x7); return len; } @@ -2413,48 +1171,52 @@ int fill_bitmap_rle_bits(struct drbd_conf *mdev, * code upon failure. */ static int -send_bitmap_rle_or_plain(struct drbd_conf *mdev, - struct p_header80 *h, struct bm_xfer_ctx *c) +send_bitmap_rle_or_plain(struct drbd_conf *mdev, struct bm_xfer_ctx *c) { - struct p_compressed_bm *p = (void*)h; - unsigned long num_words; - int len; - int ok; - - len = fill_bitmap_rle_bits(mdev, p, c); + struct drbd_socket *sock = &mdev->tconn->data; + unsigned int header_size = drbd_header_size(mdev->tconn); + struct p_compressed_bm *p = sock->sbuf + header_size; + int len, err; + len = fill_bitmap_rle_bits(mdev, p, + DRBD_SOCKET_BUFFER_SIZE - header_size - sizeof(*p), c); if (len < 0) return -EIO; if (len) { - DCBP_set_code(p, RLE_VLI_Bits); - ok = _drbd_send_cmd(mdev, mdev->data.socket, P_COMPRESSED_BITMAP, h, - sizeof(*p) + len, 0); - + dcbp_set_code(p, RLE_VLI_Bits); + err = __send_command(mdev->tconn, mdev->vnr, sock, + P_COMPRESSED_BITMAP, sizeof(*p) + len, + NULL, 0); c->packets[0]++; - c->bytes[0] += sizeof(*p) + len; + c->bytes[0] += header_size + sizeof(*p) + len; if (c->bit_offset >= c->bm_bits) len = 0; /* DONE */ } else { /* was not compressible. * send a buffer full of plain text bits instead. */ - num_words = min_t(size_t, BM_PACKET_WORDS, c->bm_words - c->word_offset); - len = num_words * sizeof(long); + unsigned int data_size; + unsigned long num_words; + unsigned long *p = sock->sbuf + header_size; + + data_size = DRBD_SOCKET_BUFFER_SIZE - header_size; + num_words = min_t(size_t, data_size / sizeof(*p), + c->bm_words - c->word_offset); + len = num_words * sizeof(*p); if (len) - drbd_bm_get_lel(mdev, c->word_offset, num_words, (unsigned long*)h->payload); - ok = _drbd_send_cmd(mdev, mdev->data.socket, P_BITMAP, - h, sizeof(struct p_header80) + len, 0); + drbd_bm_get_lel(mdev, c->word_offset, num_words, p); + err = __send_command(mdev->tconn, mdev->vnr, sock, P_BITMAP, len, NULL, 0); c->word_offset += num_words; c->bit_offset = c->word_offset * BITS_PER_LONG; c->packets[1]++; - c->bytes[1] += sizeof(struct p_header80) + len; + c->bytes[1] += header_size + len; if (c->bit_offset > c->bm_bits) c->bit_offset = c->bm_bits; } - if (ok) { + if (!err) { if (len == 0) { INFO_bm_xfer_stats(mdev, "send", c); return 0; @@ -2465,21 +1227,13 @@ send_bitmap_rle_or_plain(struct drbd_conf *mdev, } /* See the comment at receive_bitmap() */ -int _drbd_send_bitmap(struct drbd_conf *mdev) +static int _drbd_send_bitmap(struct drbd_conf *mdev) { struct bm_xfer_ctx c; - struct p_header80 *p; int err; - ERR_IF(!mdev->bitmap) return false; - - /* maybe we should use some per thread scratch page, - * and allocate that during initial device creation? */ - p = (struct p_header80 *) __get_free_page(GFP_NOIO); - if (!p) { - dev_err(DEV, "failed to allocate one page buffer in %s\n", __func__); + if (!expect(mdev->bitmap)) return false; - } if (get_ldev(mdev)) { if (drbd_md_test_flag(mdev->ldev, MDF_FULL_SYNC)) { @@ -2504,37 +1258,39 @@ int _drbd_send_bitmap(struct drbd_conf *mdev) }; do { - err = send_bitmap_rle_or_plain(mdev, p, &c); + err = send_bitmap_rle_or_plain(mdev, &c); } while (err > 0); - free_page((unsigned long) p); return err == 0; } int drbd_send_bitmap(struct drbd_conf *mdev) { - int err; + struct drbd_socket *sock = &mdev->tconn->data; + int err = -1; - if (!drbd_get_data_sock(mdev)) - return -1; - err = !_drbd_send_bitmap(mdev); - drbd_put_data_sock(mdev); + mutex_lock(&sock->mutex); + if (sock->socket) + err = !_drbd_send_bitmap(mdev); + mutex_unlock(&sock->mutex); return err; } -int drbd_send_b_ack(struct drbd_conf *mdev, u32 barrier_nr, u32 set_size) +void drbd_send_b_ack(struct drbd_tconn *tconn, u32 barrier_nr, u32 set_size) { - int ok; - struct p_barrier_ack p; + struct drbd_socket *sock; + struct p_barrier_ack *p; - p.barrier = barrier_nr; - p.set_size = cpu_to_be32(set_size); + if (tconn->cstate < C_WF_REPORT_PARAMS) + return; - if (mdev->state.conn < C_CONNECTED) - return false; - ok = drbd_send_cmd(mdev, USE_META_SOCKET, P_BARRIER_ACK, - (struct p_header80 *)&p, sizeof(p)); - return ok; + sock = &tconn->meta; + p = conn_prepare_command(tconn, sock); + if (!p) + return; + p->barrier = barrier_nr; + p->set_size = cpu_to_be32(set_size); + conn_send_command(tconn, sock, P_BARRIER_ACK, sizeof(*p), NULL, 0); } /** @@ -2545,62 +1301,62 @@ int drbd_send_b_ack(struct drbd_conf *mdev, u32 barrier_nr, u32 set_size) * @blksize: size in byte, needs to be in big endian byte order * @block_id: Id, big endian byte order */ -static int _drbd_send_ack(struct drbd_conf *mdev, enum drbd_packets cmd, - u64 sector, - u32 blksize, - u64 block_id) +static int _drbd_send_ack(struct drbd_conf *mdev, enum drbd_packet cmd, + u64 sector, u32 blksize, u64 block_id) { - int ok; - struct p_block_ack p; + struct drbd_socket *sock; + struct p_block_ack *p; - p.sector = sector; - p.block_id = block_id; - p.blksize = blksize; - p.seq_num = cpu_to_be32(atomic_add_return(1, &mdev->packet_seq)); + if (mdev->state.conn < C_CONNECTED) + return -EIO; - if (!mdev->meta.socket || mdev->state.conn < C_CONNECTED) - return false; - ok = drbd_send_cmd(mdev, USE_META_SOCKET, cmd, - (struct p_header80 *)&p, sizeof(p)); - return ok; + sock = &mdev->tconn->meta; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + p->sector = sector; + p->block_id = block_id; + p->blksize = blksize; + p->seq_num = cpu_to_be32(atomic_inc_return(&mdev->packet_seq)); + return drbd_send_command(mdev, sock, cmd, sizeof(*p), NULL, 0); } /* dp->sector and dp->block_id already/still in network byte order, * data_size is payload size according to dp->head, * and may need to be corrected for digest size. */ -int drbd_send_ack_dp(struct drbd_conf *mdev, enum drbd_packets cmd, - struct p_data *dp, int data_size) +void drbd_send_ack_dp(struct drbd_conf *mdev, enum drbd_packet cmd, + struct p_data *dp, int data_size) { - data_size -= (mdev->agreed_pro_version >= 87 && mdev->integrity_r_tfm) ? - crypto_hash_digestsize(mdev->integrity_r_tfm) : 0; - return _drbd_send_ack(mdev, cmd, dp->sector, cpu_to_be32(data_size), - dp->block_id); + if (mdev->tconn->peer_integrity_tfm) + data_size -= crypto_hash_digestsize(mdev->tconn->peer_integrity_tfm); + _drbd_send_ack(mdev, cmd, dp->sector, cpu_to_be32(data_size), + dp->block_id); } -int drbd_send_ack_rp(struct drbd_conf *mdev, enum drbd_packets cmd, - struct p_block_req *rp) +void drbd_send_ack_rp(struct drbd_conf *mdev, enum drbd_packet cmd, + struct p_block_req *rp) { - return _drbd_send_ack(mdev, cmd, rp->sector, rp->blksize, rp->block_id); + _drbd_send_ack(mdev, cmd, rp->sector, rp->blksize, rp->block_id); } /** * drbd_send_ack() - Sends an ack packet - * @mdev: DRBD device. - * @cmd: Packet command code. - * @e: Epoch entry. + * @mdev: DRBD device + * @cmd: packet command code + * @peer_req: peer request */ -int drbd_send_ack(struct drbd_conf *mdev, - enum drbd_packets cmd, struct drbd_epoch_entry *e) +int drbd_send_ack(struct drbd_conf *mdev, enum drbd_packet cmd, + struct drbd_peer_request *peer_req) { return _drbd_send_ack(mdev, cmd, - cpu_to_be64(e->sector), - cpu_to_be32(e->size), - e->block_id); + cpu_to_be64(peer_req->i.sector), + cpu_to_be32(peer_req->i.size), + peer_req->block_id); } /* This function misuses the block_id field to signal if the blocks * are is sync or not. */ -int drbd_send_ack_ex(struct drbd_conf *mdev, enum drbd_packets cmd, +int drbd_send_ack_ex(struct drbd_conf *mdev, enum drbd_packet cmd, sector_t sector, int blksize, u64 block_id) { return _drbd_send_ack(mdev, cmd, @@ -2612,85 +1368,87 @@ int drbd_send_ack_ex(struct drbd_conf *mdev, enum drbd_packets cmd, int drbd_send_drequest(struct drbd_conf *mdev, int cmd, sector_t sector, int size, u64 block_id) { - int ok; - struct p_block_req p; - - p.sector = cpu_to_be64(sector); - p.block_id = block_id; - p.blksize = cpu_to_be32(size); + struct drbd_socket *sock; + struct p_block_req *p; - ok = drbd_send_cmd(mdev, USE_DATA_SOCKET, cmd, - (struct p_header80 *)&p, sizeof(p)); - return ok; + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + p->sector = cpu_to_be64(sector); + p->block_id = block_id; + p->blksize = cpu_to_be32(size); + return drbd_send_command(mdev, sock, cmd, sizeof(*p), NULL, 0); } -int drbd_send_drequest_csum(struct drbd_conf *mdev, - sector_t sector, int size, - void *digest, int digest_size, - enum drbd_packets cmd) +int drbd_send_drequest_csum(struct drbd_conf *mdev, sector_t sector, int size, + void *digest, int digest_size, enum drbd_packet cmd) { - int ok; - struct p_block_req p; - - p.sector = cpu_to_be64(sector); - p.block_id = BE_DRBD_MAGIC + 0xbeef; - p.blksize = cpu_to_be32(size); - - p.head.magic = BE_DRBD_MAGIC; - p.head.command = cpu_to_be16(cmd); - p.head.length = cpu_to_be16(sizeof(p) - sizeof(struct p_header80) + digest_size); + struct drbd_socket *sock; + struct p_block_req *p; - mutex_lock(&mdev->data.mutex); + /* FIXME: Put the digest into the preallocated socket buffer. */ - ok = (sizeof(p) == drbd_send(mdev, mdev->data.socket, &p, sizeof(p), 0)); - ok = ok && (digest_size == drbd_send(mdev, mdev->data.socket, digest, digest_size, 0)); - - mutex_unlock(&mdev->data.mutex); - - return ok; + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + p->sector = cpu_to_be64(sector); + p->block_id = ID_SYNCER /* unused */; + p->blksize = cpu_to_be32(size); + return drbd_send_command(mdev, sock, cmd, sizeof(*p), + digest, digest_size); } int drbd_send_ov_request(struct drbd_conf *mdev, sector_t sector, int size) { - int ok; - struct p_block_req p; + struct drbd_socket *sock; + struct p_block_req *p; - p.sector = cpu_to_be64(sector); - p.block_id = BE_DRBD_MAGIC + 0xbabe; - p.blksize = cpu_to_be32(size); - - ok = drbd_send_cmd(mdev, USE_DATA_SOCKET, P_OV_REQUEST, - (struct p_header80 *)&p, sizeof(p)); - return ok; + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + p->sector = cpu_to_be64(sector); + p->block_id = ID_SYNCER /* unused */; + p->blksize = cpu_to_be32(size); + return drbd_send_command(mdev, sock, P_OV_REQUEST, sizeof(*p), NULL, 0); } /* called on sndtimeo * returns false if we should retry, * true if we think connection is dead */ -static int we_should_drop_the_connection(struct drbd_conf *mdev, struct socket *sock) +static int we_should_drop_the_connection(struct drbd_tconn *tconn, struct socket *sock) { int drop_it; /* long elapsed = (long)(jiffies - mdev->last_received); */ - drop_it = mdev->meta.socket == sock - || !mdev->asender.task - || get_t_state(&mdev->asender) != Running - || mdev->state.conn < C_CONNECTED; + drop_it = tconn->meta.socket == sock + || !tconn->asender.task + || get_t_state(&tconn->asender) != RUNNING + || tconn->cstate < C_WF_REPORT_PARAMS; if (drop_it) return true; - drop_it = !--mdev->ko_count; + drop_it = !--tconn->ko_count; if (!drop_it) { - dev_err(DEV, "[%s/%d] sock_sendmsg time expired, ko = %u\n", - current->comm, current->pid, mdev->ko_count); - request_ping(mdev); + conn_err(tconn, "[%s/%d] sock_sendmsg time expired, ko = %u\n", + current->comm, current->pid, tconn->ko_count); + request_ping(tconn); } return drop_it; /* && (mdev->state == R_PRIMARY) */; } +static void drbd_update_congested(struct drbd_tconn *tconn) +{ + struct sock *sk = tconn->data.socket->sk; + if (sk->sk_wmem_queued > sk->sk_sndbuf * 4 / 5) + set_bit(NET_CONGESTED, &tconn->flags); +} + /* The idea of sendpage seems to be to put some kind of reference * to the page into the skb, and to hand it over to the NIC. In * this process get_page() gets called. @@ -2713,21 +1471,28 @@ static int we_should_drop_the_connection(struct drbd_conf *mdev, struct socket * * with page_count == 0 or PageSlab. */ static int _drbd_no_send_page(struct drbd_conf *mdev, struct page *page, - int offset, size_t size, unsigned msg_flags) + int offset, size_t size, unsigned msg_flags) { - int sent = drbd_send(mdev, mdev->data.socket, kmap(page) + offset, size, msg_flags); + struct socket *socket; + void *addr; + int err; + + socket = mdev->tconn->data.socket; + addr = kmap(page) + offset; + err = drbd_send_all(mdev->tconn, socket, addr, size, msg_flags); kunmap(page); - if (sent == size) - mdev->send_cnt += size>>9; - return sent == size; + if (!err) + mdev->send_cnt += size >> 9; + return err; } static int _drbd_send_page(struct drbd_conf *mdev, struct page *page, int offset, size_t size, unsigned msg_flags) { + struct socket *socket = mdev->tconn->data.socket; mm_segment_t oldfs = get_fs(); - int sent, ok; int len = size; + int err = -EIO; /* e.g. XFS meta- & log-data is in slab pages, which have a * page_count of 0 and/or have PageSlab() set. @@ -2739,34 +1504,35 @@ static int _drbd_send_page(struct drbd_conf *mdev, struct page *page, return _drbd_no_send_page(mdev, page, offset, size, msg_flags); msg_flags |= MSG_NOSIGNAL; - drbd_update_congested(mdev); + drbd_update_congested(mdev->tconn); set_fs(KERNEL_DS); do { - sent = mdev->data.socket->ops->sendpage(mdev->data.socket, page, - offset, len, - msg_flags); - if (sent == -EAGAIN) { - if (we_should_drop_the_connection(mdev, - mdev->data.socket)) - break; - else - continue; - } + int sent; + + sent = socket->ops->sendpage(socket, page, offset, len, msg_flags); if (sent <= 0) { + if (sent == -EAGAIN) { + if (we_should_drop_the_connection(mdev->tconn, socket)) + break; + continue; + } dev_warn(DEV, "%s: size=%d len=%d sent=%d\n", __func__, (int)size, len, sent); + if (sent < 0) + err = sent; break; } len -= sent; offset += sent; } while (len > 0 /* THINK && mdev->cstate >= C_CONNECTED*/); set_fs(oldfs); - clear_bit(NET_CONGESTED, &mdev->flags); + clear_bit(NET_CONGESTED, &mdev->tconn->flags); - ok = (len == 0); - if (likely(ok)) - mdev->send_cnt += size>>9; - return ok; + if (len == 0) { + err = 0; + mdev->send_cnt += size >> 9; + } + return err; } static int _drbd_send_bio(struct drbd_conf *mdev, struct bio *bio) @@ -2775,12 +1541,15 @@ static int _drbd_send_bio(struct drbd_conf *mdev, struct bio *bio) int i; /* hint all but last page with MSG_MORE */ bio_for_each_segment(bvec, bio, i) { - if (!_drbd_no_send_page(mdev, bvec->bv_page, - bvec->bv_offset, bvec->bv_len, - i == bio->bi_vcnt -1 ? 0 : MSG_MORE)) - return 0; + int err; + + err = _drbd_no_send_page(mdev, bvec->bv_page, + bvec->bv_offset, bvec->bv_len, + i == bio->bi_vcnt - 1 ? 0 : MSG_MORE); + if (err) + return err; } - return 1; + return 0; } static int _drbd_send_zc_bio(struct drbd_conf *mdev, struct bio *bio) @@ -2789,32 +1558,40 @@ static int _drbd_send_zc_bio(struct drbd_conf *mdev, struct bio *bio) int i; /* hint all but last page with MSG_MORE */ bio_for_each_segment(bvec, bio, i) { - if (!_drbd_send_page(mdev, bvec->bv_page, - bvec->bv_offset, bvec->bv_len, - i == bio->bi_vcnt -1 ? 0 : MSG_MORE)) - return 0; + int err; + + err = _drbd_send_page(mdev, bvec->bv_page, + bvec->bv_offset, bvec->bv_len, + i == bio->bi_vcnt - 1 ? 0 : MSG_MORE); + if (err) + return err; } - return 1; + return 0; } -static int _drbd_send_zc_ee(struct drbd_conf *mdev, struct drbd_epoch_entry *e) +static int _drbd_send_zc_ee(struct drbd_conf *mdev, + struct drbd_peer_request *peer_req) { - struct page *page = e->pages; - unsigned len = e->size; + struct page *page = peer_req->pages; + unsigned len = peer_req->i.size; + int err; + /* hint all but last page with MSG_MORE */ page_chain_for_each(page) { unsigned l = min_t(unsigned, len, PAGE_SIZE); - if (!_drbd_send_page(mdev, page, 0, l, - page_chain_next(page) ? MSG_MORE : 0)) - return 0; + + err = _drbd_send_page(mdev, page, 0, l, + page_chain_next(page) ? MSG_MORE : 0); + if (err) + return err; len -= l; } - return 1; + return 0; } static u32 bio_flags_to_wire(struct drbd_conf *mdev, unsigned long bi_rw) { - if (mdev->agreed_pro_version >= 95) + if (mdev->tconn->agreed_pro_version >= 95) return (bi_rw & REQ_SYNC ? DP_RW_SYNC : 0) | (bi_rw & REQ_FUA ? DP_FUA : 0) | (bi_rw & REQ_FLUSH ? DP_FLUSH : 0) | @@ -2828,50 +1605,36 @@ static u32 bio_flags_to_wire(struct drbd_conf *mdev, unsigned long bi_rw) */ int drbd_send_dblock(struct drbd_conf *mdev, struct drbd_request *req) { - int ok = 1; - struct p_data p; + struct drbd_socket *sock; + struct p_data *p; unsigned int dp_flags = 0; - void *dgb; int dgs; + int err; - if (!drbd_get_data_sock(mdev)) - return 0; - - dgs = (mdev->agreed_pro_version >= 87 && mdev->integrity_w_tfm) ? - crypto_hash_digestsize(mdev->integrity_w_tfm) : 0; - - if (req->size <= DRBD_MAX_SIZE_H80_PACKET) { - p.head.h80.magic = BE_DRBD_MAGIC; - p.head.h80.command = cpu_to_be16(P_DATA); - p.head.h80.length = - cpu_to_be16(sizeof(p) - sizeof(union p_header) + dgs + req->size); - } else { - p.head.h95.magic = BE_DRBD_MAGIC_BIG; - p.head.h95.command = cpu_to_be16(P_DATA); - p.head.h95.length = - cpu_to_be32(sizeof(p) - sizeof(union p_header) + dgs + req->size); - } - - p.sector = cpu_to_be64(req->sector); - p.block_id = (unsigned long)req; - p.seq_num = cpu_to_be32(atomic_add_return(1, &mdev->packet_seq)); + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + dgs = mdev->tconn->integrity_tfm ? crypto_hash_digestsize(mdev->tconn->integrity_tfm) : 0; + if (!p) + return -EIO; + p->sector = cpu_to_be64(req->i.sector); + p->block_id = (unsigned long)req; + p->seq_num = cpu_to_be32(atomic_inc_return(&mdev->packet_seq)); dp_flags = bio_flags_to_wire(mdev, req->master_bio->bi_rw); - if (mdev->state.conn >= C_SYNC_SOURCE && mdev->state.conn <= C_PAUSED_SYNC_T) dp_flags |= DP_MAY_SET_IN_SYNC; - - p.dp_flags = cpu_to_be32(dp_flags); - set_bit(UNPLUG_REMOTE, &mdev->flags); - ok = (sizeof(p) == - drbd_send(mdev, mdev->data.socket, &p, sizeof(p), dgs ? MSG_MORE : 0)); - if (ok && dgs) { - dgb = mdev->int_dig_out; - drbd_csum_bio(mdev, mdev->integrity_w_tfm, req->master_bio, dgb); - ok = dgs == drbd_send(mdev, mdev->data.socket, dgb, dgs, 0); - } - if (ok) { + if (mdev->tconn->agreed_pro_version >= 100) { + if (req->rq_state & RQ_EXP_RECEIVE_ACK) + dp_flags |= DP_SEND_RECEIVE_ACK; + if (req->rq_state & RQ_EXP_WRITE_ACK) + dp_flags |= DP_SEND_WRITE_ACK; + } + p->dp_flags = cpu_to_be32(dp_flags); + if (dgs) + drbd_csum_bio(mdev, mdev->tconn->integrity_tfm, req->master_bio, p + 1); + err = __send_command(mdev->tconn, mdev->vnr, sock, P_DATA, sizeof(*p) + dgs, NULL, req->i.size); + if (!err) { /* For protocol A, we have to memcpy the payload into * socket buffers, as we may complete right away * as soon as we handed it over to tcp, at which point the data @@ -2883,92 +1646,76 @@ int drbd_send_dblock(struct drbd_conf *mdev, struct drbd_request *req) * out ok after sending on this side, but does not fit on the * receiving side, we sure have detected corruption elsewhere. */ - if (mdev->net_conf->wire_protocol == DRBD_PROT_A || dgs) - ok = _drbd_send_bio(mdev, req->master_bio); + if (!(req->rq_state & (RQ_EXP_RECEIVE_ACK | RQ_EXP_WRITE_ACK)) || dgs) + err = _drbd_send_bio(mdev, req->master_bio); else - ok = _drbd_send_zc_bio(mdev, req->master_bio); + err = _drbd_send_zc_bio(mdev, req->master_bio); /* double check digest, sometimes buffers have been modified in flight. */ if (dgs > 0 && dgs <= 64) { /* 64 byte, 512 bit, is the largest digest size * currently supported in kernel crypto. */ unsigned char digest[64]; - drbd_csum_bio(mdev, mdev->integrity_w_tfm, req->master_bio, digest); - if (memcmp(mdev->int_dig_out, digest, dgs)) { + drbd_csum_bio(mdev, mdev->tconn->integrity_tfm, req->master_bio, digest); + if (memcmp(p + 1, digest, dgs)) { dev_warn(DEV, "Digest mismatch, buffer modified by upper layers during write: %llus +%u\n", - (unsigned long long)req->sector, req->size); + (unsigned long long)req->i.sector, req->i.size); } } /* else if (dgs > 64) { ... Be noisy about digest too large ... } */ } + mutex_unlock(&sock->mutex); /* locked by drbd_prepare_command() */ - drbd_put_data_sock(mdev); - - return ok; + return err; } /* answer packet, used to send data back for read requests: * Peer -> (diskless) R_PRIMARY (P_DATA_REPLY) * C_SYNC_SOURCE -> C_SYNC_TARGET (P_RS_DATA_REPLY) */ -int drbd_send_block(struct drbd_conf *mdev, enum drbd_packets cmd, - struct drbd_epoch_entry *e) +int drbd_send_block(struct drbd_conf *mdev, enum drbd_packet cmd, + struct drbd_peer_request *peer_req) { - int ok; - struct p_data p; - void *dgb; + struct drbd_socket *sock; + struct p_data *p; + int err; int dgs; - dgs = (mdev->agreed_pro_version >= 87 && mdev->integrity_w_tfm) ? - crypto_hash_digestsize(mdev->integrity_w_tfm) : 0; - - if (e->size <= DRBD_MAX_SIZE_H80_PACKET) { - p.head.h80.magic = BE_DRBD_MAGIC; - p.head.h80.command = cpu_to_be16(cmd); - p.head.h80.length = - cpu_to_be16(sizeof(p) - sizeof(struct p_header80) + dgs + e->size); - } else { - p.head.h95.magic = BE_DRBD_MAGIC_BIG; - p.head.h95.command = cpu_to_be16(cmd); - p.head.h95.length = - cpu_to_be32(sizeof(p) - sizeof(struct p_header80) + dgs + e->size); - } - - p.sector = cpu_to_be64(e->sector); - p.block_id = e->block_id; - /* p.seq_num = 0; No sequence numbers here.. */ - - /* Only called by our kernel thread. - * This one may be interrupted by DRBD_SIG and/or DRBD_SIGKILL - * in response to admin command or module unload. - */ - if (!drbd_get_data_sock(mdev)) - return 0; + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); - ok = sizeof(p) == drbd_send(mdev, mdev->data.socket, &p, sizeof(p), dgs ? MSG_MORE : 0); - if (ok && dgs) { - dgb = mdev->int_dig_out; - drbd_csum_ee(mdev, mdev->integrity_w_tfm, e, dgb); - ok = dgs == drbd_send(mdev, mdev->data.socket, dgb, dgs, 0); - } - if (ok) - ok = _drbd_send_zc_ee(mdev, e); + dgs = mdev->tconn->integrity_tfm ? crypto_hash_digestsize(mdev->tconn->integrity_tfm) : 0; - drbd_put_data_sock(mdev); + if (!p) + return -EIO; + p->sector = cpu_to_be64(peer_req->i.sector); + p->block_id = peer_req->block_id; + p->seq_num = 0; /* unused */ + p->dp_flags = 0; + if (dgs) + drbd_csum_ee(mdev, mdev->tconn->integrity_tfm, peer_req, p + 1); + err = __send_command(mdev->tconn, mdev->vnr, sock, cmd, sizeof(*p) + dgs, NULL, peer_req->i.size); + if (!err) + err = _drbd_send_zc_ee(mdev, peer_req); + mutex_unlock(&sock->mutex); /* locked by drbd_prepare_command() */ - return ok; + return err; } -int drbd_send_oos(struct drbd_conf *mdev, struct drbd_request *req) +int drbd_send_out_of_sync(struct drbd_conf *mdev, struct drbd_request *req) { - struct p_block_desc p; - - p.sector = cpu_to_be64(req->sector); - p.blksize = cpu_to_be32(req->size); + struct drbd_socket *sock; + struct p_block_desc *p; - return drbd_send_cmd(mdev, USE_DATA_SOCKET, P_OUT_OF_SYNC, &p.head, sizeof(p)); + sock = &mdev->tconn->data; + p = drbd_prepare_command(mdev, sock); + if (!p) + return -EIO; + p->sector = cpu_to_be64(req->i.sector); + p->blksize = cpu_to_be32(req->i.size); + return drbd_send_command(mdev, sock, P_OUT_OF_SYNC, sizeof(*p), NULL, 0); } /* @@ -2987,7 +1734,7 @@ int drbd_send_oos(struct drbd_conf *mdev, struct drbd_request *req) /* * you must have down()ed the appropriate [m]sock_mutex elsewhere! */ -int drbd_send(struct drbd_conf *mdev, struct socket *sock, +int drbd_send(struct drbd_tconn *tconn, struct socket *sock, void *buf, size_t size, unsigned msg_flags) { struct kvec iov; @@ -2995,7 +1742,7 @@ int drbd_send(struct drbd_conf *mdev, struct socket *sock, int rv, sent = 0; if (!sock) - return -1000; + return -EBADR; /* THINK if (signal_pending) return ... ? */ @@ -3008,9 +1755,11 @@ int drbd_send(struct drbd_conf *mdev, struct socket *sock, msg.msg_controllen = 0; msg.msg_flags = msg_flags | MSG_NOSIGNAL; - if (sock == mdev->data.socket) { - mdev->ko_count = mdev->net_conf->ko_count; - drbd_update_congested(mdev); + if (sock == tconn->data.socket) { + rcu_read_lock(); + tconn->ko_count = rcu_dereference(tconn->net_conf)->ko_count; + rcu_read_unlock(); + drbd_update_congested(tconn); } do { /* STRANGE @@ -3024,12 +1773,11 @@ int drbd_send(struct drbd_conf *mdev, struct socket *sock, */ rv = kernel_sendmsg(sock, &msg, &iov, 1, size); if (rv == -EAGAIN) { - if (we_should_drop_the_connection(mdev, sock)) + if (we_should_drop_the_connection(tconn, sock)) break; else continue; } - D_ASSERT(rv != 0); if (rv == -EINTR) { flush_signals(current); rv = 0; @@ -3041,22 +1789,40 @@ int drbd_send(struct drbd_conf *mdev, struct socket *sock, iov.iov_len -= rv; } while (sent < size); - if (sock == mdev->data.socket) - clear_bit(NET_CONGESTED, &mdev->flags); + if (sock == tconn->data.socket) + clear_bit(NET_CONGESTED, &tconn->flags); if (rv <= 0) { if (rv != -EAGAIN) { - dev_err(DEV, "%s_sendmsg returned %d\n", - sock == mdev->meta.socket ? "msock" : "sock", - rv); - drbd_force_state(mdev, NS(conn, C_BROKEN_PIPE)); + conn_err(tconn, "%s_sendmsg returned %d\n", + sock == tconn->meta.socket ? "msock" : "sock", + rv); + conn_request_state(tconn, NS(conn, C_BROKEN_PIPE), CS_HARD); } else - drbd_force_state(mdev, NS(conn, C_TIMEOUT)); + conn_request_state(tconn, NS(conn, C_TIMEOUT), CS_HARD); } return sent; } +/** + * drbd_send_all - Send an entire buffer + * + * Returns 0 upon success and a negative error value otherwise. + */ +int drbd_send_all(struct drbd_tconn *tconn, struct socket *sock, void *buffer, + size_t size, unsigned msg_flags) +{ + int err; + + err = drbd_send(tconn, sock, buffer, size, msg_flags); + if (err < 0) + return err; + if (err != size) + return -EIO; + return 0; +} + static int drbd_open(struct block_device *bdev, fmode_t mode) { struct drbd_conf *mdev = bdev->bd_disk->private_data; @@ -3064,7 +1830,7 @@ static int drbd_open(struct block_device *bdev, fmode_t mode) int rv = 0; mutex_lock(&drbd_main_mutex); - spin_lock_irqsave(&mdev->req_lock, flags); + spin_lock_irqsave(&mdev->tconn->req_lock, flags); /* to have a stable mdev->state.role * and no race with updating open_cnt */ @@ -3077,7 +1843,7 @@ static int drbd_open(struct block_device *bdev, fmode_t mode) if (!rv) mdev->open_cnt++; - spin_unlock_irqrestore(&mdev->req_lock, flags); + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); mutex_unlock(&drbd_main_mutex); return rv; @@ -3094,35 +1860,14 @@ static int drbd_release(struct gendisk *gd, fmode_t mode) static void drbd_set_defaults(struct drbd_conf *mdev) { - /* This way we get a compile error when sync_conf grows, - and we forgot to initialize it here */ - mdev->sync_conf = (struct syncer_conf) { - /* .rate = */ DRBD_RATE_DEF, - /* .after = */ DRBD_AFTER_DEF, - /* .al_extents = */ DRBD_AL_EXTENTS_DEF, - /* .verify_alg = */ {}, 0, - /* .cpu_mask = */ {}, 0, - /* .csums_alg = */ {}, 0, - /* .use_rle = */ 0, - /* .on_no_data = */ DRBD_ON_NO_DATA_DEF, - /* .c_plan_ahead = */ DRBD_C_PLAN_AHEAD_DEF, - /* .c_delay_target = */ DRBD_C_DELAY_TARGET_DEF, - /* .c_fill_target = */ DRBD_C_FILL_TARGET_DEF, - /* .c_max_rate = */ DRBD_C_MAX_RATE_DEF, - /* .c_min_rate = */ DRBD_C_MIN_RATE_DEF - }; - - /* Have to use that way, because the layout differs between - big endian and little endian */ - mdev->state = (union drbd_state) { + /* Beware! The actual layout differs + * between big endian and little endian */ + mdev->state = (union drbd_dev_state) { { .role = R_SECONDARY, .peer = R_UNKNOWN, .conn = C_STANDALONE, .disk = D_DISKLESS, .pdsk = D_UNKNOWN, - .susp = 0, - .susp_nod = 0, - .susp_fen = 0 } }; } @@ -3138,28 +1883,17 @@ void drbd_init_set_defaults(struct drbd_conf *mdev) atomic_set(&mdev->rs_pending_cnt, 0); atomic_set(&mdev->unacked_cnt, 0); atomic_set(&mdev->local_cnt, 0); - atomic_set(&mdev->net_cnt, 0); - atomic_set(&mdev->packet_seq, 0); - atomic_set(&mdev->pp_in_use, 0); atomic_set(&mdev->pp_in_use_by_net, 0); atomic_set(&mdev->rs_sect_in, 0); atomic_set(&mdev->rs_sect_ev, 0); atomic_set(&mdev->ap_in_flight, 0); atomic_set(&mdev->md_io_in_use, 0); - mutex_init(&mdev->data.mutex); - mutex_init(&mdev->meta.mutex); - sema_init(&mdev->data.work.s, 0); - sema_init(&mdev->meta.work.s, 0); - mutex_init(&mdev->state_mutex); - - spin_lock_init(&mdev->data.work.q_lock); - spin_lock_init(&mdev->meta.work.q_lock); + mutex_init(&mdev->own_state_mutex); + mdev->state_mutex = &mdev->own_state_mutex; spin_lock_init(&mdev->al_lock); - spin_lock_init(&mdev->req_lock); spin_lock_init(&mdev->peer_seq_lock); - spin_lock_init(&mdev->epoch_lock); INIT_LIST_HEAD(&mdev->active_ee); INIT_LIST_HEAD(&mdev->sync_ee); @@ -3167,8 +1901,6 @@ void drbd_init_set_defaults(struct drbd_conf *mdev) INIT_LIST_HEAD(&mdev->read_ee); INIT_LIST_HEAD(&mdev->net_ee); INIT_LIST_HEAD(&mdev->resync_reads); - INIT_LIST_HEAD(&mdev->data.work.q); - INIT_LIST_HEAD(&mdev->meta.work.q); INIT_LIST_HEAD(&mdev->resync_work.list); INIT_LIST_HEAD(&mdev->unplug_work.list); INIT_LIST_HEAD(&mdev->go_diskless.list); @@ -3182,6 +1914,14 @@ void drbd_init_set_defaults(struct drbd_conf *mdev) mdev->md_sync_work.cb = w_md_sync; mdev->bm_io_work.w.cb = w_bitmap_io; mdev->start_resync_work.cb = w_start_resync; + + mdev->resync_work.mdev = mdev; + mdev->unplug_work.mdev = mdev; + mdev->go_diskless.mdev = mdev; + mdev->md_sync_work.mdev = mdev; + mdev->bm_io_work.w.mdev = mdev; + mdev->start_resync_work.mdev = mdev; + init_timer(&mdev->resync_timer); init_timer(&mdev->md_sync_timer); init_timer(&mdev->start_resync_timer); @@ -3197,17 +1937,10 @@ void drbd_init_set_defaults(struct drbd_conf *mdev) init_waitqueue_head(&mdev->misc_wait); init_waitqueue_head(&mdev->state_wait); - init_waitqueue_head(&mdev->net_cnt_wait); init_waitqueue_head(&mdev->ee_wait); init_waitqueue_head(&mdev->al_wait); init_waitqueue_head(&mdev->seq_wait); - drbd_thread_init(mdev, &mdev->receiver, drbdd_init); - drbd_thread_init(mdev, &mdev->worker, drbd_worker); - drbd_thread_init(mdev, &mdev->asender, drbd_asender); - - mdev->agreed_pro_version = PRO_VERSION_MAX; - mdev->write_ordering = WO_bdev_flush; mdev->resync_wenr = LC_FREE; mdev->peer_max_bio_size = DRBD_MAX_BIO_SIZE_SAFE; mdev->local_max_bio_size = DRBD_MAX_BIO_SIZE_SAFE; @@ -3216,13 +1949,10 @@ void drbd_init_set_defaults(struct drbd_conf *mdev) void drbd_mdev_cleanup(struct drbd_conf *mdev) { int i; - if (mdev->receiver.t_state != None) + if (mdev->tconn->receiver.t_state != NONE) dev_err(DEV, "ASSERT FAILED: receiver t_state == %d expected 0.\n", - mdev->receiver.t_state); + mdev->tconn->receiver.t_state); - /* no need to lock it, I'm the only thread alive */ - if (atomic_read(&mdev->current_epoch->epoch_size) != 0) - dev_err(DEV, "epoch_size:%d\n", atomic_read(&mdev->current_epoch->epoch_size)); mdev->al_writ_cnt = mdev->bm_writ_cnt = mdev->read_cnt = @@ -3239,7 +1969,7 @@ void drbd_mdev_cleanup(struct drbd_conf *mdev) mdev->rs_mark_left[i] = 0; mdev->rs_mark_time[i] = 0; } - D_ASSERT(mdev->net_conf == NULL); + D_ASSERT(mdev->tconn->net_conf == NULL); drbd_set_my_capacity(mdev, 0); if (mdev->bitmap) { @@ -3248,21 +1978,18 @@ void drbd_mdev_cleanup(struct drbd_conf *mdev) drbd_bm_cleanup(mdev); } - drbd_free_resources(mdev); + drbd_free_bc(mdev->ldev); + mdev->ldev = NULL; + clear_bit(AL_SUSPENDED, &mdev->flags); - /* - * currently we drbd_init_ee only on module load, so - * we may do drbd_release_ee only on module unload! - */ D_ASSERT(list_empty(&mdev->active_ee)); D_ASSERT(list_empty(&mdev->sync_ee)); D_ASSERT(list_empty(&mdev->done_ee)); D_ASSERT(list_empty(&mdev->read_ee)); D_ASSERT(list_empty(&mdev->net_ee)); D_ASSERT(list_empty(&mdev->resync_reads)); - D_ASSERT(list_empty(&mdev->data.work.q)); - D_ASSERT(list_empty(&mdev->meta.work.q)); + D_ASSERT(list_empty(&mdev->tconn->sender_work.q)); D_ASSERT(list_empty(&mdev->resync_work.list)); D_ASSERT(list_empty(&mdev->unplug_work.list)); D_ASSERT(list_empty(&mdev->go_diskless.list)); @@ -3336,7 +2063,7 @@ static int drbd_create_mempools(void) goto Enomem; drbd_ee_cache = kmem_cache_create( - "drbd_ee", sizeof(struct drbd_epoch_entry), 0, 0, NULL); + "drbd_ee", sizeof(struct drbd_peer_request), 0, 0, NULL); if (drbd_ee_cache == NULL) goto Enomem; @@ -3351,11 +2078,9 @@ static int drbd_create_mempools(void) goto Enomem; /* mempools */ -#ifdef COMPAT_HAVE_BIOSET_CREATE drbd_md_io_bio_set = bioset_create(DRBD_MIN_POOL_PAGES, 0); if (drbd_md_io_bio_set == NULL) goto Enomem; -#endif drbd_md_io_page_pool = mempool_create_page_pool(DRBD_MIN_POOL_PAGES, 0); if (drbd_md_io_page_pool == NULL) @@ -3404,73 +2129,53 @@ static struct notifier_block drbd_notifier = { .notifier_call = drbd_notify_sys, }; -static void drbd_release_ee_lists(struct drbd_conf *mdev) +static void drbd_release_all_peer_reqs(struct drbd_conf *mdev) { int rr; - rr = drbd_release_ee(mdev, &mdev->active_ee); + rr = drbd_free_peer_reqs(mdev, &mdev->active_ee); if (rr) dev_err(DEV, "%d EEs in active list found!\n", rr); - rr = drbd_release_ee(mdev, &mdev->sync_ee); + rr = drbd_free_peer_reqs(mdev, &mdev->sync_ee); if (rr) dev_err(DEV, "%d EEs in sync list found!\n", rr); - rr = drbd_release_ee(mdev, &mdev->read_ee); + rr = drbd_free_peer_reqs(mdev, &mdev->read_ee); if (rr) dev_err(DEV, "%d EEs in read list found!\n", rr); - rr = drbd_release_ee(mdev, &mdev->done_ee); + rr = drbd_free_peer_reqs(mdev, &mdev->done_ee); if (rr) dev_err(DEV, "%d EEs in done list found!\n", rr); - rr = drbd_release_ee(mdev, &mdev->net_ee); + rr = drbd_free_peer_reqs(mdev, &mdev->net_ee); if (rr) dev_err(DEV, "%d EEs in net list found!\n", rr); } -/* caution. no locking. - * currently only used from module cleanup code. */ -static void drbd_delete_device(unsigned int minor) +/* caution. no locking. */ +void drbd_minor_destroy(struct kref *kref) { - struct drbd_conf *mdev = minor_to_mdev(minor); - - if (!mdev) - return; + struct drbd_conf *mdev = container_of(kref, struct drbd_conf, kref); + struct drbd_tconn *tconn = mdev->tconn; del_timer_sync(&mdev->request_timer); /* paranoia asserts */ - if (mdev->open_cnt != 0) - dev_err(DEV, "open_cnt = %d in %s:%u", mdev->open_cnt, - __FILE__ , __LINE__); - - ERR_IF (!list_empty(&mdev->data.work.q)) { - struct list_head *lp; - list_for_each(lp, &mdev->data.work.q) { - dev_err(DEV, "lp = %p\n", lp); - } - }; + D_ASSERT(mdev->open_cnt == 0); /* end paranoia asserts */ - del_gendisk(mdev->vdisk); - /* cleanup stuff that may have been allocated during * device (re-)configuration or state changes */ if (mdev->this_bdev) bdput(mdev->this_bdev); - drbd_free_resources(mdev); + drbd_free_bc(mdev->ldev); + mdev->ldev = NULL; - drbd_release_ee_lists(mdev); - - /* should be freed on disconnect? */ - kfree(mdev->ee_hash); - /* - mdev->ee_hash_s = 0; - mdev->ee_hash = NULL; - */ + drbd_release_all_peer_reqs(mdev); lc_destroy(mdev->act_log); lc_destroy(mdev->resync); @@ -3478,19 +2183,101 @@ static void drbd_delete_device(unsigned int minor) kfree(mdev->p_uuid); /* mdev->p_uuid = NULL; */ - kfree(mdev->int_dig_out); - kfree(mdev->int_dig_in); - kfree(mdev->int_dig_vv); + if (mdev->bitmap) /* should no longer be there. */ + drbd_bm_cleanup(mdev); + __free_page(mdev->md_io_page); + put_disk(mdev->vdisk); + blk_cleanup_queue(mdev->rq_queue); + kfree(mdev->rs_plan_s); + kfree(mdev); - /* cleanup the rest that has been - * allocated from drbd_new_device - * and actually free the mdev itself */ - drbd_free_mdev(mdev); + kref_put(&tconn->kref, &conn_destroy); } +/* One global retry thread, if we need to push back some bio and have it + * reinserted through our make request function. + */ +static struct retry_worker { + struct workqueue_struct *wq; + struct work_struct worker; + + spinlock_t lock; + struct list_head writes; +} retry; + +static void do_retry(struct work_struct *ws) +{ + struct retry_worker *retry = container_of(ws, struct retry_worker, worker); + LIST_HEAD(writes); + struct drbd_request *req, *tmp; + + spin_lock_irq(&retry->lock); + list_splice_init(&retry->writes, &writes); + spin_unlock_irq(&retry->lock); + + list_for_each_entry_safe(req, tmp, &writes, tl_requests) { + struct drbd_conf *mdev = req->w.mdev; + struct bio *bio = req->master_bio; + unsigned long start_time = req->start_time; + bool expected; + + expected = + expect(atomic_read(&req->completion_ref) == 0) && + expect(req->rq_state & RQ_POSTPONED) && + expect((req->rq_state & RQ_LOCAL_PENDING) == 0 || + (req->rq_state & RQ_LOCAL_ABORTED) != 0); + + if (!expected) + dev_err(DEV, "req=%p completion_ref=%d rq_state=%x\n", + req, atomic_read(&req->completion_ref), + req->rq_state); + + /* We still need to put one kref associated with the + * "completion_ref" going zero in the code path that queued it + * here. The request object may still be referenced by a + * frozen local req->private_bio, in case we force-detached. + */ + kref_put(&req->kref, drbd_req_destroy); + + /* A single suspended or otherwise blocking device may stall + * all others as well. Fortunately, this code path is to + * recover from a situation that "should not happen": + * concurrent writes in multi-primary setup. + * In a "normal" lifecycle, this workqueue is supposed to be + * destroyed without ever doing anything. + * If it turns out to be an issue anyways, we can do per + * resource (replication group) or per device (minor) retry + * workqueues instead. + */ + + /* We are not just doing generic_make_request(), + * as we want to keep the start_time information. */ + inc_ap_bio(mdev); + __drbd_make_request(mdev, bio, start_time); + } +} + +void drbd_restart_request(struct drbd_request *req) +{ + unsigned long flags; + spin_lock_irqsave(&retry.lock, flags); + list_move_tail(&req->tl_requests, &retry.writes); + spin_unlock_irqrestore(&retry.lock, flags); + + /* Drop the extra reference that would otherwise + * have been dropped by complete_master_bio. + * do_retry() needs to grab a new one. */ + dec_ap_bio(req->w.mdev); + + queue_work(retry.wq, &retry.worker); +} + + static void drbd_cleanup(void) { unsigned int i; + struct drbd_conf *mdev; + struct drbd_tconn *tconn, *tmp; unregister_reboot_notifier(&drbd_notifier); @@ -3505,19 +2292,31 @@ static void drbd_cleanup(void) if (drbd_proc) remove_proc_entry("drbd", NULL); - drbd_nl_cleanup(); + if (retry.wq) + destroy_workqueue(retry.wq); + + drbd_genl_unregister(); - if (minor_table) { - i = minor_count; - while (i--) - drbd_delete_device(i); - drbd_destroy_mempools(); + idr_for_each_entry(&minors, mdev, i) { + idr_remove(&minors, mdev_to_minor(mdev)); + idr_remove(&mdev->tconn->volumes, mdev->vnr); + del_gendisk(mdev->vdisk); + /* synchronize_rcu(); No other threads running at this point */ + kref_put(&mdev->kref, &drbd_minor_destroy); } - kfree(minor_table); + /* not _rcu since, no other updater anymore. Genl already unregistered */ + list_for_each_entry_safe(tconn, tmp, &drbd_tconns, all_tconn) { + list_del(&tconn->all_tconn); /* not _rcu no proc, not other threads */ + /* synchronize_rcu(); */ + kref_put(&tconn->kref, &conn_destroy); + } + drbd_destroy_mempools(); unregister_blkdev(DRBD_MAJOR, "drbd"); + idr_destroy(&minors); + printk(KERN_INFO "drbd: module cleanup done.\n"); } @@ -3542,7 +2341,7 @@ static int drbd_congested(void *congested_data, int bdi_bits) goto out; } - if (test_bit(CALLBACK_PENDING, &mdev->flags)) { + if (test_bit(CALLBACK_PENDING, &mdev->tconn->flags)) { r |= (1 << BDI_async_congested); /* Without good local data, we would need to read from remote, * and that would need the worker thread as well, which is @@ -3566,7 +2365,7 @@ static int drbd_congested(void *congested_data, int bdi_bits) reason = 'b'; } - if (bdi_bits & (1 << BDI_async_congested) && test_bit(NET_CONGESTED, &mdev->flags)) { + if (bdi_bits & (1 << BDI_async_congested) && test_bit(NET_CONGESTED, &mdev->tconn->flags)) { r |= (1 << BDI_async_congested); reason = reason == 'b' ? 'a' : 'n'; } @@ -3576,20 +2375,243 @@ out: return r; } -struct drbd_conf *drbd_new_device(unsigned int minor) +static void drbd_init_workqueue(struct drbd_work_queue* wq) +{ + spin_lock_init(&wq->q_lock); + INIT_LIST_HEAD(&wq->q); + init_waitqueue_head(&wq->q_wait); +} + +struct drbd_tconn *conn_get_by_name(const char *name) +{ + struct drbd_tconn *tconn; + + if (!name || !name[0]) + return NULL; + + rcu_read_lock(); + list_for_each_entry_rcu(tconn, &drbd_tconns, all_tconn) { + if (!strcmp(tconn->name, name)) { + kref_get(&tconn->kref); + goto found; + } + } + tconn = NULL; +found: + rcu_read_unlock(); + return tconn; +} + +struct drbd_tconn *conn_get_by_addrs(void *my_addr, int my_addr_len, + void *peer_addr, int peer_addr_len) +{ + struct drbd_tconn *tconn; + + rcu_read_lock(); + list_for_each_entry_rcu(tconn, &drbd_tconns, all_tconn) { + if (tconn->my_addr_len == my_addr_len && + tconn->peer_addr_len == peer_addr_len && + !memcmp(&tconn->my_addr, my_addr, my_addr_len) && + !memcmp(&tconn->peer_addr, peer_addr, peer_addr_len)) { + kref_get(&tconn->kref); + goto found; + } + } + tconn = NULL; +found: + rcu_read_unlock(); + return tconn; +} + +static int drbd_alloc_socket(struct drbd_socket *socket) +{ + socket->rbuf = (void *) __get_free_page(GFP_KERNEL); + if (!socket->rbuf) + return -ENOMEM; + socket->sbuf = (void *) __get_free_page(GFP_KERNEL); + if (!socket->sbuf) + return -ENOMEM; + return 0; +} + +static void drbd_free_socket(struct drbd_socket *socket) +{ + free_page((unsigned long) socket->sbuf); + free_page((unsigned long) socket->rbuf); +} + +void conn_free_crypto(struct drbd_tconn *tconn) +{ + drbd_free_sock(tconn); + + crypto_free_hash(tconn->csums_tfm); + crypto_free_hash(tconn->verify_tfm); + crypto_free_hash(tconn->cram_hmac_tfm); + crypto_free_hash(tconn->integrity_tfm); + crypto_free_hash(tconn->peer_integrity_tfm); + kfree(tconn->int_dig_in); + kfree(tconn->int_dig_vv); + + tconn->csums_tfm = NULL; + tconn->verify_tfm = NULL; + tconn->cram_hmac_tfm = NULL; + tconn->integrity_tfm = NULL; + tconn->peer_integrity_tfm = NULL; + tconn->int_dig_in = NULL; + tconn->int_dig_vv = NULL; +} + +int set_resource_options(struct drbd_tconn *tconn, struct res_opts *res_opts) +{ + cpumask_var_t new_cpu_mask; + int err; + + if (!zalloc_cpumask_var(&new_cpu_mask, GFP_KERNEL)) + return -ENOMEM; + /* + retcode = ERR_NOMEM; + drbd_msg_put_info("unable to allocate cpumask"); + */ + + /* silently ignore cpu mask on UP kernel */ + if (nr_cpu_ids > 1 && res_opts->cpu_mask[0] != 0) { + /* FIXME: Get rid of constant 32 here */ + err = bitmap_parse(res_opts->cpu_mask, 32, + cpumask_bits(new_cpu_mask), nr_cpu_ids); + if (err) { + conn_warn(tconn, "bitmap_parse() failed with %d\n", err); + /* retcode = ERR_CPU_MASK_PARSE; */ + goto fail; + } + } + tconn->res_opts = *res_opts; + if (!cpumask_equal(tconn->cpu_mask, new_cpu_mask)) { + cpumask_copy(tconn->cpu_mask, new_cpu_mask); + drbd_calc_cpu_mask(tconn); + tconn->receiver.reset_cpu_mask = 1; + tconn->asender.reset_cpu_mask = 1; + tconn->worker.reset_cpu_mask = 1; + } + err = 0; + +fail: + free_cpumask_var(new_cpu_mask); + return err; + +} + +/* caller must be under genl_lock() */ +struct drbd_tconn *conn_create(const char *name, struct res_opts *res_opts) +{ + struct drbd_tconn *tconn; + + tconn = kzalloc(sizeof(struct drbd_tconn), GFP_KERNEL); + if (!tconn) + return NULL; + + tconn->name = kstrdup(name, GFP_KERNEL); + if (!tconn->name) + goto fail; + + if (drbd_alloc_socket(&tconn->data)) + goto fail; + if (drbd_alloc_socket(&tconn->meta)) + goto fail; + + if (!zalloc_cpumask_var(&tconn->cpu_mask, GFP_KERNEL)) + goto fail; + + if (set_resource_options(tconn, res_opts)) + goto fail; + + tconn->current_epoch = kzalloc(sizeof(struct drbd_epoch), GFP_KERNEL); + if (!tconn->current_epoch) + goto fail; + + INIT_LIST_HEAD(&tconn->transfer_log); + + INIT_LIST_HEAD(&tconn->current_epoch->list); + tconn->epochs = 1; + spin_lock_init(&tconn->epoch_lock); + tconn->write_ordering = WO_bdev_flush; + + tconn->send.seen_any_write_yet = false; + tconn->send.current_epoch_nr = 0; + tconn->send.current_epoch_writes = 0; + + tconn->cstate = C_STANDALONE; + mutex_init(&tconn->cstate_mutex); + spin_lock_init(&tconn->req_lock); + mutex_init(&tconn->conf_update); + init_waitqueue_head(&tconn->ping_wait); + idr_init(&tconn->volumes); + + drbd_init_workqueue(&tconn->sender_work); + mutex_init(&tconn->data.mutex); + mutex_init(&tconn->meta.mutex); + + drbd_thread_init(tconn, &tconn->receiver, drbdd_init, "receiver"); + drbd_thread_init(tconn, &tconn->worker, drbd_worker, "worker"); + drbd_thread_init(tconn, &tconn->asender, drbd_asender, "asender"); + + kref_init(&tconn->kref); + list_add_tail_rcu(&tconn->all_tconn, &drbd_tconns); + + return tconn; + +fail: + kfree(tconn->current_epoch); + free_cpumask_var(tconn->cpu_mask); + drbd_free_socket(&tconn->meta); + drbd_free_socket(&tconn->data); + kfree(tconn->name); + kfree(tconn); + + return NULL; +} + +void conn_destroy(struct kref *kref) +{ + struct drbd_tconn *tconn = container_of(kref, struct drbd_tconn, kref); + + if (atomic_read(&tconn->current_epoch->epoch_size) != 0) + conn_err(tconn, "epoch_size:%d\n", atomic_read(&tconn->current_epoch->epoch_size)); + kfree(tconn->current_epoch); + + idr_destroy(&tconn->volumes); + + free_cpumask_var(tconn->cpu_mask); + drbd_free_socket(&tconn->meta); + drbd_free_socket(&tconn->data); + kfree(tconn->name); + kfree(tconn->int_dig_in); + kfree(tconn->int_dig_vv); + kfree(tconn); +} + +enum drbd_ret_code conn_new_minor(struct drbd_tconn *tconn, unsigned int minor, int vnr) { struct drbd_conf *mdev; struct gendisk *disk; struct request_queue *q; + int vnr_got = vnr; + int minor_got = minor; + enum drbd_ret_code err = ERR_NOMEM; + + mdev = minor_to_mdev(minor); + if (mdev) + return ERR_MINOR_EXISTS; /* GFP_KERNEL, we are outside of all write-out paths */ mdev = kzalloc(sizeof(struct drbd_conf), GFP_KERNEL); if (!mdev) - return NULL; - if (!zalloc_cpumask_var(&mdev->cpu_mask, GFP_KERNEL)) - goto out_no_cpumask; + return ERR_NOMEM; + + kref_get(&tconn->kref); + mdev->tconn = tconn; mdev->minor = minor; + mdev->vnr = vnr; drbd_init_set_defaults(mdev); @@ -3627,7 +2649,7 @@ struct drbd_conf *drbd_new_device(unsigned int minor) blk_queue_max_hw_sectors(q, DRBD_MAX_BIO_SIZE_SAFE >> 8); blk_queue_bounce_limit(q, BLK_BOUNCE_ANY); blk_queue_merge_bvec(q, drbd_merge_bvec); - q->queue_lock = &mdev->req_lock; + q->queue_lock = &mdev->tconn->req_lock; /* needed since we use */ mdev->md_io_page = alloc_page(GFP_KERNEL); if (!mdev->md_io_page) @@ -3635,30 +2657,44 @@ struct drbd_conf *drbd_new_device(unsigned int minor) if (drbd_bm_init(mdev)) goto out_no_bitmap; - /* no need to lock access, we are still initializing this minor device. */ - if (!tl_init(mdev)) - goto out_no_tl; - - mdev->app_reads_hash = kzalloc(APP_R_HSIZE*sizeof(void *), GFP_KERNEL); - if (!mdev->app_reads_hash) - goto out_no_app_reads; - - mdev->current_epoch = kzalloc(sizeof(struct drbd_epoch), GFP_KERNEL); - if (!mdev->current_epoch) - goto out_no_epoch; - - INIT_LIST_HEAD(&mdev->current_epoch->list); - mdev->epochs = 1; - - return mdev; - -/* out_whatever_else: - kfree(mdev->current_epoch); */ -out_no_epoch: - kfree(mdev->app_reads_hash); -out_no_app_reads: - tl_cleanup(mdev); -out_no_tl: + mdev->read_requests = RB_ROOT; + mdev->write_requests = RB_ROOT; + + if (!idr_pre_get(&minors, GFP_KERNEL)) + goto out_no_minor_idr; + if (idr_get_new_above(&minors, mdev, minor, &minor_got)) + goto out_no_minor_idr; + if (minor_got != minor) { + err = ERR_MINOR_EXISTS; + drbd_msg_put_info("requested minor exists already"); + goto out_idr_remove_minor; + } + + if (!idr_pre_get(&tconn->volumes, GFP_KERNEL)) + goto out_idr_remove_minor; + if (idr_get_new_above(&tconn->volumes, mdev, vnr, &vnr_got)) + goto out_idr_remove_minor; + if (vnr_got != vnr) { + err = ERR_INVALID_REQUEST; + drbd_msg_put_info("requested volume exists already"); + goto out_idr_remove_vol; + } + add_disk(disk); + kref_init(&mdev->kref); /* one ref for both idrs and the the add_disk */ + + /* inherit the connection state */ + mdev->state.conn = tconn->cstate; + if (mdev->state.conn == C_WF_REPORT_PARAMS) + drbd_connected(mdev); + + return NO_ERROR; + +out_idr_remove_vol: + idr_remove(&tconn->volumes, vnr_got); +out_idr_remove_minor: + idr_remove(&minors, minor_got); + synchronize_rcu(); +out_no_minor_idr: drbd_bm_cleanup(mdev); out_no_bitmap: __free_page(mdev->md_io_page); @@ -3667,55 +2703,25 @@ out_no_io_page: out_no_disk: blk_cleanup_queue(q); out_no_q: - free_cpumask_var(mdev->cpu_mask); -out_no_cpumask: - kfree(mdev); - return NULL; -} - -/* counterpart of drbd_new_device. - * last part of drbd_delete_device. */ -void drbd_free_mdev(struct drbd_conf *mdev) -{ - kfree(mdev->current_epoch); - kfree(mdev->app_reads_hash); - tl_cleanup(mdev); - if (mdev->bitmap) /* should no longer be there. */ - drbd_bm_cleanup(mdev); - __free_page(mdev->md_io_page); - put_disk(mdev->vdisk); - blk_cleanup_queue(mdev->rq_queue); - free_cpumask_var(mdev->cpu_mask); - drbd_free_tl_hash(mdev); kfree(mdev); + kref_put(&tconn->kref, &conn_destroy); + return err; } - int __init drbd_init(void) { int err; - if (sizeof(struct p_handshake) != 80) { - printk(KERN_ERR - "drbd: never change the size or layout " - "of the HandShake packet.\n"); - return -EINVAL; - } - if (minor_count < DRBD_MINOR_COUNT_MIN || minor_count > DRBD_MINOR_COUNT_MAX) { printk(KERN_ERR - "drbd: invalid minor_count (%d)\n", minor_count); + "drbd: invalid minor_count (%d)\n", minor_count); #ifdef MODULE return -EINVAL; #else - minor_count = 8; + minor_count = DRBD_MINOR_COUNT_DEF; #endif } - err = drbd_nl_init(); - if (err) - return err; - err = register_blkdev(DRBD_MAJOR, "drbd"); if (err) { printk(KERN_ERR @@ -3724,6 +2730,13 @@ int __init drbd_init(void) return err; } + err = drbd_genl_register(); + if (err) { + printk(KERN_ERR "drbd: unable to register generic netlink family\n"); + goto fail; + } + + register_reboot_notifier(&drbd_notifier); /* @@ -3734,22 +2747,29 @@ int __init drbd_init(void) init_waitqueue_head(&drbd_pp_wait); drbd_proc = NULL; /* play safe for drbd_cleanup */ - minor_table = kzalloc(sizeof(struct drbd_conf *)*minor_count, - GFP_KERNEL); - if (!minor_table) - goto Enomem; + idr_init(&minors); err = drbd_create_mempools(); if (err) - goto Enomem; + goto fail; drbd_proc = proc_create_data("drbd", S_IFREG | S_IRUGO , NULL, &drbd_proc_fops, NULL); if (!drbd_proc) { printk(KERN_ERR "drbd: unable to register proc file\n"); - goto Enomem; + goto fail; } rwlock_init(&global_state_lock); + INIT_LIST_HEAD(&drbd_tconns); + + retry.wq = create_singlethread_workqueue("drbd-reissue"); + if (!retry.wq) { + printk(KERN_ERR "drbd: unable to create retry workqueue\n"); + goto fail; + } + INIT_WORK(&retry.worker, do_retry); + spin_lock_init(&retry.lock); + INIT_LIST_HEAD(&retry.writes); printk(KERN_INFO "drbd: initialized. " "Version: " REL_VERSION " (api:%d/proto:%d-%d)\n", @@ -3757,11 +2777,10 @@ int __init drbd_init(void) printk(KERN_INFO "drbd: %s\n", drbd_buildtag()); printk(KERN_INFO "drbd: registered as block device major %d\n", DRBD_MAJOR); - printk(KERN_INFO "drbd: minor_table @ 0x%p\n", minor_table); return 0; /* Success! */ -Enomem: +fail: drbd_cleanup(); if (err == -ENOMEM) /* currently always the case */ @@ -3782,47 +2801,42 @@ void drbd_free_bc(struct drbd_backing_dev *ldev) kfree(ldev); } -void drbd_free_sock(struct drbd_conf *mdev) +void drbd_free_sock(struct drbd_tconn *tconn) { - if (mdev->data.socket) { - mutex_lock(&mdev->data.mutex); - kernel_sock_shutdown(mdev->data.socket, SHUT_RDWR); - sock_release(mdev->data.socket); - mdev->data.socket = NULL; - mutex_unlock(&mdev->data.mutex); + if (tconn->data.socket) { + mutex_lock(&tconn->data.mutex); + kernel_sock_shutdown(tconn->data.socket, SHUT_RDWR); + sock_release(tconn->data.socket); + tconn->data.socket = NULL; + mutex_unlock(&tconn->data.mutex); } - if (mdev->meta.socket) { - mutex_lock(&mdev->meta.mutex); - kernel_sock_shutdown(mdev->meta.socket, SHUT_RDWR); - sock_release(mdev->meta.socket); - mdev->meta.socket = NULL; - mutex_unlock(&mdev->meta.mutex); + if (tconn->meta.socket) { + mutex_lock(&tconn->meta.mutex); + kernel_sock_shutdown(tconn->meta.socket, SHUT_RDWR); + sock_release(tconn->meta.socket); + tconn->meta.socket = NULL; + mutex_unlock(&tconn->meta.mutex); } } +/* meta data management */ -void drbd_free_resources(struct drbd_conf *mdev) +void conn_md_sync(struct drbd_tconn *tconn) { - crypto_free_hash(mdev->csums_tfm); - mdev->csums_tfm = NULL; - crypto_free_hash(mdev->verify_tfm); - mdev->verify_tfm = NULL; - crypto_free_hash(mdev->cram_hmac_tfm); - mdev->cram_hmac_tfm = NULL; - crypto_free_hash(mdev->integrity_w_tfm); - mdev->integrity_w_tfm = NULL; - crypto_free_hash(mdev->integrity_r_tfm); - mdev->integrity_r_tfm = NULL; - - drbd_free_sock(mdev); + struct drbd_conf *mdev; + int vnr; - __no_warn(local, - drbd_free_bc(mdev->ldev); - mdev->ldev = NULL;); + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + kref_get(&mdev->kref); + rcu_read_unlock(); + drbd_md_sync(mdev); + kref_put(&mdev->kref, &drbd_minor_destroy); + rcu_read_lock(); + } + rcu_read_unlock(); } -/* meta data management */ - struct meta_data_on_disk { u64 la_size; /* last agreed size. */ u64 uuid[UI_SIZE]; /* UUIDs. */ @@ -3833,7 +2847,7 @@ struct meta_data_on_disk { u32 md_size_sect; u32 al_offset; /* offset to this block */ u32 al_nr_extents; /* important for restoring the AL */ - /* `-- act_log->nr_elements <-- sync_conf.al_extents */ + /* `-- act_log->nr_elements <-- ldev->dc.al_extents */ u32 bm_offset; /* offset to the bitmap, from here */ u32 bm_bytes_per_bit; /* BM_BLOCK_SIZE */ u32 la_peer_max_bio_size; /* last peer max_bio_size */ @@ -3871,7 +2885,7 @@ void drbd_md_sync(struct drbd_conf *mdev) for (i = UI_CURRENT; i < UI_SIZE; i++) buffer->uuid[i] = cpu_to_be64(mdev->ldev->md.uuid[i]); buffer->flags = cpu_to_be32(mdev->ldev->md.flags); - buffer->magic = cpu_to_be32(DRBD_MD_MAGIC); + buffer->magic = cpu_to_be32(DRBD_MD_MAGIC_84_UNCLEAN); buffer->md_size_sect = cpu_to_be32(mdev->ldev->md.md_size_sect); buffer->al_offset = cpu_to_be32(mdev->ldev->md.al_offset); @@ -3885,7 +2899,7 @@ void drbd_md_sync(struct drbd_conf *mdev) D_ASSERT(drbd_md_ss__(mdev, mdev->ldev) == mdev->ldev->md.md_offset); sector = mdev->ldev->md.md_offset; - if (!drbd_md_sync_page_io(mdev, mdev->ldev, sector, WRITE)) { + if (drbd_md_sync_page_io(mdev, mdev->ldev, sector, WRITE)) { /* this was a try anyways ... */ dev_err(DEV, "meta data update failed!\n"); drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR); @@ -3906,11 +2920,12 @@ out: * @bdev: Device from which the meta data should be read in. * * Return 0 (NO_ERROR) on success, and an enum drbd_ret_code in case - * something goes wrong. Currently only: ERR_IO_MD_DISK, ERR_MD_INVALID. + * something goes wrong. */ int drbd_md_read(struct drbd_conf *mdev, struct drbd_backing_dev *bdev) { struct meta_data_on_disk *buffer; + u32 magic, flags; int i, rv = NO_ERROR; if (!get_ldev_if_state(mdev, D_ATTACHING)) @@ -3920,7 +2935,7 @@ int drbd_md_read(struct drbd_conf *mdev, struct drbd_backing_dev *bdev) if (!buffer) goto out; - if (!drbd_md_sync_page_io(mdev, bdev, bdev->md.md_offset, READ)) { + if (drbd_md_sync_page_io(mdev, bdev, bdev->md.md_offset, READ)) { /* NOTE: can't do normal error processing here as this is called BEFORE disk is attached */ dev_err(DEV, "Error while reading metadata.\n"); @@ -3928,8 +2943,20 @@ int drbd_md_read(struct drbd_conf *mdev, struct drbd_backing_dev *bdev) goto err; } - if (be32_to_cpu(buffer->magic) != DRBD_MD_MAGIC) { - dev_err(DEV, "Error while reading metadata, magic not found.\n"); + magic = be32_to_cpu(buffer->magic); + flags = be32_to_cpu(buffer->flags); + if (magic == DRBD_MD_MAGIC_84_UNCLEAN || + (magic == DRBD_MD_MAGIC_08 && !(flags & MDF_AL_CLEAN))) { + /* btw: that's Activity Log clean, not "all" clean. */ + dev_err(DEV, "Found unclean meta data. Did you \"drbdadm apply-al\"?\n"); + rv = ERR_MD_UNCLEAN; + goto err; + } + if (magic != DRBD_MD_MAGIC_08) { + if (magic == DRBD_MD_MAGIC_07) + dev_err(DEV, "Found old (0.7) meta data magic. Did you \"drbdadm create-md\"?\n"); + else + dev_err(DEV, "Meta data magic not found. Did you \"drbdadm create-md\"?\n"); rv = ERR_MD_INVALID; goto err; } @@ -3963,20 +2990,16 @@ int drbd_md_read(struct drbd_conf *mdev, struct drbd_backing_dev *bdev) for (i = UI_CURRENT; i < UI_SIZE; i++) bdev->md.uuid[i] = be64_to_cpu(buffer->uuid[i]); bdev->md.flags = be32_to_cpu(buffer->flags); - mdev->sync_conf.al_extents = be32_to_cpu(buffer->al_nr_extents); bdev->md.device_uuid = be64_to_cpu(buffer->device_uuid); - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); if (mdev->state.conn < C_CONNECTED) { unsigned int peer; peer = be32_to_cpu(buffer->la_peer_max_bio_size); peer = max(peer, DRBD_MAX_BIO_SIZE_SAFE); mdev->peer_max_bio_size = peer; } - spin_unlock_irq(&mdev->req_lock); - - if (mdev->sync_conf.al_extents < 7) - mdev->sync_conf.al_extents = 127; + spin_unlock_irq(&mdev->tconn->req_lock); err: drbd_md_put_buffer(mdev); @@ -4011,7 +3034,7 @@ void drbd_md_mark_dirty(struct drbd_conf *mdev) } #endif -static void drbd_uuid_move_history(struct drbd_conf *mdev) __must_hold(local) +void drbd_uuid_move_history(struct drbd_conf *mdev) __must_hold(local) { int i; @@ -4019,7 +3042,7 @@ static void drbd_uuid_move_history(struct drbd_conf *mdev) __must_hold(local) mdev->ldev->md.uuid[i+1] = mdev->ldev->md.uuid[i]; } -void _drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local) +void __drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local) { if (idx == UI_CURRENT) { if (mdev->state.role == R_PRIMARY) @@ -4034,14 +3057,24 @@ void _drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local) drbd_md_mark_dirty(mdev); } +void _drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local) +{ + unsigned long flags; + spin_lock_irqsave(&mdev->ldev->md.uuid_lock, flags); + __drbd_uuid_set(mdev, idx, val); + spin_unlock_irqrestore(&mdev->ldev->md.uuid_lock, flags); +} void drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local) { + unsigned long flags; + spin_lock_irqsave(&mdev->ldev->md.uuid_lock, flags); if (mdev->ldev->md.uuid[idx]) { drbd_uuid_move_history(mdev); mdev->ldev->md.uuid[UI_HISTORY_START] = mdev->ldev->md.uuid[idx]; } - _drbd_uuid_set(mdev, idx, val); + __drbd_uuid_set(mdev, idx, val); + spin_unlock_irqrestore(&mdev->ldev->md.uuid_lock, flags); } /** @@ -4054,15 +3087,20 @@ void drbd_uuid_set(struct drbd_conf *mdev, int idx, u64 val) __must_hold(local) void drbd_uuid_new_current(struct drbd_conf *mdev) __must_hold(local) { u64 val; - unsigned long long bm_uuid = mdev->ldev->md.uuid[UI_BITMAP]; + unsigned long long bm_uuid; + + get_random_bytes(&val, sizeof(u64)); + + spin_lock_irq(&mdev->ldev->md.uuid_lock); + bm_uuid = mdev->ldev->md.uuid[UI_BITMAP]; if (bm_uuid) dev_warn(DEV, "bm UUID was already set: %llX\n", bm_uuid); mdev->ldev->md.uuid[UI_BITMAP] = mdev->ldev->md.uuid[UI_CURRENT]; + __drbd_uuid_set(mdev, UI_CURRENT, val); + spin_unlock_irq(&mdev->ldev->md.uuid_lock); - get_random_bytes(&val, sizeof(u64)); - _drbd_uuid_set(mdev, UI_CURRENT, val); drbd_print_uuids(mdev, "new current UUID"); /* get it to stable storage _now_ */ drbd_md_sync(mdev); @@ -4070,9 +3108,11 @@ void drbd_uuid_new_current(struct drbd_conf *mdev) __must_hold(local) void drbd_uuid_set_bm(struct drbd_conf *mdev, u64 val) __must_hold(local) { + unsigned long flags; if (mdev->ldev->md.uuid[UI_BITMAP] == 0 && val == 0) return; + spin_lock_irqsave(&mdev->ldev->md.uuid_lock, flags); if (val == 0) { drbd_uuid_move_history(mdev); mdev->ldev->md.uuid[UI_HISTORY_START] = mdev->ldev->md.uuid[UI_BITMAP]; @@ -4084,6 +3124,8 @@ void drbd_uuid_set_bm(struct drbd_conf *mdev, u64 val) __must_hold(local) mdev->ldev->md.uuid[UI_BITMAP] = val & ~((u64)1); } + spin_unlock_irqrestore(&mdev->ldev->md.uuid_lock, flags); + drbd_md_mark_dirty(mdev); } @@ -4135,9 +3177,10 @@ int drbd_bmio_clear_n_write(struct drbd_conf *mdev) return rv; } -static int w_bitmap_io(struct drbd_conf *mdev, struct drbd_work *w, int unused) +static int w_bitmap_io(struct drbd_work *w, int unused) { struct bm_io_work *work = container_of(w, struct bm_io_work, w); + struct drbd_conf *mdev = w->mdev; int rv = -EIO; D_ASSERT(atomic_read(&mdev->ap_bio_cnt) == 0); @@ -4149,8 +3192,7 @@ static int w_bitmap_io(struct drbd_conf *mdev, struct drbd_work *w, int unused) put_ldev(mdev); } - clear_bit(BITMAP_IO, &mdev->flags); - smp_mb__after_clear_bit(); + clear_bit_unlock(BITMAP_IO, &mdev->flags); wake_up(&mdev->misc_wait); if (work->done) @@ -4160,7 +3202,7 @@ static int w_bitmap_io(struct drbd_conf *mdev, struct drbd_work *w, int unused) work->why = NULL; work->flags = 0; - return 1; + return 0; } void drbd_ldev_destroy(struct drbd_conf *mdev) @@ -4173,29 +3215,51 @@ void drbd_ldev_destroy(struct drbd_conf *mdev) drbd_free_bc(mdev->ldev); mdev->ldev = NULL;); - if (mdev->md_io_tmpp) { - __free_page(mdev->md_io_tmpp); - mdev->md_io_tmpp = NULL; - } clear_bit(GO_DISKLESS, &mdev->flags); } -static int w_go_diskless(struct drbd_conf *mdev, struct drbd_work *w, int unused) +static int w_go_diskless(struct drbd_work *w, int unused) { + struct drbd_conf *mdev = w->mdev; + D_ASSERT(mdev->state.disk == D_FAILED); /* we cannot assert local_cnt == 0 here, as get_ldev_if_state will * inc/dec it frequently. Once we are D_DISKLESS, no one will touch * the protected members anymore, though, so once put_ldev reaches zero * again, it will be safe to free them. */ + + /* Try to write changed bitmap pages, read errors may have just + * set some bits outside the area covered by the activity log. + * + * If we have an IO error during the bitmap writeout, + * we will want a full sync next time, just in case. + * (Do we want a specific meta data flag for this?) + * + * If that does not make it to stable storage either, + * we cannot do anything about that anymore. + * + * We still need to check if both bitmap and ldev are present, we may + * end up here after a failed attach, before ldev was even assigned. + */ + if (mdev->bitmap && mdev->ldev) { + if (drbd_bitmap_io_from_worker(mdev, drbd_bm_write, + "detach", BM_LOCKED_MASK)) { + if (test_bit(WAS_READ_ERROR, &mdev->flags)) { + drbd_md_set_flag(mdev, MDF_FULL_SYNC); + drbd_md_sync(mdev); + } + } + } + drbd_force_state(mdev, NS(disk, D_DISKLESS)); - return 1; + return 0; } void drbd_go_diskless(struct drbd_conf *mdev) { D_ASSERT(mdev->state.disk == D_FAILED); if (!test_and_set_bit(GO_DISKLESS, &mdev->flags)) - drbd_queue_work(&mdev->data.work, &mdev->go_diskless); + drbd_queue_work(&mdev->tconn->sender_work, &mdev->go_diskless); } /** @@ -4215,7 +3279,7 @@ void drbd_queue_bitmap_io(struct drbd_conf *mdev, void (*done)(struct drbd_conf *, int), char *why, enum bm_flag flags) { - D_ASSERT(current == mdev->worker.task); + D_ASSERT(current == mdev->tconn->worker.task); D_ASSERT(!test_bit(BITMAP_IO_QUEUED, &mdev->flags)); D_ASSERT(!test_bit(BITMAP_IO, &mdev->flags)); @@ -4229,13 +3293,13 @@ void drbd_queue_bitmap_io(struct drbd_conf *mdev, mdev->bm_io_work.why = why; mdev->bm_io_work.flags = flags; - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); set_bit(BITMAP_IO, &mdev->flags); if (atomic_read(&mdev->ap_bio_cnt) == 0) { if (!test_and_set_bit(BITMAP_IO_QUEUED, &mdev->flags)) - drbd_queue_work(&mdev->data.work, &mdev->bm_io_work.w); + drbd_queue_work(&mdev->tconn->sender_work, &mdev->bm_io_work.w); } - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); } /** @@ -4252,7 +3316,7 @@ int drbd_bitmap_io(struct drbd_conf *mdev, int (*io_fn)(struct drbd_conf *), { int rv; - D_ASSERT(current != mdev->worker.task); + D_ASSERT(current != mdev->tconn->worker.task); if ((flags & BM_LOCKED_SET_ALLOWED) == 0) drbd_suspend_io(mdev); @@ -4291,18 +3355,127 @@ static void md_sync_timer_fn(unsigned long data) { struct drbd_conf *mdev = (struct drbd_conf *) data; - drbd_queue_work_front(&mdev->data.work, &mdev->md_sync_work); + /* must not double-queue! */ + if (list_empty(&mdev->md_sync_work.list)) + drbd_queue_work_front(&mdev->tconn->sender_work, &mdev->md_sync_work); } -static int w_md_sync(struct drbd_conf *mdev, struct drbd_work *w, int unused) +static int w_md_sync(struct drbd_work *w, int unused) { + struct drbd_conf *mdev = w->mdev; + dev_warn(DEV, "md_sync_timer expired! Worker calls drbd_md_sync().\n"); #ifdef DEBUG dev_warn(DEV, "last md_mark_dirty: %s:%u\n", mdev->last_md_mark_dirty.func, mdev->last_md_mark_dirty.line); #endif drbd_md_sync(mdev); - return 1; + return 0; +} + +const char *cmdname(enum drbd_packet cmd) +{ + /* THINK may need to become several global tables + * when we want to support more than + * one PRO_VERSION */ + static const char *cmdnames[] = { + [P_DATA] = "Data", + [P_DATA_REPLY] = "DataReply", + [P_RS_DATA_REPLY] = "RSDataReply", + [P_BARRIER] = "Barrier", + [P_BITMAP] = "ReportBitMap", + [P_BECOME_SYNC_TARGET] = "BecomeSyncTarget", + [P_BECOME_SYNC_SOURCE] = "BecomeSyncSource", + [P_UNPLUG_REMOTE] = "UnplugRemote", + [P_DATA_REQUEST] = "DataRequest", + [P_RS_DATA_REQUEST] = "RSDataRequest", + [P_SYNC_PARAM] = "SyncParam", + [P_SYNC_PARAM89] = "SyncParam89", + [P_PROTOCOL] = "ReportProtocol", + [P_UUIDS] = "ReportUUIDs", + [P_SIZES] = "ReportSizes", + [P_STATE] = "ReportState", + [P_SYNC_UUID] = "ReportSyncUUID", + [P_AUTH_CHALLENGE] = "AuthChallenge", + [P_AUTH_RESPONSE] = "AuthResponse", + [P_PING] = "Ping", + [P_PING_ACK] = "PingAck", + [P_RECV_ACK] = "RecvAck", + [P_WRITE_ACK] = "WriteAck", + [P_RS_WRITE_ACK] = "RSWriteAck", + [P_SUPERSEDED] = "Superseded", + [P_NEG_ACK] = "NegAck", + [P_NEG_DREPLY] = "NegDReply", + [P_NEG_RS_DREPLY] = "NegRSDReply", + [P_BARRIER_ACK] = "BarrierAck", + [P_STATE_CHG_REQ] = "StateChgRequest", + [P_STATE_CHG_REPLY] = "StateChgReply", + [P_OV_REQUEST] = "OVRequest", + [P_OV_REPLY] = "OVReply", + [P_OV_RESULT] = "OVResult", + [P_CSUM_RS_REQUEST] = "CsumRSRequest", + [P_RS_IS_IN_SYNC] = "CsumRSIsInSync", + [P_COMPRESSED_BITMAP] = "CBitmap", + [P_DELAY_PROBE] = "DelayProbe", + [P_OUT_OF_SYNC] = "OutOfSync", + [P_RETRY_WRITE] = "RetryWrite", + [P_RS_CANCEL] = "RSCancel", + [P_CONN_ST_CHG_REQ] = "conn_st_chg_req", + [P_CONN_ST_CHG_REPLY] = "conn_st_chg_reply", + [P_RETRY_WRITE] = "retry_write", + [P_PROTOCOL_UPDATE] = "protocol_update", + + /* enum drbd_packet, but not commands - obsoleted flags: + * P_MAY_IGNORE + * P_MAX_OPT_CMD + */ + }; + + /* too big for the array: 0xfffX */ + if (cmd == P_INITIAL_META) + return "InitialMeta"; + if (cmd == P_INITIAL_DATA) + return "InitialData"; + if (cmd == P_CONNECTION_FEATURES) + return "ConnectionFeatures"; + if (cmd >= ARRAY_SIZE(cmdnames)) + return "Unknown"; + return cmdnames[cmd]; +} + +/** + * drbd_wait_misc - wait for a request to make progress + * @mdev: device associated with the request + * @i: the struct drbd_interval embedded in struct drbd_request or + * struct drbd_peer_request + */ +int drbd_wait_misc(struct drbd_conf *mdev, struct drbd_interval *i) +{ + struct net_conf *nc; + DEFINE_WAIT(wait); + long timeout; + + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + if (!nc) { + rcu_read_unlock(); + return -ETIMEDOUT; + } + timeout = nc->ko_count ? nc->timeout * HZ / 10 * nc->ko_count : MAX_SCHEDULE_TIMEOUT; + rcu_read_unlock(); + + /* Indicate to wake up mdev->misc_wait on progress. */ + i->waiting = true; + prepare_to_wait(&mdev->misc_wait, &wait, TASK_INTERRUPTIBLE); + spin_unlock_irq(&mdev->tconn->req_lock); + timeout = schedule_timeout(timeout); + finish_wait(&mdev->misc_wait, &wait); + spin_lock_irq(&mdev->tconn->req_lock); + if (!timeout || mdev->state.conn < C_CONNECTED) + return -ETIMEDOUT; + if (signal_pending(current)) + return -ERESTARTSYS; + return 0; } #ifdef CONFIG_DRBD_FAULT_INJECTION diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c index edb490aad8b4..2af26fc95280 100644 --- a/drivers/block/drbd/drbd_nl.c +++ b/drivers/block/drbd/drbd_nl.c @@ -29,159 +29,317 @@ #include <linux/fs.h> #include <linux/file.h> #include <linux/slab.h> -#include <linux/connector.h> #include <linux/blkpg.h> #include <linux/cpumask.h> #include "drbd_int.h" #include "drbd_req.h" #include "drbd_wrappers.h" #include <asm/unaligned.h> -#include <linux/drbd_tag_magic.h> #include <linux/drbd_limits.h> -#include <linux/compiler.h> #include <linux/kthread.h> -static unsigned short *tl_add_blob(unsigned short *, enum drbd_tags, const void *, int); -static unsigned short *tl_add_str(unsigned short *, enum drbd_tags, const char *); -static unsigned short *tl_add_int(unsigned short *, enum drbd_tags, const void *); - -/* see get_sb_bdev and bd_claim */ +#include <net/genetlink.h> + +/* .doit */ +// int drbd_adm_create_resource(struct sk_buff *skb, struct genl_info *info); +// int drbd_adm_delete_resource(struct sk_buff *skb, struct genl_info *info); + +int drbd_adm_add_minor(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_delete_minor(struct sk_buff *skb, struct genl_info *info); + +int drbd_adm_new_resource(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_del_resource(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_down(struct sk_buff *skb, struct genl_info *info); + +int drbd_adm_set_role(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_attach(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_detach(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_connect(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_net_opts(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_resize(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_start_ov(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_new_c_uuid(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_disconnect(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_invalidate(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_invalidate_peer(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_pause_sync(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_resume_sync(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_suspend_io(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_resume_io(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_outdate(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_resource_opts(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_get_status(struct sk_buff *skb, struct genl_info *info); +int drbd_adm_get_timeout_type(struct sk_buff *skb, struct genl_info *info); +/* .dumpit */ +int drbd_adm_get_status_all(struct sk_buff *skb, struct netlink_callback *cb); + +#include <linux/drbd_genl_api.h> +#include "drbd_nla.h" +#include <linux/genl_magic_func.h> + +/* used blkdev_get_by_path, to claim our meta data device(s) */ static char *drbd_m_holder = "Hands off! this is DRBD's meta data device."; -/* Generate the tag_list to struct functions */ -#define NL_PACKET(name, number, fields) \ -static int name ## _from_tags(struct drbd_conf *mdev, \ - unsigned short *tags, struct name *arg) __attribute__ ((unused)); \ -static int name ## _from_tags(struct drbd_conf *mdev, \ - unsigned short *tags, struct name *arg) \ -{ \ - int tag; \ - int dlen; \ - \ - while ((tag = get_unaligned(tags++)) != TT_END) { \ - dlen = get_unaligned(tags++); \ - switch (tag_number(tag)) { \ - fields \ - default: \ - if (tag & T_MANDATORY) { \ - dev_err(DEV, "Unknown tag: %d\n", tag_number(tag)); \ - return 0; \ - } \ - } \ - tags = (unsigned short *)((char *)tags + dlen); \ - } \ - return 1; \ -} -#define NL_INTEGER(pn, pr, member) \ - case pn: /* D_ASSERT( tag_type(tag) == TT_INTEGER ); */ \ - arg->member = get_unaligned((int *)(tags)); \ - break; -#define NL_INT64(pn, pr, member) \ - case pn: /* D_ASSERT( tag_type(tag) == TT_INT64 ); */ \ - arg->member = get_unaligned((u64 *)(tags)); \ +/* Configuration is strictly serialized, because generic netlink message + * processing is strictly serialized by the genl_lock(). + * Which means we can use one static global drbd_config_context struct. + */ +static struct drbd_config_context { + /* assigned from drbd_genlmsghdr */ + unsigned int minor; + /* assigned from request attributes, if present */ + unsigned int volume; +#define VOLUME_UNSPECIFIED (-1U) + /* pointer into the request skb, + * limited lifetime! */ + char *resource_name; + struct nlattr *my_addr; + struct nlattr *peer_addr; + + /* reply buffer */ + struct sk_buff *reply_skb; + /* pointer into reply buffer */ + struct drbd_genlmsghdr *reply_dh; + /* resolved from attributes, if possible */ + struct drbd_conf *mdev; + struct drbd_tconn *tconn; +} adm_ctx; + +static void drbd_adm_send_reply(struct sk_buff *skb, struct genl_info *info) +{ + genlmsg_end(skb, genlmsg_data(nlmsg_data(nlmsg_hdr(skb)))); + if (genlmsg_reply(skb, info)) + printk(KERN_ERR "drbd: error sending genl reply\n"); +} + +/* Used on a fresh "drbd_adm_prepare"d reply_skb, this cannot fail: The only + * reason it could fail was no space in skb, and there are 4k available. */ +int drbd_msg_put_info(const char *info) +{ + struct sk_buff *skb = adm_ctx.reply_skb; + struct nlattr *nla; + int err = -EMSGSIZE; + + if (!info || !info[0]) + return 0; + + nla = nla_nest_start(skb, DRBD_NLA_CFG_REPLY); + if (!nla) + return err; + + err = nla_put_string(skb, T_info_text, info); + if (err) { + nla_nest_cancel(skb, nla); + return err; + } else + nla_nest_end(skb, nla); + return 0; +} + +/* This would be a good candidate for a "pre_doit" hook, + * and per-family private info->pointers. + * But we need to stay compatible with older kernels. + * If it returns successfully, adm_ctx members are valid. + */ +#define DRBD_ADM_NEED_MINOR 1 +#define DRBD_ADM_NEED_RESOURCE 2 +#define DRBD_ADM_NEED_CONNECTION 4 +static int drbd_adm_prepare(struct sk_buff *skb, struct genl_info *info, + unsigned flags) +{ + struct drbd_genlmsghdr *d_in = info->userhdr; + const u8 cmd = info->genlhdr->cmd; + int err; + + memset(&adm_ctx, 0, sizeof(adm_ctx)); + + /* genl_rcv_msg only checks for CAP_NET_ADMIN on "GENL_ADMIN_PERM" :( */ + if (cmd != DRBD_ADM_GET_STATUS && !capable(CAP_NET_ADMIN)) + return -EPERM; + + adm_ctx.reply_skb = genlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!adm_ctx.reply_skb) { + err = -ENOMEM; + goto fail; + } + + adm_ctx.reply_dh = genlmsg_put_reply(adm_ctx.reply_skb, + info, &drbd_genl_family, 0, cmd); + /* put of a few bytes into a fresh skb of >= 4k will always succeed. + * but anyways */ + if (!adm_ctx.reply_dh) { + err = -ENOMEM; + goto fail; + } + + adm_ctx.reply_dh->minor = d_in->minor; + adm_ctx.reply_dh->ret_code = NO_ERROR; + + adm_ctx.volume = VOLUME_UNSPECIFIED; + if (info->attrs[DRBD_NLA_CFG_CONTEXT]) { + struct nlattr *nla; + /* parse and validate only */ + err = drbd_cfg_context_from_attrs(NULL, info); + if (err) + goto fail; + + /* It was present, and valid, + * copy it over to the reply skb. */ + err = nla_put_nohdr(adm_ctx.reply_skb, + info->attrs[DRBD_NLA_CFG_CONTEXT]->nla_len, + info->attrs[DRBD_NLA_CFG_CONTEXT]); + if (err) + goto fail; + + /* and assign stuff to the global adm_ctx */ + nla = nested_attr_tb[__nla_type(T_ctx_volume)]; + if (nla) + adm_ctx.volume = nla_get_u32(nla); + nla = nested_attr_tb[__nla_type(T_ctx_resource_name)]; + if (nla) + adm_ctx.resource_name = nla_data(nla); + adm_ctx.my_addr = nested_attr_tb[__nla_type(T_ctx_my_addr)]; + adm_ctx.peer_addr = nested_attr_tb[__nla_type(T_ctx_peer_addr)]; + if ((adm_ctx.my_addr && + nla_len(adm_ctx.my_addr) > sizeof(adm_ctx.tconn->my_addr)) || + (adm_ctx.peer_addr && + nla_len(adm_ctx.peer_addr) > sizeof(adm_ctx.tconn->peer_addr))) { + err = -EINVAL; + goto fail; + } + } + + adm_ctx.minor = d_in->minor; + adm_ctx.mdev = minor_to_mdev(d_in->minor); + adm_ctx.tconn = conn_get_by_name(adm_ctx.resource_name); + + if (!adm_ctx.mdev && (flags & DRBD_ADM_NEED_MINOR)) { + drbd_msg_put_info("unknown minor"); + return ERR_MINOR_INVALID; + } + if (!adm_ctx.tconn && (flags & DRBD_ADM_NEED_RESOURCE)) { + drbd_msg_put_info("unknown resource"); + return ERR_INVALID_REQUEST; + } + + if (flags & DRBD_ADM_NEED_CONNECTION) { + if (adm_ctx.tconn && !(flags & DRBD_ADM_NEED_RESOURCE)) { + drbd_msg_put_info("no resource name expected"); + return ERR_INVALID_REQUEST; + } + if (adm_ctx.mdev) { + drbd_msg_put_info("no minor number expected"); + return ERR_INVALID_REQUEST; + } + if (adm_ctx.my_addr && adm_ctx.peer_addr) + adm_ctx.tconn = conn_get_by_addrs(nla_data(adm_ctx.my_addr), + nla_len(adm_ctx.my_addr), + nla_data(adm_ctx.peer_addr), + nla_len(adm_ctx.peer_addr)); + if (!adm_ctx.tconn) { + drbd_msg_put_info("unknown connection"); + return ERR_INVALID_REQUEST; + } + } + + /* some more paranoia, if the request was over-determined */ + if (adm_ctx.mdev && adm_ctx.tconn && + adm_ctx.mdev->tconn != adm_ctx.tconn) { + pr_warning("request: minor=%u, resource=%s; but that minor belongs to connection %s\n", + adm_ctx.minor, adm_ctx.resource_name, + adm_ctx.mdev->tconn->name); + drbd_msg_put_info("minor exists in different resource"); + return ERR_INVALID_REQUEST; + } + if (adm_ctx.mdev && + adm_ctx.volume != VOLUME_UNSPECIFIED && + adm_ctx.volume != adm_ctx.mdev->vnr) { + pr_warning("request: minor=%u, volume=%u; but that minor is volume %u in %s\n", + adm_ctx.minor, adm_ctx.volume, + adm_ctx.mdev->vnr, adm_ctx.mdev->tconn->name); + drbd_msg_put_info("minor exists as different volume"); + return ERR_INVALID_REQUEST; + } + + return NO_ERROR; + +fail: + nlmsg_free(adm_ctx.reply_skb); + adm_ctx.reply_skb = NULL; + return err; +} + +static int drbd_adm_finish(struct genl_info *info, int retcode) +{ + if (adm_ctx.tconn) { + kref_put(&adm_ctx.tconn->kref, &conn_destroy); + adm_ctx.tconn = NULL; + } + + if (!adm_ctx.reply_skb) + return -ENOMEM; + + adm_ctx.reply_dh->ret_code = retcode; + drbd_adm_send_reply(adm_ctx.reply_skb, info); + return 0; +} + +static void setup_khelper_env(struct drbd_tconn *tconn, char **envp) +{ + char *afs; + + /* FIXME: A future version will not allow this case. */ + if (tconn->my_addr_len == 0 || tconn->peer_addr_len == 0) + return; + + switch (((struct sockaddr *)&tconn->peer_addr)->sa_family) { + case AF_INET6: + afs = "ipv6"; + snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI6", + &((struct sockaddr_in6 *)&tconn->peer_addr)->sin6_addr); break; -#define NL_BIT(pn, pr, member) \ - case pn: /* D_ASSERT( tag_type(tag) == TT_BIT ); */ \ - arg->member = *(char *)(tags) ? 1 : 0; \ + case AF_INET: + afs = "ipv4"; + snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI4", + &((struct sockaddr_in *)&tconn->peer_addr)->sin_addr); break; -#define NL_STRING(pn, pr, member, len) \ - case pn: /* D_ASSERT( tag_type(tag) == TT_STRING ); */ \ - if (dlen > len) { \ - dev_err(DEV, "arg too long: %s (%u wanted, max len: %u bytes)\n", \ - #member, dlen, (unsigned int)len); \ - return 0; \ - } \ - arg->member ## _len = dlen; \ - memcpy(arg->member, tags, min_t(size_t, dlen, len)); \ - break; -#include <linux/drbd_nl.h> - -/* Generate the struct to tag_list functions */ -#define NL_PACKET(name, number, fields) \ -static unsigned short* \ -name ## _to_tags(struct drbd_conf *mdev, \ - struct name *arg, unsigned short *tags) __attribute__ ((unused)); \ -static unsigned short* \ -name ## _to_tags(struct drbd_conf *mdev, \ - struct name *arg, unsigned short *tags) \ -{ \ - fields \ - return tags; \ -} - -#define NL_INTEGER(pn, pr, member) \ - put_unaligned(pn | pr | TT_INTEGER, tags++); \ - put_unaligned(sizeof(int), tags++); \ - put_unaligned(arg->member, (int *)tags); \ - tags = (unsigned short *)((char *)tags+sizeof(int)); -#define NL_INT64(pn, pr, member) \ - put_unaligned(pn | pr | TT_INT64, tags++); \ - put_unaligned(sizeof(u64), tags++); \ - put_unaligned(arg->member, (u64 *)tags); \ - tags = (unsigned short *)((char *)tags+sizeof(u64)); -#define NL_BIT(pn, pr, member) \ - put_unaligned(pn | pr | TT_BIT, tags++); \ - put_unaligned(sizeof(char), tags++); \ - *(char *)tags = arg->member; \ - tags = (unsigned short *)((char *)tags+sizeof(char)); -#define NL_STRING(pn, pr, member, len) \ - put_unaligned(pn | pr | TT_STRING, tags++); \ - put_unaligned(arg->member ## _len, tags++); \ - memcpy(tags, arg->member, arg->member ## _len); \ - tags = (unsigned short *)((char *)tags + arg->member ## _len); -#include <linux/drbd_nl.h> - -void drbd_bcast_ev_helper(struct drbd_conf *mdev, char *helper_name); -void drbd_nl_send_reply(struct cn_msg *, int); + default: + afs = "ssocks"; + snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI4", + &((struct sockaddr_in *)&tconn->peer_addr)->sin_addr); + } + snprintf(envp[3], 20, "DRBD_PEER_AF=%s", afs); +} int drbd_khelper(struct drbd_conf *mdev, char *cmd) { char *envp[] = { "HOME=/", "TERM=linux", "PATH=/sbin:/usr/sbin:/bin:/usr/bin", - NULL, /* Will be set to address family */ - NULL, /* Will be set to address */ + (char[20]) { }, /* address family */ + (char[60]) { }, /* address */ NULL }; - - char mb[12], af[20], ad[60], *afs; + char mb[12]; char *argv[] = {usermode_helper, cmd, mb, NULL }; + struct drbd_tconn *tconn = mdev->tconn; + struct sib_info sib; int ret; - if (current == mdev->worker.task) - set_bit(CALLBACK_PENDING, &mdev->flags); + if (current == tconn->worker.task) + set_bit(CALLBACK_PENDING, &tconn->flags); snprintf(mb, 12, "minor-%d", mdev_to_minor(mdev)); - - if (get_net_conf(mdev)) { - switch (((struct sockaddr *)mdev->net_conf->peer_addr)->sa_family) { - case AF_INET6: - afs = "ipv6"; - snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI6", - &((struct sockaddr_in6 *)mdev->net_conf->peer_addr)->sin6_addr); - break; - case AF_INET: - afs = "ipv4"; - snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI4", - &((struct sockaddr_in *)mdev->net_conf->peer_addr)->sin_addr); - break; - default: - afs = "ssocks"; - snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI4", - &((struct sockaddr_in *)mdev->net_conf->peer_addr)->sin_addr); - } - snprintf(af, 20, "DRBD_PEER_AF=%s", afs); - envp[3]=af; - envp[4]=ad; - put_net_conf(mdev); - } + setup_khelper_env(tconn, envp); /* The helper may take some time. * write out any unsynced meta data changes now */ drbd_md_sync(mdev); dev_info(DEV, "helper command: %s %s %s\n", usermode_helper, cmd, mb); - - drbd_bcast_ev_helper(mdev, cmd); + sib.sib_reason = SIB_HELPER_PRE; + sib.helper_name = cmd; + drbd_bcast_event(mdev, &sib); ret = call_usermodehelper(usermode_helper, argv, envp, UMH_WAIT_PROC); if (ret) dev_warn(DEV, "helper command: %s %s %s exit code %u (0x%x)\n", @@ -191,9 +349,46 @@ int drbd_khelper(struct drbd_conf *mdev, char *cmd) dev_info(DEV, "helper command: %s %s %s exit code %u (0x%x)\n", usermode_helper, cmd, mb, (ret >> 8) & 0xff, ret); + sib.sib_reason = SIB_HELPER_POST; + sib.helper_exit_code = ret; + drbd_bcast_event(mdev, &sib); + + if (current == tconn->worker.task) + clear_bit(CALLBACK_PENDING, &tconn->flags); + + if (ret < 0) /* Ignore any ERRNOs we got. */ + ret = 0; + + return ret; +} + +int conn_khelper(struct drbd_tconn *tconn, char *cmd) +{ + char *envp[] = { "HOME=/", + "TERM=linux", + "PATH=/sbin:/usr/sbin:/bin:/usr/bin", + (char[20]) { }, /* address family */ + (char[60]) { }, /* address */ + NULL }; + char *argv[] = {usermode_helper, cmd, tconn->name, NULL }; + int ret; + + setup_khelper_env(tconn, envp); + conn_md_sync(tconn); - if (current == mdev->worker.task) - clear_bit(CALLBACK_PENDING, &mdev->flags); + conn_info(tconn, "helper command: %s %s %s\n", usermode_helper, cmd, tconn->name); + /* TODO: conn_bcast_event() ?? */ + + ret = call_usermodehelper(usermode_helper, argv, envp, UMH_WAIT_PROC); + if (ret) + conn_warn(tconn, "helper command: %s %s %s exit code %u (0x%x)\n", + usermode_helper, cmd, tconn->name, + (ret >> 8) & 0xff, ret); + else + conn_info(tconn, "helper command: %s %s %s exit code %u (0x%x)\n", + usermode_helper, cmd, tconn->name, + (ret >> 8) & 0xff, ret); + /* TODO: conn_bcast_event() ?? */ if (ret < 0) /* Ignore any ERRNOs we got. */ ret = 0; @@ -201,116 +396,129 @@ int drbd_khelper(struct drbd_conf *mdev, char *cmd) return ret; } -enum drbd_disk_state drbd_try_outdate_peer(struct drbd_conf *mdev) +static enum drbd_fencing_p highest_fencing_policy(struct drbd_tconn *tconn) { + enum drbd_fencing_p fp = FP_NOT_AVAIL; + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + if (get_ldev_if_state(mdev, D_CONSISTENT)) { + fp = max_t(enum drbd_fencing_p, fp, + rcu_dereference(mdev->ldev->disk_conf)->fencing); + put_ldev(mdev); + } + } + rcu_read_unlock(); + + return fp; +} + +bool conn_try_outdate_peer(struct drbd_tconn *tconn) +{ + union drbd_state mask = { }; + union drbd_state val = { }; + enum drbd_fencing_p fp; char *ex_to_string; int r; - enum drbd_disk_state nps; - enum drbd_fencing_p fp; - D_ASSERT(mdev->state.pdsk == D_UNKNOWN); + if (tconn->cstate >= C_WF_REPORT_PARAMS) { + conn_err(tconn, "Expected cstate < C_WF_REPORT_PARAMS\n"); + return false; + } - if (get_ldev_if_state(mdev, D_CONSISTENT)) { - fp = mdev->ldev->dc.fencing; - put_ldev(mdev); - } else { - dev_warn(DEV, "Not fencing peer, I'm not even Consistent myself.\n"); - nps = mdev->state.pdsk; + fp = highest_fencing_policy(tconn); + switch (fp) { + case FP_NOT_AVAIL: + conn_warn(tconn, "Not fencing peer, I'm not even Consistent myself.\n"); goto out; + case FP_DONT_CARE: + return true; + default: ; } - r = drbd_khelper(mdev, "fence-peer"); + r = conn_khelper(tconn, "fence-peer"); switch ((r>>8) & 0xff) { case 3: /* peer is inconsistent */ ex_to_string = "peer is inconsistent or worse"; - nps = D_INCONSISTENT; + mask.pdsk = D_MASK; + val.pdsk = D_INCONSISTENT; break; case 4: /* peer got outdated, or was already outdated */ ex_to_string = "peer was fenced"; - nps = D_OUTDATED; + mask.pdsk = D_MASK; + val.pdsk = D_OUTDATED; break; case 5: /* peer was down */ - if (mdev->state.disk == D_UP_TO_DATE) { + if (conn_highest_disk(tconn) == D_UP_TO_DATE) { /* we will(have) create(d) a new UUID anyways... */ ex_to_string = "peer is unreachable, assumed to be dead"; - nps = D_OUTDATED; + mask.pdsk = D_MASK; + val.pdsk = D_OUTDATED; } else { ex_to_string = "peer unreachable, doing nothing since disk != UpToDate"; - nps = mdev->state.pdsk; } break; case 6: /* Peer is primary, voluntarily outdate myself. * This is useful when an unconnected R_SECONDARY is asked to * become R_PRIMARY, but finds the other peer being active. */ ex_to_string = "peer is active"; - dev_warn(DEV, "Peer is primary, outdating myself.\n"); - nps = D_UNKNOWN; - _drbd_request_state(mdev, NS(disk, D_OUTDATED), CS_WAIT_COMPLETE); + conn_warn(tconn, "Peer is primary, outdating myself.\n"); + mask.disk = D_MASK; + val.disk = D_OUTDATED; break; case 7: if (fp != FP_STONITH) - dev_err(DEV, "fence-peer() = 7 && fencing != Stonith !!!\n"); + conn_err(tconn, "fence-peer() = 7 && fencing != Stonith !!!\n"); ex_to_string = "peer was stonithed"; - nps = D_OUTDATED; + mask.pdsk = D_MASK; + val.pdsk = D_OUTDATED; break; default: /* The script is broken ... */ - nps = D_UNKNOWN; - dev_err(DEV, "fence-peer helper broken, returned %d\n", (r>>8)&0xff); - return nps; + conn_err(tconn, "fence-peer helper broken, returned %d\n", (r>>8)&0xff); + return false; /* Eventually leave IO frozen */ } - dev_info(DEV, "fence-peer helper returned %d (%s)\n", - (r>>8) & 0xff, ex_to_string); + conn_info(tconn, "fence-peer helper returned %d (%s)\n", + (r>>8) & 0xff, ex_to_string); -out: - if (mdev->state.susp_fen && nps >= D_UNKNOWN) { - /* The handler was not successful... unfreeze here, the - state engine can not unfreeze... */ - _drbd_request_state(mdev, NS(susp_fen, 0), CS_VERBOSE); - } + out: - return nps; + /* Not using + conn_request_state(tconn, mask, val, CS_VERBOSE); + here, because we might were able to re-establish the connection in the + meantime. */ + spin_lock_irq(&tconn->req_lock); + if (tconn->cstate < C_WF_REPORT_PARAMS && !test_bit(STATE_SENT, &tconn->flags)) + _conn_request_state(tconn, mask, val, CS_VERBOSE); + spin_unlock_irq(&tconn->req_lock); + + return conn_highest_pdsk(tconn) <= D_OUTDATED; } static int _try_outdate_peer_async(void *data) { - struct drbd_conf *mdev = (struct drbd_conf *)data; - enum drbd_disk_state nps; - union drbd_state ns; + struct drbd_tconn *tconn = (struct drbd_tconn *)data; - nps = drbd_try_outdate_peer(mdev); - - /* Not using - drbd_request_state(mdev, NS(pdsk, nps)); - here, because we might were able to re-establish the connection - in the meantime. This can only partially be solved in the state's - engine is_valid_state() and is_valid_state_transition() - functions. - - nps can be D_INCONSISTENT, D_OUTDATED or D_UNKNOWN. - pdsk == D_INCONSISTENT while conn >= C_CONNECTED is valid, - therefore we have to have the pre state change check here. - */ - spin_lock_irq(&mdev->req_lock); - ns = mdev->state; - if (ns.conn < C_WF_REPORT_PARAMS && !test_bit(STATE_SENT, &mdev->flags)) { - ns.pdsk = nps; - _drbd_set_state(mdev, ns, CS_VERBOSE, NULL); - } - spin_unlock_irq(&mdev->req_lock); + conn_try_outdate_peer(tconn); + kref_put(&tconn->kref, &conn_destroy); return 0; } -void drbd_try_outdate_peer_async(struct drbd_conf *mdev) +void conn_try_outdate_peer_async(struct drbd_tconn *tconn) { struct task_struct *opa; - opa = kthread_run(_try_outdate_peer_async, mdev, "drbd%d_a_helper", mdev_to_minor(mdev)); - if (IS_ERR(opa)) - dev_err(DEV, "out of mem, failed to invoke fence-peer helper\n"); + kref_get(&tconn->kref); + opa = kthread_run(_try_outdate_peer_async, tconn, "drbd_async_h"); + if (IS_ERR(opa)) { + conn_err(tconn, "out of mem, failed to invoke fence-peer helper\n"); + kref_put(&tconn->kref, &conn_destroy); + } } enum drbd_state_rv @@ -318,15 +526,15 @@ drbd_set_role(struct drbd_conf *mdev, enum drbd_role new_role, int force) { const int max_tries = 4; enum drbd_state_rv rv = SS_UNKNOWN_ERROR; + struct net_conf *nc; int try = 0; int forced = 0; union drbd_state mask, val; - enum drbd_disk_state nps; if (new_role == R_PRIMARY) - request_ping(mdev); /* Detect a dead peer ASAP */ + request_ping(mdev->tconn); /* Detect a dead peer ASAP */ - mutex_lock(&mdev->state_mutex); + mutex_lock(mdev->state_mutex); mask.i = 0; mask.role = R_MASK; val.i = 0; val.role = new_role; @@ -354,38 +562,34 @@ drbd_set_role(struct drbd_conf *mdev, enum drbd_role new_role, int force) if (rv == SS_NO_UP_TO_DATE_DISK && mdev->state.disk == D_CONSISTENT && mask.pdsk == 0) { D_ASSERT(mdev->state.pdsk == D_UNKNOWN); - nps = drbd_try_outdate_peer(mdev); - if (nps == D_OUTDATED || nps == D_INCONSISTENT) { + if (conn_try_outdate_peer(mdev->tconn)) { val.disk = D_UP_TO_DATE; mask.disk = D_MASK; } - - val.pdsk = nps; - mask.pdsk = D_MASK; - continue; } if (rv == SS_NOTHING_TO_DO) - goto fail; + goto out; if (rv == SS_PRIMARY_NOP && mask.pdsk == 0) { - nps = drbd_try_outdate_peer(mdev); - - if (force && nps > D_OUTDATED) { + if (!conn_try_outdate_peer(mdev->tconn) && force) { dev_warn(DEV, "Forced into split brain situation!\n"); - nps = D_OUTDATED; - } - - mask.pdsk = D_MASK; - val.pdsk = nps; + mask.pdsk = D_MASK; + val.pdsk = D_OUTDATED; + } continue; } if (rv == SS_TWO_PRIMARIES) { /* Maybe the peer is detected as dead very soon... retry at most once more in this case. */ - schedule_timeout_interruptible((mdev->net_conf->ping_timeo+1)*HZ/10); + int timeo; + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + timeo = nc ? (nc->ping_timeo + 1) * HZ / 10 : 1; + rcu_read_unlock(); + schedule_timeout_interruptible(timeo); if (try < max_tries) try = max_tries - 1; continue; @@ -394,13 +598,13 @@ drbd_set_role(struct drbd_conf *mdev, enum drbd_role new_role, int force) rv = _drbd_request_state(mdev, mask, val, CS_VERBOSE + CS_WAIT_COMPLETE); if (rv < SS_SUCCESS) - goto fail; + goto out; } break; } if (rv < SS_SUCCESS) - goto fail; + goto out; if (forced) dev_warn(DEV, "Forced to consider local data as UpToDate!\n"); @@ -408,6 +612,8 @@ drbd_set_role(struct drbd_conf *mdev, enum drbd_role new_role, int force) /* Wait until nothing is on the fly :) */ wait_event(mdev->misc_wait, atomic_read(&mdev->ap_pending_cnt) == 0); + /* FIXME also wait for all pending P_BARRIER_ACK? */ + if (new_role == R_SECONDARY) { set_disk_ro(mdev->vdisk, true); if (get_ldev(mdev)) { @@ -415,10 +621,12 @@ drbd_set_role(struct drbd_conf *mdev, enum drbd_role new_role, int force) put_ldev(mdev); } } else { - if (get_net_conf(mdev)) { - mdev->net_conf->want_lose = 0; - put_net_conf(mdev); - } + mutex_lock(&mdev->tconn->conf_update); + nc = mdev->tconn->net_conf; + if (nc) + nc->discard_my_data = 0; /* without copy; single bit op is atomic */ + mutex_unlock(&mdev->tconn->conf_update); + set_disk_ro(mdev->vdisk, false); if (get_ldev(mdev)) { if (((mdev->state.conn < C_CONNECTED || @@ -444,67 +652,47 @@ drbd_set_role(struct drbd_conf *mdev, enum drbd_role new_role, int force) drbd_md_sync(mdev); kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE); - fail: - mutex_unlock(&mdev->state_mutex); +out: + mutex_unlock(mdev->state_mutex); return rv; } -static struct drbd_conf *ensure_mdev(int minor, int create) +static const char *from_attrs_err_to_txt(int err) { - struct drbd_conf *mdev; - - if (minor >= minor_count) - return NULL; - - mdev = minor_to_mdev(minor); - - if (!mdev && create) { - struct gendisk *disk = NULL; - mdev = drbd_new_device(minor); - - spin_lock_irq(&drbd_pp_lock); - if (minor_table[minor] == NULL) { - minor_table[minor] = mdev; - disk = mdev->vdisk; - mdev = NULL; - } /* else: we lost the race */ - spin_unlock_irq(&drbd_pp_lock); - - if (disk) /* we won the race above */ - /* in case we ever add a drbd_delete_device(), - * don't forget the del_gendisk! */ - add_disk(disk); - else /* we lost the race above */ - drbd_free_mdev(mdev); - - mdev = minor_to_mdev(minor); - } - - return mdev; + return err == -ENOMSG ? "required attribute missing" : + err == -EOPNOTSUPP ? "unknown mandatory attribute" : + err == -EEXIST ? "can not change invariant setting" : + "invalid attribute value"; } -static int drbd_nl_primary(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_set_role(struct sk_buff *skb, struct genl_info *info) { - struct primary primary_args; - - memset(&primary_args, 0, sizeof(struct primary)); - if (!primary_from_tags(mdev, nlp->tag_list, &primary_args)) { - reply->ret_code = ERR_MANDATORY_TAG; - return 0; - } - - reply->ret_code = - drbd_set_role(mdev, R_PRIMARY, primary_args.primary_force); + struct set_role_parms parms; + int err; + enum drbd_ret_code retcode; - return 0; -} + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; -static int drbd_nl_secondary(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) -{ - reply->ret_code = drbd_set_role(mdev, R_SECONDARY, 0); + memset(&parms, 0, sizeof(parms)); + if (info->attrs[DRBD_NLA_SET_ROLE_PARMS]) { + err = set_role_parms_from_attrs(&parms, info); + if (err) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + goto out; + } + } + if (info->genlhdr->cmd == DRBD_ADM_PRIMARY) + retcode = drbd_set_role(adm_ctx.mdev, R_PRIMARY, parms.assume_uptodate); + else + retcode = drbd_set_role(adm_ctx.mdev, R_SECONDARY, 0); +out: + drbd_adm_finish(info, retcode); return 0; } @@ -514,7 +702,12 @@ static void drbd_md_set_sector_offsets(struct drbd_conf *mdev, struct drbd_backing_dev *bdev) { sector_t md_size_sect = 0; - switch (bdev->dc.meta_dev_idx) { + int meta_dev_idx; + + rcu_read_lock(); + meta_dev_idx = rcu_dereference(bdev->disk_conf)->meta_dev_idx; + + switch (meta_dev_idx) { default: /* v07 style fixed size indexed meta data */ bdev->md.md_size_sect = MD_RESERVED_SECT; @@ -533,7 +726,7 @@ static void drbd_md_set_sector_offsets(struct drbd_conf *mdev, case DRBD_MD_INDEX_FLEX_INT: bdev->md.md_offset = drbd_md_ss__(mdev, bdev); /* al size is still fixed */ - bdev->md.al_offset = -MD_AL_MAX_SIZE; + bdev->md.al_offset = -MD_AL_SECTORS; /* we need (slightly less than) ~ this much bitmap sectors: */ md_size_sect = drbd_get_capacity(bdev->backing_bdev); md_size_sect = ALIGN(md_size_sect, BM_SECT_PER_EXT); @@ -549,6 +742,7 @@ static void drbd_md_set_sector_offsets(struct drbd_conf *mdev, bdev->md.bm_offset = -md_size_sect + MD_AL_OFFSET; break; } + rcu_read_unlock(); } /* input size is expected to be in KB */ @@ -581,10 +775,16 @@ char *ppsize(char *buf, unsigned long long size) * R_PRIMARY D_INCONSISTENT, and C_SYNC_TARGET: * peer may not initiate a resize. */ +/* Note these are not to be confused with + * drbd_adm_suspend_io/drbd_adm_resume_io, + * which are (sub) state changes triggered by admin (drbdsetup), + * and can be long lived. + * This changes an mdev->flag, is triggered by drbd internals, + * and should be short-lived. */ void drbd_suspend_io(struct drbd_conf *mdev) { set_bit(SUSPEND_IO, &mdev->flags); - if (is_susp(mdev->state)) + if (drbd_suspended(mdev)) return; wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_bio_cnt)); } @@ -605,7 +805,7 @@ void drbd_resume_io(struct drbd_conf *mdev) enum determine_dev_size drbd_determine_dev_size(struct drbd_conf *mdev, enum dds_flags flags) __must_hold(local) { sector_t prev_first_sect, prev_size; /* previous meta location */ - sector_t la_size; + sector_t la_size, u_size; sector_t size; char ppb[10]; @@ -633,7 +833,10 @@ enum determine_dev_size drbd_determine_dev_size(struct drbd_conf *mdev, enum dds /* TODO: should only be some assert here, not (re)init... */ drbd_md_set_sector_offsets(mdev, mdev->ldev); - size = drbd_new_dev_size(mdev, mdev->ldev, flags & DDSF_FORCED); + rcu_read_lock(); + u_size = rcu_dereference(mdev->ldev->disk_conf)->disk_size; + rcu_read_unlock(); + size = drbd_new_dev_size(mdev, mdev->ldev, u_size, flags & DDSF_FORCED); if (drbd_get_capacity(mdev->this_bdev) != size || drbd_bm_capacity(mdev) != size) { @@ -696,12 +899,12 @@ out: } sector_t -drbd_new_dev_size(struct drbd_conf *mdev, struct drbd_backing_dev *bdev, int assume_peer_has_space) +drbd_new_dev_size(struct drbd_conf *mdev, struct drbd_backing_dev *bdev, + sector_t u_size, int assume_peer_has_space) { sector_t p_size = mdev->p_size; /* partner's disk size. */ sector_t la_size = bdev->md.la_size_sect; /* last agreed size. */ sector_t m_size; /* my size */ - sector_t u_size = bdev->dc.disk_size; /* size requested by user. */ sector_t size = 0; m_size = drbd_get_max_capacity(bdev); @@ -750,24 +953,21 @@ drbd_new_dev_size(struct drbd_conf *mdev, struct drbd_backing_dev *bdev, int ass * failed, and 0 on success. You should call drbd_md_sync() after you called * this function. */ -static int drbd_check_al_size(struct drbd_conf *mdev) +static int drbd_check_al_size(struct drbd_conf *mdev, struct disk_conf *dc) { struct lru_cache *n, *t; struct lc_element *e; unsigned int in_use; int i; - ERR_IF(mdev->sync_conf.al_extents < 7) - mdev->sync_conf.al_extents = 127; - if (mdev->act_log && - mdev->act_log->nr_elements == mdev->sync_conf.al_extents) + mdev->act_log->nr_elements == dc->al_extents) return 0; in_use = 0; t = mdev->act_log; - n = lc_create("act_log", drbd_al_ext_cache, - mdev->sync_conf.al_extents, sizeof(struct lc_element), 0); + n = lc_create("act_log", drbd_al_ext_cache, AL_UPDATES_PER_TRANSACTION, + dc->al_extents, sizeof(struct lc_element), 0); if (n == NULL) { dev_err(DEV, "Cannot allocate act_log lru!\n"); @@ -808,7 +1008,9 @@ static void drbd_setup_queue_param(struct drbd_conf *mdev, unsigned int max_bio_ struct request_queue * const b = mdev->ldev->backing_bdev->bd_disk->queue; max_hw_sectors = min(queue_max_hw_sectors(b), max_bio_size >> 9); - max_segments = mdev->ldev->dc.max_bio_bvecs; + rcu_read_lock(); + max_segments = rcu_dereference(mdev->ldev->disk_conf)->max_bio_bvecs; + rcu_read_unlock(); put_ldev(mdev); } @@ -852,12 +1054,14 @@ void drbd_reconsider_max_bio_size(struct drbd_conf *mdev) Because new from 8.3.8 onwards the peer can use multiple BIOs for a single peer_request */ if (mdev->state.conn >= C_CONNECTED) { - if (mdev->agreed_pro_version < 94) { - peer = min(mdev->peer_max_bio_size, DRBD_MAX_SIZE_H80_PACKET); + if (mdev->tconn->agreed_pro_version < 94) + peer = min( mdev->peer_max_bio_size, DRBD_MAX_SIZE_H80_PACKET); /* Correct old drbd (up to 8.3.7) if it believes it can do more than 32KiB */ - } else if (mdev->agreed_pro_version == 94) + else if (mdev->tconn->agreed_pro_version == 94) peer = DRBD_MAX_SIZE_H80_PACKET; - else /* drbd 8.3.8 onwards */ + else if (mdev->tconn->agreed_pro_version < 100) + peer = DRBD_MAX_BIO_SIZE_P95; /* drbd 8.3.8 onwards, before 8.4.0 */ + else peer = DRBD_MAX_BIO_SIZE; } @@ -872,36 +1076,27 @@ void drbd_reconsider_max_bio_size(struct drbd_conf *mdev) drbd_setup_queue_param(mdev, new); } -/* serialize deconfig (worker exiting, doing cleanup) - * and reconfig (drbdsetup disk, drbdsetup net) - * - * Wait for a potentially exiting worker, then restart it, - * or start a new one. Flush any pending work, there may still be an - * after_state_change queued. - */ -static void drbd_reconfig_start(struct drbd_conf *mdev) +/* Starts the worker thread */ +static void conn_reconfig_start(struct drbd_tconn *tconn) { - wait_event(mdev->state_wait, !test_and_set_bit(CONFIG_PENDING, &mdev->flags)); - wait_event(mdev->state_wait, !test_bit(DEVICE_DYING, &mdev->flags)); - drbd_thread_start(&mdev->worker); - drbd_flush_workqueue(mdev); + drbd_thread_start(&tconn->worker); + conn_flush_workqueue(tconn); } -/* if still unconfigured, stops worker again. - * if configured now, clears CONFIG_PENDING. - * wakes potential waiters */ -static void drbd_reconfig_done(struct drbd_conf *mdev) +/* if still unconfigured, stops worker again. */ +static void conn_reconfig_done(struct drbd_tconn *tconn) { - spin_lock_irq(&mdev->req_lock); - if (mdev->state.disk == D_DISKLESS && - mdev->state.conn == C_STANDALONE && - mdev->state.role == R_SECONDARY) { - set_bit(DEVICE_DYING, &mdev->flags); - drbd_thread_stop_nowait(&mdev->worker); - } else - clear_bit(CONFIG_PENDING, &mdev->flags); - spin_unlock_irq(&mdev->req_lock); - wake_up(&mdev->state_wait); + bool stop_threads; + spin_lock_irq(&tconn->req_lock); + stop_threads = conn_all_vols_unconf(tconn) && + tconn->cstate == C_STANDALONE; + spin_unlock_irq(&tconn->req_lock); + if (stop_threads) { + /* asender is implicitly stopped by receiver + * in conn_disconnect() */ + drbd_thread_stop(&tconn->receiver); + drbd_thread_stop(&tconn->worker); + } } /* Make sure IO is suspended before calling this function(). */ @@ -909,42 +1104,187 @@ static void drbd_suspend_al(struct drbd_conf *mdev) { int s = 0; - if (lc_try_lock(mdev->act_log)) { - drbd_al_shrink(mdev); - lc_unlock(mdev->act_log); - } else { + if (!lc_try_lock(mdev->act_log)) { dev_warn(DEV, "Failed to lock al in drbd_suspend_al()\n"); return; } - spin_lock_irq(&mdev->req_lock); + drbd_al_shrink(mdev); + spin_lock_irq(&mdev->tconn->req_lock); if (mdev->state.conn < C_CONNECTED) s = !test_and_set_bit(AL_SUSPENDED, &mdev->flags); - - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); + lc_unlock(mdev->act_log); if (s) dev_info(DEV, "Suspended AL updates\n"); } -/* does always return 0; - * interesting return code is in reply->ret_code */ -static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) + +static bool should_set_defaults(struct genl_info *info) +{ + unsigned flags = ((struct drbd_genlmsghdr*)info->userhdr)->flags; + return 0 != (flags & DRBD_GENL_F_SET_DEFAULTS); +} + +static void enforce_disk_conf_limits(struct disk_conf *dc) +{ + if (dc->al_extents < DRBD_AL_EXTENTS_MIN) + dc->al_extents = DRBD_AL_EXTENTS_MIN; + if (dc->al_extents > DRBD_AL_EXTENTS_MAX) + dc->al_extents = DRBD_AL_EXTENTS_MAX; + + if (dc->c_plan_ahead > DRBD_C_PLAN_AHEAD_MAX) + dc->c_plan_ahead = DRBD_C_PLAN_AHEAD_MAX; +} + +int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info) { enum drbd_ret_code retcode; + struct drbd_conf *mdev; + struct disk_conf *new_disk_conf, *old_disk_conf; + struct fifo_buffer *old_plan = NULL, *new_plan = NULL; + int err, fifo_size; + + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + + mdev = adm_ctx.mdev; + + /* we also need a disk + * to change the options on */ + if (!get_ldev(mdev)) { + retcode = ERR_NO_DISK; + goto out; + } + + new_disk_conf = kmalloc(sizeof(struct disk_conf), GFP_KERNEL); + if (!new_disk_conf) { + retcode = ERR_NOMEM; + goto fail; + } + + mutex_lock(&mdev->tconn->conf_update); + old_disk_conf = mdev->ldev->disk_conf; + *new_disk_conf = *old_disk_conf; + if (should_set_defaults(info)) + set_disk_conf_defaults(new_disk_conf); + + err = disk_conf_from_attrs_for_change(new_disk_conf, info); + if (err && err != -ENOMSG) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + } + + if (!expect(new_disk_conf->resync_rate >= 1)) + new_disk_conf->resync_rate = 1; + + enforce_disk_conf_limits(new_disk_conf); + + fifo_size = (new_disk_conf->c_plan_ahead * 10 * SLEEP_TIME) / HZ; + if (fifo_size != mdev->rs_plan_s->size) { + new_plan = fifo_alloc(fifo_size); + if (!new_plan) { + dev_err(DEV, "kmalloc of fifo_buffer failed"); + retcode = ERR_NOMEM; + goto fail_unlock; + } + } + + drbd_suspend_io(mdev); + wait_event(mdev->al_wait, lc_try_lock(mdev->act_log)); + drbd_al_shrink(mdev); + err = drbd_check_al_size(mdev, new_disk_conf); + lc_unlock(mdev->act_log); + wake_up(&mdev->al_wait); + drbd_resume_io(mdev); + + if (err) { + retcode = ERR_NOMEM; + goto fail_unlock; + } + + write_lock_irq(&global_state_lock); + retcode = drbd_resync_after_valid(mdev, new_disk_conf->resync_after); + if (retcode == NO_ERROR) { + rcu_assign_pointer(mdev->ldev->disk_conf, new_disk_conf); + drbd_resync_after_changed(mdev); + } + write_unlock_irq(&global_state_lock); + + if (retcode != NO_ERROR) + goto fail_unlock; + + if (new_plan) { + old_plan = mdev->rs_plan_s; + rcu_assign_pointer(mdev->rs_plan_s, new_plan); + } + + mutex_unlock(&mdev->tconn->conf_update); + + if (new_disk_conf->al_updates) + mdev->ldev->md.flags &= ~MDF_AL_DISABLED; + else + mdev->ldev->md.flags |= MDF_AL_DISABLED; + + if (new_disk_conf->md_flushes) + clear_bit(MD_NO_FUA, &mdev->flags); + else + set_bit(MD_NO_FUA, &mdev->flags); + + drbd_bump_write_ordering(mdev->tconn, WO_bdev_flush); + + drbd_md_sync(mdev); + + if (mdev->state.conn >= C_CONNECTED) + drbd_send_sync_param(mdev); + + synchronize_rcu(); + kfree(old_disk_conf); + kfree(old_plan); + mod_timer(&mdev->request_timer, jiffies + HZ); + goto success; + +fail_unlock: + mutex_unlock(&mdev->tconn->conf_update); + fail: + kfree(new_disk_conf); + kfree(new_plan); +success: + put_ldev(mdev); + out: + drbd_adm_finish(info, retcode); + return 0; +} + +int drbd_adm_attach(struct sk_buff *skb, struct genl_info *info) +{ + struct drbd_conf *mdev; + int err; + enum drbd_ret_code retcode; enum determine_dev_size dd; sector_t max_possible_sectors; sector_t min_md_device_sectors; struct drbd_backing_dev *nbc = NULL; /* new_backing_conf */ + struct disk_conf *new_disk_conf = NULL; struct block_device *bdev; struct lru_cache *resync_lru = NULL; + struct fifo_buffer *new_plan = NULL; union drbd_state ns, os; enum drbd_state_rv rv; - int cp_discovered = 0; - int logical_block_size; + struct net_conf *nc; - drbd_reconfig_start(mdev); + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto finish; + + mdev = adm_ctx.mdev; + conn_reconfig_start(mdev->tconn); /* if you want to reconfigure, please tear down first */ if (mdev->state.disk > D_DISKLESS) { @@ -959,47 +1299,65 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp /* make sure there is no leftover from previous force-detach attempts */ clear_bit(FORCE_DETACH, &mdev->flags); + clear_bit(WAS_IO_ERROR, &mdev->flags); + clear_bit(WAS_READ_ERROR, &mdev->flags); /* and no leftover from previously aborted resync or verify, either */ mdev->rs_total = 0; mdev->rs_failed = 0; atomic_set(&mdev->rs_pending_cnt, 0); - /* allocation not in the IO path, cqueue thread context */ + /* allocation not in the IO path, drbdsetup context */ nbc = kzalloc(sizeof(struct drbd_backing_dev), GFP_KERNEL); if (!nbc) { retcode = ERR_NOMEM; goto fail; } + spin_lock_init(&nbc->md.uuid_lock); - nbc->dc.disk_size = DRBD_DISK_SIZE_SECT_DEF; - nbc->dc.on_io_error = DRBD_ON_IO_ERROR_DEF; - nbc->dc.fencing = DRBD_FENCING_DEF; - nbc->dc.max_bio_bvecs = DRBD_MAX_BIO_BVECS_DEF; + new_disk_conf = kzalloc(sizeof(struct disk_conf), GFP_KERNEL); + if (!new_disk_conf) { + retcode = ERR_NOMEM; + goto fail; + } + nbc->disk_conf = new_disk_conf; - if (!disk_conf_from_tags(mdev, nlp->tag_list, &nbc->dc)) { + set_disk_conf_defaults(new_disk_conf); + err = disk_conf_from_attrs(new_disk_conf, info); + if (err) { retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); goto fail; } - if (nbc->dc.meta_dev_idx < DRBD_MD_INDEX_FLEX_INT) { + enforce_disk_conf_limits(new_disk_conf); + + new_plan = fifo_alloc((new_disk_conf->c_plan_ahead * 10 * SLEEP_TIME) / HZ); + if (!new_plan) { + retcode = ERR_NOMEM; + goto fail; + } + + if (new_disk_conf->meta_dev_idx < DRBD_MD_INDEX_FLEX_INT) { retcode = ERR_MD_IDX_INVALID; goto fail; } - if (get_net_conf(mdev)) { - int prot = mdev->net_conf->wire_protocol; - put_net_conf(mdev); - if (nbc->dc.fencing == FP_STONITH && prot == DRBD_PROT_A) { + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + if (nc) { + if (new_disk_conf->fencing == FP_STONITH && nc->wire_protocol == DRBD_PROT_A) { + rcu_read_unlock(); retcode = ERR_STONITH_AND_PROT_A; goto fail; } } + rcu_read_unlock(); - bdev = blkdev_get_by_path(nbc->dc.backing_dev, + bdev = blkdev_get_by_path(new_disk_conf->backing_dev, FMODE_READ | FMODE_WRITE | FMODE_EXCL, mdev); if (IS_ERR(bdev)) { - dev_err(DEV, "open(\"%s\") failed with %ld\n", nbc->dc.backing_dev, + dev_err(DEV, "open(\"%s\") failed with %ld\n", new_disk_conf->backing_dev, PTR_ERR(bdev)); retcode = ERR_OPEN_DISK; goto fail; @@ -1014,12 +1372,12 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp * should check it for you already; but if you don't, or * someone fooled it, we need to double check here) */ - bdev = blkdev_get_by_path(nbc->dc.meta_dev, + bdev = blkdev_get_by_path(new_disk_conf->meta_dev, FMODE_READ | FMODE_WRITE | FMODE_EXCL, - (nbc->dc.meta_dev_idx < 0) ? + (new_disk_conf->meta_dev_idx < 0) ? (void *)mdev : (void *)drbd_m_holder); if (IS_ERR(bdev)) { - dev_err(DEV, "open(\"%s\") failed with %ld\n", nbc->dc.meta_dev, + dev_err(DEV, "open(\"%s\") failed with %ld\n", new_disk_conf->meta_dev, PTR_ERR(bdev)); retcode = ERR_OPEN_MD_DISK; goto fail; @@ -1027,14 +1385,14 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp nbc->md_bdev = bdev; if ((nbc->backing_bdev == nbc->md_bdev) != - (nbc->dc.meta_dev_idx == DRBD_MD_INDEX_INTERNAL || - nbc->dc.meta_dev_idx == DRBD_MD_INDEX_FLEX_INT)) { + (new_disk_conf->meta_dev_idx == DRBD_MD_INDEX_INTERNAL || + new_disk_conf->meta_dev_idx == DRBD_MD_INDEX_FLEX_INT)) { retcode = ERR_MD_IDX_INVALID; goto fail; } resync_lru = lc_create("resync", drbd_bm_ext_cache, - 61, sizeof(struct bm_extent), + 1, 61, sizeof(struct bm_extent), offsetof(struct bm_extent, lce)); if (!resync_lru) { retcode = ERR_NOMEM; @@ -1044,21 +1402,21 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp /* RT - for drbd_get_max_capacity() DRBD_MD_INDEX_FLEX_INT */ drbd_md_set_sector_offsets(mdev, nbc); - if (drbd_get_max_capacity(nbc) < nbc->dc.disk_size) { + if (drbd_get_max_capacity(nbc) < new_disk_conf->disk_size) { dev_err(DEV, "max capacity %llu smaller than disk size %llu\n", (unsigned long long) drbd_get_max_capacity(nbc), - (unsigned long long) nbc->dc.disk_size); + (unsigned long long) new_disk_conf->disk_size); retcode = ERR_DISK_TOO_SMALL; goto fail; } - if (nbc->dc.meta_dev_idx < 0) { + if (new_disk_conf->meta_dev_idx < 0) { max_possible_sectors = DRBD_MAX_SECTORS_FLEX; /* at least one MB, otherwise it does not make sense */ min_md_device_sectors = (2<<10); } else { max_possible_sectors = DRBD_MAX_SECTORS; - min_md_device_sectors = MD_RESERVED_SECT * (nbc->dc.meta_dev_idx + 1); + min_md_device_sectors = MD_RESERVED_SECT * (new_disk_conf->meta_dev_idx + 1); } if (drbd_get_capacity(nbc->md_bdev) < min_md_device_sectors) { @@ -1083,14 +1441,20 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp dev_warn(DEV, "==> truncating very big lower level device " "to currently maximum possible %llu sectors <==\n", (unsigned long long) max_possible_sectors); - if (nbc->dc.meta_dev_idx >= 0) + if (new_disk_conf->meta_dev_idx >= 0) dev_warn(DEV, "==>> using internal or flexible " "meta data may help <<==\n"); } drbd_suspend_io(mdev); /* also wait for the last barrier ack. */ - wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt) || is_susp(mdev->state)); + /* FIXME see also https://daiquiri.linbit/cgi-bin/bugzilla/show_bug.cgi?id=171 + * We need a way to either ignore barrier acks for barriers sent before a device + * was attached, or a way to wait for all pending barrier acks to come in. + * As barriers are counted per resource, + * we'd need to suspend io on all devices of a resource. + */ + wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt) || drbd_suspended(mdev)); /* and for any other previously queued work */ drbd_flush_workqueue(mdev); @@ -1105,25 +1469,6 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp drbd_md_set_sector_offsets(mdev, nbc); - /* allocate a second IO page if logical_block_size != 512 */ - logical_block_size = bdev_logical_block_size(nbc->md_bdev); - if (logical_block_size == 0) - logical_block_size = MD_SECTOR_SIZE; - - if (logical_block_size != MD_SECTOR_SIZE) { - if (!mdev->md_io_tmpp) { - struct page *page = alloc_page(GFP_NOIO); - if (!page) - goto force_diskless_dec; - - dev_warn(DEV, "Meta data's bdev logical_block_size = %d != %d\n", - logical_block_size, MD_SECTOR_SIZE); - dev_warn(DEV, "Workaround engaged (has performance impact).\n"); - - mdev->md_io_tmpp = page; - } - } - if (!mdev->bitmap) { if (drbd_bm_init(mdev)) { retcode = ERR_NOMEM; @@ -1145,30 +1490,25 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp } /* Since we are diskless, fix the activity log first... */ - if (drbd_check_al_size(mdev)) { + if (drbd_check_al_size(mdev, new_disk_conf)) { retcode = ERR_NOMEM; goto force_diskless_dec; } /* Prevent shrinking of consistent devices ! */ if (drbd_md_test_flag(nbc, MDF_CONSISTENT) && - drbd_new_dev_size(mdev, nbc, 0) < nbc->md.la_size_sect) { + drbd_new_dev_size(mdev, nbc, nbc->disk_conf->disk_size, 0) < nbc->md.la_size_sect) { dev_warn(DEV, "refusing to truncate a consistent device\n"); retcode = ERR_DISK_TOO_SMALL; goto force_diskless_dec; } - if (!drbd_al_read_log(mdev, nbc)) { - retcode = ERR_IO_MD_DISK; - goto force_diskless_dec; - } - /* Reset the "barriers don't work" bits here, then force meta data to * be written, to ensure we determine if barriers are supported. */ - if (nbc->dc.no_md_flush) - set_bit(MD_NO_FUA, &mdev->flags); - else + if (new_disk_conf->md_flushes) clear_bit(MD_NO_FUA, &mdev->flags); + else + set_bit(MD_NO_FUA, &mdev->flags); /* Point of no return reached. * Devices and memory are no longer released by error cleanup below. @@ -1177,11 +1517,13 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp D_ASSERT(mdev->ldev == NULL); mdev->ldev = nbc; mdev->resync = resync_lru; + mdev->rs_plan_s = new_plan; nbc = NULL; resync_lru = NULL; + new_disk_conf = NULL; + new_plan = NULL; - mdev->write_ordering = WO_bdev_flush; - drbd_bump_write_ordering(mdev, WO_bdev_flush); + drbd_bump_write_ordering(mdev->tconn, WO_bdev_flush); if (drbd_md_test_flag(mdev->ldev, MDF_CRASHED_PRIMARY)) set_bit(CRASHED_PRIMARY, &mdev->flags); @@ -1189,10 +1531,8 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp clear_bit(CRASHED_PRIMARY, &mdev->flags); if (drbd_md_test_flag(mdev->ldev, MDF_PRIMARY_IND) && - !(mdev->state.role == R_PRIMARY && mdev->state.susp_nod)) { + !(mdev->state.role == R_PRIMARY && mdev->tconn->susp_nod)) set_bit(CRASHED_PRIMARY, &mdev->flags); - cp_discovered = 1; - } mdev->send_cnt = 0; mdev->recv_cnt = 0; @@ -1228,7 +1568,9 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp } else if (dd == grew) set_bit(RESYNC_AFTER_NEG, &mdev->flags); - if (drbd_md_test_flag(mdev->ldev, MDF_FULL_SYNC)) { + if (drbd_md_test_flag(mdev->ldev, MDF_FULL_SYNC) || + (test_bit(CRASHED_PRIMARY, &mdev->flags) && + drbd_md_test_flag(mdev->ldev, MDF_AL_DISABLED))) { dev_info(DEV, "Assuming that all blocks are out of sync " "(aka FullSync)\n"); if (drbd_bitmap_io(mdev, &drbd_bmio_set_n_write, @@ -1238,16 +1580,7 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp } } else { if (drbd_bitmap_io(mdev, &drbd_bm_read, - "read from attaching", BM_LOCKED_MASK) < 0) { - retcode = ERR_IO_MD_DISK; - goto force_diskless_dec; - } - } - - if (cp_discovered) { - drbd_al_apply_to_bm(mdev); - if (drbd_bitmap_io(mdev, &drbd_bm_write, - "crashed primary apply AL", BM_LOCKED_MASK)) { + "read from attaching", BM_LOCKED_MASK)) { retcode = ERR_IO_MD_DISK; goto force_diskless_dec; } @@ -1256,9 +1589,9 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp if (_drbd_bm_total_weight(mdev) == drbd_bm_bits(mdev)) drbd_suspend_al(mdev); /* IO is still suspended here... */ - spin_lock_irq(&mdev->req_lock); - os = mdev->state; - ns.i = os.i; + spin_lock_irq(&mdev->tconn->req_lock); + os = drbd_read_state(mdev); + ns = os; /* If MDF_CONSISTENT is not set go into inconsistent state, otherwise investigate MDF_WasUpToDate... If MDF_WAS_UP_TO_DATE is not set go into D_OUTDATED disk state, @@ -1276,8 +1609,9 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp if (drbd_md_test_flag(mdev->ldev, MDF_PEER_OUT_DATED)) ns.pdsk = D_OUTDATED; - if ( ns.disk == D_CONSISTENT && - (ns.pdsk == D_OUTDATED || mdev->ldev->dc.fencing == FP_DONT_CARE)) + rcu_read_lock(); + if (ns.disk == D_CONSISTENT && + (ns.pdsk == D_OUTDATED || rcu_dereference(mdev->ldev->disk_conf)->fencing == FP_DONT_CARE)) ns.disk = D_UP_TO_DATE; /* All tests on MDF_PRIMARY_IND, MDF_CONNECTED_IND, @@ -1285,6 +1619,13 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp this point, because drbd_request_state() modifies these flags. */ + if (rcu_dereference(mdev->ldev->disk_conf)->al_updates) + mdev->ldev->md.flags &= ~MDF_AL_DISABLED; + else + mdev->ldev->md.flags |= MDF_AL_DISABLED; + + rcu_read_unlock(); + /* In case we are C_CONNECTED postpone any decision on the new disk state after the negotiation phase. */ if (mdev->state.conn == C_CONNECTED) { @@ -1300,12 +1641,13 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp } rv = _drbd_set_state(mdev, ns, CS_VERBOSE, NULL); - ns = mdev->state; - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); if (rv < SS_SUCCESS) goto force_diskless_dec; + mod_timer(&mdev->request_timer, jiffies + HZ); + if (mdev->state.role == R_PRIMARY) mdev->ldev->md.uuid[UI_CURRENT] |= (u64)1; else @@ -1316,16 +1658,17 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE); put_ldev(mdev); - reply->ret_code = retcode; - drbd_reconfig_done(mdev); + conn_reconfig_done(mdev->tconn); + drbd_adm_finish(info, retcode); return 0; force_diskless_dec: put_ldev(mdev); force_diskless: - drbd_force_state(mdev, NS(disk, D_FAILED)); + drbd_force_state(mdev, NS(disk, D_DISKLESS)); drbd_md_sync(mdev); fail: + conn_reconfig_done(mdev->tconn); if (nbc) { if (nbc->backing_bdev) blkdev_put(nbc->backing_bdev, @@ -1335,34 +1678,24 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp FMODE_READ | FMODE_WRITE | FMODE_EXCL); kfree(nbc); } + kfree(new_disk_conf); lc_destroy(resync_lru); + kfree(new_plan); - reply->ret_code = retcode; - drbd_reconfig_done(mdev); + finish: + drbd_adm_finish(info, retcode); return 0; } -/* Detaching the disk is a process in multiple stages. First we need to lock - * out application IO, in-flight IO, IO stuck in drbd_al_begin_io. - * Then we transition to D_DISKLESS, and wait for put_ldev() to return all - * internal references as well. - * Only then we have finally detached. */ -static int drbd_nl_detach(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +static int adm_detach(struct drbd_conf *mdev, int force) { - enum drbd_ret_code retcode; + enum drbd_state_rv retcode; int ret; - struct detach dt = {}; - if (!detach_from_tags(mdev, nlp->tag_list, &dt)) { - reply->ret_code = ERR_MANDATORY_TAG; - goto out; - } - - if (dt.detach_force) { + if (force) { set_bit(FORCE_DETACH, &mdev->flags); drbd_force_state(mdev, NS(disk, D_FAILED)); - reply->ret_code = SS_SUCCESS; + retcode = SS_SUCCESS; goto out; } @@ -1374,326 +1707,529 @@ static int drbd_nl_detach(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, ret = wait_event_interruptible(mdev->misc_wait, mdev->state.disk != D_FAILED); drbd_resume_io(mdev); - if ((int)retcode == (int)SS_IS_DISKLESS) retcode = SS_NOTHING_TO_DO; if (ret) retcode = ERR_INTR; - reply->ret_code = retcode; out: - return 0; + return retcode; } -static int drbd_nl_net_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +/* Detaching the disk is a process in multiple stages. First we need to lock + * out application IO, in-flight IO, IO stuck in drbd_al_begin_io. + * Then we transition to D_DISKLESS, and wait for put_ldev() to return all + * internal references as well. + * Only then we have finally detached. */ +int drbd_adm_detach(struct sk_buff *skb, struct genl_info *info) { - int i, ns; enum drbd_ret_code retcode; - struct net_conf *new_conf = NULL; - struct crypto_hash *tfm = NULL; - struct crypto_hash *integrity_w_tfm = NULL; - struct crypto_hash *integrity_r_tfm = NULL; - struct hlist_head *new_tl_hash = NULL; - struct hlist_head *new_ee_hash = NULL; - struct drbd_conf *odev; - char hmac_name[CRYPTO_MAX_ALG_NAME]; - void *int_dig_out = NULL; - void *int_dig_in = NULL; - void *int_dig_vv = NULL; - struct sockaddr *new_my_addr, *new_peer_addr, *taken_addr; + struct detach_parms parms = { }; + int err; - drbd_reconfig_start(mdev); + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; - if (mdev->state.conn > C_STANDALONE) { - retcode = ERR_NET_CONFIGURED; - goto fail; + if (info->attrs[DRBD_NLA_DETACH_PARMS]) { + err = detach_parms_from_attrs(&parms, info); + if (err) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + goto out; + } } - /* allocation not in the IO path, cqueue thread context */ + retcode = adm_detach(adm_ctx.mdev, parms.force_detach); +out: + drbd_adm_finish(info, retcode); + return 0; +} + +static bool conn_resync_running(struct drbd_tconn *tconn) +{ + struct drbd_conf *mdev; + bool rv = false; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + if (mdev->state.conn == C_SYNC_SOURCE || + mdev->state.conn == C_SYNC_TARGET || + mdev->state.conn == C_PAUSED_SYNC_S || + mdev->state.conn == C_PAUSED_SYNC_T) { + rv = true; + break; + } + } + rcu_read_unlock(); + + return rv; +} + +static bool conn_ov_running(struct drbd_tconn *tconn) +{ + struct drbd_conf *mdev; + bool rv = false; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + if (mdev->state.conn == C_VERIFY_S || + mdev->state.conn == C_VERIFY_T) { + rv = true; + break; + } + } + rcu_read_unlock(); + + return rv; +} + +static enum drbd_ret_code +_check_net_options(struct drbd_tconn *tconn, struct net_conf *old_conf, struct net_conf *new_conf) +{ + struct drbd_conf *mdev; + int i; + + if (old_conf && tconn->cstate == C_WF_REPORT_PARAMS && tconn->agreed_pro_version < 100) { + if (new_conf->wire_protocol != old_conf->wire_protocol) + return ERR_NEED_APV_100; + + if (new_conf->two_primaries != old_conf->two_primaries) + return ERR_NEED_APV_100; + + if (strcmp(new_conf->integrity_alg, old_conf->integrity_alg)) + return ERR_NEED_APV_100; + } + + if (!new_conf->two_primaries && + conn_highest_role(tconn) == R_PRIMARY && + conn_highest_peer(tconn) == R_PRIMARY) + return ERR_NEED_ALLOW_TWO_PRI; + + if (new_conf->two_primaries && + (new_conf->wire_protocol != DRBD_PROT_C)) + return ERR_NOT_PROTO_C; + + idr_for_each_entry(&tconn->volumes, mdev, i) { + if (get_ldev(mdev)) { + enum drbd_fencing_p fp = rcu_dereference(mdev->ldev->disk_conf)->fencing; + put_ldev(mdev); + if (new_conf->wire_protocol == DRBD_PROT_A && fp == FP_STONITH) + return ERR_STONITH_AND_PROT_A; + } + if (mdev->state.role == R_PRIMARY && new_conf->discard_my_data) + return ERR_DISCARD_IMPOSSIBLE; + } + + if (new_conf->on_congestion != OC_BLOCK && new_conf->wire_protocol != DRBD_PROT_A) + return ERR_CONG_NOT_PROTO_A; + + return NO_ERROR; +} + +static enum drbd_ret_code +check_net_options(struct drbd_tconn *tconn, struct net_conf *new_conf) +{ + static enum drbd_ret_code rv; + struct drbd_conf *mdev; + int i; + + rcu_read_lock(); + rv = _check_net_options(tconn, rcu_dereference(tconn->net_conf), new_conf); + rcu_read_unlock(); + + /* tconn->volumes protected by genl_lock() here */ + idr_for_each_entry(&tconn->volumes, mdev, i) { + if (!mdev->bitmap) { + if(drbd_bm_init(mdev)) + return ERR_NOMEM; + } + } + + return rv; +} + +struct crypto { + struct crypto_hash *verify_tfm; + struct crypto_hash *csums_tfm; + struct crypto_hash *cram_hmac_tfm; + struct crypto_hash *integrity_tfm; +}; + +static int +alloc_hash(struct crypto_hash **tfm, char *tfm_name, int err_alg) +{ + if (!tfm_name[0]) + return NO_ERROR; + + *tfm = crypto_alloc_hash(tfm_name, 0, CRYPTO_ALG_ASYNC); + if (IS_ERR(*tfm)) { + *tfm = NULL; + return err_alg; + } + + return NO_ERROR; +} + +static enum drbd_ret_code +alloc_crypto(struct crypto *crypto, struct net_conf *new_conf) +{ + char hmac_name[CRYPTO_MAX_ALG_NAME]; + enum drbd_ret_code rv; + + rv = alloc_hash(&crypto->csums_tfm, new_conf->csums_alg, + ERR_CSUMS_ALG); + if (rv != NO_ERROR) + return rv; + rv = alloc_hash(&crypto->verify_tfm, new_conf->verify_alg, + ERR_VERIFY_ALG); + if (rv != NO_ERROR) + return rv; + rv = alloc_hash(&crypto->integrity_tfm, new_conf->integrity_alg, + ERR_INTEGRITY_ALG); + if (rv != NO_ERROR) + return rv; + if (new_conf->cram_hmac_alg[0] != 0) { + snprintf(hmac_name, CRYPTO_MAX_ALG_NAME, "hmac(%s)", + new_conf->cram_hmac_alg); + + rv = alloc_hash(&crypto->cram_hmac_tfm, hmac_name, + ERR_AUTH_ALG); + } + + return rv; +} + +static void free_crypto(struct crypto *crypto) +{ + crypto_free_hash(crypto->cram_hmac_tfm); + crypto_free_hash(crypto->integrity_tfm); + crypto_free_hash(crypto->csums_tfm); + crypto_free_hash(crypto->verify_tfm); +} + +int drbd_adm_net_opts(struct sk_buff *skb, struct genl_info *info) +{ + enum drbd_ret_code retcode; + struct drbd_tconn *tconn; + struct net_conf *old_conf, *new_conf = NULL; + int err; + int ovr; /* online verify running */ + int rsr; /* re-sync running */ + struct crypto crypto = { }; + + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONNECTION); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + + tconn = adm_ctx.tconn; + new_conf = kzalloc(sizeof(struct net_conf), GFP_KERNEL); if (!new_conf) { retcode = ERR_NOMEM; + goto out; + } + + conn_reconfig_start(tconn); + + mutex_lock(&tconn->data.mutex); + mutex_lock(&tconn->conf_update); + old_conf = tconn->net_conf; + + if (!old_conf) { + drbd_msg_put_info("net conf missing, try connect"); + retcode = ERR_INVALID_REQUEST; goto fail; } - new_conf->timeout = DRBD_TIMEOUT_DEF; - new_conf->try_connect_int = DRBD_CONNECT_INT_DEF; - new_conf->ping_int = DRBD_PING_INT_DEF; - new_conf->max_epoch_size = DRBD_MAX_EPOCH_SIZE_DEF; - new_conf->max_buffers = DRBD_MAX_BUFFERS_DEF; - new_conf->unplug_watermark = DRBD_UNPLUG_WATERMARK_DEF; - new_conf->sndbuf_size = DRBD_SNDBUF_SIZE_DEF; - new_conf->rcvbuf_size = DRBD_RCVBUF_SIZE_DEF; - new_conf->ko_count = DRBD_KO_COUNT_DEF; - new_conf->after_sb_0p = DRBD_AFTER_SB_0P_DEF; - new_conf->after_sb_1p = DRBD_AFTER_SB_1P_DEF; - new_conf->after_sb_2p = DRBD_AFTER_SB_2P_DEF; - new_conf->want_lose = 0; - new_conf->two_primaries = 0; - new_conf->wire_protocol = DRBD_PROT_C; - new_conf->ping_timeo = DRBD_PING_TIMEO_DEF; - new_conf->rr_conflict = DRBD_RR_CONFLICT_DEF; - new_conf->on_congestion = DRBD_ON_CONGESTION_DEF; - new_conf->cong_extents = DRBD_CONG_EXTENTS_DEF; - - if (!net_conf_from_tags(mdev, nlp->tag_list, new_conf)) { + *new_conf = *old_conf; + if (should_set_defaults(info)) + set_net_conf_defaults(new_conf); + + err = net_conf_from_attrs_for_change(new_conf, info); + if (err && err != -ENOMSG) { retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); goto fail; } - if (new_conf->two_primaries - && (new_conf->wire_protocol != DRBD_PROT_C)) { - retcode = ERR_NOT_PROTO_C; + retcode = check_net_options(tconn, new_conf); + if (retcode != NO_ERROR) goto fail; - } - if (get_ldev(mdev)) { - enum drbd_fencing_p fp = mdev->ldev->dc.fencing; - put_ldev(mdev); - if (new_conf->wire_protocol == DRBD_PROT_A && fp == FP_STONITH) { - retcode = ERR_STONITH_AND_PROT_A; - goto fail; - } + /* re-sync running */ + rsr = conn_resync_running(tconn); + if (rsr && strcmp(new_conf->csums_alg, old_conf->csums_alg)) { + retcode = ERR_CSUMS_RESYNC_RUNNING; + goto fail; } - if (new_conf->on_congestion != OC_BLOCK && new_conf->wire_protocol != DRBD_PROT_A) { - retcode = ERR_CONG_NOT_PROTO_A; + /* online verify running */ + ovr = conn_ov_running(tconn); + if (ovr && strcmp(new_conf->verify_alg, old_conf->verify_alg)) { + retcode = ERR_VERIFY_RUNNING; goto fail; } - if (mdev->state.role == R_PRIMARY && new_conf->want_lose) { - retcode = ERR_DISCARD; + retcode = alloc_crypto(&crypto, new_conf); + if (retcode != NO_ERROR) goto fail; - } - retcode = NO_ERROR; + rcu_assign_pointer(tconn->net_conf, new_conf); - new_my_addr = (struct sockaddr *)&new_conf->my_addr; - new_peer_addr = (struct sockaddr *)&new_conf->peer_addr; - for (i = 0; i < minor_count; i++) { - odev = minor_to_mdev(i); - if (!odev || odev == mdev) - continue; - if (get_net_conf(odev)) { - taken_addr = (struct sockaddr *)&odev->net_conf->my_addr; - if (new_conf->my_addr_len == odev->net_conf->my_addr_len && - !memcmp(new_my_addr, taken_addr, new_conf->my_addr_len)) - retcode = ERR_LOCAL_ADDR; - - taken_addr = (struct sockaddr *)&odev->net_conf->peer_addr; - if (new_conf->peer_addr_len == odev->net_conf->peer_addr_len && - !memcmp(new_peer_addr, taken_addr, new_conf->peer_addr_len)) - retcode = ERR_PEER_ADDR; - - put_net_conf(odev); - if (retcode != NO_ERROR) - goto fail; - } + if (!rsr) { + crypto_free_hash(tconn->csums_tfm); + tconn->csums_tfm = crypto.csums_tfm; + crypto.csums_tfm = NULL; + } + if (!ovr) { + crypto_free_hash(tconn->verify_tfm); + tconn->verify_tfm = crypto.verify_tfm; + crypto.verify_tfm = NULL; } - if (new_conf->cram_hmac_alg[0] != 0) { - snprintf(hmac_name, CRYPTO_MAX_ALG_NAME, "hmac(%s)", - new_conf->cram_hmac_alg); - tfm = crypto_alloc_hash(hmac_name, 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(tfm)) { - tfm = NULL; - retcode = ERR_AUTH_ALG; - goto fail; - } + crypto_free_hash(tconn->integrity_tfm); + tconn->integrity_tfm = crypto.integrity_tfm; + if (tconn->cstate >= C_WF_REPORT_PARAMS && tconn->agreed_pro_version >= 100) + /* Do this without trying to take tconn->data.mutex again. */ + __drbd_send_protocol(tconn, P_PROTOCOL_UPDATE); - if (!drbd_crypto_is_hash(crypto_hash_tfm(tfm))) { - retcode = ERR_AUTH_ALG_ND; - goto fail; - } - } + crypto_free_hash(tconn->cram_hmac_tfm); + tconn->cram_hmac_tfm = crypto.cram_hmac_tfm; - if (new_conf->integrity_alg[0]) { - integrity_w_tfm = crypto_alloc_hash(new_conf->integrity_alg, 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(integrity_w_tfm)) { - integrity_w_tfm = NULL; - retcode=ERR_INTEGRITY_ALG; - goto fail; - } + mutex_unlock(&tconn->conf_update); + mutex_unlock(&tconn->data.mutex); + synchronize_rcu(); + kfree(old_conf); - if (!drbd_crypto_is_hash(crypto_hash_tfm(integrity_w_tfm))) { - retcode=ERR_INTEGRITY_ALG_ND; - goto fail; - } + if (tconn->cstate >= C_WF_REPORT_PARAMS) + drbd_send_sync_param(minor_to_mdev(conn_lowest_minor(tconn))); - integrity_r_tfm = crypto_alloc_hash(new_conf->integrity_alg, 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(integrity_r_tfm)) { - integrity_r_tfm = NULL; - retcode=ERR_INTEGRITY_ALG; - goto fail; - } + goto done; + + fail: + mutex_unlock(&tconn->conf_update); + mutex_unlock(&tconn->data.mutex); + free_crypto(&crypto); + kfree(new_conf); + done: + conn_reconfig_done(tconn); + out: + drbd_adm_finish(info, retcode); + return 0; +} + +int drbd_adm_connect(struct sk_buff *skb, struct genl_info *info) +{ + struct drbd_conf *mdev; + struct net_conf *old_conf, *new_conf = NULL; + struct crypto crypto = { }; + struct drbd_tconn *tconn; + enum drbd_ret_code retcode; + int i; + int err; + + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_RESOURCE); + + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + if (!(adm_ctx.my_addr && adm_ctx.peer_addr)) { + drbd_msg_put_info("connection endpoint(s) missing"); + retcode = ERR_INVALID_REQUEST; + goto out; } - ns = new_conf->max_epoch_size/8; - if (mdev->tl_hash_s != ns) { - new_tl_hash = kzalloc(ns*sizeof(void *), GFP_KERNEL); - if (!new_tl_hash) { - retcode = ERR_NOMEM; - goto fail; + /* No need for _rcu here. All reconfiguration is + * strictly serialized on genl_lock(). We are protected against + * concurrent reconfiguration/addition/deletion */ + list_for_each_entry(tconn, &drbd_tconns, all_tconn) { + if (nla_len(adm_ctx.my_addr) == tconn->my_addr_len && + !memcmp(nla_data(adm_ctx.my_addr), &tconn->my_addr, tconn->my_addr_len)) { + retcode = ERR_LOCAL_ADDR; + goto out; } - } - ns = new_conf->max_buffers/8; - if (new_conf->two_primaries && (mdev->ee_hash_s != ns)) { - new_ee_hash = kzalloc(ns*sizeof(void *), GFP_KERNEL); - if (!new_ee_hash) { - retcode = ERR_NOMEM; - goto fail; + if (nla_len(adm_ctx.peer_addr) == tconn->peer_addr_len && + !memcmp(nla_data(adm_ctx.peer_addr), &tconn->peer_addr, tconn->peer_addr_len)) { + retcode = ERR_PEER_ADDR; + goto out; } } - ((char *)new_conf->shared_secret)[SHARED_SECRET_MAX-1] = 0; + tconn = adm_ctx.tconn; + conn_reconfig_start(tconn); - if (integrity_w_tfm) { - i = crypto_hash_digestsize(integrity_w_tfm); - int_dig_out = kmalloc(i, GFP_KERNEL); - if (!int_dig_out) { - retcode = ERR_NOMEM; - goto fail; - } - int_dig_in = kmalloc(i, GFP_KERNEL); - if (!int_dig_in) { - retcode = ERR_NOMEM; - goto fail; - } - int_dig_vv = kmalloc(i, GFP_KERNEL); - if (!int_dig_vv) { - retcode = ERR_NOMEM; - goto fail; - } + if (tconn->cstate > C_STANDALONE) { + retcode = ERR_NET_CONFIGURED; + goto fail; } - if (!mdev->bitmap) { - if(drbd_bm_init(mdev)) { - retcode = ERR_NOMEM; - goto fail; - } + /* allocation not in the IO path, drbdsetup / netlink process context */ + new_conf = kzalloc(sizeof(*new_conf), GFP_KERNEL); + if (!new_conf) { + retcode = ERR_NOMEM; + goto fail; } - drbd_flush_workqueue(mdev); - spin_lock_irq(&mdev->req_lock); - if (mdev->net_conf != NULL) { - retcode = ERR_NET_CONFIGURED; - spin_unlock_irq(&mdev->req_lock); + set_net_conf_defaults(new_conf); + + err = net_conf_from_attrs(new_conf, info); + if (err && err != -ENOMSG) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); goto fail; } - mdev->net_conf = new_conf; - mdev->send_cnt = 0; - mdev->recv_cnt = 0; + retcode = check_net_options(tconn, new_conf); + if (retcode != NO_ERROR) + goto fail; - if (new_tl_hash) { - kfree(mdev->tl_hash); - mdev->tl_hash_s = mdev->net_conf->max_epoch_size/8; - mdev->tl_hash = new_tl_hash; - } + retcode = alloc_crypto(&crypto, new_conf); + if (retcode != NO_ERROR) + goto fail; + + ((char *)new_conf->shared_secret)[SHARED_SECRET_MAX-1] = 0; + + conn_flush_workqueue(tconn); - if (new_ee_hash) { - kfree(mdev->ee_hash); - mdev->ee_hash_s = mdev->net_conf->max_buffers/8; - mdev->ee_hash = new_ee_hash; + mutex_lock(&tconn->conf_update); + old_conf = tconn->net_conf; + if (old_conf) { + retcode = ERR_NET_CONFIGURED; + mutex_unlock(&tconn->conf_update); + goto fail; } + rcu_assign_pointer(tconn->net_conf, new_conf); - crypto_free_hash(mdev->cram_hmac_tfm); - mdev->cram_hmac_tfm = tfm; + conn_free_crypto(tconn); + tconn->cram_hmac_tfm = crypto.cram_hmac_tfm; + tconn->integrity_tfm = crypto.integrity_tfm; + tconn->csums_tfm = crypto.csums_tfm; + tconn->verify_tfm = crypto.verify_tfm; - crypto_free_hash(mdev->integrity_w_tfm); - mdev->integrity_w_tfm = integrity_w_tfm; + tconn->my_addr_len = nla_len(adm_ctx.my_addr); + memcpy(&tconn->my_addr, nla_data(adm_ctx.my_addr), tconn->my_addr_len); + tconn->peer_addr_len = nla_len(adm_ctx.peer_addr); + memcpy(&tconn->peer_addr, nla_data(adm_ctx.peer_addr), tconn->peer_addr_len); - crypto_free_hash(mdev->integrity_r_tfm); - mdev->integrity_r_tfm = integrity_r_tfm; + mutex_unlock(&tconn->conf_update); - kfree(mdev->int_dig_out); - kfree(mdev->int_dig_in); - kfree(mdev->int_dig_vv); - mdev->int_dig_out=int_dig_out; - mdev->int_dig_in=int_dig_in; - mdev->int_dig_vv=int_dig_vv; - retcode = _drbd_set_state(_NS(mdev, conn, C_UNCONNECTED), CS_VERBOSE, NULL); - spin_unlock_irq(&mdev->req_lock); + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, i) { + mdev->send_cnt = 0; + mdev->recv_cnt = 0; + } + rcu_read_unlock(); - kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE); - reply->ret_code = retcode; - drbd_reconfig_done(mdev); + retcode = conn_request_state(tconn, NS(conn, C_UNCONNECTED), CS_VERBOSE); + + conn_reconfig_done(tconn); + drbd_adm_finish(info, retcode); return 0; fail: - kfree(int_dig_out); - kfree(int_dig_in); - kfree(int_dig_vv); - crypto_free_hash(tfm); - crypto_free_hash(integrity_w_tfm); - crypto_free_hash(integrity_r_tfm); - kfree(new_tl_hash); - kfree(new_ee_hash); + free_crypto(&crypto); kfree(new_conf); - reply->ret_code = retcode; - drbd_reconfig_done(mdev); + conn_reconfig_done(tconn); +out: + drbd_adm_finish(info, retcode); return 0; } -static int drbd_nl_disconnect(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +static enum drbd_state_rv conn_try_disconnect(struct drbd_tconn *tconn, bool force) { - int retcode; - struct disconnect dc; - - memset(&dc, 0, sizeof(struct disconnect)); - if (!disconnect_from_tags(mdev, nlp->tag_list, &dc)) { - retcode = ERR_MANDATORY_TAG; - goto fail; - } - - if (dc.force) { - spin_lock_irq(&mdev->req_lock); - if (mdev->state.conn >= C_WF_CONNECTION) - _drbd_set_state(_NS(mdev, conn, C_DISCONNECTING), CS_HARD, NULL); - spin_unlock_irq(&mdev->req_lock); - goto done; - } + enum drbd_state_rv rv; - retcode = _drbd_request_state(mdev, NS(conn, C_DISCONNECTING), CS_ORDERED); + rv = conn_request_state(tconn, NS(conn, C_DISCONNECTING), + force ? CS_HARD : 0); - if (retcode == SS_NOTHING_TO_DO) - goto done; - else if (retcode == SS_ALREADY_STANDALONE) - goto done; - else if (retcode == SS_PRIMARY_NOP) { - /* Our statche checking code wants to see the peer outdated. */ - retcode = drbd_request_state(mdev, NS2(conn, C_DISCONNECTING, - pdsk, D_OUTDATED)); - } else if (retcode == SS_CW_FAILED_BY_PEER) { + switch (rv) { + case SS_NOTHING_TO_DO: + break; + case SS_ALREADY_STANDALONE: + return SS_SUCCESS; + case SS_PRIMARY_NOP: + /* Our state checking code wants to see the peer outdated. */ + rv = conn_request_state(tconn, NS2(conn, C_DISCONNECTING, + pdsk, D_OUTDATED), CS_VERBOSE); + break; + case SS_CW_FAILED_BY_PEER: /* The peer probably wants to see us outdated. */ - retcode = _drbd_request_state(mdev, NS2(conn, C_DISCONNECTING, - disk, D_OUTDATED), - CS_ORDERED); - if (retcode == SS_IS_DISKLESS || retcode == SS_LOWER_THAN_OUTDATED) { - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - retcode = SS_SUCCESS; + rv = conn_request_state(tconn, NS2(conn, C_DISCONNECTING, + disk, D_OUTDATED), 0); + if (rv == SS_IS_DISKLESS || rv == SS_LOWER_THAN_OUTDATED) { + rv = conn_request_state(tconn, NS(conn, C_DISCONNECTING), + CS_HARD); } + break; + default:; + /* no special handling necessary */ + } + + if (rv >= SS_SUCCESS) { + enum drbd_state_rv rv2; + /* No one else can reconfigure the network while I am here. + * The state handling only uses drbd_thread_stop_nowait(), + * we want to really wait here until the receiver is no more. + */ + drbd_thread_stop(&adm_ctx.tconn->receiver); + + /* Race breaker. This additional state change request may be + * necessary, if this was a forced disconnect during a receiver + * restart. We may have "killed" the receiver thread just + * after drbdd_init() returned. Typically, we should be + * C_STANDALONE already, now, and this becomes a no-op. + */ + rv2 = conn_request_state(tconn, NS(conn, C_STANDALONE), + CS_VERBOSE | CS_HARD); + if (rv2 < SS_SUCCESS) + conn_err(tconn, + "unexpected rv2=%d in conn_try_disconnect()\n", + rv2); } + return rv; +} - if (retcode < SS_SUCCESS) - goto fail; +int drbd_adm_disconnect(struct sk_buff *skb, struct genl_info *info) +{ + struct disconnect_parms parms; + struct drbd_tconn *tconn; + enum drbd_state_rv rv; + enum drbd_ret_code retcode; + int err; - if (wait_event_interruptible(mdev->state_wait, - mdev->state.conn != C_DISCONNECTING)) { - /* Do not test for mdev->state.conn == C_STANDALONE, since - someone else might connect us in the mean time! */ - retcode = ERR_INTR; + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONNECTION); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) goto fail; + + tconn = adm_ctx.tconn; + memset(&parms, 0, sizeof(parms)); + if (info->attrs[DRBD_NLA_DISCONNECT_PARMS]) { + err = disconnect_parms_from_attrs(&parms, info); + if (err) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + goto fail; + } } - done: - retcode = NO_ERROR; + rv = conn_try_disconnect(tconn, parms.force_disconnect); + if (rv < SS_SUCCESS) + retcode = rv; /* FIXME: Type mismatch. */ + else + retcode = NO_ERROR; fail: - drbd_md_sync(mdev); - reply->ret_code = retcode; + drbd_adm_finish(info, retcode); return 0; } @@ -1705,7 +2241,7 @@ void resync_after_online_grow(struct drbd_conf *mdev) if (mdev->state.role != mdev->state.peer) iass = (mdev->state.role == R_PRIMARY); else - iass = test_bit(DISCARD_CONCURRENT, &mdev->flags); + iass = test_bit(RESOLVE_CONFLICTS, &mdev->tconn->flags); if (iass) drbd_start_resync(mdev, C_SYNC_SOURCE); @@ -1713,20 +2249,34 @@ void resync_after_online_grow(struct drbd_conf *mdev) _drbd_request_state(mdev, NS(conn, C_WF_SYNC_UUID), CS_VERBOSE + CS_SERIALIZE); } -static int drbd_nl_resize(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_resize(struct sk_buff *skb, struct genl_info *info) { - struct resize rs; - int retcode = NO_ERROR; + struct disk_conf *old_disk_conf, *new_disk_conf = NULL; + struct resize_parms rs; + struct drbd_conf *mdev; + enum drbd_ret_code retcode; enum determine_dev_size dd; enum dds_flags ddsf; + sector_t u_size; + int err; - memset(&rs, 0, sizeof(struct resize)); - if (!resize_from_tags(mdev, nlp->tag_list, &rs)) { - retcode = ERR_MANDATORY_TAG; + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) goto fail; + + memset(&rs, 0, sizeof(struct resize_parms)); + if (info->attrs[DRBD_NLA_RESIZE_PARMS]) { + err = resize_parms_from_attrs(&rs, info); + if (err) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + goto fail; + } } + mdev = adm_ctx.mdev; if (mdev->state.conn > C_CONNECTED) { retcode = ERR_RESIZE_RESYNC; goto fail; @@ -1743,15 +2293,36 @@ static int drbd_nl_resize(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, goto fail; } - if (rs.no_resync && mdev->agreed_pro_version < 93) { + if (rs.no_resync && mdev->tconn->agreed_pro_version < 93) { retcode = ERR_NEED_APV_93; goto fail_ldev; } + rcu_read_lock(); + u_size = rcu_dereference(mdev->ldev->disk_conf)->disk_size; + rcu_read_unlock(); + if (u_size != (sector_t)rs.resize_size) { + new_disk_conf = kmalloc(sizeof(struct disk_conf), GFP_KERNEL); + if (!new_disk_conf) { + retcode = ERR_NOMEM; + goto fail_ldev; + } + } + if (mdev->ldev->known_size != drbd_get_capacity(mdev->ldev->backing_bdev)) mdev->ldev->known_size = drbd_get_capacity(mdev->ldev->backing_bdev); - mdev->ldev->dc.disk_size = (sector_t)rs.resize_size; + if (new_disk_conf) { + mutex_lock(&mdev->tconn->conf_update); + old_disk_conf = mdev->ldev->disk_conf; + *new_disk_conf = *old_disk_conf; + new_disk_conf->disk_size = (sector_t)rs.resize_size; + rcu_assign_pointer(mdev->ldev->disk_conf, new_disk_conf); + mutex_unlock(&mdev->tconn->conf_update); + synchronize_rcu(); + kfree(old_disk_conf); + } + ddsf = (rs.resize_force ? DDSF_FORCED : 0) | (rs.no_resync ? DDSF_NO_RESYNC : 0); dd = drbd_determine_dev_size(mdev, ddsf); drbd_md_sync(mdev); @@ -1770,7 +2341,7 @@ static int drbd_nl_resize(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, } fail: - reply->ret_code = retcode; + drbd_adm_finish(info, retcode); return 0; fail_ldev: @@ -1778,204 +2349,55 @@ static int drbd_nl_resize(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, goto fail; } -static int drbd_nl_syncer_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_resource_opts(struct sk_buff *skb, struct genl_info *info) { - int retcode = NO_ERROR; + enum drbd_ret_code retcode; + struct drbd_tconn *tconn; + struct res_opts res_opts; int err; - int ovr; /* online verify running */ - int rsr; /* re-sync running */ - struct crypto_hash *verify_tfm = NULL; - struct crypto_hash *csums_tfm = NULL; - struct syncer_conf sc; - cpumask_var_t new_cpu_mask; - int *rs_plan_s = NULL; - int fifo_size; - - if (!zalloc_cpumask_var(&new_cpu_mask, GFP_KERNEL)) { - retcode = ERR_NOMEM; - goto fail; - } - if (nlp->flags & DRBD_NL_SET_DEFAULTS) { - memset(&sc, 0, sizeof(struct syncer_conf)); - sc.rate = DRBD_RATE_DEF; - sc.after = DRBD_AFTER_DEF; - sc.al_extents = DRBD_AL_EXTENTS_DEF; - sc.on_no_data = DRBD_ON_NO_DATA_DEF; - sc.c_plan_ahead = DRBD_C_PLAN_AHEAD_DEF; - sc.c_delay_target = DRBD_C_DELAY_TARGET_DEF; - sc.c_fill_target = DRBD_C_FILL_TARGET_DEF; - sc.c_max_rate = DRBD_C_MAX_RATE_DEF; - sc.c_min_rate = DRBD_C_MIN_RATE_DEF; - } else - memcpy(&sc, &mdev->sync_conf, sizeof(struct syncer_conf)); - - if (!syncer_conf_from_tags(mdev, nlp->tag_list, &sc)) { - retcode = ERR_MANDATORY_TAG; - goto fail; - } - - /* re-sync running */ - rsr = ( mdev->state.conn == C_SYNC_SOURCE || - mdev->state.conn == C_SYNC_TARGET || - mdev->state.conn == C_PAUSED_SYNC_S || - mdev->state.conn == C_PAUSED_SYNC_T ); - - if (rsr && strcmp(sc.csums_alg, mdev->sync_conf.csums_alg)) { - retcode = ERR_CSUMS_RESYNC_RUNNING; - goto fail; - } - - if (!rsr && sc.csums_alg[0]) { - csums_tfm = crypto_alloc_hash(sc.csums_alg, 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(csums_tfm)) { - csums_tfm = NULL; - retcode = ERR_CSUMS_ALG; - goto fail; - } - - if (!drbd_crypto_is_hash(crypto_hash_tfm(csums_tfm))) { - retcode = ERR_CSUMS_ALG_ND; - goto fail; - } - } - - /* online verify running */ - ovr = (mdev->state.conn == C_VERIFY_S || mdev->state.conn == C_VERIFY_T); - - if (ovr) { - if (strcmp(sc.verify_alg, mdev->sync_conf.verify_alg)) { - retcode = ERR_VERIFY_RUNNING; - goto fail; - } - } - - if (!ovr && sc.verify_alg[0]) { - verify_tfm = crypto_alloc_hash(sc.verify_alg, 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(verify_tfm)) { - verify_tfm = NULL; - retcode = ERR_VERIFY_ALG; - goto fail; - } - - if (!drbd_crypto_is_hash(crypto_hash_tfm(verify_tfm))) { - retcode = ERR_VERIFY_ALG_ND; - goto fail; - } - } - - /* silently ignore cpu mask on UP kernel */ - if (nr_cpu_ids > 1 && sc.cpu_mask[0] != 0) { - err = bitmap_parse(sc.cpu_mask, 32, - cpumask_bits(new_cpu_mask), nr_cpu_ids); - if (err) { - dev_warn(DEV, "bitmap_parse() failed with %d\n", err); - retcode = ERR_CPU_MASK_PARSE; - goto fail; - } - } - - ERR_IF (sc.rate < 1) sc.rate = 1; - ERR_IF (sc.al_extents < 7) sc.al_extents = 127; /* arbitrary minimum */ -#define AL_MAX ((MD_AL_MAX_SIZE-1) * AL_EXTENTS_PT) - if (sc.al_extents > AL_MAX) { - dev_err(DEV, "sc.al_extents > %d\n", AL_MAX); - sc.al_extents = AL_MAX; - } -#undef AL_MAX - - /* to avoid spurious errors when configuring minors before configuring - * the minors they depend on: if necessary, first create the minor we - * depend on */ - if (sc.after >= 0) - ensure_mdev(sc.after, 1); - - /* most sanity checks done, try to assign the new sync-after - * dependency. need to hold the global lock in there, - * to avoid a race in the dependency loop check. */ - retcode = drbd_alter_sa(mdev, sc.after); + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_RESOURCE); + if (!adm_ctx.reply_skb) + return retcode; if (retcode != NO_ERROR) goto fail; + tconn = adm_ctx.tconn; - fifo_size = (sc.c_plan_ahead * 10 * SLEEP_TIME) / HZ; - if (fifo_size != mdev->rs_plan_s.size && fifo_size > 0) { - rs_plan_s = kzalloc(sizeof(int) * fifo_size, GFP_KERNEL); - if (!rs_plan_s) { - dev_err(DEV, "kmalloc of fifo_buffer failed"); - retcode = ERR_NOMEM; - goto fail; - } - } + res_opts = tconn->res_opts; + if (should_set_defaults(info)) + set_res_opts_defaults(&res_opts); - /* ok, assign the rest of it as well. - * lock against receive_SyncParam() */ - spin_lock(&mdev->peer_seq_lock); - mdev->sync_conf = sc; - - if (!rsr) { - crypto_free_hash(mdev->csums_tfm); - mdev->csums_tfm = csums_tfm; - csums_tfm = NULL; - } - - if (!ovr) { - crypto_free_hash(mdev->verify_tfm); - mdev->verify_tfm = verify_tfm; - verify_tfm = NULL; - } - - if (fifo_size != mdev->rs_plan_s.size) { - kfree(mdev->rs_plan_s.values); - mdev->rs_plan_s.values = rs_plan_s; - mdev->rs_plan_s.size = fifo_size; - mdev->rs_planed = 0; - rs_plan_s = NULL; + err = res_opts_from_attrs(&res_opts, info); + if (err && err != -ENOMSG) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + goto fail; } - spin_unlock(&mdev->peer_seq_lock); - - if (get_ldev(mdev)) { - wait_event(mdev->al_wait, lc_try_lock(mdev->act_log)); - drbd_al_shrink(mdev); - err = drbd_check_al_size(mdev); - lc_unlock(mdev->act_log); - wake_up(&mdev->al_wait); - - put_ldev(mdev); - drbd_md_sync(mdev); - - if (err) { + err = set_resource_options(tconn, &res_opts); + if (err) { + retcode = ERR_INVALID_REQUEST; + if (err == -ENOMEM) retcode = ERR_NOMEM; - goto fail; - } } - if (mdev->state.conn >= C_CONNECTED) - drbd_send_sync_param(mdev, &sc); - - if (!cpumask_equal(mdev->cpu_mask, new_cpu_mask)) { - cpumask_copy(mdev->cpu_mask, new_cpu_mask); - drbd_calc_cpu_mask(mdev); - mdev->receiver.reset_cpu_mask = 1; - mdev->asender.reset_cpu_mask = 1; - mdev->worker.reset_cpu_mask = 1; - } - - kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE); fail: - kfree(rs_plan_s); - free_cpumask_var(new_cpu_mask); - crypto_free_hash(csums_tfm); - crypto_free_hash(verify_tfm); - reply->ret_code = retcode; + drbd_adm_finish(info, retcode); return 0; } -static int drbd_nl_invalidate(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_invalidate(struct sk_buff *skb, struct genl_info *info) { - int retcode; + struct drbd_conf *mdev; + int retcode; /* enum drbd_ret_code rsp. enum drbd_state_rv */ + + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + + mdev = adm_ctx.mdev; /* If there is still bitmap IO pending, probably because of a previous * resync just being finished, wait for it before requesting a new resync. @@ -1990,10 +2412,10 @@ static int drbd_nl_invalidate(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nl retcode = drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_T)); while (retcode == SS_NEED_CONNECTION) { - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); if (mdev->state.conn < C_CONNECTED) retcode = _drbd_set_state(_NS(mdev, disk, D_INCONSISTENT), CS_VERBOSE, NULL); - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); if (retcode != SS_NEED_CONNECTION) break; @@ -2002,7 +2424,25 @@ static int drbd_nl_invalidate(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nl } drbd_resume_io(mdev); - reply->ret_code = retcode; +out: + drbd_adm_finish(info, retcode); + return 0; +} + +static int drbd_adm_simple_request_state(struct sk_buff *skb, struct genl_info *info, + union drbd_state mask, union drbd_state val) +{ + enum drbd_ret_code retcode; + + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + + retcode = drbd_request_state(adm_ctx.mdev, mask, val); +out: + drbd_adm_finish(info, retcode); return 0; } @@ -2015,10 +2455,18 @@ static int drbd_bmio_set_susp_al(struct drbd_conf *mdev) return rv; } -static int drbd_nl_invalidate_peer(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_invalidate_peer(struct sk_buff *skb, struct genl_info *info) { - int retcode; + int retcode; /* drbd_ret_code, drbd_state_rv */ + struct drbd_conf *mdev; + + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + + mdev = adm_ctx.mdev; /* If there is still bitmap IO pending, probably because of a previous * resync just being finished, wait for it before requesting a new resync. @@ -2028,16 +2476,15 @@ static int drbd_nl_invalidate_peer(struct drbd_conf *mdev, struct drbd_nl_cfg_re drbd_flush_workqueue(mdev); retcode = _drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_S), CS_ORDERED); - if (retcode < SS_SUCCESS) { if (retcode == SS_NEED_CONNECTION && mdev->state.role == R_PRIMARY) { - /* The peer will get a resync upon connect anyways. Just make that - into a full resync. */ + /* The peer will get a resync upon connect anyways. + * Just make that into a full resync. */ retcode = drbd_request_state(mdev, NS(pdsk, D_INCONSISTENT)); if (retcode >= SS_SUCCESS) { if (drbd_bitmap_io(mdev, &drbd_bmio_set_susp_al, - "set_n_write from invalidate_peer", - BM_LOCKED_SET_ALLOWED)) + "set_n_write from invalidate_peer", + BM_LOCKED_SET_ALLOWED)) retcode = ERR_IO_MD_DISK; } } else @@ -2045,30 +2492,41 @@ static int drbd_nl_invalidate_peer(struct drbd_conf *mdev, struct drbd_nl_cfg_re } drbd_resume_io(mdev); - reply->ret_code = retcode; +out: + drbd_adm_finish(info, retcode); return 0; } -static int drbd_nl_pause_sync(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_pause_sync(struct sk_buff *skb, struct genl_info *info) { - int retcode = NO_ERROR; + enum drbd_ret_code retcode; - if (drbd_request_state(mdev, NS(user_isp, 1)) == SS_NOTHING_TO_DO) - retcode = ERR_PAUSE_IS_SET; + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; - reply->ret_code = retcode; + if (drbd_request_state(adm_ctx.mdev, NS(user_isp, 1)) == SS_NOTHING_TO_DO) + retcode = ERR_PAUSE_IS_SET; +out: + drbd_adm_finish(info, retcode); return 0; } -static int drbd_nl_resume_sync(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_resume_sync(struct sk_buff *skb, struct genl_info *info) { - int retcode = NO_ERROR; - union drbd_state s; + union drbd_dev_state s; + enum drbd_ret_code retcode; - if (drbd_request_state(mdev, NS(user_isp, 0)) == SS_NOTHING_TO_DO) { - s = mdev->state; + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + + if (drbd_request_state(adm_ctx.mdev, NS(user_isp, 0)) == SS_NOTHING_TO_DO) { + s = adm_ctx.mdev->state; if (s.conn == C_PAUSED_SYNC_S || s.conn == C_PAUSED_SYNC_T) { retcode = s.aftr_isp ? ERR_PIC_AFTER_DEP : s.peer_isp ? ERR_PIC_PEER_DEP : ERR_PAUSE_IS_CLEAR; @@ -2077,172 +2535,482 @@ static int drbd_nl_resume_sync(struct drbd_conf *mdev, struct drbd_nl_cfg_req *n } } - reply->ret_code = retcode; +out: + drbd_adm_finish(info, retcode); return 0; } -static int drbd_nl_suspend_io(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_suspend_io(struct sk_buff *skb, struct genl_info *info) { - reply->ret_code = drbd_request_state(mdev, NS(susp, 1)); - - return 0; + return drbd_adm_simple_request_state(skb, info, NS(susp, 1)); } -static int drbd_nl_resume_io(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_resume_io(struct sk_buff *skb, struct genl_info *info) { + struct drbd_conf *mdev; + int retcode; /* enum drbd_ret_code rsp. enum drbd_state_rv */ + + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + + mdev = adm_ctx.mdev; if (test_bit(NEW_CUR_UUID, &mdev->flags)) { drbd_uuid_new_current(mdev); clear_bit(NEW_CUR_UUID, &mdev->flags); } drbd_suspend_io(mdev); - reply->ret_code = drbd_request_state(mdev, NS3(susp, 0, susp_nod, 0, susp_fen, 0)); - if (reply->ret_code == SS_SUCCESS) { + retcode = drbd_request_state(mdev, NS3(susp, 0, susp_nod, 0, susp_fen, 0)); + if (retcode == SS_SUCCESS) { if (mdev->state.conn < C_CONNECTED) - tl_clear(mdev); + tl_clear(mdev->tconn); if (mdev->state.disk == D_DISKLESS || mdev->state.disk == D_FAILED) - tl_restart(mdev, fail_frozen_disk_io); + tl_restart(mdev->tconn, FAIL_FROZEN_DISK_IO); } drbd_resume_io(mdev); +out: + drbd_adm_finish(info, retcode); return 0; } -static int drbd_nl_outdate(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_outdate(struct sk_buff *skb, struct genl_info *info) { - reply->ret_code = drbd_request_state(mdev, NS(disk, D_OUTDATED)); - return 0; + return drbd_adm_simple_request_state(skb, info, NS(disk, D_OUTDATED)); } -static int drbd_nl_get_config(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int nla_put_drbd_cfg_context(struct sk_buff *skb, struct drbd_tconn *tconn, unsigned vnr) { - unsigned short *tl; + struct nlattr *nla; + nla = nla_nest_start(skb, DRBD_NLA_CFG_CONTEXT); + if (!nla) + goto nla_put_failure; + if (vnr != VOLUME_UNSPECIFIED && + nla_put_u32(skb, T_ctx_volume, vnr)) + goto nla_put_failure; + if (nla_put_string(skb, T_ctx_resource_name, tconn->name)) + goto nla_put_failure; + if (tconn->my_addr_len && + nla_put(skb, T_ctx_my_addr, tconn->my_addr_len, &tconn->my_addr)) + goto nla_put_failure; + if (tconn->peer_addr_len && + nla_put(skb, T_ctx_peer_addr, tconn->peer_addr_len, &tconn->peer_addr)) + goto nla_put_failure; + nla_nest_end(skb, nla); + return 0; - tl = reply->tag_list; +nla_put_failure: + if (nla) + nla_nest_cancel(skb, nla); + return -EMSGSIZE; +} - if (get_ldev(mdev)) { - tl = disk_conf_to_tags(mdev, &mdev->ldev->dc, tl); - put_ldev(mdev); - } +int nla_put_status_info(struct sk_buff *skb, struct drbd_conf *mdev, + const struct sib_info *sib) +{ + struct state_info *si = NULL; /* for sizeof(si->member); */ + struct net_conf *nc; + struct nlattr *nla; + int got_ldev; + int err = 0; + int exclude_sensitive; + + /* If sib != NULL, this is drbd_bcast_event, which anyone can listen + * to. So we better exclude_sensitive information. + * + * If sib == NULL, this is drbd_adm_get_status, executed synchronously + * in the context of the requesting user process. Exclude sensitive + * information, unless current has superuser. + * + * NOTE: for drbd_adm_get_status_all(), this is a netlink dump, and + * relies on the current implementation of netlink_dump(), which + * executes the dump callback successively from netlink_recvmsg(), + * always in the context of the receiving process */ + exclude_sensitive = sib || !capable(CAP_SYS_ADMIN); + + got_ldev = get_ldev(mdev); + + /* We need to add connection name and volume number information still. + * Minor number is in drbd_genlmsghdr. */ + if (nla_put_drbd_cfg_context(skb, mdev->tconn, mdev->vnr)) + goto nla_put_failure; + + if (res_opts_to_skb(skb, &mdev->tconn->res_opts, exclude_sensitive)) + goto nla_put_failure; + + rcu_read_lock(); + if (got_ldev) + if (disk_conf_to_skb(skb, rcu_dereference(mdev->ldev->disk_conf), exclude_sensitive)) + goto nla_put_failure; + + nc = rcu_dereference(mdev->tconn->net_conf); + if (nc) + err = net_conf_to_skb(skb, nc, exclude_sensitive); + rcu_read_unlock(); + if (err) + goto nla_put_failure; + + nla = nla_nest_start(skb, DRBD_NLA_STATE_INFO); + if (!nla) + goto nla_put_failure; + if (nla_put_u32(skb, T_sib_reason, sib ? sib->sib_reason : SIB_GET_STATUS_REPLY) || + nla_put_u32(skb, T_current_state, mdev->state.i) || + nla_put_u64(skb, T_ed_uuid, mdev->ed_uuid) || + nla_put_u64(skb, T_capacity, drbd_get_capacity(mdev->this_bdev)) || + nla_put_u64(skb, T_send_cnt, mdev->send_cnt) || + nla_put_u64(skb, T_recv_cnt, mdev->recv_cnt) || + nla_put_u64(skb, T_read_cnt, mdev->read_cnt) || + nla_put_u64(skb, T_writ_cnt, mdev->writ_cnt) || + nla_put_u64(skb, T_al_writ_cnt, mdev->al_writ_cnt) || + nla_put_u64(skb, T_bm_writ_cnt, mdev->bm_writ_cnt) || + nla_put_u32(skb, T_ap_bio_cnt, atomic_read(&mdev->ap_bio_cnt)) || + nla_put_u32(skb, T_ap_pending_cnt, atomic_read(&mdev->ap_pending_cnt)) || + nla_put_u32(skb, T_rs_pending_cnt, atomic_read(&mdev->rs_pending_cnt))) + goto nla_put_failure; + + if (got_ldev) { + int err; - if (get_net_conf(mdev)) { - tl = net_conf_to_tags(mdev, mdev->net_conf, tl); - put_net_conf(mdev); + spin_lock_irq(&mdev->ldev->md.uuid_lock); + err = nla_put(skb, T_uuids, sizeof(si->uuids), mdev->ldev->md.uuid); + spin_unlock_irq(&mdev->ldev->md.uuid_lock); + + if (err) + goto nla_put_failure; + + if (nla_put_u32(skb, T_disk_flags, mdev->ldev->md.flags) || + nla_put_u64(skb, T_bits_total, drbd_bm_bits(mdev)) || + nla_put_u64(skb, T_bits_oos, drbd_bm_total_weight(mdev))) + goto nla_put_failure; + if (C_SYNC_SOURCE <= mdev->state.conn && + C_PAUSED_SYNC_T >= mdev->state.conn) { + if (nla_put_u64(skb, T_bits_rs_total, mdev->rs_total) || + nla_put_u64(skb, T_bits_rs_failed, mdev->rs_failed)) + goto nla_put_failure; + } } - tl = syncer_conf_to_tags(mdev, &mdev->sync_conf, tl); - put_unaligned(TT_END, tl++); /* Close the tag list */ + if (sib) { + switch(sib->sib_reason) { + case SIB_SYNC_PROGRESS: + case SIB_GET_STATUS_REPLY: + break; + case SIB_STATE_CHANGE: + if (nla_put_u32(skb, T_prev_state, sib->os.i) || + nla_put_u32(skb, T_new_state, sib->ns.i)) + goto nla_put_failure; + break; + case SIB_HELPER_POST: + if (nla_put_u32(skb, T_helper_exit_code, + sib->helper_exit_code)) + goto nla_put_failure; + /* fall through */ + case SIB_HELPER_PRE: + if (nla_put_string(skb, T_helper, sib->helper_name)) + goto nla_put_failure; + break; + } + } + nla_nest_end(skb, nla); - return (int)((char *)tl - (char *)reply->tag_list); + if (0) +nla_put_failure: + err = -EMSGSIZE; + if (got_ldev) + put_ldev(mdev); + return err; } -static int drbd_nl_get_state(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_get_status(struct sk_buff *skb, struct genl_info *info) { - unsigned short *tl = reply->tag_list; - union drbd_state s = mdev->state; - unsigned long rs_left; - unsigned int res; + enum drbd_ret_code retcode; + int err; - tl = get_state_to_tags(mdev, (struct get_state *)&s, tl); + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; - /* no local ref, no bitmap, no syncer progress. */ - if (s.conn >= C_SYNC_SOURCE && s.conn <= C_PAUSED_SYNC_T) { - if (get_ldev(mdev)) { - drbd_get_syncer_progress(mdev, &rs_left, &res); - tl = tl_add_int(tl, T_sync_progress, &res); - put_ldev(mdev); - } + err = nla_put_status_info(adm_ctx.reply_skb, adm_ctx.mdev, NULL); + if (err) { + nlmsg_free(adm_ctx.reply_skb); + return err; } - put_unaligned(TT_END, tl++); /* Close the tag list */ - - return (int)((char *)tl - (char *)reply->tag_list); +out: + drbd_adm_finish(info, retcode); + return 0; } -static int drbd_nl_get_uuids(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int get_one_status(struct sk_buff *skb, struct netlink_callback *cb) { - unsigned short *tl; - - tl = reply->tag_list; + struct drbd_conf *mdev; + struct drbd_genlmsghdr *dh; + struct drbd_tconn *pos = (struct drbd_tconn*)cb->args[0]; + struct drbd_tconn *tconn = NULL; + struct drbd_tconn *tmp; + unsigned volume = cb->args[1]; + + /* Open coded, deferred, iteration: + * list_for_each_entry_safe(tconn, tmp, &drbd_tconns, all_tconn) { + * idr_for_each_entry(&tconn->volumes, mdev, i) { + * ... + * } + * } + * where tconn is cb->args[0]; + * and i is cb->args[1]; + * + * cb->args[2] indicates if we shall loop over all resources, + * or just dump all volumes of a single resource. + * + * This may miss entries inserted after this dump started, + * or entries deleted before they are reached. + * + * We need to make sure the mdev won't disappear while + * we are looking at it, and revalidate our iterators + * on each iteration. + */ - if (get_ldev(mdev)) { - tl = tl_add_blob(tl, T_uuids, mdev->ldev->md.uuid, UI_SIZE*sizeof(u64)); - tl = tl_add_int(tl, T_uuids_flags, &mdev->ldev->md.flags); - put_ldev(mdev); + /* synchronize with conn_create()/conn_destroy() */ + rcu_read_lock(); + /* revalidate iterator position */ + list_for_each_entry_rcu(tmp, &drbd_tconns, all_tconn) { + if (pos == NULL) { + /* first iteration */ + pos = tmp; + tconn = pos; + break; + } + if (tmp == pos) { + tconn = pos; + break; + } } - put_unaligned(TT_END, tl++); /* Close the tag list */ + if (tconn) { +next_tconn: + mdev = idr_get_next(&tconn->volumes, &volume); + if (!mdev) { + /* No more volumes to dump on this tconn. + * Advance tconn iterator. */ + pos = list_entry_rcu(tconn->all_tconn.next, + struct drbd_tconn, all_tconn); + /* Did we dump any volume on this tconn yet? */ + if (volume != 0) { + /* If we reached the end of the list, + * or only a single resource dump was requested, + * we are done. */ + if (&pos->all_tconn == &drbd_tconns || cb->args[2]) + goto out; + volume = 0; + tconn = pos; + goto next_tconn; + } + } + + dh = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, &drbd_genl_family, + NLM_F_MULTI, DRBD_ADM_GET_STATUS); + if (!dh) + goto out; + + if (!mdev) { + /* This is a tconn without a single volume. + * Suprisingly enough, it may have a network + * configuration. */ + struct net_conf *nc; + dh->minor = -1U; + dh->ret_code = NO_ERROR; + if (nla_put_drbd_cfg_context(skb, tconn, VOLUME_UNSPECIFIED)) + goto cancel; + nc = rcu_dereference(tconn->net_conf); + if (nc && net_conf_to_skb(skb, nc, 1) != 0) + goto cancel; + goto done; + } + + D_ASSERT(mdev->vnr == volume); + D_ASSERT(mdev->tconn == tconn); + + dh->minor = mdev_to_minor(mdev); + dh->ret_code = NO_ERROR; - return (int)((char *)tl - (char *)reply->tag_list); + if (nla_put_status_info(skb, mdev, NULL)) { +cancel: + genlmsg_cancel(skb, dh); + goto out; + } +done: + genlmsg_end(skb, dh); + } + +out: + rcu_read_unlock(); + /* where to start the next iteration */ + cb->args[0] = (long)pos; + cb->args[1] = (pos == tconn) ? volume + 1 : 0; + + /* No more tconns/volumes/minors found results in an empty skb. + * Which will terminate the dump. */ + return skb->len; } -/** - * drbd_nl_get_timeout_flag() - Used by drbdsetup to find out which timeout value to use - * @mdev: DRBD device. - * @nlp: Netlink/connector packet from drbdsetup - * @reply: Reply packet for drbdsetup +/* + * Request status of all resources, or of all volumes within a single resource. + * + * This is a dump, as the answer may not fit in a single reply skb otherwise. + * Which means we cannot use the family->attrbuf or other such members, because + * dump is NOT protected by the genl_lock(). During dump, we only have access + * to the incoming skb, and need to opencode "parsing" of the nlattr payload. + * + * Once things are setup properly, we call into get_one_status(). */ -static int drbd_nl_get_timeout_flag(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_get_status_all(struct sk_buff *skb, struct netlink_callback *cb) { - unsigned short *tl; - char rv; + const unsigned hdrlen = GENL_HDRLEN + GENL_MAGIC_FAMILY_HDRSZ; + struct nlattr *nla; + const char *resource_name; + struct drbd_tconn *tconn; + int maxtype; + + /* Is this a followup call? */ + if (cb->args[0]) { + /* ... of a single resource dump, + * and the resource iterator has been advanced already? */ + if (cb->args[2] && cb->args[2] != cb->args[0]) + return 0; /* DONE. */ + goto dump; + } + + /* First call (from netlink_dump_start). We need to figure out + * which resource(s) the user wants us to dump. */ + nla = nla_find(nlmsg_attrdata(cb->nlh, hdrlen), + nlmsg_attrlen(cb->nlh, hdrlen), + DRBD_NLA_CFG_CONTEXT); + + /* No explicit context given. Dump all. */ + if (!nla) + goto dump; + maxtype = ARRAY_SIZE(drbd_cfg_context_nl_policy) - 1; + nla = drbd_nla_find_nested(maxtype, nla, __nla_type(T_ctx_resource_name)); + if (IS_ERR(nla)) + return PTR_ERR(nla); + /* context given, but no name present? */ + if (!nla) + return -EINVAL; + resource_name = nla_data(nla); + tconn = conn_get_by_name(resource_name); + + if (!tconn) + return -ENODEV; + + kref_put(&tconn->kref, &conn_destroy); /* get_one_status() (re)validates tconn by itself */ + + /* prime iterators, and set "filter" mode mark: + * only dump this tconn. */ + cb->args[0] = (long)tconn; + /* cb->args[1] = 0; passed in this way. */ + cb->args[2] = (long)tconn; + +dump: + return get_one_status(skb, cb); +} - tl = reply->tag_list; +int drbd_adm_get_timeout_type(struct sk_buff *skb, struct genl_info *info) +{ + enum drbd_ret_code retcode; + struct timeout_parms tp; + int err; - rv = mdev->state.pdsk == D_OUTDATED ? UT_PEER_OUTDATED : - test_bit(USE_DEGR_WFC_T, &mdev->flags) ? UT_DEGRADED : UT_DEFAULT; + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; - tl = tl_add_blob(tl, T_use_degraded, &rv, sizeof(rv)); - put_unaligned(TT_END, tl++); /* Close the tag list */ + tp.timeout_type = + adm_ctx.mdev->state.pdsk == D_OUTDATED ? UT_PEER_OUTDATED : + test_bit(USE_DEGR_WFC_T, &adm_ctx.mdev->flags) ? UT_DEGRADED : + UT_DEFAULT; - return (int)((char *)tl - (char *)reply->tag_list); + err = timeout_parms_to_priv_skb(adm_ctx.reply_skb, &tp); + if (err) { + nlmsg_free(adm_ctx.reply_skb); + return err; + } +out: + drbd_adm_finish(info, retcode); + return 0; } -static int drbd_nl_start_ov(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_start_ov(struct sk_buff *skb, struct genl_info *info) { - /* default to resume from last known position, if possible */ - struct start_ov args = - { .start_sector = mdev->ov_start_sector }; + struct drbd_conf *mdev; + enum drbd_ret_code retcode; + struct start_ov_parms parms; - if (!start_ov_from_tags(mdev, nlp->tag_list, &args)) { - reply->ret_code = ERR_MANDATORY_TAG; - return 0; + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; + + mdev = adm_ctx.mdev; + + /* resume from last known position, if possible */ + parms.ov_start_sector = mdev->ov_start_sector; + parms.ov_stop_sector = ULLONG_MAX; + if (info->attrs[DRBD_NLA_START_OV_PARMS]) { + int err = start_ov_parms_from_attrs(&parms, info); + if (err) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + goto out; + } } + /* w_make_ov_request expects position to be aligned */ + mdev->ov_start_sector = parms.ov_start_sector & ~(BM_SECT_PER_BIT-1); + mdev->ov_stop_sector = parms.ov_stop_sector; /* If there is still bitmap IO pending, e.g. previous resync or verify * just being finished, wait for it before requesting a new resync. */ drbd_suspend_io(mdev); wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags)); - - /* w_make_ov_request expects position to be aligned */ - mdev->ov_start_sector = args.start_sector & ~BM_SECT_PER_BIT; - reply->ret_code = drbd_request_state(mdev,NS(conn,C_VERIFY_S)); + retcode = drbd_request_state(mdev,NS(conn,C_VERIFY_S)); drbd_resume_io(mdev); +out: + drbd_adm_finish(info, retcode); return 0; } -static int drbd_nl_new_c_uuid(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, - struct drbd_nl_cfg_reply *reply) +int drbd_adm_new_c_uuid(struct sk_buff *skb, struct genl_info *info) { - int retcode = NO_ERROR; + struct drbd_conf *mdev; + enum drbd_ret_code retcode; int skip_initial_sync = 0; int err; + struct new_c_uuid_parms args; - struct new_c_uuid args; + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out_nolock; - memset(&args, 0, sizeof(struct new_c_uuid)); - if (!new_c_uuid_from_tags(mdev, nlp->tag_list, &args)) { - reply->ret_code = ERR_MANDATORY_TAG; - return 0; + mdev = adm_ctx.mdev; + memset(&args, 0, sizeof(args)); + if (info->attrs[DRBD_NLA_NEW_C_UUID_PARMS]) { + err = new_c_uuid_parms_from_attrs(&args, info); + if (err) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + goto out_nolock; + } } - mutex_lock(&mdev->state_mutex); /* Protects us against serialized state changes. */ + mutex_lock(mdev->state_mutex); /* Protects us against serialized state changes. */ if (!get_ldev(mdev)) { retcode = ERR_NO_DISK; @@ -2250,7 +3018,7 @@ static int drbd_nl_new_c_uuid(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nl } /* this is "skip initial sync", assume to be clean */ - if (mdev->state.conn == C_CONNECTED && mdev->agreed_pro_version >= 90 && + if (mdev->state.conn == C_CONNECTED && mdev->tconn->agreed_pro_version >= 90 && mdev->ldev->md.uuid[UI_CURRENT] == UUID_JUST_CREATED && args.clear_bm) { dev_info(DEV, "Preparing to skip initial sync\n"); skip_initial_sync = 1; @@ -2273,10 +3041,10 @@ static int drbd_nl_new_c_uuid(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nl drbd_send_uuids_skip_initial_sync(mdev); _drbd_uuid_set(mdev, UI_BITMAP, 0); drbd_print_uuids(mdev, "cleared bitmap UUID"); - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); _drbd_set_state(_NS2(mdev, disk, D_UP_TO_DATE, pdsk, D_UP_TO_DATE), CS_VERBOSE, NULL); - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); } } @@ -2284,416 +3052,284 @@ static int drbd_nl_new_c_uuid(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nl out_dec: put_ldev(mdev); out: - mutex_unlock(&mdev->state_mutex); - - reply->ret_code = retcode; + mutex_unlock(mdev->state_mutex); +out_nolock: + drbd_adm_finish(info, retcode); return 0; } -struct cn_handler_struct { - int (*function)(struct drbd_conf *, - struct drbd_nl_cfg_req *, - struct drbd_nl_cfg_reply *); - int reply_body_size; -}; - -static struct cn_handler_struct cnd_table[] = { - [ P_primary ] = { &drbd_nl_primary, 0 }, - [ P_secondary ] = { &drbd_nl_secondary, 0 }, - [ P_disk_conf ] = { &drbd_nl_disk_conf, 0 }, - [ P_detach ] = { &drbd_nl_detach, 0 }, - [ P_net_conf ] = { &drbd_nl_net_conf, 0 }, - [ P_disconnect ] = { &drbd_nl_disconnect, 0 }, - [ P_resize ] = { &drbd_nl_resize, 0 }, - [ P_syncer_conf ] = { &drbd_nl_syncer_conf, 0 }, - [ P_invalidate ] = { &drbd_nl_invalidate, 0 }, - [ P_invalidate_peer ] = { &drbd_nl_invalidate_peer, 0 }, - [ P_pause_sync ] = { &drbd_nl_pause_sync, 0 }, - [ P_resume_sync ] = { &drbd_nl_resume_sync, 0 }, - [ P_suspend_io ] = { &drbd_nl_suspend_io, 0 }, - [ P_resume_io ] = { &drbd_nl_resume_io, 0 }, - [ P_outdate ] = { &drbd_nl_outdate, 0 }, - [ P_get_config ] = { &drbd_nl_get_config, - sizeof(struct syncer_conf_tag_len_struct) + - sizeof(struct disk_conf_tag_len_struct) + - sizeof(struct net_conf_tag_len_struct) }, - [ P_get_state ] = { &drbd_nl_get_state, - sizeof(struct get_state_tag_len_struct) + - sizeof(struct sync_progress_tag_len_struct) }, - [ P_get_uuids ] = { &drbd_nl_get_uuids, - sizeof(struct get_uuids_tag_len_struct) }, - [ P_get_timeout_flag ] = { &drbd_nl_get_timeout_flag, - sizeof(struct get_timeout_flag_tag_len_struct)}, - [ P_start_ov ] = { &drbd_nl_start_ov, 0 }, - [ P_new_c_uuid ] = { &drbd_nl_new_c_uuid, 0 }, -}; - -static void drbd_connector_callback(struct cn_msg *req, struct netlink_skb_parms *nsp) +static enum drbd_ret_code +drbd_check_resource_name(const char *name) { - struct drbd_nl_cfg_req *nlp = (struct drbd_nl_cfg_req *)req->data; - struct cn_handler_struct *cm; - struct cn_msg *cn_reply; - struct drbd_nl_cfg_reply *reply; - struct drbd_conf *mdev; - int retcode, rr; - int reply_size = sizeof(struct cn_msg) - + sizeof(struct drbd_nl_cfg_reply) - + sizeof(short int); - - if (!try_module_get(THIS_MODULE)) { - printk(KERN_ERR "drbd: try_module_get() failed!\n"); - return; + if (!name || !name[0]) { + drbd_msg_put_info("resource name missing"); + return ERR_MANDATORY_TAG; } - - if (!capable(CAP_SYS_ADMIN)) { - retcode = ERR_PERM; - goto fail; - } - - mdev = ensure_mdev(nlp->drbd_minor, - (nlp->flags & DRBD_NL_CREATE_DEVICE)); - if (!mdev) { - retcode = ERR_MINOR_INVALID; - goto fail; + /* if we want to use these in sysfs/configfs/debugfs some day, + * we must not allow slashes */ + if (strchr(name, '/')) { + drbd_msg_put_info("invalid resource name"); + return ERR_INVALID_REQUEST; } + return NO_ERROR; +} - if (nlp->packet_type >= P_nl_after_last_packet || - nlp->packet_type == P_return_code_only) { - retcode = ERR_PACKET_NR; - goto fail; - } +int drbd_adm_new_resource(struct sk_buff *skb, struct genl_info *info) +{ + enum drbd_ret_code retcode; + struct res_opts res_opts; + int err; - cm = cnd_table + nlp->packet_type; + retcode = drbd_adm_prepare(skb, info, 0); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; - /* This may happen if packet number is 0: */ - if (cm->function == NULL) { - retcode = ERR_PACKET_NR; - goto fail; + set_res_opts_defaults(&res_opts); + err = res_opts_from_attrs(&res_opts, info); + if (err && err != -ENOMSG) { + retcode = ERR_MANDATORY_TAG; + drbd_msg_put_info(from_attrs_err_to_txt(err)); + goto out; } - reply_size += cm->reply_body_size; + retcode = drbd_check_resource_name(adm_ctx.resource_name); + if (retcode != NO_ERROR) + goto out; - /* allocation not in the IO path, cqueue thread context */ - cn_reply = kzalloc(reply_size, GFP_KERNEL); - if (!cn_reply) { - retcode = ERR_NOMEM; - goto fail; + if (adm_ctx.tconn) { + if (info->nlhdr->nlmsg_flags & NLM_F_EXCL) { + retcode = ERR_INVALID_REQUEST; + drbd_msg_put_info("resource exists"); + } + /* else: still NO_ERROR */ + goto out; } - reply = (struct drbd_nl_cfg_reply *) cn_reply->data; - - reply->packet_type = - cm->reply_body_size ? nlp->packet_type : P_return_code_only; - reply->minor = nlp->drbd_minor; - reply->ret_code = NO_ERROR; /* Might by modified by cm->function. */ - /* reply->tag_list; might be modified by cm->function. */ - - rr = cm->function(mdev, nlp, reply); - - cn_reply->id = req->id; - cn_reply->seq = req->seq; - cn_reply->ack = req->ack + 1; - cn_reply->len = sizeof(struct drbd_nl_cfg_reply) + rr; - cn_reply->flags = 0; - - rr = cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_KERNEL); - if (rr && rr != -ESRCH) - printk(KERN_INFO "drbd: cn_netlink_send()=%d\n", rr); - kfree(cn_reply); - module_put(THIS_MODULE); - return; - fail: - drbd_nl_send_reply(req, retcode); - module_put(THIS_MODULE); + if (!conn_create(adm_ctx.resource_name, &res_opts)) + retcode = ERR_NOMEM; +out: + drbd_adm_finish(info, retcode); + return 0; } -static atomic_t drbd_nl_seq = ATOMIC_INIT(2); /* two. */ - -static unsigned short * -__tl_add_blob(unsigned short *tl, enum drbd_tags tag, const void *data, - unsigned short len, int nul_terminated) +int drbd_adm_add_minor(struct sk_buff *skb, struct genl_info *info) { - unsigned short l = tag_descriptions[tag_number(tag)].max_len; - len = (len < l) ? len : l; - put_unaligned(tag, tl++); - put_unaligned(len, tl++); - memcpy(tl, data, len); - tl = (unsigned short*)((char*)tl + len); - if (nul_terminated) - *((char*)tl - 1) = 0; - return tl; -} + struct drbd_genlmsghdr *dh = info->userhdr; + enum drbd_ret_code retcode; -static unsigned short * -tl_add_blob(unsigned short *tl, enum drbd_tags tag, const void *data, int len) -{ - return __tl_add_blob(tl, tag, data, len, 0); -} + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_RESOURCE); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; -static unsigned short * -tl_add_str(unsigned short *tl, enum drbd_tags tag, const char *str) -{ - return __tl_add_blob(tl, tag, str, strlen(str)+1, 0); -} + if (dh->minor > MINORMASK) { + drbd_msg_put_info("requested minor out of range"); + retcode = ERR_INVALID_REQUEST; + goto out; + } + if (adm_ctx.volume > DRBD_VOLUME_MAX) { + drbd_msg_put_info("requested volume id out of range"); + retcode = ERR_INVALID_REQUEST; + goto out; + } -static unsigned short * -tl_add_int(unsigned short *tl, enum drbd_tags tag, const void *val) -{ - put_unaligned(tag, tl++); - switch(tag_type(tag)) { - case TT_INTEGER: - put_unaligned(sizeof(int), tl++); - put_unaligned(*(int *)val, (int *)tl); - tl = (unsigned short*)((char*)tl+sizeof(int)); - break; - case TT_INT64: - put_unaligned(sizeof(u64), tl++); - put_unaligned(*(u64 *)val, (u64 *)tl); - tl = (unsigned short*)((char*)tl+sizeof(u64)); - break; - default: - /* someone did something stupid. */ - ; + /* drbd_adm_prepare made sure already + * that mdev->tconn and mdev->vnr match the request. */ + if (adm_ctx.mdev) { + if (info->nlhdr->nlmsg_flags & NLM_F_EXCL) + retcode = ERR_MINOR_EXISTS; + /* else: still NO_ERROR */ + goto out; } - return tl; + + retcode = conn_new_minor(adm_ctx.tconn, dh->minor, adm_ctx.volume); +out: + drbd_adm_finish(info, retcode); + return 0; } -void drbd_bcast_state(struct drbd_conf *mdev, union drbd_state state) +static enum drbd_ret_code adm_delete_minor(struct drbd_conf *mdev) { - char buffer[sizeof(struct cn_msg)+ - sizeof(struct drbd_nl_cfg_reply)+ - sizeof(struct get_state_tag_len_struct)+ - sizeof(short int)]; - struct cn_msg *cn_reply = (struct cn_msg *) buffer; - struct drbd_nl_cfg_reply *reply = - (struct drbd_nl_cfg_reply *)cn_reply->data; - unsigned short *tl = reply->tag_list; - - /* dev_warn(DEV, "drbd_bcast_state() got called\n"); */ - - tl = get_state_to_tags(mdev, (struct get_state *)&state, tl); - - put_unaligned(TT_END, tl++); /* Close the tag list */ - - cn_reply->id.idx = CN_IDX_DRBD; - cn_reply->id.val = CN_VAL_DRBD; - - cn_reply->seq = atomic_add_return(1, &drbd_nl_seq); - cn_reply->ack = 0; /* not used here. */ - cn_reply->len = sizeof(struct drbd_nl_cfg_reply) + - (int)((char *)tl - (char *)reply->tag_list); - cn_reply->flags = 0; - - reply->packet_type = P_get_state; - reply->minor = mdev_to_minor(mdev); - reply->ret_code = NO_ERROR; - - cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO); + if (mdev->state.disk == D_DISKLESS && + /* no need to be mdev->state.conn == C_STANDALONE && + * we may want to delete a minor from a live replication group. + */ + mdev->state.role == R_SECONDARY) { + _drbd_request_state(mdev, NS(conn, C_WF_REPORT_PARAMS), + CS_VERBOSE + CS_WAIT_COMPLETE); + idr_remove(&mdev->tconn->volumes, mdev->vnr); + idr_remove(&minors, mdev_to_minor(mdev)); + del_gendisk(mdev->vdisk); + synchronize_rcu(); + kref_put(&mdev->kref, &drbd_minor_destroy); + return NO_ERROR; + } else + return ERR_MINOR_CONFIGURED; } -void drbd_bcast_ev_helper(struct drbd_conf *mdev, char *helper_name) +int drbd_adm_delete_minor(struct sk_buff *skb, struct genl_info *info) { - char buffer[sizeof(struct cn_msg)+ - sizeof(struct drbd_nl_cfg_reply)+ - sizeof(struct call_helper_tag_len_struct)+ - sizeof(short int)]; - struct cn_msg *cn_reply = (struct cn_msg *) buffer; - struct drbd_nl_cfg_reply *reply = - (struct drbd_nl_cfg_reply *)cn_reply->data; - unsigned short *tl = reply->tag_list; - - /* dev_warn(DEV, "drbd_bcast_state() got called\n"); */ - - tl = tl_add_str(tl, T_helper, helper_name); - put_unaligned(TT_END, tl++); /* Close the tag list */ - - cn_reply->id.idx = CN_IDX_DRBD; - cn_reply->id.val = CN_VAL_DRBD; - - cn_reply->seq = atomic_add_return(1, &drbd_nl_seq); - cn_reply->ack = 0; /* not used here. */ - cn_reply->len = sizeof(struct drbd_nl_cfg_reply) + - (int)((char *)tl - (char *)reply->tag_list); - cn_reply->flags = 0; + enum drbd_ret_code retcode; - reply->packet_type = P_call_helper; - reply->minor = mdev_to_minor(mdev); - reply->ret_code = NO_ERROR; + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; - cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO); + retcode = adm_delete_minor(adm_ctx.mdev); +out: + drbd_adm_finish(info, retcode); + return 0; } -void drbd_bcast_ee(struct drbd_conf *mdev, - const char *reason, const int dgs, - const char* seen_hash, const char* calc_hash, - const struct drbd_epoch_entry* e) +int drbd_adm_down(struct sk_buff *skb, struct genl_info *info) { - struct cn_msg *cn_reply; - struct drbd_nl_cfg_reply *reply; - unsigned short *tl; - struct page *page; - unsigned len; + int retcode; /* enum drbd_ret_code rsp. enum drbd_state_rv */ + struct drbd_conf *mdev; + unsigned i; - if (!e) - return; - if (!reason || !reason[0]) - return; + retcode = drbd_adm_prepare(skb, info, 0); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; - /* apparently we have to memcpy twice, first to prepare the data for the - * struct cn_msg, then within cn_netlink_send from the cn_msg to the - * netlink skb. */ - /* receiver thread context, which is not in the writeout path (of this node), - * but may be in the writeout path of the _other_ node. - * GFP_NOIO to avoid potential "distributed deadlock". */ - cn_reply = kzalloc( - sizeof(struct cn_msg)+ - sizeof(struct drbd_nl_cfg_reply)+ - sizeof(struct dump_ee_tag_len_struct)+ - sizeof(short int), - GFP_NOIO); - - if (!cn_reply) { - dev_err(DEV, "could not kmalloc buffer for drbd_bcast_ee, sector %llu, size %u\n", - (unsigned long long)e->sector, e->size); - return; + if (!adm_ctx.tconn) { + retcode = ERR_RES_NOT_KNOWN; + goto out; } - reply = (struct drbd_nl_cfg_reply*)cn_reply->data; - tl = reply->tag_list; - - tl = tl_add_str(tl, T_dump_ee_reason, reason); - tl = tl_add_blob(tl, T_seen_digest, seen_hash, dgs); - tl = tl_add_blob(tl, T_calc_digest, calc_hash, dgs); - tl = tl_add_int(tl, T_ee_sector, &e->sector); - tl = tl_add_int(tl, T_ee_block_id, &e->block_id); - - /* dump the first 32k */ - len = min_t(unsigned, e->size, 32 << 10); - put_unaligned(T_ee_data, tl++); - put_unaligned(len, tl++); - - page = e->pages; - page_chain_for_each(page) { - void *d = kmap_atomic(page); - unsigned l = min_t(unsigned, len, PAGE_SIZE); - memcpy(tl, d, l); - kunmap_atomic(d); - tl = (unsigned short*)((char*)tl + l); - len -= l; - if (len == 0) - break; + /* demote */ + idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) { + retcode = drbd_set_role(mdev, R_SECONDARY, 0); + if (retcode < SS_SUCCESS) { + drbd_msg_put_info("failed to demote"); + goto out; + } } - put_unaligned(TT_END, tl++); /* Close the tag list */ - - cn_reply->id.idx = CN_IDX_DRBD; - cn_reply->id.val = CN_VAL_DRBD; - - cn_reply->seq = atomic_add_return(1,&drbd_nl_seq); - cn_reply->ack = 0; // not used here. - cn_reply->len = sizeof(struct drbd_nl_cfg_reply) + - (int)((char*)tl - (char*)reply->tag_list); - cn_reply->flags = 0; - - reply->packet_type = P_dump_ee; - reply->minor = mdev_to_minor(mdev); - reply->ret_code = NO_ERROR; - cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO); - kfree(cn_reply); -} - -void drbd_bcast_sync_progress(struct drbd_conf *mdev) -{ - char buffer[sizeof(struct cn_msg)+ - sizeof(struct drbd_nl_cfg_reply)+ - sizeof(struct sync_progress_tag_len_struct)+ - sizeof(short int)]; - struct cn_msg *cn_reply = (struct cn_msg *) buffer; - struct drbd_nl_cfg_reply *reply = - (struct drbd_nl_cfg_reply *)cn_reply->data; - unsigned short *tl = reply->tag_list; - unsigned long rs_left; - unsigned int res; + retcode = conn_try_disconnect(adm_ctx.tconn, 0); + if (retcode < SS_SUCCESS) { + drbd_msg_put_info("failed to disconnect"); + goto out; + } - /* no local ref, no bitmap, no syncer progress, no broadcast. */ - if (!get_ldev(mdev)) - return; - drbd_get_syncer_progress(mdev, &rs_left, &res); - put_ldev(mdev); + /* detach */ + idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) { + retcode = adm_detach(mdev, 0); + if (retcode < SS_SUCCESS || retcode > NO_ERROR) { + drbd_msg_put_info("failed to detach"); + goto out; + } + } - tl = tl_add_int(tl, T_sync_progress, &res); - put_unaligned(TT_END, tl++); /* Close the tag list */ + /* If we reach this, all volumes (of this tconn) are Secondary, + * Disconnected, Diskless, aka Unconfigured. Make sure all threads have + * actually stopped, state handling only does drbd_thread_stop_nowait(). */ + drbd_thread_stop(&adm_ctx.tconn->worker); - cn_reply->id.idx = CN_IDX_DRBD; - cn_reply->id.val = CN_VAL_DRBD; + /* Now, nothing can fail anymore */ - cn_reply->seq = atomic_add_return(1, &drbd_nl_seq); - cn_reply->ack = 0; /* not used here. */ - cn_reply->len = sizeof(struct drbd_nl_cfg_reply) + - (int)((char *)tl - (char *)reply->tag_list); - cn_reply->flags = 0; + /* delete volumes */ + idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) { + retcode = adm_delete_minor(mdev); + if (retcode != NO_ERROR) { + /* "can not happen" */ + drbd_msg_put_info("failed to delete volume"); + goto out; + } + } - reply->packet_type = P_sync_progress; - reply->minor = mdev_to_minor(mdev); - reply->ret_code = NO_ERROR; + /* delete connection */ + if (conn_lowest_minor(adm_ctx.tconn) < 0) { + list_del_rcu(&adm_ctx.tconn->all_tconn); + synchronize_rcu(); + kref_put(&adm_ctx.tconn->kref, &conn_destroy); - cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO); + retcode = NO_ERROR; + } else { + /* "can not happen" */ + retcode = ERR_RES_IN_USE; + drbd_msg_put_info("failed to delete connection"); + } + goto out; +out: + drbd_adm_finish(info, retcode); + return 0; } -int __init drbd_nl_init(void) +int drbd_adm_del_resource(struct sk_buff *skb, struct genl_info *info) { - static struct cb_id cn_id_drbd; - int err, try=10; + enum drbd_ret_code retcode; - cn_id_drbd.val = CN_VAL_DRBD; - do { - cn_id_drbd.idx = cn_idx; - err = cn_add_callback(&cn_id_drbd, "cn_drbd", &drbd_connector_callback); - if (!err) - break; - cn_idx = (cn_idx + CN_IDX_STEP); - } while (try--); + retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_RESOURCE); + if (!adm_ctx.reply_skb) + return retcode; + if (retcode != NO_ERROR) + goto out; - if (err) { - printk(KERN_ERR "drbd: cn_drbd failed to register\n"); - return err; + if (conn_lowest_minor(adm_ctx.tconn) < 0) { + list_del_rcu(&adm_ctx.tconn->all_tconn); + synchronize_rcu(); + kref_put(&adm_ctx.tconn->kref, &conn_destroy); + + retcode = NO_ERROR; + } else { + retcode = ERR_RES_IN_USE; } + if (retcode == NO_ERROR) + drbd_thread_stop(&adm_ctx.tconn->worker); +out: + drbd_adm_finish(info, retcode); return 0; } -void drbd_nl_cleanup(void) +void drbd_bcast_event(struct drbd_conf *mdev, const struct sib_info *sib) { - static struct cb_id cn_id_drbd; - - cn_id_drbd.idx = cn_idx; - cn_id_drbd.val = CN_VAL_DRBD; - - cn_del_callback(&cn_id_drbd); -} + static atomic_t drbd_genl_seq = ATOMIC_INIT(2); /* two. */ + struct sk_buff *msg; + struct drbd_genlmsghdr *d_out; + unsigned seq; + int err = -ENOMEM; + + if (sib->sib_reason == SIB_SYNC_PROGRESS) { + if (time_after(jiffies, mdev->rs_last_bcast + HZ)) + mdev->rs_last_bcast = jiffies; + else + return; + } -void drbd_nl_send_reply(struct cn_msg *req, int ret_code) -{ - char buffer[sizeof(struct cn_msg)+sizeof(struct drbd_nl_cfg_reply)]; - struct cn_msg *cn_reply = (struct cn_msg *) buffer; - struct drbd_nl_cfg_reply *reply = - (struct drbd_nl_cfg_reply *)cn_reply->data; - int rr; + seq = atomic_inc_return(&drbd_genl_seq); + msg = genlmsg_new(NLMSG_GOODSIZE, GFP_NOIO); + if (!msg) + goto failed; - memset(buffer, 0, sizeof(buffer)); - cn_reply->id = req->id; + err = -EMSGSIZE; + d_out = genlmsg_put(msg, 0, seq, &drbd_genl_family, 0, DRBD_EVENT); + if (!d_out) /* cannot happen, but anyways. */ + goto nla_put_failure; + d_out->minor = mdev_to_minor(mdev); + d_out->ret_code = NO_ERROR; - cn_reply->seq = req->seq; - cn_reply->ack = req->ack + 1; - cn_reply->len = sizeof(struct drbd_nl_cfg_reply); - cn_reply->flags = 0; + if (nla_put_status_info(msg, mdev, sib)) + goto nla_put_failure; + genlmsg_end(msg, d_out); + err = drbd_genl_multicast_events(msg, 0); + /* msg has been consumed or freed in netlink_broadcast() */ + if (err && err != -ESRCH) + goto failed; - reply->packet_type = P_return_code_only; - reply->minor = ((struct drbd_nl_cfg_req *)req->data)->drbd_minor; - reply->ret_code = ret_code; + return; - rr = cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO); - if (rr && rr != -ESRCH) - printk(KERN_INFO "drbd: cn_netlink_send()=%d\n", rr); +nla_put_failure: + nlmsg_free(msg); +failed: + dev_err(DEV, "Error %d while broadcasting event. " + "Event seq:%u sib_reason:%u\n", + err, seq, sib->sib_reason); } - diff --git a/drivers/block/drbd/drbd_nla.c b/drivers/block/drbd/drbd_nla.c new file mode 100644 index 000000000000..fa672b6df8d6 --- /dev/null +++ b/drivers/block/drbd/drbd_nla.c @@ -0,0 +1,55 @@ +#include "drbd_wrappers.h" +#include <linux/kernel.h> +#include <net/netlink.h> +#include <linux/drbd_genl_api.h> +#include "drbd_nla.h" + +static int drbd_nla_check_mandatory(int maxtype, struct nlattr *nla) +{ + struct nlattr *head = nla_data(nla); + int len = nla_len(nla); + int rem; + + /* + * validate_nla (called from nla_parse_nested) ignores attributes + * beyond maxtype, and does not understand the DRBD_GENLA_F_MANDATORY flag. + * In order to have it validate attributes with the DRBD_GENLA_F_MANDATORY + * flag set also, check and remove that flag before calling + * nla_parse_nested. + */ + + nla_for_each_attr(nla, head, len, rem) { + if (nla->nla_type & DRBD_GENLA_F_MANDATORY) { + nla->nla_type &= ~DRBD_GENLA_F_MANDATORY; + if (nla_type(nla) > maxtype) + return -EOPNOTSUPP; + } + } + return 0; +} + +int drbd_nla_parse_nested(struct nlattr *tb[], int maxtype, struct nlattr *nla, + const struct nla_policy *policy) +{ + int err; + + err = drbd_nla_check_mandatory(maxtype, nla); + if (!err) + err = nla_parse_nested(tb, maxtype, nla, policy); + + return err; +} + +struct nlattr *drbd_nla_find_nested(int maxtype, struct nlattr *nla, int attrtype) +{ + int err; + /* + * If any nested attribute has the DRBD_GENLA_F_MANDATORY flag set and + * we don't know about that attribute, reject all the nested + * attributes. + */ + err = drbd_nla_check_mandatory(maxtype, nla); + if (err) + return ERR_PTR(err); + return nla_find_nested(nla, attrtype); +} diff --git a/drivers/block/drbd/drbd_nla.h b/drivers/block/drbd/drbd_nla.h new file mode 100644 index 000000000000..679c2d5b4535 --- /dev/null +++ b/drivers/block/drbd/drbd_nla.h @@ -0,0 +1,8 @@ +#ifndef __DRBD_NLA_H +#define __DRBD_NLA_H + +extern int drbd_nla_parse_nested(struct nlattr *tb[], int maxtype, struct nlattr *nla, + const struct nla_policy *policy); +extern struct nlattr *drbd_nla_find_nested(int maxtype, struct nlattr *nla, int attrtype); + +#endif /* __DRBD_NLA_H */ diff --git a/drivers/block/drbd/drbd_proc.c b/drivers/block/drbd/drbd_proc.c index 5496104f90b9..56672a61eb94 100644 --- a/drivers/block/drbd/drbd_proc.c +++ b/drivers/block/drbd/drbd_proc.c @@ -167,18 +167,24 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq) * we convert to sectors in the display below. */ unsigned long bm_bits = drbd_bm_bits(mdev); unsigned long bit_pos; + unsigned long long stop_sector = 0; if (mdev->state.conn == C_VERIFY_S || - mdev->state.conn == C_VERIFY_T) + mdev->state.conn == C_VERIFY_T) { bit_pos = bm_bits - mdev->ov_left; - else + if (verify_can_do_stop_sector(mdev)) + stop_sector = mdev->ov_stop_sector; + } else bit_pos = mdev->bm_resync_fo; /* Total sectors may be slightly off for oddly * sized devices. So what. */ seq_printf(seq, - "\t%3d%% sector pos: %llu/%llu\n", + "\t%3d%% sector pos: %llu/%llu", (int)(bit_pos / (bm_bits/100+1)), (unsigned long long)bit_pos * BM_SECT_PER_BIT, (unsigned long long)bm_bits * BM_SECT_PER_BIT); + if (stop_sector != 0 && stop_sector != ULLONG_MAX) + seq_printf(seq, " stop sector: %llu", stop_sector); + seq_printf(seq, "\n"); } } @@ -194,9 +200,11 @@ static void resync_dump_detail(struct seq_file *seq, struct lc_element *e) static int drbd_seq_show(struct seq_file *seq, void *v) { - int i, hole = 0; + int i, prev_i = -1; const char *sn; struct drbd_conf *mdev; + struct net_conf *nc; + char wp; static char write_ordering_chars[] = { [WO_none] = 'n', @@ -227,16 +235,11 @@ static int drbd_seq_show(struct seq_file *seq, void *v) oos .. known out-of-sync kB */ - for (i = 0; i < minor_count; i++) { - mdev = minor_to_mdev(i); - if (!mdev) { - hole = 1; - continue; - } - if (hole) { - hole = 0; + rcu_read_lock(); + idr_for_each_entry(&minors, mdev, i) { + if (prev_i != i - 1) seq_printf(seq, "\n"); - } + prev_i = i; sn = drbd_conn_str(mdev->state.conn); @@ -248,6 +251,8 @@ static int drbd_seq_show(struct seq_file *seq, void *v) /* reset mdev->congestion_reason */ bdi_rw_congested(&mdev->rq_queue->backing_dev_info); + nc = rcu_dereference(mdev->tconn->net_conf); + wp = nc ? nc->wire_protocol - DRBD_PROT_A + 'A' : ' '; seq_printf(seq, "%2d: cs:%s ro:%s/%s ds:%s/%s %c %c%c%c%c%c%c\n" " ns:%u nr:%u dw:%u dr:%u al:%u bm:%u " @@ -257,9 +262,8 @@ static int drbd_seq_show(struct seq_file *seq, void *v) drbd_role_str(mdev->state.peer), drbd_disk_str(mdev->state.disk), drbd_disk_str(mdev->state.pdsk), - (mdev->net_conf == NULL ? ' ' : - (mdev->net_conf->wire_protocol - DRBD_PROT_A+'A')), - is_susp(mdev->state) ? 's' : 'r', + wp, + drbd_suspended(mdev) ? 's' : 'r', mdev->state.aftr_isp ? 'a' : '-', mdev->state.peer_isp ? 'p' : '-', mdev->state.user_isp ? 'u' : '-', @@ -276,8 +280,8 @@ static int drbd_seq_show(struct seq_file *seq, void *v) atomic_read(&mdev->rs_pending_cnt), atomic_read(&mdev->unacked_cnt), atomic_read(&mdev->ap_bio_cnt), - mdev->epochs, - write_ordering_chars[mdev->write_ordering] + mdev->tconn->epochs, + write_ordering_chars[mdev->tconn->write_ordering] ); seq_printf(seq, " oos:%llu\n", Bit2KB((unsigned long long) @@ -302,6 +306,7 @@ static int drbd_seq_show(struct seq_file *seq, void *v) } } } + rcu_read_unlock(); return 0; } diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index c74ca2df7431..a9eccfc6079b 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -48,17 +48,25 @@ #include "drbd_vli.h" +struct packet_info { + enum drbd_packet cmd; + unsigned int size; + unsigned int vnr; + void *data; +}; + enum finish_epoch { FE_STILL_LIVE, FE_DESTROYED, FE_RECYCLED, }; -static int drbd_do_handshake(struct drbd_conf *mdev); -static int drbd_do_auth(struct drbd_conf *mdev); +static int drbd_do_features(struct drbd_tconn *tconn); +static int drbd_do_auth(struct drbd_tconn *tconn); +static int drbd_disconnected(struct drbd_conf *mdev); -static enum finish_epoch drbd_may_finish_epoch(struct drbd_conf *, struct drbd_epoch *, enum epoch_event); -static int e_end_block(struct drbd_conf *, struct drbd_work *, int); +static enum finish_epoch drbd_may_finish_epoch(struct drbd_tconn *, struct drbd_epoch *, enum epoch_event); +static int e_end_block(struct drbd_work *, int); #define GFP_TRY (__GFP_HIGHMEM | __GFP_NOWARN) @@ -142,11 +150,12 @@ static void page_chain_add(struct page **head, *head = chain_first; } -static struct page *drbd_pp_first_pages_or_try_alloc(struct drbd_conf *mdev, int number) +static struct page *__drbd_alloc_pages(struct drbd_conf *mdev, + unsigned int number) { struct page *page = NULL; struct page *tmp = NULL; - int i = 0; + unsigned int i = 0; /* Yes, testing drbd_pp_vacant outside the lock is racy. * So what. It saves a spin_lock. */ @@ -175,7 +184,7 @@ static struct page *drbd_pp_first_pages_or_try_alloc(struct drbd_conf *mdev, int return page; /* Not enough pages immediately available this time. - * No need to jump around here, drbd_pp_alloc will retry this + * No need to jump around here, drbd_alloc_pages will retry this * function "soon". */ if (page) { tmp = page_chain_tail(page, NULL); @@ -187,9 +196,10 @@ static struct page *drbd_pp_first_pages_or_try_alloc(struct drbd_conf *mdev, int return NULL; } -static void reclaim_net_ee(struct drbd_conf *mdev, struct list_head *to_be_freed) +static void reclaim_finished_net_peer_reqs(struct drbd_conf *mdev, + struct list_head *to_be_freed) { - struct drbd_epoch_entry *e; + struct drbd_peer_request *peer_req; struct list_head *le, *tle; /* The EEs are always appended to the end of the list. Since @@ -198,8 +208,8 @@ static void reclaim_net_ee(struct drbd_conf *mdev, struct list_head *to_be_freed stop to examine the list... */ list_for_each_safe(le, tle, &mdev->net_ee) { - e = list_entry(le, struct drbd_epoch_entry, w.list); - if (drbd_ee_has_active_page(e)) + peer_req = list_entry(le, struct drbd_peer_request, w.list); + if (drbd_peer_req_has_active_page(peer_req)) break; list_move(le, to_be_freed); } @@ -208,18 +218,18 @@ static void reclaim_net_ee(struct drbd_conf *mdev, struct list_head *to_be_freed static void drbd_kick_lo_and_reclaim_net(struct drbd_conf *mdev) { LIST_HEAD(reclaimed); - struct drbd_epoch_entry *e, *t; + struct drbd_peer_request *peer_req, *t; - spin_lock_irq(&mdev->req_lock); - reclaim_net_ee(mdev, &reclaimed); - spin_unlock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); + reclaim_finished_net_peer_reqs(mdev, &reclaimed); + spin_unlock_irq(&mdev->tconn->req_lock); - list_for_each_entry_safe(e, t, &reclaimed, w.list) - drbd_free_net_ee(mdev, e); + list_for_each_entry_safe(peer_req, t, &reclaimed, w.list) + drbd_free_net_peer_req(mdev, peer_req); } /** - * drbd_pp_alloc() - Returns @number pages, retries forever (or until signalled) + * drbd_alloc_pages() - Returns @number pages, retries forever (or until signalled) * @mdev: DRBD device. * @number: number of pages requested * @retry: whether to retry, if not enough pages are available right now @@ -230,23 +240,31 @@ static void drbd_kick_lo_and_reclaim_net(struct drbd_conf *mdev) * * Returns a page chain linked via page->private. */ -static struct page *drbd_pp_alloc(struct drbd_conf *mdev, unsigned number, bool retry) +struct page *drbd_alloc_pages(struct drbd_conf *mdev, unsigned int number, + bool retry) { struct page *page = NULL; + struct net_conf *nc; DEFINE_WAIT(wait); + int mxb; /* Yes, we may run up to @number over max_buffers. If we * follow it strictly, the admin will get it wrong anyways. */ - if (atomic_read(&mdev->pp_in_use) < mdev->net_conf->max_buffers) - page = drbd_pp_first_pages_or_try_alloc(mdev, number); + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + mxb = nc ? nc->max_buffers : 1000000; + rcu_read_unlock(); + + if (atomic_read(&mdev->pp_in_use) < mxb) + page = __drbd_alloc_pages(mdev, number); while (page == NULL) { prepare_to_wait(&drbd_pp_wait, &wait, TASK_INTERRUPTIBLE); drbd_kick_lo_and_reclaim_net(mdev); - if (atomic_read(&mdev->pp_in_use) < mdev->net_conf->max_buffers) { - page = drbd_pp_first_pages_or_try_alloc(mdev, number); + if (atomic_read(&mdev->pp_in_use) < mxb) { + page = __drbd_alloc_pages(mdev, number); if (page) break; } @@ -255,7 +273,7 @@ static struct page *drbd_pp_alloc(struct drbd_conf *mdev, unsigned number, bool break; if (signal_pending(current)) { - dev_warn(DEV, "drbd_pp_alloc interrupted!\n"); + dev_warn(DEV, "drbd_alloc_pages interrupted!\n"); break; } @@ -268,11 +286,11 @@ static struct page *drbd_pp_alloc(struct drbd_conf *mdev, unsigned number, bool return page; } -/* Must not be used from irq, as that may deadlock: see drbd_pp_alloc. - * Is also used from inside an other spin_lock_irq(&mdev->req_lock); +/* Must not be used from irq, as that may deadlock: see drbd_alloc_pages. + * Is also used from inside an other spin_lock_irq(&mdev->tconn->req_lock); * Either links the page chain back to the global pool, * or returns all pages to the system. */ -static void drbd_pp_free(struct drbd_conf *mdev, struct page *page, int is_net) +static void drbd_free_pages(struct drbd_conf *mdev, struct page *page, int is_net) { atomic_t *a = is_net ? &mdev->pp_in_use_by_net : &mdev->pp_in_use; int i; @@ -280,7 +298,7 @@ static void drbd_pp_free(struct drbd_conf *mdev, struct page *page, int is_net) if (page == NULL) return; - if (drbd_pp_vacant > (DRBD_MAX_BIO_SIZE/PAGE_SIZE)*minor_count) + if (drbd_pp_vacant > (DRBD_MAX_BIO_SIZE/PAGE_SIZE) * minor_count) i = page_chain_free(page); else { struct page *tmp; @@ -302,127 +320,130 @@ You need to hold the req_lock: _drbd_wait_ee_list_empty() You must not have the req_lock: - drbd_free_ee() - drbd_alloc_ee() - drbd_init_ee() - drbd_release_ee() + drbd_free_peer_req() + drbd_alloc_peer_req() + drbd_free_peer_reqs() drbd_ee_fix_bhs() - drbd_process_done_ee() + drbd_finish_peer_reqs() drbd_clear_done_ee() drbd_wait_ee_list_empty() */ -struct drbd_epoch_entry *drbd_alloc_ee(struct drbd_conf *mdev, - u64 id, - sector_t sector, - unsigned int data_size, - gfp_t gfp_mask) __must_hold(local) +struct drbd_peer_request * +drbd_alloc_peer_req(struct drbd_conf *mdev, u64 id, sector_t sector, + unsigned int data_size, gfp_t gfp_mask) __must_hold(local) { - struct drbd_epoch_entry *e; + struct drbd_peer_request *peer_req; struct page *page = NULL; unsigned nr_pages = (data_size + PAGE_SIZE -1) >> PAGE_SHIFT; if (drbd_insert_fault(mdev, DRBD_FAULT_AL_EE)) return NULL; - e = mempool_alloc(drbd_ee_mempool, gfp_mask & ~__GFP_HIGHMEM); - if (!e) { + peer_req = mempool_alloc(drbd_ee_mempool, gfp_mask & ~__GFP_HIGHMEM); + if (!peer_req) { if (!(gfp_mask & __GFP_NOWARN)) - dev_err(DEV, "alloc_ee: Allocation of an EE failed\n"); + dev_err(DEV, "%s: allocation failed\n", __func__); return NULL; } if (data_size) { - page = drbd_pp_alloc(mdev, nr_pages, (gfp_mask & __GFP_WAIT)); + page = drbd_alloc_pages(mdev, nr_pages, (gfp_mask & __GFP_WAIT)); if (!page) goto fail; } - INIT_HLIST_NODE(&e->collision); - e->epoch = NULL; - e->mdev = mdev; - e->pages = page; - atomic_set(&e->pending_bios, 0); - e->size = data_size; - e->flags = 0; - e->sector = sector; - e->block_id = id; + drbd_clear_interval(&peer_req->i); + peer_req->i.size = data_size; + peer_req->i.sector = sector; + peer_req->i.local = false; + peer_req->i.waiting = false; + + peer_req->epoch = NULL; + peer_req->w.mdev = mdev; + peer_req->pages = page; + atomic_set(&peer_req->pending_bios, 0); + peer_req->flags = 0; + /* + * The block_id is opaque to the receiver. It is not endianness + * converted, and sent back to the sender unchanged. + */ + peer_req->block_id = id; - return e; + return peer_req; fail: - mempool_free(e, drbd_ee_mempool); + mempool_free(peer_req, drbd_ee_mempool); return NULL; } -void drbd_free_some_ee(struct drbd_conf *mdev, struct drbd_epoch_entry *e, int is_net) +void __drbd_free_peer_req(struct drbd_conf *mdev, struct drbd_peer_request *peer_req, + int is_net) { - if (e->flags & EE_HAS_DIGEST) - kfree(e->digest); - drbd_pp_free(mdev, e->pages, is_net); - D_ASSERT(atomic_read(&e->pending_bios) == 0); - D_ASSERT(hlist_unhashed(&e->collision)); - mempool_free(e, drbd_ee_mempool); + if (peer_req->flags & EE_HAS_DIGEST) + kfree(peer_req->digest); + drbd_free_pages(mdev, peer_req->pages, is_net); + D_ASSERT(atomic_read(&peer_req->pending_bios) == 0); + D_ASSERT(drbd_interval_empty(&peer_req->i)); + mempool_free(peer_req, drbd_ee_mempool); } -int drbd_release_ee(struct drbd_conf *mdev, struct list_head *list) +int drbd_free_peer_reqs(struct drbd_conf *mdev, struct list_head *list) { LIST_HEAD(work_list); - struct drbd_epoch_entry *e, *t; + struct drbd_peer_request *peer_req, *t; int count = 0; int is_net = list == &mdev->net_ee; - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); list_splice_init(list, &work_list); - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); - list_for_each_entry_safe(e, t, &work_list, w.list) { - drbd_free_some_ee(mdev, e, is_net); + list_for_each_entry_safe(peer_req, t, &work_list, w.list) { + __drbd_free_peer_req(mdev, peer_req, is_net); count++; } return count; } - /* - * This function is called from _asender only_ - * but see also comments in _req_mod(,barrier_acked) - * and receive_Barrier. - * - * Move entries from net_ee to done_ee, if ready. - * Grab done_ee, call all callbacks, free the entries. - * The callbacks typically send out ACKs. + * See also comments in _req_mod(,BARRIER_ACKED) and receive_Barrier. */ -static int drbd_process_done_ee(struct drbd_conf *mdev) +static int drbd_finish_peer_reqs(struct drbd_conf *mdev) { LIST_HEAD(work_list); LIST_HEAD(reclaimed); - struct drbd_epoch_entry *e, *t; - int ok = (mdev->state.conn >= C_WF_REPORT_PARAMS); + struct drbd_peer_request *peer_req, *t; + int err = 0; - spin_lock_irq(&mdev->req_lock); - reclaim_net_ee(mdev, &reclaimed); + spin_lock_irq(&mdev->tconn->req_lock); + reclaim_finished_net_peer_reqs(mdev, &reclaimed); list_splice_init(&mdev->done_ee, &work_list); - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); - list_for_each_entry_safe(e, t, &reclaimed, w.list) - drbd_free_net_ee(mdev, e); + list_for_each_entry_safe(peer_req, t, &reclaimed, w.list) + drbd_free_net_peer_req(mdev, peer_req); /* possible callbacks here: - * e_end_block, and e_end_resync_block, e_send_discard_ack. + * e_end_block, and e_end_resync_block, e_send_superseded. * all ignore the last argument. */ - list_for_each_entry_safe(e, t, &work_list, w.list) { + list_for_each_entry_safe(peer_req, t, &work_list, w.list) { + int err2; + /* list_del not necessary, next/prev members not touched */ - ok = e->w.cb(mdev, &e->w, !ok) && ok; - drbd_free_ee(mdev, e); + err2 = peer_req->w.cb(&peer_req->w, !!err); + if (!err) + err = err2; + drbd_free_peer_req(mdev, peer_req); } wake_up(&mdev->ee_wait); - return ok; + return err; } -void _drbd_wait_ee_list_empty(struct drbd_conf *mdev, struct list_head *head) +static void _drbd_wait_ee_list_empty(struct drbd_conf *mdev, + struct list_head *head) { DEFINE_WAIT(wait); @@ -430,55 +451,22 @@ void _drbd_wait_ee_list_empty(struct drbd_conf *mdev, struct list_head *head) * and calling prepare_to_wait in the fast path */ while (!list_empty(head)) { prepare_to_wait(&mdev->ee_wait, &wait, TASK_UNINTERRUPTIBLE); - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); io_schedule(); finish_wait(&mdev->ee_wait, &wait); - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); } } -void drbd_wait_ee_list_empty(struct drbd_conf *mdev, struct list_head *head) +static void drbd_wait_ee_list_empty(struct drbd_conf *mdev, + struct list_head *head) { - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); _drbd_wait_ee_list_empty(mdev, head); - spin_unlock_irq(&mdev->req_lock); -} - -/* see also kernel_accept; which is only present since 2.6.18. - * also we want to log which part of it failed, exactly */ -static int drbd_accept(struct drbd_conf *mdev, const char **what, - struct socket *sock, struct socket **newsock) -{ - struct sock *sk = sock->sk; - int err = 0; - - *what = "listen"; - err = sock->ops->listen(sock, 5); - if (err < 0) - goto out; - - *what = "sock_create_lite"; - err = sock_create_lite(sk->sk_family, sk->sk_type, sk->sk_protocol, - newsock); - if (err < 0) - goto out; - - *what = "accept"; - err = sock->ops->accept(sock, *newsock, 0); - if (err < 0) { - sock_release(*newsock); - *newsock = NULL; - goto out; - } - (*newsock)->ops = sock->ops; - __module_get((*newsock)->ops->owner); - -out: - return err; + spin_unlock_irq(&mdev->tconn->req_lock); } -static int drbd_recv_short(struct drbd_conf *mdev, struct socket *sock, - void *buf, size_t size, int flags) +static int drbd_recv_short(struct socket *sock, void *buf, size_t size, int flags) { mm_segment_t oldfs; struct kvec iov = { @@ -500,59 +488,62 @@ static int drbd_recv_short(struct drbd_conf *mdev, struct socket *sock, return rv; } -static int drbd_recv(struct drbd_conf *mdev, void *buf, size_t size) +static int drbd_recv(struct drbd_tconn *tconn, void *buf, size_t size) { - mm_segment_t oldfs; - struct kvec iov = { - .iov_base = buf, - .iov_len = size, - }; - struct msghdr msg = { - .msg_iovlen = 1, - .msg_iov = (struct iovec *)&iov, - .msg_flags = MSG_WAITALL | MSG_NOSIGNAL - }; int rv; - oldfs = get_fs(); - set_fs(KERNEL_DS); + rv = drbd_recv_short(tconn->data.socket, buf, size, 0); - for (;;) { - rv = sock_recvmsg(mdev->data.socket, &msg, size, msg.msg_flags); - if (rv == size) - break; + if (rv < 0) { + if (rv == -ECONNRESET) + conn_info(tconn, "sock was reset by peer\n"); + else if (rv != -ERESTARTSYS) + conn_err(tconn, "sock_recvmsg returned %d\n", rv); + } else if (rv == 0) { + if (test_bit(DISCONNECT_SENT, &tconn->flags)) { + long t; + rcu_read_lock(); + t = rcu_dereference(tconn->net_conf)->ping_timeo * HZ/10; + rcu_read_unlock(); - /* Note: - * ECONNRESET other side closed the connection - * ERESTARTSYS (on sock) we got a signal - */ + t = wait_event_timeout(tconn->ping_wait, tconn->cstate < C_WF_REPORT_PARAMS, t); - if (rv < 0) { - if (rv == -ECONNRESET) - dev_info(DEV, "sock was reset by peer\n"); - else if (rv != -ERESTARTSYS) - dev_err(DEV, "sock_recvmsg returned %d\n", rv); - break; - } else if (rv == 0) { - dev_info(DEV, "sock was shut down by peer\n"); - break; - } else { - /* signal came in, or peer/link went down, - * after we read a partial message - */ - /* D_ASSERT(signal_pending(current)); */ - break; + if (t) + goto out; } - }; - - set_fs(oldfs); + conn_info(tconn, "sock was shut down by peer\n"); + } if (rv != size) - drbd_force_state(mdev, NS(conn, C_BROKEN_PIPE)); + conn_request_state(tconn, NS(conn, C_BROKEN_PIPE), CS_HARD); +out: return rv; } +static int drbd_recv_all(struct drbd_tconn *tconn, void *buf, size_t size) +{ + int err; + + err = drbd_recv(tconn, buf, size); + if (err != size) { + if (err >= 0) + err = -EIO; + } else + err = 0; + return err; +} + +static int drbd_recv_all_warn(struct drbd_tconn *tconn, void *buf, size_t size) +{ + int err; + + err = drbd_recv_all(tconn, buf, size); + if (err && !signal_pending(current)) + conn_warn(tconn, "short read (expected size %d)\n", (int)size); + return err; +} + /* quoting tcp(7): * On individual connections, the socket buffer size must be set prior to the * listen(2) or connect(2) calls in order to have it take effect. @@ -572,29 +563,50 @@ static void drbd_setbufsize(struct socket *sock, unsigned int snd, } } -static struct socket *drbd_try_connect(struct drbd_conf *mdev) +static struct socket *drbd_try_connect(struct drbd_tconn *tconn) { const char *what; struct socket *sock; struct sockaddr_in6 src_in6; - int err; + struct sockaddr_in6 peer_in6; + struct net_conf *nc; + int err, peer_addr_len, my_addr_len; + int sndbuf_size, rcvbuf_size, connect_int; int disconnect_on_error = 1; - if (!get_net_conf(mdev)) + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); + if (!nc) { + rcu_read_unlock(); return NULL; + } + sndbuf_size = nc->sndbuf_size; + rcvbuf_size = nc->rcvbuf_size; + connect_int = nc->connect_int; + rcu_read_unlock(); + + my_addr_len = min_t(int, tconn->my_addr_len, sizeof(src_in6)); + memcpy(&src_in6, &tconn->my_addr, my_addr_len); + + if (((struct sockaddr *)&tconn->my_addr)->sa_family == AF_INET6) + src_in6.sin6_port = 0; + else + ((struct sockaddr_in *)&src_in6)->sin_port = 0; /* AF_INET & AF_SCI */ + + peer_addr_len = min_t(int, tconn->peer_addr_len, sizeof(src_in6)); + memcpy(&peer_in6, &tconn->peer_addr, peer_addr_len); what = "sock_create_kern"; - err = sock_create_kern(((struct sockaddr *)mdev->net_conf->my_addr)->sa_family, - SOCK_STREAM, IPPROTO_TCP, &sock); + err = sock_create_kern(((struct sockaddr *)&src_in6)->sa_family, + SOCK_STREAM, IPPROTO_TCP, &sock); if (err < 0) { sock = NULL; goto out; } sock->sk->sk_rcvtimeo = - sock->sk->sk_sndtimeo = mdev->net_conf->try_connect_int*HZ; - drbd_setbufsize(sock, mdev->net_conf->sndbuf_size, - mdev->net_conf->rcvbuf_size); + sock->sk->sk_sndtimeo = connect_int * HZ; + drbd_setbufsize(sock, sndbuf_size, rcvbuf_size); /* explicitly bind to the configured IP as source IP * for the outgoing connections. @@ -603,17 +615,8 @@ static struct socket *drbd_try_connect(struct drbd_conf *mdev) * Make sure to use 0 as port number, so linux selects * a free one dynamically. */ - memcpy(&src_in6, mdev->net_conf->my_addr, - min_t(int, mdev->net_conf->my_addr_len, sizeof(src_in6))); - if (((struct sockaddr *)mdev->net_conf->my_addr)->sa_family == AF_INET6) - src_in6.sin6_port = 0; - else - ((struct sockaddr_in *)&src_in6)->sin_port = 0; /* AF_INET & AF_SCI */ - what = "bind before connect"; - err = sock->ops->bind(sock, - (struct sockaddr *) &src_in6, - mdev->net_conf->my_addr_len); + err = sock->ops->bind(sock, (struct sockaddr *) &src_in6, my_addr_len); if (err < 0) goto out; @@ -621,9 +624,7 @@ static struct socket *drbd_try_connect(struct drbd_conf *mdev) * stay C_WF_CONNECTION, don't go Disconnecting! */ disconnect_on_error = 0; what = "connect"; - err = sock->ops->connect(sock, - (struct sockaddr *)mdev->net_conf->peer_addr, - mdev->net_conf->peer_addr_len, 0); + err = sock->ops->connect(sock, (struct sockaddr *) &peer_in6, peer_addr_len, 0); out: if (err < 0) { @@ -641,91 +642,174 @@ out: disconnect_on_error = 0; break; default: - dev_err(DEV, "%s failed, err = %d\n", what, err); + conn_err(tconn, "%s failed, err = %d\n", what, err); } if (disconnect_on_error) - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); + conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD); } - put_net_conf(mdev); + return sock; } -static struct socket *drbd_wait_for_connect(struct drbd_conf *mdev) +struct accept_wait_data { + struct drbd_tconn *tconn; + struct socket *s_listen; + struct completion door_bell; + void (*original_sk_state_change)(struct sock *sk); + +}; + +static void drbd_incoming_connection(struct sock *sk) { - int timeo, err; - struct socket *s_estab = NULL, *s_listen; + struct accept_wait_data *ad = sk->sk_user_data; + void (*state_change)(struct sock *sk); + + state_change = ad->original_sk_state_change; + if (sk->sk_state == TCP_ESTABLISHED) + complete(&ad->door_bell); + state_change(sk); +} + +static int prepare_listen_socket(struct drbd_tconn *tconn, struct accept_wait_data *ad) +{ + int err, sndbuf_size, rcvbuf_size, my_addr_len; + struct sockaddr_in6 my_addr; + struct socket *s_listen; + struct net_conf *nc; const char *what; - if (!get_net_conf(mdev)) - return NULL; + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); + if (!nc) { + rcu_read_unlock(); + return -EIO; + } + sndbuf_size = nc->sndbuf_size; + rcvbuf_size = nc->rcvbuf_size; + rcu_read_unlock(); + + my_addr_len = min_t(int, tconn->my_addr_len, sizeof(struct sockaddr_in6)); + memcpy(&my_addr, &tconn->my_addr, my_addr_len); what = "sock_create_kern"; - err = sock_create_kern(((struct sockaddr *)mdev->net_conf->my_addr)->sa_family, - SOCK_STREAM, IPPROTO_TCP, &s_listen); + err = sock_create_kern(((struct sockaddr *)&my_addr)->sa_family, + SOCK_STREAM, IPPROTO_TCP, &s_listen); if (err) { s_listen = NULL; goto out; } - timeo = mdev->net_conf->try_connect_int * HZ; - timeo += (random32() & 1) ? timeo / 7 : -timeo / 7; /* 28.5% random jitter */ - - s_listen->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ - s_listen->sk->sk_rcvtimeo = timeo; - s_listen->sk->sk_sndtimeo = timeo; - drbd_setbufsize(s_listen, mdev->net_conf->sndbuf_size, - mdev->net_conf->rcvbuf_size); + s_listen->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ + drbd_setbufsize(s_listen, sndbuf_size, rcvbuf_size); what = "bind before listen"; - err = s_listen->ops->bind(s_listen, - (struct sockaddr *) mdev->net_conf->my_addr, - mdev->net_conf->my_addr_len); + err = s_listen->ops->bind(s_listen, (struct sockaddr *)&my_addr, my_addr_len); if (err < 0) goto out; - err = drbd_accept(mdev, &what, s_listen, &s_estab); + ad->s_listen = s_listen; + write_lock_bh(&s_listen->sk->sk_callback_lock); + ad->original_sk_state_change = s_listen->sk->sk_state_change; + s_listen->sk->sk_state_change = drbd_incoming_connection; + s_listen->sk->sk_user_data = ad; + write_unlock_bh(&s_listen->sk->sk_callback_lock); + + what = "listen"; + err = s_listen->ops->listen(s_listen, 5); + if (err < 0) + goto out; + return 0; out: if (s_listen) sock_release(s_listen); if (err < 0) { if (err != -EAGAIN && err != -EINTR && err != -ERESTARTSYS) { - dev_err(DEV, "%s failed, err = %d\n", what, err); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); + conn_err(tconn, "%s failed, err = %d\n", what, err); + conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD); } } - put_net_conf(mdev); - return s_estab; + return -EIO; } -static int drbd_send_fp(struct drbd_conf *mdev, - struct socket *sock, enum drbd_packets cmd) +static void unregister_state_change(struct sock *sk, struct accept_wait_data *ad) { - struct p_header80 *h = &mdev->data.sbuf.header.h80; - - return _drbd_send_cmd(mdev, sock, cmd, h, sizeof(*h), 0); + write_lock_bh(&sk->sk_callback_lock); + sk->sk_state_change = ad->original_sk_state_change; + sk->sk_user_data = NULL; + write_unlock_bh(&sk->sk_callback_lock); } -static enum drbd_packets drbd_recv_fp(struct drbd_conf *mdev, struct socket *sock) +static struct socket *drbd_wait_for_connect(struct drbd_tconn *tconn, struct accept_wait_data *ad) { - struct p_header80 *h = &mdev->data.rbuf.header.h80; - int rr; + int timeo, connect_int, err = 0; + struct socket *s_estab = NULL; + struct net_conf *nc; + + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); + if (!nc) { + rcu_read_unlock(); + return NULL; + } + connect_int = nc->connect_int; + rcu_read_unlock(); + + timeo = connect_int * HZ; + timeo += (random32() & 1) ? timeo / 7 : -timeo / 7; /* 28.5% random jitter */ + + err = wait_for_completion_interruptible_timeout(&ad->door_bell, timeo); + if (err <= 0) + return NULL; + + err = kernel_accept(ad->s_listen, &s_estab, 0); + if (err < 0) { + if (err != -EAGAIN && err != -EINTR && err != -ERESTARTSYS) { + conn_err(tconn, "accept failed, err = %d\n", err); + conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD); + } + } + + if (s_estab) + unregister_state_change(s_estab->sk, ad); + + return s_estab; +} - rr = drbd_recv_short(mdev, sock, h, sizeof(*h), 0); +static int decode_header(struct drbd_tconn *, void *, struct packet_info *); - if (rr == sizeof(*h) && h->magic == BE_DRBD_MAGIC) - return be16_to_cpu(h->command); +static int send_first_packet(struct drbd_tconn *tconn, struct drbd_socket *sock, + enum drbd_packet cmd) +{ + if (!conn_prepare_command(tconn, sock)) + return -EIO; + return conn_send_command(tconn, sock, cmd, 0, NULL, 0); +} - return 0xffff; +static int receive_first_packet(struct drbd_tconn *tconn, struct socket *sock) +{ + unsigned int header_size = drbd_header_size(tconn); + struct packet_info pi; + int err; + + err = drbd_recv_short(sock, tconn->data.rbuf, header_size, 0); + if (err != header_size) { + if (err >= 0) + err = -EIO; + return err; + } + err = decode_header(tconn, tconn->data.rbuf, &pi); + if (err) + return err; + return pi.cmd; } /** * drbd_socket_okay() - Free the socket if its connection is not okay - * @mdev: DRBD device. * @sock: pointer to the pointer to the socket. */ -static int drbd_socket_okay(struct drbd_conf *mdev, struct socket **sock) +static int drbd_socket_okay(struct socket **sock) { int rr; char tb[4]; @@ -733,7 +817,7 @@ static int drbd_socket_okay(struct drbd_conf *mdev, struct socket **sock) if (!*sock) return false; - rr = drbd_recv_short(mdev, *sock, tb, 4, MSG_DONTWAIT | MSG_PEEK); + rr = drbd_recv_short(*sock, tb, 4, MSG_DONTWAIT | MSG_PEEK); if (rr > 0 || rr == -EAGAIN) { return true; @@ -743,6 +827,31 @@ static int drbd_socket_okay(struct drbd_conf *mdev, struct socket **sock) return false; } } +/* Gets called if a connection is established, or if a new minor gets created + in a connection */ +int drbd_connected(struct drbd_conf *mdev) +{ + int err; + + atomic_set(&mdev->packet_seq, 0); + mdev->peer_seq = 0; + + mdev->state_mutex = mdev->tconn->agreed_pro_version < 100 ? + &mdev->tconn->cstate_mutex : + &mdev->own_state_mutex; + + err = drbd_send_sync_param(mdev); + if (!err) + err = drbd_send_sizes(mdev, 0, 0); + if (!err) + err = drbd_send_uuids(mdev); + if (!err) + err = drbd_send_current_state(mdev); + clear_bit(USE_DEGR_WFC_T, &mdev->flags); + clear_bit(RESIZE_PENDING, &mdev->flags); + mod_timer(&mdev->request_timer, jiffies + HZ); /* just start it here. */ + return err; +} /* * return values: @@ -752,232 +861,315 @@ static int drbd_socket_okay(struct drbd_conf *mdev, struct socket **sock) * no point in trying again, please go standalone. * -2 We do not have a network config... */ -static int drbd_connect(struct drbd_conf *mdev) +static int conn_connect(struct drbd_tconn *tconn) { - struct socket *s, *sock, *msock; - int try, h, ok; + struct drbd_socket sock, msock; + struct drbd_conf *mdev; + struct net_conf *nc; + int vnr, timeout, h, ok; + bool discard_my_data; enum drbd_state_rv rv; + struct accept_wait_data ad = { + .tconn = tconn, + .door_bell = COMPLETION_INITIALIZER_ONSTACK(ad.door_bell), + }; - D_ASSERT(!mdev->data.socket); - - if (drbd_request_state(mdev, NS(conn, C_WF_CONNECTION)) < SS_SUCCESS) + clear_bit(DISCONNECT_SENT, &tconn->flags); + if (conn_request_state(tconn, NS(conn, C_WF_CONNECTION), CS_VERBOSE) < SS_SUCCESS) return -2; - clear_bit(DISCARD_CONCURRENT, &mdev->flags); + mutex_init(&sock.mutex); + sock.sbuf = tconn->data.sbuf; + sock.rbuf = tconn->data.rbuf; + sock.socket = NULL; + mutex_init(&msock.mutex); + msock.sbuf = tconn->meta.sbuf; + msock.rbuf = tconn->meta.rbuf; + msock.socket = NULL; + + /* Assume that the peer only understands protocol 80 until we know better. */ + tconn->agreed_pro_version = 80; - sock = NULL; - msock = NULL; + if (prepare_listen_socket(tconn, &ad)) + return 0; do { - for (try = 0;;) { - /* 3 tries, this should take less than a second! */ - s = drbd_try_connect(mdev); - if (s || ++try >= 3) - break; - /* give the other side time to call bind() & listen() */ - schedule_timeout_interruptible(HZ / 10); - } + struct socket *s; + s = drbd_try_connect(tconn); if (s) { - if (!sock) { - drbd_send_fp(mdev, s, P_HAND_SHAKE_S); - sock = s; - s = NULL; - } else if (!msock) { - drbd_send_fp(mdev, s, P_HAND_SHAKE_M); - msock = s; - s = NULL; + if (!sock.socket) { + sock.socket = s; + send_first_packet(tconn, &sock, P_INITIAL_DATA); + } else if (!msock.socket) { + clear_bit(RESOLVE_CONFLICTS, &tconn->flags); + msock.socket = s; + send_first_packet(tconn, &msock, P_INITIAL_META); } else { - dev_err(DEV, "Logic error in drbd_connect()\n"); + conn_err(tconn, "Logic error in conn_connect()\n"); goto out_release_sockets; } } - if (sock && msock) { - schedule_timeout_interruptible(mdev->net_conf->ping_timeo*HZ/10); - ok = drbd_socket_okay(mdev, &sock); - ok = drbd_socket_okay(mdev, &msock) && ok; + if (sock.socket && msock.socket) { + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); + timeout = nc->ping_timeo * HZ / 10; + rcu_read_unlock(); + schedule_timeout_interruptible(timeout); + ok = drbd_socket_okay(&sock.socket); + ok = drbd_socket_okay(&msock.socket) && ok; if (ok) break; } retry: - s = drbd_wait_for_connect(mdev); + s = drbd_wait_for_connect(tconn, &ad); if (s) { - try = drbd_recv_fp(mdev, s); - drbd_socket_okay(mdev, &sock); - drbd_socket_okay(mdev, &msock); - switch (try) { - case P_HAND_SHAKE_S: - if (sock) { - dev_warn(DEV, "initial packet S crossed\n"); - sock_release(sock); + int fp = receive_first_packet(tconn, s); + drbd_socket_okay(&sock.socket); + drbd_socket_okay(&msock.socket); + switch (fp) { + case P_INITIAL_DATA: + if (sock.socket) { + conn_warn(tconn, "initial packet S crossed\n"); + sock_release(sock.socket); + sock.socket = s; + goto randomize; } - sock = s; + sock.socket = s; break; - case P_HAND_SHAKE_M: - if (msock) { - dev_warn(DEV, "initial packet M crossed\n"); - sock_release(msock); + case P_INITIAL_META: + set_bit(RESOLVE_CONFLICTS, &tconn->flags); + if (msock.socket) { + conn_warn(tconn, "initial packet M crossed\n"); + sock_release(msock.socket); + msock.socket = s; + goto randomize; } - msock = s; - set_bit(DISCARD_CONCURRENT, &mdev->flags); + msock.socket = s; break; default: - dev_warn(DEV, "Error receiving initial packet\n"); + conn_warn(tconn, "Error receiving initial packet\n"); sock_release(s); +randomize: if (random32() & 1) goto retry; } } - if (mdev->state.conn <= C_DISCONNECTING) + if (tconn->cstate <= C_DISCONNECTING) goto out_release_sockets; if (signal_pending(current)) { flush_signals(current); smp_rmb(); - if (get_t_state(&mdev->receiver) == Exiting) + if (get_t_state(&tconn->receiver) == EXITING) goto out_release_sockets; } - if (sock && msock) { - ok = drbd_socket_okay(mdev, &sock); - ok = drbd_socket_okay(mdev, &msock) && ok; - if (ok) - break; - } - } while (1); + ok = drbd_socket_okay(&sock.socket); + ok = drbd_socket_okay(&msock.socket) && ok; + } while (!ok); + + if (ad.s_listen) + sock_release(ad.s_listen); - msock->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ - sock->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ + sock.socket->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ + msock.socket->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ - sock->sk->sk_allocation = GFP_NOIO; - msock->sk->sk_allocation = GFP_NOIO; + sock.socket->sk->sk_allocation = GFP_NOIO; + msock.socket->sk->sk_allocation = GFP_NOIO; - sock->sk->sk_priority = TC_PRIO_INTERACTIVE_BULK; - msock->sk->sk_priority = TC_PRIO_INTERACTIVE; + sock.socket->sk->sk_priority = TC_PRIO_INTERACTIVE_BULK; + msock.socket->sk->sk_priority = TC_PRIO_INTERACTIVE; /* NOT YET ... - * sock->sk->sk_sndtimeo = mdev->net_conf->timeout*HZ/10; - * sock->sk->sk_rcvtimeo = MAX_SCHEDULE_TIMEOUT; - * first set it to the P_HAND_SHAKE timeout, + * sock.socket->sk->sk_sndtimeo = tconn->net_conf->timeout*HZ/10; + * sock.socket->sk->sk_rcvtimeo = MAX_SCHEDULE_TIMEOUT; + * first set it to the P_CONNECTION_FEATURES timeout, * which we set to 4x the configured ping_timeout. */ - sock->sk->sk_sndtimeo = - sock->sk->sk_rcvtimeo = mdev->net_conf->ping_timeo*4*HZ/10; + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); - msock->sk->sk_sndtimeo = mdev->net_conf->timeout*HZ/10; - msock->sk->sk_rcvtimeo = mdev->net_conf->ping_int*HZ; + sock.socket->sk->sk_sndtimeo = + sock.socket->sk->sk_rcvtimeo = nc->ping_timeo*4*HZ/10; + + msock.socket->sk->sk_rcvtimeo = nc->ping_int*HZ; + timeout = nc->timeout * HZ / 10; + discard_my_data = nc->discard_my_data; + rcu_read_unlock(); + + msock.socket->sk->sk_sndtimeo = timeout; /* we don't want delays. * we use TCP_CORK where appropriate, though */ - drbd_tcp_nodelay(sock); - drbd_tcp_nodelay(msock); - - mdev->data.socket = sock; - mdev->meta.socket = msock; - mdev->last_received = jiffies; + drbd_tcp_nodelay(sock.socket); + drbd_tcp_nodelay(msock.socket); - D_ASSERT(mdev->asender.task == NULL); + tconn->data.socket = sock.socket; + tconn->meta.socket = msock.socket; + tconn->last_received = jiffies; - h = drbd_do_handshake(mdev); + h = drbd_do_features(tconn); if (h <= 0) return h; - if (mdev->cram_hmac_tfm) { + if (tconn->cram_hmac_tfm) { /* drbd_request_state(mdev, NS(conn, WFAuth)); */ - switch (drbd_do_auth(mdev)) { + switch (drbd_do_auth(tconn)) { case -1: - dev_err(DEV, "Authentication of peer failed\n"); + conn_err(tconn, "Authentication of peer failed\n"); return -1; case 0: - dev_err(DEV, "Authentication of peer failed, trying again.\n"); + conn_err(tconn, "Authentication of peer failed, trying again.\n"); return 0; } } - sock->sk->sk_sndtimeo = mdev->net_conf->timeout*HZ/10; - sock->sk->sk_rcvtimeo = MAX_SCHEDULE_TIMEOUT; + tconn->data.socket->sk->sk_sndtimeo = timeout; + tconn->data.socket->sk->sk_rcvtimeo = MAX_SCHEDULE_TIMEOUT; - atomic_set(&mdev->packet_seq, 0); - mdev->peer_seq = 0; - - if (drbd_send_protocol(mdev) == -1) + if (drbd_send_protocol(tconn) == -EOPNOTSUPP) return -1; - set_bit(STATE_SENT, &mdev->flags); - drbd_send_sync_param(mdev, &mdev->sync_conf); - drbd_send_sizes(mdev, 0, 0); - drbd_send_uuids(mdev); - drbd_send_current_state(mdev); - clear_bit(USE_DEGR_WFC_T, &mdev->flags); - clear_bit(RESIZE_PENDING, &mdev->flags); - spin_lock_irq(&mdev->req_lock); - rv = _drbd_set_state(_NS(mdev, conn, C_WF_REPORT_PARAMS), CS_VERBOSE, NULL); - if (mdev->state.conn != C_WF_REPORT_PARAMS) - clear_bit(STATE_SENT, &mdev->flags); - spin_unlock_irq(&mdev->req_lock); + set_bit(STATE_SENT, &tconn->flags); + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + kref_get(&mdev->kref); + /* Prevent a race between resync-handshake and + * being promoted to Primary. + * + * Grab and release the state mutex, so we know that any current + * drbd_set_role() is finished, and any incoming drbd_set_role + * will see the STATE_SENT flag, and wait for it to be cleared. + */ + mutex_lock(mdev->state_mutex); + mutex_unlock(mdev->state_mutex); + + rcu_read_unlock(); + + if (discard_my_data) + set_bit(DISCARD_MY_DATA, &mdev->flags); + else + clear_bit(DISCARD_MY_DATA, &mdev->flags); + + drbd_connected(mdev); + kref_put(&mdev->kref, &drbd_minor_destroy); + rcu_read_lock(); + } + rcu_read_unlock(); - if (rv < SS_SUCCESS) + rv = conn_request_state(tconn, NS(conn, C_WF_REPORT_PARAMS), CS_VERBOSE); + if (rv < SS_SUCCESS || tconn->cstate != C_WF_REPORT_PARAMS) { + clear_bit(STATE_SENT, &tconn->flags); return 0; + } - drbd_thread_start(&mdev->asender); - mod_timer(&mdev->request_timer, jiffies + HZ); /* just start it here. */ + drbd_thread_start(&tconn->asender); - return 1; + mutex_lock(&tconn->conf_update); + /* The discard_my_data flag is a single-shot modifier to the next + * connection attempt, the handshake of which is now well underway. + * No need for rcu style copying of the whole struct + * just to clear a single value. */ + tconn->net_conf->discard_my_data = 0; + mutex_unlock(&tconn->conf_update); + + return h; out_release_sockets: - if (sock) - sock_release(sock); - if (msock) - sock_release(msock); + if (ad.s_listen) + sock_release(ad.s_listen); + if (sock.socket) + sock_release(sock.socket); + if (msock.socket) + sock_release(msock.socket); return -1; } -static int drbd_recv_header(struct drbd_conf *mdev, enum drbd_packets *cmd, unsigned int *packet_size) +static int decode_header(struct drbd_tconn *tconn, void *header, struct packet_info *pi) { - union p_header *h = &mdev->data.rbuf.header; - int r; - - r = drbd_recv(mdev, h, sizeof(*h)); - if (unlikely(r != sizeof(*h))) { - if (!signal_pending(current)) - dev_warn(DEV, "short read expecting header on sock: r=%d\n", r); - return false; - } - - if (likely(h->h80.magic == BE_DRBD_MAGIC)) { - *cmd = be16_to_cpu(h->h80.command); - *packet_size = be16_to_cpu(h->h80.length); - } else if (h->h95.magic == BE_DRBD_MAGIC_BIG) { - *cmd = be16_to_cpu(h->h95.command); - *packet_size = be32_to_cpu(h->h95.length); + unsigned int header_size = drbd_header_size(tconn); + + if (header_size == sizeof(struct p_header100) && + *(__be32 *)header == cpu_to_be32(DRBD_MAGIC_100)) { + struct p_header100 *h = header; + if (h->pad != 0) { + conn_err(tconn, "Header padding is not zero\n"); + return -EINVAL; + } + pi->vnr = be16_to_cpu(h->volume); + pi->cmd = be16_to_cpu(h->command); + pi->size = be32_to_cpu(h->length); + } else if (header_size == sizeof(struct p_header95) && + *(__be16 *)header == cpu_to_be16(DRBD_MAGIC_BIG)) { + struct p_header95 *h = header; + pi->cmd = be16_to_cpu(h->command); + pi->size = be32_to_cpu(h->length); + pi->vnr = 0; + } else if (header_size == sizeof(struct p_header80) && + *(__be32 *)header == cpu_to_be32(DRBD_MAGIC)) { + struct p_header80 *h = header; + pi->cmd = be16_to_cpu(h->command); + pi->size = be16_to_cpu(h->length); + pi->vnr = 0; } else { - dev_err(DEV, "magic?? on data m: 0x%08x c: %d l: %d\n", - be32_to_cpu(h->h80.magic), - be16_to_cpu(h->h80.command), - be16_to_cpu(h->h80.length)); - return false; + conn_err(tconn, "Wrong magic value 0x%08x in protocol version %d\n", + be32_to_cpu(*(__be32 *)header), + tconn->agreed_pro_version); + return -EINVAL; } - mdev->last_received = jiffies; + pi->data = header + header_size; + return 0; +} - return true; +static int drbd_recv_header(struct drbd_tconn *tconn, struct packet_info *pi) +{ + void *buffer = tconn->data.rbuf; + int err; + + err = drbd_recv_all_warn(tconn, buffer, drbd_header_size(tconn)); + if (err) + return err; + + err = decode_header(tconn, buffer, pi); + tconn->last_received = jiffies; + + return err; } -static void drbd_flush(struct drbd_conf *mdev) +static void drbd_flush(struct drbd_tconn *tconn) { int rv; + struct drbd_conf *mdev; + int vnr; - if (mdev->write_ordering >= WO_bdev_flush && get_ldev(mdev)) { - rv = blkdev_issue_flush(mdev->ldev->backing_bdev, GFP_KERNEL, - NULL); - if (rv) { - dev_info(DEV, "local disk flush failed with status %d\n", rv); - /* would rather check on EOPNOTSUPP, but that is not reliable. - * don't try again for ANY return value != 0 - * if (rv == -EOPNOTSUPP) */ - drbd_bump_write_ordering(mdev, WO_drain_io); + if (tconn->write_ordering >= WO_bdev_flush) { + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + if (!get_ldev(mdev)) + continue; + kref_get(&mdev->kref); + rcu_read_unlock(); + + rv = blkdev_issue_flush(mdev->ldev->backing_bdev, + GFP_NOIO, NULL); + if (rv) { + dev_info(DEV, "local disk flush failed with status %d\n", rv); + /* would rather check on EOPNOTSUPP, but that is not reliable. + * don't try again for ANY return value != 0 + * if (rv == -EOPNOTSUPP) */ + drbd_bump_write_ordering(tconn, WO_drain_io); + } + put_ldev(mdev); + kref_put(&mdev->kref, &drbd_minor_destroy); + + rcu_read_lock(); + if (rv) + break; } - put_ldev(mdev); + rcu_read_unlock(); } } @@ -987,7 +1179,7 @@ static void drbd_flush(struct drbd_conf *mdev) * @epoch: Epoch object. * @ev: Epoch event. */ -static enum finish_epoch drbd_may_finish_epoch(struct drbd_conf *mdev, +static enum finish_epoch drbd_may_finish_epoch(struct drbd_tconn *tconn, struct drbd_epoch *epoch, enum epoch_event ev) { @@ -995,7 +1187,7 @@ static enum finish_epoch drbd_may_finish_epoch(struct drbd_conf *mdev, struct drbd_epoch *next_epoch; enum finish_epoch rv = FE_STILL_LIVE; - spin_lock(&mdev->epoch_lock); + spin_lock(&tconn->epoch_lock); do { next_epoch = NULL; @@ -1017,18 +1209,22 @@ static enum finish_epoch drbd_may_finish_epoch(struct drbd_conf *mdev, atomic_read(&epoch->active) == 0 && (test_bit(DE_HAVE_BARRIER_NUMBER, &epoch->flags) || ev & EV_CLEANUP)) { if (!(ev & EV_CLEANUP)) { - spin_unlock(&mdev->epoch_lock); - drbd_send_b_ack(mdev, epoch->barrier_nr, epoch_size); - spin_lock(&mdev->epoch_lock); + spin_unlock(&tconn->epoch_lock); + drbd_send_b_ack(epoch->tconn, epoch->barrier_nr, epoch_size); + spin_lock(&tconn->epoch_lock); } +#if 0 + /* FIXME: dec unacked on connection, once we have + * something to count pending connection packets in. */ if (test_bit(DE_HAVE_BARRIER_NUMBER, &epoch->flags)) - dec_unacked(mdev); + dec_unacked(epoch->tconn); +#endif - if (mdev->current_epoch != epoch) { + if (tconn->current_epoch != epoch) { next_epoch = list_entry(epoch->list.next, struct drbd_epoch, list); list_del(&epoch->list); ev = EV_BECAME_LAST | (ev & EV_CLEANUP); - mdev->epochs--; + tconn->epochs--; kfree(epoch); if (rv == FE_STILL_LIVE) @@ -1039,7 +1235,6 @@ static enum finish_epoch drbd_may_finish_epoch(struct drbd_conf *mdev, /* atomic_set(&epoch->active, 0); is already zero */ if (rv == FE_STILL_LIVE) rv = FE_RECYCLED; - wake_up(&mdev->ee_wait); } } @@ -1049,40 +1244,52 @@ static enum finish_epoch drbd_may_finish_epoch(struct drbd_conf *mdev, epoch = next_epoch; } while (1); - spin_unlock(&mdev->epoch_lock); + spin_unlock(&tconn->epoch_lock); return rv; } /** * drbd_bump_write_ordering() - Fall back to an other write ordering method - * @mdev: DRBD device. + * @tconn: DRBD connection. * @wo: Write ordering method to try. */ -void drbd_bump_write_ordering(struct drbd_conf *mdev, enum write_ordering_e wo) __must_hold(local) +void drbd_bump_write_ordering(struct drbd_tconn *tconn, enum write_ordering_e wo) { + struct disk_conf *dc; + struct drbd_conf *mdev; enum write_ordering_e pwo; + int vnr; static char *write_ordering_str[] = { [WO_none] = "none", [WO_drain_io] = "drain", [WO_bdev_flush] = "flush", }; - pwo = mdev->write_ordering; + pwo = tconn->write_ordering; wo = min(pwo, wo); - if (wo == WO_bdev_flush && mdev->ldev->dc.no_disk_flush) - wo = WO_drain_io; - if (wo == WO_drain_io && mdev->ldev->dc.no_disk_drain) - wo = WO_none; - mdev->write_ordering = wo; - if (pwo != mdev->write_ordering || wo == WO_bdev_flush) - dev_info(DEV, "Method to ensure write ordering: %s\n", write_ordering_str[mdev->write_ordering]); + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + if (!get_ldev_if_state(mdev, D_ATTACHING)) + continue; + dc = rcu_dereference(mdev->ldev->disk_conf); + + if (wo == WO_bdev_flush && !dc->disk_flushes) + wo = WO_drain_io; + if (wo == WO_drain_io && !dc->disk_drain) + wo = WO_none; + put_ldev(mdev); + } + rcu_read_unlock(); + tconn->write_ordering = wo; + if (pwo != tconn->write_ordering || wo == WO_bdev_flush) + conn_info(tconn, "Method to ensure write ordering: %s\n", write_ordering_str[tconn->write_ordering]); } /** - * drbd_submit_ee() + * drbd_submit_peer_request() * @mdev: DRBD device. - * @e: epoch entry + * @peer_req: peer request * @rw: flag field, see bio->bi_rw * * May spread the pages to multiple bios, @@ -1096,14 +1303,15 @@ void drbd_bump_write_ordering(struct drbd_conf *mdev, enum write_ordering_e wo) * on certain Xen deployments. */ /* TODO allocate from our own bio_set. */ -int drbd_submit_ee(struct drbd_conf *mdev, struct drbd_epoch_entry *e, - const unsigned rw, const int fault_type) +int drbd_submit_peer_request(struct drbd_conf *mdev, + struct drbd_peer_request *peer_req, + const unsigned rw, const int fault_type) { struct bio *bios = NULL; struct bio *bio; - struct page *page = e->pages; - sector_t sector = e->sector; - unsigned ds = e->size; + struct page *page = peer_req->pages; + sector_t sector = peer_req->i.sector; + unsigned ds = peer_req->i.size; unsigned n_bios = 0; unsigned nr_pages = (ds + PAGE_SIZE -1) >> PAGE_SHIFT; int err = -ENOMEM; @@ -1122,12 +1330,12 @@ next_bio: dev_err(DEV, "submit_ee: Allocation of a bio failed\n"); goto fail; } - /* > e->sector, unless this is the first bio */ + /* > peer_req->i.sector, unless this is the first bio */ bio->bi_sector = sector; bio->bi_bdev = mdev->ldev->backing_bdev; bio->bi_rw = rw; - bio->bi_private = e; - bio->bi_end_io = drbd_endio_sec; + bio->bi_private = peer_req; + bio->bi_end_io = drbd_peer_request_endio; bio->bi_next = bios; bios = bio; @@ -1156,7 +1364,7 @@ next_bio: D_ASSERT(page == NULL); D_ASSERT(ds == 0); - atomic_set(&e->pending_bios, n_bios); + atomic_set(&peer_req->pending_bios, n_bios); do { bio = bios; bios = bios->bi_next; @@ -1175,26 +1383,57 @@ fail: return err; } -static int receive_Barrier(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static void drbd_remove_epoch_entry_interval(struct drbd_conf *mdev, + struct drbd_peer_request *peer_req) +{ + struct drbd_interval *i = &peer_req->i; + + drbd_remove_interval(&mdev->write_requests, i); + drbd_clear_interval(i); + + /* Wake up any processes waiting for this peer request to complete. */ + if (i->waiting) + wake_up(&mdev->misc_wait); +} + +void conn_wait_active_ee_empty(struct drbd_tconn *tconn) +{ + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + kref_get(&mdev->kref); + rcu_read_unlock(); + drbd_wait_ee_list_empty(mdev, &mdev->active_ee); + kref_put(&mdev->kref, &drbd_minor_destroy); + rcu_read_lock(); + } + rcu_read_unlock(); +} + +static int receive_Barrier(struct drbd_tconn *tconn, struct packet_info *pi) { int rv; - struct p_barrier *p = &mdev->data.rbuf.barrier; + struct p_barrier *p = pi->data; struct drbd_epoch *epoch; - inc_unacked(mdev); - - mdev->current_epoch->barrier_nr = p->barrier; - rv = drbd_may_finish_epoch(mdev, mdev->current_epoch, EV_GOT_BARRIER_NR); + /* FIXME these are unacked on connection, + * not a specific (peer)device. + */ + tconn->current_epoch->barrier_nr = p->barrier; + tconn->current_epoch->tconn = tconn; + rv = drbd_may_finish_epoch(tconn, tconn->current_epoch, EV_GOT_BARRIER_NR); /* P_BARRIER_ACK may imply that the corresponding extent is dropped from * the activity log, which means it would not be resynced in case the * R_PRIMARY crashes now. * Therefore we must send the barrier_ack after the barrier request was * completed. */ - switch (mdev->write_ordering) { + switch (tconn->write_ordering) { case WO_none: if (rv == FE_RECYCLED) - return true; + return 0; /* receiver context, in the writeout path of the other node. * avoid potential distributed deadlock */ @@ -1202,81 +1441,75 @@ static int receive_Barrier(struct drbd_conf *mdev, enum drbd_packets cmd, unsign if (epoch) break; else - dev_warn(DEV, "Allocation of an epoch failed, slowing down\n"); + conn_warn(tconn, "Allocation of an epoch failed, slowing down\n"); /* Fall through */ case WO_bdev_flush: case WO_drain_io: - drbd_wait_ee_list_empty(mdev, &mdev->active_ee); - drbd_flush(mdev); + conn_wait_active_ee_empty(tconn); + drbd_flush(tconn); - if (atomic_read(&mdev->current_epoch->epoch_size)) { + if (atomic_read(&tconn->current_epoch->epoch_size)) { epoch = kmalloc(sizeof(struct drbd_epoch), GFP_NOIO); if (epoch) break; } - epoch = mdev->current_epoch; - wait_event(mdev->ee_wait, atomic_read(&epoch->epoch_size) == 0); - - D_ASSERT(atomic_read(&epoch->active) == 0); - D_ASSERT(epoch->flags == 0); - - return true; + return 0; default: - dev_err(DEV, "Strangeness in mdev->write_ordering %d\n", mdev->write_ordering); - return false; + conn_err(tconn, "Strangeness in tconn->write_ordering %d\n", tconn->write_ordering); + return -EIO; } epoch->flags = 0; atomic_set(&epoch->epoch_size, 0); atomic_set(&epoch->active, 0); - spin_lock(&mdev->epoch_lock); - if (atomic_read(&mdev->current_epoch->epoch_size)) { - list_add(&epoch->list, &mdev->current_epoch->list); - mdev->current_epoch = epoch; - mdev->epochs++; + spin_lock(&tconn->epoch_lock); + if (atomic_read(&tconn->current_epoch->epoch_size)) { + list_add(&epoch->list, &tconn->current_epoch->list); + tconn->current_epoch = epoch; + tconn->epochs++; } else { /* The current_epoch got recycled while we allocated this one... */ kfree(epoch); } - spin_unlock(&mdev->epoch_lock); + spin_unlock(&tconn->epoch_lock); - return true; + return 0; } /* used from receive_RSDataReply (recv_resync_read) * and from receive_Data */ -static struct drbd_epoch_entry * -read_in_block(struct drbd_conf *mdev, u64 id, sector_t sector, int data_size) __must_hold(local) +static struct drbd_peer_request * +read_in_block(struct drbd_conf *mdev, u64 id, sector_t sector, + int data_size) __must_hold(local) { const sector_t capacity = drbd_get_capacity(mdev->this_bdev); - struct drbd_epoch_entry *e; + struct drbd_peer_request *peer_req; struct page *page; - int dgs, ds, rr; - void *dig_in = mdev->int_dig_in; - void *dig_vv = mdev->int_dig_vv; + int dgs, ds, err; + void *dig_in = mdev->tconn->int_dig_in; + void *dig_vv = mdev->tconn->int_dig_vv; unsigned long *data; - dgs = (mdev->agreed_pro_version >= 87 && mdev->integrity_r_tfm) ? - crypto_hash_digestsize(mdev->integrity_r_tfm) : 0; - - if (dgs) { - rr = drbd_recv(mdev, dig_in, dgs); - if (rr != dgs) { - if (!signal_pending(current)) - dev_warn(DEV, - "short read receiving data digest: read %d expected %d\n", - rr, dgs); + dgs = 0; + if (mdev->tconn->peer_integrity_tfm) { + dgs = crypto_hash_digestsize(mdev->tconn->peer_integrity_tfm); + /* + * FIXME: Receive the incoming digest into the receive buffer + * here, together with its struct p_data? + */ + err = drbd_recv_all_warn(mdev->tconn, dig_in, dgs); + if (err) return NULL; - } + data_size -= dgs; } - data_size -= dgs; - - ERR_IF(data_size & 0x1ff) return NULL; - ERR_IF(data_size > DRBD_MAX_BIO_SIZE) return NULL; + if (!expect(IS_ALIGNED(data_size, 512))) + return NULL; + if (!expect(data_size <= DRBD_MAX_BIO_SIZE)) + return NULL; /* even though we trust out peer, * we sometimes have to double check. */ @@ -1291,47 +1524,42 @@ read_in_block(struct drbd_conf *mdev, u64 id, sector_t sector, int data_size) __ /* GFP_NOIO, because we must not cause arbitrary write-out: in a DRBD * "criss-cross" setup, that might cause write-out on some other DRBD, * which in turn might block on the other node at this very place. */ - e = drbd_alloc_ee(mdev, id, sector, data_size, GFP_NOIO); - if (!e) + peer_req = drbd_alloc_peer_req(mdev, id, sector, data_size, GFP_NOIO); + if (!peer_req) return NULL; if (!data_size) - return e; + return peer_req; ds = data_size; - page = e->pages; + page = peer_req->pages; page_chain_for_each(page) { unsigned len = min_t(int, ds, PAGE_SIZE); data = kmap(page); - rr = drbd_recv(mdev, data, len); + err = drbd_recv_all_warn(mdev->tconn, data, len); if (drbd_insert_fault(mdev, DRBD_FAULT_RECEIVE)) { dev_err(DEV, "Fault injection: Corrupting data on receive\n"); data[0] = data[0] ^ (unsigned long)-1; } kunmap(page); - if (rr != len) { - drbd_free_ee(mdev, e); - if (!signal_pending(current)) - dev_warn(DEV, "short read receiving data: read %d expected %d\n", - rr, len); + if (err) { + drbd_free_peer_req(mdev, peer_req); return NULL; } - ds -= rr; + ds -= len; } if (dgs) { - drbd_csum_ee(mdev, mdev->integrity_r_tfm, e, dig_vv); + drbd_csum_ee(mdev, mdev->tconn->peer_integrity_tfm, peer_req, dig_vv); if (memcmp(dig_in, dig_vv, dgs)) { dev_err(DEV, "Digest integrity check FAILED: %llus +%u\n", (unsigned long long)sector, data_size); - drbd_bcast_ee(mdev, "digest failed", - dgs, dig_in, dig_vv, e); - drbd_free_ee(mdev, e); + drbd_free_peer_req(mdev, peer_req); return NULL; } } mdev->recv_cnt += data_size>>9; - return e; + return peer_req; } /* drbd_drain_block() just takes a data block @@ -1340,30 +1568,26 @@ read_in_block(struct drbd_conf *mdev, u64 id, sector_t sector, int data_size) __ static int drbd_drain_block(struct drbd_conf *mdev, int data_size) { struct page *page; - int rr, rv = 1; + int err = 0; void *data; if (!data_size) - return true; + return 0; - page = drbd_pp_alloc(mdev, 1, 1); + page = drbd_alloc_pages(mdev, 1, 1); data = kmap(page); while (data_size) { - rr = drbd_recv(mdev, data, min_t(int, data_size, PAGE_SIZE)); - if (rr != min_t(int, data_size, PAGE_SIZE)) { - rv = 0; - if (!signal_pending(current)) - dev_warn(DEV, - "short read receiving data: read %d expected %d\n", - rr, min_t(int, data_size, PAGE_SIZE)); + unsigned int len = min_t(int, data_size, PAGE_SIZE); + + err = drbd_recv_all_warn(mdev->tconn, data, len); + if (err) break; - } - data_size -= rr; + data_size -= len; } kunmap(page); - drbd_pp_free(mdev, page, 0); - return rv; + drbd_free_pages(mdev, page, 0); + return err; } static int recv_dless_read(struct drbd_conf *mdev, struct drbd_request *req, @@ -1371,26 +1595,19 @@ static int recv_dless_read(struct drbd_conf *mdev, struct drbd_request *req, { struct bio_vec *bvec; struct bio *bio; - int dgs, rr, i, expect; - void *dig_in = mdev->int_dig_in; - void *dig_vv = mdev->int_dig_vv; - - dgs = (mdev->agreed_pro_version >= 87 && mdev->integrity_r_tfm) ? - crypto_hash_digestsize(mdev->integrity_r_tfm) : 0; + int dgs, err, i, expect; + void *dig_in = mdev->tconn->int_dig_in; + void *dig_vv = mdev->tconn->int_dig_vv; - if (dgs) { - rr = drbd_recv(mdev, dig_in, dgs); - if (rr != dgs) { - if (!signal_pending(current)) - dev_warn(DEV, - "short read receiving data reply digest: read %d expected %d\n", - rr, dgs); - return 0; - } + dgs = 0; + if (mdev->tconn->peer_integrity_tfm) { + dgs = crypto_hash_digestsize(mdev->tconn->peer_integrity_tfm); + err = drbd_recv_all_warn(mdev->tconn, dig_in, dgs); + if (err) + return err; + data_size -= dgs; } - data_size -= dgs; - /* optimistically update recv_cnt. if receiving fails below, * we disconnect anyways, and counters will be reset. */ mdev->recv_cnt += data_size>>9; @@ -1399,63 +1616,61 @@ static int recv_dless_read(struct drbd_conf *mdev, struct drbd_request *req, D_ASSERT(sector == bio->bi_sector); bio_for_each_segment(bvec, bio, i) { + void *mapped = kmap(bvec->bv_page) + bvec->bv_offset; expect = min_t(int, data_size, bvec->bv_len); - rr = drbd_recv(mdev, - kmap(bvec->bv_page)+bvec->bv_offset, - expect); + err = drbd_recv_all_warn(mdev->tconn, mapped, expect); kunmap(bvec->bv_page); - if (rr != expect) { - if (!signal_pending(current)) - dev_warn(DEV, "short read receiving data reply: " - "read %d expected %d\n", - rr, expect); - return 0; - } - data_size -= rr; + if (err) + return err; + data_size -= expect; } if (dgs) { - drbd_csum_bio(mdev, mdev->integrity_r_tfm, bio, dig_vv); + drbd_csum_bio(mdev, mdev->tconn->peer_integrity_tfm, bio, dig_vv); if (memcmp(dig_in, dig_vv, dgs)) { dev_err(DEV, "Digest integrity check FAILED. Broken NICs?\n"); - return 0; + return -EINVAL; } } D_ASSERT(data_size == 0); - return 1; + return 0; } -/* e_end_resync_block() is called via - * drbd_process_done_ee() by asender only */ -static int e_end_resync_block(struct drbd_conf *mdev, struct drbd_work *w, int unused) +/* + * e_end_resync_block() is called in asender context via + * drbd_finish_peer_reqs(). + */ +static int e_end_resync_block(struct drbd_work *w, int unused) { - struct drbd_epoch_entry *e = (struct drbd_epoch_entry *)w; - sector_t sector = e->sector; - int ok; + struct drbd_peer_request *peer_req = + container_of(w, struct drbd_peer_request, w); + struct drbd_conf *mdev = w->mdev; + sector_t sector = peer_req->i.sector; + int err; - D_ASSERT(hlist_unhashed(&e->collision)); + D_ASSERT(drbd_interval_empty(&peer_req->i)); - if (likely((e->flags & EE_WAS_ERROR) == 0)) { - drbd_set_in_sync(mdev, sector, e->size); - ok = drbd_send_ack(mdev, P_RS_WRITE_ACK, e); + if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) { + drbd_set_in_sync(mdev, sector, peer_req->i.size); + err = drbd_send_ack(mdev, P_RS_WRITE_ACK, peer_req); } else { /* Record failure to sync */ - drbd_rs_failed_io(mdev, sector, e->size); + drbd_rs_failed_io(mdev, sector, peer_req->i.size); - ok = drbd_send_ack(mdev, P_NEG_ACK, e); + err = drbd_send_ack(mdev, P_NEG_ACK, peer_req); } dec_unacked(mdev); - return ok; + return err; } static int recv_resync_read(struct drbd_conf *mdev, sector_t sector, int data_size) __releases(local) { - struct drbd_epoch_entry *e; + struct drbd_peer_request *peer_req; - e = read_in_block(mdev, ID_SYNCER, sector, data_size); - if (!e) + peer_req = read_in_block(mdev, ID_SYNCER, sector, data_size); + if (!peer_req) goto fail; dec_rs_pending(mdev); @@ -1464,64 +1679,88 @@ static int recv_resync_read(struct drbd_conf *mdev, sector_t sector, int data_si /* corresponding dec_unacked() in e_end_resync_block() * respective _drbd_clear_done_ee */ - e->w.cb = e_end_resync_block; + peer_req->w.cb = e_end_resync_block; - spin_lock_irq(&mdev->req_lock); - list_add(&e->w.list, &mdev->sync_ee); - spin_unlock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); + list_add(&peer_req->w.list, &mdev->sync_ee); + spin_unlock_irq(&mdev->tconn->req_lock); atomic_add(data_size >> 9, &mdev->rs_sect_ev); - if (drbd_submit_ee(mdev, e, WRITE, DRBD_FAULT_RS_WR) == 0) - return true; + if (drbd_submit_peer_request(mdev, peer_req, WRITE, DRBD_FAULT_RS_WR) == 0) + return 0; /* don't care for the reason here */ dev_err(DEV, "submit failed, triggering re-connect\n"); - spin_lock_irq(&mdev->req_lock); - list_del(&e->w.list); - spin_unlock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); + list_del(&peer_req->w.list); + spin_unlock_irq(&mdev->tconn->req_lock); - drbd_free_ee(mdev, e); + drbd_free_peer_req(mdev, peer_req); fail: put_ldev(mdev); - return false; + return -EIO; } -static int receive_DataReply(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static struct drbd_request * +find_request(struct drbd_conf *mdev, struct rb_root *root, u64 id, + sector_t sector, bool missing_ok, const char *func) { struct drbd_request *req; + + /* Request object according to our peer */ + req = (struct drbd_request *)(unsigned long)id; + if (drbd_contains_interval(root, sector, &req->i) && req->i.local) + return req; + if (!missing_ok) { + dev_err(DEV, "%s: failed to find request 0x%lx, sector %llus\n", func, + (unsigned long)id, (unsigned long long)sector); + } + return NULL; +} + +static int receive_DataReply(struct drbd_tconn *tconn, struct packet_info *pi) +{ + struct drbd_conf *mdev; + struct drbd_request *req; sector_t sector; - int ok; - struct p_data *p = &mdev->data.rbuf.data; + int err; + struct p_data *p = pi->data; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; sector = be64_to_cpu(p->sector); - spin_lock_irq(&mdev->req_lock); - req = _ar_id_to_req(mdev, p->block_id, sector); - spin_unlock_irq(&mdev->req_lock); - if (unlikely(!req)) { - dev_err(DEV, "Got a corrupt block_id/sector pair(1).\n"); - return false; - } + spin_lock_irq(&mdev->tconn->req_lock); + req = find_request(mdev, &mdev->read_requests, p->block_id, sector, false, __func__); + spin_unlock_irq(&mdev->tconn->req_lock); + if (unlikely(!req)) + return -EIO; /* hlist_del(&req->collision) is done in _req_may_be_done, to avoid * special casing it there for the various failure cases. * still no race with drbd_fail_pending_reads */ - ok = recv_dless_read(mdev, req, sector, data_size); - - if (ok) - req_mod(req, data_received); + err = recv_dless_read(mdev, req, sector, pi->size); + if (!err) + req_mod(req, DATA_RECEIVED); /* else: nothing. handled from drbd_disconnect... * I don't think we may complete this just yet * in case we are "on-disconnect: freeze" */ - return ok; + return err; } -static int receive_RSDataReply(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_RSDataReply(struct drbd_tconn *tconn, struct packet_info *pi) { + struct drbd_conf *mdev; sector_t sector; - int ok; - struct p_data *p = &mdev->data.rbuf.data; + int err; + struct p_data *p = pi->data; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; sector = be64_to_cpu(p->sector); D_ASSERT(p->block_id == ID_SYNCER); @@ -1529,42 +1768,63 @@ static int receive_RSDataReply(struct drbd_conf *mdev, enum drbd_packets cmd, un if (get_ldev(mdev)) { /* data is submitted to disk within recv_resync_read. * corresponding put_ldev done below on error, - * or in drbd_endio_write_sec. */ - ok = recv_resync_read(mdev, sector, data_size); + * or in drbd_peer_request_endio. */ + err = recv_resync_read(mdev, sector, pi->size); } else { if (__ratelimit(&drbd_ratelimit_state)) dev_err(DEV, "Can not write resync data to local disk.\n"); - ok = drbd_drain_block(mdev, data_size); + err = drbd_drain_block(mdev, pi->size); - drbd_send_ack_dp(mdev, P_NEG_ACK, p, data_size); + drbd_send_ack_dp(mdev, P_NEG_ACK, p, pi->size); } - atomic_add(data_size >> 9, &mdev->rs_sect_in); + atomic_add(pi->size >> 9, &mdev->rs_sect_in); - return ok; + return err; } -/* e_end_block() is called via drbd_process_done_ee(). - * this means this function only runs in the asender thread - */ -static int e_end_block(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +static void restart_conflicting_writes(struct drbd_conf *mdev, + sector_t sector, int size) { - struct drbd_epoch_entry *e = (struct drbd_epoch_entry *)w; - sector_t sector = e->sector; - int ok = 1, pcmd; + struct drbd_interval *i; + struct drbd_request *req; - if (mdev->net_conf->wire_protocol == DRBD_PROT_C) { - if (likely((e->flags & EE_WAS_ERROR) == 0)) { + drbd_for_each_overlap(i, &mdev->write_requests, sector, size) { + if (!i->local) + continue; + req = container_of(i, struct drbd_request, i); + if (req->rq_state & RQ_LOCAL_PENDING || + !(req->rq_state & RQ_POSTPONED)) + continue; + /* as it is RQ_POSTPONED, this will cause it to + * be queued on the retry workqueue. */ + __req_mod(req, CONFLICT_RESOLVED, NULL); + } +} + +/* + * e_end_block() is called in asender context via drbd_finish_peer_reqs(). + */ +static int e_end_block(struct drbd_work *w, int cancel) +{ + struct drbd_peer_request *peer_req = + container_of(w, struct drbd_peer_request, w); + struct drbd_conf *mdev = w->mdev; + sector_t sector = peer_req->i.sector; + int err = 0, pcmd; + + if (peer_req->flags & EE_SEND_WRITE_ACK) { + if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) { pcmd = (mdev->state.conn >= C_SYNC_SOURCE && mdev->state.conn <= C_PAUSED_SYNC_T && - e->flags & EE_MAY_SET_IN_SYNC) ? + peer_req->flags & EE_MAY_SET_IN_SYNC) ? P_RS_WRITE_ACK : P_WRITE_ACK; - ok &= drbd_send_ack(mdev, pcmd, e); + err = drbd_send_ack(mdev, pcmd, peer_req); if (pcmd == P_RS_WRITE_ACK) - drbd_set_in_sync(mdev, sector, e->size); + drbd_set_in_sync(mdev, sector, peer_req->i.size); } else { - ok = drbd_send_ack(mdev, P_NEG_ACK, e); + err = drbd_send_ack(mdev, P_NEG_ACK, peer_req); /* we expect it to be marked out of sync anyways... * maybe assert this? */ } @@ -1572,52 +1832,115 @@ static int e_end_block(struct drbd_conf *mdev, struct drbd_work *w, int cancel) } /* we delete from the conflict detection hash _after_ we sent out the * P_WRITE_ACK / P_NEG_ACK, to get the sequence number right. */ - if (mdev->net_conf->two_primaries) { - spin_lock_irq(&mdev->req_lock); - D_ASSERT(!hlist_unhashed(&e->collision)); - hlist_del_init(&e->collision); - spin_unlock_irq(&mdev->req_lock); - } else { - D_ASSERT(hlist_unhashed(&e->collision)); - } + if (peer_req->flags & EE_IN_INTERVAL_TREE) { + spin_lock_irq(&mdev->tconn->req_lock); + D_ASSERT(!drbd_interval_empty(&peer_req->i)); + drbd_remove_epoch_entry_interval(mdev, peer_req); + if (peer_req->flags & EE_RESTART_REQUESTS) + restart_conflicting_writes(mdev, sector, peer_req->i.size); + spin_unlock_irq(&mdev->tconn->req_lock); + } else + D_ASSERT(drbd_interval_empty(&peer_req->i)); - drbd_may_finish_epoch(mdev, e->epoch, EV_PUT + (cancel ? EV_CLEANUP : 0)); + drbd_may_finish_epoch(mdev->tconn, peer_req->epoch, EV_PUT + (cancel ? EV_CLEANUP : 0)); - return ok; + return err; } -static int e_send_discard_ack(struct drbd_conf *mdev, struct drbd_work *w, int unused) +static int e_send_ack(struct drbd_work *w, enum drbd_packet ack) { - struct drbd_epoch_entry *e = (struct drbd_epoch_entry *)w; - int ok = 1; + struct drbd_conf *mdev = w->mdev; + struct drbd_peer_request *peer_req = + container_of(w, struct drbd_peer_request, w); + int err; - D_ASSERT(mdev->net_conf->wire_protocol == DRBD_PROT_C); - ok = drbd_send_ack(mdev, P_DISCARD_ACK, e); + err = drbd_send_ack(mdev, ack, peer_req); + dec_unacked(mdev); - spin_lock_irq(&mdev->req_lock); - D_ASSERT(!hlist_unhashed(&e->collision)); - hlist_del_init(&e->collision); - spin_unlock_irq(&mdev->req_lock); + return err; +} - dec_unacked(mdev); +static int e_send_superseded(struct drbd_work *w, int unused) +{ + return e_send_ack(w, P_SUPERSEDED); +} + +static int e_send_retry_write(struct drbd_work *w, int unused) +{ + struct drbd_tconn *tconn = w->mdev->tconn; + + return e_send_ack(w, tconn->agreed_pro_version >= 100 ? + P_RETRY_WRITE : P_SUPERSEDED); +} + +static bool seq_greater(u32 a, u32 b) +{ + /* + * We assume 32-bit wrap-around here. + * For 24-bit wrap-around, we would have to shift: + * a <<= 8; b <<= 8; + */ + return (s32)a - (s32)b > 0; +} + +static u32 seq_max(u32 a, u32 b) +{ + return seq_greater(a, b) ? a : b; +} + +static bool need_peer_seq(struct drbd_conf *mdev) +{ + struct drbd_tconn *tconn = mdev->tconn; + int tp; - return ok; + /* + * We only need to keep track of the last packet_seq number of our peer + * if we are in dual-primary mode and we have the resolve-conflicts flag set; see + * handle_write_conflicts(). + */ + + rcu_read_lock(); + tp = rcu_dereference(mdev->tconn->net_conf)->two_primaries; + rcu_read_unlock(); + + return tp && test_bit(RESOLVE_CONFLICTS, &tconn->flags); } -static bool overlapping_resync_write(struct drbd_conf *mdev, struct drbd_epoch_entry *data_e) +static void update_peer_seq(struct drbd_conf *mdev, unsigned int peer_seq) { + unsigned int newest_peer_seq; - struct drbd_epoch_entry *rs_e; + if (need_peer_seq(mdev)) { + spin_lock(&mdev->peer_seq_lock); + newest_peer_seq = seq_max(mdev->peer_seq, peer_seq); + mdev->peer_seq = newest_peer_seq; + spin_unlock(&mdev->peer_seq_lock); + /* wake up only if we actually changed mdev->peer_seq */ + if (peer_seq == newest_peer_seq) + wake_up(&mdev->seq_wait); + } +} + +static inline int overlaps(sector_t s1, int l1, sector_t s2, int l2) +{ + return !((s1 + (l1>>9) <= s2) || (s1 >= s2 + (l2>>9))); +} + +/* maybe change sync_ee into interval trees as well? */ +static bool overlapping_resync_write(struct drbd_conf *mdev, struct drbd_peer_request *peer_req) +{ + struct drbd_peer_request *rs_req; bool rv = 0; - spin_lock_irq(&mdev->req_lock); - list_for_each_entry(rs_e, &mdev->sync_ee, w.list) { - if (overlaps(data_e->sector, data_e->size, rs_e->sector, rs_e->size)) { + spin_lock_irq(&mdev->tconn->req_lock); + list_for_each_entry(rs_req, &mdev->sync_ee, w.list) { + if (overlaps(peer_req->i.sector, peer_req->i.size, + rs_req->i.sector, rs_req->i.size)) { rv = 1; break; } } - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); return rv; } @@ -1643,35 +1966,41 @@ static bool overlapping_resync_write(struct drbd_conf *mdev, struct drbd_epoch_e * * returns 0 if we may process the packet, * -ERESTARTSYS if we were interrupted (by disconnect signal). */ -static int drbd_wait_peer_seq(struct drbd_conf *mdev, const u32 packet_seq) +static int wait_for_and_update_peer_seq(struct drbd_conf *mdev, const u32 peer_seq) { DEFINE_WAIT(wait); - unsigned int p_seq; long timeout; - int ret = 0; + int ret; + + if (!need_peer_seq(mdev)) + return 0; + spin_lock(&mdev->peer_seq_lock); for (;;) { - prepare_to_wait(&mdev->seq_wait, &wait, TASK_INTERRUPTIBLE); - if (seq_le(packet_seq, mdev->peer_seq+1)) + if (!seq_greater(peer_seq - 1, mdev->peer_seq)) { + mdev->peer_seq = seq_max(mdev->peer_seq, peer_seq); + ret = 0; break; + } if (signal_pending(current)) { ret = -ERESTARTSYS; break; } - p_seq = mdev->peer_seq; + prepare_to_wait(&mdev->seq_wait, &wait, TASK_INTERRUPTIBLE); spin_unlock(&mdev->peer_seq_lock); - timeout = schedule_timeout(30*HZ); + rcu_read_lock(); + timeout = rcu_dereference(mdev->tconn->net_conf)->ping_timeo*HZ/10; + rcu_read_unlock(); + timeout = schedule_timeout(timeout); spin_lock(&mdev->peer_seq_lock); - if (timeout == 0 && p_seq == mdev->peer_seq) { + if (!timeout) { ret = -ETIMEDOUT; - dev_err(DEV, "ASSERT FAILED waited 30 seconds for sequence update, forcing reconnect\n"); + dev_err(DEV, "Timed out waiting for missing ack packets; disconnecting\n"); break; } } - finish_wait(&mdev->seq_wait, &wait); - if (mdev->peer_seq+1 == packet_seq) - mdev->peer_seq++; spin_unlock(&mdev->peer_seq_lock); + finish_wait(&mdev->seq_wait, &wait); return ret; } @@ -1686,233 +2015,277 @@ static unsigned long wire_flags_to_bio(struct drbd_conf *mdev, u32 dpf) (dpf & DP_DISCARD ? REQ_DISCARD : 0); } +static void fail_postponed_requests(struct drbd_conf *mdev, sector_t sector, + unsigned int size) +{ + struct drbd_interval *i; + + repeat: + drbd_for_each_overlap(i, &mdev->write_requests, sector, size) { + struct drbd_request *req; + struct bio_and_error m; + + if (!i->local) + continue; + req = container_of(i, struct drbd_request, i); + if (!(req->rq_state & RQ_POSTPONED)) + continue; + req->rq_state &= ~RQ_POSTPONED; + __req_mod(req, NEG_ACKED, &m); + spin_unlock_irq(&mdev->tconn->req_lock); + if (m.bio) + complete_master_bio(mdev, &m); + spin_lock_irq(&mdev->tconn->req_lock); + goto repeat; + } +} + +static int handle_write_conflicts(struct drbd_conf *mdev, + struct drbd_peer_request *peer_req) +{ + struct drbd_tconn *tconn = mdev->tconn; + bool resolve_conflicts = test_bit(RESOLVE_CONFLICTS, &tconn->flags); + sector_t sector = peer_req->i.sector; + const unsigned int size = peer_req->i.size; + struct drbd_interval *i; + bool equal; + int err; + + /* + * Inserting the peer request into the write_requests tree will prevent + * new conflicting local requests from being added. + */ + drbd_insert_interval(&mdev->write_requests, &peer_req->i); + + repeat: + drbd_for_each_overlap(i, &mdev->write_requests, sector, size) { + if (i == &peer_req->i) + continue; + + if (!i->local) { + /* + * Our peer has sent a conflicting remote request; this + * should not happen in a two-node setup. Wait for the + * earlier peer request to complete. + */ + err = drbd_wait_misc(mdev, i); + if (err) + goto out; + goto repeat; + } + + equal = i->sector == sector && i->size == size; + if (resolve_conflicts) { + /* + * If the peer request is fully contained within the + * overlapping request, it can be considered overwritten + * and thus superseded; otherwise, it will be retried + * once all overlapping requests have completed. + */ + bool superseded = i->sector <= sector && i->sector + + (i->size >> 9) >= sector + (size >> 9); + + if (!equal) + dev_alert(DEV, "Concurrent writes detected: " + "local=%llus +%u, remote=%llus +%u, " + "assuming %s came first\n", + (unsigned long long)i->sector, i->size, + (unsigned long long)sector, size, + superseded ? "local" : "remote"); + + inc_unacked(mdev); + peer_req->w.cb = superseded ? e_send_superseded : + e_send_retry_write; + list_add_tail(&peer_req->w.list, &mdev->done_ee); + wake_asender(mdev->tconn); + + err = -ENOENT; + goto out; + } else { + struct drbd_request *req = + container_of(i, struct drbd_request, i); + + if (!equal) + dev_alert(DEV, "Concurrent writes detected: " + "local=%llus +%u, remote=%llus +%u\n", + (unsigned long long)i->sector, i->size, + (unsigned long long)sector, size); + + if (req->rq_state & RQ_LOCAL_PENDING || + !(req->rq_state & RQ_POSTPONED)) { + /* + * Wait for the node with the discard flag to + * decide if this request has been superseded + * or needs to be retried. + * Requests that have been superseded will + * disappear from the write_requests tree. + * + * In addition, wait for the conflicting + * request to finish locally before submitting + * the conflicting peer request. + */ + err = drbd_wait_misc(mdev, &req->i); + if (err) { + _conn_request_state(mdev->tconn, + NS(conn, C_TIMEOUT), + CS_HARD); + fail_postponed_requests(mdev, sector, size); + goto out; + } + goto repeat; + } + /* + * Remember to restart the conflicting requests after + * the new peer request has completed. + */ + peer_req->flags |= EE_RESTART_REQUESTS; + } + } + err = 0; + + out: + if (err) + drbd_remove_epoch_entry_interval(mdev, peer_req); + return err; +} + /* mirrored write */ -static int receive_Data(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_Data(struct drbd_tconn *tconn, struct packet_info *pi) { + struct drbd_conf *mdev; sector_t sector; - struct drbd_epoch_entry *e; - struct p_data *p = &mdev->data.rbuf.data; + struct drbd_peer_request *peer_req; + struct p_data *p = pi->data; + u32 peer_seq = be32_to_cpu(p->seq_num); int rw = WRITE; u32 dp_flags; + int err, tp; - if (!get_ldev(mdev)) { - spin_lock(&mdev->peer_seq_lock); - if (mdev->peer_seq+1 == be32_to_cpu(p->seq_num)) - mdev->peer_seq++; - spin_unlock(&mdev->peer_seq_lock); + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; - drbd_send_ack_dp(mdev, P_NEG_ACK, p, data_size); - atomic_inc(&mdev->current_epoch->epoch_size); - return drbd_drain_block(mdev, data_size); + if (!get_ldev(mdev)) { + int err2; + + err = wait_for_and_update_peer_seq(mdev, peer_seq); + drbd_send_ack_dp(mdev, P_NEG_ACK, p, pi->size); + atomic_inc(&tconn->current_epoch->epoch_size); + err2 = drbd_drain_block(mdev, pi->size); + if (!err) + err = err2; + return err; } - /* get_ldev(mdev) successful. - * Corresponding put_ldev done either below (on various errors), - * or in drbd_endio_write_sec, if we successfully submit the data at - * the end of this function. */ + /* + * Corresponding put_ldev done either below (on various errors), or in + * drbd_peer_request_endio, if we successfully submit the data at the + * end of this function. + */ sector = be64_to_cpu(p->sector); - e = read_in_block(mdev, p->block_id, sector, data_size); - if (!e) { + peer_req = read_in_block(mdev, p->block_id, sector, pi->size); + if (!peer_req) { put_ldev(mdev); - return false; + return -EIO; } - e->w.cb = e_end_block; + peer_req->w.cb = e_end_block; dp_flags = be32_to_cpu(p->dp_flags); rw |= wire_flags_to_bio(mdev, dp_flags); - if (e->pages == NULL) { - D_ASSERT(e->size == 0); + if (peer_req->pages == NULL) { + D_ASSERT(peer_req->i.size == 0); D_ASSERT(dp_flags & DP_FLUSH); } if (dp_flags & DP_MAY_SET_IN_SYNC) - e->flags |= EE_MAY_SET_IN_SYNC; - - spin_lock(&mdev->epoch_lock); - e->epoch = mdev->current_epoch; - atomic_inc(&e->epoch->epoch_size); - atomic_inc(&e->epoch->active); - spin_unlock(&mdev->epoch_lock); - - /* I'm the receiver, I do hold a net_cnt reference. */ - if (!mdev->net_conf->two_primaries) { - spin_lock_irq(&mdev->req_lock); - } else { - /* don't get the req_lock yet, - * we may sleep in drbd_wait_peer_seq */ - const int size = e->size; - const int discard = test_bit(DISCARD_CONCURRENT, &mdev->flags); - DEFINE_WAIT(wait); - struct drbd_request *i; - struct hlist_node *n; - struct hlist_head *slot; - int first; - - D_ASSERT(mdev->net_conf->wire_protocol == DRBD_PROT_C); - BUG_ON(mdev->ee_hash == NULL); - BUG_ON(mdev->tl_hash == NULL); - - /* conflict detection and handling: - * 1. wait on the sequence number, - * in case this data packet overtook ACK packets. - * 2. check our hash tables for conflicting requests. - * we only need to walk the tl_hash, since an ee can not - * have a conflict with an other ee: on the submitting - * node, the corresponding req had already been conflicting, - * and a conflicting req is never sent. - * - * Note: for two_primaries, we are protocol C, - * so there cannot be any request that is DONE - * but still on the transfer log. - * - * unconditionally add to the ee_hash. - * - * if no conflicting request is found: - * submit. - * - * if any conflicting request is found - * that has not yet been acked, - * AND I have the "discard concurrent writes" flag: - * queue (via done_ee) the P_DISCARD_ACK; OUT. - * - * if any conflicting request is found: - * block the receiver, waiting on misc_wait - * until no more conflicting requests are there, - * or we get interrupted (disconnect). - * - * we do not just write after local io completion of those - * requests, but only after req is done completely, i.e. - * we wait for the P_DISCARD_ACK to arrive! - * - * then proceed normally, i.e. submit. - */ - if (drbd_wait_peer_seq(mdev, be32_to_cpu(p->seq_num))) + peer_req->flags |= EE_MAY_SET_IN_SYNC; + + spin_lock(&tconn->epoch_lock); + peer_req->epoch = tconn->current_epoch; + atomic_inc(&peer_req->epoch->epoch_size); + atomic_inc(&peer_req->epoch->active); + spin_unlock(&tconn->epoch_lock); + + rcu_read_lock(); + tp = rcu_dereference(mdev->tconn->net_conf)->two_primaries; + rcu_read_unlock(); + if (tp) { + peer_req->flags |= EE_IN_INTERVAL_TREE; + err = wait_for_and_update_peer_seq(mdev, peer_seq); + if (err) goto out_interrupted; - - spin_lock_irq(&mdev->req_lock); - - hlist_add_head(&e->collision, ee_hash_slot(mdev, sector)); - -#define OVERLAPS overlaps(i->sector, i->size, sector, size) - slot = tl_hash_slot(mdev, sector); - first = 1; - for (;;) { - int have_unacked = 0; - int have_conflict = 0; - prepare_to_wait(&mdev->misc_wait, &wait, - TASK_INTERRUPTIBLE); - hlist_for_each_entry(i, n, slot, collision) { - if (OVERLAPS) { - /* only ALERT on first iteration, - * we may be woken up early... */ - if (first) - dev_alert(DEV, "%s[%u] Concurrent local write detected!" - " new: %llus +%u; pending: %llus +%u\n", - current->comm, current->pid, - (unsigned long long)sector, size, - (unsigned long long)i->sector, i->size); - if (i->rq_state & RQ_NET_PENDING) - ++have_unacked; - ++have_conflict; - } - } -#undef OVERLAPS - if (!have_conflict) - break; - - /* Discard Ack only for the _first_ iteration */ - if (first && discard && have_unacked) { - dev_alert(DEV, "Concurrent write! [DISCARD BY FLAG] sec=%llus\n", - (unsigned long long)sector); - inc_unacked(mdev); - e->w.cb = e_send_discard_ack; - list_add_tail(&e->w.list, &mdev->done_ee); - - spin_unlock_irq(&mdev->req_lock); - - /* we could probably send that P_DISCARD_ACK ourselves, - * but I don't like the receiver using the msock */ - + spin_lock_irq(&mdev->tconn->req_lock); + err = handle_write_conflicts(mdev, peer_req); + if (err) { + spin_unlock_irq(&mdev->tconn->req_lock); + if (err == -ENOENT) { put_ldev(mdev); - wake_asender(mdev); - finish_wait(&mdev->misc_wait, &wait); - return true; + return 0; } + goto out_interrupted; + } + } else + spin_lock_irq(&mdev->tconn->req_lock); + list_add(&peer_req->w.list, &mdev->active_ee); + spin_unlock_irq(&mdev->tconn->req_lock); - if (signal_pending(current)) { - hlist_del_init(&e->collision); - - spin_unlock_irq(&mdev->req_lock); - - finish_wait(&mdev->misc_wait, &wait); - goto out_interrupted; - } + if (mdev->state.conn == C_SYNC_TARGET) + wait_event(mdev->ee_wait, !overlapping_resync_write(mdev, peer_req)); - spin_unlock_irq(&mdev->req_lock); - if (first) { - first = 0; - dev_alert(DEV, "Concurrent write! [W AFTERWARDS] " - "sec=%llus\n", (unsigned long long)sector); - } else if (discard) { - /* we had none on the first iteration. - * there must be none now. */ - D_ASSERT(have_unacked == 0); - } - schedule(); - spin_lock_irq(&mdev->req_lock); + if (mdev->tconn->agreed_pro_version < 100) { + rcu_read_lock(); + switch (rcu_dereference(mdev->tconn->net_conf)->wire_protocol) { + case DRBD_PROT_C: + dp_flags |= DP_SEND_WRITE_ACK; + break; + case DRBD_PROT_B: + dp_flags |= DP_SEND_RECEIVE_ACK; + break; } - finish_wait(&mdev->misc_wait, &wait); + rcu_read_unlock(); } - list_add(&e->w.list, &mdev->active_ee); - spin_unlock_irq(&mdev->req_lock); - - if (mdev->state.conn == C_SYNC_TARGET) - wait_event(mdev->ee_wait, !overlapping_resync_write(mdev, e)); - - switch (mdev->net_conf->wire_protocol) { - case DRBD_PROT_C: + if (dp_flags & DP_SEND_WRITE_ACK) { + peer_req->flags |= EE_SEND_WRITE_ACK; inc_unacked(mdev); /* corresponding dec_unacked() in e_end_block() * respective _drbd_clear_done_ee */ - break; - case DRBD_PROT_B: + } + + if (dp_flags & DP_SEND_RECEIVE_ACK) { /* I really don't like it that the receiver thread * sends on the msock, but anyways */ - drbd_send_ack(mdev, P_RECV_ACK, e); - break; - case DRBD_PROT_A: - /* nothing to do */ - break; + drbd_send_ack(mdev, P_RECV_ACK, peer_req); } if (mdev->state.pdsk < D_INCONSISTENT) { /* In case we have the only disk of the cluster, */ - drbd_set_out_of_sync(mdev, e->sector, e->size); - e->flags |= EE_CALL_AL_COMPLETE_IO; - e->flags &= ~EE_MAY_SET_IN_SYNC; - drbd_al_begin_io(mdev, e->sector); + drbd_set_out_of_sync(mdev, peer_req->i.sector, peer_req->i.size); + peer_req->flags |= EE_CALL_AL_COMPLETE_IO; + peer_req->flags &= ~EE_MAY_SET_IN_SYNC; + drbd_al_begin_io(mdev, &peer_req->i); } - if (drbd_submit_ee(mdev, e, rw, DRBD_FAULT_DT_WR) == 0) - return true; + err = drbd_submit_peer_request(mdev, peer_req, rw, DRBD_FAULT_DT_WR); + if (!err) + return 0; /* don't care for the reason here */ dev_err(DEV, "submit failed, triggering re-connect\n"); - spin_lock_irq(&mdev->req_lock); - list_del(&e->w.list); - hlist_del_init(&e->collision); - spin_unlock_irq(&mdev->req_lock); - if (e->flags & EE_CALL_AL_COMPLETE_IO) - drbd_al_complete_io(mdev, e->sector); + spin_lock_irq(&mdev->tconn->req_lock); + list_del(&peer_req->w.list); + drbd_remove_epoch_entry_interval(mdev, peer_req); + spin_unlock_irq(&mdev->tconn->req_lock); + if (peer_req->flags & EE_CALL_AL_COMPLETE_IO) + drbd_al_complete_io(mdev, &peer_req->i); out_interrupted: - drbd_may_finish_epoch(mdev, e->epoch, EV_PUT + EV_CLEANUP); + drbd_may_finish_epoch(tconn, peer_req->epoch, EV_PUT + EV_CLEANUP); put_ldev(mdev); - drbd_free_ee(mdev, e); - return false; + drbd_free_peer_req(mdev, peer_req); + return err; } /* We may throttle resync, if the lower device seems to be busy, @@ -1933,9 +2306,14 @@ int drbd_rs_should_slow_down(struct drbd_conf *mdev, sector_t sector) struct lc_element *tmp; int curr_events; int throttle = 0; + unsigned int c_min_rate; + + rcu_read_lock(); + c_min_rate = rcu_dereference(mdev->ldev->disk_conf)->c_min_rate; + rcu_read_unlock(); /* feature disabled? */ - if (mdev->sync_conf.c_min_rate == 0) + if (c_min_rate == 0) return 0; spin_lock_irq(&mdev->al_lock); @@ -1975,40 +2353,46 @@ int drbd_rs_should_slow_down(struct drbd_conf *mdev, sector_t sector) db = mdev->rs_mark_left[i] - rs_left; dbdt = Bit2KB(db/dt); - if (dbdt > mdev->sync_conf.c_min_rate) + if (dbdt > c_min_rate) throttle = 1; } return throttle; } -static int receive_DataRequest(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int digest_size) +static int receive_DataRequest(struct drbd_tconn *tconn, struct packet_info *pi) { + struct drbd_conf *mdev; sector_t sector; - const sector_t capacity = drbd_get_capacity(mdev->this_bdev); - struct drbd_epoch_entry *e; + sector_t capacity; + struct drbd_peer_request *peer_req; struct digest_info *di = NULL; int size, verb; unsigned int fault_type; - struct p_block_req *p = &mdev->data.rbuf.block_req; + struct p_block_req *p = pi->data; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; + capacity = drbd_get_capacity(mdev->this_bdev); sector = be64_to_cpu(p->sector); size = be32_to_cpu(p->blksize); - if (size <= 0 || (size & 0x1ff) != 0 || size > DRBD_MAX_BIO_SIZE) { + if (size <= 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_BIO_SIZE) { dev_err(DEV, "%s:%d: sector: %llus, size: %u\n", __FILE__, __LINE__, (unsigned long long)sector, size); - return false; + return -EINVAL; } if (sector + (size>>9) > capacity) { dev_err(DEV, "%s:%d: sector: %llus, size: %u\n", __FILE__, __LINE__, (unsigned long long)sector, size); - return false; + return -EINVAL; } if (!get_ldev_if_state(mdev, D_UP_TO_DATE)) { verb = 1; - switch (cmd) { + switch (pi->cmd) { case P_DATA_REQUEST: drbd_send_ack_rp(mdev, P_NEG_DREPLY, p); break; @@ -2023,35 +2407,34 @@ static int receive_DataRequest(struct drbd_conf *mdev, enum drbd_packets cmd, un drbd_send_ack_ex(mdev, P_OV_RESULT, sector, size, ID_IN_SYNC); break; default: - dev_err(DEV, "unexpected command (%s) in receive_DataRequest\n", - cmdname(cmd)); + BUG(); } if (verb && __ratelimit(&drbd_ratelimit_state)) dev_err(DEV, "Can not satisfy peer's read request, " "no local data.\n"); /* drain possibly payload */ - return drbd_drain_block(mdev, digest_size); + return drbd_drain_block(mdev, pi->size); } /* GFP_NOIO, because we must not cause arbitrary write-out: in a DRBD * "criss-cross" setup, that might cause write-out on some other DRBD, * which in turn might block on the other node at this very place. */ - e = drbd_alloc_ee(mdev, p->block_id, sector, size, GFP_NOIO); - if (!e) { + peer_req = drbd_alloc_peer_req(mdev, p->block_id, sector, size, GFP_NOIO); + if (!peer_req) { put_ldev(mdev); - return false; + return -ENOMEM; } - switch (cmd) { + switch (pi->cmd) { case P_DATA_REQUEST: - e->w.cb = w_e_end_data_req; + peer_req->w.cb = w_e_end_data_req; fault_type = DRBD_FAULT_DT_RD; /* application IO, don't drbd_rs_begin_io */ goto submit; case P_RS_DATA_REQUEST: - e->w.cb = w_e_end_rsdata_req; + peer_req->w.cb = w_e_end_rsdata_req; fault_type = DRBD_FAULT_RS_RD; /* used in the sector offset progress display */ mdev->bm_resync_fo = BM_SECT_TO_BIT(sector); @@ -2060,28 +2443,28 @@ static int receive_DataRequest(struct drbd_conf *mdev, enum drbd_packets cmd, un case P_OV_REPLY: case P_CSUM_RS_REQUEST: fault_type = DRBD_FAULT_RS_RD; - di = kmalloc(sizeof(*di) + digest_size, GFP_NOIO); + di = kmalloc(sizeof(*di) + pi->size, GFP_NOIO); if (!di) goto out_free_e; - di->digest_size = digest_size; + di->digest_size = pi->size; di->digest = (((char *)di)+sizeof(struct digest_info)); - e->digest = di; - e->flags |= EE_HAS_DIGEST; + peer_req->digest = di; + peer_req->flags |= EE_HAS_DIGEST; - if (drbd_recv(mdev, di->digest, digest_size) != digest_size) + if (drbd_recv_all(mdev->tconn, di->digest, pi->size)) goto out_free_e; - if (cmd == P_CSUM_RS_REQUEST) { - D_ASSERT(mdev->agreed_pro_version >= 89); - e->w.cb = w_e_end_csum_rs_req; + if (pi->cmd == P_CSUM_RS_REQUEST) { + D_ASSERT(mdev->tconn->agreed_pro_version >= 89); + peer_req->w.cb = w_e_end_csum_rs_req; /* used in the sector offset progress display */ mdev->bm_resync_fo = BM_SECT_TO_BIT(sector); - } else if (cmd == P_OV_REPLY) { + } else if (pi->cmd == P_OV_REPLY) { /* track progress, we may need to throttle */ atomic_add(size >> 9, &mdev->rs_sect_in); - e->w.cb = w_e_end_ov_reply; + peer_req->w.cb = w_e_end_ov_reply; dec_rs_pending(mdev); /* drbd_rs_begin_io done when we sent this request, * but accounting still needs to be done. */ @@ -2091,7 +2474,7 @@ static int receive_DataRequest(struct drbd_conf *mdev, enum drbd_packets cmd, un case P_OV_REQUEST: if (mdev->ov_start_sector == ~(sector_t)0 && - mdev->agreed_pro_version >= 90) { + mdev->tconn->agreed_pro_version >= 90) { unsigned long now = jiffies; int i; mdev->ov_start_sector = sector; @@ -2105,15 +2488,12 @@ static int receive_DataRequest(struct drbd_conf *mdev, enum drbd_packets cmd, un dev_info(DEV, "Online Verify start sector: %llu\n", (unsigned long long)sector); } - e->w.cb = w_e_end_ov_req; + peer_req->w.cb = w_e_end_ov_req; fault_type = DRBD_FAULT_RS_RD; break; default: - dev_err(DEV, "unexpected command (%s) in receive_DataRequest\n", - cmdname(cmd)); - fault_type = DRBD_FAULT_MAX; - goto out_free_e; + BUG(); } /* Throttle, drbd_rs_begin_io and submit should become asynchronous @@ -2148,30 +2528,31 @@ submit_for_resync: submit: inc_unacked(mdev); - spin_lock_irq(&mdev->req_lock); - list_add_tail(&e->w.list, &mdev->read_ee); - spin_unlock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); + list_add_tail(&peer_req->w.list, &mdev->read_ee); + spin_unlock_irq(&mdev->tconn->req_lock); - if (drbd_submit_ee(mdev, e, READ, fault_type) == 0) - return true; + if (drbd_submit_peer_request(mdev, peer_req, READ, fault_type) == 0) + return 0; /* don't care for the reason here */ dev_err(DEV, "submit failed, triggering re-connect\n"); - spin_lock_irq(&mdev->req_lock); - list_del(&e->w.list); - spin_unlock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); + list_del(&peer_req->w.list); + spin_unlock_irq(&mdev->tconn->req_lock); /* no drbd_rs_complete_io(), we are dropping the connection anyways */ out_free_e: put_ldev(mdev); - drbd_free_ee(mdev, e); - return false; + drbd_free_peer_req(mdev, peer_req); + return -EIO; } static int drbd_asb_recover_0p(struct drbd_conf *mdev) __must_hold(local) { int self, peer, rv = -100; unsigned long ch_self, ch_peer; + enum drbd_after_sb_p after_sb_0p; self = mdev->ldev->md.uuid[UI_BITMAP] & 1; peer = mdev->p_uuid[UI_BITMAP] & 1; @@ -2179,10 +2560,14 @@ static int drbd_asb_recover_0p(struct drbd_conf *mdev) __must_hold(local) ch_peer = mdev->p_uuid[UI_SIZE]; ch_self = mdev->comm_bm_set; - switch (mdev->net_conf->after_sb_0p) { + rcu_read_lock(); + after_sb_0p = rcu_dereference(mdev->tconn->net_conf)->after_sb_0p; + rcu_read_unlock(); + switch (after_sb_0p) { case ASB_CONSENSUS: case ASB_DISCARD_SECONDARY: case ASB_CALL_HELPER: + case ASB_VIOLENTLY: dev_err(DEV, "Configuration error.\n"); break; case ASB_DISCONNECT: @@ -2211,14 +2596,14 @@ static int drbd_asb_recover_0p(struct drbd_conf *mdev) __must_hold(local) "Using discard-least-changes instead\n"); case ASB_DISCARD_ZERO_CHG: if (ch_peer == 0 && ch_self == 0) { - rv = test_bit(DISCARD_CONCURRENT, &mdev->flags) + rv = test_bit(RESOLVE_CONFLICTS, &mdev->tconn->flags) ? -1 : 1; break; } else { if (ch_peer == 0) { rv = 1; break; } if (ch_self == 0) { rv = -1; break; } } - if (mdev->net_conf->after_sb_0p == ASB_DISCARD_ZERO_CHG) + if (after_sb_0p == ASB_DISCARD_ZERO_CHG) break; case ASB_DISCARD_LEAST_CHG: if (ch_self < ch_peer) @@ -2227,7 +2612,7 @@ static int drbd_asb_recover_0p(struct drbd_conf *mdev) __must_hold(local) rv = 1; else /* ( ch_self == ch_peer ) */ /* Well, then use something else. */ - rv = test_bit(DISCARD_CONCURRENT, &mdev->flags) + rv = test_bit(RESOLVE_CONFLICTS, &mdev->tconn->flags) ? -1 : 1; break; case ASB_DISCARD_LOCAL: @@ -2243,13 +2628,18 @@ static int drbd_asb_recover_0p(struct drbd_conf *mdev) __must_hold(local) static int drbd_asb_recover_1p(struct drbd_conf *mdev) __must_hold(local) { int hg, rv = -100; + enum drbd_after_sb_p after_sb_1p; - switch (mdev->net_conf->after_sb_1p) { + rcu_read_lock(); + after_sb_1p = rcu_dereference(mdev->tconn->net_conf)->after_sb_1p; + rcu_read_unlock(); + switch (after_sb_1p) { case ASB_DISCARD_YOUNGER_PRI: case ASB_DISCARD_OLDER_PRI: case ASB_DISCARD_LEAST_CHG: case ASB_DISCARD_LOCAL: case ASB_DISCARD_REMOTE: + case ASB_DISCARD_ZERO_CHG: dev_err(DEV, "Configuration error.\n"); break; case ASB_DISCONNECT: @@ -2292,8 +2682,12 @@ static int drbd_asb_recover_1p(struct drbd_conf *mdev) __must_hold(local) static int drbd_asb_recover_2p(struct drbd_conf *mdev) __must_hold(local) { int hg, rv = -100; + enum drbd_after_sb_p after_sb_2p; - switch (mdev->net_conf->after_sb_2p) { + rcu_read_lock(); + after_sb_2p = rcu_dereference(mdev->tconn->net_conf)->after_sb_2p; + rcu_read_unlock(); + switch (after_sb_2p) { case ASB_DISCARD_YOUNGER_PRI: case ASB_DISCARD_OLDER_PRI: case ASB_DISCARD_LEAST_CHG: @@ -2301,6 +2695,7 @@ static int drbd_asb_recover_2p(struct drbd_conf *mdev) __must_hold(local) case ASB_DISCARD_REMOTE: case ASB_CONSENSUS: case ASB_DISCARD_SECONDARY: + case ASB_DISCARD_ZERO_CHG: dev_err(DEV, "Configuration error.\n"); break; case ASB_VIOLENTLY: @@ -2386,13 +2781,15 @@ static int drbd_uuid_compare(struct drbd_conf *mdev, int *rule_nr) __must_hold(l if (mdev->p_uuid[UI_BITMAP] == (u64)0 && mdev->ldev->md.uuid[UI_BITMAP] != (u64)0) { - if (mdev->agreed_pro_version < 91) + if (mdev->tconn->agreed_pro_version < 91) return -1091; if ((mdev->ldev->md.uuid[UI_BITMAP] & ~((u64)1)) == (mdev->p_uuid[UI_HISTORY_START] & ~((u64)1)) && (mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == (mdev->p_uuid[UI_HISTORY_START + 1] & ~((u64)1))) { dev_info(DEV, "was SyncSource, missed the resync finished event, corrected myself:\n"); - drbd_uuid_set_bm(mdev, 0UL); + drbd_uuid_move_history(mdev); + mdev->ldev->md.uuid[UI_HISTORY_START] = mdev->ldev->md.uuid[UI_BITMAP]; + mdev->ldev->md.uuid[UI_BITMAP] = 0; drbd_uuid_dump(mdev, "self", mdev->ldev->md.uuid, mdev->state.disk >= D_NEGOTIATING ? drbd_bm_total_weight(mdev) : 0, 0); @@ -2407,7 +2804,7 @@ static int drbd_uuid_compare(struct drbd_conf *mdev, int *rule_nr) __must_hold(l if (mdev->ldev->md.uuid[UI_BITMAP] == (u64)0 && mdev->p_uuid[UI_BITMAP] != (u64)0) { - if (mdev->agreed_pro_version < 91) + if (mdev->tconn->agreed_pro_version < 91) return -1091; if ((mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == (mdev->p_uuid[UI_BITMAP] & ~((u64)1)) && @@ -2440,7 +2837,7 @@ static int drbd_uuid_compare(struct drbd_conf *mdev, int *rule_nr) __must_hold(l case 1: /* self_pri && !peer_pri */ return 1; case 2: /* !self_pri && peer_pri */ return -1; case 3: /* self_pri && peer_pri */ - dc = test_bit(DISCARD_CONCURRENT, &mdev->flags); + dc = test_bit(RESOLVE_CONFLICTS, &mdev->tconn->flags); return dc ? -1 : 1; } } @@ -2453,14 +2850,14 @@ static int drbd_uuid_compare(struct drbd_conf *mdev, int *rule_nr) __must_hold(l *rule_nr = 51; peer = mdev->p_uuid[UI_HISTORY_START] & ~((u64)1); if (self == peer) { - if (mdev->agreed_pro_version < 96 ? + if (mdev->tconn->agreed_pro_version < 96 ? (mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == (mdev->p_uuid[UI_HISTORY_START + 1] & ~((u64)1)) : peer + UUID_NEW_BM_OFFSET == (mdev->p_uuid[UI_BITMAP] & ~((u64)1))) { /* The last P_SYNC_UUID did not get though. Undo the last start of resync as sync source modifications of the peer's UUIDs. */ - if (mdev->agreed_pro_version < 91) + if (mdev->tconn->agreed_pro_version < 91) return -1091; mdev->p_uuid[UI_BITMAP] = mdev->p_uuid[UI_HISTORY_START]; @@ -2490,18 +2887,18 @@ static int drbd_uuid_compare(struct drbd_conf *mdev, int *rule_nr) __must_hold(l *rule_nr = 71; self = mdev->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1); if (self == peer) { - if (mdev->agreed_pro_version < 96 ? + if (mdev->tconn->agreed_pro_version < 96 ? (mdev->ldev->md.uuid[UI_HISTORY_START + 1] & ~((u64)1)) == (mdev->p_uuid[UI_HISTORY_START] & ~((u64)1)) : self + UUID_NEW_BM_OFFSET == (mdev->ldev->md.uuid[UI_BITMAP] & ~((u64)1))) { /* The last P_SYNC_UUID did not get though. Undo the last start of resync as sync source modifications of our UUIDs. */ - if (mdev->agreed_pro_version < 91) + if (mdev->tconn->agreed_pro_version < 91) return -1091; - _drbd_uuid_set(mdev, UI_BITMAP, mdev->ldev->md.uuid[UI_HISTORY_START]); - _drbd_uuid_set(mdev, UI_HISTORY_START, mdev->ldev->md.uuid[UI_HISTORY_START + 1]); + __drbd_uuid_set(mdev, UI_BITMAP, mdev->ldev->md.uuid[UI_HISTORY_START]); + __drbd_uuid_set(mdev, UI_HISTORY_START, mdev->ldev->md.uuid[UI_HISTORY_START + 1]); dev_info(DEV, "Last syncUUID did not get through, corrected:\n"); drbd_uuid_dump(mdev, "self", mdev->ldev->md.uuid, @@ -2545,20 +2942,24 @@ static int drbd_uuid_compare(struct drbd_conf *mdev, int *rule_nr) __must_hold(l static enum drbd_conns drbd_sync_handshake(struct drbd_conf *mdev, enum drbd_role peer_role, enum drbd_disk_state peer_disk) __must_hold(local) { - int hg, rule_nr; enum drbd_conns rv = C_MASK; enum drbd_disk_state mydisk; + struct net_conf *nc; + int hg, rule_nr, rr_conflict, tentative; mydisk = mdev->state.disk; if (mydisk == D_NEGOTIATING) mydisk = mdev->new_state_tmp.disk; dev_info(DEV, "drbd_sync_handshake:\n"); + + spin_lock_irq(&mdev->ldev->md.uuid_lock); drbd_uuid_dump(mdev, "self", mdev->ldev->md.uuid, mdev->comm_bm_set, 0); drbd_uuid_dump(mdev, "peer", mdev->p_uuid, mdev->p_uuid[UI_SIZE], mdev->p_uuid[UI_FLAGS]); hg = drbd_uuid_compare(mdev, &rule_nr); + spin_unlock_irq(&mdev->ldev->md.uuid_lock); dev_info(DEV, "uuid_compare()=%d by rule %d\n", hg, rule_nr); @@ -2584,7 +2985,10 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_conf *mdev, enum drbd_rol if (abs(hg) == 100) drbd_khelper(mdev, "initial-split-brain"); - if (hg == 100 || (hg == -100 && mdev->net_conf->always_asbp)) { + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + + if (hg == 100 || (hg == -100 && nc->always_asbp)) { int pcount = (mdev->state.role == R_PRIMARY) + (peer_role == R_PRIMARY); int forced = (hg == -100); @@ -2613,9 +3017,9 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_conf *mdev, enum drbd_rol } if (hg == -100) { - if (mdev->net_conf->want_lose && !(mdev->p_uuid[UI_FLAGS]&1)) + if (test_bit(DISCARD_MY_DATA, &mdev->flags) && !(mdev->p_uuid[UI_FLAGS]&1)) hg = -1; - if (!mdev->net_conf->want_lose && (mdev->p_uuid[UI_FLAGS]&1)) + if (!test_bit(DISCARD_MY_DATA, &mdev->flags) && (mdev->p_uuid[UI_FLAGS]&1)) hg = 1; if (abs(hg) < 100) @@ -2623,6 +3027,9 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_conf *mdev, enum drbd_rol "Sync from %s node\n", (hg < 0) ? "peer" : "this"); } + rr_conflict = nc->rr_conflict; + tentative = nc->tentative; + rcu_read_unlock(); if (hg == -100) { /* FIXME this log message is not correct if we end up here @@ -2641,7 +3048,7 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_conf *mdev, enum drbd_rol if (hg < 0 && /* by intention we do not use mydisk here. */ mdev->state.role == R_PRIMARY && mdev->state.disk >= D_CONSISTENT) { - switch (mdev->net_conf->rr_conflict) { + switch (rr_conflict) { case ASB_CALL_HELPER: drbd_khelper(mdev, "pri-lost"); /* fall through */ @@ -2654,7 +3061,7 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_conf *mdev, enum drbd_rol } } - if (mdev->net_conf->dry_run || test_bit(CONN_DRY_RUN, &mdev->flags)) { + if (tentative || test_bit(CONN_DRY_RUN, &mdev->tconn->flags)) { if (hg == 0) dev_info(DEV, "dry-run connect: No resync, would become Connected immediately.\n"); else @@ -2686,33 +3093,29 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_conf *mdev, enum drbd_rol return rv; } -/* returns 1 if invalid */ -static int cmp_after_sb(enum drbd_after_sb_p peer, enum drbd_after_sb_p self) +static enum drbd_after_sb_p convert_after_sb(enum drbd_after_sb_p peer) { /* ASB_DISCARD_REMOTE - ASB_DISCARD_LOCAL is valid */ - if ((peer == ASB_DISCARD_REMOTE && self == ASB_DISCARD_LOCAL) || - (self == ASB_DISCARD_REMOTE && peer == ASB_DISCARD_LOCAL)) - return 0; + if (peer == ASB_DISCARD_REMOTE) + return ASB_DISCARD_LOCAL; /* any other things with ASB_DISCARD_REMOTE or ASB_DISCARD_LOCAL are invalid */ - if (peer == ASB_DISCARD_REMOTE || peer == ASB_DISCARD_LOCAL || - self == ASB_DISCARD_REMOTE || self == ASB_DISCARD_LOCAL) - return 1; + if (peer == ASB_DISCARD_LOCAL) + return ASB_DISCARD_REMOTE; /* everything else is valid if they are equal on both sides. */ - if (peer == self) - return 0; - - /* everything es is invalid. */ - return 1; + return peer; } -static int receive_protocol(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_protocol(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_protocol *p = &mdev->data.rbuf.protocol; - int p_proto, p_after_sb_0p, p_after_sb_1p, p_after_sb_2p; - int p_want_lose, p_two_primaries, cf; - char p_integrity_alg[SHARED_SECRET_MAX] = ""; + struct p_protocol *p = pi->data; + enum drbd_after_sb_p p_after_sb_0p, p_after_sb_1p, p_after_sb_2p; + int p_proto, p_discard_my_data, p_two_primaries, cf; + struct net_conf *nc, *old_net_conf, *new_net_conf = NULL; + char integrity_alg[SHARED_SECRET_MAX] = ""; + struct crypto_hash *peer_integrity_tfm = NULL; + void *int_dig_in = NULL, *int_dig_vv = NULL; p_proto = be32_to_cpu(p->protocol); p_after_sb_0p = be32_to_cpu(p->after_sb_0p); @@ -2720,63 +3123,138 @@ static int receive_protocol(struct drbd_conf *mdev, enum drbd_packets cmd, unsig p_after_sb_2p = be32_to_cpu(p->after_sb_2p); p_two_primaries = be32_to_cpu(p->two_primaries); cf = be32_to_cpu(p->conn_flags); - p_want_lose = cf & CF_WANT_LOSE; - - clear_bit(CONN_DRY_RUN, &mdev->flags); + p_discard_my_data = cf & CF_DISCARD_MY_DATA; - if (cf & CF_DRY_RUN) - set_bit(CONN_DRY_RUN, &mdev->flags); + if (tconn->agreed_pro_version >= 87) { + int err; - if (p_proto != mdev->net_conf->wire_protocol) { - dev_err(DEV, "incompatible communication protocols\n"); - goto disconnect; + if (pi->size > sizeof(integrity_alg)) + return -EIO; + err = drbd_recv_all(tconn, integrity_alg, pi->size); + if (err) + return err; + integrity_alg[SHARED_SECRET_MAX - 1] = 0; } - if (cmp_after_sb(p_after_sb_0p, mdev->net_conf->after_sb_0p)) { - dev_err(DEV, "incompatible after-sb-0pri settings\n"); - goto disconnect; - } + if (pi->cmd != P_PROTOCOL_UPDATE) { + clear_bit(CONN_DRY_RUN, &tconn->flags); - if (cmp_after_sb(p_after_sb_1p, mdev->net_conf->after_sb_1p)) { - dev_err(DEV, "incompatible after-sb-1pri settings\n"); - goto disconnect; - } + if (cf & CF_DRY_RUN) + set_bit(CONN_DRY_RUN, &tconn->flags); - if (cmp_after_sb(p_after_sb_2p, mdev->net_conf->after_sb_2p)) { - dev_err(DEV, "incompatible after-sb-2pri settings\n"); - goto disconnect; - } + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); - if (p_want_lose && mdev->net_conf->want_lose) { - dev_err(DEV, "both sides have the 'want_lose' flag set\n"); - goto disconnect; - } + if (p_proto != nc->wire_protocol) { + conn_err(tconn, "incompatible %s settings\n", "protocol"); + goto disconnect_rcu_unlock; + } - if (p_two_primaries != mdev->net_conf->two_primaries) { - dev_err(DEV, "incompatible setting of the two-primaries options\n"); - goto disconnect; + if (convert_after_sb(p_after_sb_0p) != nc->after_sb_0p) { + conn_err(tconn, "incompatible %s settings\n", "after-sb-0pri"); + goto disconnect_rcu_unlock; + } + + if (convert_after_sb(p_after_sb_1p) != nc->after_sb_1p) { + conn_err(tconn, "incompatible %s settings\n", "after-sb-1pri"); + goto disconnect_rcu_unlock; + } + + if (convert_after_sb(p_after_sb_2p) != nc->after_sb_2p) { + conn_err(tconn, "incompatible %s settings\n", "after-sb-2pri"); + goto disconnect_rcu_unlock; + } + + if (p_discard_my_data && nc->discard_my_data) { + conn_err(tconn, "incompatible %s settings\n", "discard-my-data"); + goto disconnect_rcu_unlock; + } + + if (p_two_primaries != nc->two_primaries) { + conn_err(tconn, "incompatible %s settings\n", "allow-two-primaries"); + goto disconnect_rcu_unlock; + } + + if (strcmp(integrity_alg, nc->integrity_alg)) { + conn_err(tconn, "incompatible %s settings\n", "data-integrity-alg"); + goto disconnect_rcu_unlock; + } + + rcu_read_unlock(); } - if (mdev->agreed_pro_version >= 87) { - unsigned char *my_alg = mdev->net_conf->integrity_alg; + if (integrity_alg[0]) { + int hash_size; - if (drbd_recv(mdev, p_integrity_alg, data_size) != data_size) - return false; + /* + * We can only change the peer data integrity algorithm + * here. Changing our own data integrity algorithm + * requires that we send a P_PROTOCOL_UPDATE packet at + * the same time; otherwise, the peer has no way to + * tell between which packets the algorithm should + * change. + */ - p_integrity_alg[SHARED_SECRET_MAX-1] = 0; - if (strcmp(p_integrity_alg, my_alg)) { - dev_err(DEV, "incompatible setting of the data-integrity-alg\n"); + peer_integrity_tfm = crypto_alloc_hash(integrity_alg, 0, CRYPTO_ALG_ASYNC); + if (!peer_integrity_tfm) { + conn_err(tconn, "peer data-integrity-alg %s not supported\n", + integrity_alg); goto disconnect; } - dev_info(DEV, "data-integrity-alg: %s\n", - my_alg[0] ? my_alg : (unsigned char *)"<not-used>"); + + hash_size = crypto_hash_digestsize(peer_integrity_tfm); + int_dig_in = kmalloc(hash_size, GFP_KERNEL); + int_dig_vv = kmalloc(hash_size, GFP_KERNEL); + if (!(int_dig_in && int_dig_vv)) { + conn_err(tconn, "Allocation of buffers for data integrity checking failed\n"); + goto disconnect; + } + } + + new_net_conf = kmalloc(sizeof(struct net_conf), GFP_KERNEL); + if (!new_net_conf) { + conn_err(tconn, "Allocation of new net_conf failed\n"); + goto disconnect; } - return true; + mutex_lock(&tconn->data.mutex); + mutex_lock(&tconn->conf_update); + old_net_conf = tconn->net_conf; + *new_net_conf = *old_net_conf; + + new_net_conf->wire_protocol = p_proto; + new_net_conf->after_sb_0p = convert_after_sb(p_after_sb_0p); + new_net_conf->after_sb_1p = convert_after_sb(p_after_sb_1p); + new_net_conf->after_sb_2p = convert_after_sb(p_after_sb_2p); + new_net_conf->two_primaries = p_two_primaries; + rcu_assign_pointer(tconn->net_conf, new_net_conf); + mutex_unlock(&tconn->conf_update); + mutex_unlock(&tconn->data.mutex); + + crypto_free_hash(tconn->peer_integrity_tfm); + kfree(tconn->int_dig_in); + kfree(tconn->int_dig_vv); + tconn->peer_integrity_tfm = peer_integrity_tfm; + tconn->int_dig_in = int_dig_in; + tconn->int_dig_vv = int_dig_vv; + + if (strcmp(old_net_conf->integrity_alg, integrity_alg)) + conn_info(tconn, "peer data-integrity-alg: %s\n", + integrity_alg[0] ? integrity_alg : "(none)"); + + synchronize_rcu(); + kfree(old_net_conf); + return 0; + +disconnect_rcu_unlock: + rcu_read_unlock(); disconnect: - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - return false; + crypto_free_hash(peer_integrity_tfm); + kfree(int_dig_in); + kfree(int_dig_vv); + conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD); + return -EIO; } /* helper function @@ -2798,24 +3276,64 @@ struct crypto_hash *drbd_crypto_alloc_digest_safe(const struct drbd_conf *mdev, alg, name, PTR_ERR(tfm)); return tfm; } - if (!drbd_crypto_is_hash(crypto_hash_tfm(tfm))) { - crypto_free_hash(tfm); - dev_err(DEV, "\"%s\" is not a digest (%s)\n", alg, name); - return ERR_PTR(-EINVAL); - } return tfm; } -static int receive_SyncParam(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int packet_size) +static int ignore_remaining_packet(struct drbd_tconn *tconn, struct packet_info *pi) +{ + void *buffer = tconn->data.rbuf; + int size = pi->size; + + while (size) { + int s = min_t(int, size, DRBD_SOCKET_BUFFER_SIZE); + s = drbd_recv(tconn, buffer, s); + if (s <= 0) { + if (s < 0) + return s; + break; + } + size -= s; + } + if (size) + return -EIO; + return 0; +} + +/* + * config_unknown_volume - device configuration command for unknown volume + * + * When a device is added to an existing connection, the node on which the + * device is added first will send configuration commands to its peer but the + * peer will not know about the device yet. It will warn and ignore these + * commands. Once the device is added on the second node, the second node will + * send the same device configuration commands, but in the other direction. + * + * (We can also end up here if drbd is misconfigured.) + */ +static int config_unknown_volume(struct drbd_tconn *tconn, struct packet_info *pi) { - int ok = true; - struct p_rs_param_95 *p = &mdev->data.rbuf.rs_param_95; + conn_warn(tconn, "%s packet received for volume %u, which is not configured locally\n", + cmdname(pi->cmd), pi->vnr); + return ignore_remaining_packet(tconn, pi); +} + +static int receive_SyncParam(struct drbd_tconn *tconn, struct packet_info *pi) +{ + struct drbd_conf *mdev; + struct p_rs_param_95 *p; unsigned int header_size, data_size, exp_max_sz; struct crypto_hash *verify_tfm = NULL; struct crypto_hash *csums_tfm = NULL; - const int apv = mdev->agreed_pro_version; - int *rs_plan_s = NULL; + struct net_conf *old_net_conf, *new_net_conf = NULL; + struct disk_conf *old_disk_conf = NULL, *new_disk_conf = NULL; + const int apv = tconn->agreed_pro_version; + struct fifo_buffer *old_plan = NULL, *new_plan = NULL; int fifo_size = 0; + int err; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return config_unknown_volume(tconn, pi); exp_max_sz = apv <= 87 ? sizeof(struct p_rs_param) : apv == 88 ? sizeof(struct p_rs_param) @@ -2823,32 +3341,49 @@ static int receive_SyncParam(struct drbd_conf *mdev, enum drbd_packets cmd, unsi : apv <= 94 ? sizeof(struct p_rs_param_89) : /* apv >= 95 */ sizeof(struct p_rs_param_95); - if (packet_size > exp_max_sz) { + if (pi->size > exp_max_sz) { dev_err(DEV, "SyncParam packet too long: received %u, expected <= %u bytes\n", - packet_size, exp_max_sz); - return false; + pi->size, exp_max_sz); + return -EIO; } if (apv <= 88) { - header_size = sizeof(struct p_rs_param) - sizeof(struct p_header80); - data_size = packet_size - header_size; + header_size = sizeof(struct p_rs_param); + data_size = pi->size - header_size; } else if (apv <= 94) { - header_size = sizeof(struct p_rs_param_89) - sizeof(struct p_header80); - data_size = packet_size - header_size; + header_size = sizeof(struct p_rs_param_89); + data_size = pi->size - header_size; D_ASSERT(data_size == 0); } else { - header_size = sizeof(struct p_rs_param_95) - sizeof(struct p_header80); - data_size = packet_size - header_size; + header_size = sizeof(struct p_rs_param_95); + data_size = pi->size - header_size; D_ASSERT(data_size == 0); } /* initialize verify_alg and csums_alg */ + p = pi->data; memset(p->verify_alg, 0, 2 * SHARED_SECRET_MAX); - if (drbd_recv(mdev, &p->head.payload, header_size) != header_size) - return false; + err = drbd_recv_all(mdev->tconn, p, header_size); + if (err) + return err; - mdev->sync_conf.rate = be32_to_cpu(p->rate); + mutex_lock(&mdev->tconn->conf_update); + old_net_conf = mdev->tconn->net_conf; + if (get_ldev(mdev)) { + new_disk_conf = kzalloc(sizeof(struct disk_conf), GFP_KERNEL); + if (!new_disk_conf) { + put_ldev(mdev); + mutex_unlock(&mdev->tconn->conf_update); + dev_err(DEV, "Allocation of new disk_conf failed\n"); + return -ENOMEM; + } + + old_disk_conf = mdev->ldev->disk_conf; + *new_disk_conf = *old_disk_conf; + + new_disk_conf->resync_rate = be32_to_cpu(p->resync_rate); + } if (apv >= 88) { if (apv == 88) { @@ -2856,12 +3391,13 @@ static int receive_SyncParam(struct drbd_conf *mdev, enum drbd_packets cmd, unsi dev_err(DEV, "verify-alg of wrong size, " "peer wants %u, accepting only up to %u byte\n", data_size, SHARED_SECRET_MAX); - return false; + err = -EIO; + goto reconnect; } - if (drbd_recv(mdev, p->verify_alg, data_size) != data_size) - return false; - + err = drbd_recv_all(mdev->tconn, p->verify_alg, data_size); + if (err) + goto reconnect; /* we expect NUL terminated string */ /* but just in case someone tries to be evil */ D_ASSERT(p->verify_alg[data_size-1] == 0); @@ -2876,10 +3412,10 @@ static int receive_SyncParam(struct drbd_conf *mdev, enum drbd_packets cmd, unsi p->csums_alg[SHARED_SECRET_MAX-1] = 0; } - if (strcmp(mdev->sync_conf.verify_alg, p->verify_alg)) { + if (strcmp(old_net_conf->verify_alg, p->verify_alg)) { if (mdev->state.conn == C_WF_REPORT_PARAMS) { dev_err(DEV, "Different verify-alg settings. me=\"%s\" peer=\"%s\"\n", - mdev->sync_conf.verify_alg, p->verify_alg); + old_net_conf->verify_alg, p->verify_alg); goto disconnect; } verify_tfm = drbd_crypto_alloc_digest_safe(mdev, @@ -2890,10 +3426,10 @@ static int receive_SyncParam(struct drbd_conf *mdev, enum drbd_packets cmd, unsi } } - if (apv >= 89 && strcmp(mdev->sync_conf.csums_alg, p->csums_alg)) { + if (apv >= 89 && strcmp(old_net_conf->csums_alg, p->csums_alg)) { if (mdev->state.conn == C_WF_REPORT_PARAMS) { dev_err(DEV, "Different csums-alg settings. me=\"%s\" peer=\"%s\"\n", - mdev->sync_conf.csums_alg, p->csums_alg); + old_net_conf->csums_alg, p->csums_alg); goto disconnect; } csums_tfm = drbd_crypto_alloc_digest_safe(mdev, @@ -2904,57 +3440,91 @@ static int receive_SyncParam(struct drbd_conf *mdev, enum drbd_packets cmd, unsi } } - if (apv > 94) { - mdev->sync_conf.rate = be32_to_cpu(p->rate); - mdev->sync_conf.c_plan_ahead = be32_to_cpu(p->c_plan_ahead); - mdev->sync_conf.c_delay_target = be32_to_cpu(p->c_delay_target); - mdev->sync_conf.c_fill_target = be32_to_cpu(p->c_fill_target); - mdev->sync_conf.c_max_rate = be32_to_cpu(p->c_max_rate); - - fifo_size = (mdev->sync_conf.c_plan_ahead * 10 * SLEEP_TIME) / HZ; - if (fifo_size != mdev->rs_plan_s.size && fifo_size > 0) { - rs_plan_s = kzalloc(sizeof(int) * fifo_size, GFP_KERNEL); - if (!rs_plan_s) { + if (apv > 94 && new_disk_conf) { + new_disk_conf->c_plan_ahead = be32_to_cpu(p->c_plan_ahead); + new_disk_conf->c_delay_target = be32_to_cpu(p->c_delay_target); + new_disk_conf->c_fill_target = be32_to_cpu(p->c_fill_target); + new_disk_conf->c_max_rate = be32_to_cpu(p->c_max_rate); + + fifo_size = (new_disk_conf->c_plan_ahead * 10 * SLEEP_TIME) / HZ; + if (fifo_size != mdev->rs_plan_s->size) { + new_plan = fifo_alloc(fifo_size); + if (!new_plan) { dev_err(DEV, "kmalloc of fifo_buffer failed"); + put_ldev(mdev); goto disconnect; } } } - spin_lock(&mdev->peer_seq_lock); - /* lock against drbd_nl_syncer_conf() */ - if (verify_tfm) { - strcpy(mdev->sync_conf.verify_alg, p->verify_alg); - mdev->sync_conf.verify_alg_len = strlen(p->verify_alg) + 1; - crypto_free_hash(mdev->verify_tfm); - mdev->verify_tfm = verify_tfm; - dev_info(DEV, "using verify-alg: \"%s\"\n", p->verify_alg); - } - if (csums_tfm) { - strcpy(mdev->sync_conf.csums_alg, p->csums_alg); - mdev->sync_conf.csums_alg_len = strlen(p->csums_alg) + 1; - crypto_free_hash(mdev->csums_tfm); - mdev->csums_tfm = csums_tfm; - dev_info(DEV, "using csums-alg: \"%s\"\n", p->csums_alg); - } - if (fifo_size != mdev->rs_plan_s.size) { - kfree(mdev->rs_plan_s.values); - mdev->rs_plan_s.values = rs_plan_s; - mdev->rs_plan_s.size = fifo_size; - mdev->rs_planed = 0; + if (verify_tfm || csums_tfm) { + new_net_conf = kzalloc(sizeof(struct net_conf), GFP_KERNEL); + if (!new_net_conf) { + dev_err(DEV, "Allocation of new net_conf failed\n"); + goto disconnect; + } + + *new_net_conf = *old_net_conf; + + if (verify_tfm) { + strcpy(new_net_conf->verify_alg, p->verify_alg); + new_net_conf->verify_alg_len = strlen(p->verify_alg) + 1; + crypto_free_hash(mdev->tconn->verify_tfm); + mdev->tconn->verify_tfm = verify_tfm; + dev_info(DEV, "using verify-alg: \"%s\"\n", p->verify_alg); + } + if (csums_tfm) { + strcpy(new_net_conf->csums_alg, p->csums_alg); + new_net_conf->csums_alg_len = strlen(p->csums_alg) + 1; + crypto_free_hash(mdev->tconn->csums_tfm); + mdev->tconn->csums_tfm = csums_tfm; + dev_info(DEV, "using csums-alg: \"%s\"\n", p->csums_alg); + } + rcu_assign_pointer(tconn->net_conf, new_net_conf); } - spin_unlock(&mdev->peer_seq_lock); } - return ok; + if (new_disk_conf) { + rcu_assign_pointer(mdev->ldev->disk_conf, new_disk_conf); + put_ldev(mdev); + } + + if (new_plan) { + old_plan = mdev->rs_plan_s; + rcu_assign_pointer(mdev->rs_plan_s, new_plan); + } + + mutex_unlock(&mdev->tconn->conf_update); + synchronize_rcu(); + if (new_net_conf) + kfree(old_net_conf); + kfree(old_disk_conf); + kfree(old_plan); + + return 0; + +reconnect: + if (new_disk_conf) { + put_ldev(mdev); + kfree(new_disk_conf); + } + mutex_unlock(&mdev->tconn->conf_update); + return -EIO; + disconnect: + kfree(new_plan); + if (new_disk_conf) { + put_ldev(mdev); + kfree(new_disk_conf); + } + mutex_unlock(&mdev->tconn->conf_update); /* just for completeness: actually not needed, * as this is not reached if csums_tfm was ok. */ crypto_free_hash(csums_tfm); /* but free the verify_tfm again, if csums_tfm did not work out */ crypto_free_hash(verify_tfm); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - return false; + conn_request_state(mdev->tconn, NS(conn, C_DISCONNECTING), CS_HARD); + return -EIO; } /* warn if the arguments differ by more than 12.5% */ @@ -2970,59 +3540,77 @@ static void warn_if_differ_considerably(struct drbd_conf *mdev, (unsigned long long)a, (unsigned long long)b); } -static int receive_sizes(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_sizes(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_sizes *p = &mdev->data.rbuf.sizes; + struct drbd_conf *mdev; + struct p_sizes *p = pi->data; enum determine_dev_size dd = unchanged; sector_t p_size, p_usize, my_usize; int ldsc = 0; /* local disk size changed */ enum dds_flags ddsf; + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return config_unknown_volume(tconn, pi); + p_size = be64_to_cpu(p->d_size); p_usize = be64_to_cpu(p->u_size); - if (p_size == 0 && mdev->state.disk == D_DISKLESS) { - dev_err(DEV, "some backing storage is needed\n"); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - return false; - } - /* just store the peer's disk size for now. * we still need to figure out whether we accept that. */ mdev->p_size = p_size; if (get_ldev(mdev)) { + rcu_read_lock(); + my_usize = rcu_dereference(mdev->ldev->disk_conf)->disk_size; + rcu_read_unlock(); + warn_if_differ_considerably(mdev, "lower level device sizes", p_size, drbd_get_max_capacity(mdev->ldev)); warn_if_differ_considerably(mdev, "user requested size", - p_usize, mdev->ldev->dc.disk_size); + p_usize, my_usize); /* if this is the first connect, or an otherwise expected * param exchange, choose the minimum */ if (mdev->state.conn == C_WF_REPORT_PARAMS) - p_usize = min_not_zero((sector_t)mdev->ldev->dc.disk_size, - p_usize); - - my_usize = mdev->ldev->dc.disk_size; - - if (mdev->ldev->dc.disk_size != p_usize) { - mdev->ldev->dc.disk_size = p_usize; - dev_info(DEV, "Peer sets u_size to %lu sectors\n", - (unsigned long)mdev->ldev->dc.disk_size); - } + p_usize = min_not_zero(my_usize, p_usize); /* Never shrink a device with usable data during connect. But allow online shrinking if we are connected. */ - if (drbd_new_dev_size(mdev, mdev->ldev, 0) < - drbd_get_capacity(mdev->this_bdev) && - mdev->state.disk >= D_OUTDATED && - mdev->state.conn < C_CONNECTED) { + if (drbd_new_dev_size(mdev, mdev->ldev, p_usize, 0) < + drbd_get_capacity(mdev->this_bdev) && + mdev->state.disk >= D_OUTDATED && + mdev->state.conn < C_CONNECTED) { dev_err(DEV, "The peer's disk size is too small!\n"); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - mdev->ldev->dc.disk_size = my_usize; + conn_request_state(mdev->tconn, NS(conn, C_DISCONNECTING), CS_HARD); put_ldev(mdev); - return false; + return -EIO; + } + + if (my_usize != p_usize) { + struct disk_conf *old_disk_conf, *new_disk_conf = NULL; + + new_disk_conf = kzalloc(sizeof(struct disk_conf), GFP_KERNEL); + if (!new_disk_conf) { + dev_err(DEV, "Allocation of new disk_conf failed\n"); + put_ldev(mdev); + return -ENOMEM; + } + + mutex_lock(&mdev->tconn->conf_update); + old_disk_conf = mdev->ldev->disk_conf; + *new_disk_conf = *old_disk_conf; + new_disk_conf->disk_size = p_usize; + + rcu_assign_pointer(mdev->ldev->disk_conf, new_disk_conf); + mutex_unlock(&mdev->tconn->conf_update); + synchronize_rcu(); + kfree(old_disk_conf); + + dev_info(DEV, "Peer sets u_size to %lu sectors\n", + (unsigned long)my_usize); } + put_ldev(mdev); } @@ -3031,7 +3619,7 @@ static int receive_sizes(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned dd = drbd_determine_dev_size(mdev, ddsf); put_ldev(mdev); if (dd == dev_size_error) - return false; + return -EIO; drbd_md_sync(mdev); } else { /* I am diskless, need to accept the peer's size. */ @@ -3070,16 +3658,25 @@ static int receive_sizes(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned } } - return true; + return 0; } -static int receive_uuids(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_uuids(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_uuids *p = &mdev->data.rbuf.uuids; + struct drbd_conf *mdev; + struct p_uuids *p = pi->data; u64 *p_uuid; int i, updated_uuids = 0; + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return config_unknown_volume(tconn, pi); + p_uuid = kmalloc(sizeof(u64)*UI_EXTENDED_SIZE, GFP_NOIO); + if (!p_uuid) { + dev_err(DEV, "kmalloc of p_uuid failed\n"); + return false; + } for (i = UI_CURRENT; i < UI_EXTENDED_SIZE; i++) p_uuid[i] = be64_to_cpu(p->uuid[i]); @@ -3093,14 +3690,14 @@ static int receive_uuids(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned (mdev->ed_uuid & ~((u64)1)) != (p_uuid[UI_CURRENT] & ~((u64)1))) { dev_err(DEV, "Can only connect to data with current UUID=%016llX\n", (unsigned long long)mdev->ed_uuid); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - return false; + conn_request_state(mdev->tconn, NS(conn, C_DISCONNECTING), CS_HARD); + return -EIO; } if (get_ldev(mdev)) { int skip_initial_sync = mdev->state.conn == C_CONNECTED && - mdev->agreed_pro_version >= 90 && + mdev->tconn->agreed_pro_version >= 90 && mdev->ldev->md.uuid[UI_CURRENT] == UUID_JUST_CREATED && (p_uuid[UI_FLAGS] & 8); if (skip_initial_sync) { @@ -3127,14 +3724,15 @@ static int receive_uuids(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned ongoing cluster wide state change is finished. That is important if we are primary and are detaching from our disk. We need to see the new disk state... */ - wait_event(mdev->misc_wait, !test_bit(CLUSTER_ST_CHANGE, &mdev->flags)); + mutex_lock(mdev->state_mutex); + mutex_unlock(mdev->state_mutex); if (mdev->state.conn >= C_CONNECTED && mdev->state.disk < D_INCONSISTENT) updated_uuids |= drbd_set_ed_uuid(mdev, p_uuid[UI_CURRENT]); if (updated_uuids) drbd_print_uuids(mdev, "receiver updated UUIDs to"); - return true; + return 0; } /** @@ -3146,6 +3744,7 @@ static union drbd_state convert_state(union drbd_state ps) union drbd_state ms; static enum drbd_conns c_tab[] = { + [C_WF_REPORT_PARAMS] = C_WF_REPORT_PARAMS, [C_CONNECTED] = C_CONNECTED, [C_STARTING_SYNC_S] = C_STARTING_SYNC_T, @@ -3167,40 +3766,74 @@ static union drbd_state convert_state(union drbd_state ps) return ms; } -static int receive_req_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_req_state(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_req_state *p = &mdev->data.rbuf.req_state; + struct drbd_conf *mdev; + struct p_req_state *p = pi->data; union drbd_state mask, val; enum drbd_state_rv rv; + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; + mask.i = be32_to_cpu(p->mask); val.i = be32_to_cpu(p->val); - if (test_bit(DISCARD_CONCURRENT, &mdev->flags) && - test_bit(CLUSTER_ST_CHANGE, &mdev->flags)) { + if (test_bit(RESOLVE_CONFLICTS, &mdev->tconn->flags) && + mutex_is_locked(mdev->state_mutex)) { drbd_send_sr_reply(mdev, SS_CONCURRENT_ST_CHG); - return true; + return 0; } mask = convert_state(mask); val = convert_state(val); rv = drbd_change_state(mdev, CS_VERBOSE, mask, val); - drbd_send_sr_reply(mdev, rv); + drbd_md_sync(mdev); - return true; + return 0; } -static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_req_conn_state(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_state *p = &mdev->data.rbuf.state; + struct p_req_state *p = pi->data; + union drbd_state mask, val; + enum drbd_state_rv rv; + + mask.i = be32_to_cpu(p->mask); + val.i = be32_to_cpu(p->val); + + if (test_bit(RESOLVE_CONFLICTS, &tconn->flags) && + mutex_is_locked(&tconn->cstate_mutex)) { + conn_send_sr_reply(tconn, SS_CONCURRENT_ST_CHG); + return 0; + } + + mask = convert_state(mask); + val = convert_state(val); + + rv = conn_request_state(tconn, mask, val, CS_VERBOSE | CS_LOCAL_ONLY | CS_IGN_OUTD_FAIL); + conn_send_sr_reply(tconn, rv); + + return 0; +} + +static int receive_state(struct drbd_tconn *tconn, struct packet_info *pi) +{ + struct drbd_conf *mdev; + struct p_state *p = pi->data; union drbd_state os, ns, peer_state; enum drbd_disk_state real_peer_disk; enum chg_state_flags cs_flags; int rv; + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return config_unknown_volume(tconn, pi); + peer_state.i = be32_to_cpu(p->state); real_peer_disk = peer_state.disk; @@ -3209,16 +3842,16 @@ static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned dev_info(DEV, "real peer disk state = %s\n", drbd_disk_str(real_peer_disk)); } - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); retry: - os = ns = mdev->state; - spin_unlock_irq(&mdev->req_lock); + os = ns = drbd_read_state(mdev); + spin_unlock_irq(&mdev->tconn->req_lock); /* If some other part of the code (asender thread, timeout) * already decided to close the connection again, * we must not "re-establish" it here. */ if (os.conn <= C_TEAR_DOWN) - return false; + return -ECONNRESET; /* If this is the "end of sync" confirmation, usually the peer disk * transitions from D_INCONSISTENT to D_UP_TO_DATE. For empty (0 bits @@ -3246,10 +3879,18 @@ static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned peer_state.conn == C_CONNECTED) { if (drbd_bm_total_weight(mdev) <= mdev->rs_failed) drbd_resync_finished(mdev); - return true; + return 0; } } + /* explicit verify finished notification, stop sector reached. */ + if (os.conn == C_VERIFY_T && os.disk == D_UP_TO_DATE && + peer_state.conn == C_CONNECTED && real_peer_disk == D_UP_TO_DATE) { + ov_out_of_sync_print(mdev); + drbd_resync_finished(mdev); + return 0; + } + /* peer says his disk is inconsistent, while we think it is uptodate, * and this happens while the peer still thinks we have a sync going on, * but we think we are already done with the sync. @@ -3298,17 +3939,17 @@ static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned peer_state.disk = D_DISKLESS; real_peer_disk = D_DISKLESS; } else { - if (test_and_clear_bit(CONN_DRY_RUN, &mdev->flags)) - return false; + if (test_and_clear_bit(CONN_DRY_RUN, &mdev->tconn->flags)) + return -EIO; D_ASSERT(os.conn == C_WF_REPORT_PARAMS); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - return false; + conn_request_state(mdev->tconn, NS(conn, C_DISCONNECTING), CS_HARD); + return -EIO; } } } - spin_lock_irq(&mdev->req_lock); - if (mdev->state.i != os.i) + spin_lock_irq(&mdev->tconn->req_lock); + if (os.i != drbd_read_state(mdev).i) goto retry; clear_bit(CONSIDER_RESYNC, &mdev->flags); ns.peer = peer_state.role; @@ -3317,25 +3958,25 @@ static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned if ((ns.conn == C_CONNECTED || ns.conn == C_WF_BITMAP_S) && ns.disk == D_NEGOTIATING) ns.disk = mdev->new_state_tmp.disk; cs_flags = CS_VERBOSE + (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED ? 0 : CS_HARD); - if (ns.pdsk == D_CONSISTENT && is_susp(ns) && ns.conn == C_CONNECTED && os.conn < C_CONNECTED && + if (ns.pdsk == D_CONSISTENT && drbd_suspended(mdev) && ns.conn == C_CONNECTED && os.conn < C_CONNECTED && test_bit(NEW_CUR_UUID, &mdev->flags)) { - /* Do not allow tl_restart(resend) for a rebooted peer. We can only allow this + /* Do not allow tl_restart(RESEND) for a rebooted peer. We can only allow this for temporal network outages! */ - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); dev_err(DEV, "Aborting Connect, can not thaw IO with an only Consistent peer\n"); - tl_clear(mdev); + tl_clear(mdev->tconn); drbd_uuid_new_current(mdev); clear_bit(NEW_CUR_UUID, &mdev->flags); - drbd_force_state(mdev, NS2(conn, C_PROTOCOL_ERROR, susp, 0)); - return false; + conn_request_state(mdev->tconn, NS2(conn, C_PROTOCOL_ERROR, susp, 0), CS_HARD); + return -EIO; } rv = _drbd_set_state(mdev, ns, cs_flags, NULL); - ns = mdev->state; - spin_unlock_irq(&mdev->req_lock); + ns = drbd_read_state(mdev); + spin_unlock_irq(&mdev->tconn->req_lock); if (rv < SS_SUCCESS) { - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - return false; + conn_request_state(mdev->tconn, NS(conn, C_DISCONNECTING), CS_HARD); + return -EIO; } if (os.conn > C_WF_REPORT_PARAMS) { @@ -3349,16 +3990,21 @@ static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned } } - mdev->net_conf->want_lose = 0; + clear_bit(DISCARD_MY_DATA, &mdev->flags); drbd_md_sync(mdev); /* update connected indicator, la_size, ... */ - return true; + return 0; } -static int receive_sync_uuid(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_sync_uuid(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_rs_uuid *p = &mdev->data.rbuf.rs_uuid; + struct drbd_conf *mdev; + struct p_rs_uuid *p = pi->data; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; wait_event(mdev->misc_wait, mdev->state.conn == C_WF_SYNC_UUID || @@ -3381,7 +4027,7 @@ static int receive_sync_uuid(struct drbd_conf *mdev, enum drbd_packets cmd, unsi } else dev_err(DEV, "Ignoring SyncUUID packet!\n"); - return true; + return 0; } /** @@ -3391,27 +4037,27 @@ static int receive_sync_uuid(struct drbd_conf *mdev, enum drbd_packets cmd, unsi * code upon failure. */ static int -receive_bitmap_plain(struct drbd_conf *mdev, unsigned int data_size, - unsigned long *buffer, struct bm_xfer_ctx *c) +receive_bitmap_plain(struct drbd_conf *mdev, unsigned int size, + unsigned long *p, struct bm_xfer_ctx *c) { - unsigned num_words = min_t(size_t, BM_PACKET_WORDS, c->bm_words - c->word_offset); - unsigned want = num_words * sizeof(long); + unsigned int data_size = DRBD_SOCKET_BUFFER_SIZE - + drbd_header_size(mdev->tconn); + unsigned int num_words = min_t(size_t, data_size / sizeof(*p), + c->bm_words - c->word_offset); + unsigned int want = num_words * sizeof(*p); int err; - if (want != data_size) { - dev_err(DEV, "%s:want (%u) != data_size (%u)\n", __func__, want, data_size); + if (want != size) { + dev_err(DEV, "%s:want (%u) != size (%u)\n", __func__, want, size); return -EIO; } if (want == 0) return 0; - err = drbd_recv(mdev, buffer, want); - if (err != want) { - if (err >= 0) - err = -EIO; + err = drbd_recv_all(mdev->tconn, p, want); + if (err) return err; - } - drbd_bm_merge_lel(mdev, c->word_offset, num_words, buffer); + drbd_bm_merge_lel(mdev, c->word_offset, num_words, p); c->word_offset += num_words; c->bit_offset = c->word_offset * BITS_PER_LONG; @@ -3421,6 +4067,21 @@ receive_bitmap_plain(struct drbd_conf *mdev, unsigned int data_size, return 1; } +static enum drbd_bitmap_code dcbp_get_code(struct p_compressed_bm *p) +{ + return (enum drbd_bitmap_code)(p->encoding & 0x0f); +} + +static int dcbp_get_start(struct p_compressed_bm *p) +{ + return (p->encoding & 0x80) != 0; +} + +static int dcbp_get_pad_bits(struct p_compressed_bm *p) +{ + return (p->encoding >> 4) & 0x7; +} + /** * recv_bm_rle_bits * @@ -3430,7 +4091,8 @@ receive_bitmap_plain(struct drbd_conf *mdev, unsigned int data_size, static int recv_bm_rle_bits(struct drbd_conf *mdev, struct p_compressed_bm *p, - struct bm_xfer_ctx *c) + struct bm_xfer_ctx *c, + unsigned int len) { struct bitstream bs; u64 look_ahead; @@ -3438,12 +4100,11 @@ recv_bm_rle_bits(struct drbd_conf *mdev, u64 tmp; unsigned long s = c->bit_offset; unsigned long e; - int len = be16_to_cpu(p->head.length) - (sizeof(*p) - sizeof(p->head)); - int toggle = DCBP_get_start(p); + int toggle = dcbp_get_start(p); int have; int bits; - bitstream_init(&bs, p->code, len, DCBP_get_pad_bits(p)); + bitstream_init(&bs, p->code, len, dcbp_get_pad_bits(p)); bits = bitstream_get_bits(&bs, &look_ahead, 64); if (bits < 0) @@ -3495,17 +4156,18 @@ recv_bm_rle_bits(struct drbd_conf *mdev, static int decode_bitmap_c(struct drbd_conf *mdev, struct p_compressed_bm *p, - struct bm_xfer_ctx *c) + struct bm_xfer_ctx *c, + unsigned int len) { - if (DCBP_get_code(p) == RLE_VLI_Bits) - return recv_bm_rle_bits(mdev, p, c); + if (dcbp_get_code(p) == RLE_VLI_Bits) + return recv_bm_rle_bits(mdev, p, c, len - sizeof(*p)); /* other variants had been implemented for evaluation, * but have been dropped as this one turned out to be "best" * during all our tests. */ dev_err(DEV, "receive_bitmap_c: unknown encoding %u\n", p->encoding); - drbd_force_state(mdev, NS(conn, C_PROTOCOL_ERROR)); + conn_request_state(mdev->tconn, NS(conn, C_PROTOCOL_ERROR), CS_HARD); return -EIO; } @@ -3513,11 +4175,13 @@ void INFO_bm_xfer_stats(struct drbd_conf *mdev, const char *direction, struct bm_xfer_ctx *c) { /* what would it take to transfer it "plaintext" */ - unsigned plain = sizeof(struct p_header80) * - ((c->bm_words+BM_PACKET_WORDS-1)/BM_PACKET_WORDS+1) - + c->bm_words * sizeof(long); - unsigned total = c->bytes[0] + c->bytes[1]; - unsigned r; + unsigned int header_size = drbd_header_size(mdev->tconn); + unsigned int data_size = DRBD_SOCKET_BUFFER_SIZE - header_size; + unsigned int plain = + header_size * (DIV_ROUND_UP(c->bm_words, data_size) + 1) + + c->bm_words * sizeof(unsigned long); + unsigned int total = c->bytes[0] + c->bytes[1]; + unsigned int r; /* total can not be zero. but just in case: */ if (total == 0) @@ -3551,67 +4215,63 @@ void INFO_bm_xfer_stats(struct drbd_conf *mdev, in order to be agnostic to the 32 vs 64 bits issue. returns 0 on failure, 1 if we successfully received it. */ -static int receive_bitmap(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_bitmap(struct drbd_tconn *tconn, struct packet_info *pi) { + struct drbd_conf *mdev; struct bm_xfer_ctx c; - void *buffer; int err; - int ok = false; - struct p_header80 *h = &mdev->data.rbuf.header.h80; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; drbd_bm_lock(mdev, "receive bitmap", BM_LOCKED_SET_ALLOWED); /* you are supposed to send additional out-of-sync information * if you actually set bits during this phase */ - /* maybe we should use some per thread scratch page, - * and allocate that during initial device creation? */ - buffer = (unsigned long *) __get_free_page(GFP_NOIO); - if (!buffer) { - dev_err(DEV, "failed to allocate one page buffer in %s\n", __func__); - goto out; - } - c = (struct bm_xfer_ctx) { .bm_bits = drbd_bm_bits(mdev), .bm_words = drbd_bm_words(mdev), }; for(;;) { - if (cmd == P_BITMAP) { - err = receive_bitmap_plain(mdev, data_size, buffer, &c); - } else if (cmd == P_COMPRESSED_BITMAP) { + if (pi->cmd == P_BITMAP) + err = receive_bitmap_plain(mdev, pi->size, pi->data, &c); + else if (pi->cmd == P_COMPRESSED_BITMAP) { /* MAYBE: sanity check that we speak proto >= 90, * and the feature is enabled! */ - struct p_compressed_bm *p; + struct p_compressed_bm *p = pi->data; - if (data_size > BM_PACKET_PAYLOAD_BYTES) { + if (pi->size > DRBD_SOCKET_BUFFER_SIZE - drbd_header_size(tconn)) { dev_err(DEV, "ReportCBitmap packet too large\n"); + err = -EIO; goto out; } - /* use the page buff */ - p = buffer; - memcpy(p, h, sizeof(*h)); - if (drbd_recv(mdev, p->head.payload, data_size) != data_size) - goto out; - if (data_size <= (sizeof(*p) - sizeof(p->head))) { - dev_err(DEV, "ReportCBitmap packet too small (l:%u)\n", data_size); + if (pi->size <= sizeof(*p)) { + dev_err(DEV, "ReportCBitmap packet too small (l:%u)\n", pi->size); + err = -EIO; goto out; } - err = decode_bitmap_c(mdev, p, &c); + err = drbd_recv_all(mdev->tconn, p, pi->size); + if (err) + goto out; + err = decode_bitmap_c(mdev, p, &c, pi->size); } else { - dev_warn(DEV, "receive_bitmap: cmd neither ReportBitMap nor ReportCBitMap (is 0x%x)", cmd); + dev_warn(DEV, "receive_bitmap: cmd neither ReportBitMap nor ReportCBitMap (is 0x%x)", pi->cmd); + err = -EIO; goto out; } - c.packets[cmd == P_BITMAP]++; - c.bytes[cmd == P_BITMAP] += sizeof(struct p_header80) + data_size; + c.packets[pi->cmd == P_BITMAP]++; + c.bytes[pi->cmd == P_BITMAP] += drbd_header_size(tconn) + pi->size; if (err <= 0) { if (err < 0) goto out; break; } - if (!drbd_recv_header(mdev, &cmd, &data_size)) + err = drbd_recv_header(mdev->tconn, pi); + if (err) goto out; } @@ -3620,8 +4280,8 @@ static int receive_bitmap(struct drbd_conf *mdev, enum drbd_packets cmd, unsigne if (mdev->state.conn == C_WF_BITMAP_T) { enum drbd_state_rv rv; - ok = !drbd_send_bitmap(mdev); - if (!ok) + err = drbd_send_bitmap(mdev); + if (err) goto out; /* Omit CS_ORDERED with this state transition to avoid deadlocks. */ rv = _drbd_request_state(mdev, NS(conn, C_WF_SYNC_UUID), CS_VERBOSE); @@ -3632,47 +4292,40 @@ static int receive_bitmap(struct drbd_conf *mdev, enum drbd_packets cmd, unsigne dev_info(DEV, "unexpected cstate (%s) in receive_bitmap\n", drbd_conn_str(mdev->state.conn)); } + err = 0; - ok = true; out: drbd_bm_unlock(mdev); - if (ok && mdev->state.conn == C_WF_BITMAP_S) + if (!err && mdev->state.conn == C_WF_BITMAP_S) drbd_start_resync(mdev, C_SYNC_SOURCE); - free_page((unsigned long) buffer); - return ok; + return err; } -static int receive_skip(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_skip(struct drbd_tconn *tconn, struct packet_info *pi) { - /* TODO zero copy sink :) */ - static char sink[128]; - int size, want, r; + conn_warn(tconn, "skipping unknown optional packet type %d, l: %d!\n", + pi->cmd, pi->size); - dev_warn(DEV, "skipping unknown optional packet type %d, l: %d!\n", - cmd, data_size); - - size = data_size; - while (size > 0) { - want = min_t(int, size, sizeof(sink)); - r = drbd_recv(mdev, sink, want); - ERR_IF(r <= 0) break; - size -= r; - } - return size == 0; + return ignore_remaining_packet(tconn, pi); } -static int receive_UnplugRemote(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_UnplugRemote(struct drbd_tconn *tconn, struct packet_info *pi) { /* Make sure we've acked all the TCP data associated * with the data requests being unplugged */ - drbd_tcp_quickack(mdev->data.socket); + drbd_tcp_quickack(tconn->data.socket); - return true; + return 0; } -static int receive_out_of_sync(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned int data_size) +static int receive_out_of_sync(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_block_desc *p = &mdev->data.rbuf.block_desc; + struct drbd_conf *mdev; + struct p_block_desc *p = pi->data; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; switch (mdev->state.conn) { case C_WF_SYNC_UUID: @@ -3686,15 +4339,13 @@ static int receive_out_of_sync(struct drbd_conf *mdev, enum drbd_packets cmd, un drbd_set_out_of_sync(mdev, be64_to_cpu(p->sector), be32_to_cpu(p->blksize)); - return true; + return 0; } -typedef int (*drbd_cmd_handler_f)(struct drbd_conf *, enum drbd_packets cmd, unsigned int to_receive); - struct data_cmd { int expect_payload; size_t pkt_size; - drbd_cmd_handler_f function; + int (*fn)(struct drbd_tconn *, struct packet_info *); }; static struct data_cmd drbd_cmd_handler[] = { @@ -3702,13 +4353,13 @@ static struct data_cmd drbd_cmd_handler[] = { [P_DATA_REPLY] = { 1, sizeof(struct p_data), receive_DataReply }, [P_RS_DATA_REPLY] = { 1, sizeof(struct p_data), receive_RSDataReply } , [P_BARRIER] = { 0, sizeof(struct p_barrier), receive_Barrier } , - [P_BITMAP] = { 1, sizeof(struct p_header80), receive_bitmap } , - [P_COMPRESSED_BITMAP] = { 1, sizeof(struct p_header80), receive_bitmap } , - [P_UNPLUG_REMOTE] = { 0, sizeof(struct p_header80), receive_UnplugRemote }, + [P_BITMAP] = { 1, 0, receive_bitmap } , + [P_COMPRESSED_BITMAP] = { 1, 0, receive_bitmap } , + [P_UNPLUG_REMOTE] = { 0, 0, receive_UnplugRemote }, [P_DATA_REQUEST] = { 0, sizeof(struct p_block_req), receive_DataRequest }, [P_RS_DATA_REQUEST] = { 0, sizeof(struct p_block_req), receive_DataRequest }, - [P_SYNC_PARAM] = { 1, sizeof(struct p_header80), receive_SyncParam }, - [P_SYNC_PARAM89] = { 1, sizeof(struct p_header80), receive_SyncParam }, + [P_SYNC_PARAM] = { 1, 0, receive_SyncParam }, + [P_SYNC_PARAM89] = { 1, 0, receive_SyncParam }, [P_PROTOCOL] = { 1, sizeof(struct p_protocol), receive_protocol }, [P_UUIDS] = { 0, sizeof(struct p_uuids), receive_uuids }, [P_SIZES] = { 0, sizeof(struct p_sizes), receive_sizes }, @@ -3720,124 +4371,75 @@ static struct data_cmd drbd_cmd_handler[] = { [P_CSUM_RS_REQUEST] = { 1, sizeof(struct p_block_req), receive_DataRequest }, [P_DELAY_PROBE] = { 0, sizeof(struct p_delay_probe93), receive_skip }, [P_OUT_OF_SYNC] = { 0, sizeof(struct p_block_desc), receive_out_of_sync }, - /* anything missing from this table is in - * the asender_tbl, see get_asender_cmd */ - [P_MAX_CMD] = { 0, 0, NULL }, + [P_CONN_ST_CHG_REQ] = { 0, sizeof(struct p_req_state), receive_req_conn_state }, + [P_PROTOCOL_UPDATE] = { 1, sizeof(struct p_protocol), receive_protocol }, }; -/* All handler functions that expect a sub-header get that sub-heder in - mdev->data.rbuf.header.head.payload. - - Usually in mdev->data.rbuf.header.head the callback can find the usual - p_header, but they may not rely on that. Since there is also p_header95 ! - */ - -static void drbdd(struct drbd_conf *mdev) +static void drbdd(struct drbd_tconn *tconn) { - union p_header *header = &mdev->data.rbuf.header; - unsigned int packet_size; - enum drbd_packets cmd; + struct packet_info pi; size_t shs; /* sub header size */ - int rv; + int err; + + while (get_t_state(&tconn->receiver) == RUNNING) { + struct data_cmd *cmd; - while (get_t_state(&mdev->receiver) == Running) { - drbd_thread_current_set_cpu(mdev); - if (!drbd_recv_header(mdev, &cmd, &packet_size)) + drbd_thread_current_set_cpu(&tconn->receiver); + if (drbd_recv_header(tconn, &pi)) goto err_out; - if (unlikely(cmd >= P_MAX_CMD || !drbd_cmd_handler[cmd].function)) { - dev_err(DEV, "unknown packet type %d, l: %d!\n", cmd, packet_size); + cmd = &drbd_cmd_handler[pi.cmd]; + if (unlikely(pi.cmd >= ARRAY_SIZE(drbd_cmd_handler) || !cmd->fn)) { + conn_err(tconn, "Unexpected data packet %s (0x%04x)", + cmdname(pi.cmd), pi.cmd); goto err_out; } - shs = drbd_cmd_handler[cmd].pkt_size - sizeof(union p_header); - if (packet_size - shs > 0 && !drbd_cmd_handler[cmd].expect_payload) { - dev_err(DEV, "No payload expected %s l:%d\n", cmdname(cmd), packet_size); + shs = cmd->pkt_size; + if (pi.size > shs && !cmd->expect_payload) { + conn_err(tconn, "No payload expected %s l:%d\n", + cmdname(pi.cmd), pi.size); goto err_out; } if (shs) { - rv = drbd_recv(mdev, &header->h80.payload, shs); - if (unlikely(rv != shs)) { - if (!signal_pending(current)) - dev_warn(DEV, "short read while reading sub header: rv=%d\n", rv); + err = drbd_recv_all_warn(tconn, pi.data, shs); + if (err) goto err_out; - } + pi.size -= shs; } - rv = drbd_cmd_handler[cmd].function(mdev, cmd, packet_size - shs); - - if (unlikely(!rv)) { - dev_err(DEV, "error receiving %s, l: %d!\n", - cmdname(cmd), packet_size); + err = cmd->fn(tconn, &pi); + if (err) { + conn_err(tconn, "error receiving %s, e: %d l: %d!\n", + cmdname(pi.cmd), err, pi.size); goto err_out; } } + return; - if (0) { - err_out: - drbd_force_state(mdev, NS(conn, C_PROTOCOL_ERROR)); - } - /* If we leave here, we probably want to update at least the - * "Connected" indicator on stable storage. Do so explicitly here. */ - drbd_md_sync(mdev); + err_out: + conn_request_state(tconn, NS(conn, C_PROTOCOL_ERROR), CS_HARD); } -void drbd_flush_workqueue(struct drbd_conf *mdev) +void conn_flush_workqueue(struct drbd_tconn *tconn) { struct drbd_wq_barrier barr; barr.w.cb = w_prev_work_done; + barr.w.tconn = tconn; init_completion(&barr.done); - drbd_queue_work(&mdev->data.work, &barr.w); + drbd_queue_work(&tconn->sender_work, &barr.w); wait_for_completion(&barr.done); } -void drbd_free_tl_hash(struct drbd_conf *mdev) -{ - struct hlist_head *h; - - spin_lock_irq(&mdev->req_lock); - - if (!mdev->tl_hash || mdev->state.conn != C_STANDALONE) { - spin_unlock_irq(&mdev->req_lock); - return; - } - /* paranoia code */ - for (h = mdev->ee_hash; h < mdev->ee_hash + mdev->ee_hash_s; h++) - if (h->first) - dev_err(DEV, "ASSERT FAILED ee_hash[%u].first == %p, expected NULL\n", - (int)(h - mdev->ee_hash), h->first); - kfree(mdev->ee_hash); - mdev->ee_hash = NULL; - mdev->ee_hash_s = 0; - - /* We may not have had the chance to wait for all locally pending - * application requests. The hlist_add_fake() prevents access after - * free on master bio completion. */ - for (h = mdev->tl_hash; h < mdev->tl_hash + mdev->tl_hash_s; h++) { - struct drbd_request *req; - struct hlist_node *pos, *n; - hlist_for_each_entry_safe(req, pos, n, h, collision) { - hlist_del_init(&req->collision); - hlist_add_fake(&req->collision); - } - } - - kfree(mdev->tl_hash); - mdev->tl_hash = NULL; - mdev->tl_hash_s = 0; - spin_unlock_irq(&mdev->req_lock); -} - -static void drbd_disconnect(struct drbd_conf *mdev) +static void conn_disconnect(struct drbd_tconn *tconn) { - enum drbd_fencing_p fp; - union drbd_state os, ns; - int rv = SS_UNKNOWN_ERROR; - unsigned int i; + struct drbd_conf *mdev; + enum drbd_conns oc; + int vnr; - if (mdev->state.conn == C_STANDALONE) + if (tconn->cstate == C_STANDALONE) return; /* We are about to start the cleanup after connection loss. @@ -3845,18 +4447,54 @@ static void drbd_disconnect(struct drbd_conf *mdev) * Usually we should be in some network failure state already, * but just in case we are not, we fix it up here. */ - drbd_force_state(mdev, NS(conn, C_NETWORK_FAILURE)); + conn_request_state(tconn, NS(conn, C_NETWORK_FAILURE), CS_HARD); /* asender does not clean up anything. it must not interfere, either */ - drbd_thread_stop(&mdev->asender); - drbd_free_sock(mdev); + drbd_thread_stop(&tconn->asender); + drbd_free_sock(tconn); + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + kref_get(&mdev->kref); + rcu_read_unlock(); + drbd_disconnected(mdev); + kref_put(&mdev->kref, &drbd_minor_destroy); + rcu_read_lock(); + } + rcu_read_unlock(); + + if (!list_empty(&tconn->current_epoch->list)) + conn_err(tconn, "ASSERTION FAILED: tconn->current_epoch->list not empty\n"); + /* ok, no more ee's on the fly, it is safe to reset the epoch_size */ + atomic_set(&tconn->current_epoch->epoch_size, 0); + tconn->send.seen_any_write_yet = false; + + conn_info(tconn, "Connection closed\n"); + + if (conn_highest_role(tconn) == R_PRIMARY && conn_highest_pdsk(tconn) >= D_UNKNOWN) + conn_try_outdate_peer_async(tconn); + + spin_lock_irq(&tconn->req_lock); + oc = tconn->cstate; + if (oc >= C_UNCONNECTED) + _conn_request_state(tconn, NS(conn, C_UNCONNECTED), CS_VERBOSE); + + spin_unlock_irq(&tconn->req_lock); + + if (oc == C_DISCONNECTING) + conn_request_state(tconn, NS(conn, C_STANDALONE), CS_VERBOSE | CS_HARD); +} + +static int drbd_disconnected(struct drbd_conf *mdev) +{ + unsigned int i; /* wait for current activity to cease. */ - spin_lock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); _drbd_wait_ee_list_empty(mdev, &mdev->active_ee); _drbd_wait_ee_list_empty(mdev, &mdev->sync_ee); _drbd_wait_ee_list_empty(mdev, &mdev->read_ee); - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); /* We do not have data structures that would allow us to * get the rs_pending_cnt down to 0 again. @@ -3874,7 +4512,6 @@ static void drbd_disconnect(struct drbd_conf *mdev) atomic_set(&mdev->rs_pending_cnt, 0); wake_up(&mdev->misc_wait); - /* make sure syncer is stopped and w_resume_next_sg queued */ del_timer_sync(&mdev->resync_timer); resync_timer_fn((unsigned long)mdev); @@ -3883,50 +4520,25 @@ static void drbd_disconnect(struct drbd_conf *mdev) * to be "canceled" */ drbd_flush_workqueue(mdev); - /* This also does reclaim_net_ee(). If we do this too early, we might - * miss some resync ee and pages.*/ - drbd_process_done_ee(mdev); + drbd_finish_peer_reqs(mdev); + + /* This second workqueue flush is necessary, since drbd_finish_peer_reqs() + might have issued a work again. The one before drbd_finish_peer_reqs() is + necessary to reclain net_ee in drbd_finish_peer_reqs(). */ + drbd_flush_workqueue(mdev); + + /* need to do it again, drbd_finish_peer_reqs() may have populated it + * again via drbd_try_clear_on_disk_bm(). */ + drbd_rs_cancel_all(mdev); kfree(mdev->p_uuid); mdev->p_uuid = NULL; - if (!is_susp(mdev->state)) - tl_clear(mdev); - - dev_info(DEV, "Connection closed\n"); + if (!drbd_suspended(mdev)) + tl_clear(mdev->tconn); drbd_md_sync(mdev); - fp = FP_DONT_CARE; - if (get_ldev(mdev)) { - fp = mdev->ldev->dc.fencing; - put_ldev(mdev); - } - - if (mdev->state.role == R_PRIMARY && fp >= FP_RESOURCE && mdev->state.pdsk >= D_UNKNOWN) - drbd_try_outdate_peer_async(mdev); - - spin_lock_irq(&mdev->req_lock); - os = mdev->state; - if (os.conn >= C_UNCONNECTED) { - /* Do not restart in case we are C_DISCONNECTING */ - ns = os; - ns.conn = C_UNCONNECTED; - rv = _drbd_set_state(mdev, ns, CS_VERBOSE, NULL); - } - spin_unlock_irq(&mdev->req_lock); - - if (os.conn == C_DISCONNECTING) { - wait_event(mdev->net_cnt_wait, atomic_read(&mdev->net_cnt) == 0); - - crypto_free_hash(mdev->cram_hmac_tfm); - mdev->cram_hmac_tfm = NULL; - - kfree(mdev->net_conf); - mdev->net_conf = NULL; - drbd_request_state(mdev, NS(conn, C_STANDALONE)); - } - /* serialize with bitmap writeout triggered by the state change, * if any. */ wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags)); @@ -3938,7 +4550,7 @@ static void drbd_disconnect(struct drbd_conf *mdev) * Actually we don't care for exactly when the network stack does its * put_page(), but release our reference on these pages right here. */ - i = drbd_release_ee(mdev, &mdev->net_ee); + i = drbd_free_peer_reqs(mdev, &mdev->net_ee); if (i) dev_info(DEV, "net_ee not empty, killed %u entries\n", i); i = atomic_read(&mdev->pp_in_use_by_net); @@ -3953,9 +4565,7 @@ static void drbd_disconnect(struct drbd_conf *mdev) D_ASSERT(list_empty(&mdev->sync_ee)); D_ASSERT(list_empty(&mdev->done_ee)); - /* ok, no more ee's on the fly, it is safe to reset the epoch_size */ - atomic_set(&mdev->current_epoch->epoch_size, 0); - D_ASSERT(list_empty(&mdev->current_epoch->list)); + return 0; } /* @@ -3967,29 +4577,19 @@ static void drbd_disconnect(struct drbd_conf *mdev) * * for now, they are expected to be zero, but ignored. */ -static int drbd_send_handshake(struct drbd_conf *mdev) +static int drbd_send_features(struct drbd_tconn *tconn) { - /* ASSERT current == mdev->receiver ... */ - struct p_handshake *p = &mdev->data.sbuf.handshake; - int ok; - - if (mutex_lock_interruptible(&mdev->data.mutex)) { - dev_err(DEV, "interrupted during initial handshake\n"); - return 0; /* interrupted. not ok. */ - } - - if (mdev->data.socket == NULL) { - mutex_unlock(&mdev->data.mutex); - return 0; - } + struct drbd_socket *sock; + struct p_connection_features *p; + sock = &tconn->data; + p = conn_prepare_command(tconn, sock); + if (!p) + return -EIO; memset(p, 0, sizeof(*p)); p->protocol_min = cpu_to_be32(PRO_VERSION_MIN); p->protocol_max = cpu_to_be32(PRO_VERSION_MAX); - ok = _drbd_send_cmd( mdev, mdev->data.socket, P_HAND_SHAKE, - (struct p_header80 *)p, sizeof(*p), 0 ); - mutex_unlock(&mdev->data.mutex); - return ok; + return conn_send_command(tconn, sock, P_CONNECTION_FEATURES, sizeof(*p), NULL, 0); } /* @@ -3999,42 +4599,38 @@ static int drbd_send_handshake(struct drbd_conf *mdev) * -1 peer talks different language, * no point in trying again, please go standalone. */ -static int drbd_do_handshake(struct drbd_conf *mdev) +static int drbd_do_features(struct drbd_tconn *tconn) { - /* ASSERT current == mdev->receiver ... */ - struct p_handshake *p = &mdev->data.rbuf.handshake; - const int expect = sizeof(struct p_handshake) - sizeof(struct p_header80); - unsigned int length; - enum drbd_packets cmd; - int rv; + /* ASSERT current == tconn->receiver ... */ + struct p_connection_features *p; + const int expect = sizeof(struct p_connection_features); + struct packet_info pi; + int err; - rv = drbd_send_handshake(mdev); - if (!rv) + err = drbd_send_features(tconn); + if (err) return 0; - rv = drbd_recv_header(mdev, &cmd, &length); - if (!rv) + err = drbd_recv_header(tconn, &pi); + if (err) return 0; - if (cmd != P_HAND_SHAKE) { - dev_err(DEV, "expected HandShake packet, received: %s (0x%04x)\n", - cmdname(cmd), cmd); + if (pi.cmd != P_CONNECTION_FEATURES) { + conn_err(tconn, "expected ConnectionFeatures packet, received: %s (0x%04x)\n", + cmdname(pi.cmd), pi.cmd); return -1; } - if (length != expect) { - dev_err(DEV, "expected HandShake length: %u, received: %u\n", - expect, length); + if (pi.size != expect) { + conn_err(tconn, "expected ConnectionFeatures length: %u, received: %u\n", + expect, pi.size); return -1; } - rv = drbd_recv(mdev, &p->head.payload, expect); - - if (rv != expect) { - if (!signal_pending(current)) - dev_warn(DEV, "short read receiving handshake packet: l=%u\n", rv); + p = pi.data; + err = drbd_recv_all_warn(tconn, p, expect); + if (err) return 0; - } p->protocol_min = be32_to_cpu(p->protocol_min); p->protocol_max = be32_to_cpu(p->protocol_max); @@ -4045,15 +4641,15 @@ static int drbd_do_handshake(struct drbd_conf *mdev) PRO_VERSION_MIN > p->protocol_max) goto incompat; - mdev->agreed_pro_version = min_t(int, PRO_VERSION_MAX, p->protocol_max); + tconn->agreed_pro_version = min_t(int, PRO_VERSION_MAX, p->protocol_max); - dev_info(DEV, "Handshake successful: " - "Agreed network protocol version %d\n", mdev->agreed_pro_version); + conn_info(tconn, "Handshake successful: " + "Agreed network protocol version %d\n", tconn->agreed_pro_version); return 1; incompat: - dev_err(DEV, "incompatible DRBD dialects: " + conn_err(tconn, "incompatible DRBD dialects: " "I support %d-%d, peer supports %d-%d\n", PRO_VERSION_MIN, PRO_VERSION_MAX, p->protocol_min, p->protocol_max); @@ -4061,7 +4657,7 @@ static int drbd_do_handshake(struct drbd_conf *mdev) } #if !defined(CONFIG_CRYPTO_HMAC) && !defined(CONFIG_CRYPTO_HMAC_MODULE) -static int drbd_do_auth(struct drbd_conf *mdev) +static int drbd_do_auth(struct drbd_tconn *tconn) { dev_err(DEV, "This kernel was build without CONFIG_CRYPTO_HMAC.\n"); dev_err(DEV, "You need to disable 'cram-hmac-alg' in drbd.conf.\n"); @@ -4076,121 +4672,139 @@ static int drbd_do_auth(struct drbd_conf *mdev) -1 - auth failed, don't try again. */ -static int drbd_do_auth(struct drbd_conf *mdev) +static int drbd_do_auth(struct drbd_tconn *tconn) { + struct drbd_socket *sock; char my_challenge[CHALLENGE_LEN]; /* 64 Bytes... */ struct scatterlist sg; char *response = NULL; char *right_response = NULL; char *peers_ch = NULL; - unsigned int key_len = strlen(mdev->net_conf->shared_secret); + unsigned int key_len; + char secret[SHARED_SECRET_MAX]; /* 64 byte */ unsigned int resp_size; struct hash_desc desc; - enum drbd_packets cmd; - unsigned int length; - int rv; + struct packet_info pi; + struct net_conf *nc; + int err, rv; + + /* FIXME: Put the challenge/response into the preallocated socket buffer. */ - desc.tfm = mdev->cram_hmac_tfm; + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); + key_len = strlen(nc->shared_secret); + memcpy(secret, nc->shared_secret, key_len); + rcu_read_unlock(); + + desc.tfm = tconn->cram_hmac_tfm; desc.flags = 0; - rv = crypto_hash_setkey(mdev->cram_hmac_tfm, - (u8 *)mdev->net_conf->shared_secret, key_len); + rv = crypto_hash_setkey(tconn->cram_hmac_tfm, (u8 *)secret, key_len); if (rv) { - dev_err(DEV, "crypto_hash_setkey() failed with %d\n", rv); + conn_err(tconn, "crypto_hash_setkey() failed with %d\n", rv); rv = -1; goto fail; } get_random_bytes(my_challenge, CHALLENGE_LEN); - rv = drbd_send_cmd2(mdev, P_AUTH_CHALLENGE, my_challenge, CHALLENGE_LEN); + sock = &tconn->data; + if (!conn_prepare_command(tconn, sock)) { + rv = 0; + goto fail; + } + rv = !conn_send_command(tconn, sock, P_AUTH_CHALLENGE, 0, + my_challenge, CHALLENGE_LEN); if (!rv) goto fail; - rv = drbd_recv_header(mdev, &cmd, &length); - if (!rv) + err = drbd_recv_header(tconn, &pi); + if (err) { + rv = 0; goto fail; + } - if (cmd != P_AUTH_CHALLENGE) { - dev_err(DEV, "expected AuthChallenge packet, received: %s (0x%04x)\n", - cmdname(cmd), cmd); + if (pi.cmd != P_AUTH_CHALLENGE) { + conn_err(tconn, "expected AuthChallenge packet, received: %s (0x%04x)\n", + cmdname(pi.cmd), pi.cmd); rv = 0; goto fail; } - if (length > CHALLENGE_LEN * 2) { - dev_err(DEV, "expected AuthChallenge payload too big.\n"); + if (pi.size > CHALLENGE_LEN * 2) { + conn_err(tconn, "expected AuthChallenge payload too big.\n"); rv = -1; goto fail; } - peers_ch = kmalloc(length, GFP_NOIO); + peers_ch = kmalloc(pi.size, GFP_NOIO); if (peers_ch == NULL) { - dev_err(DEV, "kmalloc of peers_ch failed\n"); + conn_err(tconn, "kmalloc of peers_ch failed\n"); rv = -1; goto fail; } - rv = drbd_recv(mdev, peers_ch, length); - - if (rv != length) { - if (!signal_pending(current)) - dev_warn(DEV, "short read AuthChallenge: l=%u\n", rv); + err = drbd_recv_all_warn(tconn, peers_ch, pi.size); + if (err) { rv = 0; goto fail; } - resp_size = crypto_hash_digestsize(mdev->cram_hmac_tfm); + resp_size = crypto_hash_digestsize(tconn->cram_hmac_tfm); response = kmalloc(resp_size, GFP_NOIO); if (response == NULL) { - dev_err(DEV, "kmalloc of response failed\n"); + conn_err(tconn, "kmalloc of response failed\n"); rv = -1; goto fail; } sg_init_table(&sg, 1); - sg_set_buf(&sg, peers_ch, length); + sg_set_buf(&sg, peers_ch, pi.size); rv = crypto_hash_digest(&desc, &sg, sg.length, response); if (rv) { - dev_err(DEV, "crypto_hash_digest() failed with %d\n", rv); + conn_err(tconn, "crypto_hash_digest() failed with %d\n", rv); rv = -1; goto fail; } - rv = drbd_send_cmd2(mdev, P_AUTH_RESPONSE, response, resp_size); - if (!rv) + if (!conn_prepare_command(tconn, sock)) { + rv = 0; goto fail; - - rv = drbd_recv_header(mdev, &cmd, &length); + } + rv = !conn_send_command(tconn, sock, P_AUTH_RESPONSE, 0, + response, resp_size); if (!rv) goto fail; - if (cmd != P_AUTH_RESPONSE) { - dev_err(DEV, "expected AuthResponse packet, received: %s (0x%04x)\n", - cmdname(cmd), cmd); + err = drbd_recv_header(tconn, &pi); + if (err) { rv = 0; goto fail; } - if (length != resp_size) { - dev_err(DEV, "expected AuthResponse payload of wrong size\n"); + if (pi.cmd != P_AUTH_RESPONSE) { + conn_err(tconn, "expected AuthResponse packet, received: %s (0x%04x)\n", + cmdname(pi.cmd), pi.cmd); rv = 0; goto fail; } - rv = drbd_recv(mdev, response , resp_size); + if (pi.size != resp_size) { + conn_err(tconn, "expected AuthResponse payload of wrong size\n"); + rv = 0; + goto fail; + } - if (rv != resp_size) { - if (!signal_pending(current)) - dev_warn(DEV, "short read receiving AuthResponse: l=%u\n", rv); + err = drbd_recv_all_warn(tconn, response , resp_size); + if (err) { rv = 0; goto fail; } right_response = kmalloc(resp_size, GFP_NOIO); if (right_response == NULL) { - dev_err(DEV, "kmalloc of right_response failed\n"); + conn_err(tconn, "kmalloc of right_response failed\n"); rv = -1; goto fail; } @@ -4199,7 +4813,7 @@ static int drbd_do_auth(struct drbd_conf *mdev) rv = crypto_hash_digest(&desc, &sg, sg.length, right_response); if (rv) { - dev_err(DEV, "crypto_hash_digest() failed with %d\n", rv); + conn_err(tconn, "crypto_hash_digest() failed with %d\n", rv); rv = -1; goto fail; } @@ -4207,8 +4821,8 @@ static int drbd_do_auth(struct drbd_conf *mdev) rv = !memcmp(response, right_response, resp_size); if (rv) - dev_info(DEV, "Peer authenticated using %d bytes of '%s' HMAC\n", - resp_size, mdev->net_conf->cram_hmac_alg); + conn_info(tconn, "Peer authenticated using %d bytes HMAC\n", + resp_size); else rv = -1; @@ -4223,82 +4837,106 @@ static int drbd_do_auth(struct drbd_conf *mdev) int drbdd_init(struct drbd_thread *thi) { - struct drbd_conf *mdev = thi->mdev; - unsigned int minor = mdev_to_minor(mdev); + struct drbd_tconn *tconn = thi->tconn; int h; - sprintf(current->comm, "drbd%d_receiver", minor); - - dev_info(DEV, "receiver (re)started\n"); + conn_info(tconn, "receiver (re)started\n"); do { - h = drbd_connect(mdev); + h = conn_connect(tconn); if (h == 0) { - drbd_disconnect(mdev); + conn_disconnect(tconn); schedule_timeout_interruptible(HZ); } if (h == -1) { - dev_warn(DEV, "Discarding network configuration.\n"); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); + conn_warn(tconn, "Discarding network configuration.\n"); + conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD); } } while (h == 0); - if (h > 0) { - if (get_net_conf(mdev)) { - drbdd(mdev); - put_net_conf(mdev); - } - } + if (h > 0) + drbdd(tconn); - drbd_disconnect(mdev); + conn_disconnect(tconn); - dev_info(DEV, "receiver terminated\n"); + conn_info(tconn, "receiver terminated\n"); return 0; } /* ********* acknowledge sender ******** */ -static int got_RqSReply(struct drbd_conf *mdev, struct p_header80 *h) +static int got_conn_RqSReply(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_req_state_reply *p = (struct p_req_state_reply *)h; + struct p_req_state_reply *p = pi->data; + int retcode = be32_to_cpu(p->retcode); + + if (retcode >= SS_SUCCESS) { + set_bit(CONN_WD_ST_CHG_OKAY, &tconn->flags); + } else { + set_bit(CONN_WD_ST_CHG_FAIL, &tconn->flags); + conn_err(tconn, "Requested state change failed by peer: %s (%d)\n", + drbd_set_st_err_str(retcode), retcode); + } + wake_up(&tconn->ping_wait); + + return 0; +} +static int got_RqSReply(struct drbd_tconn *tconn, struct packet_info *pi) +{ + struct drbd_conf *mdev; + struct p_req_state_reply *p = pi->data; int retcode = be32_to_cpu(p->retcode); + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; + + if (test_bit(CONN_WD_ST_CHG_REQ, &tconn->flags)) { + D_ASSERT(tconn->agreed_pro_version < 100); + return got_conn_RqSReply(tconn, pi); + } + if (retcode >= SS_SUCCESS) { set_bit(CL_ST_CHG_SUCCESS, &mdev->flags); } else { set_bit(CL_ST_CHG_FAIL, &mdev->flags); dev_err(DEV, "Requested state change failed by peer: %s (%d)\n", - drbd_set_st_err_str(retcode), retcode); + drbd_set_st_err_str(retcode), retcode); } wake_up(&mdev->state_wait); - return true; + return 0; } -static int got_Ping(struct drbd_conf *mdev, struct p_header80 *h) +static int got_Ping(struct drbd_tconn *tconn, struct packet_info *pi) { - return drbd_send_ping_ack(mdev); + return drbd_send_ping_ack(tconn); } -static int got_PingAck(struct drbd_conf *mdev, struct p_header80 *h) +static int got_PingAck(struct drbd_tconn *tconn, struct packet_info *pi) { /* restore idle timeout */ - mdev->meta.socket->sk->sk_rcvtimeo = mdev->net_conf->ping_int*HZ; - if (!test_and_set_bit(GOT_PING_ACK, &mdev->flags)) - wake_up(&mdev->misc_wait); + tconn->meta.socket->sk->sk_rcvtimeo = tconn->net_conf->ping_int*HZ; + if (!test_and_set_bit(GOT_PING_ACK, &tconn->flags)) + wake_up(&tconn->ping_wait); - return true; + return 0; } -static int got_IsInSync(struct drbd_conf *mdev, struct p_header80 *h) +static int got_IsInSync(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_block_ack *p = (struct p_block_ack *)h; + struct drbd_conf *mdev; + struct p_block_ack *p = pi->data; sector_t sector = be64_to_cpu(p->sector); int blksize = be32_to_cpu(p->blksize); - D_ASSERT(mdev->agreed_pro_version >= 89); + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; + + D_ASSERT(mdev->tconn->agreed_pro_version >= 89); update_peer_seq(mdev, be32_to_cpu(p->seq_num)); @@ -4312,162 +4950,139 @@ static int got_IsInSync(struct drbd_conf *mdev, struct p_header80 *h) dec_rs_pending(mdev); atomic_add(blksize >> 9, &mdev->rs_sect_in); - return true; -} - -/* when we receive the ACK for a write request, - * verify that we actually know about it */ -static struct drbd_request *_ack_id_to_req(struct drbd_conf *mdev, - u64 id, sector_t sector) -{ - struct hlist_head *slot = tl_hash_slot(mdev, sector); - struct hlist_node *n; - struct drbd_request *req; - - hlist_for_each_entry(req, n, slot, collision) { - if ((unsigned long)req == (unsigned long)id) { - if (req->sector != sector) { - dev_err(DEV, "_ack_id_to_req: found req %p but it has " - "wrong sector (%llus versus %llus)\n", req, - (unsigned long long)req->sector, - (unsigned long long)sector); - break; - } - return req; - } - } - return NULL; + return 0; } -typedef struct drbd_request *(req_validator_fn) - (struct drbd_conf *mdev, u64 id, sector_t sector); - -static int validate_req_change_req_state(struct drbd_conf *mdev, - u64 id, sector_t sector, req_validator_fn validator, - const char *func, enum drbd_req_event what) +static int +validate_req_change_req_state(struct drbd_conf *mdev, u64 id, sector_t sector, + struct rb_root *root, const char *func, + enum drbd_req_event what, bool missing_ok) { struct drbd_request *req; struct bio_and_error m; - spin_lock_irq(&mdev->req_lock); - req = validator(mdev, id, sector); + spin_lock_irq(&mdev->tconn->req_lock); + req = find_request(mdev, root, id, sector, missing_ok, func); if (unlikely(!req)) { - spin_unlock_irq(&mdev->req_lock); - - dev_err(DEV, "%s: failed to find req %p, sector %llus\n", func, - (void *)(unsigned long)id, (unsigned long long)sector); - return false; + spin_unlock_irq(&mdev->tconn->req_lock); + return -EIO; } __req_mod(req, what, &m); - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); if (m.bio) complete_master_bio(mdev, &m); - return true; + return 0; } -static int got_BlockAck(struct drbd_conf *mdev, struct p_header80 *h) +static int got_BlockAck(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_block_ack *p = (struct p_block_ack *)h; + struct drbd_conf *mdev; + struct p_block_ack *p = pi->data; sector_t sector = be64_to_cpu(p->sector); int blksize = be32_to_cpu(p->blksize); enum drbd_req_event what; + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; + update_peer_seq(mdev, be32_to_cpu(p->seq_num)); - if (is_syncer_block_id(p->block_id)) { + if (p->block_id == ID_SYNCER) { drbd_set_in_sync(mdev, sector, blksize); dec_rs_pending(mdev); - return true; + return 0; } - switch (be16_to_cpu(h->command)) { + switch (pi->cmd) { case P_RS_WRITE_ACK: - D_ASSERT(mdev->net_conf->wire_protocol == DRBD_PROT_C); - what = write_acked_by_peer_and_sis; + what = WRITE_ACKED_BY_PEER_AND_SIS; break; case P_WRITE_ACK: - D_ASSERT(mdev->net_conf->wire_protocol == DRBD_PROT_C); - what = write_acked_by_peer; + what = WRITE_ACKED_BY_PEER; break; case P_RECV_ACK: - D_ASSERT(mdev->net_conf->wire_protocol == DRBD_PROT_B); - what = recv_acked_by_peer; + what = RECV_ACKED_BY_PEER; break; - case P_DISCARD_ACK: - D_ASSERT(mdev->net_conf->wire_protocol == DRBD_PROT_C); - what = conflict_discarded_by_peer; + case P_SUPERSEDED: + what = CONFLICT_RESOLVED; + break; + case P_RETRY_WRITE: + what = POSTPONE_WRITE; break; default: - D_ASSERT(0); - return false; + BUG(); } return validate_req_change_req_state(mdev, p->block_id, sector, - _ack_id_to_req, __func__ , what); + &mdev->write_requests, __func__, + what, false); } -static int got_NegAck(struct drbd_conf *mdev, struct p_header80 *h) +static int got_NegAck(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_block_ack *p = (struct p_block_ack *)h; + struct drbd_conf *mdev; + struct p_block_ack *p = pi->data; sector_t sector = be64_to_cpu(p->sector); int size = be32_to_cpu(p->blksize); - struct drbd_request *req; - struct bio_and_error m; + int err; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; update_peer_seq(mdev, be32_to_cpu(p->seq_num)); - if (is_syncer_block_id(p->block_id)) { + if (p->block_id == ID_SYNCER) { dec_rs_pending(mdev); drbd_rs_failed_io(mdev, sector, size); - return true; + return 0; } - spin_lock_irq(&mdev->req_lock); - req = _ack_id_to_req(mdev, p->block_id, sector); - if (!req) { - spin_unlock_irq(&mdev->req_lock); - if (mdev->net_conf->wire_protocol == DRBD_PROT_A || - mdev->net_conf->wire_protocol == DRBD_PROT_B) { - /* Protocol A has no P_WRITE_ACKs, but has P_NEG_ACKs. - The master bio might already be completed, therefore the - request is no longer in the collision hash. - => Do not try to validate block_id as request. */ - /* In Protocol B we might already have got a P_RECV_ACK - but then get a P_NEG_ACK after wards. */ - drbd_set_out_of_sync(mdev, sector, size); - return true; - } else { - dev_err(DEV, "%s: failed to find req %p, sector %llus\n", __func__, - (void *)(unsigned long)p->block_id, (unsigned long long)sector); - return false; - } + err = validate_req_change_req_state(mdev, p->block_id, sector, + &mdev->write_requests, __func__, + NEG_ACKED, true); + if (err) { + /* Protocol A has no P_WRITE_ACKs, but has P_NEG_ACKs. + The master bio might already be completed, therefore the + request is no longer in the collision hash. */ + /* In Protocol B we might already have got a P_RECV_ACK + but then get a P_NEG_ACK afterwards. */ + drbd_set_out_of_sync(mdev, sector, size); } - __req_mod(req, neg_acked, &m); - spin_unlock_irq(&mdev->req_lock); - - if (m.bio) - complete_master_bio(mdev, &m); - return true; + return 0; } -static int got_NegDReply(struct drbd_conf *mdev, struct p_header80 *h) +static int got_NegDReply(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_block_ack *p = (struct p_block_ack *)h; + struct drbd_conf *mdev; + struct p_block_ack *p = pi->data; sector_t sector = be64_to_cpu(p->sector); + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; + update_peer_seq(mdev, be32_to_cpu(p->seq_num)); - dev_err(DEV, "Got NegDReply; Sector %llus, len %u; Fail original request.\n", + + dev_err(DEV, "Got NegDReply; Sector %llus, len %u.\n", (unsigned long long)sector, be32_to_cpu(p->blksize)); return validate_req_change_req_state(mdev, p->block_id, sector, - _ar_id_to_req, __func__ , neg_acked); + &mdev->read_requests, __func__, + NEG_ACKED, false); } -static int got_NegRSDReply(struct drbd_conf *mdev, struct p_header80 *h) +static int got_NegRSDReply(struct drbd_tconn *tconn, struct packet_info *pi) { + struct drbd_conf *mdev; sector_t sector; int size; - struct p_block_ack *p = (struct p_block_ack *)h; + struct p_block_ack *p = pi->data; + + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; sector = be64_to_cpu(p->sector); size = be32_to_cpu(p->blksize); @@ -4478,57 +5093,66 @@ static int got_NegRSDReply(struct drbd_conf *mdev, struct p_header80 *h) if (get_ldev_if_state(mdev, D_FAILED)) { drbd_rs_complete_io(mdev, sector); - switch (be16_to_cpu(h->command)) { + switch (pi->cmd) { case P_NEG_RS_DREPLY: drbd_rs_failed_io(mdev, sector, size); case P_RS_CANCEL: break; default: - D_ASSERT(0); - put_ldev(mdev); - return false; + BUG(); } put_ldev(mdev); } - return true; + return 0; } -static int got_BarrierAck(struct drbd_conf *mdev, struct p_header80 *h) +static int got_BarrierAck(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_barrier_ack *p = (struct p_barrier_ack *)h; - - tl_release(mdev, p->barrier, be32_to_cpu(p->set_size)); - - if (mdev->state.conn == C_AHEAD && - atomic_read(&mdev->ap_in_flight) == 0 && - !test_and_set_bit(AHEAD_TO_SYNC_SOURCE, &mdev->flags)) { - mdev->start_resync_timer.expires = jiffies + HZ; - add_timer(&mdev->start_resync_timer); + struct p_barrier_ack *p = pi->data; + struct drbd_conf *mdev; + int vnr; + + tl_release(tconn, p->barrier, be32_to_cpu(p->set_size)); + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + if (mdev->state.conn == C_AHEAD && + atomic_read(&mdev->ap_in_flight) == 0 && + !test_and_set_bit(AHEAD_TO_SYNC_SOURCE, &mdev->flags)) { + mdev->start_resync_timer.expires = jiffies + HZ; + add_timer(&mdev->start_resync_timer); + } } + rcu_read_unlock(); - return true; + return 0; } -static int got_OVResult(struct drbd_conf *mdev, struct p_header80 *h) +static int got_OVResult(struct drbd_tconn *tconn, struct packet_info *pi) { - struct p_block_ack *p = (struct p_block_ack *)h; + struct drbd_conf *mdev; + struct p_block_ack *p = pi->data; struct drbd_work *w; sector_t sector; int size; + mdev = vnr_to_mdev(tconn, pi->vnr); + if (!mdev) + return -EIO; + sector = be64_to_cpu(p->sector); size = be32_to_cpu(p->blksize); update_peer_seq(mdev, be32_to_cpu(p->seq_num)); if (be64_to_cpu(p->block_id) == ID_OUT_OF_SYNC) - drbd_ov_oos_found(mdev, sector, size); + drbd_ov_out_of_sync_found(mdev, sector, size); else - ov_oos_print(mdev); + ov_out_of_sync_print(mdev); if (!get_ldev(mdev)) - return true; + return 0; drbd_rs_complete_io(mdev, sector); dec_rs_pending(mdev); @@ -4543,114 +5167,137 @@ static int got_OVResult(struct drbd_conf *mdev, struct p_header80 *h) w = kmalloc(sizeof(*w), GFP_NOIO); if (w) { w->cb = w_ov_finished; - drbd_queue_work_front(&mdev->data.work, w); + w->mdev = mdev; + drbd_queue_work(&mdev->tconn->sender_work, w); } else { dev_err(DEV, "kmalloc(w) failed."); - ov_oos_print(mdev); + ov_out_of_sync_print(mdev); drbd_resync_finished(mdev); } } put_ldev(mdev); - return true; + return 0; +} + +static int got_skip(struct drbd_tconn *tconn, struct packet_info *pi) +{ + return 0; } -static int got_skip(struct drbd_conf *mdev, struct p_header80 *h) +static int tconn_finish_peer_reqs(struct drbd_tconn *tconn) { - return true; + struct drbd_conf *mdev; + int vnr, not_empty = 0; + + do { + clear_bit(SIGNAL_ASENDER, &tconn->flags); + flush_signals(current); + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + kref_get(&mdev->kref); + rcu_read_unlock(); + if (drbd_finish_peer_reqs(mdev)) { + kref_put(&mdev->kref, &drbd_minor_destroy); + return 1; + } + kref_put(&mdev->kref, &drbd_minor_destroy); + rcu_read_lock(); + } + set_bit(SIGNAL_ASENDER, &tconn->flags); + + spin_lock_irq(&tconn->req_lock); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + not_empty = !list_empty(&mdev->done_ee); + if (not_empty) + break; + } + spin_unlock_irq(&tconn->req_lock); + rcu_read_unlock(); + } while (not_empty); + + return 0; } struct asender_cmd { size_t pkt_size; - int (*process)(struct drbd_conf *mdev, struct p_header80 *h); + int (*fn)(struct drbd_tconn *tconn, struct packet_info *); }; -static struct asender_cmd *get_asender_cmd(int cmd) -{ - static struct asender_cmd asender_tbl[] = { - /* anything missing from this table is in - * the drbd_cmd_handler (drbd_default_handler) table, - * see the beginning of drbdd() */ - [P_PING] = { sizeof(struct p_header80), got_Ping }, - [P_PING_ACK] = { sizeof(struct p_header80), got_PingAck }, +static struct asender_cmd asender_tbl[] = { + [P_PING] = { 0, got_Ping }, + [P_PING_ACK] = { 0, got_PingAck }, [P_RECV_ACK] = { sizeof(struct p_block_ack), got_BlockAck }, [P_WRITE_ACK] = { sizeof(struct p_block_ack), got_BlockAck }, [P_RS_WRITE_ACK] = { sizeof(struct p_block_ack), got_BlockAck }, - [P_DISCARD_ACK] = { sizeof(struct p_block_ack), got_BlockAck }, + [P_SUPERSEDED] = { sizeof(struct p_block_ack), got_BlockAck }, [P_NEG_ACK] = { sizeof(struct p_block_ack), got_NegAck }, [P_NEG_DREPLY] = { sizeof(struct p_block_ack), got_NegDReply }, - [P_NEG_RS_DREPLY] = { sizeof(struct p_block_ack), got_NegRSDReply}, + [P_NEG_RS_DREPLY] = { sizeof(struct p_block_ack), got_NegRSDReply }, [P_OV_RESULT] = { sizeof(struct p_block_ack), got_OVResult }, [P_BARRIER_ACK] = { sizeof(struct p_barrier_ack), got_BarrierAck }, [P_STATE_CHG_REPLY] = { sizeof(struct p_req_state_reply), got_RqSReply }, [P_RS_IS_IN_SYNC] = { sizeof(struct p_block_ack), got_IsInSync }, [P_DELAY_PROBE] = { sizeof(struct p_delay_probe93), got_skip }, - [P_RS_CANCEL] = { sizeof(struct p_block_ack), got_NegRSDReply}, - [P_MAX_CMD] = { 0, NULL }, - }; - if (cmd > P_MAX_CMD || asender_tbl[cmd].process == NULL) - return NULL; - return &asender_tbl[cmd]; -} + [P_RS_CANCEL] = { sizeof(struct p_block_ack), got_NegRSDReply }, + [P_CONN_ST_CHG_REPLY]={ sizeof(struct p_req_state_reply), got_conn_RqSReply }, + [P_RETRY_WRITE] = { sizeof(struct p_block_ack), got_BlockAck }, +}; int drbd_asender(struct drbd_thread *thi) { - struct drbd_conf *mdev = thi->mdev; - struct p_header80 *h = &mdev->meta.rbuf.header.h80; + struct drbd_tconn *tconn = thi->tconn; struct asender_cmd *cmd = NULL; - - int rv, len; - void *buf = h; + struct packet_info pi; + int rv; + void *buf = tconn->meta.rbuf; int received = 0; - int expect = sizeof(struct p_header80); - int empty; - int ping_timeout_active = 0; - - sprintf(current->comm, "drbd%d_asender", mdev_to_minor(mdev)); + unsigned int header_size = drbd_header_size(tconn); + int expect = header_size; + bool ping_timeout_active = false; + struct net_conf *nc; + int ping_timeo, tcp_cork, ping_int; current->policy = SCHED_RR; /* Make this a realtime task! */ current->rt_priority = 2; /* more important than all other tasks */ - while (get_t_state(thi) == Running) { - drbd_thread_current_set_cpu(mdev); - if (test_and_clear_bit(SEND_PING, &mdev->flags)) { - ERR_IF(!drbd_send_ping(mdev)) goto reconnect; - mdev->meta.socket->sk->sk_rcvtimeo = - mdev->net_conf->ping_timeo*HZ/10; - ping_timeout_active = 1; - } + while (get_t_state(thi) == RUNNING) { + drbd_thread_current_set_cpu(thi); - /* conditionally cork; - * it may hurt latency if we cork without much to send */ - if (!mdev->net_conf->no_cork && - 3 < atomic_read(&mdev->unacked_cnt)) - drbd_tcp_cork(mdev->meta.socket); - while (1) { - clear_bit(SIGNAL_ASENDER, &mdev->flags); - flush_signals(current); - if (!drbd_process_done_ee(mdev)) + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); + ping_timeo = nc->ping_timeo; + tcp_cork = nc->tcp_cork; + ping_int = nc->ping_int; + rcu_read_unlock(); + + if (test_and_clear_bit(SEND_PING, &tconn->flags)) { + if (drbd_send_ping(tconn)) { + conn_err(tconn, "drbd_send_ping has failed\n"); goto reconnect; - /* to avoid race with newly queued ACKs */ - set_bit(SIGNAL_ASENDER, &mdev->flags); - spin_lock_irq(&mdev->req_lock); - empty = list_empty(&mdev->done_ee); - spin_unlock_irq(&mdev->req_lock); - /* new ack may have been queued right here, - * but then there is also a signal pending, - * and we start over... */ - if (empty) - break; + } + tconn->meta.socket->sk->sk_rcvtimeo = ping_timeo * HZ / 10; + ping_timeout_active = true; + } + + /* TODO: conditionally cork; it may hurt latency if we cork without + much to send */ + if (tcp_cork) + drbd_tcp_cork(tconn->meta.socket); + if (tconn_finish_peer_reqs(tconn)) { + conn_err(tconn, "tconn_finish_peer_reqs() failed\n"); + goto reconnect; } /* but unconditionally uncork unless disabled */ - if (!mdev->net_conf->no_cork) - drbd_tcp_uncork(mdev->meta.socket); + if (tcp_cork) + drbd_tcp_uncork(tconn->meta.socket); /* short circuit, recv_msg would return EINTR anyways. */ if (signal_pending(current)) continue; - rv = drbd_recv_short(mdev, mdev->meta.socket, - buf, expect-received, 0); - clear_bit(SIGNAL_ASENDER, &mdev->flags); + rv = drbd_recv_short(tconn->meta.socket, buf, expect-received, 0); + clear_bit(SIGNAL_ASENDER, &tconn->flags); flush_signals(current); @@ -4668,80 +5315,91 @@ int drbd_asender(struct drbd_thread *thi) received += rv; buf += rv; } else if (rv == 0) { - dev_err(DEV, "meta connection shut down by peer.\n"); + if (test_bit(DISCONNECT_SENT, &tconn->flags)) { + long t; + rcu_read_lock(); + t = rcu_dereference(tconn->net_conf)->ping_timeo * HZ/10; + rcu_read_unlock(); + + t = wait_event_timeout(tconn->ping_wait, + tconn->cstate < C_WF_REPORT_PARAMS, + t); + if (t) + break; + } + conn_err(tconn, "meta connection shut down by peer.\n"); goto reconnect; } else if (rv == -EAGAIN) { /* If the data socket received something meanwhile, * that is good enough: peer is still alive. */ - if (time_after(mdev->last_received, - jiffies - mdev->meta.socket->sk->sk_rcvtimeo)) + if (time_after(tconn->last_received, + jiffies - tconn->meta.socket->sk->sk_rcvtimeo)) continue; if (ping_timeout_active) { - dev_err(DEV, "PingAck did not arrive in time.\n"); + conn_err(tconn, "PingAck did not arrive in time.\n"); goto reconnect; } - set_bit(SEND_PING, &mdev->flags); + set_bit(SEND_PING, &tconn->flags); continue; } else if (rv == -EINTR) { continue; } else { - dev_err(DEV, "sock_recvmsg returned %d\n", rv); + conn_err(tconn, "sock_recvmsg returned %d\n", rv); goto reconnect; } if (received == expect && cmd == NULL) { - if (unlikely(h->magic != BE_DRBD_MAGIC)) { - dev_err(DEV, "magic?? on meta m: 0x%08x c: %d l: %d\n", - be32_to_cpu(h->magic), - be16_to_cpu(h->command), - be16_to_cpu(h->length)); + if (decode_header(tconn, tconn->meta.rbuf, &pi)) goto reconnect; - } - cmd = get_asender_cmd(be16_to_cpu(h->command)); - len = be16_to_cpu(h->length); - if (unlikely(cmd == NULL)) { - dev_err(DEV, "unknown command?? on meta m: 0x%08x c: %d l: %d\n", - be32_to_cpu(h->magic), - be16_to_cpu(h->command), - be16_to_cpu(h->length)); + cmd = &asender_tbl[pi.cmd]; + if (pi.cmd >= ARRAY_SIZE(asender_tbl) || !cmd->fn) { + conn_err(tconn, "Unexpected meta packet %s (0x%04x)\n", + cmdname(pi.cmd), pi.cmd); goto disconnect; } - expect = cmd->pkt_size; - ERR_IF(len != expect-sizeof(struct p_header80)) + expect = header_size + cmd->pkt_size; + if (pi.size != expect - header_size) { + conn_err(tconn, "Wrong packet size on meta (c: %d, l: %d)\n", + pi.cmd, pi.size); goto reconnect; + } } if (received == expect) { - mdev->last_received = jiffies; - D_ASSERT(cmd != NULL); - if (!cmd->process(mdev, h)) + bool err; + + err = cmd->fn(tconn, &pi); + if (err) { + conn_err(tconn, "%pf failed\n", cmd->fn); goto reconnect; + } + + tconn->last_received = jiffies; - /* the idle_timeout (ping-int) - * has been restored in got_PingAck() */ - if (cmd == get_asender_cmd(P_PING_ACK)) - ping_timeout_active = 0; + if (cmd == &asender_tbl[P_PING_ACK]) { + /* restore idle timeout */ + tconn->meta.socket->sk->sk_rcvtimeo = ping_int * HZ; + ping_timeout_active = false; + } - buf = h; + buf = tconn->meta.rbuf; received = 0; - expect = sizeof(struct p_header80); + expect = header_size; cmd = NULL; } } if (0) { reconnect: - drbd_force_state(mdev, NS(conn, C_NETWORK_FAILURE)); - drbd_md_sync(mdev); + conn_request_state(tconn, NS(conn, C_NETWORK_FAILURE), CS_HARD); + conn_md_sync(tconn); } if (0) { disconnect: - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - drbd_md_sync(mdev); + conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD); } - clear_bit(SIGNAL_ASENDER, &mdev->flags); + clear_bit(SIGNAL_ASENDER, &tconn->flags); - D_ASSERT(mdev->state.conn < C_CONNECTED); - dev_info(DEV, "asender terminated\n"); + conn_info(tconn, "asender terminated\n"); return 0; } diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c index 01b2ac641c7b..f58a4a4b4dfb 100644 --- a/drivers/block/drbd/drbd_req.c +++ b/drivers/block/drbd/drbd_req.c @@ -31,6 +31,8 @@ #include "drbd_req.h" +static bool drbd_may_do_local_read(struct drbd_conf *mdev, sector_t sector, int size); + /* Update disk stats at start of I/O request */ static void _drbd_start_io_acct(struct drbd_conf *mdev, struct drbd_request *req, struct bio *bio) { @@ -40,6 +42,8 @@ static void _drbd_start_io_acct(struct drbd_conf *mdev, struct drbd_request *req part_round_stats(cpu, &mdev->vdisk->part0); part_stat_inc(cpu, &mdev->vdisk->part0, ios[rw]); part_stat_add(cpu, &mdev->vdisk->part0, sectors[rw], bio_sectors(bio)); + (void) cpu; /* The macro invocations above want the cpu argument, I do not like + the compiler warning about cpu only assigned but never used... */ part_inc_in_flight(&mdev->vdisk->part0, rw); part_stat_unlock(); } @@ -57,9 +61,51 @@ static void _drbd_end_io_acct(struct drbd_conf *mdev, struct drbd_request *req) part_stat_unlock(); } -static void _req_is_done(struct drbd_conf *mdev, struct drbd_request *req, const int rw) +static struct drbd_request *drbd_req_new(struct drbd_conf *mdev, + struct bio *bio_src) +{ + struct drbd_request *req; + + req = mempool_alloc(drbd_request_mempool, GFP_NOIO); + if (!req) + return NULL; + + drbd_req_make_private_bio(req, bio_src); + req->rq_state = bio_data_dir(bio_src) == WRITE ? RQ_WRITE : 0; + req->w.mdev = mdev; + req->master_bio = bio_src; + req->epoch = 0; + + drbd_clear_interval(&req->i); + req->i.sector = bio_src->bi_sector; + req->i.size = bio_src->bi_size; + req->i.local = true; + req->i.waiting = false; + + INIT_LIST_HEAD(&req->tl_requests); + INIT_LIST_HEAD(&req->w.list); + + /* one reference to be put by __drbd_make_request */ + atomic_set(&req->completion_ref, 1); + /* one kref as long as completion_ref > 0 */ + kref_init(&req->kref); + return req; +} + +void drbd_req_destroy(struct kref *kref) { - const unsigned long s = req->rq_state; + struct drbd_request *req = container_of(kref, struct drbd_request, kref); + struct drbd_conf *mdev = req->w.mdev; + const unsigned s = req->rq_state; + + if ((req->master_bio && !(s & RQ_POSTPONED)) || + atomic_read(&req->completion_ref) || + (s & RQ_LOCAL_PENDING) || + ((s & RQ_NET_MASK) && !(s & RQ_NET_DONE))) { + dev_err(DEV, "drbd_req_destroy: Logic BUG rq_state = 0x%x, completion_ref = %d\n", + s, atomic_read(&req->completion_ref)); + return; + } /* remove it from the transfer log. * well, only if it had been there in the first @@ -67,24 +113,33 @@ static void _req_is_done(struct drbd_conf *mdev, struct drbd_request *req, const * and never sent), it should still be "empty" as * initialized in drbd_req_new(), so we can list_del() it * here unconditionally */ - list_del(&req->tl_requests); + list_del_init(&req->tl_requests); /* if it was a write, we may have to set the corresponding * bit(s) out-of-sync first. If it had a local part, we need to * release the reference to the activity log. */ - if (rw == WRITE) { + if (s & RQ_WRITE) { /* Set out-of-sync unless both OK flags are set * (local only or remote failed). * Other places where we set out-of-sync: * READ with local io-error */ - if (!(s & RQ_NET_OK) || !(s & RQ_LOCAL_OK)) - drbd_set_out_of_sync(mdev, req->sector, req->size); - if ((s & RQ_NET_OK) && (s & RQ_LOCAL_OK) && (s & RQ_NET_SIS)) - drbd_set_in_sync(mdev, req->sector, req->size); + /* There is a special case: + * we may notice late that IO was suspended, + * and postpone, or schedule for retry, a write, + * before it even was submitted or sent. + * In that case we do not want to touch the bitmap at all. + */ + if ((s & (RQ_POSTPONED|RQ_LOCAL_MASK|RQ_NET_MASK)) != RQ_POSTPONED) { + if (!(s & RQ_NET_OK) || !(s & RQ_LOCAL_OK)) + drbd_set_out_of_sync(mdev, req->i.sector, req->i.size); + + if ((s & RQ_NET_OK) && (s & RQ_LOCAL_OK) && (s & RQ_NET_SIS)) + drbd_set_in_sync(mdev, req->i.sector, req->i.size); + } /* one might be tempted to move the drbd_al_complete_io - * to the local io completion callback drbd_endio_pri. + * to the local io completion callback drbd_request_endio. * but, if this was a mirror write, we may only * drbd_al_complete_io after this is RQ_NET_DONE, * otherwise the extent could be dropped from the al @@ -93,109 +148,35 @@ static void _req_is_done(struct drbd_conf *mdev, struct drbd_request *req, const * but after the extent has been dropped from the al, * we would forget to resync the corresponding extent. */ - if (s & RQ_LOCAL_MASK) { + if (s & RQ_IN_ACT_LOG) { if (get_ldev_if_state(mdev, D_FAILED)) { - if (s & RQ_IN_ACT_LOG) - drbd_al_complete_io(mdev, req->sector); + drbd_al_complete_io(mdev, &req->i); put_ldev(mdev); } else if (__ratelimit(&drbd_ratelimit_state)) { - dev_warn(DEV, "Should have called drbd_al_complete_io(, %llu), " - "but my Disk seems to have failed :(\n", - (unsigned long long) req->sector); + dev_warn(DEV, "Should have called drbd_al_complete_io(, %llu, %u), " + "but my Disk seems to have failed :(\n", + (unsigned long long) req->i.sector, req->i.size); } } } - drbd_req_free(req); + mempool_free(req, drbd_request_mempool); } -static void queue_barrier(struct drbd_conf *mdev) -{ - struct drbd_tl_epoch *b; - - /* We are within the req_lock. Once we queued the barrier for sending, - * we set the CREATE_BARRIER bit. It is cleared as soon as a new - * barrier/epoch object is added. This is the only place this bit is - * set. It indicates that the barrier for this epoch is already queued, - * and no new epoch has been created yet. */ - if (test_bit(CREATE_BARRIER, &mdev->flags)) - return; - - b = mdev->newest_tle; - b->w.cb = w_send_barrier; - /* inc_ap_pending done here, so we won't - * get imbalanced on connection loss. - * dec_ap_pending will be done in got_BarrierAck - * or (on connection loss) in tl_clear. */ - inc_ap_pending(mdev); - drbd_queue_work(&mdev->data.work, &b->w); - set_bit(CREATE_BARRIER, &mdev->flags); +static void wake_all_senders(struct drbd_tconn *tconn) { + wake_up(&tconn->sender_work.q_wait); } -static void _about_to_complete_local_write(struct drbd_conf *mdev, - struct drbd_request *req) +/* must hold resource->req_lock */ +static void start_new_tl_epoch(struct drbd_tconn *tconn) { - const unsigned long s = req->rq_state; - struct drbd_request *i; - struct drbd_epoch_entry *e; - struct hlist_node *n; - struct hlist_head *slot; - - /* Before we can signal completion to the upper layers, - * we may need to close the current epoch. - * We can skip this, if this request has not even been sent, because we - * did not have a fully established connection yet/anymore, during - * bitmap exchange, or while we are C_AHEAD due to congestion policy. - */ - if (mdev->state.conn >= C_CONNECTED && - (s & RQ_NET_SENT) != 0 && - req->epoch == mdev->newest_tle->br_number) - queue_barrier(mdev); - - /* we need to do the conflict detection stuff, - * if we have the ee_hash (two_primaries) and - * this has been on the network */ - if ((s & RQ_NET_DONE) && mdev->ee_hash != NULL) { - const sector_t sector = req->sector; - const int size = req->size; - - /* ASSERT: - * there must be no conflicting requests, since - * they must have been failed on the spot */ -#define OVERLAPS overlaps(sector, size, i->sector, i->size) - slot = tl_hash_slot(mdev, sector); - hlist_for_each_entry(i, n, slot, collision) { - if (OVERLAPS) { - dev_alert(DEV, "LOGIC BUG: completed: %p %llus +%u; " - "other: %p %llus +%u\n", - req, (unsigned long long)sector, size, - i, (unsigned long long)i->sector, i->size); - } - } + /* no point closing an epoch, if it is empty, anyways. */ + if (tconn->current_tle_writes == 0) + return; - /* maybe "wake" those conflicting epoch entries - * that wait for this request to finish. - * - * currently, there can be only _one_ such ee - * (well, or some more, which would be pending - * P_DISCARD_ACK not yet sent by the asender...), - * since we block the receiver thread upon the - * first conflict detection, which will wait on - * misc_wait. maybe we want to assert that? - * - * anyways, if we found one, - * we just have to do a wake_up. */ -#undef OVERLAPS -#define OVERLAPS overlaps(sector, size, e->sector, e->size) - slot = ee_hash_slot(mdev, req->sector); - hlist_for_each_entry(e, n, slot, collision) { - if (OVERLAPS) { - wake_up(&mdev->misc_wait); - break; - } - } - } -#undef OVERLAPS + tconn->current_tle_writes = 0; + atomic_inc(&tconn->current_tle_nr); + wake_all_senders(tconn); } void complete_master_bio(struct drbd_conf *mdev, @@ -205,17 +186,33 @@ void complete_master_bio(struct drbd_conf *mdev, dec_ap_bio(mdev); } + +static void drbd_remove_request_interval(struct rb_root *root, + struct drbd_request *req) +{ + struct drbd_conf *mdev = req->w.mdev; + struct drbd_interval *i = &req->i; + + drbd_remove_interval(root, i); + + /* Wake up any processes waiting for this request to complete. */ + if (i->waiting) + wake_up(&mdev->misc_wait); +} + /* Helper for __req_mod(). * Set m->bio to the master bio, if it is fit to be completed, * or leave it alone (it is initialized to NULL in __req_mod), * if it has already been completed, or cannot be completed yet. * If m->bio is set, the error status to be returned is placed in m->error. */ -void _req_may_be_done(struct drbd_request *req, struct bio_and_error *m) +static +void drbd_req_complete(struct drbd_request *req, struct bio_and_error *m) { - const unsigned long s = req->rq_state; - struct drbd_conf *mdev = req->mdev; - int rw = req->rq_state & RQ_WRITE ? WRITE : READ; + const unsigned s = req->rq_state; + struct drbd_conf *mdev = req->w.mdev; + int rw; + int error, ok; /* we must not complete the master bio, while it is * still being processed by _drbd_send_zc_bio (drbd_send_dblock) @@ -226,165 +223,220 @@ void _req_may_be_done(struct drbd_request *req, struct bio_and_error *m) * the receiver, * the bio_endio completion callbacks. */ - if (s & RQ_NET_QUEUED) - return; - if (s & RQ_NET_PENDING) + if ((s & RQ_LOCAL_PENDING && !(s & RQ_LOCAL_ABORTED)) || + (s & RQ_NET_QUEUED) || (s & RQ_NET_PENDING) || + (s & RQ_COMPLETION_SUSP)) { + dev_err(DEV, "drbd_req_complete: Logic BUG rq_state = 0x%x\n", s); return; - if (s & RQ_LOCAL_PENDING && !(s & RQ_LOCAL_ABORTED)) + } + + if (!req->master_bio) { + dev_err(DEV, "drbd_req_complete: Logic BUG, master_bio == NULL!\n"); return; + } - if (req->master_bio) { - /* this is data_received (remote read) - * or protocol C P_WRITE_ACK - * or protocol B P_RECV_ACK - * or protocol A "handed_over_to_network" (SendAck) - * or canceled or failed, - * or killed from the transfer log due to connection loss. - */ + rw = bio_rw(req->master_bio); - /* - * figure out whether to report success or failure. - * - * report success when at least one of the operations succeeded. - * or, to put the other way, - * only report failure, when both operations failed. - * - * what to do about the failures is handled elsewhere. - * what we need to do here is just: complete the master_bio. - * - * local completion error, if any, has been stored as ERR_PTR - * in private_bio within drbd_endio_pri. - */ - int ok = (s & RQ_LOCAL_OK) || (s & RQ_NET_OK); - int error = PTR_ERR(req->private_bio); + /* + * figure out whether to report success or failure. + * + * report success when at least one of the operations succeeded. + * or, to put the other way, + * only report failure, when both operations failed. + * + * what to do about the failures is handled elsewhere. + * what we need to do here is just: complete the master_bio. + * + * local completion error, if any, has been stored as ERR_PTR + * in private_bio within drbd_request_endio. + */ + ok = (s & RQ_LOCAL_OK) || (s & RQ_NET_OK); + error = PTR_ERR(req->private_bio); - /* remove the request from the conflict detection - * respective block_id verification hash */ - if (!hlist_unhashed(&req->collision)) - hlist_del(&req->collision); - else - D_ASSERT((s & (RQ_NET_MASK & ~RQ_NET_DONE)) == 0); + /* remove the request from the conflict detection + * respective block_id verification hash */ + if (!drbd_interval_empty(&req->i)) { + struct rb_root *root; - /* for writes we need to do some extra housekeeping */ if (rw == WRITE) - _about_to_complete_local_write(mdev, req); + root = &mdev->write_requests; + else + root = &mdev->read_requests; + drbd_remove_request_interval(root, req); + } else if (!(s & RQ_POSTPONED)) + D_ASSERT((s & (RQ_NET_MASK & ~RQ_NET_DONE)) == 0); - /* Update disk stats */ - _drbd_end_io_acct(mdev, req); + /* Before we can signal completion to the upper layers, + * we may need to close the current transfer log epoch. + * We are within the request lock, so we can simply compare + * the request epoch number with the current transfer log + * epoch number. If they match, increase the current_tle_nr, + * and reset the transfer log epoch write_cnt. + */ + if (rw == WRITE && + req->epoch == atomic_read(&mdev->tconn->current_tle_nr)) + start_new_tl_epoch(mdev->tconn); + + /* Update disk stats */ + _drbd_end_io_acct(mdev, req); + + /* If READ failed, + * have it be pushed back to the retry work queue, + * so it will re-enter __drbd_make_request(), + * and be re-assigned to a suitable local or remote path, + * or failed if we do not have access to good data anymore. + * + * Unless it was failed early by __drbd_make_request(), + * because no path was available, in which case + * it was not even added to the transfer_log. + * + * READA may fail, and will not be retried. + * + * WRITE should have used all available paths already. + */ + if (!ok && rw == READ && !list_empty(&req->tl_requests)) + req->rq_state |= RQ_POSTPONED; + if (!(req->rq_state & RQ_POSTPONED)) { m->error = ok ? 0 : (error ?: -EIO); m->bio = req->master_bio; req->master_bio = NULL; } +} - if (s & RQ_LOCAL_PENDING) - return; +static int drbd_req_put_completion_ref(struct drbd_request *req, struct bio_and_error *m, int put) +{ + struct drbd_conf *mdev = req->w.mdev; + D_ASSERT(m || (req->rq_state & RQ_POSTPONED)); + + if (!atomic_sub_and_test(put, &req->completion_ref)) + return 0; - if ((s & RQ_NET_MASK) == 0 || (s & RQ_NET_DONE)) { - /* this is disconnected (local only) operation, - * or protocol C P_WRITE_ACK, - * or protocol A or B P_BARRIER_ACK, - * or killed from the transfer log due to connection loss. */ - _req_is_done(mdev, req, rw); + drbd_req_complete(req, m); + + if (req->rq_state & RQ_POSTPONED) { + /* don't destroy the req object just yet, + * but queue it for retry */ + drbd_restart_request(req); + return 0; } - /* else: network part and not DONE yet. that is - * protocol A or B, barrier ack still pending... */ + + return 1; } -static void _req_may_be_done_not_susp(struct drbd_request *req, struct bio_and_error *m) +/* I'd like this to be the only place that manipulates + * req->completion_ref and req->kref. */ +static void mod_rq_state(struct drbd_request *req, struct bio_and_error *m, + int clear, int set) { - struct drbd_conf *mdev = req->mdev; + struct drbd_conf *mdev = req->w.mdev; + unsigned s = req->rq_state; + int c_put = 0; + int k_put = 0; - if (!is_susp(mdev->state)) - _req_may_be_done(req, m); -} + if (drbd_suspended(mdev) && !((s | clear) & RQ_COMPLETION_SUSP)) + set |= RQ_COMPLETION_SUSP; -/* - * checks whether there was an overlapping request - * or ee already registered. - * - * if so, return 1, in which case this request is completed on the spot, - * without ever being submitted or send. - * - * return 0 if it is ok to submit this request. - * - * NOTE: - * paranoia: assume something above us is broken, and issues different write - * requests for the same block simultaneously... - * - * To ensure these won't be reordered differently on both nodes, resulting in - * diverging data sets, we discard the later one(s). Not that this is supposed - * to happen, but this is the rationale why we also have to check for - * conflicting requests with local origin, and why we have to do so regardless - * of whether we allowed multiple primaries. - * - * BTW, in case we only have one primary, the ee_hash is empty anyways, and the - * second hlist_for_each_entry becomes a noop. This is even simpler than to - * grab a reference on the net_conf, and check for the two_primaries flag... - */ -static int _req_conflicts(struct drbd_request *req) -{ - struct drbd_conf *mdev = req->mdev; - const sector_t sector = req->sector; - const int size = req->size; - struct drbd_request *i; - struct drbd_epoch_entry *e; - struct hlist_node *n; - struct hlist_head *slot; + /* apply */ - D_ASSERT(hlist_unhashed(&req->collision)); + req->rq_state &= ~clear; + req->rq_state |= set; - if (!get_net_conf(mdev)) - return 0; + /* no change? */ + if (req->rq_state == s) + return; - /* BUG_ON */ - ERR_IF (mdev->tl_hash_s == 0) - goto out_no_conflict; - BUG_ON(mdev->tl_hash == NULL); - -#define OVERLAPS overlaps(i->sector, i->size, sector, size) - slot = tl_hash_slot(mdev, sector); - hlist_for_each_entry(i, n, slot, collision) { - if (OVERLAPS) { - dev_alert(DEV, "%s[%u] Concurrent local write detected! " - "[DISCARD L] new: %llus +%u; " - "pending: %llus +%u\n", - current->comm, current->pid, - (unsigned long long)sector, size, - (unsigned long long)i->sector, i->size); - goto out_conflict; - } + /* intent: get references */ + + if (!(s & RQ_LOCAL_PENDING) && (set & RQ_LOCAL_PENDING)) + atomic_inc(&req->completion_ref); + + if (!(s & RQ_NET_PENDING) && (set & RQ_NET_PENDING)) { + inc_ap_pending(mdev); + atomic_inc(&req->completion_ref); } - if (mdev->ee_hash_s) { - /* now, check for overlapping requests with remote origin */ - BUG_ON(mdev->ee_hash == NULL); -#undef OVERLAPS -#define OVERLAPS overlaps(e->sector, e->size, sector, size) - slot = ee_hash_slot(mdev, sector); - hlist_for_each_entry(e, n, slot, collision) { - if (OVERLAPS) { - dev_alert(DEV, "%s[%u] Concurrent remote write detected!" - " [DISCARD L] new: %llus +%u; " - "pending: %llus +%u\n", - current->comm, current->pid, - (unsigned long long)sector, size, - (unsigned long long)e->sector, e->size); - goto out_conflict; - } - } + if (!(s & RQ_NET_QUEUED) && (set & RQ_NET_QUEUED)) + atomic_inc(&req->completion_ref); + + if (!(s & RQ_EXP_BARR_ACK) && (set & RQ_EXP_BARR_ACK)) + kref_get(&req->kref); /* wait for the DONE */ + + if (!(s & RQ_NET_SENT) && (set & RQ_NET_SENT)) + atomic_add(req->i.size >> 9, &mdev->ap_in_flight); + + if (!(s & RQ_COMPLETION_SUSP) && (set & RQ_COMPLETION_SUSP)) + atomic_inc(&req->completion_ref); + + /* progress: put references */ + + if ((s & RQ_COMPLETION_SUSP) && (clear & RQ_COMPLETION_SUSP)) + ++c_put; + + if (!(s & RQ_LOCAL_ABORTED) && (set & RQ_LOCAL_ABORTED)) { + D_ASSERT(req->rq_state & RQ_LOCAL_PENDING); + /* local completion may still come in later, + * we need to keep the req object around. */ + kref_get(&req->kref); + ++c_put; + } + + if ((s & RQ_LOCAL_PENDING) && (clear & RQ_LOCAL_PENDING)) { + if (req->rq_state & RQ_LOCAL_ABORTED) + ++k_put; + else + ++c_put; } -#undef OVERLAPS -out_no_conflict: - /* this is like it should be, and what we expected. - * our users do behave after all... */ - put_net_conf(mdev); - return 0; + if ((s & RQ_NET_PENDING) && (clear & RQ_NET_PENDING)) { + dec_ap_pending(mdev); + ++c_put; + } -out_conflict: - put_net_conf(mdev); - return 1; + if ((s & RQ_NET_QUEUED) && (clear & RQ_NET_QUEUED)) + ++c_put; + + if ((s & RQ_EXP_BARR_ACK) && !(s & RQ_NET_DONE) && (set & RQ_NET_DONE)) { + if (req->rq_state & RQ_NET_SENT) + atomic_sub(req->i.size >> 9, &mdev->ap_in_flight); + ++k_put; + } + + /* potentially complete and destroy */ + + if (k_put || c_put) { + /* Completion does it's own kref_put. If we are going to + * kref_sub below, we need req to be still around then. */ + int at_least = k_put + !!c_put; + int refcount = atomic_read(&req->kref.refcount); + if (refcount < at_least) + dev_err(DEV, + "mod_rq_state: Logic BUG: %x -> %x: refcount = %d, should be >= %d\n", + s, req->rq_state, refcount, at_least); + } + + /* If we made progress, retry conflicting peer requests, if any. */ + if (req->i.waiting) + wake_up(&mdev->misc_wait); + + if (c_put) + k_put += drbd_req_put_completion_ref(req, m, c_put); + if (k_put) + kref_sub(&req->kref, k_put, drbd_req_destroy); +} + +static void drbd_report_io_error(struct drbd_conf *mdev, struct drbd_request *req) +{ + char b[BDEVNAME_SIZE]; + + if (!__ratelimit(&drbd_ratelimit_state)) + return; + + dev_warn(DEV, "local %s IO error sector %llu+%u on %s\n", + (req->rq_state & RQ_WRITE) ? "WRITE" : "READ", + (unsigned long long)req->i.sector, + req->i.size >> 9, + bdevname(mdev->ldev->backing_bdev, b)); } /* obviously this could be coded as many single functions @@ -402,9 +454,12 @@ out_conflict: int __req_mod(struct drbd_request *req, enum drbd_req_event what, struct bio_and_error *m) { - struct drbd_conf *mdev = req->mdev; - int rv = 0; - m->bio = NULL; + struct drbd_conf *mdev = req->w.mdev; + struct net_conf *nc; + int p, rv = 0; + + if (m) + m->bio = NULL; switch (what) { default: @@ -413,116 +468,91 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what, /* does not happen... * initialization done in drbd_req_new - case created: + case CREATED: break; */ - case to_be_send: /* via network */ - /* reached via drbd_make_request_common + case TO_BE_SENT: /* via network */ + /* reached via __drbd_make_request * and from w_read_retry_remote */ D_ASSERT(!(req->rq_state & RQ_NET_MASK)); - req->rq_state |= RQ_NET_PENDING; - inc_ap_pending(mdev); + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + p = nc->wire_protocol; + rcu_read_unlock(); + req->rq_state |= + p == DRBD_PROT_C ? RQ_EXP_WRITE_ACK : + p == DRBD_PROT_B ? RQ_EXP_RECEIVE_ACK : 0; + mod_rq_state(req, m, 0, RQ_NET_PENDING); break; - case to_be_submitted: /* locally */ - /* reached via drbd_make_request_common */ + case TO_BE_SUBMITTED: /* locally */ + /* reached via __drbd_make_request */ D_ASSERT(!(req->rq_state & RQ_LOCAL_MASK)); - req->rq_state |= RQ_LOCAL_PENDING; + mod_rq_state(req, m, 0, RQ_LOCAL_PENDING); break; - case completed_ok: + case COMPLETED_OK: if (req->rq_state & RQ_WRITE) - mdev->writ_cnt += req->size>>9; + mdev->writ_cnt += req->i.size >> 9; else - mdev->read_cnt += req->size>>9; + mdev->read_cnt += req->i.size >> 9; - req->rq_state |= (RQ_LOCAL_COMPLETED|RQ_LOCAL_OK); - req->rq_state &= ~RQ_LOCAL_PENDING; - - _req_may_be_done_not_susp(req, m); + mod_rq_state(req, m, RQ_LOCAL_PENDING, + RQ_LOCAL_COMPLETED|RQ_LOCAL_OK); break; - case abort_disk_io: - req->rq_state |= RQ_LOCAL_ABORTED; - if (req->rq_state & RQ_WRITE) - _req_may_be_done_not_susp(req, m); - else - goto goto_queue_for_net_read; + case ABORT_DISK_IO: + mod_rq_state(req, m, 0, RQ_LOCAL_ABORTED); break; - case write_completed_with_error: - req->rq_state |= RQ_LOCAL_COMPLETED; - req->rq_state &= ~RQ_LOCAL_PENDING; - - __drbd_chk_io_error(mdev, DRBD_IO_ERROR); - _req_may_be_done_not_susp(req, m); + case WRITE_COMPLETED_WITH_ERROR: + drbd_report_io_error(mdev, req); + __drbd_chk_io_error(mdev, DRBD_WRITE_ERROR); + mod_rq_state(req, m, RQ_LOCAL_PENDING, RQ_LOCAL_COMPLETED); break; - case read_ahead_completed_with_error: - /* it is legal to fail READA */ - req->rq_state |= RQ_LOCAL_COMPLETED; - req->rq_state &= ~RQ_LOCAL_PENDING; - _req_may_be_done_not_susp(req, m); + case READ_COMPLETED_WITH_ERROR: + drbd_set_out_of_sync(mdev, req->i.sector, req->i.size); + drbd_report_io_error(mdev, req); + __drbd_chk_io_error(mdev, DRBD_READ_ERROR); + /* fall through. */ + case READ_AHEAD_COMPLETED_WITH_ERROR: + /* it is legal to fail READA, no __drbd_chk_io_error in that case. */ + mod_rq_state(req, m, RQ_LOCAL_PENDING, RQ_LOCAL_COMPLETED); break; - case read_completed_with_error: - drbd_set_out_of_sync(mdev, req->sector, req->size); - - req->rq_state |= RQ_LOCAL_COMPLETED; - req->rq_state &= ~RQ_LOCAL_PENDING; - - if (req->rq_state & RQ_LOCAL_ABORTED) { - _req_may_be_done(req, m); - break; - } - - __drbd_chk_io_error(mdev, DRBD_IO_ERROR); - - goto_queue_for_net_read: - - D_ASSERT(!(req->rq_state & RQ_NET_MASK)); - - /* no point in retrying if there is no good remote data, - * or we have no connection. */ - if (mdev->state.pdsk != D_UP_TO_DATE) { - _req_may_be_done_not_susp(req, m); - break; - } - - /* _req_mod(req,to_be_send); oops, recursion... */ - req->rq_state |= RQ_NET_PENDING; - inc_ap_pending(mdev); - /* fall through: _req_mod(req,queue_for_net_read); */ - - case queue_for_net_read: + case QUEUE_FOR_NET_READ: /* READ or READA, and * no local disk, * or target area marked as invalid, * or just got an io-error. */ - /* from drbd_make_request_common + /* from __drbd_make_request * or from bio_endio during read io-error recovery */ - /* so we can verify the handle in the answer packet - * corresponding hlist_del is in _req_may_be_done() */ - hlist_add_head(&req->collision, ar_hash_slot(mdev, req->sector)); + /* So we can verify the handle in the answer packet. + * Corresponding drbd_remove_request_interval is in + * drbd_req_complete() */ + D_ASSERT(drbd_interval_empty(&req->i)); + drbd_insert_interval(&mdev->read_requests, &req->i); set_bit(UNPLUG_REMOTE, &mdev->flags); D_ASSERT(req->rq_state & RQ_NET_PENDING); - req->rq_state |= RQ_NET_QUEUED; - req->w.cb = (req->rq_state & RQ_LOCAL_MASK) - ? w_read_retry_remote - : w_send_read_req; - drbd_queue_work(&mdev->data.work, &req->w); + D_ASSERT((req->rq_state & RQ_LOCAL_MASK) == 0); + mod_rq_state(req, m, 0, RQ_NET_QUEUED); + req->w.cb = w_send_read_req; + drbd_queue_work(&mdev->tconn->sender_work, &req->w); break; - case queue_for_net_write: + case QUEUE_FOR_NET_WRITE: /* assert something? */ - /* from drbd_make_request_common only */ + /* from __drbd_make_request only */ - hlist_add_head(&req->collision, tl_hash_slot(mdev, req->sector)); - /* corresponding hlist_del is in _req_may_be_done() */ + /* Corresponding drbd_remove_request_interval is in + * drbd_req_complete() */ + D_ASSERT(drbd_interval_empty(&req->i)); + drbd_insert_interval(&mdev->write_requests, &req->i); /* NOTE * In case the req ended up on the transfer log before being @@ -533,7 +563,7 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what, * * _req_add_to_epoch(req); this has to be after the * _maybe_start_new_epoch(req); which happened in - * drbd_make_request_common, because we now may set the bit + * __drbd_make_request, because we now may set the bit * again ourselves to close the current epoch. * * Add req to the (now) current epoch (barrier). */ @@ -543,202 +573,187 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what, * hurting performance. */ set_bit(UNPLUG_REMOTE, &mdev->flags); - /* see drbd_make_request_common, - * just after it grabs the req_lock */ - D_ASSERT(test_bit(CREATE_BARRIER, &mdev->flags) == 0); - - req->epoch = mdev->newest_tle->br_number; - - /* increment size of current epoch */ - mdev->newest_tle->n_writes++; - /* queue work item to send data */ D_ASSERT(req->rq_state & RQ_NET_PENDING); - req->rq_state |= RQ_NET_QUEUED; + mod_rq_state(req, m, 0, RQ_NET_QUEUED|RQ_EXP_BARR_ACK); req->w.cb = w_send_dblock; - drbd_queue_work(&mdev->data.work, &req->w); + drbd_queue_work(&mdev->tconn->sender_work, &req->w); /* close the epoch, in case it outgrew the limit */ - if (mdev->newest_tle->n_writes >= mdev->net_conf->max_epoch_size) - queue_barrier(mdev); + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + p = nc->max_epoch_size; + rcu_read_unlock(); + if (mdev->tconn->current_tle_writes >= p) + start_new_tl_epoch(mdev->tconn); break; - case queue_for_send_oos: - req->rq_state |= RQ_NET_QUEUED; - req->w.cb = w_send_oos; - drbd_queue_work(&mdev->data.work, &req->w); + case QUEUE_FOR_SEND_OOS: + mod_rq_state(req, m, 0, RQ_NET_QUEUED); + req->w.cb = w_send_out_of_sync; + drbd_queue_work(&mdev->tconn->sender_work, &req->w); break; - case read_retry_remote_canceled: - case send_canceled: - case send_failed: + case READ_RETRY_REMOTE_CANCELED: + case SEND_CANCELED: + case SEND_FAILED: /* real cleanup will be done from tl_clear. just update flags * so it is no longer marked as on the worker queue */ - req->rq_state &= ~RQ_NET_QUEUED; - /* if we did it right, tl_clear should be scheduled only after - * this, so this should not be necessary! */ - _req_may_be_done_not_susp(req, m); + mod_rq_state(req, m, RQ_NET_QUEUED, 0); break; - case handed_over_to_network: + case HANDED_OVER_TO_NETWORK: /* assert something? */ - if (bio_data_dir(req->master_bio) == WRITE) - atomic_add(req->size>>9, &mdev->ap_in_flight); - if (bio_data_dir(req->master_bio) == WRITE && - mdev->net_conf->wire_protocol == DRBD_PROT_A) { + !(req->rq_state & (RQ_EXP_RECEIVE_ACK | RQ_EXP_WRITE_ACK))) { /* this is what is dangerous about protocol A: * pretend it was successfully written on the peer. */ - if (req->rq_state & RQ_NET_PENDING) { - dec_ap_pending(mdev); - req->rq_state &= ~RQ_NET_PENDING; - req->rq_state |= RQ_NET_OK; - } /* else: neg-ack was faster... */ + if (req->rq_state & RQ_NET_PENDING) + mod_rq_state(req, m, RQ_NET_PENDING, RQ_NET_OK); + /* else: neg-ack was faster... */ /* it is still not yet RQ_NET_DONE until the * corresponding epoch barrier got acked as well, * so we know what to dirty on connection loss */ } - req->rq_state &= ~RQ_NET_QUEUED; - req->rq_state |= RQ_NET_SENT; - _req_may_be_done_not_susp(req, m); + mod_rq_state(req, m, RQ_NET_QUEUED, RQ_NET_SENT); break; - case oos_handed_to_network: + case OOS_HANDED_TO_NETWORK: /* Was not set PENDING, no longer QUEUED, so is now DONE * as far as this connection is concerned. */ - req->rq_state &= ~RQ_NET_QUEUED; - req->rq_state |= RQ_NET_DONE; - _req_may_be_done_not_susp(req, m); + mod_rq_state(req, m, RQ_NET_QUEUED, RQ_NET_DONE); break; - case connection_lost_while_pending: + case CONNECTION_LOST_WHILE_PENDING: /* transfer log cleanup after connection loss */ - /* assert something? */ - if (req->rq_state & RQ_NET_PENDING) - dec_ap_pending(mdev); - req->rq_state &= ~(RQ_NET_OK|RQ_NET_PENDING); - req->rq_state |= RQ_NET_DONE; - if (req->rq_state & RQ_NET_SENT && req->rq_state & RQ_WRITE) - atomic_sub(req->size>>9, &mdev->ap_in_flight); - - /* if it is still queued, we may not complete it here. - * it will be canceled soon. */ - if (!(req->rq_state & RQ_NET_QUEUED)) - _req_may_be_done(req, m); /* Allowed while state.susp */ + mod_rq_state(req, m, + RQ_NET_OK|RQ_NET_PENDING|RQ_COMPLETION_SUSP, + RQ_NET_DONE); break; - case conflict_discarded_by_peer: - /* for discarded conflicting writes of multiple primaries, + case CONFLICT_RESOLVED: + /* for superseded conflicting writes of multiple primaries, * there is no need to keep anything in the tl, potential - * node crashes are covered by the activity log. */ - if (what == conflict_discarded_by_peer) - dev_alert(DEV, "Got DiscardAck packet %llus +%u!" - " DRBD is not a random data generator!\n", - (unsigned long long)req->sector, req->size); - req->rq_state |= RQ_NET_DONE; - /* fall through */ - case write_acked_by_peer_and_sis: - case write_acked_by_peer: - if (what == write_acked_by_peer_and_sis) - req->rq_state |= RQ_NET_SIS; + * node crashes are covered by the activity log. + * + * If this request had been marked as RQ_POSTPONED before, + * it will actually not be completed, but "restarted", + * resubmitted from the retry worker context. */ + D_ASSERT(req->rq_state & RQ_NET_PENDING); + D_ASSERT(req->rq_state & RQ_EXP_WRITE_ACK); + mod_rq_state(req, m, RQ_NET_PENDING, RQ_NET_DONE|RQ_NET_OK); + break; + + case WRITE_ACKED_BY_PEER_AND_SIS: + req->rq_state |= RQ_NET_SIS; + case WRITE_ACKED_BY_PEER: + D_ASSERT(req->rq_state & RQ_EXP_WRITE_ACK); /* protocol C; successfully written on peer. * Nothing more to do here. * We want to keep the tl in place for all protocols, to cater * for volatile write-back caches on lower level devices. */ - case recv_acked_by_peer: + goto ack_common; + case RECV_ACKED_BY_PEER: + D_ASSERT(req->rq_state & RQ_EXP_RECEIVE_ACK); /* protocol B; pretends to be successfully written on peer. - * see also notes above in handed_over_to_network about + * see also notes above in HANDED_OVER_TO_NETWORK about * protocol != C */ - req->rq_state |= RQ_NET_OK; + ack_common: D_ASSERT(req->rq_state & RQ_NET_PENDING); - dec_ap_pending(mdev); - atomic_sub(req->size>>9, &mdev->ap_in_flight); - req->rq_state &= ~RQ_NET_PENDING; - _req_may_be_done_not_susp(req, m); + mod_rq_state(req, m, RQ_NET_PENDING, RQ_NET_OK); break; - case neg_acked: - /* assert something? */ - if (req->rq_state & RQ_NET_PENDING) { - dec_ap_pending(mdev); - atomic_sub(req->size>>9, &mdev->ap_in_flight); - } - req->rq_state &= ~(RQ_NET_OK|RQ_NET_PENDING); + case POSTPONE_WRITE: + D_ASSERT(req->rq_state & RQ_EXP_WRITE_ACK); + /* If this node has already detected the write conflict, the + * worker will be waiting on misc_wait. Wake it up once this + * request has completed locally. + */ + D_ASSERT(req->rq_state & RQ_NET_PENDING); + req->rq_state |= RQ_POSTPONED; + if (req->i.waiting) + wake_up(&mdev->misc_wait); + /* Do not clear RQ_NET_PENDING. This request will make further + * progress via restart_conflicting_writes() or + * fail_postponed_requests(). Hopefully. */ + break; - req->rq_state |= RQ_NET_DONE; - _req_may_be_done_not_susp(req, m); - /* else: done by handed_over_to_network */ + case NEG_ACKED: + mod_rq_state(req, m, RQ_NET_OK|RQ_NET_PENDING, 0); break; - case fail_frozen_disk_io: + case FAIL_FROZEN_DISK_IO: if (!(req->rq_state & RQ_LOCAL_COMPLETED)) break; - - _req_may_be_done(req, m); /* Allowed while state.susp */ + mod_rq_state(req, m, RQ_COMPLETION_SUSP, 0); break; - case restart_frozen_disk_io: + case RESTART_FROZEN_DISK_IO: if (!(req->rq_state & RQ_LOCAL_COMPLETED)) break; - req->rq_state &= ~RQ_LOCAL_COMPLETED; + mod_rq_state(req, m, + RQ_COMPLETION_SUSP|RQ_LOCAL_COMPLETED, + RQ_LOCAL_PENDING); rv = MR_READ; if (bio_data_dir(req->master_bio) == WRITE) rv = MR_WRITE; - get_ldev(mdev); + get_ldev(mdev); /* always succeeds in this call path */ req->w.cb = w_restart_disk_io; - drbd_queue_work(&mdev->data.work, &req->w); + drbd_queue_work(&mdev->tconn->sender_work, &req->w); break; - case resend: + case RESEND: /* Simply complete (local only) READs. */ if (!(req->rq_state & RQ_WRITE) && !req->w.cb) { - _req_may_be_done(req, m); + mod_rq_state(req, m, RQ_COMPLETION_SUSP, 0); break; } /* If RQ_NET_OK is already set, we got a P_WRITE_ACK or P_RECV_ACK - before the connection loss (B&C only); only P_BARRIER_ACK was missing. - Trowing them out of the TL here by pretending we got a BARRIER_ACK - We ensure that the peer was not rebooted */ + before the connection loss (B&C only); only P_BARRIER_ACK + (or the local completion?) was missing when we suspended. + Throwing them out of the TL here by pretending we got a BARRIER_ACK. + During connection handshake, we ensure that the peer was not rebooted. */ if (!(req->rq_state & RQ_NET_OK)) { + /* FIXME could this possibly be a req->w.cb == w_send_out_of_sync? + * in that case we must not set RQ_NET_PENDING. */ + + mod_rq_state(req, m, RQ_COMPLETION_SUSP, RQ_NET_QUEUED|RQ_NET_PENDING); if (req->w.cb) { - drbd_queue_work(&mdev->data.work, &req->w); + drbd_queue_work(&mdev->tconn->sender_work, &req->w); rv = req->rq_state & RQ_WRITE ? MR_WRITE : MR_READ; - } + } /* else: FIXME can this happen? */ break; } - /* else, fall through to barrier_acked */ + /* else, fall through to BARRIER_ACKED */ - case barrier_acked: + case BARRIER_ACKED: + /* barrier ack for READ requests does not make sense */ if (!(req->rq_state & RQ_WRITE)) break; if (req->rq_state & RQ_NET_PENDING) { - /* barrier came in before all requests have been acked. + /* barrier came in before all requests were acked. * this is bad, because if the connection is lost now, * we won't be able to clean them up... */ - dev_err(DEV, "FIXME (barrier_acked but pending)\n"); - list_move(&req->tl_requests, &mdev->out_of_sequence_requests); + dev_err(DEV, "FIXME (BARRIER_ACKED but pending)\n"); } - if ((req->rq_state & RQ_NET_MASK) != 0) { - req->rq_state |= RQ_NET_DONE; - if (mdev->net_conf->wire_protocol == DRBD_PROT_A) - atomic_sub(req->size>>9, &mdev->ap_in_flight); - } - _req_may_be_done(req, m); /* Allowed while state.susp */ + /* Allowed to complete requests, even while suspended. + * As this is called for all requests within a matching epoch, + * we need to filter, and only set RQ_NET_DONE for those that + * have actually been on the wire. */ + mod_rq_state(req, m, RQ_COMPLETION_SUSP, + (req->rq_state & RQ_NET_MASK) ? RQ_NET_DONE : 0); break; - case data_received: + case DATA_RECEIVED: D_ASSERT(req->rq_state & RQ_NET_PENDING); - dec_ap_pending(mdev); - req->rq_state &= ~RQ_NET_PENDING; - req->rq_state |= (RQ_NET_OK|RQ_NET_DONE); - _req_may_be_done_not_susp(req, m); + mod_rq_state(req, m, RQ_NET_PENDING, RQ_NET_OK|RQ_NET_DONE); break; }; @@ -752,75 +767,265 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what, * since size may be bigger than BM_BLOCK_SIZE, * we may need to check several bits. */ -static int drbd_may_do_local_read(struct drbd_conf *mdev, sector_t sector, int size) +static bool drbd_may_do_local_read(struct drbd_conf *mdev, sector_t sector, int size) { unsigned long sbnr, ebnr; sector_t esector, nr_sectors; if (mdev->state.disk == D_UP_TO_DATE) - return 1; - if (mdev->state.disk >= D_OUTDATED) - return 0; - if (mdev->state.disk < D_INCONSISTENT) - return 0; - /* state.disk == D_INCONSISTENT We will have a look at the BitMap */ - nr_sectors = drbd_get_capacity(mdev->this_bdev); + return true; + if (mdev->state.disk != D_INCONSISTENT) + return false; esector = sector + (size >> 9) - 1; - + nr_sectors = drbd_get_capacity(mdev->this_bdev); D_ASSERT(sector < nr_sectors); D_ASSERT(esector < nr_sectors); sbnr = BM_SECT_TO_BIT(sector); ebnr = BM_SECT_TO_BIT(esector); - return 0 == drbd_bm_count_bits(mdev, sbnr, ebnr); + return drbd_bm_count_bits(mdev, sbnr, ebnr) == 0; +} + +static bool remote_due_to_read_balancing(struct drbd_conf *mdev, sector_t sector, + enum drbd_read_balancing rbm) +{ + struct backing_dev_info *bdi; + int stripe_shift; + + switch (rbm) { + case RB_CONGESTED_REMOTE: + bdi = &mdev->ldev->backing_bdev->bd_disk->queue->backing_dev_info; + return bdi_read_congested(bdi); + case RB_LEAST_PENDING: + return atomic_read(&mdev->local_cnt) > + atomic_read(&mdev->ap_pending_cnt) + atomic_read(&mdev->rs_pending_cnt); + case RB_32K_STRIPING: /* stripe_shift = 15 */ + case RB_64K_STRIPING: + case RB_128K_STRIPING: + case RB_256K_STRIPING: + case RB_512K_STRIPING: + case RB_1M_STRIPING: /* stripe_shift = 20 */ + stripe_shift = (rbm - RB_32K_STRIPING + 15); + return (sector >> (stripe_shift - 9)) & 1; + case RB_ROUND_ROBIN: + return test_and_change_bit(READ_BALANCE_RR, &mdev->flags); + case RB_PREFER_REMOTE: + return true; + case RB_PREFER_LOCAL: + default: + return false; + } +} + +/* + * complete_conflicting_writes - wait for any conflicting write requests + * + * The write_requests tree contains all active write requests which we + * currently know about. Wait for any requests to complete which conflict with + * the new one. + * + * Only way out: remove the conflicting intervals from the tree. + */ +static void complete_conflicting_writes(struct drbd_request *req) +{ + DEFINE_WAIT(wait); + struct drbd_conf *mdev = req->w.mdev; + struct drbd_interval *i; + sector_t sector = req->i.sector; + int size = req->i.size; + + i = drbd_find_overlap(&mdev->write_requests, sector, size); + if (!i) + return; + + for (;;) { + prepare_to_wait(&mdev->misc_wait, &wait, TASK_UNINTERRUPTIBLE); + i = drbd_find_overlap(&mdev->write_requests, sector, size); + if (!i) + break; + /* Indicate to wake up device->misc_wait on progress. */ + i->waiting = true; + spin_unlock_irq(&mdev->tconn->req_lock); + schedule(); + spin_lock_irq(&mdev->tconn->req_lock); + } + finish_wait(&mdev->misc_wait, &wait); } +/* called within req_lock and rcu_read_lock() */ static void maybe_pull_ahead(struct drbd_conf *mdev) { - int congested = 0; + struct drbd_tconn *tconn = mdev->tconn; + struct net_conf *nc; + bool congested = false; + enum drbd_on_congestion on_congestion; + + nc = rcu_dereference(tconn->net_conf); + on_congestion = nc ? nc->on_congestion : OC_BLOCK; + if (on_congestion == OC_BLOCK || + tconn->agreed_pro_version < 96) + return; /* If I don't even have good local storage, we can not reasonably try * to pull ahead of the peer. We also need the local reference to make * sure mdev->act_log is there. - * Note: caller has to make sure that net_conf is there. */ if (!get_ldev_if_state(mdev, D_UP_TO_DATE)) return; - if (mdev->net_conf->cong_fill && - atomic_read(&mdev->ap_in_flight) >= mdev->net_conf->cong_fill) { + if (nc->cong_fill && + atomic_read(&mdev->ap_in_flight) >= nc->cong_fill) { dev_info(DEV, "Congestion-fill threshold reached\n"); - congested = 1; + congested = true; } - if (mdev->act_log->used >= mdev->net_conf->cong_extents) { + if (mdev->act_log->used >= nc->cong_extents) { dev_info(DEV, "Congestion-extents threshold reached\n"); - congested = 1; + congested = true; } if (congested) { - queue_barrier(mdev); /* last barrier, after mirrored writes */ + /* start a new epoch for non-mirrored writes */ + start_new_tl_epoch(mdev->tconn); - if (mdev->net_conf->on_congestion == OC_PULL_AHEAD) + if (on_congestion == OC_PULL_AHEAD) _drbd_set_state(_NS(mdev, conn, C_AHEAD), 0, NULL); - else /*mdev->net_conf->on_congestion == OC_DISCONNECT */ + else /*nc->on_congestion == OC_DISCONNECT */ _drbd_set_state(_NS(mdev, conn, C_DISCONNECTING), 0, NULL); } put_ldev(mdev); } -static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio, unsigned long start_time) +/* If this returns false, and req->private_bio is still set, + * this should be submitted locally. + * + * If it returns false, but req->private_bio is not set, + * we do not have access to good data :( + * + * Otherwise, this destroys req->private_bio, if any, + * and returns true. + */ +static bool do_remote_read(struct drbd_request *req) +{ + struct drbd_conf *mdev = req->w.mdev; + enum drbd_read_balancing rbm; + + if (req->private_bio) { + if (!drbd_may_do_local_read(mdev, + req->i.sector, req->i.size)) { + bio_put(req->private_bio); + req->private_bio = NULL; + put_ldev(mdev); + } + } + + if (mdev->state.pdsk != D_UP_TO_DATE) + return false; + + if (req->private_bio == NULL) + return true; + + /* TODO: improve read balancing decisions, take into account drbd + * protocol, pending requests etc. */ + + rcu_read_lock(); + rbm = rcu_dereference(mdev->ldev->disk_conf)->read_balancing; + rcu_read_unlock(); + + if (rbm == RB_PREFER_LOCAL && req->private_bio) + return false; /* submit locally */ + + if (remote_due_to_read_balancing(mdev, req->i.sector, rbm)) { + if (req->private_bio) { + bio_put(req->private_bio); + req->private_bio = NULL; + put_ldev(mdev); + } + return true; + } + + return false; +} + +/* returns number of connections (== 1, for drbd 8.4) + * expected to actually write this data, + * which does NOT include those that we are L_AHEAD for. */ +static int drbd_process_write_request(struct drbd_request *req) +{ + struct drbd_conf *mdev = req->w.mdev; + int remote, send_oos; + + rcu_read_lock(); + remote = drbd_should_do_remote(mdev->state); + if (remote) { + maybe_pull_ahead(mdev); + remote = drbd_should_do_remote(mdev->state); + } + send_oos = drbd_should_send_out_of_sync(mdev->state); + rcu_read_unlock(); + + /* Need to replicate writes. Unless it is an empty flush, + * which is better mapped to a DRBD P_BARRIER packet, + * also for drbd wire protocol compatibility reasons. + * If this was a flush, just start a new epoch. + * Unless the current epoch was empty anyways, or we are not currently + * replicating, in which case there is no point. */ + if (unlikely(req->i.size == 0)) { + /* The only size==0 bios we expect are empty flushes. */ + D_ASSERT(req->master_bio->bi_rw & REQ_FLUSH); + if (remote) + start_new_tl_epoch(mdev->tconn); + return 0; + } + + if (!remote && !send_oos) + return 0; + + D_ASSERT(!(remote && send_oos)); + + if (remote) { + _req_mod(req, TO_BE_SENT); + _req_mod(req, QUEUE_FOR_NET_WRITE); + } else if (drbd_set_out_of_sync(mdev, req->i.sector, req->i.size)) + _req_mod(req, QUEUE_FOR_SEND_OOS); + + return remote; +} + +static void +drbd_submit_req_private_bio(struct drbd_request *req) +{ + struct drbd_conf *mdev = req->w.mdev; + struct bio *bio = req->private_bio; + const int rw = bio_rw(bio); + + bio->bi_bdev = mdev->ldev->backing_bdev; + + /* State may have changed since we grabbed our reference on the + * ->ldev member. Double check, and short-circuit to endio. + * In case the last activity log transaction failed to get on + * stable storage, and this is a WRITE, we may not even submit + * this bio. */ + if (get_ldev(mdev)) { + if (drbd_insert_fault(mdev, + rw == WRITE ? DRBD_FAULT_DT_WR + : rw == READ ? DRBD_FAULT_DT_RD + : DRBD_FAULT_DT_RA)) + bio_endio(bio, -EIO); + else + generic_make_request(bio); + put_ldev(mdev); + } else + bio_endio(bio, -EIO); +} + +void __drbd_make_request(struct drbd_conf *mdev, struct bio *bio, unsigned long start_time) { const int rw = bio_rw(bio); - const int size = bio->bi_size; - const sector_t sector = bio->bi_sector; - struct drbd_tl_epoch *b = NULL; + struct bio_and_error m = { NULL, }; struct drbd_request *req; - int local, remote, send_oos = 0; - int err = -EIO; - int ret = 0; - union drbd_state s; + bool no_remote = false; /* allocate outside of all locks; */ req = drbd_req_new(mdev, bio); @@ -830,55 +1035,14 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio, uns * if user cannot handle io errors, that's not our business. */ dev_err(DEV, "could not kmalloc() req\n"); bio_endio(bio, -ENOMEM); - return 0; + return; } req->start_time = start_time; - local = get_ldev(mdev); - if (!local) { - bio_put(req->private_bio); /* or we get a bio leak */ + if (!get_ldev(mdev)) { + bio_put(req->private_bio); req->private_bio = NULL; } - if (rw == WRITE) { - /* Need to replicate writes. Unless it is an empty flush, - * which is better mapped to a DRBD P_BARRIER packet, - * also for drbd wire protocol compatibility reasons. */ - if (unlikely(size == 0)) { - /* The only size==0 bios we expect are empty flushes. */ - D_ASSERT(bio->bi_rw & REQ_FLUSH); - remote = 0; - } else - remote = 1; - } else { - /* READ || READA */ - if (local) { - if (!drbd_may_do_local_read(mdev, sector, size)) { - /* we could kick the syncer to - * sync this extent asap, wait for - * it, then continue locally. - * Or just issue the request remotely. - */ - local = 0; - bio_put(req->private_bio); - req->private_bio = NULL; - put_ldev(mdev); - } - } - remote = !local && mdev->state.pdsk >= D_UP_TO_DATE; - } - - /* If we have a disk, but a READA request is mapped to remote, - * we are R_PRIMARY, D_INCONSISTENT, SyncTarget. - * Just fail that READA request right here. - * - * THINK: maybe fail all READA when not local? - * or make this configurable... - * if network is slow, READA won't do any good. - */ - if (rw == READA && mdev->state.disk >= D_INCONSISTENT && !local) { - err = -EWOULDBLOCK; - goto fail_and_free_req; - } /* For WRITES going to the local disk, grab a reference on the target * extent. This waits for any resync activity in the corresponding @@ -887,348 +1051,131 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio, uns * of transactional on-disk meta data updates. * Empty flushes don't need to go into the activity log, they can only * flush data for pending writes which are already in there. */ - if (rw == WRITE && local && size + if (rw == WRITE && req->private_bio && req->i.size && !test_bit(AL_SUSPENDED, &mdev->flags)) { req->rq_state |= RQ_IN_ACT_LOG; - drbd_al_begin_io(mdev, sector); - } - - s = mdev->state; - remote = remote && drbd_should_do_remote(s); - send_oos = rw == WRITE && drbd_should_send_oos(s); - D_ASSERT(!(remote && send_oos)); - - if (!(local || remote) && !is_susp(mdev->state)) { - if (__ratelimit(&drbd_ratelimit_state)) - dev_err(DEV, "IO ERROR: neither local nor remote disk\n"); - goto fail_free_complete; + drbd_al_begin_io(mdev, &req->i); } - /* For WRITE request, we have to make sure that we have an - * unused_spare_tle, in case we need to start a new epoch. - * I try to be smart and avoid to pre-allocate always "just in case", - * but there is a race between testing the bit and pointer outside the - * spinlock, and grabbing the spinlock. - * if we lost that race, we retry. */ - if (rw == WRITE && (remote || send_oos) && - mdev->unused_spare_tle == NULL && - test_bit(CREATE_BARRIER, &mdev->flags)) { -allocate_barrier: - b = kmalloc(sizeof(struct drbd_tl_epoch), GFP_NOIO); - if (!b) { - dev_err(DEV, "Failed to alloc barrier.\n"); - err = -ENOMEM; - goto fail_free_complete; - } + spin_lock_irq(&mdev->tconn->req_lock); + if (rw == WRITE) { + /* This may temporarily give up the req_lock, + * but will re-aquire it before it returns here. + * Needs to be before the check on drbd_suspended() */ + complete_conflicting_writes(req); } - /* GOOD, everything prepared, grab the spin_lock */ - spin_lock_irq(&mdev->req_lock); - - if (is_susp(mdev->state)) { - /* If we got suspended, use the retry mechanism of - drbd_make_request() to restart processing of this - bio. In the next call to drbd_make_request - we sleep in inc_ap_bio() */ - ret = 1; - spin_unlock_irq(&mdev->req_lock); - goto fail_free_complete; - } + /* no more giving up req_lock from now on! */ - if (remote || send_oos) { - remote = drbd_should_do_remote(mdev->state); - send_oos = rw == WRITE && drbd_should_send_oos(mdev->state); - D_ASSERT(!(remote && send_oos)); - - if (!(remote || send_oos)) - dev_warn(DEV, "lost connection while grabbing the req_lock!\n"); - if (!(local || remote)) { - dev_err(DEV, "IO ERROR: neither local nor remote disk\n"); - spin_unlock_irq(&mdev->req_lock); - goto fail_free_complete; + if (drbd_suspended(mdev)) { + /* push back and retry: */ + req->rq_state |= RQ_POSTPONED; + if (req->private_bio) { + bio_put(req->private_bio); + req->private_bio = NULL; + put_ldev(mdev); } + goto out; } - if (b && mdev->unused_spare_tle == NULL) { - mdev->unused_spare_tle = b; - b = NULL; - } - if (rw == WRITE && (remote || send_oos) && - mdev->unused_spare_tle == NULL && - test_bit(CREATE_BARRIER, &mdev->flags)) { - /* someone closed the current epoch - * while we were grabbing the spinlock */ - spin_unlock_irq(&mdev->req_lock); - goto allocate_barrier; - } - - /* Update disk stats */ _drbd_start_io_acct(mdev, req, bio); - /* _maybe_start_new_epoch(mdev); - * If we need to generate a write barrier packet, we have to add the - * new epoch (barrier) object, and queue the barrier packet for sending, - * and queue the req's data after it _within the same lock_, otherwise - * we have race conditions were the reorder domains could be mixed up. - * - * Even read requests may start a new epoch and queue the corresponding - * barrier packet. To get the write ordering right, we only have to - * make sure that, if this is a write request and it triggered a - * barrier packet, this request is queued within the same spinlock. */ - if ((remote || send_oos) && mdev->unused_spare_tle && - test_and_clear_bit(CREATE_BARRIER, &mdev->flags)) { - _tl_add_barrier(mdev, mdev->unused_spare_tle); - mdev->unused_spare_tle = NULL; - } else { - D_ASSERT(!(remote && rw == WRITE && - test_bit(CREATE_BARRIER, &mdev->flags))); + /* We fail READ/READA early, if we can not serve it. + * We must do this before req is registered on any lists. + * Otherwise, drbd_req_complete() will queue failed READ for retry. */ + if (rw != WRITE) { + if (!do_remote_read(req) && !req->private_bio) + goto nodata; } - /* NOTE - * Actually, 'local' may be wrong here already, since we may have failed - * to write to the meta data, and may become wrong anytime because of - * local io-error for some other request, which would lead to us - * "detaching" the local disk. - * - * 'remote' may become wrong any time because the network could fail. - * - * This is a harmless race condition, though, since it is handled - * correctly at the appropriate places; so it just defers the failure - * of the respective operation. - */ - - /* mark them early for readability. - * this just sets some state flags. */ - if (remote) - _req_mod(req, to_be_send); - if (local) - _req_mod(req, to_be_submitted); - - /* check this request on the collision detection hash tables. - * if we have a conflict, just complete it here. - * THINK do we want to check reads, too? (I don't think so...) */ - if (rw == WRITE && _req_conflicts(req)) - goto fail_conflicting; + /* which transfer log epoch does this belong to? */ + req->epoch = atomic_read(&mdev->tconn->current_tle_nr); /* no point in adding empty flushes to the transfer log, * they are mapped to drbd barriers already. */ - if (likely(size!=0)) - list_add_tail(&req->tl_requests, &mdev->newest_tle->requests); + if (likely(req->i.size!=0)) { + if (rw == WRITE) + mdev->tconn->current_tle_writes++; - /* NOTE remote first: to get the concurrent write detection right, - * we must register the request before start of local IO. */ - if (remote) { - /* either WRITE and C_CONNECTED, - * or READ, and no local disk, - * or READ, but not in sync. - */ - _req_mod(req, (rw == WRITE) - ? queue_for_net_write - : queue_for_net_read); + list_add_tail(&req->tl_requests, &mdev->tconn->transfer_log); } - if (send_oos && drbd_set_out_of_sync(mdev, sector, size)) - _req_mod(req, queue_for_send_oos); - if (remote && - mdev->net_conf->on_congestion != OC_BLOCK && mdev->agreed_pro_version >= 96) - maybe_pull_ahead(mdev); - - /* If this was a flush, queue a drbd barrier/start a new epoch. - * Unless the current epoch was empty anyways, or we are not currently - * replicating, in which case there is no point. */ - if (unlikely(bio->bi_rw & REQ_FLUSH) - && mdev->newest_tle->n_writes - && drbd_should_do_remote(mdev->state)) - queue_barrier(mdev); - - spin_unlock_irq(&mdev->req_lock); - kfree(b); /* if someone else has beaten us to it... */ - - if (local) { - req->private_bio->bi_bdev = mdev->ldev->backing_bdev; - - /* State may have changed since we grabbed our reference on the - * mdev->ldev member. Double check, and short-circuit to endio. - * In case the last activity log transaction failed to get on - * stable storage, and this is a WRITE, we may not even submit - * this bio. */ - if (get_ldev(mdev)) { - if (drbd_insert_fault(mdev, rw == WRITE ? DRBD_FAULT_DT_WR - : rw == READ ? DRBD_FAULT_DT_RD - : DRBD_FAULT_DT_RA)) - bio_endio(req->private_bio, -EIO); - else - generic_make_request(req->private_bio); - put_ldev(mdev); + if (rw == WRITE) { + if (!drbd_process_write_request(req)) + no_remote = true; + } else { + /* We either have a private_bio, or we can read from remote. + * Otherwise we had done the goto nodata above. */ + if (req->private_bio == NULL) { + _req_mod(req, TO_BE_SENT); + _req_mod(req, QUEUE_FOR_NET_READ); } else - bio_endio(req->private_bio, -EIO); + no_remote = true; } - return 0; - -fail_conflicting: - /* this is a conflicting request. - * even though it may have been only _partially_ - * overlapping with one of the currently pending requests, - * without even submitting or sending it, we will - * pretend that it was successfully served right now. - */ - _drbd_end_io_acct(mdev, req); - spin_unlock_irq(&mdev->req_lock); - if (remote) - dec_ap_pending(mdev); - /* THINK: do we want to fail it (-EIO), or pretend success? - * this pretends success. */ - err = 0; - -fail_free_complete: - if (req->rq_state & RQ_IN_ACT_LOG) - drbd_al_complete_io(mdev, sector); -fail_and_free_req: - if (local) { - bio_put(req->private_bio); - req->private_bio = NULL; - put_ldev(mdev); + if (req->private_bio) { + /* needs to be marked within the same spinlock */ + _req_mod(req, TO_BE_SUBMITTED); + /* but we need to give up the spinlock to submit */ + spin_unlock_irq(&mdev->tconn->req_lock); + drbd_submit_req_private_bio(req); + spin_lock_irq(&mdev->tconn->req_lock); + } else if (no_remote) { +nodata: + if (__ratelimit(&drbd_ratelimit_state)) + dev_err(DEV, "IO ERROR: neither local nor remote data, sector %llu+%u\n", + (unsigned long long)req->i.sector, req->i.size >> 9); + /* A write may have been queued for send_oos, however. + * So we can not simply free it, we must go through drbd_req_put_completion_ref() */ } - if (!ret) - bio_endio(bio, err); - - drbd_req_free(req); - dec_ap_bio(mdev); - kfree(b); - - return ret; -} -/* helper function for drbd_make_request - * if we can determine just by the mdev (state) that this request will fail, - * return 1 - * otherwise return 0 - */ -static int drbd_fail_request_early(struct drbd_conf *mdev, int is_write) -{ - if (mdev->state.role != R_PRIMARY && - (!allow_oos || is_write)) { - if (__ratelimit(&drbd_ratelimit_state)) { - dev_err(DEV, "Process %s[%u] tried to %s; " - "since we are not in Primary state, " - "we cannot allow this\n", - current->comm, current->pid, - is_write ? "WRITE" : "READ"); - } - return 1; - } +out: + if (drbd_req_put_completion_ref(req, &m, 1)) + kref_put(&req->kref, drbd_req_destroy); + spin_unlock_irq(&mdev->tconn->req_lock); - return 0; + if (m.bio) + complete_master_bio(mdev, &m); + return; } void drbd_make_request(struct request_queue *q, struct bio *bio) { - unsigned int s_enr, e_enr; struct drbd_conf *mdev = (struct drbd_conf *) q->queuedata; unsigned long start_time; - if (drbd_fail_request_early(mdev, bio_data_dir(bio) & WRITE)) { - bio_endio(bio, -EPERM); - return; - } - start_time = jiffies; /* * what we "blindly" assume: */ - D_ASSERT((bio->bi_size & 0x1ff) == 0); - - /* to make some things easier, force alignment of requests within the - * granularity of our hash tables */ - s_enr = bio->bi_sector >> HT_SHIFT; - e_enr = bio->bi_size ? (bio->bi_sector+(bio->bi_size>>9)-1) >> HT_SHIFT : s_enr; - - if (likely(s_enr == e_enr)) { - do { - inc_ap_bio(mdev, 1); - } while (drbd_make_request_common(mdev, bio, start_time)); - return; - } - - /* can this bio be split generically? - * Maybe add our own split-arbitrary-bios function. */ - if (bio->bi_vcnt != 1 || bio->bi_idx != 0 || bio->bi_size > DRBD_MAX_BIO_SIZE) { - /* rather error out here than BUG in bio_split */ - dev_err(DEV, "bio would need to, but cannot, be split: " - "(vcnt=%u,idx=%u,size=%u,sector=%llu)\n", - bio->bi_vcnt, bio->bi_idx, bio->bi_size, - (unsigned long long)bio->bi_sector); - bio_endio(bio, -EINVAL); - } else { - /* This bio crosses some boundary, so we have to split it. */ - struct bio_pair *bp; - /* works for the "do not cross hash slot boundaries" case - * e.g. sector 262269, size 4096 - * s_enr = 262269 >> 6 = 4097 - * e_enr = (262269+8-1) >> 6 = 4098 - * HT_SHIFT = 6 - * sps = 64, mask = 63 - * first_sectors = 64 - (262269 & 63) = 3 - */ - const sector_t sect = bio->bi_sector; - const int sps = 1 << HT_SHIFT; /* sectors per slot */ - const int mask = sps - 1; - const sector_t first_sectors = sps - (sect & mask); - bp = bio_split(bio, first_sectors); + D_ASSERT(IS_ALIGNED(bio->bi_size, 512)); - /* we need to get a "reference count" (ap_bio_cnt) - * to avoid races with the disconnect/reconnect/suspend code. - * In case we need to split the bio here, we need to get three references - * atomically, otherwise we might deadlock when trying to submit the - * second one! */ - inc_ap_bio(mdev, 3); - - D_ASSERT(e_enr == s_enr + 1); - - while (drbd_make_request_common(mdev, &bp->bio1, start_time)) - inc_ap_bio(mdev, 1); - - while (drbd_make_request_common(mdev, &bp->bio2, start_time)) - inc_ap_bio(mdev, 1); - - dec_ap_bio(mdev); - - bio_pair_release(bp); - } + inc_ap_bio(mdev); + __drbd_make_request(mdev, bio, start_time); } -/* This is called by bio_add_page(). With this function we reduce - * the number of BIOs that span over multiple DRBD_MAX_BIO_SIZEs - * units (was AL_EXTENTs). +/* This is called by bio_add_page(). + * + * q->max_hw_sectors and other global limits are already enforced there. * - * we do the calculation within the lower 32bit of the byte offsets, - * since we don't care for actual offset, but only check whether it - * would cross "activity log extent" boundaries. + * We need to call down to our lower level device, + * in case it has special restrictions. + * + * We also may need to enforce configured max-bio-bvecs limits. * * As long as the BIO is empty we have to allow at least one bvec, - * regardless of size and offset. so the resulting bio may still - * cross extent boundaries. those are dealt with (bio_split) in - * drbd_make_request. + * regardless of size and offset, so no need to ask lower levels. */ int drbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bvm, struct bio_vec *bvec) { struct drbd_conf *mdev = (struct drbd_conf *) q->queuedata; - unsigned int bio_offset = - (unsigned int)bvm->bi_sector << 9; /* 32 bit */ unsigned int bio_size = bvm->bi_size; - int limit, backing_limit; - - limit = DRBD_MAX_BIO_SIZE - - ((bio_offset & (DRBD_MAX_BIO_SIZE-1)) + bio_size); - if (limit < 0) - limit = 0; - if (bio_size == 0) { - if (limit <= bvec->bv_len) - limit = bvec->bv_len; - } else if (limit && get_ldev(mdev)) { + int limit = DRBD_MAX_BIO_SIZE; + int backing_limit; + + if (bio_size && get_ldev(mdev)) { struct request_queue * const b = mdev->ldev->backing_bdev->bd_disk->queue; if (b->merge_bvec_fn) { @@ -1240,24 +1187,38 @@ int drbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bvm, struct return limit; } +struct drbd_request *find_oldest_request(struct drbd_tconn *tconn) +{ + /* Walk the transfer log, + * and find the oldest not yet completed request */ + struct drbd_request *r; + list_for_each_entry(r, &tconn->transfer_log, tl_requests) { + if (atomic_read(&r->completion_ref)) + return r; + } + return NULL; +} + void request_timer_fn(unsigned long data) { struct drbd_conf *mdev = (struct drbd_conf *) data; + struct drbd_tconn *tconn = mdev->tconn; struct drbd_request *req; /* oldest request */ - struct list_head *le; + struct net_conf *nc; unsigned long ent = 0, dt = 0, et, nt; /* effective timeout = ko_count * timeout */ unsigned long now; - if (get_net_conf(mdev)) { - if (mdev->state.conn >= C_WF_REPORT_PARAMS) - ent = mdev->net_conf->timeout*HZ/10 - * mdev->net_conf->ko_count; - put_net_conf(mdev); - } + rcu_read_lock(); + nc = rcu_dereference(tconn->net_conf); + if (nc && mdev->state.conn >= C_WF_REPORT_PARAMS) + ent = nc->timeout * HZ/10 * nc->ko_count; + if (get_ldev(mdev)) { /* implicit state.disk >= D_INCONSISTENT */ - dt = mdev->ldev->dc.disk_timeout * HZ / 10; + dt = rcu_dereference(mdev->ldev->disk_conf)->disk_timeout * HZ / 10; put_ldev(mdev); } + rcu_read_unlock(); + et = min_not_zero(dt, ent); if (!et) @@ -1265,17 +1226,14 @@ void request_timer_fn(unsigned long data) now = jiffies; - spin_lock_irq(&mdev->req_lock); - le = &mdev->oldest_tle->requests; - if (list_empty(le)) { - spin_unlock_irq(&mdev->req_lock); + spin_lock_irq(&tconn->req_lock); + req = find_oldest_request(tconn); + if (!req) { + spin_unlock_irq(&tconn->req_lock); mod_timer(&mdev->request_timer, now + et); return; } - le = le->prev; - req = list_entry(le, struct drbd_request, tl_requests); - /* The request is considered timed out, if * - we have some effective timeout from the configuration, * with above state restrictions applied, @@ -1294,17 +1252,17 @@ void request_timer_fn(unsigned long data) */ if (ent && req->rq_state & RQ_NET_PENDING && time_after(now, req->start_time + ent) && - !time_in_range(now, mdev->last_reconnect_jif, mdev->last_reconnect_jif + ent)) { + !time_in_range(now, tconn->last_reconnect_jif, tconn->last_reconnect_jif + ent)) { dev_warn(DEV, "Remote failed to finish a request within ko-count * timeout\n"); _drbd_set_state(_NS(mdev, conn, C_TIMEOUT), CS_VERBOSE | CS_HARD, NULL); } - if (dt && req->rq_state & RQ_LOCAL_PENDING && + if (dt && req->rq_state & RQ_LOCAL_PENDING && req->w.mdev == mdev && time_after(now, req->start_time + dt) && !time_in_range(now, mdev->last_reattach_jif, mdev->last_reattach_jif + dt)) { dev_warn(DEV, "Local backing device failed to meet the disk-timeout\n"); __drbd_chk_io_error(mdev, DRBD_FORCE_DETACH); } nt = (time_after(now, req->start_time + et) ? now : req->start_time) + et; - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&tconn->req_lock); mod_timer(&mdev->request_timer, nt); } diff --git a/drivers/block/drbd/drbd_req.h b/drivers/block/drbd/drbd_req.h index 3d2111919486..016de6b8bb57 100644 --- a/drivers/block/drbd/drbd_req.h +++ b/drivers/block/drbd/drbd_req.h @@ -77,40 +77,41 @@ */ enum drbd_req_event { - created, - to_be_send, - to_be_submitted, + CREATED, + TO_BE_SENT, + TO_BE_SUBMITTED, /* XXX yes, now I am inconsistent... * these are not "events" but "actions" * oh, well... */ - queue_for_net_write, - queue_for_net_read, - queue_for_send_oos, - - send_canceled, - send_failed, - handed_over_to_network, - oos_handed_to_network, - connection_lost_while_pending, - read_retry_remote_canceled, - recv_acked_by_peer, - write_acked_by_peer, - write_acked_by_peer_and_sis, /* and set_in_sync */ - conflict_discarded_by_peer, - neg_acked, - barrier_acked, /* in protocol A and B */ - data_received, /* (remote read) */ - - read_completed_with_error, - read_ahead_completed_with_error, - write_completed_with_error, - abort_disk_io, - completed_ok, - resend, - fail_frozen_disk_io, - restart_frozen_disk_io, - nothing, /* for tracing only */ + QUEUE_FOR_NET_WRITE, + QUEUE_FOR_NET_READ, + QUEUE_FOR_SEND_OOS, + + SEND_CANCELED, + SEND_FAILED, + HANDED_OVER_TO_NETWORK, + OOS_HANDED_TO_NETWORK, + CONNECTION_LOST_WHILE_PENDING, + READ_RETRY_REMOTE_CANCELED, + RECV_ACKED_BY_PEER, + WRITE_ACKED_BY_PEER, + WRITE_ACKED_BY_PEER_AND_SIS, /* and set_in_sync */ + CONFLICT_RESOLVED, + POSTPONE_WRITE, + NEG_ACKED, + BARRIER_ACKED, /* in protocol A and B */ + DATA_RECEIVED, /* (remote read) */ + + READ_COMPLETED_WITH_ERROR, + READ_AHEAD_COMPLETED_WITH_ERROR, + WRITE_COMPLETED_WITH_ERROR, + ABORT_DISK_IO, + COMPLETED_OK, + RESEND, + FAIL_FROZEN_DISK_IO, + RESTART_FROZEN_DISK_IO, + NOTHING, }; /* encoding of request states for now. we don't actually need that many bits. @@ -142,8 +143,8 @@ enum drbd_req_state_bits { * recv_ack (B) or implicit "ack" (A), * still waiting for the barrier ack. * master_bio may already be completed and invalidated. - * 11100: write_acked (C), - * data_received (for remote read, any protocol) + * 11100: write acked (C), + * data received (for remote read, any protocol) * or finally the barrier ack has arrived (B,A)... * request can be freed * 01100: neg-acked (write, protocol C) @@ -198,6 +199,22 @@ enum drbd_req_state_bits { /* Should call drbd_al_complete_io() for this request... */ __RQ_IN_ACT_LOG, + + /* The peer has sent a retry ACK */ + __RQ_POSTPONED, + + /* would have been completed, + * but was not, because of drbd_suspended() */ + __RQ_COMPLETION_SUSP, + + /* We expect a receive ACK (wire proto B) */ + __RQ_EXP_RECEIVE_ACK, + + /* We expect a write ACK (wite proto C) */ + __RQ_EXP_WRITE_ACK, + + /* waiting for a barrier ack, did an extra kref_get */ + __RQ_EXP_BARR_ACK, }; #define RQ_LOCAL_PENDING (1UL << __RQ_LOCAL_PENDING) @@ -219,56 +236,16 @@ enum drbd_req_state_bits { #define RQ_WRITE (1UL << __RQ_WRITE) #define RQ_IN_ACT_LOG (1UL << __RQ_IN_ACT_LOG) +#define RQ_POSTPONED (1UL << __RQ_POSTPONED) +#define RQ_COMPLETION_SUSP (1UL << __RQ_COMPLETION_SUSP) +#define RQ_EXP_RECEIVE_ACK (1UL << __RQ_EXP_RECEIVE_ACK) +#define RQ_EXP_WRITE_ACK (1UL << __RQ_EXP_WRITE_ACK) +#define RQ_EXP_BARR_ACK (1UL << __RQ_EXP_BARR_ACK) /* For waking up the frozen transfer log mod_req() has to return if the request should be counted in the epoch object*/ -#define MR_WRITE_SHIFT 0 -#define MR_WRITE (1 << MR_WRITE_SHIFT) -#define MR_READ_SHIFT 1 -#define MR_READ (1 << MR_READ_SHIFT) - -/* epoch entries */ -static inline -struct hlist_head *ee_hash_slot(struct drbd_conf *mdev, sector_t sector) -{ - BUG_ON(mdev->ee_hash_s == 0); - return mdev->ee_hash + - ((unsigned int)(sector>>HT_SHIFT) % mdev->ee_hash_s); -} - -/* transfer log (drbd_request objects) */ -static inline -struct hlist_head *tl_hash_slot(struct drbd_conf *mdev, sector_t sector) -{ - BUG_ON(mdev->tl_hash_s == 0); - return mdev->tl_hash + - ((unsigned int)(sector>>HT_SHIFT) % mdev->tl_hash_s); -} - -/* application reads (drbd_request objects) */ -static struct hlist_head *ar_hash_slot(struct drbd_conf *mdev, sector_t sector) -{ - return mdev->app_reads_hash - + ((unsigned int)(sector) % APP_R_HSIZE); -} - -/* when we receive the answer for a read request, - * verify that we actually know about it */ -static inline struct drbd_request *_ar_id_to_req(struct drbd_conf *mdev, - u64 id, sector_t sector) -{ - struct hlist_head *slot = ar_hash_slot(mdev, sector); - struct hlist_node *n; - struct drbd_request *req; - - hlist_for_each_entry(req, n, slot, collision) { - if ((unsigned long)req == (unsigned long)id) { - D_ASSERT(req->sector == sector); - return req; - } - } - return NULL; -} +#define MR_WRITE 1 +#define MR_READ 2 static inline void drbd_req_make_private_bio(struct drbd_request *req, struct bio *bio_src) { @@ -278,41 +255,10 @@ static inline void drbd_req_make_private_bio(struct drbd_request *req, struct bi req->private_bio = bio; bio->bi_private = req; - bio->bi_end_io = drbd_endio_pri; + bio->bi_end_io = drbd_request_endio; bio->bi_next = NULL; } -static inline struct drbd_request *drbd_req_new(struct drbd_conf *mdev, - struct bio *bio_src) -{ - struct drbd_request *req = - mempool_alloc(drbd_request_mempool, GFP_NOIO); - if (likely(req)) { - drbd_req_make_private_bio(req, bio_src); - - req->rq_state = bio_data_dir(bio_src) == WRITE ? RQ_WRITE : 0; - req->mdev = mdev; - req->master_bio = bio_src; - req->epoch = 0; - req->sector = bio_src->bi_sector; - req->size = bio_src->bi_size; - INIT_HLIST_NODE(&req->collision); - INIT_LIST_HEAD(&req->tl_requests); - INIT_LIST_HEAD(&req->w.list); - } - return req; -} - -static inline void drbd_req_free(struct drbd_request *req) -{ - mempool_free(req, drbd_request_mempool); -} - -static inline int overlaps(sector_t s1, int l1, sector_t s2, int l2) -{ - return !((s1 + (l1>>9) <= s2) || (s1 >= s2 + (l2>>9))); -} - /* Short lived temporary struct on the stack. * We could squirrel the error to be returned into * bio->bi_size, or similar. But that would be too ugly. */ @@ -321,6 +267,7 @@ struct bio_and_error { int error; }; +extern void drbd_req_destroy(struct kref *kref); extern void _req_may_be_done(struct drbd_request *req, struct bio_and_error *m); extern int __req_mod(struct drbd_request *req, enum drbd_req_event what, @@ -328,13 +275,17 @@ extern int __req_mod(struct drbd_request *req, enum drbd_req_event what, extern void complete_master_bio(struct drbd_conf *mdev, struct bio_and_error *m); extern void request_timer_fn(unsigned long data); -extern void tl_restart(struct drbd_conf *mdev, enum drbd_req_event what); +extern void tl_restart(struct drbd_tconn *tconn, enum drbd_req_event what); +extern void _tl_restart(struct drbd_tconn *tconn, enum drbd_req_event what); + +/* this is in drbd_main.c */ +extern void drbd_restart_request(struct drbd_request *req); /* use this if you don't want to deal with calling complete_master_bio() * outside the spinlock, e.g. when walking some list on cleanup. */ static inline int _req_mod(struct drbd_request *req, enum drbd_req_event what) { - struct drbd_conf *mdev = req->mdev; + struct drbd_conf *mdev = req->w.mdev; struct bio_and_error m; int rv; @@ -354,13 +305,13 @@ static inline int req_mod(struct drbd_request *req, enum drbd_req_event what) { unsigned long flags; - struct drbd_conf *mdev = req->mdev; + struct drbd_conf *mdev = req->w.mdev; struct bio_and_error m; int rv; - spin_lock_irqsave(&mdev->req_lock, flags); + spin_lock_irqsave(&mdev->tconn->req_lock, flags); rv = __req_mod(req, what, &m); - spin_unlock_irqrestore(&mdev->req_lock, flags); + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); if (m.bio) complete_master_bio(mdev, &m); @@ -368,7 +319,7 @@ static inline int req_mod(struct drbd_request *req, return rv; } -static inline bool drbd_should_do_remote(union drbd_state s) +static inline bool drbd_should_do_remote(union drbd_dev_state s) { return s.pdsk == D_UP_TO_DATE || (s.pdsk >= D_INCONSISTENT && @@ -378,7 +329,7 @@ static inline bool drbd_should_do_remote(union drbd_state s) That is equivalent since before 96 IO was frozen in the C_WF_BITMAP* states. */ } -static inline bool drbd_should_send_oos(union drbd_state s) +static inline bool drbd_should_send_out_of_sync(union drbd_dev_state s) { return s.conn == C_AHEAD || s.conn == C_WF_BITMAP_S; /* pdsk = D_INCONSISTENT as a consequence. Protocol 96 check not necessary diff --git a/drivers/block/drbd/drbd_state.c b/drivers/block/drbd/drbd_state.c new file mode 100644 index 000000000000..53bf6182bac4 --- /dev/null +++ b/drivers/block/drbd/drbd_state.c @@ -0,0 +1,1856 @@ +/* + drbd_state.c + + This file is part of DRBD by Philipp Reisner and Lars Ellenberg. + + Copyright (C) 2001-2008, LINBIT Information Technologies GmbH. + Copyright (C) 1999-2008, Philipp Reisner <philipp.reisner@linbit.com>. + Copyright (C) 2002-2008, Lars Ellenberg <lars.ellenberg@linbit.com>. + + Thanks to Carter Burden, Bart Grantham and Gennadiy Nerubayev + from Logicworks, Inc. for making SDP replication support possible. + + drbd is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + drbd is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with drbd; see the file COPYING. If not, write to + the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include <linux/drbd_limits.h> +#include "drbd_int.h" +#include "drbd_req.h" + +/* in drbd_main.c */ +extern void tl_abort_disk_io(struct drbd_conf *mdev); + +struct after_state_chg_work { + struct drbd_work w; + union drbd_state os; + union drbd_state ns; + enum chg_state_flags flags; + struct completion *done; +}; + +enum sanitize_state_warnings { + NO_WARNING, + ABORTED_ONLINE_VERIFY, + ABORTED_RESYNC, + CONNECTION_LOST_NEGOTIATING, + IMPLICITLY_UPGRADED_DISK, + IMPLICITLY_UPGRADED_PDSK, +}; + +static int w_after_state_ch(struct drbd_work *w, int unused); +static void after_state_ch(struct drbd_conf *mdev, union drbd_state os, + union drbd_state ns, enum chg_state_flags flags); +static enum drbd_state_rv is_valid_state(struct drbd_conf *, union drbd_state); +static enum drbd_state_rv is_valid_soft_transition(union drbd_state, union drbd_state, struct drbd_tconn *); +static enum drbd_state_rv is_valid_transition(union drbd_state os, union drbd_state ns); +static union drbd_state sanitize_state(struct drbd_conf *mdev, union drbd_state ns, + enum sanitize_state_warnings *warn); + +static inline bool is_susp(union drbd_state s) +{ + return s.susp || s.susp_nod || s.susp_fen; +} + +bool conn_all_vols_unconf(struct drbd_tconn *tconn) +{ + struct drbd_conf *mdev; + bool rv = true; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + if (mdev->state.disk != D_DISKLESS || + mdev->state.conn != C_STANDALONE || + mdev->state.role != R_SECONDARY) { + rv = false; + break; + } + } + rcu_read_unlock(); + + return rv; +} + +/* Unfortunately the states where not correctly ordered, when + they where defined. therefore can not use max_t() here. */ +static enum drbd_role max_role(enum drbd_role role1, enum drbd_role role2) +{ + if (role1 == R_PRIMARY || role2 == R_PRIMARY) + return R_PRIMARY; + if (role1 == R_SECONDARY || role2 == R_SECONDARY) + return R_SECONDARY; + return R_UNKNOWN; +} +static enum drbd_role min_role(enum drbd_role role1, enum drbd_role role2) +{ + if (role1 == R_UNKNOWN || role2 == R_UNKNOWN) + return R_UNKNOWN; + if (role1 == R_SECONDARY || role2 == R_SECONDARY) + return R_SECONDARY; + return R_PRIMARY; +} + +enum drbd_role conn_highest_role(struct drbd_tconn *tconn) +{ + enum drbd_role role = R_UNKNOWN; + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) + role = max_role(role, mdev->state.role); + rcu_read_unlock(); + + return role; +} + +enum drbd_role conn_highest_peer(struct drbd_tconn *tconn) +{ + enum drbd_role peer = R_UNKNOWN; + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) + peer = max_role(peer, mdev->state.peer); + rcu_read_unlock(); + + return peer; +} + +enum drbd_disk_state conn_highest_disk(struct drbd_tconn *tconn) +{ + enum drbd_disk_state ds = D_DISKLESS; + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) + ds = max_t(enum drbd_disk_state, ds, mdev->state.disk); + rcu_read_unlock(); + + return ds; +} + +enum drbd_disk_state conn_lowest_disk(struct drbd_tconn *tconn) +{ + enum drbd_disk_state ds = D_MASK; + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) + ds = min_t(enum drbd_disk_state, ds, mdev->state.disk); + rcu_read_unlock(); + + return ds; +} + +enum drbd_disk_state conn_highest_pdsk(struct drbd_tconn *tconn) +{ + enum drbd_disk_state ds = D_DISKLESS; + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) + ds = max_t(enum drbd_disk_state, ds, mdev->state.pdsk); + rcu_read_unlock(); + + return ds; +} + +enum drbd_conns conn_lowest_conn(struct drbd_tconn *tconn) +{ + enum drbd_conns conn = C_MASK; + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) + conn = min_t(enum drbd_conns, conn, mdev->state.conn); + rcu_read_unlock(); + + return conn; +} + +static bool no_peer_wf_report_params(struct drbd_tconn *tconn) +{ + struct drbd_conf *mdev; + int vnr; + bool rv = true; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) + if (mdev->state.conn == C_WF_REPORT_PARAMS) { + rv = false; + break; + } + rcu_read_unlock(); + + return rv; +} + + +/** + * cl_wide_st_chg() - true if the state change is a cluster wide one + * @mdev: DRBD device. + * @os: old (current) state. + * @ns: new (wanted) state. + */ +static int cl_wide_st_chg(struct drbd_conf *mdev, + union drbd_state os, union drbd_state ns) +{ + return (os.conn >= C_CONNECTED && ns.conn >= C_CONNECTED && + ((os.role != R_PRIMARY && ns.role == R_PRIMARY) || + (os.conn != C_STARTING_SYNC_T && ns.conn == C_STARTING_SYNC_T) || + (os.conn != C_STARTING_SYNC_S && ns.conn == C_STARTING_SYNC_S) || + (os.disk != D_FAILED && ns.disk == D_FAILED))) || + (os.conn >= C_CONNECTED && ns.conn == C_DISCONNECTING) || + (os.conn == C_CONNECTED && ns.conn == C_VERIFY_S) || + (os.conn == C_CONNECTED && ns.conn == C_WF_REPORT_PARAMS); +} + +static union drbd_state +apply_mask_val(union drbd_state os, union drbd_state mask, union drbd_state val) +{ + union drbd_state ns; + ns.i = (os.i & ~mask.i) | val.i; + return ns; +} + +enum drbd_state_rv +drbd_change_state(struct drbd_conf *mdev, enum chg_state_flags f, + union drbd_state mask, union drbd_state val) +{ + unsigned long flags; + union drbd_state ns; + enum drbd_state_rv rv; + + spin_lock_irqsave(&mdev->tconn->req_lock, flags); + ns = apply_mask_val(drbd_read_state(mdev), mask, val); + rv = _drbd_set_state(mdev, ns, f, NULL); + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); + + return rv; +} + +/** + * drbd_force_state() - Impose a change which happens outside our control on our state + * @mdev: DRBD device. + * @mask: mask of state bits to change. + * @val: value of new state bits. + */ +void drbd_force_state(struct drbd_conf *mdev, + union drbd_state mask, union drbd_state val) +{ + drbd_change_state(mdev, CS_HARD, mask, val); +} + +static enum drbd_state_rv +_req_st_cond(struct drbd_conf *mdev, union drbd_state mask, + union drbd_state val) +{ + union drbd_state os, ns; + unsigned long flags; + enum drbd_state_rv rv; + + if (test_and_clear_bit(CL_ST_CHG_SUCCESS, &mdev->flags)) + return SS_CW_SUCCESS; + + if (test_and_clear_bit(CL_ST_CHG_FAIL, &mdev->flags)) + return SS_CW_FAILED_BY_PEER; + + spin_lock_irqsave(&mdev->tconn->req_lock, flags); + os = drbd_read_state(mdev); + ns = sanitize_state(mdev, apply_mask_val(os, mask, val), NULL); + rv = is_valid_transition(os, ns); + if (rv >= SS_SUCCESS) + rv = SS_UNKNOWN_ERROR; /* cont waiting, otherwise fail. */ + + if (!cl_wide_st_chg(mdev, os, ns)) + rv = SS_CW_NO_NEED; + if (rv == SS_UNKNOWN_ERROR) { + rv = is_valid_state(mdev, ns); + if (rv >= SS_SUCCESS) { + rv = is_valid_soft_transition(os, ns, mdev->tconn); + if (rv >= SS_SUCCESS) + rv = SS_UNKNOWN_ERROR; /* cont waiting, otherwise fail. */ + } + } + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); + + return rv; +} + +/** + * drbd_req_state() - Perform an eventually cluster wide state change + * @mdev: DRBD device. + * @mask: mask of state bits to change. + * @val: value of new state bits. + * @f: flags + * + * Should not be called directly, use drbd_request_state() or + * _drbd_request_state(). + */ +static enum drbd_state_rv +drbd_req_state(struct drbd_conf *mdev, union drbd_state mask, + union drbd_state val, enum chg_state_flags f) +{ + struct completion done; + unsigned long flags; + union drbd_state os, ns; + enum drbd_state_rv rv; + + init_completion(&done); + + if (f & CS_SERIALIZE) + mutex_lock(mdev->state_mutex); + + spin_lock_irqsave(&mdev->tconn->req_lock, flags); + os = drbd_read_state(mdev); + ns = sanitize_state(mdev, apply_mask_val(os, mask, val), NULL); + rv = is_valid_transition(os, ns); + if (rv < SS_SUCCESS) { + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); + goto abort; + } + + if (cl_wide_st_chg(mdev, os, ns)) { + rv = is_valid_state(mdev, ns); + if (rv == SS_SUCCESS) + rv = is_valid_soft_transition(os, ns, mdev->tconn); + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); + + if (rv < SS_SUCCESS) { + if (f & CS_VERBOSE) + print_st_err(mdev, os, ns, rv); + goto abort; + } + + if (drbd_send_state_req(mdev, mask, val)) { + rv = SS_CW_FAILED_BY_PEER; + if (f & CS_VERBOSE) + print_st_err(mdev, os, ns, rv); + goto abort; + } + + wait_event(mdev->state_wait, + (rv = _req_st_cond(mdev, mask, val))); + + if (rv < SS_SUCCESS) { + if (f & CS_VERBOSE) + print_st_err(mdev, os, ns, rv); + goto abort; + } + spin_lock_irqsave(&mdev->tconn->req_lock, flags); + ns = apply_mask_val(drbd_read_state(mdev), mask, val); + rv = _drbd_set_state(mdev, ns, f, &done); + } else { + rv = _drbd_set_state(mdev, ns, f, &done); + } + + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); + + if (f & CS_WAIT_COMPLETE && rv == SS_SUCCESS) { + D_ASSERT(current != mdev->tconn->worker.task); + wait_for_completion(&done); + } + +abort: + if (f & CS_SERIALIZE) + mutex_unlock(mdev->state_mutex); + + return rv; +} + +/** + * _drbd_request_state() - Request a state change (with flags) + * @mdev: DRBD device. + * @mask: mask of state bits to change. + * @val: value of new state bits. + * @f: flags + * + * Cousin of drbd_request_state(), useful with the CS_WAIT_COMPLETE + * flag, or when logging of failed state change requests is not desired. + */ +enum drbd_state_rv +_drbd_request_state(struct drbd_conf *mdev, union drbd_state mask, + union drbd_state val, enum chg_state_flags f) +{ + enum drbd_state_rv rv; + + wait_event(mdev->state_wait, + (rv = drbd_req_state(mdev, mask, val, f)) != SS_IN_TRANSIENT_STATE); + + return rv; +} + +static void print_st(struct drbd_conf *mdev, char *name, union drbd_state ns) +{ + dev_err(DEV, " %s = { cs:%s ro:%s/%s ds:%s/%s %c%c%c%c%c%c }\n", + name, + drbd_conn_str(ns.conn), + drbd_role_str(ns.role), + drbd_role_str(ns.peer), + drbd_disk_str(ns.disk), + drbd_disk_str(ns.pdsk), + is_susp(ns) ? 's' : 'r', + ns.aftr_isp ? 'a' : '-', + ns.peer_isp ? 'p' : '-', + ns.user_isp ? 'u' : '-', + ns.susp_fen ? 'F' : '-', + ns.susp_nod ? 'N' : '-' + ); +} + +void print_st_err(struct drbd_conf *mdev, union drbd_state os, + union drbd_state ns, enum drbd_state_rv err) +{ + if (err == SS_IN_TRANSIENT_STATE) + return; + dev_err(DEV, "State change failed: %s\n", drbd_set_st_err_str(err)); + print_st(mdev, " state", os); + print_st(mdev, "wanted", ns); +} + +static long print_state_change(char *pb, union drbd_state os, union drbd_state ns, + enum chg_state_flags flags) +{ + char *pbp; + pbp = pb; + *pbp = 0; + + if (ns.role != os.role && flags & CS_DC_ROLE) + pbp += sprintf(pbp, "role( %s -> %s ) ", + drbd_role_str(os.role), + drbd_role_str(ns.role)); + if (ns.peer != os.peer && flags & CS_DC_PEER) + pbp += sprintf(pbp, "peer( %s -> %s ) ", + drbd_role_str(os.peer), + drbd_role_str(ns.peer)); + if (ns.conn != os.conn && flags & CS_DC_CONN) + pbp += sprintf(pbp, "conn( %s -> %s ) ", + drbd_conn_str(os.conn), + drbd_conn_str(ns.conn)); + if (ns.disk != os.disk && flags & CS_DC_DISK) + pbp += sprintf(pbp, "disk( %s -> %s ) ", + drbd_disk_str(os.disk), + drbd_disk_str(ns.disk)); + if (ns.pdsk != os.pdsk && flags & CS_DC_PDSK) + pbp += sprintf(pbp, "pdsk( %s -> %s ) ", + drbd_disk_str(os.pdsk), + drbd_disk_str(ns.pdsk)); + + return pbp - pb; +} + +static void drbd_pr_state_change(struct drbd_conf *mdev, union drbd_state os, union drbd_state ns, + enum chg_state_flags flags) +{ + char pb[300]; + char *pbp = pb; + + pbp += print_state_change(pbp, os, ns, flags ^ CS_DC_MASK); + + if (ns.aftr_isp != os.aftr_isp) + pbp += sprintf(pbp, "aftr_isp( %d -> %d ) ", + os.aftr_isp, + ns.aftr_isp); + if (ns.peer_isp != os.peer_isp) + pbp += sprintf(pbp, "peer_isp( %d -> %d ) ", + os.peer_isp, + ns.peer_isp); + if (ns.user_isp != os.user_isp) + pbp += sprintf(pbp, "user_isp( %d -> %d ) ", + os.user_isp, + ns.user_isp); + + if (pbp != pb) + dev_info(DEV, "%s\n", pb); +} + +static void conn_pr_state_change(struct drbd_tconn *tconn, union drbd_state os, union drbd_state ns, + enum chg_state_flags flags) +{ + char pb[300]; + char *pbp = pb; + + pbp += print_state_change(pbp, os, ns, flags); + + if (is_susp(ns) != is_susp(os) && flags & CS_DC_SUSP) + pbp += sprintf(pbp, "susp( %d -> %d ) ", + is_susp(os), + is_susp(ns)); + + if (pbp != pb) + conn_info(tconn, "%s\n", pb); +} + + +/** + * is_valid_state() - Returns an SS_ error code if ns is not valid + * @mdev: DRBD device. + * @ns: State to consider. + */ +static enum drbd_state_rv +is_valid_state(struct drbd_conf *mdev, union drbd_state ns) +{ + /* See drbd_state_sw_errors in drbd_strings.c */ + + enum drbd_fencing_p fp; + enum drbd_state_rv rv = SS_SUCCESS; + struct net_conf *nc; + + rcu_read_lock(); + fp = FP_DONT_CARE; + if (get_ldev(mdev)) { + fp = rcu_dereference(mdev->ldev->disk_conf)->fencing; + put_ldev(mdev); + } + + nc = rcu_dereference(mdev->tconn->net_conf); + if (nc) { + if (!nc->two_primaries && ns.role == R_PRIMARY) { + if (ns.peer == R_PRIMARY) + rv = SS_TWO_PRIMARIES; + else if (conn_highest_peer(mdev->tconn) == R_PRIMARY) + rv = SS_O_VOL_PEER_PRI; + } + } + + if (rv <= 0) + /* already found a reason to abort */; + else if (ns.role == R_SECONDARY && mdev->open_cnt) + rv = SS_DEVICE_IN_USE; + + else if (ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.disk < D_UP_TO_DATE) + rv = SS_NO_UP_TO_DATE_DISK; + + else if (fp >= FP_RESOURCE && + ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.pdsk >= D_UNKNOWN) + rv = SS_PRIMARY_NOP; + + else if (ns.role == R_PRIMARY && ns.disk <= D_INCONSISTENT && ns.pdsk <= D_INCONSISTENT) + rv = SS_NO_UP_TO_DATE_DISK; + + else if (ns.conn > C_CONNECTED && ns.disk < D_INCONSISTENT) + rv = SS_NO_LOCAL_DISK; + + else if (ns.conn > C_CONNECTED && ns.pdsk < D_INCONSISTENT) + rv = SS_NO_REMOTE_DISK; + + else if (ns.conn > C_CONNECTED && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE) + rv = SS_NO_UP_TO_DATE_DISK; + + else if ((ns.conn == C_CONNECTED || + ns.conn == C_WF_BITMAP_S || + ns.conn == C_SYNC_SOURCE || + ns.conn == C_PAUSED_SYNC_S) && + ns.disk == D_OUTDATED) + rv = SS_CONNECTED_OUTDATES; + + else if ((ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T) && + (nc->verify_alg[0] == 0)) + rv = SS_NO_VERIFY_ALG; + + else if ((ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T) && + mdev->tconn->agreed_pro_version < 88) + rv = SS_NOT_SUPPORTED; + + else if (ns.conn >= C_CONNECTED && ns.pdsk == D_UNKNOWN) + rv = SS_CONNECTED_OUTDATES; + + rcu_read_unlock(); + + return rv; +} + +/** + * is_valid_soft_transition() - Returns an SS_ error code if the state transition is not possible + * This function limits state transitions that may be declined by DRBD. I.e. + * user requests (aka soft transitions). + * @mdev: DRBD device. + * @ns: new state. + * @os: old state. + */ +static enum drbd_state_rv +is_valid_soft_transition(union drbd_state os, union drbd_state ns, struct drbd_tconn *tconn) +{ + enum drbd_state_rv rv = SS_SUCCESS; + + if ((ns.conn == C_STARTING_SYNC_T || ns.conn == C_STARTING_SYNC_S) && + os.conn > C_CONNECTED) + rv = SS_RESYNC_RUNNING; + + if (ns.conn == C_DISCONNECTING && os.conn == C_STANDALONE) + rv = SS_ALREADY_STANDALONE; + + if (ns.disk > D_ATTACHING && os.disk == D_DISKLESS) + rv = SS_IS_DISKLESS; + + if (ns.conn == C_WF_CONNECTION && os.conn < C_UNCONNECTED) + rv = SS_NO_NET_CONFIG; + + if (ns.disk == D_OUTDATED && os.disk < D_OUTDATED && os.disk != D_ATTACHING) + rv = SS_LOWER_THAN_OUTDATED; + + if (ns.conn == C_DISCONNECTING && os.conn == C_UNCONNECTED) + rv = SS_IN_TRANSIENT_STATE; + + /* if (ns.conn == os.conn && ns.conn == C_WF_REPORT_PARAMS) + rv = SS_IN_TRANSIENT_STATE; */ + + /* While establishing a connection only allow cstate to change. + Delay/refuse role changes, detach attach etc... */ + if (test_bit(STATE_SENT, &tconn->flags) && + !(os.conn == C_WF_REPORT_PARAMS || + (ns.conn == C_WF_REPORT_PARAMS && os.conn == C_WF_CONNECTION))) + rv = SS_IN_TRANSIENT_STATE; + + if ((ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T) && os.conn < C_CONNECTED) + rv = SS_NEED_CONNECTION; + + if ((ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T) && + ns.conn != os.conn && os.conn > C_CONNECTED) + rv = SS_RESYNC_RUNNING; + + if ((ns.conn == C_STARTING_SYNC_S || ns.conn == C_STARTING_SYNC_T) && + os.conn < C_CONNECTED) + rv = SS_NEED_CONNECTION; + + if ((ns.conn == C_SYNC_TARGET || ns.conn == C_SYNC_SOURCE) + && os.conn < C_WF_REPORT_PARAMS) + rv = SS_NEED_CONNECTION; /* No NetworkFailure -> SyncTarget etc... */ + + return rv; +} + +static enum drbd_state_rv +is_valid_conn_transition(enum drbd_conns oc, enum drbd_conns nc) +{ + /* no change -> nothing to do, at least for the connection part */ + if (oc == nc) + return SS_NOTHING_TO_DO; + + /* disconnect of an unconfigured connection does not make sense */ + if (oc == C_STANDALONE && nc == C_DISCONNECTING) + return SS_ALREADY_STANDALONE; + + /* from C_STANDALONE, we start with C_UNCONNECTED */ + if (oc == C_STANDALONE && nc != C_UNCONNECTED) + return SS_NEED_CONNECTION; + + /* When establishing a connection we need to go through WF_REPORT_PARAMS! + Necessary to do the right thing upon invalidate-remote on a disconnected resource */ + if (oc < C_WF_REPORT_PARAMS && nc >= C_CONNECTED) + return SS_NEED_CONNECTION; + + /* After a network error only C_UNCONNECTED or C_DISCONNECTING may follow. */ + if (oc >= C_TIMEOUT && oc <= C_TEAR_DOWN && nc != C_UNCONNECTED && nc != C_DISCONNECTING) + return SS_IN_TRANSIENT_STATE; + + /* After C_DISCONNECTING only C_STANDALONE may follow */ + if (oc == C_DISCONNECTING && nc != C_STANDALONE) + return SS_IN_TRANSIENT_STATE; + + return SS_SUCCESS; +} + + +/** + * is_valid_transition() - Returns an SS_ error code if the state transition is not possible + * This limits hard state transitions. Hard state transitions are facts there are + * imposed on DRBD by the environment. E.g. disk broke or network broke down. + * But those hard state transitions are still not allowed to do everything. + * @ns: new state. + * @os: old state. + */ +static enum drbd_state_rv +is_valid_transition(union drbd_state os, union drbd_state ns) +{ + enum drbd_state_rv rv; + + rv = is_valid_conn_transition(os.conn, ns.conn); + + /* we cannot fail (again) if we already detached */ + if (ns.disk == D_FAILED && os.disk == D_DISKLESS) + rv = SS_IS_DISKLESS; + + return rv; +} + +static void print_sanitize_warnings(struct drbd_conf *mdev, enum sanitize_state_warnings warn) +{ + static const char *msg_table[] = { + [NO_WARNING] = "", + [ABORTED_ONLINE_VERIFY] = "Online-verify aborted.", + [ABORTED_RESYNC] = "Resync aborted.", + [CONNECTION_LOST_NEGOTIATING] = "Connection lost while negotiating, no data!", + [IMPLICITLY_UPGRADED_DISK] = "Implicitly upgraded disk", + [IMPLICITLY_UPGRADED_PDSK] = "Implicitly upgraded pdsk", + }; + + if (warn != NO_WARNING) + dev_warn(DEV, "%s\n", msg_table[warn]); +} + +/** + * sanitize_state() - Resolves implicitly necessary additional changes to a state transition + * @mdev: DRBD device. + * @os: old state. + * @ns: new state. + * @warn_sync_abort: + * + * When we loose connection, we have to set the state of the peers disk (pdsk) + * to D_UNKNOWN. This rule and many more along those lines are in this function. + */ +static union drbd_state sanitize_state(struct drbd_conf *mdev, union drbd_state ns, + enum sanitize_state_warnings *warn) +{ + enum drbd_fencing_p fp; + enum drbd_disk_state disk_min, disk_max, pdsk_min, pdsk_max; + + if (warn) + *warn = NO_WARNING; + + fp = FP_DONT_CARE; + if (get_ldev(mdev)) { + rcu_read_lock(); + fp = rcu_dereference(mdev->ldev->disk_conf)->fencing; + rcu_read_unlock(); + put_ldev(mdev); + } + + /* Implications from connection to peer and peer_isp */ + if (ns.conn < C_CONNECTED) { + ns.peer_isp = 0; + ns.peer = R_UNKNOWN; + if (ns.pdsk > D_UNKNOWN || ns.pdsk < D_INCONSISTENT) + ns.pdsk = D_UNKNOWN; + } + + /* Clear the aftr_isp when becoming unconfigured */ + if (ns.conn == C_STANDALONE && ns.disk == D_DISKLESS && ns.role == R_SECONDARY) + ns.aftr_isp = 0; + + /* An implication of the disk states onto the connection state */ + /* Abort resync if a disk fails/detaches */ + if (ns.conn > C_CONNECTED && (ns.disk <= D_FAILED || ns.pdsk <= D_FAILED)) { + if (warn) + *warn = ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T ? + ABORTED_ONLINE_VERIFY : ABORTED_RESYNC; + ns.conn = C_CONNECTED; + } + + /* Connection breaks down before we finished "Negotiating" */ + if (ns.conn < C_CONNECTED && ns.disk == D_NEGOTIATING && + get_ldev_if_state(mdev, D_NEGOTIATING)) { + if (mdev->ed_uuid == mdev->ldev->md.uuid[UI_CURRENT]) { + ns.disk = mdev->new_state_tmp.disk; + ns.pdsk = mdev->new_state_tmp.pdsk; + } else { + if (warn) + *warn = CONNECTION_LOST_NEGOTIATING; + ns.disk = D_DISKLESS; + ns.pdsk = D_UNKNOWN; + } + put_ldev(mdev); + } + + /* D_CONSISTENT and D_OUTDATED vanish when we get connected */ + if (ns.conn >= C_CONNECTED && ns.conn < C_AHEAD) { + if (ns.disk == D_CONSISTENT || ns.disk == D_OUTDATED) + ns.disk = D_UP_TO_DATE; + if (ns.pdsk == D_CONSISTENT || ns.pdsk == D_OUTDATED) + ns.pdsk = D_UP_TO_DATE; + } + + /* Implications of the connection stat on the disk states */ + disk_min = D_DISKLESS; + disk_max = D_UP_TO_DATE; + pdsk_min = D_INCONSISTENT; + pdsk_max = D_UNKNOWN; + switch ((enum drbd_conns)ns.conn) { + case C_WF_BITMAP_T: + case C_PAUSED_SYNC_T: + case C_STARTING_SYNC_T: + case C_WF_SYNC_UUID: + case C_BEHIND: + disk_min = D_INCONSISTENT; + disk_max = D_OUTDATED; + pdsk_min = D_UP_TO_DATE; + pdsk_max = D_UP_TO_DATE; + break; + case C_VERIFY_S: + case C_VERIFY_T: + disk_min = D_UP_TO_DATE; + disk_max = D_UP_TO_DATE; + pdsk_min = D_UP_TO_DATE; + pdsk_max = D_UP_TO_DATE; + break; + case C_CONNECTED: + disk_min = D_DISKLESS; + disk_max = D_UP_TO_DATE; + pdsk_min = D_DISKLESS; + pdsk_max = D_UP_TO_DATE; + break; + case C_WF_BITMAP_S: + case C_PAUSED_SYNC_S: + case C_STARTING_SYNC_S: + case C_AHEAD: + disk_min = D_UP_TO_DATE; + disk_max = D_UP_TO_DATE; + pdsk_min = D_INCONSISTENT; + pdsk_max = D_CONSISTENT; /* D_OUTDATED would be nice. But explicit outdate necessary*/ + break; + case C_SYNC_TARGET: + disk_min = D_INCONSISTENT; + disk_max = D_INCONSISTENT; + pdsk_min = D_UP_TO_DATE; + pdsk_max = D_UP_TO_DATE; + break; + case C_SYNC_SOURCE: + disk_min = D_UP_TO_DATE; + disk_max = D_UP_TO_DATE; + pdsk_min = D_INCONSISTENT; + pdsk_max = D_INCONSISTENT; + break; + case C_STANDALONE: + case C_DISCONNECTING: + case C_UNCONNECTED: + case C_TIMEOUT: + case C_BROKEN_PIPE: + case C_NETWORK_FAILURE: + case C_PROTOCOL_ERROR: + case C_TEAR_DOWN: + case C_WF_CONNECTION: + case C_WF_REPORT_PARAMS: + case C_MASK: + break; + } + if (ns.disk > disk_max) + ns.disk = disk_max; + + if (ns.disk < disk_min) { + if (warn) + *warn = IMPLICITLY_UPGRADED_DISK; + ns.disk = disk_min; + } + if (ns.pdsk > pdsk_max) + ns.pdsk = pdsk_max; + + if (ns.pdsk < pdsk_min) { + if (warn) + *warn = IMPLICITLY_UPGRADED_PDSK; + ns.pdsk = pdsk_min; + } + + if (fp == FP_STONITH && + (ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.pdsk > D_OUTDATED)) + ns.susp_fen = 1; /* Suspend IO while fence-peer handler runs (peer lost) */ + + if (mdev->tconn->res_opts.on_no_data == OND_SUSPEND_IO && + (ns.role == R_PRIMARY && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE)) + ns.susp_nod = 1; /* Suspend IO while no data available (no accessible data available) */ + + if (ns.aftr_isp || ns.peer_isp || ns.user_isp) { + if (ns.conn == C_SYNC_SOURCE) + ns.conn = C_PAUSED_SYNC_S; + if (ns.conn == C_SYNC_TARGET) + ns.conn = C_PAUSED_SYNC_T; + } else { + if (ns.conn == C_PAUSED_SYNC_S) + ns.conn = C_SYNC_SOURCE; + if (ns.conn == C_PAUSED_SYNC_T) + ns.conn = C_SYNC_TARGET; + } + + return ns; +} + +void drbd_resume_al(struct drbd_conf *mdev) +{ + if (test_and_clear_bit(AL_SUSPENDED, &mdev->flags)) + dev_info(DEV, "Resumed AL updates\n"); +} + +/* helper for __drbd_set_state */ +static void set_ov_position(struct drbd_conf *mdev, enum drbd_conns cs) +{ + if (mdev->tconn->agreed_pro_version < 90) + mdev->ov_start_sector = 0; + mdev->rs_total = drbd_bm_bits(mdev); + mdev->ov_position = 0; + if (cs == C_VERIFY_T) { + /* starting online verify from an arbitrary position + * does not fit well into the existing protocol. + * on C_VERIFY_T, we initialize ov_left and friends + * implicitly in receive_DataRequest once the + * first P_OV_REQUEST is received */ + mdev->ov_start_sector = ~(sector_t)0; + } else { + unsigned long bit = BM_SECT_TO_BIT(mdev->ov_start_sector); + if (bit >= mdev->rs_total) { + mdev->ov_start_sector = + BM_BIT_TO_SECT(mdev->rs_total - 1); + mdev->rs_total = 1; + } else + mdev->rs_total -= bit; + mdev->ov_position = mdev->ov_start_sector; + } + mdev->ov_left = mdev->rs_total; +} + +/** + * __drbd_set_state() - Set a new DRBD state + * @mdev: DRBD device. + * @ns: new state. + * @flags: Flags + * @done: Optional completion, that will get completed after the after_state_ch() finished + * + * Caller needs to hold req_lock, and global_state_lock. Do not call directly. + */ +enum drbd_state_rv +__drbd_set_state(struct drbd_conf *mdev, union drbd_state ns, + enum chg_state_flags flags, struct completion *done) +{ + union drbd_state os; + enum drbd_state_rv rv = SS_SUCCESS; + enum sanitize_state_warnings ssw; + struct after_state_chg_work *ascw; + + os = drbd_read_state(mdev); + + ns = sanitize_state(mdev, ns, &ssw); + if (ns.i == os.i) + return SS_NOTHING_TO_DO; + + rv = is_valid_transition(os, ns); + if (rv < SS_SUCCESS) + return rv; + + if (!(flags & CS_HARD)) { + /* pre-state-change checks ; only look at ns */ + /* See drbd_state_sw_errors in drbd_strings.c */ + + rv = is_valid_state(mdev, ns); + if (rv < SS_SUCCESS) { + /* If the old state was illegal as well, then let + this happen...*/ + + if (is_valid_state(mdev, os) == rv) + rv = is_valid_soft_transition(os, ns, mdev->tconn); + } else + rv = is_valid_soft_transition(os, ns, mdev->tconn); + } + + if (rv < SS_SUCCESS) { + if (flags & CS_VERBOSE) + print_st_err(mdev, os, ns, rv); + return rv; + } + + print_sanitize_warnings(mdev, ssw); + + drbd_pr_state_change(mdev, os, ns, flags); + + /* Display changes to the susp* flags that where caused by the call to + sanitize_state(). Only display it here if we where not called from + _conn_request_state() */ + if (!(flags & CS_DC_SUSP)) + conn_pr_state_change(mdev->tconn, os, ns, (flags & ~CS_DC_MASK) | CS_DC_SUSP); + + /* if we are going -> D_FAILED or D_DISKLESS, grab one extra reference + * on the ldev here, to be sure the transition -> D_DISKLESS resp. + * drbd_ldev_destroy() won't happen before our corresponding + * after_state_ch works run, where we put_ldev again. */ + if ((os.disk != D_FAILED && ns.disk == D_FAILED) || + (os.disk != D_DISKLESS && ns.disk == D_DISKLESS)) + atomic_inc(&mdev->local_cnt); + + mdev->state.i = ns.i; + mdev->tconn->susp = ns.susp; + mdev->tconn->susp_nod = ns.susp_nod; + mdev->tconn->susp_fen = ns.susp_fen; + + if (os.disk == D_ATTACHING && ns.disk >= D_NEGOTIATING) + drbd_print_uuids(mdev, "attached to UUIDs"); + + /* Wake up role changes, that were delayed because of connection establishing */ + if (os.conn == C_WF_REPORT_PARAMS && ns.conn != C_WF_REPORT_PARAMS && + no_peer_wf_report_params(mdev->tconn)) + clear_bit(STATE_SENT, &mdev->tconn->flags); + + wake_up(&mdev->misc_wait); + wake_up(&mdev->state_wait); + wake_up(&mdev->tconn->ping_wait); + + /* Aborted verify run, or we reached the stop sector. + * Log the last position, unless end-of-device. */ + if ((os.conn == C_VERIFY_S || os.conn == C_VERIFY_T) && + ns.conn <= C_CONNECTED) { + mdev->ov_start_sector = + BM_BIT_TO_SECT(drbd_bm_bits(mdev) - mdev->ov_left); + if (mdev->ov_left) + dev_info(DEV, "Online Verify reached sector %llu\n", + (unsigned long long)mdev->ov_start_sector); + } + + if ((os.conn == C_PAUSED_SYNC_T || os.conn == C_PAUSED_SYNC_S) && + (ns.conn == C_SYNC_TARGET || ns.conn == C_SYNC_SOURCE)) { + dev_info(DEV, "Syncer continues.\n"); + mdev->rs_paused += (long)jiffies + -(long)mdev->rs_mark_time[mdev->rs_last_mark]; + if (ns.conn == C_SYNC_TARGET) + mod_timer(&mdev->resync_timer, jiffies); + } + + if ((os.conn == C_SYNC_TARGET || os.conn == C_SYNC_SOURCE) && + (ns.conn == C_PAUSED_SYNC_T || ns.conn == C_PAUSED_SYNC_S)) { + dev_info(DEV, "Resync suspended\n"); + mdev->rs_mark_time[mdev->rs_last_mark] = jiffies; + } + + if (os.conn == C_CONNECTED && + (ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T)) { + unsigned long now = jiffies; + int i; + + set_ov_position(mdev, ns.conn); + mdev->rs_start = now; + mdev->rs_last_events = 0; + mdev->rs_last_sect_ev = 0; + mdev->ov_last_oos_size = 0; + mdev->ov_last_oos_start = 0; + + for (i = 0; i < DRBD_SYNC_MARKS; i++) { + mdev->rs_mark_left[i] = mdev->ov_left; + mdev->rs_mark_time[i] = now; + } + + drbd_rs_controller_reset(mdev); + + if (ns.conn == C_VERIFY_S) { + dev_info(DEV, "Starting Online Verify from sector %llu\n", + (unsigned long long)mdev->ov_position); + mod_timer(&mdev->resync_timer, jiffies); + } + } + + if (get_ldev(mdev)) { + u32 mdf = mdev->ldev->md.flags & ~(MDF_CONSISTENT|MDF_PRIMARY_IND| + MDF_CONNECTED_IND|MDF_WAS_UP_TO_DATE| + MDF_PEER_OUT_DATED|MDF_CRASHED_PRIMARY); + + mdf &= ~MDF_AL_CLEAN; + if (test_bit(CRASHED_PRIMARY, &mdev->flags)) + mdf |= MDF_CRASHED_PRIMARY; + if (mdev->state.role == R_PRIMARY || + (mdev->state.pdsk < D_INCONSISTENT && mdev->state.peer == R_PRIMARY)) + mdf |= MDF_PRIMARY_IND; + if (mdev->state.conn > C_WF_REPORT_PARAMS) + mdf |= MDF_CONNECTED_IND; + if (mdev->state.disk > D_INCONSISTENT) + mdf |= MDF_CONSISTENT; + if (mdev->state.disk > D_OUTDATED) + mdf |= MDF_WAS_UP_TO_DATE; + if (mdev->state.pdsk <= D_OUTDATED && mdev->state.pdsk >= D_INCONSISTENT) + mdf |= MDF_PEER_OUT_DATED; + if (mdf != mdev->ldev->md.flags) { + mdev->ldev->md.flags = mdf; + drbd_md_mark_dirty(mdev); + } + if (os.disk < D_CONSISTENT && ns.disk >= D_CONSISTENT) + drbd_set_ed_uuid(mdev, mdev->ldev->md.uuid[UI_CURRENT]); + put_ldev(mdev); + } + + /* Peer was forced D_UP_TO_DATE & R_PRIMARY, consider to resync */ + if (os.disk == D_INCONSISTENT && os.pdsk == D_INCONSISTENT && + os.peer == R_SECONDARY && ns.peer == R_PRIMARY) + set_bit(CONSIDER_RESYNC, &mdev->flags); + + /* Receiver should clean up itself */ + if (os.conn != C_DISCONNECTING && ns.conn == C_DISCONNECTING) + drbd_thread_stop_nowait(&mdev->tconn->receiver); + + /* Now the receiver finished cleaning up itself, it should die */ + if (os.conn != C_STANDALONE && ns.conn == C_STANDALONE) + drbd_thread_stop_nowait(&mdev->tconn->receiver); + + /* Upon network failure, we need to restart the receiver. */ + if (os.conn > C_WF_CONNECTION && + ns.conn <= C_TEAR_DOWN && ns.conn >= C_TIMEOUT) + drbd_thread_restart_nowait(&mdev->tconn->receiver); + + /* Resume AL writing if we get a connection */ + if (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED) + drbd_resume_al(mdev); + + /* remember last attach time so request_timer_fn() won't + * kill newly established sessions while we are still trying to thaw + * previously frozen IO */ + if ((os.disk == D_ATTACHING || os.disk == D_NEGOTIATING) && + ns.disk > D_NEGOTIATING) + mdev->last_reattach_jif = jiffies; + + ascw = kmalloc(sizeof(*ascw), GFP_ATOMIC); + if (ascw) { + ascw->os = os; + ascw->ns = ns; + ascw->flags = flags; + ascw->w.cb = w_after_state_ch; + ascw->w.mdev = mdev; + ascw->done = done; + drbd_queue_work(&mdev->tconn->sender_work, &ascw->w); + } else { + dev_err(DEV, "Could not kmalloc an ascw\n"); + } + + return rv; +} + +static int w_after_state_ch(struct drbd_work *w, int unused) +{ + struct after_state_chg_work *ascw = + container_of(w, struct after_state_chg_work, w); + struct drbd_conf *mdev = w->mdev; + + after_state_ch(mdev, ascw->os, ascw->ns, ascw->flags); + if (ascw->flags & CS_WAIT_COMPLETE) { + D_ASSERT(ascw->done != NULL); + complete(ascw->done); + } + kfree(ascw); + + return 0; +} + +static void abw_start_sync(struct drbd_conf *mdev, int rv) +{ + if (rv) { + dev_err(DEV, "Writing the bitmap failed not starting resync.\n"); + _drbd_request_state(mdev, NS(conn, C_CONNECTED), CS_VERBOSE); + return; + } + + switch (mdev->state.conn) { + case C_STARTING_SYNC_T: + _drbd_request_state(mdev, NS(conn, C_WF_SYNC_UUID), CS_VERBOSE); + break; + case C_STARTING_SYNC_S: + drbd_start_resync(mdev, C_SYNC_SOURCE); + break; + } +} + +int drbd_bitmap_io_from_worker(struct drbd_conf *mdev, + int (*io_fn)(struct drbd_conf *), + char *why, enum bm_flag flags) +{ + int rv; + + D_ASSERT(current == mdev->tconn->worker.task); + + /* open coded non-blocking drbd_suspend_io(mdev); */ + set_bit(SUSPEND_IO, &mdev->flags); + + drbd_bm_lock(mdev, why, flags); + rv = io_fn(mdev); + drbd_bm_unlock(mdev); + + drbd_resume_io(mdev); + + return rv; +} + +/** + * after_state_ch() - Perform after state change actions that may sleep + * @mdev: DRBD device. + * @os: old state. + * @ns: new state. + * @flags: Flags + */ +static void after_state_ch(struct drbd_conf *mdev, union drbd_state os, + union drbd_state ns, enum chg_state_flags flags) +{ + struct sib_info sib; + + sib.sib_reason = SIB_STATE_CHANGE; + sib.os = os; + sib.ns = ns; + + if (os.conn != C_CONNECTED && ns.conn == C_CONNECTED) { + clear_bit(CRASHED_PRIMARY, &mdev->flags); + if (mdev->p_uuid) + mdev->p_uuid[UI_FLAGS] &= ~((u64)2); + } + + /* Inform userspace about the change... */ + drbd_bcast_event(mdev, &sib); + + if (!(os.role == R_PRIMARY && os.disk < D_UP_TO_DATE && os.pdsk < D_UP_TO_DATE) && + (ns.role == R_PRIMARY && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE)) + drbd_khelper(mdev, "pri-on-incon-degr"); + + /* Here we have the actions that are performed after a + state change. This function might sleep */ + + if (ns.susp_nod) { + struct drbd_tconn *tconn = mdev->tconn; + enum drbd_req_event what = NOTHING; + + spin_lock_irq(&tconn->req_lock); + if (os.conn < C_CONNECTED && conn_lowest_conn(tconn) >= C_CONNECTED) + what = RESEND; + + if ((os.disk == D_ATTACHING || os.disk == D_NEGOTIATING) && + conn_lowest_disk(tconn) > D_NEGOTIATING) + what = RESTART_FROZEN_DISK_IO; + + if (tconn->susp_nod && what != NOTHING) { + _tl_restart(tconn, what); + _conn_request_state(tconn, + (union drbd_state) { { .susp_nod = 1 } }, + (union drbd_state) { { .susp_nod = 0 } }, + CS_VERBOSE); + } + spin_unlock_irq(&tconn->req_lock); + } + + if (ns.susp_fen) { + struct drbd_tconn *tconn = mdev->tconn; + + spin_lock_irq(&tconn->req_lock); + if (tconn->susp_fen && conn_lowest_conn(tconn) >= C_CONNECTED) { + /* case2: The connection was established again: */ + struct drbd_conf *odev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, odev, vnr) + clear_bit(NEW_CUR_UUID, &odev->flags); + rcu_read_unlock(); + _tl_restart(tconn, RESEND); + _conn_request_state(tconn, + (union drbd_state) { { .susp_fen = 1 } }, + (union drbd_state) { { .susp_fen = 0 } }, + CS_VERBOSE); + } + spin_unlock_irq(&tconn->req_lock); + } + + /* Became sync source. With protocol >= 96, we still need to send out + * the sync uuid now. Need to do that before any drbd_send_state, or + * the other side may go "paused sync" before receiving the sync uuids, + * which is unexpected. */ + if ((os.conn != C_SYNC_SOURCE && os.conn != C_PAUSED_SYNC_S) && + (ns.conn == C_SYNC_SOURCE || ns.conn == C_PAUSED_SYNC_S) && + mdev->tconn->agreed_pro_version >= 96 && get_ldev(mdev)) { + drbd_gen_and_send_sync_uuid(mdev); + put_ldev(mdev); + } + + /* Do not change the order of the if above and the two below... */ + if (os.pdsk == D_DISKLESS && + ns.pdsk > D_DISKLESS && ns.pdsk != D_UNKNOWN) { /* attach on the peer */ + /* we probably will start a resync soon. + * make sure those things are properly reset. */ + mdev->rs_total = 0; + mdev->rs_failed = 0; + atomic_set(&mdev->rs_pending_cnt, 0); + drbd_rs_cancel_all(mdev); + + drbd_send_uuids(mdev); + drbd_send_state(mdev, ns); + } + /* No point in queuing send_bitmap if we don't have a connection + * anymore, so check also the _current_ state, not only the new state + * at the time this work was queued. */ + if (os.conn != C_WF_BITMAP_S && ns.conn == C_WF_BITMAP_S && + mdev->state.conn == C_WF_BITMAP_S) + drbd_queue_bitmap_io(mdev, &drbd_send_bitmap, NULL, + "send_bitmap (WFBitMapS)", + BM_LOCKED_TEST_ALLOWED); + + /* Lost contact to peer's copy of the data */ + if ((os.pdsk >= D_INCONSISTENT && + os.pdsk != D_UNKNOWN && + os.pdsk != D_OUTDATED) + && (ns.pdsk < D_INCONSISTENT || + ns.pdsk == D_UNKNOWN || + ns.pdsk == D_OUTDATED)) { + if (get_ldev(mdev)) { + if ((ns.role == R_PRIMARY || ns.peer == R_PRIMARY) && + mdev->ldev->md.uuid[UI_BITMAP] == 0 && ns.disk >= D_UP_TO_DATE) { + if (drbd_suspended(mdev)) { + set_bit(NEW_CUR_UUID, &mdev->flags); + } else { + drbd_uuid_new_current(mdev); + drbd_send_uuids(mdev); + } + } + put_ldev(mdev); + } + } + + if (ns.pdsk < D_INCONSISTENT && get_ldev(mdev)) { + if (os.peer == R_SECONDARY && ns.peer == R_PRIMARY && + mdev->ldev->md.uuid[UI_BITMAP] == 0 && ns.disk >= D_UP_TO_DATE) { + drbd_uuid_new_current(mdev); + drbd_send_uuids(mdev); + } + /* D_DISKLESS Peer becomes secondary */ + if (os.peer == R_PRIMARY && ns.peer == R_SECONDARY) + /* We may still be Primary ourselves. + * No harm done if the bitmap still changes, + * redirtied pages will follow later. */ + drbd_bitmap_io_from_worker(mdev, &drbd_bm_write, + "demote diskless peer", BM_LOCKED_SET_ALLOWED); + put_ldev(mdev); + } + + /* Write out all changed bits on demote. + * Though, no need to da that just yet + * if there is a resync going on still */ + if (os.role == R_PRIMARY && ns.role == R_SECONDARY && + mdev->state.conn <= C_CONNECTED && get_ldev(mdev)) { + /* No changes to the bitmap expected this time, so assert that, + * even though no harm was done if it did change. */ + drbd_bitmap_io_from_worker(mdev, &drbd_bm_write, + "demote", BM_LOCKED_TEST_ALLOWED); + put_ldev(mdev); + } + + /* Last part of the attaching process ... */ + if (ns.conn >= C_CONNECTED && + os.disk == D_ATTACHING && ns.disk == D_NEGOTIATING) { + drbd_send_sizes(mdev, 0, 0); /* to start sync... */ + drbd_send_uuids(mdev); + drbd_send_state(mdev, ns); + } + + /* We want to pause/continue resync, tell peer. */ + if (ns.conn >= C_CONNECTED && + ((os.aftr_isp != ns.aftr_isp) || + (os.user_isp != ns.user_isp))) + drbd_send_state(mdev, ns); + + /* In case one of the isp bits got set, suspend other devices. */ + if ((!os.aftr_isp && !os.peer_isp && !os.user_isp) && + (ns.aftr_isp || ns.peer_isp || ns.user_isp)) + suspend_other_sg(mdev); + + /* Make sure the peer gets informed about eventual state + changes (ISP bits) while we were in WFReportParams. */ + if (os.conn == C_WF_REPORT_PARAMS && ns.conn >= C_CONNECTED) + drbd_send_state(mdev, ns); + + if (os.conn != C_AHEAD && ns.conn == C_AHEAD) + drbd_send_state(mdev, ns); + + /* We are in the progress to start a full sync... */ + if ((os.conn != C_STARTING_SYNC_T && ns.conn == C_STARTING_SYNC_T) || + (os.conn != C_STARTING_SYNC_S && ns.conn == C_STARTING_SYNC_S)) + /* no other bitmap changes expected during this phase */ + drbd_queue_bitmap_io(mdev, + &drbd_bmio_set_n_write, &abw_start_sync, + "set_n_write from StartingSync", BM_LOCKED_TEST_ALLOWED); + + /* We are invalidating our self... */ + if (os.conn < C_CONNECTED && ns.conn < C_CONNECTED && + os.disk > D_INCONSISTENT && ns.disk == D_INCONSISTENT) + /* other bitmap operation expected during this phase */ + drbd_queue_bitmap_io(mdev, &drbd_bmio_set_n_write, NULL, + "set_n_write from invalidate", BM_LOCKED_MASK); + + /* first half of local IO error, failure to attach, + * or administrative detach */ + if (os.disk != D_FAILED && ns.disk == D_FAILED) { + enum drbd_io_error_p eh = EP_PASS_ON; + int was_io_error = 0; + /* corresponding get_ldev was in __drbd_set_state, to serialize + * our cleanup here with the transition to D_DISKLESS. + * But is is still not save to dreference ldev here, since + * we might come from an failed Attach before ldev was set. */ + if (mdev->ldev) { + rcu_read_lock(); + eh = rcu_dereference(mdev->ldev->disk_conf)->on_io_error; + rcu_read_unlock(); + + was_io_error = test_and_clear_bit(WAS_IO_ERROR, &mdev->flags); + + if (was_io_error && eh == EP_CALL_HELPER) + drbd_khelper(mdev, "local-io-error"); + + /* Immediately allow completion of all application IO, + * that waits for completion from the local disk, + * if this was a force-detach due to disk_timeout + * or administrator request (drbdsetup detach --force). + * Do NOT abort otherwise. + * Aborting local requests may cause serious problems, + * if requests are completed to upper layers already, + * and then later the already submitted local bio completes. + * This can cause DMA into former bio pages that meanwhile + * have been re-used for other things. + * So aborting local requests may cause crashes, + * or even worse, silent data corruption. + */ + if (test_and_clear_bit(FORCE_DETACH, &mdev->flags)) + tl_abort_disk_io(mdev); + + /* current state still has to be D_FAILED, + * there is only one way out: to D_DISKLESS, + * and that may only happen after our put_ldev below. */ + if (mdev->state.disk != D_FAILED) + dev_err(DEV, + "ASSERT FAILED: disk is %s during detach\n", + drbd_disk_str(mdev->state.disk)); + + if (ns.conn >= C_CONNECTED) + drbd_send_state(mdev, ns); + + drbd_rs_cancel_all(mdev); + + /* In case we want to get something to stable storage still, + * this may be the last chance. + * Following put_ldev may transition to D_DISKLESS. */ + drbd_md_sync(mdev); + } + put_ldev(mdev); + } + + /* second half of local IO error, failure to attach, + * or administrative detach, + * after local_cnt references have reached zero again */ + if (os.disk != D_DISKLESS && ns.disk == D_DISKLESS) { + /* We must still be diskless, + * re-attach has to be serialized with this! */ + if (mdev->state.disk != D_DISKLESS) + dev_err(DEV, + "ASSERT FAILED: disk is %s while going diskless\n", + drbd_disk_str(mdev->state.disk)); + + if (ns.conn >= C_CONNECTED) + drbd_send_state(mdev, ns); + /* corresponding get_ldev in __drbd_set_state + * this may finally trigger drbd_ldev_destroy. */ + put_ldev(mdev); + } + + /* Notify peer that I had a local IO error, and did not detached.. */ + if (os.disk == D_UP_TO_DATE && ns.disk == D_INCONSISTENT && ns.conn >= C_CONNECTED) + drbd_send_state(mdev, ns); + + /* Disks got bigger while they were detached */ + if (ns.disk > D_NEGOTIATING && ns.pdsk > D_NEGOTIATING && + test_and_clear_bit(RESYNC_AFTER_NEG, &mdev->flags)) { + if (ns.conn == C_CONNECTED) + resync_after_online_grow(mdev); + } + + /* A resync finished or aborted, wake paused devices... */ + if ((os.conn > C_CONNECTED && ns.conn <= C_CONNECTED) || + (os.peer_isp && !ns.peer_isp) || + (os.user_isp && !ns.user_isp)) + resume_next_sg(mdev); + + /* sync target done with resync. Explicitly notify peer, even though + * it should (at least for non-empty resyncs) already know itself. */ + if (os.disk < D_UP_TO_DATE && os.conn >= C_SYNC_SOURCE && ns.conn == C_CONNECTED) + drbd_send_state(mdev, ns); + + /* Verify finished, or reached stop sector. Peer did not know about + * the stop sector, and we may even have changed the stop sector during + * verify to interrupt/stop early. Send the new state. */ + if (os.conn == C_VERIFY_S && ns.conn == C_CONNECTED + && verify_can_do_stop_sector(mdev)) + drbd_send_state(mdev, ns); + + /* This triggers bitmap writeout of potentially still unwritten pages + * if the resync finished cleanly, or aborted because of peer disk + * failure, or because of connection loss. + * For resync aborted because of local disk failure, we cannot do + * any bitmap writeout anymore. + * No harm done if some bits change during this phase. + */ + if (os.conn > C_CONNECTED && ns.conn <= C_CONNECTED && get_ldev(mdev)) { + drbd_queue_bitmap_io(mdev, &drbd_bm_write_copy_pages, NULL, + "write from resync_finished", BM_LOCKED_CHANGE_ALLOWED); + put_ldev(mdev); + } + + if (ns.disk == D_DISKLESS && + ns.conn == C_STANDALONE && + ns.role == R_SECONDARY) { + if (os.aftr_isp != ns.aftr_isp) + resume_next_sg(mdev); + } + + drbd_md_sync(mdev); +} + +struct after_conn_state_chg_work { + struct drbd_work w; + enum drbd_conns oc; + union drbd_state ns_min; + union drbd_state ns_max; /* new, max state, over all mdevs */ + enum chg_state_flags flags; +}; + +static int w_after_conn_state_ch(struct drbd_work *w, int unused) +{ + struct after_conn_state_chg_work *acscw = + container_of(w, struct after_conn_state_chg_work, w); + struct drbd_tconn *tconn = w->tconn; + enum drbd_conns oc = acscw->oc; + union drbd_state ns_max = acscw->ns_max; + struct drbd_conf *mdev; + int vnr; + + kfree(acscw); + + /* Upon network configuration, we need to start the receiver */ + if (oc == C_STANDALONE && ns_max.conn == C_UNCONNECTED) + drbd_thread_start(&tconn->receiver); + + if (oc == C_DISCONNECTING && ns_max.conn == C_STANDALONE) { + struct net_conf *old_conf; + + mutex_lock(&tconn->conf_update); + old_conf = tconn->net_conf; + tconn->my_addr_len = 0; + tconn->peer_addr_len = 0; + rcu_assign_pointer(tconn->net_conf, NULL); + conn_free_crypto(tconn); + mutex_unlock(&tconn->conf_update); + + synchronize_rcu(); + kfree(old_conf); + } + + if (ns_max.susp_fen) { + /* case1: The outdate peer handler is successful: */ + if (ns_max.pdsk <= D_OUTDATED) { + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + if (test_bit(NEW_CUR_UUID, &mdev->flags)) { + drbd_uuid_new_current(mdev); + clear_bit(NEW_CUR_UUID, &mdev->flags); + } + } + rcu_read_unlock(); + spin_lock_irq(&tconn->req_lock); + _tl_restart(tconn, CONNECTION_LOST_WHILE_PENDING); + _conn_request_state(tconn, + (union drbd_state) { { .susp_fen = 1 } }, + (union drbd_state) { { .susp_fen = 0 } }, + CS_VERBOSE); + spin_unlock_irq(&tconn->req_lock); + } + } + kref_put(&tconn->kref, &conn_destroy); + + conn_md_sync(tconn); + + return 0; +} + +void conn_old_common_state(struct drbd_tconn *tconn, union drbd_state *pcs, enum chg_state_flags *pf) +{ + enum chg_state_flags flags = ~0; + struct drbd_conf *mdev; + int vnr, first_vol = 1; + union drbd_dev_state os, cs = { + { .role = R_SECONDARY, + .peer = R_UNKNOWN, + .conn = tconn->cstate, + .disk = D_DISKLESS, + .pdsk = D_UNKNOWN, + } }; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + os = mdev->state; + + if (first_vol) { + cs = os; + first_vol = 0; + continue; + } + + if (cs.role != os.role) + flags &= ~CS_DC_ROLE; + + if (cs.peer != os.peer) + flags &= ~CS_DC_PEER; + + if (cs.conn != os.conn) + flags &= ~CS_DC_CONN; + + if (cs.disk != os.disk) + flags &= ~CS_DC_DISK; + + if (cs.pdsk != os.pdsk) + flags &= ~CS_DC_PDSK; + } + rcu_read_unlock(); + + *pf |= CS_DC_MASK; + *pf &= flags; + (*pcs).i = cs.i; +} + +static enum drbd_state_rv +conn_is_valid_transition(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val, + enum chg_state_flags flags) +{ + enum drbd_state_rv rv = SS_SUCCESS; + union drbd_state ns, os; + struct drbd_conf *mdev; + int vnr; + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + os = drbd_read_state(mdev); + ns = sanitize_state(mdev, apply_mask_val(os, mask, val), NULL); + + if (flags & CS_IGN_OUTD_FAIL && ns.disk == D_OUTDATED && os.disk < D_OUTDATED) + ns.disk = os.disk; + + if (ns.i == os.i) + continue; + + rv = is_valid_transition(os, ns); + if (rv < SS_SUCCESS) + break; + + if (!(flags & CS_HARD)) { + rv = is_valid_state(mdev, ns); + if (rv < SS_SUCCESS) { + if (is_valid_state(mdev, os) == rv) + rv = is_valid_soft_transition(os, ns, tconn); + } else + rv = is_valid_soft_transition(os, ns, tconn); + } + if (rv < SS_SUCCESS) + break; + } + rcu_read_unlock(); + + if (rv < SS_SUCCESS && flags & CS_VERBOSE) + print_st_err(mdev, os, ns, rv); + + return rv; +} + +void +conn_set_state(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val, + union drbd_state *pns_min, union drbd_state *pns_max, enum chg_state_flags flags) +{ + union drbd_state ns, os, ns_max = { }; + union drbd_state ns_min = { + { .role = R_MASK, + .peer = R_MASK, + .conn = val.conn, + .disk = D_MASK, + .pdsk = D_MASK + } }; + struct drbd_conf *mdev; + enum drbd_state_rv rv; + int vnr, number_of_volumes = 0; + + if (mask.conn == C_MASK) { + /* remember last connect time so request_timer_fn() won't + * kill newly established sessions while we are still trying to thaw + * previously frozen IO */ + if (tconn->cstate != C_WF_REPORT_PARAMS && val.conn == C_WF_REPORT_PARAMS) + tconn->last_reconnect_jif = jiffies; + + tconn->cstate = val.conn; + } + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + number_of_volumes++; + os = drbd_read_state(mdev); + ns = apply_mask_val(os, mask, val); + ns = sanitize_state(mdev, ns, NULL); + + if (flags & CS_IGN_OUTD_FAIL && ns.disk == D_OUTDATED && os.disk < D_OUTDATED) + ns.disk = os.disk; + + rv = __drbd_set_state(mdev, ns, flags, NULL); + if (rv < SS_SUCCESS) + BUG(); + + ns.i = mdev->state.i; + ns_max.role = max_role(ns.role, ns_max.role); + ns_max.peer = max_role(ns.peer, ns_max.peer); + ns_max.conn = max_t(enum drbd_conns, ns.conn, ns_max.conn); + ns_max.disk = max_t(enum drbd_disk_state, ns.disk, ns_max.disk); + ns_max.pdsk = max_t(enum drbd_disk_state, ns.pdsk, ns_max.pdsk); + + ns_min.role = min_role(ns.role, ns_min.role); + ns_min.peer = min_role(ns.peer, ns_min.peer); + ns_min.conn = min_t(enum drbd_conns, ns.conn, ns_min.conn); + ns_min.disk = min_t(enum drbd_disk_state, ns.disk, ns_min.disk); + ns_min.pdsk = min_t(enum drbd_disk_state, ns.pdsk, ns_min.pdsk); + } + rcu_read_unlock(); + + if (number_of_volumes == 0) { + ns_min = ns_max = (union drbd_state) { { + .role = R_SECONDARY, + .peer = R_UNKNOWN, + .conn = val.conn, + .disk = D_DISKLESS, + .pdsk = D_UNKNOWN + } }; + } + + ns_min.susp = ns_max.susp = tconn->susp; + ns_min.susp_nod = ns_max.susp_nod = tconn->susp_nod; + ns_min.susp_fen = ns_max.susp_fen = tconn->susp_fen; + + *pns_min = ns_min; + *pns_max = ns_max; +} + +static enum drbd_state_rv +_conn_rq_cond(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val) +{ + enum drbd_state_rv rv; + + if (test_and_clear_bit(CONN_WD_ST_CHG_OKAY, &tconn->flags)) + return SS_CW_SUCCESS; + + if (test_and_clear_bit(CONN_WD_ST_CHG_FAIL, &tconn->flags)) + return SS_CW_FAILED_BY_PEER; + + rv = tconn->cstate != C_WF_REPORT_PARAMS ? SS_CW_NO_NEED : SS_UNKNOWN_ERROR; + + if (rv == SS_UNKNOWN_ERROR) + rv = conn_is_valid_transition(tconn, mask, val, 0); + + if (rv == SS_SUCCESS) + rv = SS_UNKNOWN_ERROR; /* cont waiting, otherwise fail. */ + + return rv; +} + +enum drbd_state_rv +_conn_request_state(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val, + enum chg_state_flags flags) +{ + enum drbd_state_rv rv = SS_SUCCESS; + struct after_conn_state_chg_work *acscw; + enum drbd_conns oc = tconn->cstate; + union drbd_state ns_max, ns_min, os; + bool have_mutex = false; + + if (mask.conn) { + rv = is_valid_conn_transition(oc, val.conn); + if (rv < SS_SUCCESS) + goto abort; + } + + rv = conn_is_valid_transition(tconn, mask, val, flags); + if (rv < SS_SUCCESS) + goto abort; + + if (oc == C_WF_REPORT_PARAMS && val.conn == C_DISCONNECTING && + !(flags & (CS_LOCAL_ONLY | CS_HARD))) { + + /* This will be a cluster-wide state change. + * Need to give up the spinlock, grab the mutex, + * then send the state change request, ... */ + spin_unlock_irq(&tconn->req_lock); + mutex_lock(&tconn->cstate_mutex); + have_mutex = true; + + set_bit(CONN_WD_ST_CHG_REQ, &tconn->flags); + if (conn_send_state_req(tconn, mask, val)) { + /* sending failed. */ + clear_bit(CONN_WD_ST_CHG_REQ, &tconn->flags); + rv = SS_CW_FAILED_BY_PEER; + /* need to re-aquire the spin lock, though */ + goto abort_unlocked; + } + + if (val.conn == C_DISCONNECTING) + set_bit(DISCONNECT_SENT, &tconn->flags); + + /* ... and re-aquire the spinlock. + * If _conn_rq_cond() returned >= SS_SUCCESS, we must call + * conn_set_state() within the same spinlock. */ + spin_lock_irq(&tconn->req_lock); + wait_event_lock_irq(tconn->ping_wait, + (rv = _conn_rq_cond(tconn, mask, val)), + tconn->req_lock); + clear_bit(CONN_WD_ST_CHG_REQ, &tconn->flags); + if (rv < SS_SUCCESS) + goto abort; + } + + conn_old_common_state(tconn, &os, &flags); + flags |= CS_DC_SUSP; + conn_set_state(tconn, mask, val, &ns_min, &ns_max, flags); + conn_pr_state_change(tconn, os, ns_max, flags); + + acscw = kmalloc(sizeof(*acscw), GFP_ATOMIC); + if (acscw) { + acscw->oc = os.conn; + acscw->ns_min = ns_min; + acscw->ns_max = ns_max; + acscw->flags = flags; + acscw->w.cb = w_after_conn_state_ch; + kref_get(&tconn->kref); + acscw->w.tconn = tconn; + drbd_queue_work(&tconn->sender_work, &acscw->w); + } else { + conn_err(tconn, "Could not kmalloc an acscw\n"); + } + + abort: + if (have_mutex) { + /* mutex_unlock() "... must not be used in interrupt context.", + * so give up the spinlock, then re-aquire it */ + spin_unlock_irq(&tconn->req_lock); + abort_unlocked: + mutex_unlock(&tconn->cstate_mutex); + spin_lock_irq(&tconn->req_lock); + } + if (rv < SS_SUCCESS && flags & CS_VERBOSE) { + conn_err(tconn, "State change failed: %s\n", drbd_set_st_err_str(rv)); + conn_err(tconn, " mask = 0x%x val = 0x%x\n", mask.i, val.i); + conn_err(tconn, " old_conn:%s wanted_conn:%s\n", drbd_conn_str(oc), drbd_conn_str(val.conn)); + } + return rv; +} + +enum drbd_state_rv +conn_request_state(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val, + enum chg_state_flags flags) +{ + enum drbd_state_rv rv; + + spin_lock_irq(&tconn->req_lock); + rv = _conn_request_state(tconn, mask, val, flags); + spin_unlock_irq(&tconn->req_lock); + + return rv; +} diff --git a/drivers/block/drbd/drbd_state.h b/drivers/block/drbd/drbd_state.h new file mode 100644 index 000000000000..a3c361bbc4b6 --- /dev/null +++ b/drivers/block/drbd/drbd_state.h @@ -0,0 +1,161 @@ +#ifndef DRBD_STATE_H +#define DRBD_STATE_H + +struct drbd_conf; +struct drbd_tconn; + +/** + * DOC: DRBD State macros + * + * These macros are used to express state changes in easily readable form. + * + * The NS macros expand to a mask and a value, that can be bit ored onto the + * current state as soon as the spinlock (req_lock) was taken. + * + * The _NS macros are used for state functions that get called with the + * spinlock. These macros expand directly to the new state value. + * + * Besides the basic forms NS() and _NS() additional _?NS[23] are defined + * to express state changes that affect more than one aspect of the state. + * + * E.g. NS2(conn, C_CONNECTED, peer, R_SECONDARY) + * Means that the network connection was established and that the peer + * is in secondary role. + */ +#define role_MASK R_MASK +#define peer_MASK R_MASK +#define disk_MASK D_MASK +#define pdsk_MASK D_MASK +#define conn_MASK C_MASK +#define susp_MASK 1 +#define user_isp_MASK 1 +#define aftr_isp_MASK 1 +#define susp_nod_MASK 1 +#define susp_fen_MASK 1 + +#define NS(T, S) \ + ({ union drbd_state mask; mask.i = 0; mask.T = T##_MASK; mask; }), \ + ({ union drbd_state val; val.i = 0; val.T = (S); val; }) +#define NS2(T1, S1, T2, S2) \ + ({ union drbd_state mask; mask.i = 0; mask.T1 = T1##_MASK; \ + mask.T2 = T2##_MASK; mask; }), \ + ({ union drbd_state val; val.i = 0; val.T1 = (S1); \ + val.T2 = (S2); val; }) +#define NS3(T1, S1, T2, S2, T3, S3) \ + ({ union drbd_state mask; mask.i = 0; mask.T1 = T1##_MASK; \ + mask.T2 = T2##_MASK; mask.T3 = T3##_MASK; mask; }), \ + ({ union drbd_state val; val.i = 0; val.T1 = (S1); \ + val.T2 = (S2); val.T3 = (S3); val; }) + +#define _NS(D, T, S) \ + D, ({ union drbd_state __ns; __ns = drbd_read_state(D); __ns.T = (S); __ns; }) +#define _NS2(D, T1, S1, T2, S2) \ + D, ({ union drbd_state __ns; __ns = drbd_read_state(D); __ns.T1 = (S1); \ + __ns.T2 = (S2); __ns; }) +#define _NS3(D, T1, S1, T2, S2, T3, S3) \ + D, ({ union drbd_state __ns; __ns = drbd_read_state(D); __ns.T1 = (S1); \ + __ns.T2 = (S2); __ns.T3 = (S3); __ns; }) + +enum chg_state_flags { + CS_HARD = 1 << 0, + CS_VERBOSE = 1 << 1, + CS_WAIT_COMPLETE = 1 << 2, + CS_SERIALIZE = 1 << 3, + CS_ORDERED = CS_WAIT_COMPLETE + CS_SERIALIZE, + CS_LOCAL_ONLY = 1 << 4, /* Do not consider a device pair wide state change */ + CS_DC_ROLE = 1 << 5, /* DC = display as connection state change */ + CS_DC_PEER = 1 << 6, + CS_DC_CONN = 1 << 7, + CS_DC_DISK = 1 << 8, + CS_DC_PDSK = 1 << 9, + CS_DC_SUSP = 1 << 10, + CS_DC_MASK = CS_DC_ROLE + CS_DC_PEER + CS_DC_CONN + CS_DC_DISK + CS_DC_PDSK, + CS_IGN_OUTD_FAIL = 1 << 11, +}; + +/* drbd_dev_state and drbd_state are different types. This is to stress the + small difference. There is no suspended flag (.susp), and no suspended + while fence handler runs flas (susp_fen). */ +union drbd_dev_state { + struct { +#if defined(__LITTLE_ENDIAN_BITFIELD) + unsigned role:2 ; /* 3/4 primary/secondary/unknown */ + unsigned peer:2 ; /* 3/4 primary/secondary/unknown */ + unsigned conn:5 ; /* 17/32 cstates */ + unsigned disk:4 ; /* 8/16 from D_DISKLESS to D_UP_TO_DATE */ + unsigned pdsk:4 ; /* 8/16 from D_DISKLESS to D_UP_TO_DATE */ + unsigned _unused:1 ; + unsigned aftr_isp:1 ; /* isp .. imposed sync pause */ + unsigned peer_isp:1 ; + unsigned user_isp:1 ; + unsigned _pad:11; /* 0 unused */ +#elif defined(__BIG_ENDIAN_BITFIELD) + unsigned _pad:11; + unsigned user_isp:1 ; + unsigned peer_isp:1 ; + unsigned aftr_isp:1 ; /* isp .. imposed sync pause */ + unsigned _unused:1 ; + unsigned pdsk:4 ; /* 8/16 from D_DISKLESS to D_UP_TO_DATE */ + unsigned disk:4 ; /* 8/16 from D_DISKLESS to D_UP_TO_DATE */ + unsigned conn:5 ; /* 17/32 cstates */ + unsigned peer:2 ; /* 3/4 primary/secondary/unknown */ + unsigned role:2 ; /* 3/4 primary/secondary/unknown */ +#else +# error "this endianess is not supported" +#endif + }; + unsigned int i; +}; + +extern enum drbd_state_rv drbd_change_state(struct drbd_conf *mdev, + enum chg_state_flags f, + union drbd_state mask, + union drbd_state val); +extern void drbd_force_state(struct drbd_conf *, union drbd_state, + union drbd_state); +extern enum drbd_state_rv _drbd_request_state(struct drbd_conf *, + union drbd_state, + union drbd_state, + enum chg_state_flags); +extern enum drbd_state_rv __drbd_set_state(struct drbd_conf *, union drbd_state, + enum chg_state_flags, + struct completion *done); +extern void print_st_err(struct drbd_conf *, union drbd_state, + union drbd_state, int); + +enum drbd_state_rv +_conn_request_state(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val, + enum chg_state_flags flags); + +enum drbd_state_rv +conn_request_state(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val, + enum chg_state_flags flags); + +extern void drbd_resume_al(struct drbd_conf *mdev); +extern bool conn_all_vols_unconf(struct drbd_tconn *tconn); + +/** + * drbd_request_state() - Reqest a state change + * @mdev: DRBD device. + * @mask: mask of state bits to change. + * @val: value of new state bits. + * + * This is the most graceful way of requesting a state change. It is verbose + * quite verbose in case the state change is not possible, and all those + * state changes are globally serialized. + */ +static inline int drbd_request_state(struct drbd_conf *mdev, + union drbd_state mask, + union drbd_state val) +{ + return _drbd_request_state(mdev, mask, val, CS_VERBOSE + CS_ORDERED); +} + +enum drbd_role conn_highest_role(struct drbd_tconn *tconn); +enum drbd_role conn_highest_peer(struct drbd_tconn *tconn); +enum drbd_disk_state conn_highest_disk(struct drbd_tconn *tconn); +enum drbd_disk_state conn_lowest_disk(struct drbd_tconn *tconn); +enum drbd_disk_state conn_highest_pdsk(struct drbd_tconn *tconn); +enum drbd_conns conn_lowest_conn(struct drbd_tconn *tconn); + +#endif diff --git a/drivers/block/drbd/drbd_strings.c b/drivers/block/drbd/drbd_strings.c index c44a2a602772..9a664bd27404 100644 --- a/drivers/block/drbd/drbd_strings.c +++ b/drivers/block/drbd/drbd_strings.c @@ -89,6 +89,7 @@ static const char *drbd_state_sw_errors[] = { [-SS_LOWER_THAN_OUTDATED] = "Disk state is lower than outdated", [-SS_IN_TRANSIENT_STATE] = "In transient state, retry after next state change", [-SS_CONCURRENT_ST_CHG] = "Concurrent state changes detected and aborted", + [-SS_O_VOL_PEER_PRI] = "Other vol primary on peer not allowed by config", }; const char *drbd_conn_str(enum drbd_conns s) diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c index 6bce2cc179d4..424dc7bdf9b7 100644 --- a/drivers/block/drbd/drbd_worker.c +++ b/drivers/block/drbd/drbd_worker.c @@ -38,16 +38,13 @@ #include "drbd_int.h" #include "drbd_req.h" -static int w_make_ov_request(struct drbd_conf *mdev, struct drbd_work *w, int cancel); -static int w_make_resync_request(struct drbd_conf *mdev, - struct drbd_work *w, int cancel); - +static int w_make_ov_request(struct drbd_work *w, int cancel); /* endio handlers: * drbd_md_io_complete (defined here) - * drbd_endio_pri (defined here) - * drbd_endio_sec (defined here) + * drbd_request_endio (defined here) + * drbd_peer_request_endio (defined here) * bm_async_io_complete (defined in drbd_bitmap.c) * * For all these callbacks, note the following: @@ -60,7 +57,7 @@ static int w_make_resync_request(struct drbd_conf *mdev, /* About the global_state_lock Each state transition on an device holds a read lock. In case we have - to evaluate the sync after dependencies, we grab a write lock, because + to evaluate the resync after dependencies, we grab a write lock, because we need stable states on all devices for that. */ rwlock_t global_state_lock; @@ -98,97 +95,93 @@ void drbd_md_io_complete(struct bio *bio, int error) /* reads on behalf of the partner, * "submitted" by the receiver */ -void drbd_endio_read_sec_final(struct drbd_epoch_entry *e) __releases(local) +void drbd_endio_read_sec_final(struct drbd_peer_request *peer_req) __releases(local) { unsigned long flags = 0; - struct drbd_conf *mdev = e->mdev; - - D_ASSERT(e->block_id != ID_VACANT); + struct drbd_conf *mdev = peer_req->w.mdev; - spin_lock_irqsave(&mdev->req_lock, flags); - mdev->read_cnt += e->size >> 9; - list_del(&e->w.list); + spin_lock_irqsave(&mdev->tconn->req_lock, flags); + mdev->read_cnt += peer_req->i.size >> 9; + list_del(&peer_req->w.list); if (list_empty(&mdev->read_ee)) wake_up(&mdev->ee_wait); - if (test_bit(__EE_WAS_ERROR, &e->flags)) - __drbd_chk_io_error(mdev, DRBD_IO_ERROR); - spin_unlock_irqrestore(&mdev->req_lock, flags); + if (test_bit(__EE_WAS_ERROR, &peer_req->flags)) + __drbd_chk_io_error(mdev, DRBD_READ_ERROR); + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); - drbd_queue_work(&mdev->data.work, &e->w); + drbd_queue_work(&mdev->tconn->sender_work, &peer_req->w); put_ldev(mdev); } /* writes on behalf of the partner, or resync writes, * "submitted" by the receiver, final stage. */ -static void drbd_endio_write_sec_final(struct drbd_epoch_entry *e) __releases(local) +static void drbd_endio_write_sec_final(struct drbd_peer_request *peer_req) __releases(local) { unsigned long flags = 0; - struct drbd_conf *mdev = e->mdev; - sector_t e_sector; + struct drbd_conf *mdev = peer_req->w.mdev; + struct drbd_interval i; int do_wake; - int is_syncer_req; + u64 block_id; int do_al_complete_io; - D_ASSERT(e->block_id != ID_VACANT); - - /* after we moved e to done_ee, + /* after we moved peer_req to done_ee, * we may no longer access it, * it may be freed/reused already! * (as soon as we release the req_lock) */ - e_sector = e->sector; - do_al_complete_io = e->flags & EE_CALL_AL_COMPLETE_IO; - is_syncer_req = is_syncer_block_id(e->block_id); + i = peer_req->i; + do_al_complete_io = peer_req->flags & EE_CALL_AL_COMPLETE_IO; + block_id = peer_req->block_id; - spin_lock_irqsave(&mdev->req_lock, flags); - mdev->writ_cnt += e->size >> 9; - list_del(&e->w.list); /* has been on active_ee or sync_ee */ - list_add_tail(&e->w.list, &mdev->done_ee); + spin_lock_irqsave(&mdev->tconn->req_lock, flags); + mdev->writ_cnt += peer_req->i.size >> 9; + list_move_tail(&peer_req->w.list, &mdev->done_ee); - /* No hlist_del_init(&e->collision) here, we did not send the Ack yet, - * neither did we wake possibly waiting conflicting requests. - * done from "drbd_process_done_ee" within the appropriate w.cb - * (e_end_block/e_end_resync_block) or from _drbd_clear_done_ee */ + /* + * Do not remove from the write_requests tree here: we did not send the + * Ack yet and did not wake possibly waiting conflicting requests. + * Removed from the tree from "drbd_process_done_ee" within the + * appropriate w.cb (e_end_block/e_end_resync_block) or from + * _drbd_clear_done_ee. + */ - do_wake = is_syncer_req - ? list_empty(&mdev->sync_ee) - : list_empty(&mdev->active_ee); + do_wake = list_empty(block_id == ID_SYNCER ? &mdev->sync_ee : &mdev->active_ee); - if (test_bit(__EE_WAS_ERROR, &e->flags)) - __drbd_chk_io_error(mdev, DRBD_IO_ERROR); - spin_unlock_irqrestore(&mdev->req_lock, flags); + if (test_bit(__EE_WAS_ERROR, &peer_req->flags)) + __drbd_chk_io_error(mdev, DRBD_WRITE_ERROR); + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); - if (is_syncer_req) - drbd_rs_complete_io(mdev, e_sector); + if (block_id == ID_SYNCER) + drbd_rs_complete_io(mdev, i.sector); if (do_wake) wake_up(&mdev->ee_wait); if (do_al_complete_io) - drbd_al_complete_io(mdev, e_sector); + drbd_al_complete_io(mdev, &i); - wake_asender(mdev); + wake_asender(mdev->tconn); put_ldev(mdev); } /* writes on behalf of the partner, or resync writes, * "submitted" by the receiver. */ -void drbd_endio_sec(struct bio *bio, int error) +void drbd_peer_request_endio(struct bio *bio, int error) { - struct drbd_epoch_entry *e = bio->bi_private; - struct drbd_conf *mdev = e->mdev; + struct drbd_peer_request *peer_req = bio->bi_private; + struct drbd_conf *mdev = peer_req->w.mdev; int uptodate = bio_flagged(bio, BIO_UPTODATE); int is_write = bio_data_dir(bio) == WRITE; if (error && __ratelimit(&drbd_ratelimit_state)) dev_warn(DEV, "%s: error=%d s=%llus\n", is_write ? "write" : "read", error, - (unsigned long long)e->sector); + (unsigned long long)peer_req->i.sector); if (!error && !uptodate) { if (__ratelimit(&drbd_ratelimit_state)) dev_warn(DEV, "%s: setting error to -EIO s=%llus\n", is_write ? "write" : "read", - (unsigned long long)e->sector); + (unsigned long long)peer_req->i.sector); /* strange behavior of some lower level drivers... * fail the request by clearing the uptodate flag, * but do not return any error?! */ @@ -196,24 +189,24 @@ void drbd_endio_sec(struct bio *bio, int error) } if (error) - set_bit(__EE_WAS_ERROR, &e->flags); + set_bit(__EE_WAS_ERROR, &peer_req->flags); bio_put(bio); /* no need for the bio anymore */ - if (atomic_dec_and_test(&e->pending_bios)) { + if (atomic_dec_and_test(&peer_req->pending_bios)) { if (is_write) - drbd_endio_write_sec_final(e); + drbd_endio_write_sec_final(peer_req); else - drbd_endio_read_sec_final(e); + drbd_endio_read_sec_final(peer_req); } } /* read, readA or write requests on R_PRIMARY coming from drbd_make_request */ -void drbd_endio_pri(struct bio *bio, int error) +void drbd_request_endio(struct bio *bio, int error) { unsigned long flags; struct drbd_request *req = bio->bi_private; - struct drbd_conf *mdev = req->mdev; + struct drbd_conf *mdev = req->w.mdev; struct bio_and_error m; enum drbd_req_event what; int uptodate = bio_flagged(bio, BIO_UPTODATE); @@ -227,53 +220,72 @@ void drbd_endio_pri(struct bio *bio, int error) error = -EIO; } + + /* If this request was aborted locally before, + * but now was completed "successfully", + * chances are that this caused arbitrary data corruption. + * + * "aborting" requests, or force-detaching the disk, is intended for + * completely blocked/hung local backing devices which do no longer + * complete requests at all, not even do error completions. In this + * situation, usually a hard-reset and failover is the only way out. + * + * By "aborting", basically faking a local error-completion, + * we allow for a more graceful swichover by cleanly migrating services. + * Still the affected node has to be rebooted "soon". + * + * By completing these requests, we allow the upper layers to re-use + * the associated data pages. + * + * If later the local backing device "recovers", and now DMAs some data + * from disk into the original request pages, in the best case it will + * just put random data into unused pages; but typically it will corrupt + * meanwhile completely unrelated data, causing all sorts of damage. + * + * Which means delayed successful completion, + * especially for READ requests, + * is a reason to panic(). + * + * We assume that a delayed *error* completion is OK, + * though we still will complain noisily about it. + */ + if (unlikely(req->rq_state & RQ_LOCAL_ABORTED)) { + if (__ratelimit(&drbd_ratelimit_state)) + dev_emerg(DEV, "delayed completion of aborted local request; disk-timeout may be too aggressive\n"); + + if (!error) + panic("possible random memory corruption caused by delayed completion of aborted local request\n"); + } + /* to avoid recursion in __req_mod */ if (unlikely(error)) { what = (bio_data_dir(bio) == WRITE) - ? write_completed_with_error + ? WRITE_COMPLETED_WITH_ERROR : (bio_rw(bio) == READ) - ? read_completed_with_error - : read_ahead_completed_with_error; + ? READ_COMPLETED_WITH_ERROR + : READ_AHEAD_COMPLETED_WITH_ERROR; } else - what = completed_ok; + what = COMPLETED_OK; bio_put(req->private_bio); req->private_bio = ERR_PTR(error); /* not req_mod(), we need irqsave here! */ - spin_lock_irqsave(&mdev->req_lock, flags); + spin_lock_irqsave(&mdev->tconn->req_lock, flags); __req_mod(req, what, &m); - spin_unlock_irqrestore(&mdev->req_lock, flags); + spin_unlock_irqrestore(&mdev->tconn->req_lock, flags); put_ldev(mdev); if (m.bio) complete_master_bio(mdev, &m); } -int w_read_retry_remote(struct drbd_conf *mdev, struct drbd_work *w, int cancel) -{ - struct drbd_request *req = container_of(w, struct drbd_request, w); - - /* We should not detach for read io-error, - * but try to WRITE the P_DATA_REPLY to the failed location, - * to give the disk the chance to relocate that block */ - - spin_lock_irq(&mdev->req_lock); - if (cancel || mdev->state.pdsk != D_UP_TO_DATE) { - _req_mod(req, read_retry_remote_canceled); - spin_unlock_irq(&mdev->req_lock); - return 1; - } - spin_unlock_irq(&mdev->req_lock); - - return w_send_read_req(mdev, w, 0); -} - -void drbd_csum_ee(struct drbd_conf *mdev, struct crypto_hash *tfm, struct drbd_epoch_entry *e, void *digest) +void drbd_csum_ee(struct drbd_conf *mdev, struct crypto_hash *tfm, + struct drbd_peer_request *peer_req, void *digest) { struct hash_desc desc; struct scatterlist sg; - struct page *page = e->pages; + struct page *page = peer_req->pages; struct page *tmp; unsigned len; @@ -290,7 +302,7 @@ void drbd_csum_ee(struct drbd_conf *mdev, struct crypto_hash *tfm, struct drbd_e page = tmp; } /* and now the last, possibly only partially used page */ - len = e->size & (PAGE_SIZE - 1); + len = peer_req->i.size & (PAGE_SIZE - 1); sg_set_page(&sg, page, len ?: PAGE_SIZE, 0); crypto_hash_update(&desc, &sg, sg.length); crypto_hash_final(&desc, digest); @@ -316,59 +328,58 @@ void drbd_csum_bio(struct drbd_conf *mdev, struct crypto_hash *tfm, struct bio * crypto_hash_final(&desc, digest); } -/* TODO merge common code with w_e_end_ov_req */ -int w_e_send_csum(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +/* MAYBE merge common code with w_e_end_ov_req */ +static int w_e_send_csum(struct drbd_work *w, int cancel) { - struct drbd_epoch_entry *e = container_of(w, struct drbd_epoch_entry, w); + struct drbd_peer_request *peer_req = container_of(w, struct drbd_peer_request, w); + struct drbd_conf *mdev = w->mdev; int digest_size; void *digest; - int ok = 1; - - D_ASSERT(e->block_id == DRBD_MAGIC + 0xbeef); + int err = 0; if (unlikely(cancel)) goto out; - if (likely((e->flags & EE_WAS_ERROR) != 0)) + if (unlikely((peer_req->flags & EE_WAS_ERROR) != 0)) goto out; - digest_size = crypto_hash_digestsize(mdev->csums_tfm); + digest_size = crypto_hash_digestsize(mdev->tconn->csums_tfm); digest = kmalloc(digest_size, GFP_NOIO); if (digest) { - sector_t sector = e->sector; - unsigned int size = e->size; - drbd_csum_ee(mdev, mdev->csums_tfm, e, digest); - /* Free e and pages before send. + sector_t sector = peer_req->i.sector; + unsigned int size = peer_req->i.size; + drbd_csum_ee(mdev, mdev->tconn->csums_tfm, peer_req, digest); + /* Free peer_req and pages before send. * In case we block on congestion, we could otherwise run into * some distributed deadlock, if the other side blocks on * congestion as well, because our receiver blocks in - * drbd_pp_alloc due to pp_in_use > max_buffers. */ - drbd_free_ee(mdev, e); - e = NULL; + * drbd_alloc_pages due to pp_in_use > max_buffers. */ + drbd_free_peer_req(mdev, peer_req); + peer_req = NULL; inc_rs_pending(mdev); - ok = drbd_send_drequest_csum(mdev, sector, size, - digest, digest_size, - P_CSUM_RS_REQUEST); + err = drbd_send_drequest_csum(mdev, sector, size, + digest, digest_size, + P_CSUM_RS_REQUEST); kfree(digest); } else { dev_err(DEV, "kmalloc() of digest failed.\n"); - ok = 0; + err = -ENOMEM; } out: - if (e) - drbd_free_ee(mdev, e); + if (peer_req) + drbd_free_peer_req(mdev, peer_req); - if (unlikely(!ok)) + if (unlikely(err)) dev_err(DEV, "drbd_send_drequest(..., csum) failed\n"); - return ok; + return err; } #define GFP_TRY (__GFP_HIGHMEM | __GFP_NOWARN) static int read_for_csum(struct drbd_conf *mdev, sector_t sector, int size) { - struct drbd_epoch_entry *e; + struct drbd_peer_request *peer_req; if (!get_ldev(mdev)) return -EIO; @@ -378,45 +389,47 @@ static int read_for_csum(struct drbd_conf *mdev, sector_t sector, int size) /* GFP_TRY, because if there is no memory available right now, this may * be rescheduled for later. It is "only" background resync, after all. */ - e = drbd_alloc_ee(mdev, DRBD_MAGIC+0xbeef, sector, size, GFP_TRY); - if (!e) + peer_req = drbd_alloc_peer_req(mdev, ID_SYNCER /* unused */, sector, + size, GFP_TRY); + if (!peer_req) goto defer; - e->w.cb = w_e_send_csum; - spin_lock_irq(&mdev->req_lock); - list_add(&e->w.list, &mdev->read_ee); - spin_unlock_irq(&mdev->req_lock); + peer_req->w.cb = w_e_send_csum; + spin_lock_irq(&mdev->tconn->req_lock); + list_add(&peer_req->w.list, &mdev->read_ee); + spin_unlock_irq(&mdev->tconn->req_lock); atomic_add(size >> 9, &mdev->rs_sect_ev); - if (drbd_submit_ee(mdev, e, READ, DRBD_FAULT_RS_RD) == 0) + if (drbd_submit_peer_request(mdev, peer_req, READ, DRBD_FAULT_RS_RD) == 0) return 0; /* If it failed because of ENOMEM, retry should help. If it failed * because bio_add_page failed (probably broken lower level driver), * retry may or may not help. * If it does not, you may need to force disconnect. */ - spin_lock_irq(&mdev->req_lock); - list_del(&e->w.list); - spin_unlock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); + list_del(&peer_req->w.list); + spin_unlock_irq(&mdev->tconn->req_lock); - drbd_free_ee(mdev, e); + drbd_free_peer_req(mdev, peer_req); defer: put_ldev(mdev); return -EAGAIN; } -int w_resync_timer(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_resync_timer(struct drbd_work *w, int cancel) { + struct drbd_conf *mdev = w->mdev; switch (mdev->state.conn) { case C_VERIFY_S: - w_make_ov_request(mdev, w, cancel); + w_make_ov_request(w, cancel); break; case C_SYNC_TARGET: - w_make_resync_request(mdev, w, cancel); + w_make_resync_request(w, cancel); break; } - return 1; + return 0; } void resync_timer_fn(unsigned long data) @@ -424,7 +437,7 @@ void resync_timer_fn(unsigned long data) struct drbd_conf *mdev = (struct drbd_conf *) data; if (list_empty(&mdev->resync_work.list)) - drbd_queue_work(&mdev->data.work, &mdev->resync_work); + drbd_queue_work(&mdev->tconn->sender_work, &mdev->resync_work); } static void fifo_set(struct fifo_buffer *fb, int value) @@ -456,8 +469,24 @@ static void fifo_add_val(struct fifo_buffer *fb, int value) fb->values[i] += value; } +struct fifo_buffer *fifo_alloc(int fifo_size) +{ + struct fifo_buffer *fb; + + fb = kzalloc(sizeof(struct fifo_buffer) + sizeof(int) * fifo_size, GFP_NOIO); + if (!fb) + return NULL; + + fb->head_index = 0; + fb->size = fifo_size; + fb->total = 0; + + return fb; +} + static int drbd_rs_controller(struct drbd_conf *mdev) { + struct disk_conf *dc; unsigned int sect_in; /* Number of sectors that came in since the last turn */ unsigned int want; /* The number of sectors we want in the proxy */ int req_sect; /* Number of sectors to request in this turn */ @@ -466,38 +495,39 @@ static int drbd_rs_controller(struct drbd_conf *mdev) int steps; /* Number of time steps to plan ahead */ int curr_corr; int max_sect; + struct fifo_buffer *plan; sect_in = atomic_xchg(&mdev->rs_sect_in, 0); /* Number of sectors that came in */ mdev->rs_in_flight -= sect_in; - spin_lock(&mdev->peer_seq_lock); /* get an atomic view on mdev->rs_plan_s */ + dc = rcu_dereference(mdev->ldev->disk_conf); + plan = rcu_dereference(mdev->rs_plan_s); - steps = mdev->rs_plan_s.size; /* (mdev->sync_conf.c_plan_ahead * 10 * SLEEP_TIME) / HZ; */ + steps = plan->size; /* (dc->c_plan_ahead * 10 * SLEEP_TIME) / HZ; */ if (mdev->rs_in_flight + sect_in == 0) { /* At start of resync */ - want = ((mdev->sync_conf.rate * 2 * SLEEP_TIME) / HZ) * steps; + want = ((dc->resync_rate * 2 * SLEEP_TIME) / HZ) * steps; } else { /* normal path */ - want = mdev->sync_conf.c_fill_target ? mdev->sync_conf.c_fill_target : - sect_in * mdev->sync_conf.c_delay_target * HZ / (SLEEP_TIME * 10); + want = dc->c_fill_target ? dc->c_fill_target : + sect_in * dc->c_delay_target * HZ / (SLEEP_TIME * 10); } - correction = want - mdev->rs_in_flight - mdev->rs_planed; + correction = want - mdev->rs_in_flight - plan->total; /* Plan ahead */ cps = correction / steps; - fifo_add_val(&mdev->rs_plan_s, cps); - mdev->rs_planed += cps * steps; + fifo_add_val(plan, cps); + plan->total += cps * steps; /* What we do in this step */ - curr_corr = fifo_push(&mdev->rs_plan_s, 0); - spin_unlock(&mdev->peer_seq_lock); - mdev->rs_planed -= curr_corr; + curr_corr = fifo_push(plan, 0); + plan->total -= curr_corr; req_sect = sect_in + curr_corr; if (req_sect < 0) req_sect = 0; - max_sect = (mdev->sync_conf.c_max_rate * 2 * SLEEP_TIME) / HZ; + max_sect = (dc->c_max_rate * 2 * SLEEP_TIME) / HZ; if (req_sect > max_sect) req_sect = max_sect; @@ -513,22 +543,25 @@ static int drbd_rs_controller(struct drbd_conf *mdev) static int drbd_rs_number_requests(struct drbd_conf *mdev) { int number; - if (mdev->rs_plan_s.size) { /* mdev->sync_conf.c_plan_ahead */ + + rcu_read_lock(); + if (rcu_dereference(mdev->rs_plan_s)->size) { number = drbd_rs_controller(mdev) >> (BM_BLOCK_SHIFT - 9); mdev->c_sync_rate = number * HZ * (BM_BLOCK_SIZE / 1024) / SLEEP_TIME; } else { - mdev->c_sync_rate = mdev->sync_conf.rate; + mdev->c_sync_rate = rcu_dereference(mdev->ldev->disk_conf)->resync_rate; number = SLEEP_TIME * mdev->c_sync_rate / ((BM_BLOCK_SIZE / 1024) * HZ); } + rcu_read_unlock(); /* ignore the amount of pending requests, the resync controller should * throttle down to incoming reply rate soon enough anyways. */ return number; } -static int w_make_resync_request(struct drbd_conf *mdev, - struct drbd_work *w, int cancel) +int w_make_resync_request(struct drbd_work *w, int cancel) { + struct drbd_conf *mdev = w->mdev; unsigned long bit; sector_t sector; const sector_t capacity = drbd_get_capacity(mdev->this_bdev); @@ -538,12 +571,12 @@ static int w_make_resync_request(struct drbd_conf *mdev, int i = 0; if (unlikely(cancel)) - return 1; + return 0; if (mdev->rs_total == 0) { /* empty resync? */ drbd_resync_finished(mdev); - return 1; + return 0; } if (!get_ldev(mdev)) { @@ -552,7 +585,7 @@ static int w_make_resync_request(struct drbd_conf *mdev, to continue resync with a broken disk makes no sense at all */ dev_err(DEV, "Disk broke down during resync!\n"); - return 1; + return 0; } max_bio_size = queue_max_hw_sectors(mdev->rq_queue) << 9; @@ -562,15 +595,15 @@ static int w_make_resync_request(struct drbd_conf *mdev, for (i = 0; i < number; i++) { /* Stop generating RS requests, when half of the send buffer is filled */ - mutex_lock(&mdev->data.mutex); - if (mdev->data.socket) { - queued = mdev->data.socket->sk->sk_wmem_queued; - sndbuf = mdev->data.socket->sk->sk_sndbuf; + mutex_lock(&mdev->tconn->data.mutex); + if (mdev->tconn->data.socket) { + queued = mdev->tconn->data.socket->sk->sk_wmem_queued; + sndbuf = mdev->tconn->data.socket->sk->sk_sndbuf; } else { queued = 1; sndbuf = 0; } - mutex_unlock(&mdev->data.mutex); + mutex_unlock(&mdev->tconn->data.mutex); if (queued > sndbuf / 2) goto requeue; @@ -581,7 +614,7 @@ next_sector: if (bit == DRBD_END_OF_BITMAP) { mdev->bm_resync_fo = drbd_bm_bits(mdev); put_ldev(mdev); - return 1; + return 0; } sector = BM_BIT_TO_SECT(bit); @@ -640,11 +673,11 @@ next_sector: /* adjust very last sectors, in case we are oddly sized */ if (sector + (size>>9) > capacity) size = (capacity-sector)<<9; - if (mdev->agreed_pro_version >= 89 && mdev->csums_tfm) { + if (mdev->tconn->agreed_pro_version >= 89 && mdev->tconn->csums_tfm) { switch (read_for_csum(mdev, sector, size)) { case -EIO: /* Disk failure */ put_ldev(mdev); - return 0; + return -EIO; case -EAGAIN: /* allocation failed, or ldev busy */ drbd_rs_complete_io(mdev, sector); mdev->bm_resync_fo = BM_SECT_TO_BIT(sector); @@ -657,13 +690,16 @@ next_sector: BUG(); } } else { + int err; + inc_rs_pending(mdev); - if (!drbd_send_drequest(mdev, P_RS_DATA_REQUEST, - sector, size, ID_SYNCER)) { + err = drbd_send_drequest(mdev, P_RS_DATA_REQUEST, + sector, size, ID_SYNCER); + if (err) { dev_err(DEV, "drbd_send_drequest() failed, aborting...\n"); dec_rs_pending(mdev); put_ldev(mdev); - return 0; + return err; } } } @@ -676,21 +712,23 @@ next_sector: * until then resync "work" is "inactive" ... */ put_ldev(mdev); - return 1; + return 0; } requeue: mdev->rs_in_flight += (i << (BM_BLOCK_SHIFT - 9)); mod_timer(&mdev->resync_timer, jiffies + SLEEP_TIME); put_ldev(mdev); - return 1; + return 0; } -static int w_make_ov_request(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +static int w_make_ov_request(struct drbd_work *w, int cancel) { + struct drbd_conf *mdev = w->mdev; int number, i, size; sector_t sector; const sector_t capacity = drbd_get_capacity(mdev->this_bdev); + bool stop_sector_reached = false; if (unlikely(cancel)) return 1; @@ -699,9 +737,17 @@ static int w_make_ov_request(struct drbd_conf *mdev, struct drbd_work *w, int ca sector = mdev->ov_position; for (i = 0; i < number; i++) { - if (sector >= capacity) { + if (sector >= capacity) return 1; - } + + /* We check for "finished" only in the reply path: + * w_e_end_ov_reply(). + * We need to send at least one request out. */ + stop_sector_reached = i > 0 + && verify_can_do_stop_sector(mdev) + && sector >= mdev->ov_stop_sector; + if (stop_sector_reached) + break; size = BM_BLOCK_SIZE; @@ -715,7 +761,7 @@ static int w_make_ov_request(struct drbd_conf *mdev, struct drbd_work *w, int ca size = (capacity-sector)<<9; inc_rs_pending(mdev); - if (!drbd_send_ov_request(mdev, sector, size)) { + if (drbd_send_ov_request(mdev, sector, size)) { dec_rs_pending(mdev); return 0; } @@ -725,56 +771,39 @@ static int w_make_ov_request(struct drbd_conf *mdev, struct drbd_work *w, int ca requeue: mdev->rs_in_flight += (i << (BM_BLOCK_SHIFT - 9)); - mod_timer(&mdev->resync_timer, jiffies + SLEEP_TIME); - return 1; -} - - -void start_resync_timer_fn(unsigned long data) -{ - struct drbd_conf *mdev = (struct drbd_conf *) data; - - drbd_queue_work(&mdev->data.work, &mdev->start_resync_work); -} - -int w_start_resync(struct drbd_conf *mdev, struct drbd_work *w, int cancel) -{ - if (atomic_read(&mdev->unacked_cnt) || atomic_read(&mdev->rs_pending_cnt)) { - dev_warn(DEV, "w_start_resync later...\n"); - mdev->start_resync_timer.expires = jiffies + HZ/10; - add_timer(&mdev->start_resync_timer); - return 1; - } - - drbd_start_resync(mdev, C_SYNC_SOURCE); - clear_bit(AHEAD_TO_SYNC_SOURCE, &mdev->flags); + if (i == 0 || !stop_sector_reached) + mod_timer(&mdev->resync_timer, jiffies + SLEEP_TIME); return 1; } -int w_ov_finished(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_ov_finished(struct drbd_work *w, int cancel) { + struct drbd_conf *mdev = w->mdev; kfree(w); - ov_oos_print(mdev); + ov_out_of_sync_print(mdev); drbd_resync_finished(mdev); - return 1; + return 0; } -static int w_resync_finished(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +static int w_resync_finished(struct drbd_work *w, int cancel) { + struct drbd_conf *mdev = w->mdev; kfree(w); drbd_resync_finished(mdev); - return 1; + return 0; } static void ping_peer(struct drbd_conf *mdev) { - clear_bit(GOT_PING_ACK, &mdev->flags); - request_ping(mdev); - wait_event(mdev->misc_wait, - test_bit(GOT_PING_ACK, &mdev->flags) || mdev->state.conn < C_CONNECTED); + struct drbd_tconn *tconn = mdev->tconn; + + clear_bit(GOT_PING_ACK, &tconn->flags); + request_ping(tconn); + wait_event(tconn->ping_wait, + test_bit(GOT_PING_ACK, &tconn->flags) || mdev->state.conn < C_CONNECTED); } int drbd_resync_finished(struct drbd_conf *mdev) @@ -799,7 +828,8 @@ int drbd_resync_finished(struct drbd_conf *mdev) w = kmalloc(sizeof(struct drbd_work), GFP_ATOMIC); if (w) { w->cb = w_resync_finished; - drbd_queue_work(&mdev->data.work, w); + w->mdev = mdev; + drbd_queue_work(&mdev->tconn->sender_work, w); return 1; } dev_err(DEV, "Warn failed to drbd_rs_del_all() and to kmalloc(w).\n"); @@ -808,7 +838,12 @@ int drbd_resync_finished(struct drbd_conf *mdev) dt = (jiffies - mdev->rs_start - mdev->rs_paused) / HZ; if (dt <= 0) dt = 1; + db = mdev->rs_total; + /* adjust for verify start and stop sectors, respective reached position */ + if (mdev->state.conn == C_VERIFY_S || mdev->state.conn == C_VERIFY_T) + db -= mdev->ov_left; + dbdt = Bit2KB(db/dt); mdev->rs_paused /= HZ; @@ -817,8 +852,8 @@ int drbd_resync_finished(struct drbd_conf *mdev) ping_peer(mdev); - spin_lock_irq(&mdev->req_lock); - os = mdev->state; + spin_lock_irq(&mdev->tconn->req_lock); + os = drbd_read_state(mdev); verify_done = (os.conn == C_VERIFY_S || os.conn == C_VERIFY_T); @@ -831,7 +866,7 @@ int drbd_resync_finished(struct drbd_conf *mdev) ns.conn = C_CONNECTED; dev_info(DEV, "%s done (total %lu sec; paused %lu sec; %lu K/sec)\n", - verify_done ? "Online verify " : "Resync", + verify_done ? "Online verify" : "Resync", dt + mdev->rs_paused, mdev->rs_paused, dbdt); n_oos = drbd_bm_total_weight(mdev); @@ -848,7 +883,7 @@ int drbd_resync_finished(struct drbd_conf *mdev) if (os.conn == C_SYNC_TARGET || os.conn == C_PAUSED_SYNC_T) khelper_cmd = "after-resync-target"; - if (mdev->csums_tfm && mdev->rs_total) { + if (mdev->tconn->csums_tfm && mdev->rs_total) { const unsigned long s = mdev->rs_same_csum; const unsigned long t = mdev->rs_total; const int ratio = @@ -906,13 +941,15 @@ int drbd_resync_finished(struct drbd_conf *mdev) _drbd_set_state(mdev, ns, CS_VERBOSE, NULL); out_unlock: - spin_unlock_irq(&mdev->req_lock); + spin_unlock_irq(&mdev->tconn->req_lock); put_ldev(mdev); out: mdev->rs_total = 0; mdev->rs_failed = 0; mdev->rs_paused = 0; - if (verify_done) + + /* reset start sector, if we reached end of device */ + if (verify_done && mdev->ov_left == 0) mdev->ov_start_sector = 0; drbd_md_sync(mdev); @@ -924,19 +961,19 @@ out: } /* helper */ -static void move_to_net_ee_or_free(struct drbd_conf *mdev, struct drbd_epoch_entry *e) +static void move_to_net_ee_or_free(struct drbd_conf *mdev, struct drbd_peer_request *peer_req) { - if (drbd_ee_has_active_page(e)) { + if (drbd_peer_req_has_active_page(peer_req)) { /* This might happen if sendpage() has not finished */ - int i = (e->size + PAGE_SIZE -1) >> PAGE_SHIFT; + int i = (peer_req->i.size + PAGE_SIZE -1) >> PAGE_SHIFT; atomic_add(i, &mdev->pp_in_use_by_net); atomic_sub(i, &mdev->pp_in_use); - spin_lock_irq(&mdev->req_lock); - list_add_tail(&e->w.list, &mdev->net_ee); - spin_unlock_irq(&mdev->req_lock); + spin_lock_irq(&mdev->tconn->req_lock); + list_add_tail(&peer_req->w.list, &mdev->net_ee); + spin_unlock_irq(&mdev->tconn->req_lock); wake_up(&drbd_pp_wait); } else - drbd_free_ee(mdev, e); + drbd_free_peer_req(mdev, peer_req); } /** @@ -945,174 +982,177 @@ static void move_to_net_ee_or_free(struct drbd_conf *mdev, struct drbd_epoch_ent * @w: work object. * @cancel: The connection will be closed anyways */ -int w_e_end_data_req(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_e_end_data_req(struct drbd_work *w, int cancel) { - struct drbd_epoch_entry *e = container_of(w, struct drbd_epoch_entry, w); - int ok; + struct drbd_peer_request *peer_req = container_of(w, struct drbd_peer_request, w); + struct drbd_conf *mdev = w->mdev; + int err; if (unlikely(cancel)) { - drbd_free_ee(mdev, e); + drbd_free_peer_req(mdev, peer_req); dec_unacked(mdev); - return 1; + return 0; } - if (likely((e->flags & EE_WAS_ERROR) == 0)) { - ok = drbd_send_block(mdev, P_DATA_REPLY, e); + if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) { + err = drbd_send_block(mdev, P_DATA_REPLY, peer_req); } else { if (__ratelimit(&drbd_ratelimit_state)) dev_err(DEV, "Sending NegDReply. sector=%llus.\n", - (unsigned long long)e->sector); + (unsigned long long)peer_req->i.sector); - ok = drbd_send_ack(mdev, P_NEG_DREPLY, e); + err = drbd_send_ack(mdev, P_NEG_DREPLY, peer_req); } dec_unacked(mdev); - move_to_net_ee_or_free(mdev, e); + move_to_net_ee_or_free(mdev, peer_req); - if (unlikely(!ok)) + if (unlikely(err)) dev_err(DEV, "drbd_send_block() failed\n"); - return ok; + return err; } /** - * w_e_end_rsdata_req() - Worker callback to send a P_RS_DATA_REPLY packet in response to a P_RS_DATA_REQUESTRS + * w_e_end_rsdata_req() - Worker callback to send a P_RS_DATA_REPLY packet in response to a P_RS_DATA_REQUEST * @mdev: DRBD device. * @w: work object. * @cancel: The connection will be closed anyways */ -int w_e_end_rsdata_req(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_e_end_rsdata_req(struct drbd_work *w, int cancel) { - struct drbd_epoch_entry *e = container_of(w, struct drbd_epoch_entry, w); - int ok; + struct drbd_peer_request *peer_req = container_of(w, struct drbd_peer_request, w); + struct drbd_conf *mdev = w->mdev; + int err; if (unlikely(cancel)) { - drbd_free_ee(mdev, e); + drbd_free_peer_req(mdev, peer_req); dec_unacked(mdev); - return 1; + return 0; } if (get_ldev_if_state(mdev, D_FAILED)) { - drbd_rs_complete_io(mdev, e->sector); + drbd_rs_complete_io(mdev, peer_req->i.sector); put_ldev(mdev); } if (mdev->state.conn == C_AHEAD) { - ok = drbd_send_ack(mdev, P_RS_CANCEL, e); - } else if (likely((e->flags & EE_WAS_ERROR) == 0)) { + err = drbd_send_ack(mdev, P_RS_CANCEL, peer_req); + } else if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) { if (likely(mdev->state.pdsk >= D_INCONSISTENT)) { inc_rs_pending(mdev); - ok = drbd_send_block(mdev, P_RS_DATA_REPLY, e); + err = drbd_send_block(mdev, P_RS_DATA_REPLY, peer_req); } else { if (__ratelimit(&drbd_ratelimit_state)) dev_err(DEV, "Not sending RSDataReply, " "partner DISKLESS!\n"); - ok = 1; + err = 0; } } else { if (__ratelimit(&drbd_ratelimit_state)) dev_err(DEV, "Sending NegRSDReply. sector %llus.\n", - (unsigned long long)e->sector); + (unsigned long long)peer_req->i.sector); - ok = drbd_send_ack(mdev, P_NEG_RS_DREPLY, e); + err = drbd_send_ack(mdev, P_NEG_RS_DREPLY, peer_req); /* update resync data with failure */ - drbd_rs_failed_io(mdev, e->sector, e->size); + drbd_rs_failed_io(mdev, peer_req->i.sector, peer_req->i.size); } dec_unacked(mdev); - move_to_net_ee_or_free(mdev, e); + move_to_net_ee_or_free(mdev, peer_req); - if (unlikely(!ok)) + if (unlikely(err)) dev_err(DEV, "drbd_send_block() failed\n"); - return ok; + return err; } -int w_e_end_csum_rs_req(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_e_end_csum_rs_req(struct drbd_work *w, int cancel) { - struct drbd_epoch_entry *e = container_of(w, struct drbd_epoch_entry, w); + struct drbd_peer_request *peer_req = container_of(w, struct drbd_peer_request, w); + struct drbd_conf *mdev = w->mdev; struct digest_info *di; int digest_size; void *digest = NULL; - int ok, eq = 0; + int err, eq = 0; if (unlikely(cancel)) { - drbd_free_ee(mdev, e); + drbd_free_peer_req(mdev, peer_req); dec_unacked(mdev); - return 1; + return 0; } if (get_ldev(mdev)) { - drbd_rs_complete_io(mdev, e->sector); + drbd_rs_complete_io(mdev, peer_req->i.sector); put_ldev(mdev); } - di = e->digest; + di = peer_req->digest; - if (likely((e->flags & EE_WAS_ERROR) == 0)) { + if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) { /* quick hack to try to avoid a race against reconfiguration. * a real fix would be much more involved, * introducing more locking mechanisms */ - if (mdev->csums_tfm) { - digest_size = crypto_hash_digestsize(mdev->csums_tfm); + if (mdev->tconn->csums_tfm) { + digest_size = crypto_hash_digestsize(mdev->tconn->csums_tfm); D_ASSERT(digest_size == di->digest_size); digest = kmalloc(digest_size, GFP_NOIO); } if (digest) { - drbd_csum_ee(mdev, mdev->csums_tfm, e, digest); + drbd_csum_ee(mdev, mdev->tconn->csums_tfm, peer_req, digest); eq = !memcmp(digest, di->digest, digest_size); kfree(digest); } if (eq) { - drbd_set_in_sync(mdev, e->sector, e->size); + drbd_set_in_sync(mdev, peer_req->i.sector, peer_req->i.size); /* rs_same_csums unit is BM_BLOCK_SIZE */ - mdev->rs_same_csum += e->size >> BM_BLOCK_SHIFT; - ok = drbd_send_ack(mdev, P_RS_IS_IN_SYNC, e); + mdev->rs_same_csum += peer_req->i.size >> BM_BLOCK_SHIFT; + err = drbd_send_ack(mdev, P_RS_IS_IN_SYNC, peer_req); } else { inc_rs_pending(mdev); - e->block_id = ID_SYNCER; /* By setting block_id, digest pointer becomes invalid! */ - e->flags &= ~EE_HAS_DIGEST; /* This e no longer has a digest pointer */ + peer_req->block_id = ID_SYNCER; /* By setting block_id, digest pointer becomes invalid! */ + peer_req->flags &= ~EE_HAS_DIGEST; /* This peer request no longer has a digest pointer */ kfree(di); - ok = drbd_send_block(mdev, P_RS_DATA_REPLY, e); + err = drbd_send_block(mdev, P_RS_DATA_REPLY, peer_req); } } else { - ok = drbd_send_ack(mdev, P_NEG_RS_DREPLY, e); + err = drbd_send_ack(mdev, P_NEG_RS_DREPLY, peer_req); if (__ratelimit(&drbd_ratelimit_state)) dev_err(DEV, "Sending NegDReply. I guess it gets messy.\n"); } dec_unacked(mdev); - move_to_net_ee_or_free(mdev, e); + move_to_net_ee_or_free(mdev, peer_req); - if (unlikely(!ok)) + if (unlikely(err)) dev_err(DEV, "drbd_send_block/ack() failed\n"); - return ok; + return err; } -/* TODO merge common code with w_e_send_csum */ -int w_e_end_ov_req(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_e_end_ov_req(struct drbd_work *w, int cancel) { - struct drbd_epoch_entry *e = container_of(w, struct drbd_epoch_entry, w); - sector_t sector = e->sector; - unsigned int size = e->size; + struct drbd_peer_request *peer_req = container_of(w, struct drbd_peer_request, w); + struct drbd_conf *mdev = w->mdev; + sector_t sector = peer_req->i.sector; + unsigned int size = peer_req->i.size; int digest_size; void *digest; - int ok = 1; + int err = 0; if (unlikely(cancel)) goto out; - digest_size = crypto_hash_digestsize(mdev->verify_tfm); + digest_size = crypto_hash_digestsize(mdev->tconn->verify_tfm); digest = kmalloc(digest_size, GFP_NOIO); if (!digest) { - ok = 0; /* terminate the connection in case the allocation failed */ + err = 1; /* terminate the connection in case the allocation failed */ goto out; } - if (likely(!(e->flags & EE_WAS_ERROR))) - drbd_csum_ee(mdev, mdev->verify_tfm, e, digest); + if (likely(!(peer_req->flags & EE_WAS_ERROR))) + drbd_csum_ee(mdev, mdev->tconn->verify_tfm, peer_req, digest); else memset(digest, 0, digest_size); @@ -1120,25 +1160,23 @@ int w_e_end_ov_req(struct drbd_conf *mdev, struct drbd_work *w, int cancel) * In case we block on congestion, we could otherwise run into * some distributed deadlock, if the other side blocks on * congestion as well, because our receiver blocks in - * drbd_pp_alloc due to pp_in_use > max_buffers. */ - drbd_free_ee(mdev, e); - e = NULL; + * drbd_alloc_pages due to pp_in_use > max_buffers. */ + drbd_free_peer_req(mdev, peer_req); + peer_req = NULL; inc_rs_pending(mdev); - ok = drbd_send_drequest_csum(mdev, sector, size, - digest, digest_size, - P_OV_REPLY); - if (!ok) + err = drbd_send_drequest_csum(mdev, sector, size, digest, digest_size, P_OV_REPLY); + if (err) dec_rs_pending(mdev); kfree(digest); out: - if (e) - drbd_free_ee(mdev, e); + if (peer_req) + drbd_free_peer_req(mdev, peer_req); dec_unacked(mdev); - return ok; + return err; } -void drbd_ov_oos_found(struct drbd_conf *mdev, sector_t sector, int size) +void drbd_ov_out_of_sync_found(struct drbd_conf *mdev, sector_t sector, int size) { if (mdev->ov_last_oos_start + mdev->ov_last_oos_size == sector) { mdev->ov_last_oos_size += size>>9; @@ -1149,36 +1187,38 @@ void drbd_ov_oos_found(struct drbd_conf *mdev, sector_t sector, int size) drbd_set_out_of_sync(mdev, sector, size); } -int w_e_end_ov_reply(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_e_end_ov_reply(struct drbd_work *w, int cancel) { - struct drbd_epoch_entry *e = container_of(w, struct drbd_epoch_entry, w); + struct drbd_peer_request *peer_req = container_of(w, struct drbd_peer_request, w); + struct drbd_conf *mdev = w->mdev; struct digest_info *di; void *digest; - sector_t sector = e->sector; - unsigned int size = e->size; + sector_t sector = peer_req->i.sector; + unsigned int size = peer_req->i.size; int digest_size; - int ok, eq = 0; + int err, eq = 0; + bool stop_sector_reached = false; if (unlikely(cancel)) { - drbd_free_ee(mdev, e); + drbd_free_peer_req(mdev, peer_req); dec_unacked(mdev); - return 1; + return 0; } /* after "cancel", because after drbd_disconnect/drbd_rs_cancel_all * the resync lru has been cleaned up already */ if (get_ldev(mdev)) { - drbd_rs_complete_io(mdev, e->sector); + drbd_rs_complete_io(mdev, peer_req->i.sector); put_ldev(mdev); } - di = e->digest; + di = peer_req->digest; - if (likely((e->flags & EE_WAS_ERROR) == 0)) { - digest_size = crypto_hash_digestsize(mdev->verify_tfm); + if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) { + digest_size = crypto_hash_digestsize(mdev->tconn->verify_tfm); digest = kmalloc(digest_size, GFP_NOIO); if (digest) { - drbd_csum_ee(mdev, mdev->verify_tfm, e, digest); + drbd_csum_ee(mdev, mdev->tconn->verify_tfm, peer_req, digest); D_ASSERT(digest_size == di->digest_size); eq = !memcmp(digest, di->digest, digest_size); @@ -1186,19 +1226,19 @@ int w_e_end_ov_reply(struct drbd_conf *mdev, struct drbd_work *w, int cancel) } } - /* Free e and pages before send. - * In case we block on congestion, we could otherwise run into - * some distributed deadlock, if the other side blocks on - * congestion as well, because our receiver blocks in - * drbd_pp_alloc due to pp_in_use > max_buffers. */ - drbd_free_ee(mdev, e); + /* Free peer_req and pages before send. + * In case we block on congestion, we could otherwise run into + * some distributed deadlock, if the other side blocks on + * congestion as well, because our receiver blocks in + * drbd_alloc_pages due to pp_in_use > max_buffers. */ + drbd_free_peer_req(mdev, peer_req); if (!eq) - drbd_ov_oos_found(mdev, sector, size); + drbd_ov_out_of_sync_found(mdev, sector, size); else - ov_oos_print(mdev); + ov_out_of_sync_print(mdev); - ok = drbd_send_ack_ex(mdev, P_OV_RESULT, sector, size, - eq ? ID_IN_SYNC : ID_OUT_OF_SYNC); + err = drbd_send_ack_ex(mdev, P_OV_RESULT, sector, size, + eq ? ID_IN_SYNC : ID_OUT_OF_SYNC); dec_unacked(mdev); @@ -1208,73 +1248,102 @@ int w_e_end_ov_reply(struct drbd_conf *mdev, struct drbd_work *w, int cancel) if ((mdev->ov_left & 0x200) == 0x200) drbd_advance_rs_marks(mdev, mdev->ov_left); - if (mdev->ov_left == 0) { - ov_oos_print(mdev); + stop_sector_reached = verify_can_do_stop_sector(mdev) && + (sector + (size>>9)) >= mdev->ov_stop_sector; + + if (mdev->ov_left == 0 || stop_sector_reached) { + ov_out_of_sync_print(mdev); drbd_resync_finished(mdev); } - return ok; + return err; } -int w_prev_work_done(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_prev_work_done(struct drbd_work *w, int cancel) { struct drbd_wq_barrier *b = container_of(w, struct drbd_wq_barrier, w); + complete(&b->done); - return 1; + return 0; } -int w_send_barrier(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +/* FIXME + * We need to track the number of pending barrier acks, + * and to be able to wait for them. + * See also comment in drbd_adm_attach before drbd_suspend_io. + */ +int drbd_send_barrier(struct drbd_tconn *tconn) { - struct drbd_tl_epoch *b = container_of(w, struct drbd_tl_epoch, w); - struct p_barrier *p = &mdev->data.sbuf.barrier; - int ok = 1; - - /* really avoid racing with tl_clear. w.cb may have been referenced - * just before it was reassigned and re-queued, so double check that. - * actually, this race was harmless, since we only try to send the - * barrier packet here, and otherwise do nothing with the object. - * but compare with the head of w_clear_epoch */ - spin_lock_irq(&mdev->req_lock); - if (w->cb != w_send_barrier || mdev->state.conn < C_CONNECTED) - cancel = 1; - spin_unlock_irq(&mdev->req_lock); - if (cancel) - return 1; + struct p_barrier *p; + struct drbd_socket *sock; - if (!drbd_get_data_sock(mdev)) - return 0; - p->barrier = b->br_number; - /* inc_ap_pending was done where this was queued. - * dec_ap_pending will be done in got_BarrierAck - * or (on connection loss) in w_clear_epoch. */ - ok = _drbd_send_cmd(mdev, mdev->data.socket, P_BARRIER, - (struct p_header80 *)p, sizeof(*p), 0); - drbd_put_data_sock(mdev); - - return ok; + sock = &tconn->data; + p = conn_prepare_command(tconn, sock); + if (!p) + return -EIO; + p->barrier = tconn->send.current_epoch_nr; + p->pad = 0; + tconn->send.current_epoch_writes = 0; + + return conn_send_command(tconn, sock, P_BARRIER, sizeof(*p), NULL, 0); } -int w_send_write_hint(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_send_write_hint(struct drbd_work *w, int cancel) { + struct drbd_conf *mdev = w->mdev; + struct drbd_socket *sock; + if (cancel) - return 1; - return drbd_send_short_cmd(mdev, P_UNPLUG_REMOTE); + return 0; + sock = &mdev->tconn->data; + if (!drbd_prepare_command(mdev, sock)) + return -EIO; + return drbd_send_command(mdev, sock, P_UNPLUG_REMOTE, 0, NULL, 0); } -int w_send_oos(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +static void re_init_if_first_write(struct drbd_tconn *tconn, unsigned int epoch) +{ + if (!tconn->send.seen_any_write_yet) { + tconn->send.seen_any_write_yet = true; + tconn->send.current_epoch_nr = epoch; + tconn->send.current_epoch_writes = 0; + } +} + +static void maybe_send_barrier(struct drbd_tconn *tconn, unsigned int epoch) +{ + /* re-init if first write on this connection */ + if (!tconn->send.seen_any_write_yet) + return; + if (tconn->send.current_epoch_nr != epoch) { + if (tconn->send.current_epoch_writes) + drbd_send_barrier(tconn); + tconn->send.current_epoch_nr = epoch; + } +} + +int w_send_out_of_sync(struct drbd_work *w, int cancel) { struct drbd_request *req = container_of(w, struct drbd_request, w); - int ok; + struct drbd_conf *mdev = w->mdev; + struct drbd_tconn *tconn = mdev->tconn; + int err; if (unlikely(cancel)) { - req_mod(req, send_canceled); - return 1; + req_mod(req, SEND_CANCELED); + return 0; } - ok = drbd_send_oos(mdev, req); - req_mod(req, oos_handed_to_network); + /* this time, no tconn->send.current_epoch_writes++; + * If it was sent, it was the closing barrier for the last + * replicated epoch, before we went into AHEAD mode. + * No more barriers will be sent, until we leave AHEAD mode again. */ + maybe_send_barrier(tconn, req->epoch); + + err = drbd_send_out_of_sync(mdev, req); + req_mod(req, OOS_HANDED_TO_NETWORK); - return ok; + return err; } /** @@ -1283,20 +1352,26 @@ int w_send_oos(struct drbd_conf *mdev, struct drbd_work *w, int cancel) * @w: work object. * @cancel: The connection will be closed anyways */ -int w_send_dblock(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_send_dblock(struct drbd_work *w, int cancel) { struct drbd_request *req = container_of(w, struct drbd_request, w); - int ok; + struct drbd_conf *mdev = w->mdev; + struct drbd_tconn *tconn = mdev->tconn; + int err; if (unlikely(cancel)) { - req_mod(req, send_canceled); - return 1; + req_mod(req, SEND_CANCELED); + return 0; } - ok = drbd_send_dblock(mdev, req); - req_mod(req, ok ? handed_over_to_network : send_failed); + re_init_if_first_write(tconn, req->epoch); + maybe_send_barrier(tconn, req->epoch); + tconn->send.current_epoch_writes++; + + err = drbd_send_dblock(mdev, req); + req_mod(req, err ? SEND_FAILED : HANDED_OVER_TO_NETWORK); - return ok; + return err; } /** @@ -1305,57 +1380,61 @@ int w_send_dblock(struct drbd_conf *mdev, struct drbd_work *w, int cancel) * @w: work object. * @cancel: The connection will be closed anyways */ -int w_send_read_req(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_send_read_req(struct drbd_work *w, int cancel) { struct drbd_request *req = container_of(w, struct drbd_request, w); - int ok; + struct drbd_conf *mdev = w->mdev; + struct drbd_tconn *tconn = mdev->tconn; + int err; if (unlikely(cancel)) { - req_mod(req, send_canceled); - return 1; + req_mod(req, SEND_CANCELED); + return 0; } - ok = drbd_send_drequest(mdev, P_DATA_REQUEST, req->sector, req->size, - (unsigned long)req); + /* Even read requests may close a write epoch, + * if there was any yet. */ + maybe_send_barrier(tconn, req->epoch); - if (!ok) { - /* ?? we set C_TIMEOUT or C_BROKEN_PIPE in drbd_send(); - * so this is probably redundant */ - if (mdev->state.conn >= C_CONNECTED) - drbd_force_state(mdev, NS(conn, C_NETWORK_FAILURE)); - } - req_mod(req, ok ? handed_over_to_network : send_failed); + err = drbd_send_drequest(mdev, P_DATA_REQUEST, req->i.sector, req->i.size, + (unsigned long)req); + + req_mod(req, err ? SEND_FAILED : HANDED_OVER_TO_NETWORK); - return ok; + return err; } -int w_restart_disk_io(struct drbd_conf *mdev, struct drbd_work *w, int cancel) +int w_restart_disk_io(struct drbd_work *w, int cancel) { struct drbd_request *req = container_of(w, struct drbd_request, w); + struct drbd_conf *mdev = w->mdev; if (bio_data_dir(req->master_bio) == WRITE && req->rq_state & RQ_IN_ACT_LOG) - drbd_al_begin_io(mdev, req->sector); - /* Calling drbd_al_begin_io() out of the worker might deadlocks - theoretically. Practically it can not deadlock, since this is - only used when unfreezing IOs. All the extents of the requests - that made it into the TL are already active */ + drbd_al_begin_io(mdev, &req->i); drbd_req_make_private_bio(req, req->master_bio); req->private_bio->bi_bdev = mdev->ldev->backing_bdev; generic_make_request(req->private_bio); - return 1; + return 0; } static int _drbd_may_sync_now(struct drbd_conf *mdev) { struct drbd_conf *odev = mdev; + int resync_after; while (1) { - if (odev->sync_conf.after == -1) + if (!odev->ldev) + return 1; + rcu_read_lock(); + resync_after = rcu_dereference(odev->ldev->disk_conf)->resync_after; + rcu_read_unlock(); + if (resync_after == -1) + return 1; + odev = minor_to_mdev(resync_after); + if (!expect(odev)) return 1; - odev = minor_to_mdev(odev->sync_conf.after); - ERR_IF(!odev) return 1; if ((odev->state.conn >= C_SYNC_SOURCE && odev->state.conn <= C_PAUSED_SYNC_T) || odev->state.aftr_isp || odev->state.peer_isp || @@ -1375,16 +1454,15 @@ static int _drbd_pause_after(struct drbd_conf *mdev) struct drbd_conf *odev; int i, rv = 0; - for (i = 0; i < minor_count; i++) { - odev = minor_to_mdev(i); - if (!odev) - continue; + rcu_read_lock(); + idr_for_each_entry(&minors, odev, i) { if (odev->state.conn == C_STANDALONE && odev->state.disk == D_DISKLESS) continue; if (!_drbd_may_sync_now(odev)) rv |= (__drbd_set_state(_NS(odev, aftr_isp, 1), CS_HARD, NULL) != SS_NOTHING_TO_DO); } + rcu_read_unlock(); return rv; } @@ -1400,10 +1478,8 @@ static int _drbd_resume_next(struct drbd_conf *mdev) struct drbd_conf *odev; int i, rv = 0; - for (i = 0; i < minor_count; i++) { - odev = minor_to_mdev(i); - if (!odev) - continue; + rcu_read_lock(); + idr_for_each_entry(&minors, odev, i) { if (odev->state.conn == C_STANDALONE && odev->state.disk == D_DISKLESS) continue; if (odev->state.aftr_isp) { @@ -1413,6 +1489,7 @@ static int _drbd_resume_next(struct drbd_conf *mdev) != SS_NOTHING_TO_DO) ; } } + rcu_read_unlock(); return rv; } @@ -1430,57 +1507,86 @@ void suspend_other_sg(struct drbd_conf *mdev) write_unlock_irq(&global_state_lock); } -static int sync_after_error(struct drbd_conf *mdev, int o_minor) +/* caller must hold global_state_lock */ +enum drbd_ret_code drbd_resync_after_valid(struct drbd_conf *mdev, int o_minor) { struct drbd_conf *odev; + int resync_after; if (o_minor == -1) return NO_ERROR; if (o_minor < -1 || minor_to_mdev(o_minor) == NULL) - return ERR_SYNC_AFTER; + return ERR_RESYNC_AFTER; /* check for loops */ odev = minor_to_mdev(o_minor); while (1) { if (odev == mdev) - return ERR_SYNC_AFTER_CYCLE; + return ERR_RESYNC_AFTER_CYCLE; + rcu_read_lock(); + resync_after = rcu_dereference(odev->ldev->disk_conf)->resync_after; + rcu_read_unlock(); /* dependency chain ends here, no cycles. */ - if (odev->sync_conf.after == -1) + if (resync_after == -1) return NO_ERROR; /* follow the dependency chain */ - odev = minor_to_mdev(odev->sync_conf.after); + odev = minor_to_mdev(resync_after); } } -int drbd_alter_sa(struct drbd_conf *mdev, int na) +/* caller must hold global_state_lock */ +void drbd_resync_after_changed(struct drbd_conf *mdev) { int changes; - int retcode; - write_lock_irq(&global_state_lock); - retcode = sync_after_error(mdev, na); - if (retcode == NO_ERROR) { - mdev->sync_conf.after = na; - do { - changes = _drbd_pause_after(mdev); - changes |= _drbd_resume_next(mdev); - } while (changes); - } - write_unlock_irq(&global_state_lock); - return retcode; + do { + changes = _drbd_pause_after(mdev); + changes |= _drbd_resume_next(mdev); + } while (changes); } void drbd_rs_controller_reset(struct drbd_conf *mdev) { + struct fifo_buffer *plan; + atomic_set(&mdev->rs_sect_in, 0); atomic_set(&mdev->rs_sect_ev, 0); mdev->rs_in_flight = 0; - mdev->rs_planed = 0; - spin_lock(&mdev->peer_seq_lock); - fifo_set(&mdev->rs_plan_s, 0); - spin_unlock(&mdev->peer_seq_lock); + + /* Updating the RCU protected object in place is necessary since + this function gets called from atomic context. + It is valid since all other updates also lead to an completely + empty fifo */ + rcu_read_lock(); + plan = rcu_dereference(mdev->rs_plan_s); + plan->total = 0; + fifo_set(plan, 0); + rcu_read_unlock(); +} + +void start_resync_timer_fn(unsigned long data) +{ + struct drbd_conf *mdev = (struct drbd_conf *) data; + + drbd_queue_work(&mdev->tconn->sender_work, &mdev->start_resync_work); +} + +int w_start_resync(struct drbd_work *w, int cancel) +{ + struct drbd_conf *mdev = w->mdev; + + if (atomic_read(&mdev->unacked_cnt) || atomic_read(&mdev->rs_pending_cnt)) { + dev_warn(DEV, "w_start_resync later...\n"); + mdev->start_resync_timer.expires = jiffies + HZ/10; + add_timer(&mdev->start_resync_timer); + return 0; + } + + drbd_start_resync(mdev, C_SYNC_SOURCE); + clear_bit(AHEAD_TO_SYNC_SOURCE, &mdev->flags); + return 0; } /** @@ -1501,43 +1607,58 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side) return; } - if (side == C_SYNC_TARGET) { - /* Since application IO was locked out during C_WF_BITMAP_T and - C_WF_SYNC_UUID we are still unmodified. Before going to C_SYNC_TARGET - we check that we might make the data inconsistent. */ - r = drbd_khelper(mdev, "before-resync-target"); - r = (r >> 8) & 0xff; - if (r > 0) { - dev_info(DEV, "before-resync-target handler returned %d, " - "dropping connection.\n", r); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); - return; - } - } else /* C_SYNC_SOURCE */ { - r = drbd_khelper(mdev, "before-resync-source"); - r = (r >> 8) & 0xff; - if (r > 0) { - if (r == 3) { - dev_info(DEV, "before-resync-source handler returned %d, " - "ignoring. Old userland tools?", r); - } else { - dev_info(DEV, "before-resync-source handler returned %d, " + if (!test_bit(B_RS_H_DONE, &mdev->flags)) { + if (side == C_SYNC_TARGET) { + /* Since application IO was locked out during C_WF_BITMAP_T and + C_WF_SYNC_UUID we are still unmodified. Before going to C_SYNC_TARGET + we check that we might make the data inconsistent. */ + r = drbd_khelper(mdev, "before-resync-target"); + r = (r >> 8) & 0xff; + if (r > 0) { + dev_info(DEV, "before-resync-target handler returned %d, " "dropping connection.\n", r); - drbd_force_state(mdev, NS(conn, C_DISCONNECTING)); + conn_request_state(mdev->tconn, NS(conn, C_DISCONNECTING), CS_HARD); return; } + } else /* C_SYNC_SOURCE */ { + r = drbd_khelper(mdev, "before-resync-source"); + r = (r >> 8) & 0xff; + if (r > 0) { + if (r == 3) { + dev_info(DEV, "before-resync-source handler returned %d, " + "ignoring. Old userland tools?", r); + } else { + dev_info(DEV, "before-resync-source handler returned %d, " + "dropping connection.\n", r); + conn_request_state(mdev->tconn, NS(conn, C_DISCONNECTING), CS_HARD); + return; + } + } } } - drbd_state_lock(mdev); + if (current == mdev->tconn->worker.task) { + /* The worker should not sleep waiting for state_mutex, + that can take long */ + if (!mutex_trylock(mdev->state_mutex)) { + set_bit(B_RS_H_DONE, &mdev->flags); + mdev->start_resync_timer.expires = jiffies + HZ/5; + add_timer(&mdev->start_resync_timer); + return; + } + } else { + mutex_lock(mdev->state_mutex); + } + clear_bit(B_RS_H_DONE, &mdev->flags); + write_lock_irq(&global_state_lock); if (!get_ldev_if_state(mdev, D_NEGOTIATING)) { write_unlock_irq(&global_state_lock); - drbd_state_unlock(mdev); + mutex_unlock(mdev->state_mutex); return; } - ns.i = mdev->state.i; + ns = drbd_read_state(mdev); ns.aftr_isp = !_drbd_may_sync_now(mdev); @@ -1549,7 +1670,7 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side) ns.pdsk = D_INCONSISTENT; r = __drbd_set_state(mdev, ns, CS_VERBOSE, NULL); - ns = mdev->state; + ns = drbd_read_state(mdev); if (ns.conn < C_CONNECTED) r = SS_UNKNOWN_ERROR; @@ -1575,6 +1696,10 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side) write_unlock_irq(&global_state_lock); if (r == SS_SUCCESS) { + /* reset rs_last_bcast when a resync or verify is started, + * to deal with potential jiffies wrap. */ + mdev->rs_last_bcast = jiffies - HZ; + dev_info(DEV, "Began resync as %s (will sync %lu KB [%lu bits set]).\n", drbd_conn_str(ns.conn), (unsigned long) mdev->rs_total << (BM_BLOCK_SHIFT-10), @@ -1589,10 +1714,10 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side) * drbd_resync_finished from here in that case. * We drbd_gen_and_send_sync_uuid here for protocol < 96, * and from after_state_ch otherwise. */ - if (side == C_SYNC_SOURCE && mdev->agreed_pro_version < 96) + if (side == C_SYNC_SOURCE && mdev->tconn->agreed_pro_version < 96) drbd_gen_and_send_sync_uuid(mdev); - if (mdev->agreed_pro_version < 95 && mdev->rs_total == 0) { + if (mdev->tconn->agreed_pro_version < 95 && mdev->rs_total == 0) { /* This still has a race (about when exactly the peers * detect connection loss) that can lead to a full sync * on next handshake. In 8.3.9 we fixed this with explicit @@ -1603,10 +1728,16 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side) * detect connection loss, then waiting for a ping * response (implicit in drbd_resync_finished) reduces * the race considerably, but does not solve it. */ - if (side == C_SYNC_SOURCE) - schedule_timeout_interruptible( - mdev->net_conf->ping_int * HZ + - mdev->net_conf->ping_timeo*HZ/9); + if (side == C_SYNC_SOURCE) { + struct net_conf *nc; + int timeo; + + rcu_read_lock(); + nc = rcu_dereference(mdev->tconn->net_conf); + timeo = nc->ping_int * HZ + nc->ping_timeo * HZ / 9; + rcu_read_unlock(); + schedule_timeout_interruptible(timeo); + } drbd_resync_finished(mdev); } @@ -1621,114 +1752,180 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side) drbd_md_sync(mdev); } put_ldev(mdev); - drbd_state_unlock(mdev); + mutex_unlock(mdev->state_mutex); } -int drbd_worker(struct drbd_thread *thi) +/* If the resource already closed the current epoch, but we did not + * (because we have not yet seen new requests), we should send the + * corresponding barrier now. Must be checked within the same spinlock + * that is used to check for new requests. */ +bool need_to_send_barrier(struct drbd_tconn *connection) { - struct drbd_conf *mdev = thi->mdev; - struct drbd_work *w = NULL; - LIST_HEAD(work_list); - int intr = 0, i; + if (!connection->send.seen_any_write_yet) + return false; + + /* Skip barriers that do not contain any writes. + * This may happen during AHEAD mode. */ + if (!connection->send.current_epoch_writes) + return false; + + /* ->req_lock is held when requests are queued on + * connection->sender_work, and put into ->transfer_log. + * It is also held when ->current_tle_nr is increased. + * So either there are already new requests queued, + * and corresponding barriers will be send there. + * Or nothing new is queued yet, so the difference will be 1. + */ + if (atomic_read(&connection->current_tle_nr) != + connection->send.current_epoch_nr + 1) + return false; + + return true; +} + +bool dequeue_work_batch(struct drbd_work_queue *queue, struct list_head *work_list) +{ + spin_lock_irq(&queue->q_lock); + list_splice_init(&queue->q, work_list); + spin_unlock_irq(&queue->q_lock); + return !list_empty(work_list); +} - sprintf(current->comm, "drbd%d_worker", mdev_to_minor(mdev)); +bool dequeue_work_item(struct drbd_work_queue *queue, struct list_head *work_list) +{ + spin_lock_irq(&queue->q_lock); + if (!list_empty(&queue->q)) + list_move(queue->q.next, work_list); + spin_unlock_irq(&queue->q_lock); + return !list_empty(work_list); +} - while (get_t_state(thi) == Running) { - drbd_thread_current_set_cpu(mdev); +void wait_for_work(struct drbd_tconn *connection, struct list_head *work_list) +{ + DEFINE_WAIT(wait); + struct net_conf *nc; + int uncork, cork; - if (down_trylock(&mdev->data.work.s)) { - mutex_lock(&mdev->data.mutex); - if (mdev->data.socket && !mdev->net_conf->no_cork) - drbd_tcp_uncork(mdev->data.socket); - mutex_unlock(&mdev->data.mutex); + dequeue_work_item(&connection->sender_work, work_list); + if (!list_empty(work_list)) + return; - intr = down_interruptible(&mdev->data.work.s); + /* Still nothing to do? + * Maybe we still need to close the current epoch, + * even if no new requests are queued yet. + * + * Also, poke TCP, just in case. + * Then wait for new work (or signal). */ + rcu_read_lock(); + nc = rcu_dereference(connection->net_conf); + uncork = nc ? nc->tcp_cork : 0; + rcu_read_unlock(); + if (uncork) { + mutex_lock(&connection->data.mutex); + if (connection->data.socket) + drbd_tcp_uncork(connection->data.socket); + mutex_unlock(&connection->data.mutex); + } - mutex_lock(&mdev->data.mutex); - if (mdev->data.socket && !mdev->net_conf->no_cork) - drbd_tcp_cork(mdev->data.socket); - mutex_unlock(&mdev->data.mutex); + for (;;) { + int send_barrier; + prepare_to_wait(&connection->sender_work.q_wait, &wait, TASK_INTERRUPTIBLE); + spin_lock_irq(&connection->req_lock); + spin_lock(&connection->sender_work.q_lock); /* FIXME get rid of this one? */ + /* dequeue single item only, + * we still use drbd_queue_work_front() in some places */ + if (!list_empty(&connection->sender_work.q)) + list_move(connection->sender_work.q.next, work_list); + spin_unlock(&connection->sender_work.q_lock); /* FIXME get rid of this one? */ + if (!list_empty(work_list) || signal_pending(current)) { + spin_unlock_irq(&connection->req_lock); + break; } + send_barrier = need_to_send_barrier(connection); + spin_unlock_irq(&connection->req_lock); + if (send_barrier) { + drbd_send_barrier(connection); + connection->send.current_epoch_nr++; + } + schedule(); + /* may be woken up for other things but new work, too, + * e.g. if the current epoch got closed. + * In which case we send the barrier above. */ + } + finish_wait(&connection->sender_work.q_wait, &wait); + + /* someone may have changed the config while we have been waiting above. */ + rcu_read_lock(); + nc = rcu_dereference(connection->net_conf); + cork = nc ? nc->tcp_cork : 0; + rcu_read_unlock(); + mutex_lock(&connection->data.mutex); + if (connection->data.socket) { + if (cork) + drbd_tcp_cork(connection->data.socket); + else if (!uncork) + drbd_tcp_uncork(connection->data.socket); + } + mutex_unlock(&connection->data.mutex); +} - if (intr) { - D_ASSERT(intr == -EINTR); +int drbd_worker(struct drbd_thread *thi) +{ + struct drbd_tconn *tconn = thi->tconn; + struct drbd_work *w = NULL; + struct drbd_conf *mdev; + LIST_HEAD(work_list); + int vnr; + + while (get_t_state(thi) == RUNNING) { + drbd_thread_current_set_cpu(thi); + + /* as long as we use drbd_queue_work_front(), + * we may only dequeue single work items here, not batches. */ + if (list_empty(&work_list)) + wait_for_work(tconn, &work_list); + + if (signal_pending(current)) { flush_signals(current); - ERR_IF (get_t_state(thi) == Running) + if (get_t_state(thi) == RUNNING) { + conn_warn(tconn, "Worker got an unexpected signal\n"); continue; + } break; } - if (get_t_state(thi) != Running) + if (get_t_state(thi) != RUNNING) break; - /* With this break, we have done a down() but not consumed - the entry from the list. The cleanup code takes care of - this... */ - - w = NULL; - spin_lock_irq(&mdev->data.work.q_lock); - ERR_IF(list_empty(&mdev->data.work.q)) { - /* something terribly wrong in our logic. - * we were able to down() the semaphore, - * but the list is empty... doh. - * - * what is the best thing to do now? - * try again from scratch, restarting the receiver, - * asender, whatnot? could break even more ugly, - * e.g. when we are primary, but no good local data. - * - * I'll try to get away just starting over this loop. - */ - spin_unlock_irq(&mdev->data.work.q_lock); - continue; - } - w = list_entry(mdev->data.work.q.next, struct drbd_work, list); - list_del_init(&w->list); - spin_unlock_irq(&mdev->data.work.q_lock); - - if (!w->cb(mdev, w, mdev->state.conn < C_CONNECTED)) { - /* dev_warn(DEV, "worker: a callback failed! \n"); */ - if (mdev->state.conn >= C_CONNECTED) - drbd_force_state(mdev, - NS(conn, C_NETWORK_FAILURE)); + + while (!list_empty(&work_list)) { + w = list_first_entry(&work_list, struct drbd_work, list); + list_del_init(&w->list); + if (w->cb(w, tconn->cstate < C_WF_REPORT_PARAMS) == 0) + continue; + if (tconn->cstate >= C_WF_REPORT_PARAMS) + conn_request_state(tconn, NS(conn, C_NETWORK_FAILURE), CS_HARD); } } - D_ASSERT(test_bit(DEVICE_DYING, &mdev->flags)); - D_ASSERT(test_bit(CONFIG_PENDING, &mdev->flags)); - - spin_lock_irq(&mdev->data.work.q_lock); - i = 0; - while (!list_empty(&mdev->data.work.q)) { - list_splice_init(&mdev->data.work.q, &work_list); - spin_unlock_irq(&mdev->data.work.q_lock); + do { while (!list_empty(&work_list)) { - w = list_entry(work_list.next, struct drbd_work, list); + w = list_first_entry(&work_list, struct drbd_work, list); list_del_init(&w->list); - w->cb(mdev, w, 1); - i++; /* dead debugging code */ + w->cb(w, 1); } - - spin_lock_irq(&mdev->data.work.q_lock); + dequeue_work_batch(&tconn->sender_work, &work_list); + } while (!list_empty(&work_list)); + + rcu_read_lock(); + idr_for_each_entry(&tconn->volumes, mdev, vnr) { + D_ASSERT(mdev->state.disk == D_DISKLESS && mdev->state.conn == C_STANDALONE); + kref_get(&mdev->kref); + rcu_read_unlock(); + drbd_mdev_cleanup(mdev); + kref_put(&mdev->kref, &drbd_minor_destroy); + rcu_read_lock(); } - sema_init(&mdev->data.work.s, 0); - /* DANGEROUS race: if someone did queue his work within the spinlock, - * but up() ed outside the spinlock, we could get an up() on the - * semaphore without corresponding list entry. - * So don't do that. - */ - spin_unlock_irq(&mdev->data.work.q_lock); - - D_ASSERT(mdev->state.disk == D_DISKLESS && mdev->state.conn == C_STANDALONE); - /* _drbd_set_state only uses stop_nowait. - * wait here for the Exiting receiver. */ - drbd_thread_stop(&mdev->receiver); - drbd_mdev_cleanup(mdev); - - dev_info(DEV, "worker terminated\n"); - - clear_bit(DEVICE_DYING, &mdev->flags); - clear_bit(CONFIG_PENDING, &mdev->flags); - wake_up(&mdev->state_wait); + rcu_read_unlock(); return 0; } diff --git a/drivers/block/drbd/drbd_wrappers.h b/drivers/block/drbd/drbd_wrappers.h index 151f1a37478f..328f18e4b4ee 100644 --- a/drivers/block/drbd/drbd_wrappers.h +++ b/drivers/block/drbd/drbd_wrappers.h @@ -3,6 +3,7 @@ #include <linux/ctype.h> #include <linux/mm.h> +#include "drbd_int.h" /* see get_sb_bdev and bd_claim */ extern char *drbd_sec_holder; @@ -20,8 +21,8 @@ static inline void drbd_set_my_capacity(struct drbd_conf *mdev, /* bi_end_io handlers */ extern void drbd_md_io_complete(struct bio *bio, int error); -extern void drbd_endio_sec(struct bio *bio, int error); -extern void drbd_endio_pri(struct bio *bio, int error); +extern void drbd_peer_request_endio(struct bio *bio, int error); +extern void drbd_request_endio(struct bio *bio, int error); /* * used to submit our private bio @@ -45,12 +46,6 @@ static inline void drbd_generic_make_request(struct drbd_conf *mdev, generic_make_request(bio); } -static inline int drbd_crypto_is_hash(struct crypto_tfm *tfm) -{ - return (crypto_tfm_alg_type(tfm) & CRYPTO_ALG_TYPE_HASH_MASK) - == CRYPTO_ALG_TYPE_HASH; -} - #ifndef __CHECKER__ # undef __cond_lock # define __cond_lock(x,c) (c) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 54046e51160a..ae1251270624 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -463,6 +463,7 @@ out: */ static void loop_add_bio(struct loop_device *lo, struct bio *bio) { + lo->lo_bio_count++; bio_list_add(&lo->lo_bio_list, bio); } @@ -471,6 +472,7 @@ static void loop_add_bio(struct loop_device *lo, struct bio *bio) */ static struct bio *loop_get_bio(struct loop_device *lo) { + lo->lo_bio_count--; return bio_list_pop(&lo->lo_bio_list); } @@ -489,6 +491,10 @@ static void loop_make_request(struct request_queue *q, struct bio *old_bio) goto out; if (unlikely(rw == WRITE && (lo->lo_flags & LO_FLAGS_READ_ONLY))) goto out; + if (lo->lo_bio_count >= q->nr_congestion_on) + wait_event_lock_irq(lo->lo_req_wait, + lo->lo_bio_count < q->nr_congestion_off, + lo->lo_lock); loop_add_bio(lo, old_bio); wake_up(&lo->lo_event); spin_unlock_irq(&lo->lo_lock); @@ -546,6 +552,8 @@ static int loop_thread(void *data) continue; spin_lock_irq(&lo->lo_lock); bio = loop_get_bio(lo); + if (lo->lo_bio_count < lo->lo_queue->nr_congestion_off) + wake_up(&lo->lo_req_wait); spin_unlock_irq(&lo->lo_lock); BUG_ON(!bio); @@ -873,6 +881,7 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, lo->transfer = transfer_none; lo->ioctl = NULL; lo->lo_sizelimit = 0; + lo->lo_bio_count = 0; lo->old_gfp_mask = mapping_gfp_mask(mapping); mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS)); @@ -1673,6 +1682,7 @@ static int loop_add(struct loop_device **l, int i) lo->lo_number = i; lo->lo_thread = NULL; init_waitqueue_head(&lo->lo_event); + init_waitqueue_head(&lo->lo_req_wait); spin_lock_init(&lo->lo_lock); disk->major = LOOP_MAJOR; disk->first_minor = i << part_shift; diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 280a13846e6c..74374fb762aa 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -39,6 +39,7 @@ #include <linux/list.h> #include <linux/delay.h> #include <linux/freezer.h> +#include <linux/bitmap.h> #include <xen/events.h> #include <xen/page.h> @@ -79,6 +80,7 @@ struct pending_req { unsigned short operation; int status; struct list_head free_list; + DECLARE_BITMAP(unmap_seg, BLKIF_MAX_SEGMENTS_PER_REQUEST); }; #define BLKBACK_INVALID_HANDLE (~0) @@ -99,6 +101,36 @@ struct xen_blkbk { static struct xen_blkbk *blkbk; /* + * Maximum number of grant pages that can be mapped in blkback. + * BLKIF_MAX_SEGMENTS_PER_REQUEST * RING_SIZE is the maximum number of + * pages that blkback will persistently map. + * Currently, this is: + * RING_SIZE = 32 (for all known ring types) + * BLKIF_MAX_SEGMENTS_PER_REQUEST = 11 + * sizeof(struct persistent_gnt) = 48 + * So the maximum memory used to store the grants is: + * 32 * 11 * 48 = 16896 bytes + */ +static inline unsigned int max_mapped_grant_pages(enum blkif_protocol protocol) +{ + switch (protocol) { + case BLKIF_PROTOCOL_NATIVE: + return __CONST_RING_SIZE(blkif, PAGE_SIZE) * + BLKIF_MAX_SEGMENTS_PER_REQUEST; + case BLKIF_PROTOCOL_X86_32: + return __CONST_RING_SIZE(blkif_x86_32, PAGE_SIZE) * + BLKIF_MAX_SEGMENTS_PER_REQUEST; + case BLKIF_PROTOCOL_X86_64: + return __CONST_RING_SIZE(blkif_x86_64, PAGE_SIZE) * + BLKIF_MAX_SEGMENTS_PER_REQUEST; + default: + BUG(); + } + return 0; +} + + +/* * Little helpful macro to figure out the index and virtual address of the * pending_pages[..]. For each 'pending_req' we have have up to * BLKIF_MAX_SEGMENTS_PER_REQUEST (11) pages. The seg would be from 0 through @@ -129,6 +161,90 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif, static void make_response(struct xen_blkif *blkif, u64 id, unsigned short op, int st); +#define foreach_grant(pos, rbtree, node) \ + for ((pos) = container_of(rb_first((rbtree)), typeof(*(pos)), node); \ + &(pos)->node != NULL; \ + (pos) = container_of(rb_next(&(pos)->node), typeof(*(pos)), node)) + + +static void add_persistent_gnt(struct rb_root *root, + struct persistent_gnt *persistent_gnt) +{ + struct rb_node **new = &(root->rb_node), *parent = NULL; + struct persistent_gnt *this; + + /* Figure out where to put new node */ + while (*new) { + this = container_of(*new, struct persistent_gnt, node); + + parent = *new; + if (persistent_gnt->gnt < this->gnt) + new = &((*new)->rb_left); + else if (persistent_gnt->gnt > this->gnt) + new = &((*new)->rb_right); + else { + pr_alert(DRV_PFX " trying to add a gref that's already in the tree\n"); + BUG(); + } + } + + /* Add new node and rebalance tree. */ + rb_link_node(&(persistent_gnt->node), parent, new); + rb_insert_color(&(persistent_gnt->node), root); +} + +static struct persistent_gnt *get_persistent_gnt(struct rb_root *root, + grant_ref_t gref) +{ + struct persistent_gnt *data; + struct rb_node *node = root->rb_node; + + while (node) { + data = container_of(node, struct persistent_gnt, node); + + if (gref < data->gnt) + node = node->rb_left; + else if (gref > data->gnt) + node = node->rb_right; + else + return data; + } + return NULL; +} + +static void free_persistent_gnts(struct rb_root *root, unsigned int num) +{ + struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + struct page *pages[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + struct persistent_gnt *persistent_gnt; + int ret = 0; + int segs_to_unmap = 0; + + foreach_grant(persistent_gnt, root, node) { + BUG_ON(persistent_gnt->handle == + BLKBACK_INVALID_HANDLE); + gnttab_set_unmap_op(&unmap[segs_to_unmap], + (unsigned long) pfn_to_kaddr(page_to_pfn( + persistent_gnt->page)), + GNTMAP_host_map, + persistent_gnt->handle); + + pages[segs_to_unmap] = persistent_gnt->page; + rb_erase(&persistent_gnt->node, root); + kfree(persistent_gnt); + num--; + + if (++segs_to_unmap == BLKIF_MAX_SEGMENTS_PER_REQUEST || + !rb_next(&persistent_gnt->node)) { + ret = gnttab_unmap_refs(unmap, NULL, pages, + segs_to_unmap); + BUG_ON(ret); + segs_to_unmap = 0; + } + } + BUG_ON(num != 0); +} + /* * Retrieve from the 'pending_reqs' a free pending_req structure to be used. */ @@ -302,6 +418,14 @@ int xen_blkif_schedule(void *arg) print_stats(blkif); } + /* Free all persistent grant pages */ + if (!RB_EMPTY_ROOT(&blkif->persistent_gnts)) + free_persistent_gnts(&blkif->persistent_gnts, + blkif->persistent_gnt_c); + + BUG_ON(!RB_EMPTY_ROOT(&blkif->persistent_gnts)); + blkif->persistent_gnt_c = 0; + if (log_stats) print_stats(blkif); @@ -328,6 +452,8 @@ static void xen_blkbk_unmap(struct pending_req *req) int ret; for (i = 0; i < req->nr_pages; i++) { + if (!test_bit(i, req->unmap_seg)) + continue; handle = pending_handle(req, i); if (handle == BLKBACK_INVALID_HANDLE) continue; @@ -344,12 +470,26 @@ static void xen_blkbk_unmap(struct pending_req *req) static int xen_blkbk_map(struct blkif_request *req, struct pending_req *pending_req, - struct seg_buf seg[]) + struct seg_buf seg[], + struct page *pages[]) { struct gnttab_map_grant_ref map[BLKIF_MAX_SEGMENTS_PER_REQUEST]; - int i; + struct persistent_gnt *persistent_gnts[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + struct page *pages_to_gnt[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + struct persistent_gnt *persistent_gnt = NULL; + struct xen_blkif *blkif = pending_req->blkif; + phys_addr_t addr = 0; + int i, j; + bool new_map; int nseg = req->u.rw.nr_segments; + int segs_to_map = 0; int ret = 0; + int use_persistent_gnts; + + use_persistent_gnts = (blkif->vbd.feature_gnt_persistent); + + BUG_ON(blkif->persistent_gnt_c > + max_mapped_grant_pages(pending_req->blkif->blk_protocol)); /* * Fill out preq.nr_sects with proper amount of sectors, and setup @@ -359,36 +499,146 @@ static int xen_blkbk_map(struct blkif_request *req, for (i = 0; i < nseg; i++) { uint32_t flags; - flags = GNTMAP_host_map; - if (pending_req->operation != BLKIF_OP_READ) - flags |= GNTMAP_readonly; - gnttab_set_map_op(&map[i], vaddr(pending_req, i), flags, - req->u.rw.seg[i].gref, - pending_req->blkif->domid); + if (use_persistent_gnts) + persistent_gnt = get_persistent_gnt( + &blkif->persistent_gnts, + req->u.rw.seg[i].gref); + + if (persistent_gnt) { + /* + * We are using persistent grants and + * the grant is already mapped + */ + new_map = false; + } else if (use_persistent_gnts && + blkif->persistent_gnt_c < + max_mapped_grant_pages(blkif->blk_protocol)) { + /* + * We are using persistent grants, the grant is + * not mapped but we have room for it + */ + new_map = true; + persistent_gnt = kmalloc( + sizeof(struct persistent_gnt), + GFP_KERNEL); + if (!persistent_gnt) + return -ENOMEM; + persistent_gnt->page = alloc_page(GFP_KERNEL); + if (!persistent_gnt->page) { + kfree(persistent_gnt); + return -ENOMEM; + } + persistent_gnt->gnt = req->u.rw.seg[i].gref; + persistent_gnt->handle = BLKBACK_INVALID_HANDLE; + + pages_to_gnt[segs_to_map] = + persistent_gnt->page; + addr = (unsigned long) pfn_to_kaddr( + page_to_pfn(persistent_gnt->page)); + + add_persistent_gnt(&blkif->persistent_gnts, + persistent_gnt); + blkif->persistent_gnt_c++; + pr_debug(DRV_PFX " grant %u added to the tree of persistent grants, using %u/%u\n", + persistent_gnt->gnt, blkif->persistent_gnt_c, + max_mapped_grant_pages(blkif->blk_protocol)); + } else { + /* + * We are either using persistent grants and + * hit the maximum limit of grants mapped, + * or we are not using persistent grants. + */ + if (use_persistent_gnts && + !blkif->vbd.overflow_max_grants) { + blkif->vbd.overflow_max_grants = 1; + pr_alert(DRV_PFX " domain %u, device %#x is using maximum number of persistent grants\n", + blkif->domid, blkif->vbd.handle); + } + new_map = true; + pages[i] = blkbk->pending_page(pending_req, i); + addr = vaddr(pending_req, i); + pages_to_gnt[segs_to_map] = + blkbk->pending_page(pending_req, i); + } + + if (persistent_gnt) { + pages[i] = persistent_gnt->page; + persistent_gnts[i] = persistent_gnt; + } else { + persistent_gnts[i] = NULL; + } + + if (new_map) { + flags = GNTMAP_host_map; + if (!persistent_gnt && + (pending_req->operation != BLKIF_OP_READ)) + flags |= GNTMAP_readonly; + gnttab_set_map_op(&map[segs_to_map++], addr, + flags, req->u.rw.seg[i].gref, + blkif->domid); + } } - ret = gnttab_map_refs(map, NULL, &blkbk->pending_page(pending_req, 0), nseg); - BUG_ON(ret); + if (segs_to_map) { + ret = gnttab_map_refs(map, NULL, pages_to_gnt, segs_to_map); + BUG_ON(ret); + } /* * Now swizzle the MFN in our domain with the MFN from the other domain * so that when we access vaddr(pending_req,i) it has the contents of * the page from the other domain. */ - for (i = 0; i < nseg; i++) { - if (unlikely(map[i].status != 0)) { - pr_debug(DRV_PFX "invalid buffer -- could not remap it\n"); - map[i].handle = BLKBACK_INVALID_HANDLE; - ret |= 1; + bitmap_zero(pending_req->unmap_seg, BLKIF_MAX_SEGMENTS_PER_REQUEST); + for (i = 0, j = 0; i < nseg; i++) { + if (!persistent_gnts[i] || + persistent_gnts[i]->handle == BLKBACK_INVALID_HANDLE) { + /* This is a newly mapped grant */ + BUG_ON(j >= segs_to_map); + if (unlikely(map[j].status != 0)) { + pr_debug(DRV_PFX "invalid buffer -- could not remap it\n"); + map[j].handle = BLKBACK_INVALID_HANDLE; + ret |= 1; + if (persistent_gnts[i]) { + rb_erase(&persistent_gnts[i]->node, + &blkif->persistent_gnts); + blkif->persistent_gnt_c--; + kfree(persistent_gnts[i]); + persistent_gnts[i] = NULL; + } + } + } + if (persistent_gnts[i]) { + if (persistent_gnts[i]->handle == + BLKBACK_INVALID_HANDLE) { + /* + * If this is a new persistent grant + * save the handler + */ + persistent_gnts[i]->handle = map[j].handle; + persistent_gnts[i]->dev_bus_addr = + map[j++].dev_bus_addr; + } + pending_handle(pending_req, i) = + persistent_gnts[i]->handle; + + if (ret) + continue; + + seg[i].buf = persistent_gnts[i]->dev_bus_addr | + (req->u.rw.seg[i].first_sect << 9); + } else { + pending_handle(pending_req, i) = map[j].handle; + bitmap_set(pending_req->unmap_seg, i, 1); + + if (ret) { + j++; + continue; + } + + seg[i].buf = map[j++].dev_bus_addr | + (req->u.rw.seg[i].first_sect << 9); } - - pending_handle(pending_req, i) = map[i].handle; - - if (ret) - continue; - - seg[i].buf = map[i].dev_bus_addr | - (req->u.rw.seg[i].first_sect << 9); } return ret; } @@ -591,6 +841,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif, int operation; struct blk_plug plug; bool drain = false; + struct page *pages[BLKIF_MAX_SEGMENTS_PER_REQUEST]; switch (req->operation) { case BLKIF_OP_READ: @@ -677,7 +928,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif, * the hypercall to unmap the grants - that is all done in * xen_blkbk_unmap. */ - if (xen_blkbk_map(req, pending_req, seg)) + if (xen_blkbk_map(req, pending_req, seg, pages)) goto fail_flush; /* @@ -689,7 +940,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif, for (i = 0; i < nseg; i++) { while ((bio == NULL) || (bio_add_page(bio, - blkbk->pending_page(pending_req, i), + pages[i], seg[i].nsec << 9, seg[i].buf & ~PAGE_MASK) == 0)) { diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h index 9a54623e52d7..6072390c7f57 100644 --- a/drivers/block/xen-blkback/common.h +++ b/drivers/block/xen-blkback/common.h @@ -34,6 +34,7 @@ #include <linux/vmalloc.h> #include <linux/wait.h> #include <linux/io.h> +#include <linux/rbtree.h> #include <asm/setup.h> #include <asm/pgalloc.h> #include <asm/hypervisor.h> @@ -160,10 +161,21 @@ struct xen_vbd { sector_t size; unsigned int flush_support:1; unsigned int discard_secure:1; + unsigned int feature_gnt_persistent:1; + unsigned int overflow_max_grants:1; }; struct backend_info; + +struct persistent_gnt { + struct page *page; + grant_ref_t gnt; + grant_handle_t handle; + uint64_t dev_bus_addr; + struct rb_node node; +}; + struct xen_blkif { /* Unique identifier for this interface. */ domid_t domid; @@ -190,6 +202,10 @@ struct xen_blkif { struct task_struct *xenblkd; unsigned int waiting_reqs; + /* tree to store persistent grants */ + struct rb_root persistent_gnts; + unsigned int persistent_gnt_c; + /* statistics */ unsigned long st_print; int st_rd_req; diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c index f58434c2617c..63980722db41 100644 --- a/drivers/block/xen-blkback/xenbus.c +++ b/drivers/block/xen-blkback/xenbus.c @@ -117,6 +117,7 @@ static struct xen_blkif *xen_blkif_alloc(domid_t domid) atomic_set(&blkif->drain, 0); blkif->st_print = jiffies; init_waitqueue_head(&blkif->waiting_to_free); + blkif->persistent_gnts.rb_node = NULL; return blkif; } @@ -672,6 +673,13 @@ again: xen_blkbk_barrier(xbt, be, be->blkif->vbd.flush_support); + err = xenbus_printf(xbt, dev->nodename, "feature-persistent", "%u", 1); + if (err) { + xenbus_dev_fatal(dev, err, "writing %s/feature-persistent", + dev->nodename); + goto abort; + } + err = xenbus_printf(xbt, dev->nodename, "sectors", "%llu", (unsigned long long)vbd_sz(&be->blkif->vbd)); if (err) { @@ -720,6 +728,7 @@ static int connect_ring(struct backend_info *be) struct xenbus_device *dev = be->dev; unsigned long ring_ref; unsigned int evtchn; + unsigned int pers_grants; char protocol[64] = ""; int err; @@ -749,8 +758,18 @@ static int connect_ring(struct backend_info *be) xenbus_dev_fatal(dev, err, "unknown fe protocol %s", protocol); return -1; } - pr_info(DRV_PFX "ring-ref %ld, event-channel %d, protocol %d (%s)\n", - ring_ref, evtchn, be->blkif->blk_protocol, protocol); + err = xenbus_gather(XBT_NIL, dev->otherend, + "feature-persistent", "%u", + &pers_grants, NULL); + if (err) + pers_grants = 0; + + be->blkif->vbd.feature_gnt_persistent = pers_grants; + be->blkif->vbd.overflow_max_grants = 0; + + pr_info(DRV_PFX "ring-ref %ld, event-channel %d, protocol %d (%s) %s\n", + ring_ref, evtchn, be->blkif->blk_protocol, protocol, + pers_grants ? "persistent grants" : ""); /* Map the shared frame, irq etc. */ err = xen_blkif_map(be->blkif, ring_ref, evtchn); diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 007db8986e84..96e9b00db081 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -44,6 +44,7 @@ #include <linux/mutex.h> #include <linux/scatterlist.h> #include <linux/bitmap.h> +#include <linux/llist.h> #include <xen/xen.h> #include <xen/xenbus.h> @@ -64,10 +65,17 @@ enum blkif_state { BLKIF_STATE_SUSPENDED, }; +struct grant { + grant_ref_t gref; + unsigned long pfn; + struct llist_node node; +}; + struct blk_shadow { struct blkif_request req; struct request *request; unsigned long frame[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + struct grant *grants_used[BLKIF_MAX_SEGMENTS_PER_REQUEST]; }; static DEFINE_MUTEX(blkfront_mutex); @@ -97,6 +105,8 @@ struct blkfront_info struct work_struct work; struct gnttab_free_callback callback; struct blk_shadow shadow[BLK_RING_SIZE]; + struct llist_head persistent_gnts; + unsigned int persistent_gnts_c; unsigned long shadow_free; unsigned int feature_flush; unsigned int flush_op; @@ -104,6 +114,7 @@ struct blkfront_info unsigned int feature_secdiscard:1; unsigned int discard_granularity; unsigned int discard_alignment; + unsigned int feature_persistent:1; int is_ready; }; @@ -287,21 +298,36 @@ static int blkif_queue_request(struct request *req) unsigned long id; unsigned int fsect, lsect; int i, ref; + + /* + * Used to store if we are able to queue the request by just using + * existing persistent grants, or if we have to get new grants, + * as there are not sufficiently many free. + */ + bool new_persistent_gnts; grant_ref_t gref_head; + struct page *granted_page; + struct grant *gnt_list_entry = NULL; struct scatterlist *sg; if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) return 1; - if (gnttab_alloc_grant_references( - BLKIF_MAX_SEGMENTS_PER_REQUEST, &gref_head) < 0) { - gnttab_request_free_callback( - &info->callback, - blkif_restart_queue_callback, - info, - BLKIF_MAX_SEGMENTS_PER_REQUEST); - return 1; - } + /* Check if we have enought grants to allocate a requests */ + if (info->persistent_gnts_c < BLKIF_MAX_SEGMENTS_PER_REQUEST) { + new_persistent_gnts = 1; + if (gnttab_alloc_grant_references( + BLKIF_MAX_SEGMENTS_PER_REQUEST - info->persistent_gnts_c, + &gref_head) < 0) { + gnttab_request_free_callback( + &info->callback, + blkif_restart_queue_callback, + info, + BLKIF_MAX_SEGMENTS_PER_REQUEST); + return 1; + } + } else + new_persistent_gnts = 0; /* Fill out a communications ring structure. */ ring_req = RING_GET_REQUEST(&info->ring, info->ring.req_prod_pvt); @@ -341,18 +367,73 @@ static int blkif_queue_request(struct request *req) BLKIF_MAX_SEGMENTS_PER_REQUEST); for_each_sg(info->sg, sg, ring_req->u.rw.nr_segments, i) { - buffer_mfn = pfn_to_mfn(page_to_pfn(sg_page(sg))); fsect = sg->offset >> 9; lsect = fsect + (sg->length >> 9) - 1; - /* install a grant reference. */ - ref = gnttab_claim_grant_reference(&gref_head); - BUG_ON(ref == -ENOSPC); - gnttab_grant_foreign_access_ref( - ref, + if (info->persistent_gnts_c) { + BUG_ON(llist_empty(&info->persistent_gnts)); + gnt_list_entry = llist_entry( + llist_del_first(&info->persistent_gnts), + struct grant, node); + + ref = gnt_list_entry->gref; + buffer_mfn = pfn_to_mfn(gnt_list_entry->pfn); + info->persistent_gnts_c--; + } else { + ref = gnttab_claim_grant_reference(&gref_head); + BUG_ON(ref == -ENOSPC); + + gnt_list_entry = + kmalloc(sizeof(struct grant), + GFP_ATOMIC); + if (!gnt_list_entry) + return -ENOMEM; + + granted_page = alloc_page(GFP_ATOMIC); + if (!granted_page) { + kfree(gnt_list_entry); + return -ENOMEM; + } + + gnt_list_entry->pfn = + page_to_pfn(granted_page); + gnt_list_entry->gref = ref; + + buffer_mfn = pfn_to_mfn(page_to_pfn( + granted_page)); + gnttab_grant_foreign_access_ref(ref, info->xbdev->otherend_id, - buffer_mfn, - rq_data_dir(req)); + buffer_mfn, 0); + } + + info->shadow[id].grants_used[i] = gnt_list_entry; + + if (rq_data_dir(req)) { + char *bvec_data; + void *shared_data; + + BUG_ON(sg->offset + sg->length > PAGE_SIZE); + + shared_data = kmap_atomic( + pfn_to_page(gnt_list_entry->pfn)); + bvec_data = kmap_atomic(sg_page(sg)); + + /* + * this does not wipe data stored outside the + * range sg->offset..sg->offset+sg->length. + * Therefore, blkback *could* see data from + * previous requests. This is OK as long as + * persistent grants are shared with just one + * domain. It may need refactoring if this + * changes + */ + memcpy(shared_data + sg->offset, + bvec_data + sg->offset, + sg->length); + + kunmap_atomic(bvec_data); + kunmap_atomic(shared_data); + } info->shadow[id].frame[i] = mfn_to_pfn(buffer_mfn); ring_req->u.rw.seg[i] = @@ -368,7 +449,8 @@ static int blkif_queue_request(struct request *req) /* Keep a private copy so we can reissue requests when recovering. */ info->shadow[id].req = *ring_req; - gnttab_free_grant_references(gref_head); + if (new_persistent_gnts) + gnttab_free_grant_references(gref_head); return 0; } @@ -480,12 +562,13 @@ static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size) static void xlvbd_flush(struct blkfront_info *info) { blk_queue_flush(info->rq, info->feature_flush); - printk(KERN_INFO "blkfront: %s: %s: %s\n", + printk(KERN_INFO "blkfront: %s: %s: %s %s\n", info->gd->disk_name, info->flush_op == BLKIF_OP_WRITE_BARRIER ? "barrier" : (info->flush_op == BLKIF_OP_FLUSH_DISKCACHE ? "flush diskcache" : "barrier or flush"), - info->feature_flush ? "enabled" : "disabled"); + info->feature_flush ? "enabled" : "disabled", + info->feature_persistent ? "using persistent grants" : ""); } static int xen_translate_vdev(int vdevice, int *minor, unsigned int *offset) @@ -707,6 +790,9 @@ static void blkif_restart_queue(struct work_struct *work) static void blkif_free(struct blkfront_info *info, int suspend) { + struct llist_node *all_gnts; + struct grant *persistent_gnt; + /* Prevent new requests being issued until we fix things up. */ spin_lock_irq(&info->io_lock); info->connected = suspend ? @@ -714,6 +800,18 @@ static void blkif_free(struct blkfront_info *info, int suspend) /* No more blkif_request(). */ if (info->rq) blk_stop_queue(info->rq); + + /* Remove all persistent grants */ + if (info->persistent_gnts_c) { + all_gnts = llist_del_all(&info->persistent_gnts); + llist_for_each_entry(persistent_gnt, all_gnts, node) { + gnttab_end_foreign_access(persistent_gnt->gref, 0, 0UL); + __free_page(pfn_to_page(persistent_gnt->pfn)); + kfree(persistent_gnt); + } + info->persistent_gnts_c = 0; + } + /* No more gnttab callback work. */ gnttab_cancel_free_callback(&info->callback); spin_unlock_irq(&info->io_lock); @@ -734,13 +832,43 @@ static void blkif_free(struct blkfront_info *info, int suspend) } -static void blkif_completion(struct blk_shadow *s) +static void blkif_completion(struct blk_shadow *s, struct blkfront_info *info, + struct blkif_response *bret) { int i; - /* Do not let BLKIF_OP_DISCARD as nr_segment is in the same place - * flag. */ - for (i = 0; i < s->req.u.rw.nr_segments; i++) - gnttab_end_foreign_access(s->req.u.rw.seg[i].gref, 0, 0UL); + struct bio_vec *bvec; + struct req_iterator iter; + unsigned long flags; + char *bvec_data; + void *shared_data; + unsigned int offset = 0; + + if (bret->operation == BLKIF_OP_READ) { + /* + * Copy the data received from the backend into the bvec. + * Since bv_offset can be different than 0, and bv_len different + * than PAGE_SIZE, we have to keep track of the current offset, + * to be sure we are copying the data from the right shared page. + */ + rq_for_each_segment(bvec, s->request, iter) { + BUG_ON((bvec->bv_offset + bvec->bv_len) > PAGE_SIZE); + i = offset >> PAGE_SHIFT; + BUG_ON(i >= s->req.u.rw.nr_segments); + shared_data = kmap_atomic( + pfn_to_page(s->grants_used[i]->pfn)); + bvec_data = bvec_kmap_irq(bvec, &flags); + memcpy(bvec_data, shared_data + bvec->bv_offset, + bvec->bv_len); + bvec_kunmap_irq(bvec_data, &flags); + kunmap_atomic(shared_data); + offset += bvec->bv_len; + } + } + /* Add the persistent grant into the list of free grants */ + for (i = 0; i < s->req.u.rw.nr_segments; i++) { + llist_add(&s->grants_used[i]->node, &info->persistent_gnts); + info->persistent_gnts_c++; + } } static irqreturn_t blkif_interrupt(int irq, void *dev_id) @@ -783,7 +911,7 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) req = info->shadow[id].request; if (bret->operation != BLKIF_OP_DISCARD) - blkif_completion(&info->shadow[id]); + blkif_completion(&info->shadow[id], info, bret); if (add_id_to_freelist(info, id)) { WARN(1, "%s: response to %s (id %ld) couldn't be recycled!\n", @@ -942,6 +1070,11 @@ again: message = "writing protocol"; goto abort_transaction; } + err = xenbus_printf(xbt, dev->nodename, + "feature-persistent", "%u", 1); + if (err) + dev_warn(&dev->dev, + "writing persistent grants feature to xenbus"); err = xenbus_transaction_end(xbt, 0); if (err) { @@ -1029,6 +1162,8 @@ static int blkfront_probe(struct xenbus_device *dev, spin_lock_init(&info->io_lock); info->xbdev = dev; info->vdevice = vdevice; + init_llist_head(&info->persistent_gnts); + info->persistent_gnts_c = 0; info->connected = BLKIF_STATE_DISCONNECTED; INIT_WORK(&info->work, blkif_restart_queue); @@ -1093,7 +1228,7 @@ static int blkif_recover(struct blkfront_info *info) req->u.rw.seg[j].gref, info->xbdev->otherend_id, pfn_to_mfn(info->shadow[req->u.rw.id].frame[j]), - rq_data_dir(info->shadow[req->u.rw.id].request)); + 0); } info->shadow[req->u.rw.id].req = *req; @@ -1225,7 +1360,7 @@ static void blkfront_connect(struct blkfront_info *info) unsigned long sector_size; unsigned int binfo; int err; - int barrier, flush, discard; + int barrier, flush, discard, persistent; switch (info->connected) { case BLKIF_STATE_CONNECTED: @@ -1303,6 +1438,14 @@ static void blkfront_connect(struct blkfront_info *info) if (!err && discard) blkfront_setup_discard(info); + err = xenbus_gather(XBT_NIL, info->xbdev->otherend, + "feature-persistent", "%u", &persistent, + NULL); + if (err) + info->feature_persistent = 0; + else + info->feature_persistent = persistent; + err = xlvbd_alloc_gendisk(sectors, info, binfo, sector_size); if (err) { xenbus_dev_fatal(info->xbdev, err, "xlvbd_add at %s", diff --git a/drivers/char/random.c b/drivers/char/random.c index b86eae9b77df..85e81ec1451e 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -399,7 +399,6 @@ static DECLARE_WAIT_QUEUE_HEAD(random_read_wait); static DECLARE_WAIT_QUEUE_HEAD(random_write_wait); static struct fasync_struct *fasync; -#if 0 static bool debug; module_param(debug, bool, 0644); #define DEBUG_ENT(fmt, arg...) do { \ @@ -410,9 +409,6 @@ module_param(debug, bool, 0644); blocking_pool.entropy_count,\ nonblocking_pool.entropy_count,\ ## arg); } while (0) -#else -#define DEBUG_ENT(fmt, arg...) do {} while (0) -#endif /********************************************************************** * @@ -437,6 +433,7 @@ struct entropy_store { int entropy_count; int entropy_total; unsigned int initialized:1; + bool last_data_init; __u8 last_data[EXTRACT_SIZE]; }; @@ -829,7 +826,7 @@ static void xfer_secondary_pool(struct entropy_store *r, size_t nbytes) bytes = min_t(int, bytes, sizeof(tmp)); DEBUG_ENT("going to reseed %s with %d bits " - "(%d of %d requested)\n", + "(%zu of %d requested)\n", r->name, bytes * 8, nbytes * 8, r->entropy_count); bytes = extract_entropy(r->pull, tmp, bytes, @@ -860,7 +857,7 @@ static size_t account(struct entropy_store *r, size_t nbytes, int min, spin_lock_irqsave(&r->lock, flags); BUG_ON(r->entropy_count > r->poolinfo->POOLBITS); - DEBUG_ENT("trying to extract %d bits from %s\n", + DEBUG_ENT("trying to extract %zu bits from %s\n", nbytes * 8, r->name); /* Can we pull enough? */ @@ -882,7 +879,7 @@ static size_t account(struct entropy_store *r, size_t nbytes, int min, } } - DEBUG_ENT("debiting %d entropy credits from %s%s\n", + DEBUG_ENT("debiting %zu entropy credits from %s%s\n", nbytes * 8, r->name, r->limit ? "" : " (unlimited)"); spin_unlock_irqrestore(&r->lock, flags); @@ -957,6 +954,10 @@ static ssize_t extract_entropy(struct entropy_store *r, void *buf, ssize_t ret = 0, i; __u8 tmp[EXTRACT_SIZE]; + /* if last_data isn't primed, we need EXTRACT_SIZE extra bytes */ + if (fips_enabled && !r->last_data_init) + nbytes += EXTRACT_SIZE; + trace_extract_entropy(r->name, nbytes, r->entropy_count, _RET_IP_); xfer_secondary_pool(r, nbytes); nbytes = account(r, nbytes, min, reserved); @@ -967,6 +968,17 @@ static ssize_t extract_entropy(struct entropy_store *r, void *buf, if (fips_enabled) { unsigned long flags; + + /* prime last_data value if need be, per fips 140-2 */ + if (!r->last_data_init) { + spin_lock_irqsave(&r->lock, flags); + memcpy(r->last_data, tmp, EXTRACT_SIZE); + r->last_data_init = true; + nbytes -= EXTRACT_SIZE; + spin_unlock_irqrestore(&r->lock, flags); + extract_buf(r, tmp); + } + spin_lock_irqsave(&r->lock, flags); if (!memcmp(tmp, r->last_data, EXTRACT_SIZE)) panic("Hardware RNG duplicated output!\n"); @@ -1086,6 +1098,7 @@ static void init_std_data(struct entropy_store *r) r->entropy_count = 0; r->entropy_total = 0; + r->last_data_init = false; mix_pool_bytes(r, &now, sizeof(now), NULL); for (i = r->poolinfo->POOLBYTES; i > 0; i -= sizeof(rv)) { if (!arch_get_random_long(&rv)) @@ -1142,11 +1155,16 @@ random_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos) if (n > SEC_XFER_SIZE) n = SEC_XFER_SIZE; - DEBUG_ENT("reading %d bits\n", n*8); + DEBUG_ENT("reading %zu bits\n", n*8); n = extract_entropy_user(&blocking_pool, buf, n); - DEBUG_ENT("read got %d bits (%d still needed)\n", + if (n < 0) { + retval = n; + break; + } + + DEBUG_ENT("read got %zd bits (%zd still needed)\n", n*8, (nbytes-n)*8); if (n == 0) { @@ -1171,10 +1189,6 @@ random_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos) continue; } - if (n < 0) { - retval = n; - break; - } count += n; buf += n; nbytes -= n; diff --git a/drivers/clk/clk-nomadik.c b/drivers/clk/clk-nomadik.c index 517a8ff7121e..6b4c70f7d23d 100644 --- a/drivers/clk/clk-nomadik.c +++ b/drivers/clk/clk-nomadik.c @@ -20,6 +20,7 @@ void __init nomadik_clk_init(void) clk_register_clkdev(clk, NULL, "gpio.2"); clk_register_clkdev(clk, NULL, "gpio.3"); clk_register_clkdev(clk, NULL, "rng"); + clk_register_clkdev(clk, NULL, "fsmc-nand"); /* * The 2.4 MHz TIMCLK reference clock is active at boot time, this is diff --git a/drivers/crypto/nx/nx-842.c b/drivers/crypto/nx/nx-842.c index 0ce625738677..6c4c000671c5 100644 --- a/drivers/crypto/nx/nx-842.c +++ b/drivers/crypto/nx/nx-842.c @@ -28,7 +28,6 @@ #include <linux/slab.h> #include <asm/page.h> -#include <asm/pSeries_reconfig.h> #include <asm/vio.h> #include "nx_csbcpb.h" /* struct nx_csbcpb */ @@ -1014,26 +1013,23 @@ error_out: * NOTIFY_BAD encoded with error number on failure, use * notifier_to_errno() to decode this value */ -static int nx842_OF_notifier(struct notifier_block *np, - unsigned long action, - void *update) +static int nx842_OF_notifier(struct notifier_block *np, unsigned long action, + void *update) { - struct pSeries_reconfig_prop_update *upd; + struct of_prop_reconfig *upd = update; struct nx842_devdata *local_devdata; struct device_node *node = NULL; - upd = (struct pSeries_reconfig_prop_update *)update; - rcu_read_lock(); local_devdata = rcu_dereference(devdata); if (local_devdata) node = local_devdata->dev->of_node; if (local_devdata && - action == PSERIES_UPDATE_PROPERTY && - !strcmp(upd->node->name, node->name)) { + action == OF_RECONFIG_UPDATE_PROPERTY && + !strcmp(upd->dn->name, node->name)) { rcu_read_unlock(); - nx842_OF_upd(upd->property); + nx842_OF_upd(upd->prop); } else rcu_read_unlock(); @@ -1182,7 +1178,7 @@ static int __init nx842_probe(struct vio_dev *viodev, synchronize_rcu(); kfree(old_devdata); - pSeries_reconfig_notifier_register(&nx842_of_nb); + of_reconfig_notifier_register(&nx842_of_nb); ret = nx842_OF_upd(NULL); if (ret && ret != -ENODEV) { @@ -1228,7 +1224,7 @@ static int __exit nx842_remove(struct vio_dev *viodev) spin_lock_irqsave(&devdata_mutex, flags); old_devdata = rcu_dereference_check(devdata, lockdep_is_held(&devdata_mutex)); - pSeries_reconfig_notifier_unregister(&nx842_of_nb); + of_reconfig_notifier_unregister(&nx842_of_nb); rcu_assign_pointer(devdata, NULL); spin_unlock_irqrestore(&devdata_mutex, flags); synchronize_rcu(); diff --git a/drivers/crypto/nx/nx.c b/drivers/crypto/nx/nx.c index 638110efae9b..f7a8a16aa7d3 100644 --- a/drivers/crypto/nx/nx.c +++ b/drivers/crypto/nx/nx.c @@ -33,7 +33,6 @@ #include <linux/scatterlist.h> #include <linux/device.h> #include <linux/of.h> -#include <asm/pSeries_reconfig.h> #include <asm/hvcall.h> #include <asm/vio.h> diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c index 24225f0fdcd8..64b048d7fba7 100644 --- a/drivers/dma/dmatest.c +++ b/drivers/dma/dmatest.c @@ -228,6 +228,20 @@ static void dmatest_callback(void *arg) wake_up_all(done->wait); } +static inline void unmap_src(struct device *dev, dma_addr_t *addr, size_t len, + unsigned int count) +{ + while (count--) + dma_unmap_single(dev, addr[count], len, DMA_TO_DEVICE); +} + +static inline void unmap_dst(struct device *dev, dma_addr_t *addr, size_t len, + unsigned int count) +{ + while (count--) + dma_unmap_single(dev, addr[count], len, DMA_BIDIRECTIONAL); +} + /* * This function repeatedly tests DMA transfers of various lengths and * offsets for a given operation type until it is told to exit by @@ -353,15 +367,35 @@ static int dmatest_func(void *data) dma_srcs[i] = dma_map_single(dev->dev, buf, len, DMA_TO_DEVICE); + ret = dma_mapping_error(dev->dev, dma_srcs[i]); + if (ret) { + unmap_src(dev->dev, dma_srcs, len, i); + pr_warn("%s: #%u: mapping error %d with " + "src_off=0x%x len=0x%x\n", + thread_name, total_tests - 1, ret, + src_off, len); + failed_tests++; + continue; + } } /* map with DMA_BIDIRECTIONAL to force writeback/invalidate */ for (i = 0; i < dst_cnt; i++) { dma_dsts[i] = dma_map_single(dev->dev, thread->dsts[i], test_buf_size, DMA_BIDIRECTIONAL); + ret = dma_mapping_error(dev->dev, dma_dsts[i]); + if (ret) { + unmap_src(dev->dev, dma_srcs, len, src_cnt); + unmap_dst(dev->dev, dma_dsts, test_buf_size, i); + pr_warn("%s: #%u: mapping error %d with " + "dst_off=0x%x len=0x%x\n", + thread_name, total_tests - 1, ret, + dst_off, test_buf_size); + failed_tests++; + continue; + } } - if (thread->type == DMA_MEMCPY) tx = dev->device_prep_dma_memcpy(chan, dma_dsts[0] + dst_off, @@ -383,13 +417,8 @@ static int dmatest_func(void *data) } if (!tx) { - for (i = 0; i < src_cnt; i++) - dma_unmap_single(dev->dev, dma_srcs[i], len, - DMA_TO_DEVICE); - for (i = 0; i < dst_cnt; i++) - dma_unmap_single(dev->dev, dma_dsts[i], - test_buf_size, - DMA_BIDIRECTIONAL); + unmap_src(dev->dev, dma_srcs, len, src_cnt); + unmap_dst(dev->dev, dma_dsts, test_buf_size, dst_cnt); pr_warning("%s: #%u: prep error with src_off=0x%x " "dst_off=0x%x len=0x%x\n", thread_name, total_tests - 1, @@ -443,9 +472,7 @@ static int dmatest_func(void *data) } /* Unmap by myself (see DMA_COMPL_SKIP_DEST_UNMAP above) */ - for (i = 0; i < dst_cnt; i++) - dma_unmap_single(dev->dev, dma_dsts[i], test_buf_size, - DMA_BIDIRECTIONAL); + unmap_dst(dev->dev, dma_dsts, test_buf_size, dst_cnt); error_count = 0; diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c index 52c5d8956d7d..7b1c37497c9a 100644 --- a/drivers/firmware/efivars.c +++ b/drivers/firmware/efivars.c @@ -883,7 +883,6 @@ static struct inode *efivarfs_get_inode(struct super_block *sb, if (inode) { inode->i_ino = get_next_ino(); - inode->i_uid = inode->i_gid = 0; inode->i_mode = mode; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch (mode & S_IFMT) { diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig index 8ae1f5b19669..682de754d63f 100644 --- a/drivers/gpio/Kconfig +++ b/drivers/gpio/Kconfig @@ -172,6 +172,7 @@ config GPIO_MSM_V2 config GPIO_MVEBU def_bool y depends on PLAT_ORION + depends on OF select GPIO_GENERIC select GENERIC_IRQ_CHIP diff --git a/drivers/gpio/gpio-ich.c b/drivers/gpio/gpio-ich.c index 6cc87ac8e019..6f2306db8591 100644 --- a/drivers/gpio/gpio-ich.c +++ b/drivers/gpio/gpio-ich.c @@ -390,6 +390,7 @@ static int ichx_gpio_probe(struct platform_device *pdev) return -ENODEV; } + spin_lock_init(&ichx_priv.lock); res_base = platform_get_resource(pdev, IORESOURCE_IO, ICH_RES_GPIO); ichx_priv.use_gpio = ich_info->use_gpio; err = ichx_gpio_request_regions(res_base, pdev->name, diff --git a/drivers/gpio/gpio-mvebu.c b/drivers/gpio/gpio-mvebu.c index d767b534c4af..7d9bd94be8d2 100644 --- a/drivers/gpio/gpio-mvebu.c +++ b/drivers/gpio/gpio-mvebu.c @@ -41,7 +41,6 @@ #include <linux/io.h> #include <linux/of_irq.h> #include <linux/of_device.h> -#include <linux/platform_device.h> #include <linux/pinctrl/consumer.h> /* @@ -469,19 +468,6 @@ static void mvebu_gpio_irq_handler(unsigned int irq, struct irq_desc *desc) } } -static struct platform_device_id mvebu_gpio_ids[] = { - { - .name = "orion-gpio", - }, { - .name = "mv78200-gpio", - }, { - .name = "armadaxp-gpio", - }, { - /* sentinel */ - }, -}; -MODULE_DEVICE_TABLE(platform, mvebu_gpio_ids); - static struct of_device_id mvebu_gpio_of_match[] = { { .compatible = "marvell,orion-gpio", @@ -555,9 +541,7 @@ static int mvebu_gpio_probe(struct platform_device *pdev) mvchip->chip.base = id * MVEBU_MAX_GPIO_PER_BANK; mvchip->chip.ngpio = ngpios; mvchip->chip.can_sleep = 0; -#ifdef CONFIG_OF mvchip->chip.of_node = np; -#endif spin_lock_init(&mvchip->lock); mvchip->membase = devm_request_and_ioremap(&pdev->dev, res); @@ -698,7 +682,6 @@ static struct platform_driver mvebu_gpio_driver = { .of_match_table = mvebu_gpio_of_match, }, .probe = mvebu_gpio_probe, - .id_table = mvebu_gpio_ids, }; static int __init mvebu_gpio_init(void) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index a9151337d5b9..33d20be87db5 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -579,7 +579,7 @@ static int ttm_bo_cleanup_refs_and_unlock(struct ttm_buffer_object *bo, * at this point the buffer should be dead, so * no new sync objects can be attached. */ - sync_obj = driver->sync_obj_ref(&bo->sync_obj); + sync_obj = driver->sync_obj_ref(bo->sync_obj); spin_unlock(&bdev->fence_lock); atomic_set(&bo->reserved, 0); diff --git a/drivers/hwmon/hwmon-vid.c b/drivers/hwmon/hwmon-vid.c index 9f26400713f0..89cfd64b3373 100644 --- a/drivers/hwmon/hwmon-vid.c +++ b/drivers/hwmon/hwmon-vid.c @@ -115,6 +115,12 @@ int vid_from_reg(int val, u8 vrm) return (val < 32) ? 1550 - 25 * val : 775 - (25 * (val - 31)) / 2; + case 26: /* AMD family 10h to 15h, serial VID */ + val &= 0x7f; + if (val >= 0x7c) + return 0; + return DIV_ROUND_CLOSEST(15500 - 125 * val, 10); + case 91: /* VRM 9.1 */ case 90: /* VRM 9.0 */ val &= 0x1f; @@ -195,6 +201,10 @@ static struct vrm_model vrm_models[] = { {X86_VENDOR_AMD, 0xF, 0x40, 0x7F, ANY, 24}, /* NPT family 0Fh */ {X86_VENDOR_AMD, 0xF, 0x80, ANY, ANY, 25}, /* future fam. 0Fh */ {X86_VENDOR_AMD, 0x10, 0x0, ANY, ANY, 25}, /* NPT family 10h */ + {X86_VENDOR_AMD, 0x11, 0x0, ANY, ANY, 26}, /* family 11h */ + {X86_VENDOR_AMD, 0x12, 0x0, ANY, ANY, 26}, /* family 12h */ + {X86_VENDOR_AMD, 0x14, 0x0, ANY, ANY, 26}, /* family 14h */ + {X86_VENDOR_AMD, 0x15, 0x0, ANY, ANY, 26}, /* family 15h */ {X86_VENDOR_INTEL, 0x6, 0x0, 0x6, ANY, 82}, /* Pentium Pro, * Pentium II, Xeon, diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c index c3c471ca202f..646314f7c839 100644 --- a/drivers/hwmon/hwmon.c +++ b/drivers/hwmon/hwmon.c @@ -84,19 +84,21 @@ static void __init hwmon_pci_quirks(void) /* Open access to 0x295-0x296 on MSI MS-7031 */ sb = pci_get_device(PCI_VENDOR_ID_ATI, 0x436c, NULL); - if (sb && - (sb->subsystem_vendor == 0x1462 && /* MSI */ - sb->subsystem_device == 0x0031)) { /* MS-7031 */ - - pci_read_config_byte(sb, 0x48, &enable); - pci_read_config_word(sb, 0x64, &base); - - if (base == 0 && !(enable & BIT(2))) { - dev_info(&sb->dev, - "Opening wide generic port at 0x295\n"); - pci_write_config_word(sb, 0x64, 0x295); - pci_write_config_byte(sb, 0x48, enable | BIT(2)); + if (sb) { + if (sb->subsystem_vendor == 0x1462 && /* MSI */ + sb->subsystem_device == 0x0031) { /* MS-7031 */ + pci_read_config_byte(sb, 0x48, &enable); + pci_read_config_word(sb, 0x64, &base); + + if (base == 0 && !(enable & BIT(2))) { + dev_info(&sb->dev, + "Opening wide generic port at 0x295\n"); + pci_write_config_word(sb, 0x64, 0x295); + pci_write_config_byte(sb, 0x48, + enable | BIT(2)); + } } + pci_dev_put(sb); } #endif } diff --git a/drivers/hwmon/it87.c b/drivers/hwmon/it87.c index d32aa354cbdf..117d66fcded6 100644 --- a/drivers/hwmon/it87.c +++ b/drivers/hwmon/it87.c @@ -203,6 +203,8 @@ static const u8 IT87_REG_FAN[] = { 0x0d, 0x0e, 0x0f, 0x80, 0x82 }; static const u8 IT87_REG_FAN_MIN[] = { 0x10, 0x11, 0x12, 0x84, 0x86 }; static const u8 IT87_REG_FANX[] = { 0x18, 0x19, 0x1a, 0x81, 0x83 }; static const u8 IT87_REG_FANX_MIN[] = { 0x1b, 0x1c, 0x1d, 0x85, 0x87 }; +static const u8 IT87_REG_TEMP_OFFSET[] = { 0x56, 0x57, 0x59 }; + #define IT87_REG_FAN_MAIN_CTRL 0x13 #define IT87_REG_FAN_CTL 0x14 #define IT87_REG_PWM(nr) (0x15 + (nr)) @@ -226,6 +228,83 @@ static const u8 IT87_REG_FANX_MIN[] = { 0x1b, 0x1c, 0x1d, 0x85, 0x87 }; #define IT87_REG_AUTO_TEMP(nr, i) (0x60 + (nr) * 8 + (i)) #define IT87_REG_AUTO_PWM(nr, i) (0x65 + (nr) * 8 + (i)) +struct it87_devices { + const char *name; + u16 features; + u8 peci_mask; + u8 old_peci_mask; +}; + +#define FEAT_12MV_ADC (1 << 0) +#define FEAT_NEWER_AUTOPWM (1 << 1) +#define FEAT_OLD_AUTOPWM (1 << 2) +#define FEAT_16BIT_FANS (1 << 3) +#define FEAT_TEMP_OFFSET (1 << 4) +#define FEAT_TEMP_PECI (1 << 5) +#define FEAT_TEMP_OLD_PECI (1 << 6) + +static const struct it87_devices it87_devices[] = { + [it87] = { + .name = "it87", + .features = FEAT_OLD_AUTOPWM, /* may need to overwrite */ + }, + [it8712] = { + .name = "it8712", + .features = FEAT_OLD_AUTOPWM, /* may need to overwrite */ + }, + [it8716] = { + .name = "it8716", + .features = FEAT_16BIT_FANS | FEAT_TEMP_OFFSET, + }, + [it8718] = { + .name = "it8718", + .features = FEAT_16BIT_FANS | FEAT_TEMP_OFFSET + | FEAT_TEMP_OLD_PECI, + .old_peci_mask = 0x4, + }, + [it8720] = { + .name = "it8720", + .features = FEAT_16BIT_FANS | FEAT_TEMP_OFFSET + | FEAT_TEMP_OLD_PECI, + .old_peci_mask = 0x4, + }, + [it8721] = { + .name = "it8721", + .features = FEAT_NEWER_AUTOPWM | FEAT_12MV_ADC | FEAT_16BIT_FANS + | FEAT_TEMP_OFFSET | FEAT_TEMP_OLD_PECI | FEAT_TEMP_PECI, + .peci_mask = 0x05, + .old_peci_mask = 0x02, /* Actually reports PCH */ + }, + [it8728] = { + .name = "it8728", + .features = FEAT_NEWER_AUTOPWM | FEAT_12MV_ADC | FEAT_16BIT_FANS + | FEAT_TEMP_OFFSET | FEAT_TEMP_PECI, + .peci_mask = 0x07, + }, + [it8782] = { + .name = "it8782", + .features = FEAT_16BIT_FANS | FEAT_TEMP_OFFSET + | FEAT_TEMP_OLD_PECI, + .old_peci_mask = 0x4, + }, + [it8783] = { + .name = "it8783", + .features = FEAT_16BIT_FANS | FEAT_TEMP_OFFSET + | FEAT_TEMP_OLD_PECI, + .old_peci_mask = 0x4, + }, +}; + +#define has_16bit_fans(data) ((data)->features & FEAT_16BIT_FANS) +#define has_12mv_adc(data) ((data)->features & FEAT_12MV_ADC) +#define has_newer_autopwm(data) ((data)->features & FEAT_NEWER_AUTOPWM) +#define has_old_autopwm(data) ((data)->features & FEAT_OLD_AUTOPWM) +#define has_temp_offset(data) ((data)->features & FEAT_TEMP_OFFSET) +#define has_temp_peci(data, nr) (((data)->features & FEAT_TEMP_PECI) && \ + ((data)->peci_mask & (1 << nr))) +#define has_temp_old_peci(data, nr) \ + (((data)->features & FEAT_TEMP_OLD_PECI) && \ + ((data)->old_peci_mask & (1 << nr))) struct it87_sio_data { enum chips type; @@ -249,7 +328,9 @@ struct it87_sio_data { struct it87_data { struct device *hwmon_dev; enum chips type; - u8 revision; + u16 features; + u8 peci_mask; + u8 old_peci_mask; unsigned short addr; const char *name; @@ -258,17 +339,13 @@ struct it87_data { unsigned long last_updated; /* In jiffies */ u16 in_scaled; /* Internal voltage sensors are scaled */ - u8 in[9]; /* Register value */ - u8 in_max[8]; /* Register value */ - u8 in_min[8]; /* Register value */ + u8 in[9][3]; /* [nr][0]=in, [1]=min, [2]=max */ u8 has_fan; /* Bitfield, fans enabled */ - u16 fan[5]; /* Register values, possibly combined */ - u16 fan_min[5]; /* Register values, possibly combined */ + u16 fan[5][2]; /* Register values, [nr][0]=fan, [1]=min */ u8 has_temp; /* Bitfield, temp sensors enabled */ - s8 temp[3]; /* Register value */ - s8 temp_high[3]; /* Register value */ - s8 temp_low[3]; /* Register value */ - u8 sensor; /* Register value */ + s8 temp[3][4]; /* [nr][0]=temp, [1]=min, [2]=max, [3]=offset */ + u8 sensor; /* Register value (IT87_REG_TEMP_ENABLE) */ + u8 extra; /* Register value (IT87_REG_TEMP_EXTRA) */ u8 fan_div[3]; /* Register encoding, shifted right */ u8 vid; /* Register encoding, combined */ u8 vrm; @@ -296,26 +373,6 @@ struct it87_data { s8 auto_temp[3][5]; /* [nr][0] is point1_temp_hyst */ }; -static inline int has_12mv_adc(const struct it87_data *data) -{ - /* - * IT8721F and later have a 12 mV ADC, also with internal scaling - * on selected inputs. - */ - return data->type == it8721 - || data->type == it8728; -} - -static inline int has_newer_autopwm(const struct it87_data *data) -{ - /* - * IT8721F and later have separate registers for the temperature - * mapping and the manual duty cycle. - */ - return data->type == it8721 - || data->type == it8728; -} - static int adc_lsb(const struct it87_data *data, int nr) { int lsb = has_12mv_adc(data) ? 12 : 16; @@ -398,35 +455,6 @@ static const unsigned int pwm_freq[8] = { 750000 / 128, }; -static inline int has_16bit_fans(const struct it87_data *data) -{ - /* - * IT8705F Datasheet 0.4.1, 3h == Version G. - * IT8712F Datasheet 0.9.1, section 8.3.5 indicates 8h == Version J. - * These are the first revisions with 16-bit tachometer support. - */ - return (data->type == it87 && data->revision >= 0x03) - || (data->type == it8712 && data->revision >= 0x08) - || data->type == it8716 - || data->type == it8718 - || data->type == it8720 - || data->type == it8721 - || data->type == it8728 - || data->type == it8782 - || data->type == it8783; -} - -static inline int has_old_autopwm(const struct it87_data *data) -{ - /* - * The old automatic fan speed control interface is implemented - * by IT8705F chips up to revision F and IT8712F chips up to - * revision G. - */ - return (data->type == it87 && data->revision < 0x03) - || (data->type == it8712 && data->revision < 0x08); -} - static int it87_probe(struct platform_device *pdev); static int it87_remove(struct platform_device *pdev); @@ -447,59 +475,22 @@ static struct platform_driver it87_driver = { }; static ssize_t show_in(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - - struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", in_from_reg(data, nr, data->in[nr])); -} - -static ssize_t show_in_min(struct device *dev, struct device_attribute *attr, - char *buf) + char *buf) { - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; + struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr); + int nr = sattr->nr; + int index = sattr->index; struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", in_from_reg(data, nr, data->in_min[nr])); + return sprintf(buf, "%d\n", in_from_reg(data, nr, data->in[nr][index])); } -static ssize_t show_in_max(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - - struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", in_from_reg(data, nr, data->in_max[nr])); -} - -static ssize_t set_in_min(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - - struct it87_data *data = dev_get_drvdata(dev); - unsigned long val; - - if (kstrtoul(buf, 10, &val) < 0) - return -EINVAL; - - mutex_lock(&data->update_lock); - data->in_min[nr] = in_to_reg(data, nr, val); - it87_write_value(data, IT87_REG_VIN_MIN(nr), - data->in_min[nr]); - mutex_unlock(&data->update_lock); - return count; -} -static ssize_t set_in_max(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t set_in(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) { - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; + struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr); + int nr = sattr->nr; + int index = sattr->index; struct it87_data *data = dev_get_drvdata(dev); unsigned long val; @@ -508,140 +499,167 @@ static ssize_t set_in_max(struct device *dev, struct device_attribute *attr, return -EINVAL; mutex_lock(&data->update_lock); - data->in_max[nr] = in_to_reg(data, nr, val); - it87_write_value(data, IT87_REG_VIN_MAX(nr), - data->in_max[nr]); + data->in[nr][index] = in_to_reg(data, nr, val); + it87_write_value(data, + index == 1 ? IT87_REG_VIN_MIN(nr) + : IT87_REG_VIN_MAX(nr), + data->in[nr][index]); mutex_unlock(&data->update_lock); return count; } -#define show_in_offset(offset) \ -static SENSOR_DEVICE_ATTR(in##offset##_input, S_IRUGO, \ - show_in, NULL, offset); - -#define limit_in_offset(offset) \ -static SENSOR_DEVICE_ATTR(in##offset##_min, S_IRUGO | S_IWUSR, \ - show_in_min, set_in_min, offset); \ -static SENSOR_DEVICE_ATTR(in##offset##_max, S_IRUGO | S_IWUSR, \ - show_in_max, set_in_max, offset); - -show_in_offset(0); -limit_in_offset(0); -show_in_offset(1); -limit_in_offset(1); -show_in_offset(2); -limit_in_offset(2); -show_in_offset(3); -limit_in_offset(3); -show_in_offset(4); -limit_in_offset(4); -show_in_offset(5); -limit_in_offset(5); -show_in_offset(6); -limit_in_offset(6); -show_in_offset(7); -limit_in_offset(7); -show_in_offset(8); +static SENSOR_DEVICE_ATTR_2(in0_input, S_IRUGO, show_in, NULL, 0, 0); +static SENSOR_DEVICE_ATTR_2(in0_min, S_IRUGO | S_IWUSR, show_in, set_in, + 0, 1); +static SENSOR_DEVICE_ATTR_2(in0_max, S_IRUGO | S_IWUSR, show_in, set_in, + 0, 2); + +static SENSOR_DEVICE_ATTR_2(in1_input, S_IRUGO, show_in, NULL, 1, 0); +static SENSOR_DEVICE_ATTR_2(in1_min, S_IRUGO | S_IWUSR, show_in, set_in, + 1, 1); +static SENSOR_DEVICE_ATTR_2(in1_max, S_IRUGO | S_IWUSR, show_in, set_in, + 1, 2); + +static SENSOR_DEVICE_ATTR_2(in2_input, S_IRUGO, show_in, NULL, 2, 0); +static SENSOR_DEVICE_ATTR_2(in2_min, S_IRUGO | S_IWUSR, show_in, set_in, + 2, 1); +static SENSOR_DEVICE_ATTR_2(in2_max, S_IRUGO | S_IWUSR, show_in, set_in, + 2, 2); + +static SENSOR_DEVICE_ATTR_2(in3_input, S_IRUGO, show_in, NULL, 3, 0); +static SENSOR_DEVICE_ATTR_2(in3_min, S_IRUGO | S_IWUSR, show_in, set_in, + 3, 1); +static SENSOR_DEVICE_ATTR_2(in3_max, S_IRUGO | S_IWUSR, show_in, set_in, + 3, 2); + +static SENSOR_DEVICE_ATTR_2(in4_input, S_IRUGO, show_in, NULL, 4, 0); +static SENSOR_DEVICE_ATTR_2(in4_min, S_IRUGO | S_IWUSR, show_in, set_in, + 4, 1); +static SENSOR_DEVICE_ATTR_2(in4_max, S_IRUGO | S_IWUSR, show_in, set_in, + 4, 2); + +static SENSOR_DEVICE_ATTR_2(in5_input, S_IRUGO, show_in, NULL, 5, 0); +static SENSOR_DEVICE_ATTR_2(in5_min, S_IRUGO | S_IWUSR, show_in, set_in, + 5, 1); +static SENSOR_DEVICE_ATTR_2(in5_max, S_IRUGO | S_IWUSR, show_in, set_in, + 5, 2); + +static SENSOR_DEVICE_ATTR_2(in6_input, S_IRUGO, show_in, NULL, 6, 0); +static SENSOR_DEVICE_ATTR_2(in6_min, S_IRUGO | S_IWUSR, show_in, set_in, + 6, 1); +static SENSOR_DEVICE_ATTR_2(in6_max, S_IRUGO | S_IWUSR, show_in, set_in, + 6, 2); + +static SENSOR_DEVICE_ATTR_2(in7_input, S_IRUGO, show_in, NULL, 7, 0); +static SENSOR_DEVICE_ATTR_2(in7_min, S_IRUGO | S_IWUSR, show_in, set_in, + 7, 1); +static SENSOR_DEVICE_ATTR_2(in7_max, S_IRUGO | S_IWUSR, show_in, set_in, + 7, 2); + +static SENSOR_DEVICE_ATTR_2(in8_input, S_IRUGO, show_in, NULL, 8, 0); /* 3 temperatures */ static ssize_t show_temp(struct device *dev, struct device_attribute *attr, - char *buf) + char *buf) { - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - + struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr); + int nr = sattr->nr; + int index = sattr->index; struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", TEMP_FROM_REG(data->temp[nr])); -} -static ssize_t show_temp_max(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", TEMP_FROM_REG(data->temp_high[nr])); + return sprintf(buf, "%d\n", TEMP_FROM_REG(data->temp[nr][index])); } -static ssize_t show_temp_min(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", TEMP_FROM_REG(data->temp_low[nr])); -} -static ssize_t set_temp_max(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t set_temp(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) { - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - + struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr); + int nr = sattr->nr; + int index = sattr->index; struct it87_data *data = dev_get_drvdata(dev); long val; + u8 reg, regval; if (kstrtol(buf, 10, &val) < 0) return -EINVAL; mutex_lock(&data->update_lock); - data->temp_high[nr] = TEMP_TO_REG(val); - it87_write_value(data, IT87_REG_TEMP_HIGH(nr), data->temp_high[nr]); - mutex_unlock(&data->update_lock); - return count; -} -static ssize_t set_temp_min(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - struct it87_data *data = dev_get_drvdata(dev); - long val; - - if (kstrtol(buf, 10, &val) < 0) - return -EINVAL; + switch (index) { + default: + case 1: + reg = IT87_REG_TEMP_LOW(nr); + break; + case 2: + reg = IT87_REG_TEMP_HIGH(nr); + break; + case 3: + regval = it87_read_value(data, IT87_REG_BEEP_ENABLE); + if (!(regval & 0x80)) { + regval |= 0x80; + it87_write_value(data, IT87_REG_BEEP_ENABLE, regval); + } + data->valid = 0; + reg = IT87_REG_TEMP_OFFSET[nr]; + break; + } - mutex_lock(&data->update_lock); - data->temp_low[nr] = TEMP_TO_REG(val); - it87_write_value(data, IT87_REG_TEMP_LOW(nr), data->temp_low[nr]); + data->temp[nr][index] = TEMP_TO_REG(val); + it87_write_value(data, reg, data->temp[nr][index]); mutex_unlock(&data->update_lock); return count; } -#define show_temp_offset(offset) \ -static SENSOR_DEVICE_ATTR(temp##offset##_input, S_IRUGO, \ - show_temp, NULL, offset - 1); \ -static SENSOR_DEVICE_ATTR(temp##offset##_max, S_IRUGO | S_IWUSR, \ - show_temp_max, set_temp_max, offset - 1); \ -static SENSOR_DEVICE_ATTR(temp##offset##_min, S_IRUGO | S_IWUSR, \ - show_temp_min, set_temp_min, offset - 1); - -show_temp_offset(1); -show_temp_offset(2); -show_temp_offset(3); - -static ssize_t show_sensor(struct device *dev, struct device_attribute *attr, - char *buf) + +static SENSOR_DEVICE_ATTR_2(temp1_input, S_IRUGO, show_temp, NULL, 0, 0); +static SENSOR_DEVICE_ATTR_2(temp1_min, S_IRUGO | S_IWUSR, show_temp, set_temp, + 0, 1); +static SENSOR_DEVICE_ATTR_2(temp1_max, S_IRUGO | S_IWUSR, show_temp, set_temp, + 0, 2); +static SENSOR_DEVICE_ATTR_2(temp1_offset, S_IRUGO | S_IWUSR, show_temp, + set_temp, 0, 3); +static SENSOR_DEVICE_ATTR_2(temp2_input, S_IRUGO, show_temp, NULL, 1, 0); +static SENSOR_DEVICE_ATTR_2(temp2_min, S_IRUGO | S_IWUSR, show_temp, set_temp, + 1, 1); +static SENSOR_DEVICE_ATTR_2(temp2_max, S_IRUGO | S_IWUSR, show_temp, set_temp, + 1, 2); +static SENSOR_DEVICE_ATTR_2(temp2_offset, S_IRUGO | S_IWUSR, show_temp, + set_temp, 1, 3); +static SENSOR_DEVICE_ATTR_2(temp3_input, S_IRUGO, show_temp, NULL, 2, 0); +static SENSOR_DEVICE_ATTR_2(temp3_min, S_IRUGO | S_IWUSR, show_temp, set_temp, + 2, 1); +static SENSOR_DEVICE_ATTR_2(temp3_max, S_IRUGO | S_IWUSR, show_temp, set_temp, + 2, 2); +static SENSOR_DEVICE_ATTR_2(temp3_offset, S_IRUGO | S_IWUSR, show_temp, + set_temp, 2, 3); + +static ssize_t show_temp_type(struct device *dev, struct device_attribute *attr, + char *buf) { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; struct it87_data *data = it87_update_device(dev); u8 reg = data->sensor; /* In case value is updated while used */ + u8 extra = data->extra; + if ((has_temp_peci(data, nr) && (reg >> 6 == nr + 1)) + || (has_temp_old_peci(data, nr) && (extra & 0x80))) + return sprintf(buf, "6\n"); /* Intel PECI */ if (reg & (1 << nr)) return sprintf(buf, "3\n"); /* thermal diode */ if (reg & (8 << nr)) return sprintf(buf, "4\n"); /* thermistor */ return sprintf(buf, "0\n"); /* disabled */ } -static ssize_t set_sensor(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) + +static ssize_t set_temp_type(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); long val; - u8 reg; + u8 reg, extra; if (kstrtol(buf, 10, &val) < 0) return -EINVAL; @@ -649,33 +667,45 @@ static ssize_t set_sensor(struct device *dev, struct device_attribute *attr, reg = it87_read_value(data, IT87_REG_TEMP_ENABLE); reg &= ~(1 << nr); reg &= ~(8 << nr); + if (has_temp_peci(data, nr) && (reg >> 6 == nr + 1 || val == 6)) + reg &= 0x3f; + extra = it87_read_value(data, IT87_REG_TEMP_EXTRA); + if (has_temp_old_peci(data, nr) && ((extra & 0x80) || val == 6)) + extra &= 0x7f; if (val == 2) { /* backwards compatibility */ - dev_warn(dev, "Sensor type 2 is deprecated, please use 4 " - "instead\n"); + dev_warn(dev, + "Sensor type 2 is deprecated, please use 4 instead\n"); val = 4; } - /* 3 = thermal diode; 4 = thermistor; 0 = disabled */ + /* 3 = thermal diode; 4 = thermistor; 6 = Intel PECI; 0 = disabled */ if (val == 3) reg |= 1 << nr; else if (val == 4) reg |= 8 << nr; + else if (has_temp_peci(data, nr) && val == 6) + reg |= (nr + 1) << 6; + else if (has_temp_old_peci(data, nr) && val == 6) + extra |= 0x80; else if (val != 0) return -EINVAL; mutex_lock(&data->update_lock); data->sensor = reg; + data->extra = extra; it87_write_value(data, IT87_REG_TEMP_ENABLE, data->sensor); + if (has_temp_old_peci(data, nr)) + it87_write_value(data, IT87_REG_TEMP_EXTRA, data->extra); data->valid = 0; /* Force cache refresh */ mutex_unlock(&data->update_lock); return count; } -#define show_sensor_offset(offset) \ -static SENSOR_DEVICE_ATTR(temp##offset##_type, S_IRUGO | S_IWUSR, \ - show_sensor, set_sensor, offset - 1); -show_sensor_offset(1); -show_sensor_offset(2); -show_sensor_offset(3); +static SENSOR_DEVICE_ATTR(temp1_type, S_IRUGO | S_IWUSR, show_temp_type, + set_temp_type, 0); +static SENSOR_DEVICE_ATTR(temp2_type, S_IRUGO | S_IWUSR, show_temp_type, + set_temp_type, 1); +static SENSOR_DEVICE_ATTR(temp3_type, S_IRUGO | S_IWUSR, show_temp_type, + set_temp_type, 2); /* 3 Fans */ @@ -692,25 +722,21 @@ static int pwm_mode(const struct it87_data *data, int nr) } static ssize_t show_fan(struct device *dev, struct device_attribute *attr, - char *buf) + char *buf) { - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - + struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr); + int nr = sattr->nr; + int index = sattr->index; + int speed; struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", FAN_FROM_REG(data->fan[nr], - DIV_FROM_REG(data->fan_div[nr]))); -} -static ssize_t show_fan_min(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", FAN_FROM_REG(data->fan_min[nr], - DIV_FROM_REG(data->fan_div[nr]))); + speed = has_16bit_fans(data) ? + FAN16_FROM_REG(data->fan[nr][index]) : + FAN_FROM_REG(data->fan[nr][index], + DIV_FROM_REG(data->fan_div[nr])); + return sprintf(buf, "%d\n", speed); } + static ssize_t show_fan_div(struct device *dev, struct device_attribute *attr, char *buf) { @@ -747,11 +773,13 @@ static ssize_t show_pwm_freq(struct device *dev, struct device_attribute *attr, return sprintf(buf, "%u\n", pwm_freq[index]); } -static ssize_t set_fan_min(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) + +static ssize_t set_fan(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) { - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; + struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr); + int nr = sattr->nr; + int index = sattr->index; struct it87_data *data = dev_get_drvdata(dev); long val; @@ -761,24 +789,36 @@ static ssize_t set_fan_min(struct device *dev, struct device_attribute *attr, return -EINVAL; mutex_lock(&data->update_lock); - reg = it87_read_value(data, IT87_REG_FAN_DIV); - switch (nr) { - case 0: - data->fan_div[nr] = reg & 0x07; - break; - case 1: - data->fan_div[nr] = (reg >> 3) & 0x07; - break; - case 2: - data->fan_div[nr] = (reg & 0x40) ? 3 : 1; - break; + + if (has_16bit_fans(data)) { + data->fan[nr][index] = FAN16_TO_REG(val); + it87_write_value(data, IT87_REG_FAN_MIN[nr], + data->fan[nr][index] & 0xff); + it87_write_value(data, IT87_REG_FANX_MIN[nr], + data->fan[nr][index] >> 8); + } else { + reg = it87_read_value(data, IT87_REG_FAN_DIV); + switch (nr) { + case 0: + data->fan_div[nr] = reg & 0x07; + break; + case 1: + data->fan_div[nr] = (reg >> 3) & 0x07; + break; + case 2: + data->fan_div[nr] = (reg & 0x40) ? 3 : 1; + break; + } + data->fan[nr][index] = + FAN_TO_REG(val, DIV_FROM_REG(data->fan_div[nr])); + it87_write_value(data, IT87_REG_FAN_MIN[nr], + data->fan[nr][index]); } - data->fan_min[nr] = FAN_TO_REG(val, DIV_FROM_REG(data->fan_div[nr])); - it87_write_value(data, IT87_REG_FAN_MIN[nr], data->fan_min[nr]); mutex_unlock(&data->update_lock); return count; } + static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { @@ -797,7 +837,7 @@ static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, old = it87_read_value(data, IT87_REG_FAN_DIV); /* Save fan min limit */ - min = FAN_FROM_REG(data->fan_min[nr], DIV_FROM_REG(data->fan_div[nr])); + min = FAN_FROM_REG(data->fan[nr][1], DIV_FROM_REG(data->fan_div[nr])); switch (nr) { case 0: @@ -818,8 +858,8 @@ static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, it87_write_value(data, IT87_REG_FAN_DIV, val); /* Restore fan min limit */ - data->fan_min[nr] = FAN_TO_REG(min, DIV_FROM_REG(data->fan_div[nr])); - it87_write_value(data, IT87_REG_FAN_MIN[nr], data->fan_min[nr]); + data->fan[nr][1] = FAN_TO_REG(min, DIV_FROM_REG(data->fan_div[nr])); + it87_write_value(data, IT87_REG_FAN_MIN[nr], data->fan[nr][1]); mutex_unlock(&data->update_lock); return count; @@ -843,8 +883,8 @@ static int check_trip_points(struct device *dev, int nr) } if (err) { - dev_err(dev, "Inconsistent trip points, not switching to " - "automatic mode\n"); + dev_err(dev, + "Inconsistent trip points, not switching to automatic mode\n"); dev_err(dev, "Adjust the trip points and try again\n"); } return err; @@ -1092,118 +1132,106 @@ static ssize_t set_auto_temp(struct device *dev, return count; } -#define show_fan_offset(offset) \ -static SENSOR_DEVICE_ATTR(fan##offset##_input, S_IRUGO, \ - show_fan, NULL, offset - 1); \ -static SENSOR_DEVICE_ATTR(fan##offset##_min, S_IRUGO | S_IWUSR, \ - show_fan_min, set_fan_min, offset - 1); \ -static SENSOR_DEVICE_ATTR(fan##offset##_div, S_IRUGO | S_IWUSR, \ - show_fan_div, set_fan_div, offset - 1); - -show_fan_offset(1); -show_fan_offset(2); -show_fan_offset(3); - -#define show_pwm_offset(offset) \ -static SENSOR_DEVICE_ATTR(pwm##offset##_enable, S_IRUGO | S_IWUSR, \ - show_pwm_enable, set_pwm_enable, offset - 1); \ -static SENSOR_DEVICE_ATTR(pwm##offset, S_IRUGO | S_IWUSR, \ - show_pwm, set_pwm, offset - 1); \ -static DEVICE_ATTR(pwm##offset##_freq, \ - (offset == 1 ? S_IRUGO | S_IWUSR : S_IRUGO), \ - show_pwm_freq, (offset == 1 ? set_pwm_freq : NULL)); \ -static SENSOR_DEVICE_ATTR(pwm##offset##_auto_channels_temp, \ - S_IRUGO | S_IWUSR, show_pwm_temp_map, set_pwm_temp_map, \ - offset - 1); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point1_pwm, \ - S_IRUGO | S_IWUSR, show_auto_pwm, set_auto_pwm, \ - offset - 1, 0); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point2_pwm, \ - S_IRUGO | S_IWUSR, show_auto_pwm, set_auto_pwm, \ - offset - 1, 1); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point3_pwm, \ - S_IRUGO | S_IWUSR, show_auto_pwm, set_auto_pwm, \ - offset - 1, 2); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point4_pwm, \ - S_IRUGO, show_auto_pwm, NULL, offset - 1, 3); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point1_temp, \ - S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ - offset - 1, 1); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point1_temp_hyst, \ - S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ - offset - 1, 0); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point2_temp, \ - S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ - offset - 1, 2); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point3_temp, \ - S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ - offset - 1, 3); \ -static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point4_temp, \ - S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ - offset - 1, 4); - -show_pwm_offset(1); -show_pwm_offset(2); -show_pwm_offset(3); - -/* A different set of callbacks for 16-bit fans */ -static ssize_t show_fan16(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", FAN16_FROM_REG(data->fan[nr])); -} - -static ssize_t show_fan16_min(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - struct it87_data *data = it87_update_device(dev); - return sprintf(buf, "%d\n", FAN16_FROM_REG(data->fan_min[nr])); -} - -static ssize_t set_fan16_min(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); - int nr = sensor_attr->index; - struct it87_data *data = dev_get_drvdata(dev); - long val; - - if (kstrtol(buf, 10, &val) < 0) - return -EINVAL; - - mutex_lock(&data->update_lock); - data->fan_min[nr] = FAN16_TO_REG(val); - it87_write_value(data, IT87_REG_FAN_MIN[nr], - data->fan_min[nr] & 0xff); - it87_write_value(data, IT87_REG_FANX_MIN[nr], - data->fan_min[nr] >> 8); - mutex_unlock(&data->update_lock); - return count; -} - -/* - * We want to use the same sysfs file names as 8-bit fans, but we need - * different variable names, so we have to use SENSOR_ATTR instead of - * SENSOR_DEVICE_ATTR. - */ -#define show_fan16_offset(offset) \ -static struct sensor_device_attribute sensor_dev_attr_fan##offset##_input16 \ - = SENSOR_ATTR(fan##offset##_input, S_IRUGO, \ - show_fan16, NULL, offset - 1); \ -static struct sensor_device_attribute sensor_dev_attr_fan##offset##_min16 \ - = SENSOR_ATTR(fan##offset##_min, S_IRUGO | S_IWUSR, \ - show_fan16_min, set_fan16_min, offset - 1) - -show_fan16_offset(1); -show_fan16_offset(2); -show_fan16_offset(3); -show_fan16_offset(4); -show_fan16_offset(5); +static SENSOR_DEVICE_ATTR_2(fan1_input, S_IRUGO, show_fan, NULL, 0, 0); +static SENSOR_DEVICE_ATTR_2(fan1_min, S_IRUGO | S_IWUSR, show_fan, set_fan, + 0, 1); +static SENSOR_DEVICE_ATTR(fan1_div, S_IRUGO | S_IWUSR, show_fan_div, + set_fan_div, 0); + +static SENSOR_DEVICE_ATTR_2(fan2_input, S_IRUGO, show_fan, NULL, 1, 0); +static SENSOR_DEVICE_ATTR_2(fan2_min, S_IRUGO | S_IWUSR, show_fan, set_fan, + 1, 1); +static SENSOR_DEVICE_ATTR(fan2_div, S_IRUGO | S_IWUSR, show_fan_div, + set_fan_div, 1); + +static SENSOR_DEVICE_ATTR_2(fan3_input, S_IRUGO, show_fan, NULL, 2, 0); +static SENSOR_DEVICE_ATTR_2(fan3_min, S_IRUGO | S_IWUSR, show_fan, set_fan, + 2, 1); +static SENSOR_DEVICE_ATTR(fan3_div, S_IRUGO | S_IWUSR, show_fan_div, + set_fan_div, 2); + +static SENSOR_DEVICE_ATTR_2(fan4_input, S_IRUGO, show_fan, NULL, 3, 0); +static SENSOR_DEVICE_ATTR_2(fan4_min, S_IRUGO | S_IWUSR, show_fan, set_fan, + 3, 1); + +static SENSOR_DEVICE_ATTR_2(fan5_input, S_IRUGO, show_fan, NULL, 4, 0); +static SENSOR_DEVICE_ATTR_2(fan5_min, S_IRUGO | S_IWUSR, show_fan, set_fan, + 4, 1); + +static SENSOR_DEVICE_ATTR(pwm1_enable, S_IRUGO | S_IWUSR, + show_pwm_enable, set_pwm_enable, 0); +static SENSOR_DEVICE_ATTR(pwm1, S_IRUGO | S_IWUSR, show_pwm, set_pwm, 0); +static DEVICE_ATTR(pwm1_freq, S_IRUGO | S_IWUSR, show_pwm_freq, set_pwm_freq); +static SENSOR_DEVICE_ATTR(pwm1_auto_channels_temp, S_IRUGO | S_IWUSR, + show_pwm_temp_map, set_pwm_temp_map, 0); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point1_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 0, 0); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point2_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 0, 1); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point3_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 0, 2); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point4_pwm, S_IRUGO, + show_auto_pwm, NULL, 0, 3); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point1_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 0, 1); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point1_temp_hyst, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 0, 0); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point2_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 0, 2); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point3_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 0, 3); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point4_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 0, 4); + +static SENSOR_DEVICE_ATTR(pwm2_enable, S_IRUGO | S_IWUSR, + show_pwm_enable, set_pwm_enable, 1); +static SENSOR_DEVICE_ATTR(pwm2, S_IRUGO | S_IWUSR, show_pwm, set_pwm, 1); +static DEVICE_ATTR(pwm2_freq, S_IRUGO, show_pwm_freq, NULL); +static SENSOR_DEVICE_ATTR(pwm2_auto_channels_temp, S_IRUGO | S_IWUSR, + show_pwm_temp_map, set_pwm_temp_map, 1); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point1_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 1, 0); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point2_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 1, 1); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point3_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 1, 2); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point4_pwm, S_IRUGO, + show_auto_pwm, NULL, 1, 3); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point1_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 1, 1); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point1_temp_hyst, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 1, 0); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point2_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 1, 2); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point3_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 1, 3); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point4_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 1, 4); + +static SENSOR_DEVICE_ATTR(pwm3_enable, S_IRUGO | S_IWUSR, + show_pwm_enable, set_pwm_enable, 2); +static SENSOR_DEVICE_ATTR(pwm3, S_IRUGO | S_IWUSR, show_pwm, set_pwm, 2); +static DEVICE_ATTR(pwm3_freq, S_IRUGO, show_pwm_freq, NULL); +static SENSOR_DEVICE_ATTR(pwm3_auto_channels_temp, S_IRUGO | S_IWUSR, + show_pwm_temp_map, set_pwm_temp_map, 2); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point1_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 2, 0); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point2_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 2, 1); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point3_pwm, S_IRUGO | S_IWUSR, + show_auto_pwm, set_auto_pwm, 2, 2); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point4_pwm, S_IRUGO, + show_auto_pwm, NULL, 2, 3); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point1_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 2, 1); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point1_temp_hyst, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 2, 0); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point2_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 2, 2); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point3_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 2, 3); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point4_temp, S_IRUGO | S_IWUSR, + show_auto_temp, set_auto_temp, 2, 4); /* Alarms */ static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, @@ -1471,6 +1499,12 @@ static const struct attribute_group it87_group_temp[3] = { { .attrs = it87_attributes_temp[2] }, }; +static struct attribute *it87_attributes_temp_offset[] = { + &sensor_dev_attr_temp1_offset.dev_attr.attr, + &sensor_dev_attr_temp2_offset.dev_attr.attr, + &sensor_dev_attr_temp3_offset.dev_attr.attr, +}; + static struct attribute *it87_attributes[] = { &dev_attr_alarms.attr, &sensor_dev_attr_intrusion0_alarm.dev_attr.attr, @@ -1500,73 +1534,47 @@ static struct attribute *it87_attributes_temp_beep[] = { &sensor_dev_attr_temp3_beep.dev_attr.attr, }; -static struct attribute *it87_attributes_fan16[5][3+1] = { { - &sensor_dev_attr_fan1_input16.dev_attr.attr, - &sensor_dev_attr_fan1_min16.dev_attr.attr, +static struct attribute *it87_attributes_fan[5][3+1] = { { + &sensor_dev_attr_fan1_input.dev_attr.attr, + &sensor_dev_attr_fan1_min.dev_attr.attr, &sensor_dev_attr_fan1_alarm.dev_attr.attr, NULL }, { - &sensor_dev_attr_fan2_input16.dev_attr.attr, - &sensor_dev_attr_fan2_min16.dev_attr.attr, + &sensor_dev_attr_fan2_input.dev_attr.attr, + &sensor_dev_attr_fan2_min.dev_attr.attr, &sensor_dev_attr_fan2_alarm.dev_attr.attr, NULL }, { - &sensor_dev_attr_fan3_input16.dev_attr.attr, - &sensor_dev_attr_fan3_min16.dev_attr.attr, + &sensor_dev_attr_fan3_input.dev_attr.attr, + &sensor_dev_attr_fan3_min.dev_attr.attr, &sensor_dev_attr_fan3_alarm.dev_attr.attr, NULL }, { - &sensor_dev_attr_fan4_input16.dev_attr.attr, - &sensor_dev_attr_fan4_min16.dev_attr.attr, + &sensor_dev_attr_fan4_input.dev_attr.attr, + &sensor_dev_attr_fan4_min.dev_attr.attr, &sensor_dev_attr_fan4_alarm.dev_attr.attr, NULL }, { - &sensor_dev_attr_fan5_input16.dev_attr.attr, - &sensor_dev_attr_fan5_min16.dev_attr.attr, + &sensor_dev_attr_fan5_input.dev_attr.attr, + &sensor_dev_attr_fan5_min.dev_attr.attr, &sensor_dev_attr_fan5_alarm.dev_attr.attr, NULL } }; -static const struct attribute_group it87_group_fan16[5] = { - { .attrs = it87_attributes_fan16[0] }, - { .attrs = it87_attributes_fan16[1] }, - { .attrs = it87_attributes_fan16[2] }, - { .attrs = it87_attributes_fan16[3] }, - { .attrs = it87_attributes_fan16[4] }, +static const struct attribute_group it87_group_fan[5] = { + { .attrs = it87_attributes_fan[0] }, + { .attrs = it87_attributes_fan[1] }, + { .attrs = it87_attributes_fan[2] }, + { .attrs = it87_attributes_fan[3] }, + { .attrs = it87_attributes_fan[4] }, }; -static struct attribute *it87_attributes_fan[3][4+1] = { { - &sensor_dev_attr_fan1_input.dev_attr.attr, - &sensor_dev_attr_fan1_min.dev_attr.attr, +static const struct attribute *it87_attributes_fan_div[] = { &sensor_dev_attr_fan1_div.dev_attr.attr, - &sensor_dev_attr_fan1_alarm.dev_attr.attr, - NULL -}, { - &sensor_dev_attr_fan2_input.dev_attr.attr, - &sensor_dev_attr_fan2_min.dev_attr.attr, &sensor_dev_attr_fan2_div.dev_attr.attr, - &sensor_dev_attr_fan2_alarm.dev_attr.attr, - NULL -}, { - &sensor_dev_attr_fan3_input.dev_attr.attr, - &sensor_dev_attr_fan3_min.dev_attr.attr, &sensor_dev_attr_fan3_div.dev_attr.attr, - &sensor_dev_attr_fan3_alarm.dev_attr.attr, - NULL -} }; - -static const struct attribute_group it87_group_fan[3] = { - { .attrs = it87_attributes_fan[0] }, - { .attrs = it87_attributes_fan[1] }, - { .attrs = it87_attributes_fan[2] }, }; -static const struct attribute_group * -it87_get_fan_group(const struct it87_data *data) -{ - return has_16bit_fans(data) ? it87_group_fan16 : it87_group_fan; -} - static struct attribute *it87_attributes_pwm[3][4+1] = { { &sensor_dev_attr_pwm1_enable.dev_attr.attr, &sensor_dev_attr_pwm1.dev_attr.attr, @@ -1925,7 +1933,6 @@ static void it87_remove_files(struct device *dev) { struct it87_data *data = platform_get_drvdata(pdev); struct it87_sio_data *sio_data = dev->platform_data; - const struct attribute_group *fan_group = it87_get_fan_group(data); int i; sysfs_remove_group(&dev->kobj, &it87_group); @@ -1941,6 +1948,9 @@ static void it87_remove_files(struct device *dev) if (!(data->has_temp & (1 << i))) continue; sysfs_remove_group(&dev->kobj, &it87_group_temp[i]); + if (has_temp_offset(data)) + sysfs_remove_file(&dev->kobj, + it87_attributes_temp_offset[i]); if (sio_data->beep_pin) sysfs_remove_file(&dev->kobj, it87_attributes_temp_beep[i]); @@ -1948,10 +1958,13 @@ static void it87_remove_files(struct device *dev) for (i = 0; i < 5; i++) { if (!(data->has_fan & (1 << i))) continue; - sysfs_remove_group(&dev->kobj, &fan_group[i]); + sysfs_remove_group(&dev->kobj, &it87_group_fan[i]); if (sio_data->beep_pin) sysfs_remove_file(&dev->kobj, it87_attributes_fan_beep[i]); + if (i < 3 && !has_16bit_fans(data)) + sysfs_remove_file(&dev->kobj, + it87_attributes_fan_div[i]); } for (i = 0; i < 3; i++) { if (sio_data->skip_pwm & (1 << 0)) @@ -1972,21 +1985,9 @@ static int it87_probe(struct platform_device *pdev) struct resource *res; struct device *dev = &pdev->dev; struct it87_sio_data *sio_data = dev->platform_data; - const struct attribute_group *fan_group; int err = 0, i; int enable_pwm_interface; int fan_beep_need_rw; - static const char * const names[] = { - "it87", - "it8712", - "it8716", - "it8718", - "it8720", - "it8721", - "it8728", - "it8782", - "it8783", - }; res = platform_get_resource(pdev, IORESOURCE_IO, 0); if (!devm_request_region(&pdev->dev, res->start, IT87_EC_EXTENT, @@ -2003,8 +2004,31 @@ static int it87_probe(struct platform_device *pdev) data->addr = res->start; data->type = sio_data->type; - data->revision = sio_data->revision; - data->name = names[sio_data->type]; + data->features = it87_devices[sio_data->type].features; + data->peci_mask = it87_devices[sio_data->type].peci_mask; + data->old_peci_mask = it87_devices[sio_data->type].old_peci_mask; + data->name = it87_devices[sio_data->type].name; + /* + * IT8705F Datasheet 0.4.1, 3h == Version G. + * IT8712F Datasheet 0.9.1, section 8.3.5 indicates 8h == Version J. + * These are the first revisions with 16-bit tachometer support. + */ + switch (data->type) { + case it87: + if (sio_data->revision >= 0x03) { + data->features &= ~FEAT_OLD_AUTOPWM; + data->features |= FEAT_16BIT_FANS; + } + break; + case it8712: + if (sio_data->revision >= 0x08) { + data->features &= ~FEAT_OLD_AUTOPWM; + data->features |= FEAT_16BIT_FANS; + } + break; + default: + break; + } /* Now, we do the remaining detection. */ if ((it87_read_value(data, IT87_REG_CONFIG) & 0x80) @@ -2068,6 +2092,12 @@ static int it87_probe(struct platform_device *pdev) err = sysfs_create_group(&dev->kobj, &it87_group_temp[i]); if (err) goto error; + if (has_temp_offset(data)) { + err = sysfs_create_file(&dev->kobj, + it87_attributes_temp_offset[i]); + if (err) + goto error; + } if (sio_data->beep_pin) { err = sysfs_create_file(&dev->kobj, it87_attributes_temp_beep[i]); @@ -2077,15 +2107,21 @@ static int it87_probe(struct platform_device *pdev) } /* Do not create fan files for disabled fans */ - fan_group = it87_get_fan_group(data); fan_beep_need_rw = 1; for (i = 0; i < 5; i++) { if (!(data->has_fan & (1 << i))) continue; - err = sysfs_create_group(&dev->kobj, &fan_group[i]); + err = sysfs_create_group(&dev->kobj, &it87_group_fan[i]); if (err) goto error; + if (i < 3 && !has_16bit_fans(data)) { + err = sysfs_create_file(&dev->kobj, + it87_attributes_fan_div[i]); + if (err) + goto error; + } + if (sio_data->beep_pin) { err = sysfs_create_file(&dev->kobj, it87_attributes_fan_beep[i]); @@ -2221,8 +2257,8 @@ static int it87_check_pwm(struct device *dev) * PWM interface). */ if (!((pwm[0] | pwm[1] | pwm[2]) & 0x80)) { - dev_info(dev, "Reconfiguring PWM to " - "active high polarity\n"); + dev_info(dev, + "Reconfiguring PWM to active high polarity\n"); it87_write_value(data, IT87_REG_FAN_CTL, tmp | 0x87); for (i = 0; i < 3; i++) @@ -2232,16 +2268,16 @@ static int it87_check_pwm(struct device *dev) return 1; } - dev_info(dev, "PWM configuration is " - "too broken to be fixed\n"); + dev_info(dev, + "PWM configuration is too broken to be fixed\n"); } - dev_info(dev, "Detected broken BIOS " - "defaults, disabling PWM interface\n"); + dev_info(dev, + "Detected broken BIOS defaults, disabling PWM interface\n"); return 0; } else if (fix_pwm_polarity) { - dev_info(dev, "PWM configuration looks " - "sane, won't touch\n"); + dev_info(dev, + "PWM configuration looks sane, won't touch\n"); } return 1; @@ -2389,42 +2425,46 @@ static struct it87_data *it87_update_device(struct device *dev) it87_read_value(data, IT87_REG_CONFIG) | 0x40); } for (i = 0; i <= 7; i++) { - data->in[i] = + data->in[i][0] = it87_read_value(data, IT87_REG_VIN(i)); - data->in_min[i] = + data->in[i][1] = it87_read_value(data, IT87_REG_VIN_MIN(i)); - data->in_max[i] = + data->in[i][2] = it87_read_value(data, IT87_REG_VIN_MAX(i)); } /* in8 (battery) has no limit registers */ - data->in[8] = it87_read_value(data, IT87_REG_VIN(8)); + data->in[8][0] = it87_read_value(data, IT87_REG_VIN(8)); for (i = 0; i < 5; i++) { /* Skip disabled fans */ if (!(data->has_fan & (1 << i))) continue; - data->fan_min[i] = + data->fan[i][1] = it87_read_value(data, IT87_REG_FAN_MIN[i]); - data->fan[i] = it87_read_value(data, + data->fan[i][0] = it87_read_value(data, IT87_REG_FAN[i]); /* Add high byte if in 16-bit mode */ if (has_16bit_fans(data)) { - data->fan[i] |= it87_read_value(data, + data->fan[i][0] |= it87_read_value(data, IT87_REG_FANX[i]) << 8; - data->fan_min[i] |= it87_read_value(data, + data->fan[i][1] |= it87_read_value(data, IT87_REG_FANX_MIN[i]) << 8; } } for (i = 0; i < 3; i++) { if (!(data->has_temp & (1 << i))) continue; - data->temp[i] = + data->temp[i][0] = it87_read_value(data, IT87_REG_TEMP(i)); - data->temp_high[i] = - it87_read_value(data, IT87_REG_TEMP_HIGH(i)); - data->temp_low[i] = + data->temp[i][1] = it87_read_value(data, IT87_REG_TEMP_LOW(i)); + data->temp[i][2] = + it87_read_value(data, IT87_REG_TEMP_HIGH(i)); + if (has_temp_offset(data)) + data->temp[i][3] = + it87_read_value(data, + IT87_REG_TEMP_OFFSET[i]); } /* Newer chips don't have clock dividers */ @@ -2448,6 +2488,7 @@ static struct it87_data *it87_update_device(struct device *dev) it87_update_pwm_ctrl(data, i); data->sensor = it87_read_value(data, IT87_REG_TEMP_ENABLE); + data->extra = it87_read_value(data, IT87_REG_TEMP_EXTRA); /* * The IT8705F does not have VID capability. * The IT8718F and later don't use IT87_REG_VID for the @@ -2549,8 +2590,7 @@ static void __exit sm_it87_exit(void) } -MODULE_AUTHOR("Chris Gauthron, " - "Jean Delvare <khali@linux-fr.org>"); +MODULE_AUTHOR("Chris Gauthron, Jean Delvare <khali@linux-fr.org>"); MODULE_DESCRIPTION("IT8705F/IT871xF/IT872xF hardware monitoring driver"); module_param(update_vbat, bool, 0); MODULE_PARM_DESC(update_vbat, "Update vbat if set else return powerup value"); diff --git a/drivers/hwmon/twl4030-madc-hwmon.c b/drivers/hwmon/twl4030-madc-hwmon.c index 149d44a7c584..6c6d440bb2dd 100644 --- a/drivers/hwmon/twl4030-madc-hwmon.c +++ b/drivers/hwmon/twl4030-madc-hwmon.c @@ -130,7 +130,7 @@ static int twl4030_madc_hwmon_remove(struct platform_device *pdev) static struct platform_driver twl4030_madc_hwmon_driver = { .probe = twl4030_madc_hwmon_probe, - .remove = __exit_p(twl4030_madc_hwmon_remove), + .remove = twl4030_madc_hwmon_remove, .driver = { .name = "twl4030_madc_hwmon", .owner = THIS_MODULE, diff --git a/drivers/hwmon/w83627ehf.c b/drivers/hwmon/w83627ehf.c index 55ac41c05561..0e8ffd6059a0 100644 --- a/drivers/hwmon/w83627ehf.c +++ b/drivers/hwmon/w83627ehf.c @@ -1,7 +1,7 @@ /* * w83627ehf - Driver for the hardware monitoring functionality of * the Winbond W83627EHF Super-I/O chip - * Copyright (C) 2005-2011 Jean Delvare <khali@linux-fr.org> + * Copyright (C) 2005-2012 Jean Delvare <khali@linux-fr.org> * Copyright (C) 2006 Yuan Mu (Winbond), * Rudolf Marek <r.marek@assembler.cz> * David Hubbard <david.c.hubbard@gmail.com> @@ -502,6 +502,13 @@ struct w83627ehf_data { u16 have_temp_offset; u8 in6_skip:1; u8 temp3_val_only:1; + +#ifdef CONFIG_PM + /* Remember extra register values over suspend/resume */ + u8 vbat; + u8 fandiv1; + u8 fandiv2; +#endif }; struct w83627ehf_sio_data { @@ -898,6 +905,8 @@ static struct w83627ehf_data *w83627ehf_update_device(struct device *dev) data->temp_max_hyst[i] = w83627ehf_read_temp(data, data->reg_temp_hyst[i]); + if (i > 2) + continue; if (data->have_temp_offset & (1 << i)) data->temp_offset[i] = w83627ehf_read_value(data, @@ -2608,10 +2617,98 @@ static int w83627ehf_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM +static int w83627ehf_suspend(struct device *dev) +{ + struct w83627ehf_data *data = w83627ehf_update_device(dev); + struct w83627ehf_sio_data *sio_data = dev->platform_data; + + mutex_lock(&data->update_lock); + data->vbat = w83627ehf_read_value(data, W83627EHF_REG_VBAT); + if (sio_data->kind == nct6775) { + data->fandiv1 = w83627ehf_read_value(data, NCT6775_REG_FANDIV1); + data->fandiv2 = w83627ehf_read_value(data, NCT6775_REG_FANDIV2); + } + mutex_unlock(&data->update_lock); + + return 0; +} + +static int w83627ehf_resume(struct device *dev) +{ + struct w83627ehf_data *data = dev_get_drvdata(dev); + struct w83627ehf_sio_data *sio_data = dev->platform_data; + int i; + + mutex_lock(&data->update_lock); + data->bank = 0xff; /* Force initial bank selection */ + + /* Restore limits */ + for (i = 0; i < data->in_num; i++) { + if ((i == 6) && data->in6_skip) + continue; + + w83627ehf_write_value(data, W83627EHF_REG_IN_MIN(i), + data->in_min[i]); + w83627ehf_write_value(data, W83627EHF_REG_IN_MAX(i), + data->in_max[i]); + } + + for (i = 0; i < 5; i++) { + if (!(data->has_fan_min & (1 << i))) + continue; + + w83627ehf_write_value(data, data->REG_FAN_MIN[i], + data->fan_min[i]); + } + + for (i = 0; i < NUM_REG_TEMP; i++) { + if (!(data->have_temp & (1 << i))) + continue; + + if (data->reg_temp_over[i]) + w83627ehf_write_temp(data, data->reg_temp_over[i], + data->temp_max[i]); + if (data->reg_temp_hyst[i]) + w83627ehf_write_temp(data, data->reg_temp_hyst[i], + data->temp_max_hyst[i]); + if (i > 2) + continue; + if (data->have_temp_offset & (1 << i)) + w83627ehf_write_value(data, + W83627EHF_REG_TEMP_OFFSET[i], + data->temp_offset[i]); + } + + /* Restore other settings */ + w83627ehf_write_value(data, W83627EHF_REG_VBAT, data->vbat); + if (sio_data->kind == nct6775) { + w83627ehf_write_value(data, NCT6775_REG_FANDIV1, data->fandiv1); + w83627ehf_write_value(data, NCT6775_REG_FANDIV2, data->fandiv2); + } + + /* Force re-reading all values */ + data->valid = 0; + mutex_unlock(&data->update_lock); + + return 0; +} + +static const struct dev_pm_ops w83627ehf_dev_pm_ops = { + .suspend = w83627ehf_suspend, + .resume = w83627ehf_resume, +}; + +#define W83627EHF_DEV_PM_OPS (&w83627ehf_dev_pm_ops) +#else +#define W83627EHF_DEV_PM_OPS NULL +#endif /* CONFIG_PM */ + static struct platform_driver w83627ehf_driver = { .driver = { .owner = THIS_MODULE, .name = DRVNAME, + .pm = W83627EHF_DEV_PM_OPS, }, .probe = w83627ehf_probe, .remove = w83627ehf_remove, diff --git a/drivers/hwmon/w83627hf.c b/drivers/hwmon/w83627hf.c index 7f68b8309d10..81f486520cea 100644 --- a/drivers/hwmon/w83627hf.c +++ b/drivers/hwmon/w83627hf.c @@ -5,7 +5,7 @@ * Philip Edelbrock <phil@netroedge.com>, * and Mark Studebaker <mdsxyz123@yahoo.com> * Ported to 2.6 by Bernhard C. Schrenk <clemy@clemy.org> - * Copyright (c) 2007 Jean Delvare <khali@linux-fr.org> + * Copyright (c) 2007 - 1012 Jean Delvare <khali@linux-fr.org> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -389,6 +389,12 @@ struct w83627hf_data { */ u8 vrm; u8 vrm_ovt; /* Register value, 627THF/637HF/687THF only */ + +#ifdef CONFIG_PM + /* Remember extra register values over suspend/resume */ + u8 scfg1; + u8 scfg2; +#endif }; @@ -401,10 +407,77 @@ static void w83627hf_update_fan_div(struct w83627hf_data *data); static struct w83627hf_data *w83627hf_update_device(struct device *dev); static void w83627hf_init_device(struct platform_device *pdev); +#ifdef CONFIG_PM +static int w83627hf_suspend(struct device *dev) +{ + struct w83627hf_data *data = w83627hf_update_device(dev); + + mutex_lock(&data->update_lock); + data->scfg1 = w83627hf_read_value(data, W83781D_REG_SCFG1); + data->scfg2 = w83627hf_read_value(data, W83781D_REG_SCFG2); + mutex_unlock(&data->update_lock); + + return 0; +} + +static int w83627hf_resume(struct device *dev) +{ + struct w83627hf_data *data = dev_get_drvdata(dev); + int i, num_temps = (data->type == w83697hf) ? 2 : 3; + + /* Restore limits */ + mutex_lock(&data->update_lock); + for (i = 0; i <= 8; i++) { + /* skip missing sensors */ + if (((data->type == w83697hf) && (i == 1)) || + ((data->type != w83627hf && data->type != w83697hf) + && (i == 5 || i == 6))) + continue; + w83627hf_write_value(data, W83781D_REG_IN_MAX(i), + data->in_max[i]); + w83627hf_write_value(data, W83781D_REG_IN_MIN(i), + data->in_min[i]); + } + for (i = 0; i <= 2; i++) + w83627hf_write_value(data, W83627HF_REG_FAN_MIN(i), + data->fan_min[i]); + for (i = 0; i < num_temps; i++) { + w83627hf_write_value(data, w83627hf_reg_temp_over[i], + data->temp_max[i]); + w83627hf_write_value(data, w83627hf_reg_temp_hyst[i], + data->temp_max_hyst[i]); + } + + /* Fixup BIOS bugs */ + if (data->type == w83627thf || data->type == w83637hf || + data->type == w83687thf) + w83627hf_write_value(data, W83627THF_REG_VRM_OVT_CFG, + data->vrm_ovt); + w83627hf_write_value(data, W83781D_REG_SCFG1, data->scfg1); + w83627hf_write_value(data, W83781D_REG_SCFG2, data->scfg2); + + /* Force re-reading all values */ + data->valid = 0; + mutex_unlock(&data->update_lock); + + return 0; +} + +static const struct dev_pm_ops w83627hf_dev_pm_ops = { + .suspend = w83627hf_suspend, + .resume = w83627hf_resume, +}; + +#define W83627HF_DEV_PM_OPS (&w83627hf_dev_pm_ops) +#else +#define W83627HF_DEV_PM_OPS NULL +#endif /* CONFIG_PM */ + static struct platform_driver w83627hf_driver = { .driver = { .owner = THIS_MODULE, .name = DRVNAME, + .pm = W83627HF_DEV_PM_OPS, }, .probe = w83627hf_probe, .remove = w83627hf_remove, @@ -1659,8 +1732,10 @@ static void w83627hf_init_device(struct platform_device *pdev) /* Minimize conflicts with other winbond i2c-only clients... */ /* disable i2c subclients... how to disable main i2c client?? */ /* force i2c address to relatively uncommon address */ - w83627hf_write_value(data, W83781D_REG_I2C_SUBADDR, 0x89); - w83627hf_write_value(data, W83781D_REG_I2C_ADDR, force_i2c); + if (type == w83627hf) { + w83627hf_write_value(data, W83781D_REG_I2C_SUBADDR, 0x89); + w83627hf_write_value(data, W83781D_REG_I2C_ADDR, force_i2c); + } /* Read VID only once */ if (type == w83627hf || type == w83637hf) { diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index c7bff51fe524..bdca5111eb9d 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -337,6 +337,16 @@ config I2C_BLACKFIN_TWI_CLK_KHZ help The unit of the TWI clock is kHz. +config I2C_CBUS_GPIO + tristate "CBUS I2C driver" + depends on GENERIC_GPIO + help + Support for CBUS access using I2C API. Mostly relevant for Nokia + Internet Tablets (770, N800 and N810). + + This driver can also be built as a module. If so, the module + will be called i2c-cbus-gpio. + config I2C_CPM tristate "Freescale CPM1 or CPM2 (MPC8xx/826x)" depends on (CPM1 || CPM2) && OF_I2C diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile index e5cb209d276c..6181f3ff263f 100644 --- a/drivers/i2c/busses/Makefile +++ b/drivers/i2c/busses/Makefile @@ -31,6 +31,7 @@ obj-$(CONFIG_I2C_POWERMAC) += i2c-powermac.o obj-$(CONFIG_I2C_AT91) += i2c-at91.o obj-$(CONFIG_I2C_AU1550) += i2c-au1550.o obj-$(CONFIG_I2C_BLACKFIN_TWI) += i2c-bfin-twi.o +obj-$(CONFIG_I2C_CBUS_GPIO) += i2c-cbus-gpio.o obj-$(CONFIG_I2C_CPM) += i2c-cpm.o obj-$(CONFIG_I2C_DAVINCI) += i2c-davinci.o obj-$(CONFIG_I2C_DESIGNWARE_CORE) += i2c-designware-core.o diff --git a/drivers/i2c/busses/i2c-at91.c b/drivers/i2c/busses/i2c-at91.c index c02bf208084f..b4575ee4bdf3 100644 --- a/drivers/i2c/busses/i2c-at91.c +++ b/drivers/i2c/busses/i2c-at91.c @@ -19,6 +19,8 @@ #include <linux/clk.h> #include <linux/completion.h> +#include <linux/dma-mapping.h> +#include <linux/dmaengine.h> #include <linux/err.h> #include <linux/i2c.h> #include <linux/interrupt.h> @@ -29,9 +31,11 @@ #include <linux/of_i2c.h> #include <linux/platform_device.h> #include <linux/slab.h> +#include <linux/platform_data/dma-atmel.h> #define TWI_CLK_HZ 100000 /* max 400 Kbits/s */ #define AT91_I2C_TIMEOUT msecs_to_jiffies(100) /* transfer timeout */ +#define AT91_I2C_DMA_THRESHOLD 8 /* enable DMA if transfer size is bigger than this threshold */ /* AT91 TWI register definitions */ #define AT91_TWI_CR 0x0000 /* Control Register */ @@ -66,24 +70,39 @@ #define AT91_TWI_THR 0x0034 /* Transmit Holding Register */ struct at91_twi_pdata { - unsigned clk_max_div; - unsigned clk_offset; - bool has_unre_flag; + unsigned clk_max_div; + unsigned clk_offset; + bool has_unre_flag; + bool has_dma_support; + struct at_dma_slave dma_slave; +}; + +struct at91_twi_dma { + struct dma_chan *chan_rx; + struct dma_chan *chan_tx; + struct scatterlist sg; + struct dma_async_tx_descriptor *data_desc; + enum dma_data_direction direction; + bool buf_mapped; + bool xfer_in_progress; }; struct at91_twi_dev { - struct device *dev; - void __iomem *base; - struct completion cmd_complete; - struct clk *clk; - u8 *buf; - size_t buf_len; - struct i2c_msg *msg; - int irq; - unsigned transfer_status; - struct i2c_adapter adapter; - unsigned twi_cwgr_reg; - struct at91_twi_pdata *pdata; + struct device *dev; + void __iomem *base; + struct completion cmd_complete; + struct clk *clk; + u8 *buf; + size_t buf_len; + struct i2c_msg *msg; + int irq; + unsigned imr; + unsigned transfer_status; + struct i2c_adapter adapter; + unsigned twi_cwgr_reg; + struct at91_twi_pdata *pdata; + bool use_dma; + struct at91_twi_dma dma; }; static unsigned at91_twi_read(struct at91_twi_dev *dev, unsigned reg) @@ -102,6 +121,17 @@ static void at91_disable_twi_interrupts(struct at91_twi_dev *dev) AT91_TWI_TXCOMP | AT91_TWI_RXRDY | AT91_TWI_TXRDY); } +static void at91_twi_irq_save(struct at91_twi_dev *dev) +{ + dev->imr = at91_twi_read(dev, AT91_TWI_IMR) & 0x7; + at91_disable_twi_interrupts(dev); +} + +static void at91_twi_irq_restore(struct at91_twi_dev *dev) +{ + at91_twi_write(dev, AT91_TWI_IER, dev->imr); +} + static void at91_init_twi_bus(struct at91_twi_dev *dev) { at91_disable_twi_interrupts(dev); @@ -138,6 +168,28 @@ static void __devinit at91_calc_twi_clock(struct at91_twi_dev *dev, int twi_clk) dev_dbg(dev->dev, "cdiv %d ckdiv %d\n", cdiv, ckdiv); } +static void at91_twi_dma_cleanup(struct at91_twi_dev *dev) +{ + struct at91_twi_dma *dma = &dev->dma; + + at91_twi_irq_save(dev); + + if (dma->xfer_in_progress) { + if (dma->direction == DMA_FROM_DEVICE) + dmaengine_terminate_all(dma->chan_rx); + else + dmaengine_terminate_all(dma->chan_tx); + dma->xfer_in_progress = false; + } + if (dma->buf_mapped) { + dma_unmap_single(dev->dev, sg_dma_address(&dma->sg), + dev->buf_len, dma->direction); + dma->buf_mapped = false; + } + + at91_twi_irq_restore(dev); +} + static void at91_twi_write_next_byte(struct at91_twi_dev *dev) { if (dev->buf_len <= 0) @@ -154,6 +206,60 @@ static void at91_twi_write_next_byte(struct at91_twi_dev *dev) ++dev->buf; } +static void at91_twi_write_data_dma_callback(void *data) +{ + struct at91_twi_dev *dev = (struct at91_twi_dev *)data; + + dma_unmap_single(dev->dev, sg_dma_address(&dev->dma.sg), + dev->buf_len, DMA_MEM_TO_DEV); + + at91_twi_write(dev, AT91_TWI_CR, AT91_TWI_STOP); +} + +static void at91_twi_write_data_dma(struct at91_twi_dev *dev) +{ + dma_addr_t dma_addr; + struct dma_async_tx_descriptor *txdesc; + struct at91_twi_dma *dma = &dev->dma; + struct dma_chan *chan_tx = dma->chan_tx; + + if (dev->buf_len <= 0) + return; + + dma->direction = DMA_TO_DEVICE; + + at91_twi_irq_save(dev); + dma_addr = dma_map_single(dev->dev, dev->buf, dev->buf_len, + DMA_TO_DEVICE); + if (dma_mapping_error(dev->dev, dma_addr)) { + dev_err(dev->dev, "dma map failed\n"); + return; + } + dma->buf_mapped = true; + at91_twi_irq_restore(dev); + sg_dma_len(&dma->sg) = dev->buf_len; + sg_dma_address(&dma->sg) = dma_addr; + + txdesc = dmaengine_prep_slave_sg(chan_tx, &dma->sg, 1, DMA_MEM_TO_DEV, + DMA_PREP_INTERRUPT | DMA_CTRL_ACK); + if (!txdesc) { + dev_err(dev->dev, "dma prep slave sg failed\n"); + goto error; + } + + txdesc->callback = at91_twi_write_data_dma_callback; + txdesc->callback_param = dev; + + dma->xfer_in_progress = true; + dmaengine_submit(txdesc); + dma_async_issue_pending(chan_tx); + + return; + +error: + at91_twi_dma_cleanup(dev); +} + static void at91_twi_read_next_byte(struct at91_twi_dev *dev) { if (dev->buf_len <= 0) @@ -179,6 +285,61 @@ static void at91_twi_read_next_byte(struct at91_twi_dev *dev) ++dev->buf; } +static void at91_twi_read_data_dma_callback(void *data) +{ + struct at91_twi_dev *dev = (struct at91_twi_dev *)data; + + dma_unmap_single(dev->dev, sg_dma_address(&dev->dma.sg), + dev->buf_len, DMA_DEV_TO_MEM); + + /* The last two bytes have to be read without using dma */ + dev->buf += dev->buf_len - 2; + dev->buf_len = 2; + at91_twi_write(dev, AT91_TWI_IER, AT91_TWI_RXRDY); +} + +static void at91_twi_read_data_dma(struct at91_twi_dev *dev) +{ + dma_addr_t dma_addr; + struct dma_async_tx_descriptor *rxdesc; + struct at91_twi_dma *dma = &dev->dma; + struct dma_chan *chan_rx = dma->chan_rx; + + dma->direction = DMA_FROM_DEVICE; + + /* Keep in mind that we won't use dma to read the last two bytes */ + at91_twi_irq_save(dev); + dma_addr = dma_map_single(dev->dev, dev->buf, dev->buf_len - 2, + DMA_FROM_DEVICE); + if (dma_mapping_error(dev->dev, dma_addr)) { + dev_err(dev->dev, "dma map failed\n"); + return; + } + dma->buf_mapped = true; + at91_twi_irq_restore(dev); + dma->sg.dma_address = dma_addr; + sg_dma_len(&dma->sg) = dev->buf_len - 2; + + rxdesc = dmaengine_prep_slave_sg(chan_rx, &dma->sg, 1, DMA_DEV_TO_MEM, + DMA_PREP_INTERRUPT | DMA_CTRL_ACK); + if (!rxdesc) { + dev_err(dev->dev, "dma prep slave sg failed\n"); + goto error; + } + + rxdesc->callback = at91_twi_read_data_dma_callback; + rxdesc->callback_param = dev; + + dma->xfer_in_progress = true; + dmaengine_submit(rxdesc); + dma_async_issue_pending(dma->chan_rx); + + return; + +error: + at91_twi_dma_cleanup(dev); +} + static irqreturn_t atmel_twi_interrupt(int irq, void *dev_id) { struct at91_twi_dev *dev = dev_id; @@ -229,12 +390,36 @@ static int at91_do_twi_transfer(struct at91_twi_dev *dev) if (dev->buf_len <= 1 && !(dev->msg->flags & I2C_M_RECV_LEN)) start_flags |= AT91_TWI_STOP; at91_twi_write(dev, AT91_TWI_CR, start_flags); - at91_twi_write(dev, AT91_TWI_IER, + /* + * When using dma, the last byte has to be read manually in + * order to not send the stop command too late and then + * to receive extra data. In practice, there are some issues + * if you use the dma to read n-1 bytes because of latency. + * Reading n-2 bytes with dma and the two last ones manually + * seems to be the best solution. + */ + if (dev->use_dma && (dev->buf_len > AT91_I2C_DMA_THRESHOLD)) { + at91_twi_read_data_dma(dev); + /* + * It is important to enable TXCOMP irq here because + * doing it only when transferring the last two bytes + * will mask NACK errors since TXCOMP is set when a + * NACK occurs. + */ + at91_twi_write(dev, AT91_TWI_IER, + AT91_TWI_TXCOMP); + } else + at91_twi_write(dev, AT91_TWI_IER, AT91_TWI_TXCOMP | AT91_TWI_RXRDY); } else { - at91_twi_write_next_byte(dev); - at91_twi_write(dev, AT91_TWI_IER, - AT91_TWI_TXCOMP | AT91_TWI_TXRDY); + if (dev->use_dma && (dev->buf_len > AT91_I2C_DMA_THRESHOLD)) { + at91_twi_write_data_dma(dev); + at91_twi_write(dev, AT91_TWI_IER, AT91_TWI_TXCOMP); + } else { + at91_twi_write_next_byte(dev); + at91_twi_write(dev, AT91_TWI_IER, + AT91_TWI_TXCOMP | AT91_TWI_TXRDY); + } } ret = wait_for_completion_interruptible_timeout(&dev->cmd_complete, @@ -242,23 +427,31 @@ static int at91_do_twi_transfer(struct at91_twi_dev *dev) if (ret == 0) { dev_err(dev->dev, "controller timed out\n"); at91_init_twi_bus(dev); - return -ETIMEDOUT; + ret = -ETIMEDOUT; + goto error; } if (dev->transfer_status & AT91_TWI_NACK) { dev_dbg(dev->dev, "received nack\n"); - return -EREMOTEIO; + ret = -EREMOTEIO; + goto error; } if (dev->transfer_status & AT91_TWI_OVRE) { dev_err(dev->dev, "overrun while reading\n"); - return -EIO; + ret = -EIO; + goto error; } if (has_unre_flag && dev->transfer_status & AT91_TWI_UNRE) { dev_err(dev->dev, "underrun while writing\n"); - return -EIO; + ret = -EIO; + goto error; } dev_dbg(dev->dev, "transfer complete\n"); return 0; + +error: + at91_twi_dma_cleanup(dev); + return ret; } static int at91_twi_xfer(struct i2c_adapter *adap, struct i2c_msg *msg, int num) @@ -329,36 +522,42 @@ static struct at91_twi_pdata at91rm9200_config = { .clk_max_div = 5, .clk_offset = 3, .has_unre_flag = true, + .has_dma_support = false, }; static struct at91_twi_pdata at91sam9261_config = { .clk_max_div = 5, .clk_offset = 4, .has_unre_flag = false, + .has_dma_support = false, }; static struct at91_twi_pdata at91sam9260_config = { .clk_max_div = 7, .clk_offset = 4, .has_unre_flag = false, + .has_dma_support = false, }; static struct at91_twi_pdata at91sam9g20_config = { .clk_max_div = 7, .clk_offset = 4, .has_unre_flag = false, + .has_dma_support = false, }; static struct at91_twi_pdata at91sam9g10_config = { .clk_max_div = 7, .clk_offset = 4, .has_unre_flag = false, + .has_dma_support = false, }; static struct at91_twi_pdata at91sam9x5_config = { .clk_max_div = 7, .clk_offset = 4, .has_unre_flag = false, + .has_dma_support = true, }; static const struct platform_device_id at91_twi_devtypes[] = { @@ -405,6 +604,90 @@ MODULE_DEVICE_TABLE(of, atmel_twi_dt_ids); #define atmel_twi_dt_ids NULL #endif +static bool __devinit filter(struct dma_chan *chan, void *slave) +{ + struct at_dma_slave *sl = slave; + + if (sl->dma_dev == chan->device->dev) { + chan->private = sl; + return true; + } else { + return false; + } +} + +static int __devinit at91_twi_configure_dma(struct at91_twi_dev *dev, u32 phy_addr) +{ + int ret = 0; + struct at_dma_slave *sdata; + struct dma_slave_config slave_config; + struct at91_twi_dma *dma = &dev->dma; + + sdata = &dev->pdata->dma_slave; + + memset(&slave_config, 0, sizeof(slave_config)); + slave_config.src_addr = (dma_addr_t)phy_addr + AT91_TWI_RHR; + slave_config.src_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE; + slave_config.src_maxburst = 1; + slave_config.dst_addr = (dma_addr_t)phy_addr + AT91_TWI_THR; + slave_config.dst_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE; + slave_config.dst_maxburst = 1; + slave_config.device_fc = false; + + if (sdata && sdata->dma_dev) { + dma_cap_mask_t mask; + + dma_cap_zero(mask); + dma_cap_set(DMA_SLAVE, mask); + dma->chan_tx = dma_request_channel(mask, filter, sdata); + if (!dma->chan_tx) { + dev_err(dev->dev, "no DMA channel available for tx\n"); + ret = -EBUSY; + goto error; + } + dma->chan_rx = dma_request_channel(mask, filter, sdata); + if (!dma->chan_rx) { + dev_err(dev->dev, "no DMA channel available for rx\n"); + ret = -EBUSY; + goto error; + } + } else { + ret = -EINVAL; + goto error; + } + + slave_config.direction = DMA_MEM_TO_DEV; + if (dmaengine_slave_config(dma->chan_tx, &slave_config)) { + dev_err(dev->dev, "failed to configure tx channel\n"); + ret = -EINVAL; + goto error; + } + + slave_config.direction = DMA_DEV_TO_MEM; + if (dmaengine_slave_config(dma->chan_rx, &slave_config)) { + dev_err(dev->dev, "failed to configure rx channel\n"); + ret = -EINVAL; + goto error; + } + + sg_init_table(&dma->sg, 1); + dma->buf_mapped = false; + dma->xfer_in_progress = false; + + dev_info(dev->dev, "using %s (tx) and %s (rx) for DMA transfers\n", + dma_chan_name(dma->chan_tx), dma_chan_name(dma->chan_rx)); + + return ret; + +error: + dev_info(dev->dev, "can't use DMA\n"); + if (dma->chan_rx) + dma_release_channel(dma->chan_rx); + if (dma->chan_tx) + dma_release_channel(dma->chan_tx); + return ret; +} + static struct at91_twi_pdata * __devinit at91_twi_get_driver_data( struct platform_device *pdev) { @@ -413,7 +696,7 @@ static struct at91_twi_pdata * __devinit at91_twi_get_driver_data( match = of_match_node(atmel_twi_dt_ids, pdev->dev.of_node); if (!match) return NULL; - return match->data; + return (struct at91_twi_pdata *)match->data; } return (struct at91_twi_pdata *) platform_get_device_id(pdev)->driver_data; } @@ -423,6 +706,7 @@ static int __devinit at91_twi_probe(struct platform_device *pdev) struct at91_twi_dev *dev; struct resource *mem; int rc; + u32 phy_addr; dev = devm_kzalloc(&pdev->dev, sizeof(*dev), GFP_KERNEL); if (!dev) @@ -433,6 +717,7 @@ static int __devinit at91_twi_probe(struct platform_device *pdev) mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!mem) return -ENODEV; + phy_addr = mem->start; dev->pdata = at91_twi_get_driver_data(pdev); if (!dev->pdata) @@ -462,6 +747,11 @@ static int __devinit at91_twi_probe(struct platform_device *pdev) } clk_prepare_enable(dev->clk); + if (dev->pdata->has_dma_support) { + if (at91_twi_configure_dma(dev, phy_addr) == 0) + dev->use_dma = true; + } + at91_calc_twi_clock(dev, TWI_CLK_HZ); at91_init_twi_bus(dev); diff --git a/drivers/i2c/busses/i2c-cbus-gpio.c b/drivers/i2c/busses/i2c-cbus-gpio.c new file mode 100644 index 000000000000..98386d659318 --- /dev/null +++ b/drivers/i2c/busses/i2c-cbus-gpio.c @@ -0,0 +1,300 @@ +/* + * CBUS I2C driver for Nokia Internet Tablets. + * + * Copyright (C) 2004-2010 Nokia Corporation + * + * Based on code written by Juha Yrjölä, David Weinehall, Mikko Ylinen and + * Felipe Balbi. Converted to I2C driver by Aaro Koskinen. + * + * This file is subject to the terms and conditions of the GNU General + * Public License. See the file "COPYING" in the main directory of this + * archive for more details. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/io.h> +#include <linux/i2c.h> +#include <linux/gpio.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/delay.h> +#include <linux/errno.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/of_gpio.h> +#include <linux/interrupt.h> +#include <linux/platform_device.h> +#include <linux/platform_data/i2c-cbus-gpio.h> + +/* + * Bit counts are derived from Nokia implementation. These should be checked + * if other CBUS implementations appear. + */ +#define CBUS_ADDR_BITS 3 +#define CBUS_REG_BITS 5 + +struct cbus_host { + spinlock_t lock; /* host lock */ + struct device *dev; + int clk_gpio; + int dat_gpio; + int sel_gpio; +}; + +/** + * cbus_send_bit - sends one bit over the bus + * @host: the host we're using + * @bit: one bit of information to send + */ +static void cbus_send_bit(struct cbus_host *host, unsigned bit) +{ + gpio_set_value(host->dat_gpio, bit ? 1 : 0); + gpio_set_value(host->clk_gpio, 1); + gpio_set_value(host->clk_gpio, 0); +} + +/** + * cbus_send_data - sends @len amount of data over the bus + * @host: the host we're using + * @data: the data to send + * @len: size of the transfer + */ +static void cbus_send_data(struct cbus_host *host, unsigned data, unsigned len) +{ + int i; + + for (i = len; i > 0; i--) + cbus_send_bit(host, data & (1 << (i - 1))); +} + +/** + * cbus_receive_bit - receives one bit from the bus + * @host: the host we're using + */ +static int cbus_receive_bit(struct cbus_host *host) +{ + int ret; + + gpio_set_value(host->clk_gpio, 1); + ret = gpio_get_value(host->dat_gpio); + gpio_set_value(host->clk_gpio, 0); + return ret; +} + +/** + * cbus_receive_word - receives 16-bit word from the bus + * @host: the host we're using + */ +static int cbus_receive_word(struct cbus_host *host) +{ + int ret = 0; + int i; + + for (i = 16; i > 0; i--) { + int bit = cbus_receive_bit(host); + + if (bit < 0) + return bit; + + if (bit) + ret |= 1 << (i - 1); + } + return ret; +} + +/** + * cbus_transfer - transfers data over the bus + * @host: the host we're using + * @rw: read/write flag + * @dev: device address + * @reg: register address + * @data: if @rw == I2C_SBUS_WRITE data to send otherwise 0 + */ +static int cbus_transfer(struct cbus_host *host, char rw, unsigned dev, + unsigned reg, unsigned data) +{ + unsigned long flags; + int ret; + + /* We don't want interrupts disturbing our transfer */ + spin_lock_irqsave(&host->lock, flags); + + /* Reset state and start of transfer, SEL stays down during transfer */ + gpio_set_value(host->sel_gpio, 0); + + /* Set the DAT pin to output */ + gpio_direction_output(host->dat_gpio, 1); + + /* Send the device address */ + cbus_send_data(host, dev, CBUS_ADDR_BITS); + + /* Send the rw flag */ + cbus_send_bit(host, rw == I2C_SMBUS_READ); + + /* Send the register address */ + cbus_send_data(host, reg, CBUS_REG_BITS); + + if (rw == I2C_SMBUS_WRITE) { + cbus_send_data(host, data, 16); + ret = 0; + } else { + ret = gpio_direction_input(host->dat_gpio); + if (ret) { + dev_dbg(host->dev, "failed setting direction\n"); + goto out; + } + gpio_set_value(host->clk_gpio, 1); + + ret = cbus_receive_word(host); + if (ret < 0) { + dev_dbg(host->dev, "failed receiving data\n"); + goto out; + } + } + + /* Indicate end of transfer, SEL goes up until next transfer */ + gpio_set_value(host->sel_gpio, 1); + gpio_set_value(host->clk_gpio, 1); + gpio_set_value(host->clk_gpio, 0); + +out: + spin_unlock_irqrestore(&host->lock, flags); + + return ret; +} + +static int cbus_i2c_smbus_xfer(struct i2c_adapter *adapter, + u16 addr, + unsigned short flags, + char read_write, + u8 command, + int size, + union i2c_smbus_data *data) +{ + struct cbus_host *chost = i2c_get_adapdata(adapter); + int ret; + + if (size != I2C_SMBUS_WORD_DATA) + return -EINVAL; + + ret = cbus_transfer(chost, read_write == I2C_SMBUS_READ, addr, + command, data->word); + if (ret < 0) + return ret; + + if (read_write == I2C_SMBUS_READ) + data->word = ret; + + return 0; +} + +static u32 cbus_i2c_func(struct i2c_adapter *adapter) +{ + return I2C_FUNC_SMBUS_READ_WORD_DATA | I2C_FUNC_SMBUS_WRITE_WORD_DATA; +} + +static const struct i2c_algorithm cbus_i2c_algo = { + .smbus_xfer = cbus_i2c_smbus_xfer, + .functionality = cbus_i2c_func, +}; + +static int cbus_i2c_remove(struct platform_device *pdev) +{ + struct i2c_adapter *adapter = platform_get_drvdata(pdev); + + return i2c_del_adapter(adapter); +} + +static int cbus_i2c_probe(struct platform_device *pdev) +{ + struct i2c_adapter *adapter; + struct cbus_host *chost; + int ret; + + adapter = devm_kzalloc(&pdev->dev, sizeof(struct i2c_adapter), + GFP_KERNEL); + if (!adapter) + return -ENOMEM; + + chost = devm_kzalloc(&pdev->dev, sizeof(*chost), GFP_KERNEL); + if (!chost) + return -ENOMEM; + + if (pdev->dev.of_node) { + struct device_node *dnode = pdev->dev.of_node; + if (of_gpio_count(dnode) != 3) + return -ENODEV; + chost->clk_gpio = of_get_gpio(dnode, 0); + chost->dat_gpio = of_get_gpio(dnode, 1); + chost->sel_gpio = of_get_gpio(dnode, 2); + } else if (pdev->dev.platform_data) { + struct i2c_cbus_platform_data *pdata = pdev->dev.platform_data; + chost->clk_gpio = pdata->clk_gpio; + chost->dat_gpio = pdata->dat_gpio; + chost->sel_gpio = pdata->sel_gpio; + } else { + return -ENODEV; + } + + adapter->owner = THIS_MODULE; + adapter->class = I2C_CLASS_HWMON; + adapter->dev.parent = &pdev->dev; + adapter->nr = pdev->id; + adapter->timeout = HZ; + adapter->algo = &cbus_i2c_algo; + strlcpy(adapter->name, "CBUS I2C adapter", sizeof(adapter->name)); + + spin_lock_init(&chost->lock); + chost->dev = &pdev->dev; + + ret = devm_gpio_request_one(&pdev->dev, chost->clk_gpio, + GPIOF_OUT_INIT_LOW, "CBUS clk"); + if (ret) + return ret; + + ret = devm_gpio_request_one(&pdev->dev, chost->dat_gpio, GPIOF_IN, + "CBUS data"); + if (ret) + return ret; + + ret = devm_gpio_request_one(&pdev->dev, chost->sel_gpio, + GPIOF_OUT_INIT_HIGH, "CBUS sel"); + if (ret) + return ret; + + i2c_set_adapdata(adapter, chost); + platform_set_drvdata(pdev, adapter); + + return i2c_add_numbered_adapter(adapter); +} + +#if defined(CONFIG_OF) +static const struct of_device_id i2c_cbus_dt_ids[] = { + { .compatible = "i2c-cbus-gpio", }, + { } +}; +MODULE_DEVICE_TABLE(of, i2c_cbus_dt_ids); +#endif + +static struct platform_driver cbus_i2c_driver = { + .probe = cbus_i2c_probe, + .remove = cbus_i2c_remove, + .driver = { + .owner = THIS_MODULE, + .name = "i2c-cbus-gpio", + }, +}; +module_platform_driver(cbus_i2c_driver); + +MODULE_ALIAS("platform:i2c-cbus-gpio"); +MODULE_DESCRIPTION("CBUS I2C driver"); +MODULE_AUTHOR("Juha Yrjölä"); +MODULE_AUTHOR("David Weinehall"); +MODULE_AUTHOR("Mikko Ylinen"); +MODULE_AUTHOR("Felipe Balbi"); +MODULE_AUTHOR("Aaro Koskinen <aaro.koskinen@iki.fi>"); +MODULE_LICENSE("GPL"); diff --git a/drivers/i2c/busses/i2c-gpio.c b/drivers/i2c/busses/i2c-gpio.c index e62d2d938628..257299a92df3 100644 --- a/drivers/i2c/busses/i2c-gpio.c +++ b/drivers/i2c/busses/i2c-gpio.c @@ -184,7 +184,11 @@ static int __devinit i2c_gpio_probe(struct platform_device *pdev) bit_data->data = pdata; adap->owner = THIS_MODULE; - snprintf(adap->name, sizeof(adap->name), "i2c-gpio%d", pdev->id); + if (pdev->dev.of_node) + strlcpy(adap->name, dev_name(&pdev->dev), sizeof(adap->name)); + else + snprintf(adap->name, sizeof(adap->name), "i2c-gpio%d", pdev->id); + adap->algo_data = bit_data; adap->class = I2C_CLASS_HWMON | I2C_CLASS_SPD; adap->dev.parent = &pdev->dev; diff --git a/drivers/i2c/busses/i2c-mxs.c b/drivers/i2c/busses/i2c-mxs.c index 0670da79ee5e..6ed53da9e1f4 100644 --- a/drivers/i2c/busses/i2c-mxs.c +++ b/drivers/i2c/busses/i2c-mxs.c @@ -359,7 +359,7 @@ static int mxs_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], static u32 mxs_i2c_func(struct i2c_adapter *adap) { - return I2C_FUNC_I2C | (I2C_FUNC_SMBUS_EMUL & ~I2C_FUNC_SMBUS_QUICK); + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; } static irqreturn_t mxs_i2c_isr(int this_irq, void *dev_id) diff --git a/drivers/i2c/busses/i2c-nomadik.c b/drivers/i2c/busses/i2c-nomadik.c index 02c3115a2dfa..8b2ffcf45322 100644 --- a/drivers/i2c/busses/i2c-nomadik.c +++ b/drivers/i2c/busses/i2c-nomadik.c @@ -435,13 +435,6 @@ static int read_i2c(struct nmk_i2c_dev *dev, u16 flags) timeout = wait_for_completion_timeout( &dev->xfer_complete, dev->adap.timeout); - if (timeout < 0) { - dev_err(&dev->adev->dev, - "wait_for_completion_timeout " - "returned %d waiting for event\n", timeout); - status = timeout; - } - if (timeout == 0) { /* Controller timed out */ dev_err(&dev->adev->dev, "read from slave 0x%x timed out\n", @@ -523,13 +516,6 @@ static int write_i2c(struct nmk_i2c_dev *dev, u16 flags) timeout = wait_for_completion_timeout( &dev->xfer_complete, dev->adap.timeout); - if (timeout < 0) { - dev_err(&dev->adev->dev, - "wait_for_completion_timeout " - "returned %d waiting for event\n", timeout); - status = timeout; - } - if (timeout == 0) { /* Controller timed out */ dev_err(&dev->adev->dev, "write to slave 0x%x timed out\n", diff --git a/drivers/i2c/busses/i2c-ocores.c b/drivers/i2c/busses/i2c-ocores.c index 15da1ac7cf9e..9b35c9fbb2fe 100644 --- a/drivers/i2c/busses/i2c-ocores.c +++ b/drivers/i2c/busses/i2c-ocores.c @@ -4,6 +4,9 @@ * * Peter Korsgaard <jacmet@sunsite.dk> * + * Support for the GRLIB port of the controller by + * Andreas Larsson <andreas@gaisler.com> + * * This file is licensed under the terms of the GNU General Public License * version 2. This program is licensed "as is" without any warranty of any * kind, whether express or implied. @@ -34,6 +37,8 @@ struct ocores_i2c { int nmsgs; int state; /* see STATE_ */ int clock_khz; + void (*setreg)(struct ocores_i2c *i2c, int reg, u8 value); + u8 (*getreg)(struct ocores_i2c *i2c, int reg); }; /* registers */ @@ -67,24 +72,47 @@ struct ocores_i2c { #define STATE_READ 3 #define STATE_ERROR 4 +#define TYPE_OCORES 0 +#define TYPE_GRLIB 1 + +static void oc_setreg_8(struct ocores_i2c *i2c, int reg, u8 value) +{ + iowrite8(value, i2c->base + (reg << i2c->reg_shift)); +} + +static void oc_setreg_16(struct ocores_i2c *i2c, int reg, u8 value) +{ + iowrite16(value, i2c->base + (reg << i2c->reg_shift)); +} + +static void oc_setreg_32(struct ocores_i2c *i2c, int reg, u8 value) +{ + iowrite32(value, i2c->base + (reg << i2c->reg_shift)); +} + +static inline u8 oc_getreg_8(struct ocores_i2c *i2c, int reg) +{ + return ioread8(i2c->base + (reg << i2c->reg_shift)); +} + +static inline u8 oc_getreg_16(struct ocores_i2c *i2c, int reg) +{ + return ioread16(i2c->base + (reg << i2c->reg_shift)); +} + +static inline u8 oc_getreg_32(struct ocores_i2c *i2c, int reg) +{ + return ioread32(i2c->base + (reg << i2c->reg_shift)); +} + static inline void oc_setreg(struct ocores_i2c *i2c, int reg, u8 value) { - if (i2c->reg_io_width == 4) - iowrite32(value, i2c->base + (reg << i2c->reg_shift)); - else if (i2c->reg_io_width == 2) - iowrite16(value, i2c->base + (reg << i2c->reg_shift)); - else - iowrite8(value, i2c->base + (reg << i2c->reg_shift)); + i2c->setreg(i2c, reg, value); } static inline u8 oc_getreg(struct ocores_i2c *i2c, int reg) { - if (i2c->reg_io_width == 4) - return ioread32(i2c->base + (reg << i2c->reg_shift)); - else if (i2c->reg_io_width == 2) - return ioread16(i2c->base + (reg << i2c->reg_shift)); - else - return ioread8(i2c->base + (reg << i2c->reg_shift)); + return i2c->getreg(i2c, reg); } static void ocores_process(struct ocores_i2c *i2c) @@ -223,11 +251,59 @@ static struct i2c_adapter ocores_adapter = { .algo = &ocores_algorithm, }; +static struct of_device_id ocores_i2c_match[] = { + { + .compatible = "opencores,i2c-ocores", + .data = (void *)TYPE_OCORES, + }, + { + .compatible = "aeroflexgaisler,i2cmst", + .data = (void *)TYPE_GRLIB, + }, + {}, +}; +MODULE_DEVICE_TABLE(of, ocores_i2c_match); + #ifdef CONFIG_OF +/* Read and write functions for the GRLIB port of the controller. Registers are + * 32-bit big endian and the PRELOW and PREHIGH registers are merged into one + * register. The subsequent registers has their offset decreased accordingly. */ +static u8 oc_getreg_grlib(struct ocores_i2c *i2c, int reg) +{ + u32 rd; + int rreg = reg; + if (reg != OCI2C_PRELOW) + rreg--; + rd = ioread32be(i2c->base + (rreg << i2c->reg_shift)); + if (reg == OCI2C_PREHIGH) + return (u8)(rd >> 8); + else + return (u8)rd; +} + +static void oc_setreg_grlib(struct ocores_i2c *i2c, int reg, u8 value) +{ + u32 curr, wr; + int rreg = reg; + if (reg != OCI2C_PRELOW) + rreg--; + if (reg == OCI2C_PRELOW || reg == OCI2C_PREHIGH) { + curr = ioread32be(i2c->base + (rreg << i2c->reg_shift)); + if (reg == OCI2C_PRELOW) + wr = (curr & 0xff00) | value; + else + wr = (((u32)value) << 8) | (curr & 0xff); + } else { + wr = value; + } + iowrite32be(wr, i2c->base + (rreg << i2c->reg_shift)); +} + static int ocores_i2c_of_probe(struct platform_device *pdev, struct ocores_i2c *i2c) { struct device_node *np = pdev->dev.of_node; + const struct of_device_id *match; u32 val; if (of_property_read_u32(np, "reg-shift", &i2c->reg_shift)) { @@ -253,6 +329,14 @@ static int ocores_i2c_of_probe(struct platform_device *pdev, of_property_read_u32(pdev->dev.of_node, "reg-io-width", &i2c->reg_io_width); + + match = of_match_node(ocores_i2c_match, pdev->dev.of_node); + if (match && (int)match->data == TYPE_GRLIB) { + dev_dbg(&pdev->dev, "GRLIB variant of i2c-ocores\n"); + i2c->setreg = oc_setreg_grlib; + i2c->getreg = oc_getreg_grlib; + } + return 0; } #else @@ -263,7 +347,8 @@ static int __devinit ocores_i2c_probe(struct platform_device *pdev) { struct ocores_i2c *i2c; struct ocores_i2c_platform_data *pdata; - struct resource *res, *res2; + struct resource *res; + int irq; int ret; int i; @@ -271,26 +356,17 @@ static int __devinit ocores_i2c_probe(struct platform_device *pdev) if (!res) return -ENODEV; - res2 = platform_get_resource(pdev, IORESOURCE_IRQ, 0); - if (!res2) - return -ENODEV; + irq = platform_get_irq(pdev, 0); + if (irq < 0) + return irq; i2c = devm_kzalloc(&pdev->dev, sizeof(*i2c), GFP_KERNEL); if (!i2c) return -ENOMEM; - if (!devm_request_mem_region(&pdev->dev, res->start, - resource_size(res), pdev->name)) { - dev_err(&pdev->dev, "Memory region busy\n"); - return -EBUSY; - } - - i2c->base = devm_ioremap_nocache(&pdev->dev, res->start, - resource_size(res)); - if (!i2c->base) { - dev_err(&pdev->dev, "Unable to map registers\n"); - return -EIO; - } + i2c->base = devm_request_and_ioremap(&pdev->dev, res); + if (!i2c->base) + return -EADDRNOTAVAIL; pdata = pdev->dev.platform_data; if (pdata) { @@ -306,10 +382,34 @@ static int __devinit ocores_i2c_probe(struct platform_device *pdev) if (i2c->reg_io_width == 0) i2c->reg_io_width = 1; /* Set to default value */ + if (!i2c->setreg || !i2c->getreg) { + switch (i2c->reg_io_width) { + case 1: + i2c->setreg = oc_setreg_8; + i2c->getreg = oc_getreg_8; + break; + + case 2: + i2c->setreg = oc_setreg_16; + i2c->getreg = oc_getreg_16; + break; + + case 4: + i2c->setreg = oc_setreg_32; + i2c->getreg = oc_getreg_32; + break; + + default: + dev_err(&pdev->dev, "Unsupported I/O width (%d)\n", + i2c->reg_io_width); + return -EINVAL; + } + } + ocores_init(i2c); init_waitqueue_head(&i2c->wait); - ret = devm_request_irq(&pdev->dev, res2->start, ocores_isr, 0, + ret = devm_request_irq(&pdev->dev, irq, ocores_isr, 0, pdev->name, i2c); if (ret) { dev_err(&pdev->dev, "Cannot claim IRQ\n"); @@ -383,12 +483,6 @@ static SIMPLE_DEV_PM_OPS(ocores_i2c_pm, ocores_i2c_suspend, ocores_i2c_resume); #define OCORES_I2C_PM NULL #endif -static struct of_device_id ocores_i2c_match[] = { - { .compatible = "opencores,i2c-ocores", }, - {}, -}; -MODULE_DEVICE_TABLE(of, ocores_i2c_match); - static struct platform_driver ocores_i2c_driver = { .probe = ocores_i2c_probe, .remove = __devexit_p(ocores_i2c_remove), diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c index 3525c9e62cb0..7a62acb7d262 100644 --- a/drivers/i2c/busses/i2c-omap.c +++ b/drivers/i2c/busses/i2c-omap.c @@ -43,14 +43,16 @@ #include <linux/slab.h> #include <linux/i2c-omap.h> #include <linux/pm_runtime.h> +#include <linux/pinctrl/consumer.h> /* I2C controller revisions */ #define OMAP_I2C_OMAP1_REV_2 0x20 /* I2C controller revisions present on specific hardware */ -#define OMAP_I2C_REV_ON_2430 0x36 -#define OMAP_I2C_REV_ON_3430_3530 0x3C -#define OMAP_I2C_REV_ON_3630_4430 0x40 +#define OMAP_I2C_REV_ON_2430 0x00000036 +#define OMAP_I2C_REV_ON_3430_3530 0x0000003C +#define OMAP_I2C_REV_ON_3630 0x00000040 +#define OMAP_I2C_REV_ON_4430_PLUS 0x50400002 /* timeout waiting for the controller to respond */ #define OMAP_I2C_TIMEOUT (msecs_to_jiffies(1000)) @@ -190,7 +192,6 @@ struct omap_i2c_dev { void (*set_mpu_wkup_lat)(struct device *dev, long latency); u32 speed; /* Speed of bus in kHz */ - u32 dtrev; /* extra revision from DT */ u32 flags; u16 cmd_err; u8 *buf; @@ -202,17 +203,18 @@ struct omap_i2c_dev { * fifo_size==0 implies no fifo * if set, should be trsh+1 */ - u8 rev; + u32 rev; unsigned b_hw:1; /* bad h/w fixes */ unsigned receiver:1; /* true when we're in receiver mode */ u16 iestate; /* Saved interrupt register */ u16 pscstate; u16 scllstate; u16 sclhstate; - u16 bufstate; u16 syscstate; u16 westate; u16 errata; + + struct pinctrl *pins; }; static const u8 reg_map_ip_v1[] = { @@ -275,16 +277,39 @@ static inline u16 omap_i2c_read_reg(struct omap_i2c_dev *i2c_dev, int reg) (i2c_dev->regs[reg] << i2c_dev->reg_shift)); } -static int omap_i2c_init(struct omap_i2c_dev *dev) +static void __omap_i2c_init(struct omap_i2c_dev *dev) +{ + + omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, 0); + + /* Setup clock prescaler to obtain approx 12MHz I2C module clock: */ + omap_i2c_write_reg(dev, OMAP_I2C_PSC_REG, dev->pscstate); + + /* SCL low and high time values */ + omap_i2c_write_reg(dev, OMAP_I2C_SCLL_REG, dev->scllstate); + omap_i2c_write_reg(dev, OMAP_I2C_SCLH_REG, dev->sclhstate); + if (dev->rev >= OMAP_I2C_REV_ON_3430_3530) + omap_i2c_write_reg(dev, OMAP_I2C_WE_REG, dev->westate); + + /* Take the I2C module out of reset: */ + omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, OMAP_I2C_CON_EN); + + /* + * Don't write to this register if the IE state is 0 as it can + * cause deadlock. + */ + if (dev->iestate) + omap_i2c_write_reg(dev, OMAP_I2C_IE_REG, dev->iestate); +} + +static int omap_i2c_reset(struct omap_i2c_dev *dev) { - u16 psc = 0, scll = 0, sclh = 0, buf = 0; - u16 fsscll = 0, fssclh = 0, hsscll = 0, hssclh = 0; - unsigned long fclk_rate = 12000000; unsigned long timeout; - unsigned long internal_clk = 0; - struct clk *fclk; + u16 sysc; if (dev->rev >= OMAP_I2C_OMAP1_REV_2) { + sysc = omap_i2c_read_reg(dev, OMAP_I2C_SYSC_REG); + /* Disable I2C controller before soft reset */ omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, omap_i2c_read_reg(dev, OMAP_I2C_CON_REG) & @@ -306,32 +331,28 @@ static int omap_i2c_init(struct omap_i2c_dev *dev) } /* SYSC register is cleared by the reset; rewrite it */ - if (dev->rev == OMAP_I2C_REV_ON_2430) { - - omap_i2c_write_reg(dev, OMAP_I2C_SYSC_REG, - SYSC_AUTOIDLE_MASK); - - } else if (dev->rev >= OMAP_I2C_REV_ON_3430_3530) { - dev->syscstate = SYSC_AUTOIDLE_MASK; - dev->syscstate |= SYSC_ENAWAKEUP_MASK; - dev->syscstate |= (SYSC_IDLEMODE_SMART << - __ffs(SYSC_SIDLEMODE_MASK)); - dev->syscstate |= (SYSC_CLOCKACTIVITY_FCLK << - __ffs(SYSC_CLOCKACTIVITY_MASK)); - - omap_i2c_write_reg(dev, OMAP_I2C_SYSC_REG, - dev->syscstate); - /* - * Enabling all wakup sources to stop I2C freezing on - * WFI instruction. - * REVISIT: Some wkup sources might not be needed. - */ - dev->westate = OMAP_I2C_WE_ALL; - omap_i2c_write_reg(dev, OMAP_I2C_WE_REG, - dev->westate); - } + omap_i2c_write_reg(dev, OMAP_I2C_SYSC_REG, sysc); + + } + return 0; +} + +static int omap_i2c_init(struct omap_i2c_dev *dev) +{ + u16 psc = 0, scll = 0, sclh = 0; + u16 fsscll = 0, fssclh = 0, hsscll = 0, hssclh = 0; + unsigned long fclk_rate = 12000000; + unsigned long internal_clk = 0; + struct clk *fclk; + + if (dev->rev >= OMAP_I2C_REV_ON_3430_3530) { + /* + * Enabling all wakup sources to stop I2C freezing on + * WFI instruction. + * REVISIT: Some wkup sources might not be needed. + */ + dev->westate = OMAP_I2C_WE_ALL; } - omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, 0); if (dev->flags & OMAP_I2C_FLAG_ALWAYS_ARMXOR_CLK) { /* @@ -416,28 +437,17 @@ static int omap_i2c_init(struct omap_i2c_dev *dev) sclh = fclk_rate / (dev->speed * 2) - 7 + psc; } - /* Setup clock prescaler to obtain approx 12MHz I2C module clock: */ - omap_i2c_write_reg(dev, OMAP_I2C_PSC_REG, psc); - - /* SCL low and high time values */ - omap_i2c_write_reg(dev, OMAP_I2C_SCLL_REG, scll); - omap_i2c_write_reg(dev, OMAP_I2C_SCLH_REG, sclh); - - /* Take the I2C module out of reset: */ - omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, OMAP_I2C_CON_EN); - - /* Enable interrupts */ dev->iestate = (OMAP_I2C_IE_XRDY | OMAP_I2C_IE_RRDY | OMAP_I2C_IE_ARDY | OMAP_I2C_IE_NACK | OMAP_I2C_IE_AL) | ((dev->fifo_size) ? (OMAP_I2C_IE_RDR | OMAP_I2C_IE_XDR) : 0); - omap_i2c_write_reg(dev, OMAP_I2C_IE_REG, dev->iestate); - if (dev->flags & OMAP_I2C_FLAG_RESET_REGS_POSTIDLE) { - dev->pscstate = psc; - dev->scllstate = scll; - dev->sclhstate = sclh; - dev->bufstate = buf; - } + + dev->pscstate = psc; + dev->scllstate = scll; + dev->sclhstate = sclh; + + __omap_i2c_init(dev); + return 0; } @@ -490,7 +500,7 @@ static void omap_i2c_resize_fifo(struct omap_i2c_dev *dev, u8 size, bool is_rx) omap_i2c_write_reg(dev, OMAP_I2C_BUF_REG, buf); - if (dev->rev < OMAP_I2C_REV_ON_3630_4430) + if (dev->rev < OMAP_I2C_REV_ON_3630) dev->b_hw = 1; /* Enable hardware fixes */ /* calculate wakeup latency constraint for MPU */ @@ -586,7 +596,8 @@ static int omap_i2c_xfer_msg(struct i2c_adapter *adap, OMAP_I2C_TIMEOUT); if (timeout == 0) { dev_err(dev->dev, "controller timed out\n"); - omap_i2c_init(dev); + omap_i2c_reset(dev); + __omap_i2c_init(dev); return -ETIMEDOUT; } @@ -596,7 +607,8 @@ static int omap_i2c_xfer_msg(struct i2c_adapter *adap, /* We have an error */ if (dev->cmd_err & (OMAP_I2C_STAT_AL | OMAP_I2C_STAT_ROVR | OMAP_I2C_STAT_XUDF)) { - omap_i2c_init(dev); + omap_i2c_reset(dev); + __omap_i2c_init(dev); return -EIO; } @@ -642,13 +654,14 @@ omap_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], int num) break; } - if (dev->set_mpu_wkup_lat != NULL) - dev->set_mpu_wkup_lat(dev->dev, -1); - if (r == 0) r = num; omap_i2c_wait_for_bb(dev); + + if (dev->set_mpu_wkup_lat != NULL) + dev->set_mpu_wkup_lat(dev->dev, -1); + out: pm_runtime_mark_last_busy(dev->dev); pm_runtime_put_autosuspend(dev->dev); @@ -1025,9 +1038,7 @@ static const struct i2c_algorithm omap_i2c_algo = { #ifdef CONFIG_OF static struct omap_i2c_bus_platform_data omap3_pdata = { .rev = OMAP_I2C_IP_VERSION_1, - .flags = OMAP_I2C_FLAG_APPLY_ERRATA_I207 | - OMAP_I2C_FLAG_RESET_REGS_POSTIDLE | - OMAP_I2C_FLAG_BUS_SHIFT_2, + .flags = OMAP_I2C_FLAG_BUS_SHIFT_2, }; static struct omap_i2c_bus_platform_data omap4_pdata = { @@ -1048,6 +1059,16 @@ static const struct of_device_id omap_i2c_of_match[] = { MODULE_DEVICE_TABLE(of, omap_i2c_of_match); #endif +#define OMAP_I2C_SCHEME(rev) ((rev & 0xc000) >> 14) + +#define OMAP_I2C_REV_SCHEME_0_MAJOR(rev) (rev >> 4) +#define OMAP_I2C_REV_SCHEME_0_MINOR(rev) (rev & 0xf) + +#define OMAP_I2C_REV_SCHEME_1_MAJOR(rev) ((rev & 0x0700) >> 7) +#define OMAP_I2C_REV_SCHEME_1_MINOR(rev) (rev & 0x1f) +#define OMAP_I2C_SCHEME_0 0 +#define OMAP_I2C_SCHEME_1 1 + static int __devinit omap_i2c_probe(struct platform_device *pdev) { @@ -1060,6 +1081,8 @@ omap_i2c_probe(struct platform_device *pdev) const struct of_device_id *match; int irq; int r; + u32 rev; + u16 minor, major, scheme; /* NOTE: driver uses the static register mapping */ mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); @@ -1091,7 +1114,6 @@ omap_i2c_probe(struct platform_device *pdev) u32 freq = 100000; /* default to 100000 Hz */ pdata = match->data; - dev->dtrev = pdata->rev; dev->flags = pdata->flags; of_property_read_u32(node, "clock-frequency", &freq); @@ -1101,7 +1123,16 @@ omap_i2c_probe(struct platform_device *pdev) dev->speed = pdata->clkrate; dev->flags = pdata->flags; dev->set_mpu_wkup_lat = pdata->set_mpu_wkup_lat; - dev->dtrev = pdata->rev; + } + + dev->pins = devm_pinctrl_get_select_default(&pdev->dev); + if (IS_ERR(dev->pins)) { + if (PTR_ERR(dev->pins) == -EPROBE_DEFER) + return -EPROBE_DEFER; + + dev_warn(&pdev->dev, "did not get pins for i2c error: %li\n", + PTR_ERR(dev->pins)); + dev->pins = NULL; } dev->dev = &pdev->dev; @@ -1114,11 +1145,6 @@ omap_i2c_probe(struct platform_device *pdev) dev->reg_shift = (dev->flags >> OMAP_I2C_FLAG_BUS_SHIFT__SHIFT) & 3; - if (dev->dtrev == OMAP_I2C_IP_VERSION_2) - dev->regs = (u8 *)reg_map_ip_v2; - else - dev->regs = (u8 *)reg_map_ip_v1; - pm_runtime_enable(dev->dev); pm_runtime_set_autosuspend_delay(dev->dev, OMAP_I2C_PM_TIMEOUT); pm_runtime_use_autosuspend(dev->dev); @@ -1127,11 +1153,37 @@ omap_i2c_probe(struct platform_device *pdev) if (IS_ERR_VALUE(r)) goto err_free_mem; - dev->rev = omap_i2c_read_reg(dev, OMAP_I2C_REV_REG) & 0xff; + /* + * Read the Rev hi bit-[15:14] ie scheme this is 1 indicates ver2. + * On omap1/3/2 Offset 4 is IE Reg the bit [15:14] is 0 at reset. + * Also since the omap_i2c_read_reg uses reg_map_ip_* a + * raw_readw is done. + */ + rev = __raw_readw(dev->base + 0x04); + + scheme = OMAP_I2C_SCHEME(rev); + switch (scheme) { + case OMAP_I2C_SCHEME_0: + dev->regs = (u8 *)reg_map_ip_v1; + dev->rev = omap_i2c_read_reg(dev, OMAP_I2C_REV_REG); + minor = OMAP_I2C_REV_SCHEME_0_MAJOR(dev->rev); + major = OMAP_I2C_REV_SCHEME_0_MAJOR(dev->rev); + break; + case OMAP_I2C_SCHEME_1: + /* FALLTHROUGH */ + default: + dev->regs = (u8 *)reg_map_ip_v2; + rev = (rev << 16) | + omap_i2c_read_reg(dev, OMAP_I2C_IP_V2_REVNB_LO); + minor = OMAP_I2C_REV_SCHEME_1_MINOR(rev); + major = OMAP_I2C_REV_SCHEME_1_MAJOR(rev); + dev->rev = rev; + } dev->errata = 0; - if (dev->flags & OMAP_I2C_FLAG_APPLY_ERRATA_I207) + if (dev->rev >= OMAP_I2C_REV_ON_2430 && + dev->rev < OMAP_I2C_REV_ON_4430_PLUS) dev->errata |= I2C_OMAP_ERRATA_I207; if (dev->rev <= OMAP_I2C_REV_ON_3430_3530) @@ -1152,7 +1204,7 @@ omap_i2c_probe(struct platform_device *pdev) dev->fifo_size = (dev->fifo_size / 2); - if (dev->rev < OMAP_I2C_REV_ON_3630_4430) + if (dev->rev < OMAP_I2C_REV_ON_3630) dev->b_hw = 1; /* Enable hardware fixes */ /* calculate wakeup latency constraint for MPU */ @@ -1195,8 +1247,8 @@ omap_i2c_probe(struct platform_device *pdev) goto err_unuse_clocks; } - dev_info(dev->dev, "bus %d rev%d.%d.%d at %d kHz\n", adap->nr, - dev->dtrev, dev->rev >> 4, dev->rev & 0xf, dev->speed); + dev_info(dev->dev, "bus %d rev%d.%d at %d kHz\n", adap->nr, + major, minor, dev->speed); of_i2c_register_devices(adap); @@ -1239,14 +1291,13 @@ static int omap_i2c_runtime_suspend(struct device *dev) { struct platform_device *pdev = to_platform_device(dev); struct omap_i2c_dev *_dev = platform_get_drvdata(pdev); - u16 iv; _dev->iestate = omap_i2c_read_reg(_dev, OMAP_I2C_IE_REG); omap_i2c_write_reg(_dev, OMAP_I2C_IE_REG, 0); if (_dev->rev < OMAP_I2C_OMAP1_REV_2) { - iv = omap_i2c_read_reg(_dev, OMAP_I2C_IV_REG); /* Read clears */ + omap_i2c_read_reg(_dev, OMAP_I2C_IV_REG); /* Read clears */ } else { omap_i2c_write_reg(_dev, OMAP_I2C_STAT_REG, _dev->iestate); @@ -1262,23 +1313,10 @@ static int omap_i2c_runtime_resume(struct device *dev) struct platform_device *pdev = to_platform_device(dev); struct omap_i2c_dev *_dev = platform_get_drvdata(pdev); - if (_dev->flags & OMAP_I2C_FLAG_RESET_REGS_POSTIDLE) { - omap_i2c_write_reg(_dev, OMAP_I2C_CON_REG, 0); - omap_i2c_write_reg(_dev, OMAP_I2C_PSC_REG, _dev->pscstate); - omap_i2c_write_reg(_dev, OMAP_I2C_SCLL_REG, _dev->scllstate); - omap_i2c_write_reg(_dev, OMAP_I2C_SCLH_REG, _dev->sclhstate); - omap_i2c_write_reg(_dev, OMAP_I2C_BUF_REG, _dev->bufstate); - omap_i2c_write_reg(_dev, OMAP_I2C_SYSC_REG, _dev->syscstate); - omap_i2c_write_reg(_dev, OMAP_I2C_WE_REG, _dev->westate); - omap_i2c_write_reg(_dev, OMAP_I2C_CON_REG, OMAP_I2C_CON_EN); - } + if (!_dev->regs) + return 0; - /* - * Don't write to this register if the IE state is 0 as it can - * cause deadlock. - */ - if (_dev->iestate) - omap_i2c_write_reg(_dev, OMAP_I2C_IE_REG, _dev->iestate); + __omap_i2c_init(_dev); return 0; } diff --git a/drivers/i2c/busses/i2c-rcar.c b/drivers/i2c/busses/i2c-rcar.c index f9399d163af2..72a8071a5556 100644 --- a/drivers/i2c/busses/i2c-rcar.c +++ b/drivers/i2c/busses/i2c-rcar.c @@ -642,7 +642,7 @@ static int __devinit rcar_i2c_probe(struct platform_device *pdev) if (ret < 0) return ret; - priv->io = devm_ioremap(dev, res->start, resource_size(res)); + priv->io = devm_request_and_ioremap(dev, res); if (!priv->io) { dev_err(dev, "cannot ioremap\n"); return -ENODEV; @@ -693,7 +693,7 @@ static int __devexit rcar_i2c_remove(struct platform_device *pdev) return 0; } -static struct platform_driver rcar_i2c_drv = { +static struct platform_driver rcar_i2c_driver = { .driver = { .name = "i2c-rcar", .owner = THIS_MODULE, @@ -702,7 +702,7 @@ static struct platform_driver rcar_i2c_drv = { .remove = __devexit_p(rcar_i2c_remove), }; -module_platform_driver(rcar_i2c_drv); +module_platform_driver(rcar_i2c_driver); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("Renesas R-Car I2C bus driver"); diff --git a/drivers/i2c/busses/i2c-s3c2410.c b/drivers/i2c/busses/i2c-s3c2410.c index b33d95ebc890..a290d089ceaf 100644 --- a/drivers/i2c/busses/i2c-s3c2410.c +++ b/drivers/i2c/busses/i2c-s3c2410.c @@ -38,6 +38,7 @@ #include <linux/io.h> #include <linux/of_i2c.h> #include <linux/of_gpio.h> +#include <linux/pinctrl/consumer.h> #include <asm/irq.h> @@ -49,6 +50,9 @@ #define QUIRK_HDMIPHY (1 << 1) #define QUIRK_NO_GPIO (1 << 2) +/* Max time to wait for bus to become idle after a xfer (in us) */ +#define S3C2410_IDLE_TIMEOUT 5000 + /* i2c controller state */ enum s3c24xx_i2c_state { STATE_IDLE, @@ -59,7 +63,6 @@ enum s3c24xx_i2c_state { }; struct s3c24xx_i2c { - spinlock_t lock; wait_queue_head_t wait; unsigned int quirks; unsigned int suspended:1; @@ -78,11 +81,11 @@ struct s3c24xx_i2c { void __iomem *regs; struct clk *clk; struct device *dev; - struct resource *ioarea; struct i2c_adapter adap; struct s3c2410_platform_i2c *pdata; int gpios[2]; + struct pinctrl *pctrl; #ifdef CONFIG_CPU_FREQ struct notifier_block freq_transition; #endif @@ -235,8 +238,47 @@ static inline void s3c24xx_i2c_stop(struct s3c24xx_i2c *i2c, int ret) dev_dbg(i2c->dev, "STOP\n"); - /* stop the transfer */ - iicstat &= ~S3C2410_IICSTAT_START; + /* + * The datasheet says that the STOP sequence should be: + * 1) I2CSTAT.5 = 0 - Clear BUSY (or 'generate STOP') + * 2) I2CCON.4 = 0 - Clear IRQPEND + * 3) Wait until the stop condition takes effect. + * 4*) I2CSTAT.4 = 0 - Clear TXRXEN + * + * Where, step "4*" is only for buses with the "HDMIPHY" quirk. + * + * However, after much experimentation, it appears that: + * a) normal buses automatically clear BUSY and transition from + * Master->Slave when they complete generating a STOP condition. + * Therefore, step (3) can be done in doxfer() by polling I2CCON.4 + * after starting the STOP generation here. + * b) HDMIPHY bus does neither, so there is no way to do step 3. + * There is no indication when this bus has finished generating + * STOP. + * + * In fact, we have found that as soon as the IRQPEND bit is cleared in + * step 2, the HDMIPHY bus generates the STOP condition, and then + * immediately starts transferring another data byte, even though the + * bus is supposedly stopped. This is presumably because the bus is + * still in "Master" mode, and its BUSY bit is still set. + * + * To avoid these extra post-STOP transactions on HDMI phy devices, we + * just disable Serial Output on the bus (I2CSTAT.4 = 0) directly, + * instead of first generating a proper STOP condition. This should + * float SDA & SCK terminating the transfer. Subsequent transfers + * start with a proper START condition, and proceed normally. + * + * The HDMIPHY bus is an internal bus that always has exactly two + * devices, the host as Master and the HDMIPHY device as the slave. + * Skipping the STOP condition has been tested on this bus and works. + */ + if (i2c->quirks & QUIRK_HDMIPHY) { + /* Stop driving the I2C pins */ + iicstat &= ~S3C2410_IICSTAT_TXRXEN; + } else { + /* stop the transfer */ + iicstat &= ~S3C2410_IICSTAT_START; + } writel(iicstat, i2c->regs + S3C2410_IICSTAT); i2c->state = STATE_STOP; @@ -490,13 +532,6 @@ static int s3c24xx_i2c_set_master(struct s3c24xx_i2c *i2c) unsigned long iicstat; int timeout = 400; - /* the timeout for HDMIPHY is reduced to 10 ms because - * the hangup is expected to happen, so waiting 400 ms - * causes only unnecessary system hangup - */ - if (i2c->quirks & QUIRK_HDMIPHY) - timeout = 10; - while (timeout-- > 0) { iicstat = readl(i2c->regs + S3C2410_IICSTAT); @@ -506,16 +541,61 @@ static int s3c24xx_i2c_set_master(struct s3c24xx_i2c *i2c) msleep(1); } - /* hang-up of bus dedicated for HDMIPHY occurred, resetting */ - if (i2c->quirks & QUIRK_HDMIPHY) { - writel(0, i2c->regs + S3C2410_IICCON); - writel(0, i2c->regs + S3C2410_IICSTAT); - writel(0, i2c->regs + S3C2410_IICDS); + return -ETIMEDOUT; +} - return 0; +/* s3c24xx_i2c_wait_idle + * + * wait for the i2c bus to become idle. +*/ + +static void s3c24xx_i2c_wait_idle(struct s3c24xx_i2c *i2c) +{ + unsigned long iicstat; + ktime_t start, now; + unsigned long delay; + int spins; + + /* ensure the stop has been through the bus */ + + dev_dbg(i2c->dev, "waiting for bus idle\n"); + + start = now = ktime_get(); + + /* + * Most of the time, the bus is already idle within a few usec of the + * end of a transaction. However, really slow i2c devices can stretch + * the clock, delaying STOP generation. + * + * On slower SoCs this typically happens within a very small number of + * instructions so busy wait briefly to avoid scheduling overhead. + */ + spins = 3; + iicstat = readl(i2c->regs + S3C2410_IICSTAT); + while ((iicstat & S3C2410_IICSTAT_START) && --spins) { + cpu_relax(); + iicstat = readl(i2c->regs + S3C2410_IICSTAT); } - return -ETIMEDOUT; + /* + * If we do get an appreciable delay as a compromise between idle + * detection latency for the normal, fast case, and system load in the + * slow device case, use an exponential back off in the polling loop, + * up to 1/10th of the total timeout, then continue to poll at a + * constant rate up to the timeout. + */ + delay = 1; + while ((iicstat & S3C2410_IICSTAT_START) && + ktime_us_delta(now, start) < S3C2410_IDLE_TIMEOUT) { + usleep_range(delay, 2 * delay); + if (delay < S3C2410_IDLE_TIMEOUT / 10) + delay <<= 1; + now = ktime_get(); + iicstat = readl(i2c->regs + S3C2410_IICSTAT); + } + + if (iicstat & S3C2410_IICSTAT_START) + dev_warn(i2c->dev, "timeout waiting for bus idle\n"); } /* s3c24xx_i2c_doxfer @@ -526,8 +606,7 @@ static int s3c24xx_i2c_set_master(struct s3c24xx_i2c *i2c) static int s3c24xx_i2c_doxfer(struct s3c24xx_i2c *i2c, struct i2c_msg *msgs, int num) { - unsigned long iicstat, timeout; - int spins = 20; + unsigned long timeout; int ret; if (i2c->suspended) @@ -540,8 +619,6 @@ static int s3c24xx_i2c_doxfer(struct s3c24xx_i2c *i2c, goto out; } - spin_lock_irq(&i2c->lock); - i2c->msg = msgs; i2c->msg_num = num; i2c->msg_ptr = 0; @@ -550,7 +627,6 @@ static int s3c24xx_i2c_doxfer(struct s3c24xx_i2c *i2c, s3c24xx_i2c_enable_irq(i2c); s3c24xx_i2c_message_start(i2c, msgs); - spin_unlock_irq(&i2c->lock); timeout = wait_event_timeout(i2c->wait, i2c->msg_num == 0, HZ * 5); @@ -564,24 +640,11 @@ static int s3c24xx_i2c_doxfer(struct s3c24xx_i2c *i2c, else if (ret != num) dev_dbg(i2c->dev, "incomplete xfer (%d)\n", ret); - /* ensure the stop has been through the bus */ - - dev_dbg(i2c->dev, "waiting for bus idle\n"); - - /* first, try busy waiting briefly */ - do { - cpu_relax(); - iicstat = readl(i2c->regs + S3C2410_IICSTAT); - } while ((iicstat & S3C2410_IICSTAT_START) && --spins); - - /* if that timed out sleep */ - if (!spins) { - msleep(1); - iicstat = readl(i2c->regs + S3C2410_IICSTAT); - } + /* For QUIRK_HDMIPHY, bus is already disabled */ + if (i2c->quirks & QUIRK_HDMIPHY) + goto out; - if (iicstat & S3C2410_IICSTAT_START) - dev_warn(i2c->dev, "timeout waiting for bus idle\n"); + s3c24xx_i2c_wait_idle(i2c); out: return ret; @@ -740,7 +803,6 @@ static int s3c24xx_i2c_cpufreq_transition(struct notifier_block *nb, unsigned long val, void *data) { struct s3c24xx_i2c *i2c = freq_to_i2c(nb); - unsigned long flags; unsigned int got; int delta_f; int ret; @@ -754,9 +816,9 @@ static int s3c24xx_i2c_cpufreq_transition(struct notifier_block *nb, if ((val == CPUFREQ_POSTCHANGE && delta_f < 0) || (val == CPUFREQ_PRECHANGE && delta_f > 0)) { - spin_lock_irqsave(&i2c->lock, flags); + i2c_lock_adapter(&i2c->adap); ret = s3c24xx_i2c_clockrate(i2c, &got); - spin_unlock_irqrestore(&i2c->lock, flags); + i2c_unlock_adapter(&i2c->adap); if (ret < 0) dev_err(i2c->dev, "cannot find frequency\n"); @@ -858,14 +920,6 @@ static int s3c24xx_i2c_init(struct s3c24xx_i2c *i2c) pdata = i2c->pdata; - /* inititalise the gpio */ - - if (pdata->cfg_gpio) - pdata->cfg_gpio(to_platform_device(i2c->dev)); - else - if (s3c24xx_i2c_parse_dt_gpio(i2c)) - return -EINVAL; - /* write slave address */ writeb(pdata->slave_addr, i2c->regs + S3C2410_IICADD); @@ -963,7 +1017,6 @@ static int s3c24xx_i2c_probe(struct platform_device *pdev) i2c->adap.class = I2C_CLASS_HWMON | I2C_CLASS_SPD; i2c->tx_setup = 50; - spin_lock_init(&i2c->lock); init_waitqueue_head(&i2c->wait); /* find the clock and enable it */ @@ -989,36 +1042,38 @@ static int s3c24xx_i2c_probe(struct platform_device *pdev) goto err_clk; } - i2c->ioarea = request_mem_region(res->start, resource_size(res), - pdev->name); - - if (i2c->ioarea == NULL) { - dev_err(&pdev->dev, "cannot request IO\n"); - ret = -ENXIO; - goto err_clk; - } - - i2c->regs = ioremap(res->start, resource_size(res)); + i2c->regs = devm_request_and_ioremap(&pdev->dev, res); if (i2c->regs == NULL) { dev_err(&pdev->dev, "cannot map IO\n"); ret = -ENXIO; - goto err_ioarea; + goto err_clk; } - dev_dbg(&pdev->dev, "registers %p (%p, %p)\n", - i2c->regs, i2c->ioarea, res); + dev_dbg(&pdev->dev, "registers %p (%p)\n", + i2c->regs, res); /* setup info block for the i2c core */ i2c->adap.algo_data = i2c; i2c->adap.dev.parent = &pdev->dev; + i2c->pctrl = devm_pinctrl_get_select_default(i2c->dev); + + /* inititalise the i2c gpio lines */ + + if (i2c->pdata->cfg_gpio) { + i2c->pdata->cfg_gpio(to_platform_device(i2c->dev)); + } else if (IS_ERR(i2c->pctrl) && s3c24xx_i2c_parse_dt_gpio(i2c)) { + ret = -EINVAL; + goto err_clk; + } + /* initialise the i2c controller */ ret = s3c24xx_i2c_init(i2c); if (ret != 0) - goto err_iomap; + goto err_clk; /* find the IRQ for this unit (note, this relies on the init call to * ensure no current IRQs pending @@ -1027,7 +1082,7 @@ static int s3c24xx_i2c_probe(struct platform_device *pdev) i2c->irq = ret = platform_get_irq(pdev, 0); if (ret <= 0) { dev_err(&pdev->dev, "cannot find IRQ\n"); - goto err_iomap; + goto err_clk; } ret = request_irq(i2c->irq, s3c24xx_i2c_irq, 0, @@ -1035,7 +1090,7 @@ static int s3c24xx_i2c_probe(struct platform_device *pdev) if (ret != 0) { dev_err(&pdev->dev, "cannot claim IRQ %d\n", i2c->irq); - goto err_iomap; + goto err_clk; } ret = s3c24xx_i2c_register_cpufreq(i2c); @@ -1075,13 +1130,6 @@ static int s3c24xx_i2c_probe(struct platform_device *pdev) err_irq: free_irq(i2c->irq, i2c); - err_iomap: - iounmap(i2c->regs); - - err_ioarea: - release_resource(i2c->ioarea); - kfree(i2c->ioarea); - err_clk: clk_disable_unprepare(i2c->clk); clk_put(i2c->clk); @@ -1110,16 +1158,13 @@ static int s3c24xx_i2c_remove(struct platform_device *pdev) clk_disable_unprepare(i2c->clk); clk_put(i2c->clk); - iounmap(i2c->regs); - - release_resource(i2c->ioarea); - s3c24xx_i2c_dt_gpio_free(i2c); - kfree(i2c->ioarea); + if (pdev->dev.of_node && IS_ERR(i2c->pctrl)) + s3c24xx_i2c_dt_gpio_free(i2c); return 0; } -#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP static int s3c24xx_i2c_suspend_noirq(struct device *dev) { struct platform_device *pdev = to_platform_device(dev); @@ -1142,10 +1187,14 @@ static int s3c24xx_i2c_resume(struct device *dev) return 0; } +#endif +#ifdef CONFIG_PM static const struct dev_pm_ops s3c24xx_i2c_dev_pm_ops = { +#ifdef CONFIG_PM_SLEEP .suspend_noirq = s3c24xx_i2c_suspend_noirq, .resume = s3c24xx_i2c_resume, +#endif }; #define S3C24XX_DEV_PM_OPS (&s3c24xx_i2c_dev_pm_ops) diff --git a/drivers/i2c/busses/i2c-sh_mobile.c b/drivers/i2c/busses/i2c-sh_mobile.c index 8110ca45f342..9411c1b892c0 100644 --- a/drivers/i2c/busses/i2c-sh_mobile.c +++ b/drivers/i2c/busses/i2c-sh_mobile.c @@ -120,11 +120,12 @@ struct sh_mobile_i2c_data { void __iomem *reg; struct i2c_adapter adap; unsigned long bus_speed; + unsigned int clks_per_count; struct clk *clk; u_int8_t icic; - u_int8_t iccl; - u_int8_t icch; u_int8_t flags; + u_int16_t iccl; + u_int16_t icch; spinlock_t lock; wait_queue_head_t wait; @@ -135,7 +136,8 @@ struct sh_mobile_i2c_data { #define IIC_FLAG_HAS_ICIC67 (1 << 0) -#define NORMAL_SPEED 100000 /* FAST_SPEED 400000 */ +#define STANDARD_MODE 100000 +#define FAST_MODE 400000 /* Register offsets */ #define ICDR 0x00 @@ -187,57 +189,90 @@ static void iic_set_clr(struct sh_mobile_i2c_data *pd, int offs, iic_wr(pd, offs, (iic_rd(pd, offs) | set) & ~clr); } -static void activate_ch(struct sh_mobile_i2c_data *pd) +static u32 sh_mobile_i2c_iccl(unsigned long count_khz, u32 tLOW, u32 tf, int offset) { - unsigned long i2c_clk; - u_int32_t num; - u_int32_t denom; - u_int32_t tmp; - - /* Wake up device and enable clock */ - pm_runtime_get_sync(pd->dev); - clk_enable(pd->clk); - - /* Get clock rate after clock is enabled */ - i2c_clk = clk_get_rate(pd->clk); + /* + * Conditional expression: + * ICCL >= COUNT_CLK * (tLOW + tf) + * + * SH-Mobile IIC hardware starts counting the LOW period of + * the SCL signal (tLOW) as soon as it pulls the SCL line. + * In order to meet the tLOW timing spec, we need to take into + * account the fall time of SCL signal (tf). Default tf value + * should be 0.3 us, for safety. + */ + return (((count_khz * (tLOW + tf)) + 5000) / 10000) + offset; +} - /* Calculate the value for iccl. From the data sheet: - * iccl = (p clock / transfer rate) * (L / (L + H)) - * where L and H are the SCL low/high ratio (5/4 in this case). - * We also round off the result. +static u32 sh_mobile_i2c_icch(unsigned long count_khz, u32 tHIGH, u32 tf, int offset) +{ + /* + * Conditional expression: + * ICCH >= COUNT_CLK * (tHIGH + tf) + * + * SH-Mobile IIC hardware is aware of SCL transition period 'tr', + * and can ignore it. SH-Mobile IIC controller starts counting + * the HIGH period of the SCL signal (tHIGH) after the SCL input + * voltage increases at VIH. + * + * Afterward it turned out calculating ICCH using only tHIGH spec + * will result in violation of the tHD;STA timing spec. We need + * to take into account the fall time of SDA signal (tf) at START + * condition, in order to meet both tHIGH and tHD;STA specs. */ - num = i2c_clk * 5; - denom = pd->bus_speed * 9; - tmp = num * 10 / denom; - if (tmp % 10 >= 5) - pd->iccl = (u_int8_t)((num/denom) + 1); - else - pd->iccl = (u_int8_t)(num/denom); + return (((count_khz * (tHIGH + tf)) + 5000) / 10000) + offset; +} - /* one more bit of ICCL in ICIC */ - if (pd->flags & IIC_FLAG_HAS_ICIC67) { - if ((num/denom) > 0xff) - pd->icic |= ICIC_ICCLB8; - else - pd->icic &= ~ICIC_ICCLB8; +static void sh_mobile_i2c_init(struct sh_mobile_i2c_data *pd) +{ + unsigned long i2c_clk_khz; + u32 tHIGH, tLOW, tf; + int offset; + + /* Get clock rate after clock is enabled */ + clk_enable(pd->clk); + i2c_clk_khz = clk_get_rate(pd->clk) / 1000; + i2c_clk_khz /= pd->clks_per_count; + + if (pd->bus_speed == STANDARD_MODE) { + tLOW = 47; /* tLOW = 4.7 us */ + tHIGH = 40; /* tHD;STA = tHIGH = 4.0 us */ + tf = 3; /* tf = 0.3 us */ + offset = 0; /* No offset */ + } else if (pd->bus_speed == FAST_MODE) { + tLOW = 13; /* tLOW = 1.3 us */ + tHIGH = 6; /* tHD;STA = tHIGH = 0.6 us */ + tf = 3; /* tf = 0.3 us */ + offset = 0; /* No offset */ + } else { + dev_err(pd->dev, "unrecognized bus speed %lu Hz\n", + pd->bus_speed); + goto out; } - /* Calculate the value for icch. From the data sheet: - icch = (p clock / transfer rate) * (H / (L + H)) */ - num = i2c_clk * 4; - tmp = num * 10 / denom; - if (tmp % 10 >= 5) - pd->icch = (u_int8_t)((num/denom) + 1); + pd->iccl = sh_mobile_i2c_iccl(i2c_clk_khz, tLOW, tf, offset); + /* one more bit of ICCL in ICIC */ + if ((pd->iccl > 0xff) && (pd->flags & IIC_FLAG_HAS_ICIC67)) + pd->icic |= ICIC_ICCLB8; else - pd->icch = (u_int8_t)(num/denom); + pd->icic &= ~ICIC_ICCLB8; + pd->icch = sh_mobile_i2c_icch(i2c_clk_khz, tHIGH, tf, offset); /* one more bit of ICCH in ICIC */ - if (pd->flags & IIC_FLAG_HAS_ICIC67) { - if ((num/denom) > 0xff) - pd->icic |= ICIC_ICCHB8; - else - pd->icic &= ~ICIC_ICCHB8; - } + if ((pd->icch > 0xff) && (pd->flags & IIC_FLAG_HAS_ICIC67)) + pd->icic |= ICIC_ICCHB8; + else + pd->icic &= ~ICIC_ICCHB8; + +out: + clk_disable(pd->clk); +} + +static void activate_ch(struct sh_mobile_i2c_data *pd) +{ + /* Wake up device and enable clock */ + pm_runtime_get_sync(pd->dev); + clk_enable(pd->clk); /* Enable channel and configure rx ack */ iic_set_clr(pd, ICCR, ICCR_ICE, 0); @@ -246,8 +281,8 @@ static void activate_ch(struct sh_mobile_i2c_data *pd) iic_wr(pd, ICIC, 0); /* Set the clock */ - iic_wr(pd, ICCL, pd->iccl); - iic_wr(pd, ICCH, pd->icch); + iic_wr(pd, ICCL, pd->iccl & 0xff); + iic_wr(pd, ICCH, pd->icch & 0xff); } static void deactivate_ch(struct sh_mobile_i2c_data *pd) @@ -434,6 +469,9 @@ static irqreturn_t sh_mobile_i2c_isr(int irq, void *dev_id) wake_up(&pd->wait); } + /* defeat write posting to avoid spurious WAIT interrupts */ + iic_rd(pd, ICSR); + return IRQ_HANDLED; } @@ -451,8 +489,8 @@ static int start_ch(struct sh_mobile_i2c_data *pd, struct i2c_msg *usr_msg) iic_set_clr(pd, ICCR, ICCR_ICE, 0); /* Set the clock */ - iic_wr(pd, ICCL, pd->iccl); - iic_wr(pd, ICCH, pd->icch); + iic_wr(pd, ICCL, pd->iccl & 0xff); + iic_wr(pd, ICCH, pd->icch & 0xff); pd->msg = usr_msg; pd->pos = -1; @@ -621,10 +659,13 @@ static int sh_mobile_i2c_probe(struct platform_device *dev) goto err_irq; } - /* Use platformd data bus speed or NORMAL_SPEED */ - pd->bus_speed = NORMAL_SPEED; + /* Use platform data bus speed or STANDARD_MODE */ + pd->bus_speed = STANDARD_MODE; if (pdata && pdata->bus_speed) pd->bus_speed = pdata->bus_speed; + pd->clks_per_count = 1; + if (pdata && pdata->clks_per_count) + pd->clks_per_count = pdata->clks_per_count; /* The IIC blocks on SH-Mobile ARM processors * come with two new bits in ICIC. @@ -632,6 +673,8 @@ static int sh_mobile_i2c_probe(struct platform_device *dev) if (size > 0x17) pd->flags |= IIC_FLAG_HAS_ICIC67; + sh_mobile_i2c_init(pd); + /* Enable Runtime PM for this device. * * Also tell the Runtime PM core to ignore children @@ -667,8 +710,9 @@ static int sh_mobile_i2c_probe(struct platform_device *dev) goto err_all; } - dev_info(&dev->dev, "I2C adapter %d with bus speed %lu Hz\n", - adap->nr, pd->bus_speed); + dev_info(&dev->dev, + "I2C adapter %d with bus speed %lu Hz (L/H=%x/%x)\n", + adap->nr, pd->bus_speed, pd->iccl, pd->icch); of_i2c_register_devices(adap); return 0; diff --git a/drivers/i2c/muxes/i2c-mux-gpio.c b/drivers/i2c/muxes/i2c-mux-gpio.c index 566a6757a33d..3b7bc06fe8a6 100644 --- a/drivers/i2c/muxes/i2c-mux-gpio.c +++ b/drivers/i2c/muxes/i2c-mux-gpio.c @@ -16,6 +16,8 @@ #include <linux/module.h> #include <linux/slab.h> #include <linux/gpio.h> +#include <linux/of_i2c.h> +#include <linux/of_gpio.h> struct gpiomux { struct i2c_adapter *parent; @@ -57,29 +59,110 @@ static int __devinit match_gpio_chip_by_label(struct gpio_chip *chip, return !strcmp(chip->label, data); } +#ifdef CONFIG_OF +static int __devinit i2c_mux_gpio_probe_dt(struct gpiomux *mux, + struct platform_device *pdev) +{ + struct device_node *np = pdev->dev.of_node; + struct device_node *adapter_np, *child; + struct i2c_adapter *adapter; + unsigned *values, *gpios; + int i = 0; + + if (!np) + return -ENODEV; + + adapter_np = of_parse_phandle(np, "i2c-parent", 0); + if (!adapter_np) { + dev_err(&pdev->dev, "Cannot parse i2c-parent\n"); + return -ENODEV; + } + adapter = of_find_i2c_adapter_by_node(adapter_np); + if (!adapter) { + dev_err(&pdev->dev, "Cannot find parent bus\n"); + return -ENODEV; + } + mux->data.parent = i2c_adapter_id(adapter); + put_device(&adapter->dev); + + mux->data.n_values = of_get_child_count(np); + + values = devm_kzalloc(&pdev->dev, + sizeof(*mux->data.values) * mux->data.n_values, + GFP_KERNEL); + if (!values) { + dev_err(&pdev->dev, "Cannot allocate values array"); + return -ENOMEM; + } + + for_each_child_of_node(np, child) { + of_property_read_u32(child, "reg", values + i); + i++; + } + mux->data.values = values; + + if (of_property_read_u32(np, "idle-state", &mux->data.idle)) + mux->data.idle = I2C_MUX_GPIO_NO_IDLE; + + mux->data.n_gpios = of_gpio_named_count(np, "mux-gpios"); + if (mux->data.n_gpios < 0) { + dev_err(&pdev->dev, "Missing mux-gpios property in the DT.\n"); + return -EINVAL; + } + + gpios = devm_kzalloc(&pdev->dev, + sizeof(*mux->data.gpios) * mux->data.n_gpios, GFP_KERNEL); + if (!gpios) { + dev_err(&pdev->dev, "Cannot allocate gpios array"); + return -ENOMEM; + } + + for (i = 0; i < mux->data.n_gpios; i++) + gpios[i] = of_get_named_gpio(np, "mux-gpios", i); + + mux->data.gpios = gpios; + + return 0; +} +#else +static int __devinit i2c_mux_gpio_probe_dt(struct gpiomux *mux, + struct platform_device *pdev) +{ + return 0; +} +#endif + static int __devinit i2c_mux_gpio_probe(struct platform_device *pdev) { struct gpiomux *mux; - struct i2c_mux_gpio_platform_data *pdata; struct i2c_adapter *parent; int (*deselect) (struct i2c_adapter *, void *, u32); unsigned initial_state, gpio_base; int i, ret; - pdata = pdev->dev.platform_data; - if (!pdata) { - dev_err(&pdev->dev, "Missing platform data\n"); - return -ENODEV; + mux = devm_kzalloc(&pdev->dev, sizeof(*mux), GFP_KERNEL); + if (!mux) { + dev_err(&pdev->dev, "Cannot allocate gpiomux structure"); + return -ENOMEM; } + platform_set_drvdata(pdev, mux); + + if (!pdev->dev.platform_data) { + ret = i2c_mux_gpio_probe_dt(mux, pdev); + if (ret < 0) + return ret; + } else + memcpy(&mux->data, pdev->dev.platform_data, sizeof(mux->data)); + /* * If a GPIO chip name is provided, the GPIO pin numbers provided are * relative to its base GPIO number. Otherwise they are absolute. */ - if (pdata->gpio_chip) { + if (mux->data.gpio_chip) { struct gpio_chip *gpio; - gpio = gpiochip_find(pdata->gpio_chip, + gpio = gpiochip_find(mux->data.gpio_chip, match_gpio_chip_by_label); if (!gpio) return -EPROBE_DEFER; @@ -89,49 +172,44 @@ static int __devinit i2c_mux_gpio_probe(struct platform_device *pdev) gpio_base = 0; } - parent = i2c_get_adapter(pdata->parent); + parent = i2c_get_adapter(mux->data.parent); if (!parent) { dev_err(&pdev->dev, "Parent adapter (%d) not found\n", - pdata->parent); + mux->data.parent); return -ENODEV; } - mux = devm_kzalloc(&pdev->dev, sizeof(*mux), GFP_KERNEL); - if (!mux) { - ret = -ENOMEM; - goto alloc_failed; - } - mux->parent = parent; - mux->data = *pdata; mux->gpio_base = gpio_base; + mux->adap = devm_kzalloc(&pdev->dev, - sizeof(*mux->adap) * pdata->n_values, + sizeof(*mux->adap) * mux->data.n_values, GFP_KERNEL); if (!mux->adap) { + dev_err(&pdev->dev, "Cannot allocate i2c_adapter structure"); ret = -ENOMEM; goto alloc_failed; } - if (pdata->idle != I2C_MUX_GPIO_NO_IDLE) { - initial_state = pdata->idle; + if (mux->data.idle != I2C_MUX_GPIO_NO_IDLE) { + initial_state = mux->data.idle; deselect = i2c_mux_gpio_deselect; } else { - initial_state = pdata->values[0]; + initial_state = mux->data.values[0]; deselect = NULL; } - for (i = 0; i < pdata->n_gpios; i++) { - ret = gpio_request(gpio_base + pdata->gpios[i], "i2c-mux-gpio"); + for (i = 0; i < mux->data.n_gpios; i++) { + ret = gpio_request(gpio_base + mux->data.gpios[i], "i2c-mux-gpio"); if (ret) goto err_request_gpio; - gpio_direction_output(gpio_base + pdata->gpios[i], + gpio_direction_output(gpio_base + mux->data.gpios[i], initial_state & (1 << i)); } - for (i = 0; i < pdata->n_values; i++) { - u32 nr = pdata->base_nr ? (pdata->base_nr + i) : 0; - unsigned int class = pdata->classes ? pdata->classes[i] : 0; + for (i = 0; i < mux->data.n_values; i++) { + u32 nr = mux->data.base_nr ? (mux->data.base_nr + i) : 0; + unsigned int class = mux->data.classes ? mux->data.classes[i] : 0; mux->adap[i] = i2c_add_mux_adapter(parent, &pdev->dev, mux, nr, i, class, @@ -144,19 +222,17 @@ static int __devinit i2c_mux_gpio_probe(struct platform_device *pdev) } dev_info(&pdev->dev, "%d port mux on %s adapter\n", - pdata->n_values, parent->name); - - platform_set_drvdata(pdev, mux); + mux->data.n_values, parent->name); return 0; add_adapter_failed: for (; i > 0; i--) i2c_del_mux_adapter(mux->adap[i - 1]); - i = pdata->n_gpios; + i = mux->data.n_gpios; err_request_gpio: for (; i > 0; i--) - gpio_free(gpio_base + pdata->gpios[i - 1]); + gpio_free(gpio_base + mux->data.gpios[i - 1]); alloc_failed: i2c_put_adapter(parent); @@ -180,12 +256,19 @@ static int __devexit i2c_mux_gpio_remove(struct platform_device *pdev) return 0; } +static const struct of_device_id i2c_mux_gpio_of_match[] __devinitconst = { + { .compatible = "i2c-mux-gpio", }, + {}, +}; +MODULE_DEVICE_TABLE(of, i2c_mux_gpio_of_match); + static struct platform_driver i2c_mux_gpio_driver = { .probe = i2c_mux_gpio_probe, .remove = __devexit_p(i2c_mux_gpio_remove), .driver = { .owner = THIS_MODULE, .name = "i2c-mux-gpio", + .of_match_table = of_match_ptr(i2c_mux_gpio_of_match), }, }; diff --git a/drivers/infiniband/hw/ehca/hcp_if.c b/drivers/infiniband/hw/ehca/hcp_if.c index 2d41d04fd959..89517ffb4389 100644 --- a/drivers/infiniband/hw/ehca/hcp_if.c +++ b/drivers/infiniband/hw/ehca/hcp_if.c @@ -90,26 +90,6 @@ static DEFINE_SPINLOCK(hcall_lock); -static u32 get_longbusy_msecs(int longbusy_rc) -{ - switch (longbusy_rc) { - case H_LONG_BUSY_ORDER_1_MSEC: - return 1; - case H_LONG_BUSY_ORDER_10_MSEC: - return 10; - case H_LONG_BUSY_ORDER_100_MSEC: - return 100; - case H_LONG_BUSY_ORDER_1_SEC: - return 1000; - case H_LONG_BUSY_ORDER_10_SEC: - return 10000; - case H_LONG_BUSY_ORDER_100_SEC: - return 100000; - default: - return 1; - } -} - static long ehca_plpar_hcall_norets(unsigned long opcode, unsigned long arg1, unsigned long arg2, diff --git a/drivers/input/gameport/emu10k1-gp.c b/drivers/input/gameport/emu10k1-gp.c index daceafe7ee7d..fa7a95c1da0e 100644 --- a/drivers/input/gameport/emu10k1-gp.c +++ b/drivers/input/gameport/emu10k1-gp.c @@ -57,7 +57,7 @@ static const struct pci_device_id emu_tbl[] = { MODULE_DEVICE_TABLE(pci, emu_tbl); -static int __devinit emu_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +static int emu_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { struct emu *emu; struct gameport *port; @@ -107,7 +107,7 @@ static int __devinit emu_probe(struct pci_dev *pdev, const struct pci_device_id return error; } -static void __devexit emu_remove(struct pci_dev *pdev) +static void emu_remove(struct pci_dev *pdev) { struct emu *emu = pci_get_drvdata(pdev); @@ -122,7 +122,7 @@ static struct pci_driver emu_driver = { .name = "Emu10k1_gameport", .id_table = emu_tbl, .probe = emu_probe, - .remove = __devexit_p(emu_remove), + .remove = emu_remove, }; module_pci_driver(emu_driver); diff --git a/drivers/input/gameport/fm801-gp.c b/drivers/input/gameport/fm801-gp.c index 48ad3829ff20..ae912d3aee4e 100644 --- a/drivers/input/gameport/fm801-gp.c +++ b/drivers/input/gameport/fm801-gp.c @@ -78,7 +78,7 @@ static int fm801_gp_open(struct gameport *gameport, int mode) return 0; } -static int __devinit fm801_gp_probe(struct pci_dev *pci, const struct pci_device_id *id) +static int fm801_gp_probe(struct pci_dev *pci, const struct pci_device_id *id) { struct fm801_gp *gp; struct gameport *port; @@ -129,7 +129,7 @@ static int __devinit fm801_gp_probe(struct pci_dev *pci, const struct pci_device return error; } -static void __devexit fm801_gp_remove(struct pci_dev *pci) +static void fm801_gp_remove(struct pci_dev *pci) { struct fm801_gp *gp = pci_get_drvdata(pci); @@ -150,7 +150,7 @@ static struct pci_driver fm801_gp_driver = { .name = "FM801_gameport", .id_table = fm801_gp_id_table, .probe = fm801_gp_probe, - .remove = __devexit_p(fm801_gp_remove), + .remove = fm801_gp_remove, }; module_pci_driver(fm801_gp_driver); diff --git a/drivers/input/input-mt.c b/drivers/input/input-mt.c index 8c4b50fd9a79..47a6009dbf43 100644 --- a/drivers/input/input-mt.c +++ b/drivers/input/input-mt.c @@ -194,7 +194,7 @@ void input_mt_report_pointer_emulation(struct input_dev *dev, bool use_count) if (!mt) return; - oldest = 0; + oldest = NULL; oldid = mt->trkid; count = 0; diff --git a/drivers/input/input.c b/drivers/input/input.c index 53a0ddee7872..ce01332f7b3a 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -534,8 +534,11 @@ EXPORT_SYMBOL(input_grab_device); static void __input_release_device(struct input_handle *handle) { struct input_dev *dev = handle->dev; + struct input_handle *grabber; - if (dev->grab == handle) { + grabber = rcu_dereference_protected(dev->grab, + lockdep_is_held(&dev->mutex)); + if (grabber == handle) { rcu_assign_pointer(dev->grab, NULL); /* Make sure input_pass_event() notices that grab is gone */ synchronize_rcu(); @@ -1723,7 +1726,7 @@ EXPORT_SYMBOL_GPL(input_class); /** * input_allocate_device - allocate memory for new input device * - * Returns prepared struct input_dev or NULL. + * Returns prepared struct input_dev or %NULL. * * NOTE: Use input_free_device() to free devices that have not been * registered; input_unregister_device() should be used for already @@ -1750,6 +1753,70 @@ struct input_dev *input_allocate_device(void) } EXPORT_SYMBOL(input_allocate_device); +struct input_devres { + struct input_dev *input; +}; + +static int devm_input_device_match(struct device *dev, void *res, void *data) +{ + struct input_devres *devres = res; + + return devres->input == data; +} + +static void devm_input_device_release(struct device *dev, void *res) +{ + struct input_devres *devres = res; + struct input_dev *input = devres->input; + + dev_dbg(dev, "%s: dropping reference to %s\n", + __func__, dev_name(&input->dev)); + input_put_device(input); +} + +/** + * devm_input_allocate_device - allocate managed input device + * @dev: device owning the input device being created + * + * Returns prepared struct input_dev or %NULL. + * + * Managed input devices do not need to be explicitly unregistered or + * freed as it will be done automatically when owner device unbinds from + * its driver (or binding fails). Once managed input device is allocated, + * it is ready to be set up and registered in the same fashion as regular + * input device. There are no special devm_input_device_[un]register() + * variants, regular ones work with both managed and unmanaged devices. + * + * NOTE: the owner device is set up as parent of input device and users + * should not override it. + */ + +struct input_dev *devm_input_allocate_device(struct device *dev) +{ + struct input_dev *input; + struct input_devres *devres; + + devres = devres_alloc(devm_input_device_release, + sizeof(struct input_devres), GFP_KERNEL); + if (!devres) + return NULL; + + input = input_allocate_device(); + if (!input) { + devres_free(devres); + return NULL; + } + + input->dev.parent = dev; + input->devres_managed = true; + + devres->input = input; + devres_add(dev, devres); + + return input; +} +EXPORT_SYMBOL(devm_input_allocate_device); + /** * input_free_device - free memory occupied by input_dev structure * @dev: input device to free @@ -1766,8 +1833,14 @@ EXPORT_SYMBOL(input_allocate_device); */ void input_free_device(struct input_dev *dev) { - if (dev) + if (dev) { + if (dev->devres_managed) + WARN_ON(devres_destroy(dev->dev.parent, + devm_input_device_release, + devm_input_device_match, + dev)); input_put_device(dev); + } } EXPORT_SYMBOL(input_free_device); @@ -1888,6 +1961,38 @@ static void input_cleanse_bitmasks(struct input_dev *dev) INPUT_CLEANSE_BITMASK(dev, SW, sw); } +static void __input_unregister_device(struct input_dev *dev) +{ + struct input_handle *handle, *next; + + input_disconnect_device(dev); + + mutex_lock(&input_mutex); + + list_for_each_entry_safe(handle, next, &dev->h_list, d_node) + handle->handler->disconnect(handle); + WARN_ON(!list_empty(&dev->h_list)); + + del_timer_sync(&dev->timer); + list_del_init(&dev->node); + + input_wakeup_procfs_readers(); + + mutex_unlock(&input_mutex); + + device_del(&dev->dev); +} + +static void devm_input_device_unregister(struct device *dev, void *res) +{ + struct input_devres *devres = res; + struct input_dev *input = devres->input; + + dev_dbg(dev, "%s: unregistering device %s\n", + __func__, dev_name(&input->dev)); + __input_unregister_device(input); +} + /** * input_register_device - register device with input core * @dev: device to be registered @@ -1903,11 +2008,21 @@ static void input_cleanse_bitmasks(struct input_dev *dev) int input_register_device(struct input_dev *dev) { static atomic_t input_no = ATOMIC_INIT(0); + struct input_devres *devres = NULL; struct input_handler *handler; unsigned int packet_size; const char *path; int error; + if (dev->devres_managed) { + devres = devres_alloc(devm_input_device_unregister, + sizeof(struct input_devres), GFP_KERNEL); + if (!devres) + return -ENOMEM; + + devres->input = dev; + } + /* Every input device generates EV_SYN/SYN_REPORT events. */ __set_bit(EV_SYN, dev->evbit); @@ -1923,8 +2038,10 @@ int input_register_device(struct input_dev *dev) dev->max_vals = max(dev->hint_events_per_packet, packet_size) + 2; dev->vals = kcalloc(dev->max_vals, sizeof(*dev->vals), GFP_KERNEL); - if (!dev->vals) - return -ENOMEM; + if (!dev->vals) { + error = -ENOMEM; + goto err_devres_free; + } /* * If delay and period are pre-set by the driver, then autorepeating @@ -1949,7 +2066,7 @@ int input_register_device(struct input_dev *dev) error = device_add(&dev->dev); if (error) - return error; + goto err_free_vals; path = kobject_get_path(&dev->dev.kobj, GFP_KERNEL); pr_info("%s as %s\n", @@ -1958,10 +2075,8 @@ int input_register_device(struct input_dev *dev) kfree(path); error = mutex_lock_interruptible(&input_mutex); - if (error) { - device_del(&dev->dev); - return error; - } + if (error) + goto err_device_del; list_add_tail(&dev->node, &input_dev_list); @@ -1972,7 +2087,21 @@ int input_register_device(struct input_dev *dev) mutex_unlock(&input_mutex); + if (dev->devres_managed) { + dev_dbg(dev->dev.parent, "%s: registering %s with devres.\n", + __func__, dev_name(&dev->dev)); + devres_add(dev->dev.parent, devres); + } return 0; + +err_device_del: + device_del(&dev->dev); +err_free_vals: + kfree(dev->vals); + dev->vals = NULL; +err_devres_free: + devres_free(devres); + return error; } EXPORT_SYMBOL(input_register_device); @@ -1985,24 +2114,20 @@ EXPORT_SYMBOL(input_register_device); */ void input_unregister_device(struct input_dev *dev) { - struct input_handle *handle, *next; - - input_disconnect_device(dev); - - mutex_lock(&input_mutex); - - list_for_each_entry_safe(handle, next, &dev->h_list, d_node) - handle->handler->disconnect(handle); - WARN_ON(!list_empty(&dev->h_list)); - - del_timer_sync(&dev->timer); - list_del_init(&dev->node); - - input_wakeup_procfs_readers(); - - mutex_unlock(&input_mutex); - - device_unregister(&dev->dev); + if (dev->devres_managed) { + WARN_ON(devres_destroy(dev->dev.parent, + devm_input_device_unregister, + devm_input_device_match, + dev)); + __input_unregister_device(dev); + /* + * We do not do input_put_device() here because it will be done + * when 2nd devres fires up. + */ + } else { + __input_unregister_device(dev); + input_put_device(dev); + } } EXPORT_SYMBOL(input_unregister_device); diff --git a/drivers/input/joystick/as5011.c b/drivers/input/joystick/as5011.c index c96653b58867..121cd63d3334 100644 --- a/drivers/input/joystick/as5011.c +++ b/drivers/input/joystick/as5011.c @@ -85,7 +85,10 @@ static int as5011_i2c_write(struct i2c_client *client, { uint8_t data[2] = { aregaddr, avalue }; struct i2c_msg msg = { - client->addr, I2C_M_IGNORE_NAK, 2, (uint8_t *)data + .addr = client->addr, + .flags = I2C_M_IGNORE_NAK, + .len = 2, + .buf = (uint8_t *)data }; int error; @@ -98,8 +101,18 @@ static int as5011_i2c_read(struct i2c_client *client, { uint8_t data[2] = { aregaddr }; struct i2c_msg msg_set[2] = { - { client->addr, I2C_M_REV_DIR_ADDR, 1, (uint8_t *)data }, - { client->addr, I2C_M_RD | I2C_M_NOSTART, 1, (uint8_t *)data } + { + .addr = client->addr, + .flags = I2C_M_REV_DIR_ADDR, + .len = 1, + .buf = (uint8_t *)data + }, + { + .addr = client->addr, + .flags = I2C_M_RD | I2C_M_NOSTART, + .len = 1, + .buf = (uint8_t *)data + } }; int error; @@ -144,7 +157,7 @@ out: return IRQ_HANDLED; } -static int __devinit as5011_configure_chip(struct as5011_device *as5011, +static int as5011_configure_chip(struct as5011_device *as5011, const struct as5011_platform_data *plat_dat) { struct i2c_client *client = as5011->i2c_client; @@ -212,8 +225,8 @@ static int __devinit as5011_configure_chip(struct as5011_device *as5011, return 0; } -static int __devinit as5011_probe(struct i2c_client *client, - const struct i2c_device_id *id) +static int as5011_probe(struct i2c_client *client, + const struct i2c_device_id *id) { const struct as5011_platform_data *plat_data; struct as5011_device *as5011; @@ -328,7 +341,7 @@ err_free_mem: return error; } -static int __devexit as5011_remove(struct i2c_client *client) +static int as5011_remove(struct i2c_client *client) { struct as5011_device *as5011 = i2c_get_clientdata(client); @@ -353,7 +366,7 @@ static struct i2c_driver as5011_driver = { .name = "as5011", }, .probe = as5011_probe, - .remove = __devexit_p(as5011_remove), + .remove = as5011_remove, .id_table = as5011_id, }; diff --git a/drivers/input/joystick/maplecontrol.c b/drivers/input/joystick/maplecontrol.c index 77cfde571bd9..59c10ec5a2a1 100644 --- a/drivers/input/joystick/maplecontrol.c +++ b/drivers/input/joystick/maplecontrol.c @@ -78,7 +78,7 @@ static void dc_pad_close(struct input_dev *dev) } /* allow the controller to be used */ -static int __devinit probe_maple_controller(struct device *dev) +static int probe_maple_controller(struct device *dev) { static const short btn_bit[32] = { BTN_C, BTN_B, BTN_A, BTN_START, -1, -1, -1, -1, @@ -157,7 +157,7 @@ fail: return error; } -static int __devexit remove_maple_controller(struct device *dev) +static int remove_maple_controller(struct device *dev) { struct maple_device *mdev = to_maple_dev(dev); struct dc_pad *pad = maple_get_drvdata(mdev); @@ -175,7 +175,7 @@ static struct maple_driver dc_pad_driver = { .drv = { .name = "Dreamcast_controller", .probe = probe_maple_controller, - .remove = __devexit_p(remove_maple_controller), + .remove = remove_maple_controller, }, }; diff --git a/drivers/input/joystick/walkera0701.c b/drivers/input/joystick/walkera0701.c index 4dfa1eed4b7c..f8f892b076e8 100644 --- a/drivers/input/joystick/walkera0701.c +++ b/drivers/input/joystick/walkera0701.c @@ -196,6 +196,7 @@ static void walkera0701_close(struct input_dev *dev) struct walkera_dev *w = input_get_drvdata(dev); parport_disable_irq(w->parport); + hrtimer_cancel(&w->timer); } static int walkera0701_connect(struct walkera_dev *w, int parport) @@ -224,6 +225,9 @@ static int walkera0701_connect(struct walkera_dev *w, int parport) if (parport_claim(w->pardevice)) goto init_err1; + hrtimer_init(&w->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + w->timer.function = timer_handler; + w->input_dev = input_allocate_device(); if (!w->input_dev) goto init_err2; @@ -254,8 +258,6 @@ static int walkera0701_connect(struct walkera_dev *w, int parport) if (err) goto init_err3; - hrtimer_init(&w->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - w->timer.function = timer_handler; return 0; init_err3: @@ -271,7 +273,6 @@ static int walkera0701_connect(struct walkera_dev *w, int parport) static void walkera0701_disconnect(struct walkera_dev *w) { - hrtimer_cancel(&w->timer); input_unregister_device(w->input_dev); parport_release(w->pardevice); parport_unregister_device(w->pardevice); diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 83811e45d633..d6cbfe9df218 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -118,11 +118,12 @@ static const struct xpad_device { u8 xtype; } xpad_device[] = { { 0x045e, 0x0202, "Microsoft X-Box pad v1 (US)", 0, XTYPE_XBOX }, - { 0x045e, 0x0289, "Microsoft X-Box pad v2 (US)", 0, XTYPE_XBOX }, { 0x045e, 0x0285, "Microsoft X-Box pad (Japan)", 0, XTYPE_XBOX }, { 0x045e, 0x0287, "Microsoft Xbox Controller S", 0, XTYPE_XBOX }, + { 0x045e, 0x0289, "Microsoft X-Box pad v2 (US)", 0, XTYPE_XBOX }, + { 0x045e, 0x028e, "Microsoft X-Box 360 pad", 0, XTYPE_XBOX360 }, + { 0x045e, 0x0291, "Xbox 360 Wireless Receiver (XBOX)", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360W }, { 0x045e, 0x0719, "Xbox 360 Wireless Receiver", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360W }, - { 0x0c12, 0x8809, "RedOctane Xbox Dance Pad", DANCEPAD_MAP_CONFIG, XTYPE_XBOX }, { 0x044f, 0x0f07, "Thrustmaster, Inc. Controller", 0, XTYPE_XBOX }, { 0x046d, 0xc242, "Logitech Chillstream Controller", 0, XTYPE_XBOX360 }, { 0x046d, 0xca84, "Logitech Xbox Cordless Controller", 0, XTYPE_XBOX }, @@ -136,9 +137,12 @@ static const struct xpad_device { { 0x0738, 0x4540, "Mad Catz Beat Pad", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX }, { 0x0738, 0x4556, "Mad Catz Lynx Wireless Controller", 0, XTYPE_XBOX }, { 0x0738, 0x4716, "Mad Catz Wired Xbox 360 Controller", 0, XTYPE_XBOX360 }, + { 0x0738, 0x4728, "Mad Catz Street Fighter IV FightPad", XTYPE_XBOX360 }, { 0x0738, 0x4738, "Mad Catz Wired Xbox 360 Controller (SFIV)", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, { 0x0738, 0x6040, "Mad Catz Beat Pad Pro", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX }, + { 0x0738, 0xbeef, "Mad Catz JOYTECH NEO SE Advanced GamePad", XTYPE_XBOX360 }, { 0x0c12, 0x8802, "Zeroplus Xbox Controller", 0, XTYPE_XBOX }, + { 0x0c12, 0x8809, "RedOctane Xbox Dance Pad", DANCEPAD_MAP_CONFIG, XTYPE_XBOX }, { 0x0c12, 0x880a, "Pelican Eclipse PL-2023", 0, XTYPE_XBOX }, { 0x0c12, 0x8810, "Zeroplus Xbox Controller", 0, XTYPE_XBOX }, { 0x0c12, 0x9902, "HAMA VibraX - *FAULTY HARDWARE*", 0, XTYPE_XBOX }, @@ -148,24 +152,28 @@ static const struct xpad_device { { 0x0e6f, 0x0003, "Logic3 Freebird wireless Controller", 0, XTYPE_XBOX }, { 0x0e6f, 0x0005, "Eclipse wireless Controller", 0, XTYPE_XBOX }, { 0x0e6f, 0x0006, "Edge wireless Controller", 0, XTYPE_XBOX }, - { 0x0e6f, 0x0006, "Pelican 'TSZ' Wired Xbox 360 Controller", 0, XTYPE_XBOX360 }, + { 0x0e6f, 0x0105, "HSM3 Xbox360 dancepad", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360 }, { 0x0e6f, 0x0201, "Pelican PL-3601 'TSZ' Wired Xbox 360 Controller", 0, XTYPE_XBOX360 }, + { 0x0e6f, 0x0213, "Afterglow Gamepad for Xbox 360", 0, XTYPE_XBOX360 }, { 0x0e8f, 0x0201, "SmartJoy Frag Xpad/PS2 adaptor", 0, XTYPE_XBOX }, + { 0x0f0d, 0x000d, "Hori Fighting Stick EX2", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, + { 0x0f0d, 0x0016, "Hori Real Arcade Pro.EX", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, { 0x0f30, 0x0202, "Joytech Advanced Controller", 0, XTYPE_XBOX }, { 0x0f30, 0x8888, "BigBen XBMiniPad Controller", 0, XTYPE_XBOX }, { 0x102c, 0xff0c, "Joytech Wireless Advanced Controller", 0, XTYPE_XBOX }, - { 0x12ab, 0x8809, "Xbox DDR dancepad", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX }, { 0x12ab, 0x0004, "Honey Bee Xbox360 dancepad", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360 }, - { 0x0e6f, 0x0105, "HSM3 Xbox360 dancepad", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360 }, + { 0x12ab, 0x8809, "Xbox DDR dancepad", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX }, { 0x1430, 0x4748, "RedOctane Guitar Hero X-plorer", 0, XTYPE_XBOX360 }, { 0x1430, 0x8888, "TX6500+ Dance Pad (first generation)", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX }, { 0x146b, 0x0601, "BigBen Interactive XBOX 360 Controller", 0, XTYPE_XBOX360 }, - { 0x045e, 0x028e, "Microsoft X-Box 360 pad", 0, XTYPE_XBOX360 }, + { 0x1689, 0xfd00, "Razer Onza Tournament Edition", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360 }, { 0x1bad, 0x0002, "Harmonix Rock Band Guitar", 0, XTYPE_XBOX360 }, { 0x1bad, 0x0003, "Harmonix Rock Band Drumkit", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360 }, - { 0x0f0d, 0x0016, "Hori Real Arcade Pro.EX", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, - { 0x0f0d, 0x000d, "Hori Fighting Stick EX2", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, - { 0x1689, 0xfd00, "Razer Onza Tournament Edition", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360 }, + { 0x1bad, 0xf016, "Mad Catz Xbox 360 Controller", 0, XTYPE_XBOX360 }, + { 0x1bad, 0xf028, "Street Fighter IV FightPad", 0, XTYPE_XBOX360 }, + { 0x1bad, 0xf901, "Gamestop Xbox 360 Controller", 0, XTYPE_XBOX360 }, + { 0x1bad, 0xf903, "Tron Xbox 360 controller", 0, XTYPE_XBOX360 }, + { 0x24c6, 0x5300, "PowerA MINI PROEX Controller", 0, XTYPE_XBOX360 }, { 0xffff, 0xffff, "Chinese-made Xbox Controller", 0, XTYPE_XBOX }, { 0x0000, 0x0000, "Generic X-Box pad", 0, XTYPE_UNKNOWN } }; @@ -235,7 +243,7 @@ static const signed short xpad_abs_triggers[] = { { XPAD_XBOX360_VENDOR_PROTOCOL(vend,1) }, \ { XPAD_XBOX360_VENDOR_PROTOCOL(vend,129) } -static struct usb_device_id xpad_table [] = { +static struct usb_device_id xpad_table[] = { { USB_INTERFACE_INFO('X', 'B', 0) }, /* X-Box USB-IF not approved class */ XPAD_XBOX360_VENDOR(0x045e), /* Microsoft X-Box 360 controllers */ XPAD_XBOX360_VENDOR(0x046d), /* Logitech X-Box 360 style controllers */ @@ -248,10 +256,11 @@ static struct usb_device_id xpad_table [] = { XPAD_XBOX360_VENDOR(0x1bad), /* Harminix Rock Band Guitar and Drums */ XPAD_XBOX360_VENDOR(0x0f0d), /* Hori Controllers */ XPAD_XBOX360_VENDOR(0x1689), /* Razer Onza */ + XPAD_XBOX360_VENDOR(0x24c6), /* PowerA Controllers */ { } }; -MODULE_DEVICE_TABLE (usb, xpad_table); +MODULE_DEVICE_TABLE(usb, xpad_table); struct usb_xpad { struct input_dev *dev; /* input device interface */ @@ -783,7 +792,7 @@ static int xpad_open(struct input_dev *dev) struct usb_xpad *xpad = input_get_drvdata(dev); /* URB was submitted in probe */ - if(xpad->xtype == XTYPE_XBOX360W) + if (xpad->xtype == XTYPE_XBOX360W) return 0; xpad->irq_in->dev = xpad->udev; diff --git a/drivers/input/keyboard/Kconfig b/drivers/input/keyboard/Kconfig index febead4bf8a5..5a240c60342d 100644 --- a/drivers/input/keyboard/Kconfig +++ b/drivers/input/keyboard/Kconfig @@ -134,7 +134,7 @@ config KEYBOARD_QT1070 config KEYBOARD_QT2160 tristate "Atmel AT42QT2160 Touch Sensor Chip" - depends on I2C && EXPERIMENTAL + depends on I2C help If you say yes here you get support for Atmel AT42QT2160 Touch Sensor chip as a keyboard input. diff --git a/drivers/input/keyboard/adp5520-keys.c b/drivers/input/keyboard/adp5520-keys.c index e9e8674dfda1..ef26b17fb159 100644 --- a/drivers/input/keyboard/adp5520-keys.c +++ b/drivers/input/keyboard/adp5520-keys.c @@ -69,7 +69,7 @@ static int adp5520_keys_notifier(struct notifier_block *nb, return 0; } -static int __devinit adp5520_keys_probe(struct platform_device *pdev) +static int adp5520_keys_probe(struct platform_device *pdev) { struct adp5520_keys_platform_data *pdata = pdev->dev.platform_data; struct input_dev *input; @@ -182,7 +182,7 @@ err: return ret; } -static int __devexit adp5520_keys_remove(struct platform_device *pdev) +static int adp5520_keys_remove(struct platform_device *pdev) { struct adp5520_keys *dev = platform_get_drvdata(pdev); @@ -200,7 +200,7 @@ static struct platform_driver adp5520_keys_driver = { .owner = THIS_MODULE, }, .probe = adp5520_keys_probe, - .remove = __devexit_p(adp5520_keys_remove), + .remove = adp5520_keys_remove, }; module_platform_driver(adp5520_keys_driver); diff --git a/drivers/input/keyboard/adp5588-keys.c b/drivers/input/keyboard/adp5588-keys.c index b083bf10f139..dbd2047f1641 100644 --- a/drivers/input/keyboard/adp5588-keys.c +++ b/drivers/input/keyboard/adp5588-keys.c @@ -145,7 +145,7 @@ static int adp5588_gpio_direction_output(struct gpio_chip *chip, return ret; } -static int __devinit adp5588_build_gpiomap(struct adp5588_kpad *kpad, +static int adp5588_build_gpiomap(struct adp5588_kpad *kpad, const struct adp5588_kpad_platform_data *pdata) { bool pin_used[ADP5588_MAXGPIO]; @@ -170,7 +170,7 @@ static int __devinit adp5588_build_gpiomap(struct adp5588_kpad *kpad, return n_unused; } -static int __devinit adp5588_gpio_add(struct adp5588_kpad *kpad) +static int adp5588_gpio_add(struct adp5588_kpad *kpad) { struct device *dev = &kpad->client->dev; const struct adp5588_kpad_platform_data *pdata = dev->platform_data; @@ -224,7 +224,7 @@ static int __devinit adp5588_gpio_add(struct adp5588_kpad *kpad) return 0; } -static void __devexit adp5588_gpio_remove(struct adp5588_kpad *kpad) +static void adp5588_gpio_remove(struct adp5588_kpad *kpad) { struct device *dev = &kpad->client->dev; const struct adp5588_kpad_platform_data *pdata = dev->platform_data; @@ -319,7 +319,7 @@ static irqreturn_t adp5588_irq(int irq, void *handle) return IRQ_HANDLED; } -static int __devinit adp5588_setup(struct i2c_client *client) +static int adp5588_setup(struct i2c_client *client) { const struct adp5588_kpad_platform_data *pdata = client->dev.platform_data; const struct adp5588_gpio_platform_data *gpio_data = pdata->gpio_data; @@ -382,7 +382,7 @@ static int __devinit adp5588_setup(struct i2c_client *client) return 0; } -static void __devinit adp5588_report_switch_state(struct adp5588_kpad *kpad) +static void adp5588_report_switch_state(struct adp5588_kpad *kpad) { int gpi_stat1 = adp5588_read(kpad->client, GPIO_DAT_STAT1); int gpi_stat2 = adp5588_read(kpad->client, GPIO_DAT_STAT2); @@ -420,8 +420,8 @@ static void __devinit adp5588_report_switch_state(struct adp5588_kpad *kpad) } -static int __devinit adp5588_probe(struct i2c_client *client, - const struct i2c_device_id *id) +static int adp5588_probe(struct i2c_client *client, + const struct i2c_device_id *id) { struct adp5588_kpad *kpad; const struct adp5588_kpad_platform_data *pdata = client->dev.platform_data; @@ -587,7 +587,7 @@ static int __devinit adp5588_probe(struct i2c_client *client, return error; } -static int __devexit adp5588_remove(struct i2c_client *client) +static int adp5588_remove(struct i2c_client *client) { struct adp5588_kpad *kpad = i2c_get_clientdata(client); @@ -650,7 +650,7 @@ static struct i2c_driver adp5588_driver = { #endif }, .probe = adp5588_probe, - .remove = __devexit_p(adp5588_remove), + .remove = adp5588_remove, .id_table = adp5588_id, }; diff --git a/drivers/input/keyboard/adp5589-keys.c b/drivers/input/keyboard/adp5589-keys.c index 74e603213386..67d12b3427c9 100644 --- a/drivers/input/keyboard/adp5589-keys.c +++ b/drivers/input/keyboard/adp5589-keys.c @@ -464,7 +464,7 @@ static int adp5589_gpio_direction_output(struct gpio_chip *chip, return ret; } -static int __devinit adp5589_build_gpiomap(struct adp5589_kpad *kpad, +static int adp5589_build_gpiomap(struct adp5589_kpad *kpad, const struct adp5589_kpad_platform_data *pdata) { bool pin_used[ADP5589_MAXGPIO]; @@ -496,7 +496,7 @@ static int __devinit adp5589_build_gpiomap(struct adp5589_kpad *kpad, return n_unused; } -static int __devinit adp5589_gpio_add(struct adp5589_kpad *kpad) +static int adp5589_gpio_add(struct adp5589_kpad *kpad) { struct device *dev = &kpad->client->dev; const struct adp5589_kpad_platform_data *pdata = dev->platform_data; @@ -550,7 +550,7 @@ static int __devinit adp5589_gpio_add(struct adp5589_kpad *kpad) return 0; } -static void __devexit adp5589_gpio_remove(struct adp5589_kpad *kpad) +static void adp5589_gpio_remove(struct adp5589_kpad *kpad) { struct device *dev = &kpad->client->dev; const struct adp5589_kpad_platform_data *pdata = dev->platform_data; @@ -641,8 +641,7 @@ static irqreturn_t adp5589_irq(int irq, void *handle) return IRQ_HANDLED; } -static int __devinit adp5589_get_evcode(struct adp5589_kpad *kpad, - unsigned short key) +static int adp5589_get_evcode(struct adp5589_kpad *kpad, unsigned short key) { int i; @@ -655,7 +654,7 @@ static int __devinit adp5589_get_evcode(struct adp5589_kpad *kpad, return -EINVAL; } -static int __devinit adp5589_setup(struct adp5589_kpad *kpad) +static int adp5589_setup(struct adp5589_kpad *kpad) { struct i2c_client *client = kpad->client; const struct adp5589_kpad_platform_data *pdata = @@ -820,7 +819,7 @@ static int __devinit adp5589_setup(struct adp5589_kpad *kpad) return 0; } -static void __devinit adp5589_report_switch_state(struct adp5589_kpad *kpad) +static void adp5589_report_switch_state(struct adp5589_kpad *kpad) { int gpi_stat_tmp, pin_loc; int i; @@ -860,8 +859,8 @@ static void __devinit adp5589_report_switch_state(struct adp5589_kpad *kpad) input_sync(kpad->input); } -static int __devinit adp5589_probe(struct i2c_client *client, - const struct i2c_device_id *id) +static int adp5589_probe(struct i2c_client *client, + const struct i2c_device_id *id) { struct adp5589_kpad *kpad; const struct adp5589_kpad_platform_data *pdata = @@ -1045,7 +1044,7 @@ err_free_mem: return error; } -static int __devexit adp5589_remove(struct i2c_client *client) +static int adp5589_remove(struct i2c_client *client) { struct adp5589_kpad *kpad = i2c_get_clientdata(client); @@ -1104,7 +1103,7 @@ static struct i2c_driver adp5589_driver = { .pm = &adp5589_dev_pm_ops, }, .probe = adp5589_probe, - .remove = __devexit_p(adp5589_remove), + .remove = adp5589_remove, .id_table = adp5589_id, }; diff --git a/drivers/input/keyboard/bf54x-keys.c b/drivers/input/keyboard/bf54x-keys.c index 8eb9116e0a5f..20b9fa91fb9e 100644 --- a/drivers/input/keyboard/bf54x-keys.c +++ b/drivers/input/keyboard/bf54x-keys.c @@ -177,7 +177,7 @@ static irqreturn_t bfin_kpad_isr(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit bfin_kpad_probe(struct platform_device *pdev) +static int bfin_kpad_probe(struct platform_device *pdev) { struct bf54x_kpad *bf54x_kpad; struct bfin_kpad_platform_data *pdata = pdev->dev.platform_data; @@ -331,7 +331,7 @@ out: return error; } -static int __devexit bfin_kpad_remove(struct platform_device *pdev) +static int bfin_kpad_remove(struct platform_device *pdev) { struct bfin_kpad_platform_data *pdata = pdev->dev.platform_data; struct bf54x_kpad *bf54x_kpad = platform_get_drvdata(pdev); @@ -390,7 +390,7 @@ static struct platform_driver bfin_kpad_device_driver = { .owner = THIS_MODULE, }, .probe = bfin_kpad_probe, - .remove = __devexit_p(bfin_kpad_remove), + .remove = bfin_kpad_remove, .suspend = bfin_kpad_suspend, .resume = bfin_kpad_resume, }; diff --git a/drivers/input/keyboard/davinci_keyscan.c b/drivers/input/keyboard/davinci_keyscan.c index d5bacbb479b0..4e4e453ea15e 100644 --- a/drivers/input/keyboard/davinci_keyscan.c +++ b/drivers/input/keyboard/davinci_keyscan.c @@ -303,7 +303,7 @@ fail1: return error; } -static int __devexit davinci_ks_remove(struct platform_device *pdev) +static int davinci_ks_remove(struct platform_device *pdev) { struct davinci_ks *davinci_ks = platform_get_drvdata(pdev); @@ -326,7 +326,7 @@ static struct platform_driver davinci_ks_driver = { .name = "davinci_keyscan", .owner = THIS_MODULE, }, - .remove = __devexit_p(davinci_ks_remove), + .remove = davinci_ks_remove, }; static int __init davinci_ks_init(void) diff --git a/drivers/input/keyboard/ep93xx_keypad.c b/drivers/input/keyboard/ep93xx_keypad.c index 7363402de8d4..9857e8fd0987 100644 --- a/drivers/input/keyboard/ep93xx_keypad.c +++ b/drivers/input/keyboard/ep93xx_keypad.c @@ -232,7 +232,7 @@ static int ep93xx_keypad_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(ep93xx_keypad_pm_ops, ep93xx_keypad_suspend, ep93xx_keypad_resume); -static int __devinit ep93xx_keypad_probe(struct platform_device *pdev) +static int ep93xx_keypad_probe(struct platform_device *pdev) { struct ep93xx_keypad *keypad; const struct matrix_keymap_data *keymap_data; @@ -346,7 +346,7 @@ failed_free: return err; } -static int __devexit ep93xx_keypad_remove(struct platform_device *pdev) +static int ep93xx_keypad_remove(struct platform_device *pdev) { struct ep93xx_keypad *keypad = platform_get_drvdata(pdev); struct resource *res; @@ -380,7 +380,7 @@ static struct platform_driver ep93xx_keypad_driver = { .pm = &ep93xx_keypad_pm_ops, }, .probe = ep93xx_keypad_probe, - .remove = __devexit_p(ep93xx_keypad_remove), + .remove = ep93xx_keypad_remove, }; module_platform_driver(ep93xx_keypad_driver); diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c index 6a68041c261d..d327f5a2bb0e 100644 --- a/drivers/input/keyboard/gpio_keys.c +++ b/drivers/input/keyboard/gpio_keys.c @@ -423,10 +423,10 @@ out: return IRQ_HANDLED; } -static int __devinit gpio_keys_setup_key(struct platform_device *pdev, - struct input_dev *input, - struct gpio_button_data *bdata, - const struct gpio_keys_button *button) +static int gpio_keys_setup_key(struct platform_device *pdev, + struct input_dev *input, + struct gpio_button_data *bdata, + const struct gpio_keys_button *button) { const char *desc = button->desc ? button->desc : "gpio_keys"; struct device *dev = &pdev->dev; @@ -440,21 +440,13 @@ static int __devinit gpio_keys_setup_key(struct platform_device *pdev, if (gpio_is_valid(button->gpio)) { - error = gpio_request(button->gpio, desc); + error = gpio_request_one(button->gpio, GPIOF_IN, desc); if (error < 0) { dev_err(dev, "Failed to request GPIO %d, error %d\n", button->gpio, error); return error; } - error = gpio_direction_input(button->gpio); - if (error < 0) { - dev_err(dev, - "Failed to configure direction for GPIO %d, error %d\n", - button->gpio, error); - goto fail; - } - if (button->debounce_interval) { error = gpio_set_debounce(button->gpio, button->debounce_interval * 1000); @@ -526,12 +518,35 @@ fail: return error; } +static void gpio_keys_report_state(struct gpio_keys_drvdata *ddata) +{ + struct input_dev *input = ddata->input; + int i; + + for (i = 0; i < ddata->pdata->nbuttons; i++) { + struct gpio_button_data *bdata = &ddata->data[i]; + if (gpio_is_valid(bdata->button->gpio)) + gpio_keys_gpio_report_event(bdata); + } + input_sync(input); +} + static int gpio_keys_open(struct input_dev *input) { struct gpio_keys_drvdata *ddata = input_get_drvdata(input); const struct gpio_keys_platform_data *pdata = ddata->pdata; + int error; + + if (pdata->enable) { + error = pdata->enable(input->dev.parent); + if (error) + return error; + } + + /* Report current state of buttons that are connected to GPIOs */ + gpio_keys_report_state(ddata); - return pdata->enable ? pdata->enable(input->dev.parent) : 0; + return 0; } static void gpio_keys_close(struct input_dev *input) @@ -551,7 +566,7 @@ static void gpio_keys_close(struct input_dev *input) /* * Translate OpenFirmware node properties into platform_data */ -static struct gpio_keys_platform_data * __devinit +static struct gpio_keys_platform_data * gpio_keys_get_devtree_pdata(struct device *dev) { struct device_node *node, *pp; @@ -658,7 +673,7 @@ static void gpio_remove_key(struct gpio_button_data *bdata) gpio_free(bdata->button->gpio); } -static int __devinit gpio_keys_probe(struct platform_device *pdev) +static int gpio_keys_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; const struct gpio_keys_platform_data *pdata = dev_get_platdata(dev); @@ -731,14 +746,6 @@ static int __devinit gpio_keys_probe(struct platform_device *pdev) goto fail3; } - /* get current state of buttons that are connected to GPIOs */ - for (i = 0; i < pdata->nbuttons; i++) { - struct gpio_button_data *bdata = &ddata->data[i]; - if (gpio_is_valid(bdata->button->gpio)) - gpio_keys_gpio_report_event(bdata); - } - input_sync(input); - device_init_wakeup(&pdev->dev, wakeup); return 0; @@ -760,7 +767,7 @@ static int __devinit gpio_keys_probe(struct platform_device *pdev) return error; } -static int __devexit gpio_keys_remove(struct platform_device *pdev) +static int gpio_keys_remove(struct platform_device *pdev) { struct gpio_keys_drvdata *ddata = platform_get_drvdata(pdev); struct input_dev *input = ddata->input; @@ -788,6 +795,7 @@ static int __devexit gpio_keys_remove(struct platform_device *pdev) static int gpio_keys_suspend(struct device *dev) { struct gpio_keys_drvdata *ddata = dev_get_drvdata(dev); + struct input_dev *input = ddata->input; int i; if (device_may_wakeup(dev)) { @@ -796,6 +804,11 @@ static int gpio_keys_suspend(struct device *dev) if (bdata->button->wakeup) enable_irq_wake(bdata->irq); } + } else { + mutex_lock(&input->mutex); + if (input->users) + gpio_keys_close(input); + mutex_unlock(&input->mutex); } return 0; @@ -804,18 +817,27 @@ static int gpio_keys_suspend(struct device *dev) static int gpio_keys_resume(struct device *dev) { struct gpio_keys_drvdata *ddata = dev_get_drvdata(dev); + struct input_dev *input = ddata->input; + int error = 0; int i; - for (i = 0; i < ddata->pdata->nbuttons; i++) { - struct gpio_button_data *bdata = &ddata->data[i]; - if (bdata->button->wakeup && device_may_wakeup(dev)) - disable_irq_wake(bdata->irq); - - if (gpio_is_valid(bdata->button->gpio)) - gpio_keys_gpio_report_event(bdata); + if (device_may_wakeup(dev)) { + for (i = 0; i < ddata->pdata->nbuttons; i++) { + struct gpio_button_data *bdata = &ddata->data[i]; + if (bdata->button->wakeup) + disable_irq_wake(bdata->irq); + } + } else { + mutex_lock(&input->mutex); + if (input->users) + error = gpio_keys_open(input); + mutex_unlock(&input->mutex); } - input_sync(ddata->input); + if (error) + return error; + + gpio_keys_report_state(ddata); return 0; } #endif @@ -824,7 +846,7 @@ static SIMPLE_DEV_PM_OPS(gpio_keys_pm_ops, gpio_keys_suspend, gpio_keys_resume); static struct platform_driver gpio_keys_device_driver = { .probe = gpio_keys_probe, - .remove = __devexit_p(gpio_keys_remove), + .remove = gpio_keys_remove, .driver = { .name = "gpio-keys", .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/gpio_keys_polled.c b/drivers/input/keyboard/gpio_keys_polled.c index f2142de789e7..f686fd970553 100644 --- a/drivers/input/keyboard/gpio_keys_polled.c +++ b/drivers/input/keyboard/gpio_keys_polled.c @@ -103,8 +103,7 @@ static void gpio_keys_polled_close(struct input_polled_dev *dev) } #ifdef CONFIG_OF -static struct gpio_keys_platform_data * __devinit -gpio_keys_polled_get_devtree_pdata(struct device *dev) +static struct gpio_keys_platform_data *gpio_keys_polled_get_devtree_pdata(struct device *dev) { struct device_node *node, *pp; struct gpio_keys_platform_data *pdata; @@ -196,7 +195,7 @@ gpio_keys_polled_get_devtree_pdata(struct device *dev) } #endif -static int __devinit gpio_keys_polled_probe(struct platform_device *pdev) +static int gpio_keys_polled_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; const struct gpio_keys_platform_data *pdata = dev_get_platdata(dev); @@ -246,7 +245,6 @@ static int __devinit gpio_keys_polled_probe(struct platform_device *pdev) input = poll_dev->input; - input->evbit[0] = BIT(EV_KEY); input->name = pdev->name; input->phys = DRV_NAME"/input0"; input->dev.parent = &pdev->dev; @@ -256,6 +254,10 @@ static int __devinit gpio_keys_polled_probe(struct platform_device *pdev) input->id.product = 0x0001; input->id.version = 0x0100; + __set_bit(EV_KEY, input->evbit); + if (pdata->rep) + __set_bit(EV_REP, input->evbit); + for (i = 0; i < pdata->nbuttons; i++) { struct gpio_keys_button *button = &pdata->buttons[i]; struct gpio_keys_button_data *bdata = &bdev->data[i]; @@ -268,22 +270,14 @@ static int __devinit gpio_keys_polled_probe(struct platform_device *pdev) goto err_free_gpio; } - error = gpio_request(gpio, - button->desc ? button->desc : DRV_NAME); + error = gpio_request_one(gpio, GPIOF_IN, + button->desc ?: DRV_NAME); if (error) { dev_err(dev, "unable to claim gpio %u, err=%d\n", gpio, error); goto err_free_gpio; } - error = gpio_direction_input(gpio); - if (error) { - dev_err(dev, - "unable to set direction on gpio %u, err=%d\n", - gpio, error); - goto err_free_gpio; - } - bdata->can_sleep = gpio_cansleep(gpio); bdata->last_state = -1; bdata->threshold = DIV_ROUND_UP(button->debounce_interval, @@ -329,7 +323,7 @@ err_free_pdata: return error; } -static int __devexit gpio_keys_polled_remove(struct platform_device *pdev) +static int gpio_keys_polled_remove(struct platform_device *pdev) { struct gpio_keys_polled_dev *bdev = platform_get_drvdata(pdev); const struct gpio_keys_platform_data *pdata = bdev->pdata; @@ -357,7 +351,7 @@ static int __devexit gpio_keys_polled_remove(struct platform_device *pdev) static struct platform_driver gpio_keys_polled_driver = { .probe = gpio_keys_polled_probe, - .remove = __devexit_p(gpio_keys_polled_remove), + .remove = gpio_keys_polled_remove, .driver = { .name = DRV_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/hilkbd.c b/drivers/input/keyboard/hilkbd.c index 5f72440b50c8..198dc07a1be5 100644 --- a/drivers/input/keyboard/hilkbd.c +++ b/drivers/input/keyboard/hilkbd.c @@ -200,7 +200,7 @@ static void hil_do(unsigned char cmd, unsigned char *data, unsigned int len) /* initialize HIL */ -static int __devinit hil_keyb_init(void) +static int hil_keyb_init(void) { unsigned char c; unsigned int i, kbid; @@ -286,7 +286,7 @@ err1: return err; } -static void __devexit hil_keyb_exit(void) +static void hil_keyb_exit(void) { if (HIL_IRQ) free_irq(HIL_IRQ, hil_dev.dev_id); @@ -299,7 +299,7 @@ static void __devexit hil_keyb_exit(void) } #if defined(CONFIG_PARISC) -static int __devinit hil_probe_chip(struct parisc_device *dev) +static int hil_probe_chip(struct parisc_device *dev) { /* Only allow one HIL keyboard */ if (hil_dev.dev) @@ -320,7 +320,7 @@ static int __devinit hil_probe_chip(struct parisc_device *dev) return hil_keyb_init(); } -static int __devexit hil_remove_chip(struct parisc_device *dev) +static int hil_remove_chip(struct parisc_device *dev) { hil_keyb_exit(); @@ -341,7 +341,7 @@ static struct parisc_driver hil_driver = { .name = "hil", .id_table = hil_tbl, .probe = hil_probe_chip, - .remove = __devexit_p(hil_remove_chip), + .remove = hil_remove_chip, }; static int __init hil_init(void) diff --git a/drivers/input/keyboard/imx_keypad.c b/drivers/input/keyboard/imx_keypad.c index cdc252612c0b..6d150e3e1f55 100644 --- a/drivers/input/keyboard/imx_keypad.c +++ b/drivers/input/keyboard/imx_keypad.c @@ -362,7 +362,8 @@ static void imx_keypad_inhibit(struct imx_keypad *keypad) writew(reg_val, keypad->mmio_base + KPSR); /* Colums as open drain and disable all rows */ - writew(0xff00, keypad->mmio_base + KPCR); + reg_val = (keypad->cols_en_mask & 0xff) << 8; + writew(reg_val, keypad->mmio_base + KPCR); } static void imx_keypad_close(struct input_dev *dev) @@ -413,7 +414,7 @@ open_err: return -EIO; } -static int __devinit imx_keypad_probe(struct platform_device *pdev) +static int imx_keypad_probe(struct platform_device *pdev) { const struct matrix_keymap_data *keymap_data = pdev->dev.platform_data; struct imx_keypad *keypad; @@ -554,7 +555,7 @@ failed_rel_mem: return error; } -static int __devexit imx_keypad_remove(struct platform_device *pdev) +static int imx_keypad_remove(struct platform_device *pdev) { struct imx_keypad *keypad = platform_get_drvdata(pdev); struct resource *res; @@ -632,7 +633,7 @@ static struct platform_driver imx_keypad_driver = { .pm = &imx_kbd_pm_ops, }, .probe = imx_keypad_probe, - .remove = __devexit_p(imx_keypad_remove), + .remove = imx_keypad_remove, }; module_platform_driver(imx_keypad_driver); diff --git a/drivers/input/keyboard/jornada680_kbd.c b/drivers/input/keyboard/jornada680_kbd.c index 24f3ea01c4d5..74e75a6e8deb 100644 --- a/drivers/input/keyboard/jornada680_kbd.c +++ b/drivers/input/keyboard/jornada680_kbd.c @@ -179,7 +179,7 @@ static void jornadakbd680_poll(struct input_polled_dev *dev) memcpy(jornadakbd->old_scan, jornadakbd->new_scan, JORNADA_SCAN_SIZE); } -static int __devinit jornada680kbd_probe(struct platform_device *pdev) +static int jornada680kbd_probe(struct platform_device *pdev) { struct jornadakbd *jornadakbd; struct input_polled_dev *poll_dev; @@ -240,7 +240,7 @@ static int __devinit jornada680kbd_probe(struct platform_device *pdev) } -static int __devexit jornada680kbd_remove(struct platform_device *pdev) +static int jornada680kbd_remove(struct platform_device *pdev) { struct jornadakbd *jornadakbd = platform_get_drvdata(pdev); @@ -258,7 +258,7 @@ static struct platform_driver jornada680kbd_driver = { .owner = THIS_MODULE, }, .probe = jornada680kbd_probe, - .remove = __devexit_p(jornada680kbd_remove), + .remove = jornada680kbd_remove, }; module_platform_driver(jornada680kbd_driver); diff --git a/drivers/input/keyboard/jornada720_kbd.c b/drivers/input/keyboard/jornada720_kbd.c index 9d639fa1afbd..5ceef636df2f 100644 --- a/drivers/input/keyboard/jornada720_kbd.c +++ b/drivers/input/keyboard/jornada720_kbd.c @@ -94,7 +94,7 @@ static irqreturn_t jornada720_kbd_interrupt(int irq, void *dev_id) return IRQ_HANDLED; }; -static int __devinit jornada720_kbd_probe(struct platform_device *pdev) +static int jornada720_kbd_probe(struct platform_device *pdev) { struct jornadakbd *jornadakbd; struct input_dev *input_dev; @@ -152,7 +152,7 @@ static int __devinit jornada720_kbd_probe(struct platform_device *pdev) return err; }; -static int __devexit jornada720_kbd_remove(struct platform_device *pdev) +static int jornada720_kbd_remove(struct platform_device *pdev) { struct jornadakbd *jornadakbd = platform_get_drvdata(pdev); @@ -173,6 +173,6 @@ static struct platform_driver jornada720_kbd_driver = { .owner = THIS_MODULE, }, .probe = jornada720_kbd_probe, - .remove = __devexit_p(jornada720_kbd_remove), + .remove = jornada720_kbd_remove, }; module_platform_driver(jornada720_kbd_driver); diff --git a/drivers/input/keyboard/lm8323.c b/drivers/input/keyboard/lm8323.c index 39ac2787e275..93c812662134 100644 --- a/drivers/input/keyboard/lm8323.c +++ b/drivers/input/keyboard/lm8323.c @@ -624,7 +624,7 @@ static ssize_t lm8323_set_disable(struct device *dev, } static DEVICE_ATTR(disable_kp, 0644, lm8323_show_disable, lm8323_set_disable); -static int __devinit lm8323_probe(struct i2c_client *client, +static int lm8323_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct lm8323_platform_data *pdata = client->dev.platform_data; @@ -764,7 +764,7 @@ fail1: return err; } -static int __devexit lm8323_remove(struct i2c_client *client) +static int lm8323_remove(struct i2c_client *client) { struct lm8323_chip *lm = i2c_get_clientdata(client); int i; @@ -846,7 +846,7 @@ static struct i2c_driver lm8323_i2c_driver = { .pm = &lm8323_pm_ops, }, .probe = lm8323_probe, - .remove = __devexit_p(lm8323_remove), + .remove = lm8323_remove, .id_table = lm8323_id, }; MODULE_DEVICE_TABLE(i2c, lm8323_id); diff --git a/drivers/input/keyboard/lm8333.c b/drivers/input/keyboard/lm8333.c index 081fd9effa8c..5a8ca35dc9af 100644 --- a/drivers/input/keyboard/lm8333.c +++ b/drivers/input/keyboard/lm8333.c @@ -128,7 +128,7 @@ static irqreturn_t lm8333_irq_thread(int irq, void *data) return IRQ_HANDLED; } -static int __devinit lm8333_probe(struct i2c_client *client, +static int lm8333_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct lm8333_platform_data *pdata = client->dev.platform_data; @@ -202,7 +202,7 @@ static int __devinit lm8333_probe(struct i2c_client *client, return err; } -static int __devexit lm8333_remove(struct i2c_client *client) +static int lm8333_remove(struct i2c_client *client) { struct lm8333 *lm8333 = i2c_get_clientdata(client); @@ -225,7 +225,7 @@ static struct i2c_driver lm8333_driver = { .owner = THIS_MODULE, }, .probe = lm8333_probe, - .remove = __devexit_p(lm8333_remove), + .remove = lm8333_remove, .id_table = lm8333_id, }; module_i2c_driver(lm8333_driver); diff --git a/drivers/input/keyboard/locomokbd.c b/drivers/input/keyboard/locomokbd.c index b1ab29861e1c..c94d610b9d78 100644 --- a/drivers/input/keyboard/locomokbd.c +++ b/drivers/input/keyboard/locomokbd.c @@ -46,7 +46,7 @@ MODULE_LICENSE("GPL"); #define KEY_CENTER KEY_F15 static const unsigned char -locomokbd_keycode[LOCOMOKBD_NUMKEYS] __devinitconst = { +locomokbd_keycode[LOCOMOKBD_NUMKEYS] = { 0, KEY_ESC, KEY_ACTIVITY, 0, 0, 0, 0, 0, 0, 0, /* 0 - 9 */ 0, 0, 0, 0, 0, 0, 0, KEY_MENU, KEY_HOME, KEY_CONTACT, /* 10 - 19 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 20 - 29 */ @@ -236,7 +236,7 @@ static void locomokbd_close(struct input_dev *dev) locomo_writel(r, locomokbd->base + LOCOMO_KIC); } -static int __devinit locomokbd_probe(struct locomo_dev *dev) +static int locomokbd_probe(struct locomo_dev *dev) { struct locomokbd *locomokbd; struct input_dev *input_dev; @@ -321,7 +321,7 @@ static int __devinit locomokbd_probe(struct locomo_dev *dev) return err; } -static int __devexit locomokbd_remove(struct locomo_dev *dev) +static int locomokbd_remove(struct locomo_dev *dev) { struct locomokbd *locomokbd = locomo_get_drvdata(dev); @@ -345,7 +345,7 @@ static struct locomo_driver keyboard_driver = { }, .devid = LOCOMO_DEVID_KEYBOARD, .probe = locomokbd_probe, - .remove = __devexit_p(locomokbd_remove), + .remove = locomokbd_remove, }; static int __init locomokbd_init(void) diff --git a/drivers/input/keyboard/lpc32xx-keys.c b/drivers/input/keyboard/lpc32xx-keys.c index dd786c8a7584..1b8add6cfb9d 100644 --- a/drivers/input/keyboard/lpc32xx-keys.c +++ b/drivers/input/keyboard/lpc32xx-keys.c @@ -139,7 +139,7 @@ static void lpc32xx_kscan_close(struct input_dev *dev) clk_disable_unprepare(kscandat->clk); } -static int __devinit lpc32xx_parse_dt(struct device *dev, +static int lpc32xx_parse_dt(struct device *dev, struct lpc32xx_kscan_drv *kscandat) { struct device_node *np = dev->of_node; @@ -166,7 +166,7 @@ static int __devinit lpc32xx_parse_dt(struct device *dev, return 0; } -static int __devinit lpc32xx_kscan_probe(struct platform_device *pdev) +static int lpc32xx_kscan_probe(struct platform_device *pdev) { struct lpc32xx_kscan_drv *kscandat; struct input_dev *input; @@ -310,7 +310,7 @@ err_free_mem: return error; } -static int __devexit lpc32xx_kscan_remove(struct platform_device *pdev) +static int lpc32xx_kscan_remove(struct platform_device *pdev) { struct lpc32xx_kscan_drv *kscandat = platform_get_drvdata(pdev); @@ -377,7 +377,7 @@ MODULE_DEVICE_TABLE(of, lpc32xx_kscan_match); static struct platform_driver lpc32xx_kscan_driver = { .probe = lpc32xx_kscan_probe, - .remove = __devexit_p(lpc32xx_kscan_remove), + .remove = lpc32xx_kscan_remove, .driver = { .name = DRV_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/matrix_keypad.c b/drivers/input/keyboard/matrix_keypad.c index 18b72372028a..f4ff0dda7597 100644 --- a/drivers/input/keyboard/matrix_keypad.c +++ b/drivers/input/keyboard/matrix_keypad.c @@ -23,6 +23,9 @@ #include <linux/gpio.h> #include <linux/input/matrix_keypad.h> #include <linux/slab.h> +#include <linux/of.h> +#include <linux/of_gpio.h> +#include <linux/of_platform.h> struct matrix_keypad { const struct matrix_keypad_platform_data *pdata; @@ -37,8 +40,6 @@ struct matrix_keypad { bool scan_pending; bool stopped; bool gpio_all_disabled; - - unsigned short keycodes[]; }; /* @@ -118,6 +119,7 @@ static void matrix_keypad_scan(struct work_struct *work) struct matrix_keypad *keypad = container_of(work, struct matrix_keypad, work.work); struct input_dev *input_dev = keypad->input_dev; + const unsigned short *keycodes = input_dev->keycode; const struct matrix_keypad_platform_data *pdata = keypad->pdata; uint32_t new_state[MATRIX_MAX_COLS]; int row, col, code; @@ -153,7 +155,7 @@ static void matrix_keypad_scan(struct work_struct *work) code = MATRIX_SCAN_CODE(row, col, keypad->row_shift); input_event(input_dev, EV_MSC, MSC_SCAN, code); input_report_key(input_dev, - keypad->keycodes[code], + keycodes[code], new_state[col] & (1 << row)); } } @@ -299,8 +301,8 @@ static int matrix_keypad_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(matrix_keypad_pm_ops, matrix_keypad_suspend, matrix_keypad_resume); -static int __devinit matrix_keypad_init_gpio(struct platform_device *pdev, - struct matrix_keypad *keypad) +static int matrix_keypad_init_gpio(struct platform_device *pdev, + struct matrix_keypad *keypad) { const struct matrix_keypad_platform_data *pdata = keypad->pdata; int i, err; @@ -394,33 +396,95 @@ static void matrix_keypad_free_gpio(struct matrix_keypad *keypad) gpio_free(pdata->col_gpios[i]); } -static int __devinit matrix_keypad_probe(struct platform_device *pdev) +#ifdef CONFIG_OF +static struct matrix_keypad_platform_data * +matrix_keypad_parse_dt(struct device *dev) +{ + struct matrix_keypad_platform_data *pdata; + struct device_node *np = dev->of_node; + unsigned int *gpios; + int i; + + if (!np) { + dev_err(dev, "device lacks DT data\n"); + return ERR_PTR(-ENODEV); + } + + pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL); + if (!pdata) { + dev_err(dev, "could not allocate memory for platform data\n"); + return ERR_PTR(-ENOMEM); + } + + pdata->num_row_gpios = of_gpio_named_count(np, "row-gpios"); + pdata->num_col_gpios = of_gpio_named_count(np, "col-gpios"); + if (!pdata->num_row_gpios || !pdata->num_col_gpios) { + dev_err(dev, "number of keypad rows/columns not specified\n"); + return ERR_PTR(-EINVAL); + } + + if (of_get_property(np, "linux,no-autorepeat", NULL)) + pdata->no_autorepeat = true; + if (of_get_property(np, "linux,wakeup", NULL)) + pdata->wakeup = true; + if (of_get_property(np, "gpio-activelow", NULL)) + pdata->active_low = true; + + of_property_read_u32(np, "debounce-delay-ms", &pdata->debounce_ms); + of_property_read_u32(np, "col-scan-delay-us", + &pdata->col_scan_delay_us); + + gpios = devm_kzalloc(dev, + sizeof(unsigned int) * + (pdata->num_row_gpios + pdata->num_col_gpios), + GFP_KERNEL); + if (!gpios) { + dev_err(dev, "could not allocate memory for gpios\n"); + return ERR_PTR(-ENOMEM); + } + + for (i = 0; i < pdata->num_row_gpios; i++) + gpios[i] = of_get_named_gpio(np, "row-gpios", i); + + for (i = 0; i < pdata->num_col_gpios; i++) + gpios[pdata->num_row_gpios + i] = + of_get_named_gpio(np, "col-gpios", i); + + pdata->row_gpios = gpios; + pdata->col_gpios = &gpios[pdata->num_row_gpios]; + + return pdata; +} +#else +static inline struct matrix_keypad_platform_data * +matrix_keypad_parse_dt(struct device *dev) +{ + dev_err(dev, "no platform data defined\n"); + + return ERR_PTR(-EINVAL); +} +#endif + +static int matrix_keypad_probe(struct platform_device *pdev) { const struct matrix_keypad_platform_data *pdata; - const struct matrix_keymap_data *keymap_data; struct matrix_keypad *keypad; struct input_dev *input_dev; - unsigned int row_shift; - size_t keymap_size; int err; - pdata = pdev->dev.platform_data; + pdata = dev_get_platdata(&pdev->dev); if (!pdata) { - dev_err(&pdev->dev, "no platform data defined\n"); - return -EINVAL; - } - - keymap_data = pdata->keymap_data; - if (!keymap_data) { + pdata = matrix_keypad_parse_dt(&pdev->dev); + if (IS_ERR(pdata)) { + dev_err(&pdev->dev, "no platform data defined\n"); + return PTR_ERR(pdata); + } + } else if (!pdata->keymap_data) { dev_err(&pdev->dev, "no keymap data defined\n"); return -EINVAL; } - row_shift = get_count_order(pdata->num_col_gpios); - keymap_size = (pdata->num_row_gpios << row_shift) * - sizeof(keypad->keycodes[0]); - keypad = kzalloc(sizeof(struct matrix_keypad) + keymap_size, - GFP_KERNEL); + keypad = kzalloc(sizeof(struct matrix_keypad), GFP_KERNEL); input_dev = input_allocate_device(); if (!keypad || !input_dev) { err = -ENOMEM; @@ -429,7 +493,7 @@ static int __devinit matrix_keypad_probe(struct platform_device *pdev) keypad->input_dev = input_dev; keypad->pdata = pdata; - keypad->row_shift = row_shift; + keypad->row_shift = get_count_order(pdata->num_col_gpios); keypad->stopped = true; INIT_DELAYED_WORK(&keypad->work, matrix_keypad_scan); spin_lock_init(&keypad->lock); @@ -440,12 +504,14 @@ static int __devinit matrix_keypad_probe(struct platform_device *pdev) input_dev->open = matrix_keypad_start; input_dev->close = matrix_keypad_stop; - err = matrix_keypad_build_keymap(keymap_data, NULL, + err = matrix_keypad_build_keymap(pdata->keymap_data, NULL, pdata->num_row_gpios, pdata->num_col_gpios, - keypad->keycodes, input_dev); - if (err) + NULL, input_dev); + if (err) { + dev_err(&pdev->dev, "failed to build keymap\n"); goto err_free_mem; + } if (!pdata->no_autorepeat) __set_bit(EV_REP, input_dev->evbit); @@ -473,7 +539,7 @@ err_free_mem: return err; } -static int __devexit matrix_keypad_remove(struct platform_device *pdev) +static int matrix_keypad_remove(struct platform_device *pdev) { struct matrix_keypad *keypad = platform_get_drvdata(pdev); @@ -488,13 +554,22 @@ static int __devexit matrix_keypad_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_OF +static const struct of_device_id matrix_keypad_dt_match[] = { + { .compatible = "gpio-matrix-keypad" }, + { } +}; +MODULE_DEVICE_TABLE(of, matrix_keypad_dt_match); +#endif + static struct platform_driver matrix_keypad_driver = { .probe = matrix_keypad_probe, - .remove = __devexit_p(matrix_keypad_remove), + .remove = matrix_keypad_remove, .driver = { .name = "matrix-keypad", .owner = THIS_MODULE, .pm = &matrix_keypad_pm_ops, + .of_match_table = of_match_ptr(matrix_keypad_dt_match), }, }; module_platform_driver(matrix_keypad_driver); diff --git a/drivers/input/keyboard/max7359_keypad.c b/drivers/input/keyboard/max7359_keypad.c index 8edada8ae712..7c7af2b01e65 100644 --- a/drivers/input/keyboard/max7359_keypad.c +++ b/drivers/input/keyboard/max7359_keypad.c @@ -179,7 +179,7 @@ static void max7359_initialize(struct i2c_client *client) max7359_fall_deepsleep(client); } -static int __devinit max7359_probe(struct i2c_client *client, +static int max7359_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct matrix_keymap_data *keymap_data = client->dev.platform_data; @@ -260,7 +260,7 @@ failed_free_mem: return error; } -static int __devexit max7359_remove(struct i2c_client *client) +static int max7359_remove(struct i2c_client *client) { struct max7359_keypad *keypad = i2c_get_clientdata(client); @@ -312,7 +312,7 @@ static struct i2c_driver max7359_i2c_driver = { .pm = &max7359_pm, }, .probe = max7359_probe, - .remove = __devexit_p(max7359_remove), + .remove = max7359_remove, .id_table = max7359_ids, }; diff --git a/drivers/input/keyboard/mcs_touchkey.c b/drivers/input/keyboard/mcs_touchkey.c index 0d77f6c84950..7c236f9c6a51 100644 --- a/drivers/input/keyboard/mcs_touchkey.c +++ b/drivers/input/keyboard/mcs_touchkey.c @@ -97,7 +97,7 @@ static irqreturn_t mcs_touchkey_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit mcs_touchkey_probe(struct i2c_client *client, +static int mcs_touchkey_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct mcs_platform_data *pdata; @@ -200,7 +200,7 @@ err_free_mem: return error; } -static int __devexit mcs_touchkey_remove(struct i2c_client *client) +static int mcs_touchkey_remove(struct i2c_client *client) { struct mcs_touchkey_data *data = i2c_get_clientdata(client); @@ -270,7 +270,7 @@ static struct i2c_driver mcs_touchkey_driver = { .pm = &mcs_touchkey_pm_ops, }, .probe = mcs_touchkey_probe, - .remove = __devexit_p(mcs_touchkey_remove), + .remove = mcs_touchkey_remove, .shutdown = mcs_touchkey_shutdown, .id_table = mcs_touchkey_id, }; diff --git a/drivers/input/keyboard/mpr121_touchkey.c b/drivers/input/keyboard/mpr121_touchkey.c index 7613f1cac951..f7f3e9a9fd3f 100644 --- a/drivers/input/keyboard/mpr121_touchkey.c +++ b/drivers/input/keyboard/mpr121_touchkey.c @@ -71,7 +71,7 @@ struct mpr121_init_register { u8 val; }; -static const struct mpr121_init_register init_reg_table[] __devinitconst = { +static const struct mpr121_init_register init_reg_table[] = { { MHD_RISING_ADDR, 0x1 }, { NHD_RISING_ADDR, 0x1 }, { MHD_FALLING_ADDR, 0x1 }, @@ -123,7 +123,7 @@ out: return IRQ_HANDLED; } -static int __devinit mpr121_phys_init(const struct mpr121_platform_data *pdata, +static int mpr121_phys_init(const struct mpr121_platform_data *pdata, struct mpr121_touchkey *mpr121, struct i2c_client *client) { @@ -185,8 +185,8 @@ err_i2c_write: return ret; } -static int __devinit mpr_touchkey_probe(struct i2c_client *client, - const struct i2c_device_id *id) +static int mpr_touchkey_probe(struct i2c_client *client, + const struct i2c_device_id *id) { const struct mpr121_platform_data *pdata = client->dev.platform_data; struct mpr121_touchkey *mpr121; @@ -272,7 +272,7 @@ err_free_mem: return error; } -static int __devexit mpr_touchkey_remove(struct i2c_client *client) +static int mpr_touchkey_remove(struct i2c_client *client) { struct mpr121_touchkey *mpr121 = i2c_get_clientdata(client); @@ -327,7 +327,7 @@ static struct i2c_driver mpr_touchkey_driver = { }, .id_table = mpr121_id, .probe = mpr_touchkey_probe, - .remove = __devexit_p(mpr_touchkey_remove), + .remove = mpr_touchkey_remove, }; module_i2c_driver(mpr_touchkey_driver); diff --git a/drivers/input/keyboard/nomadik-ske-keypad.c b/drivers/input/keyboard/nomadik-ske-keypad.c index 49f5fa64e0b1..0e6a8151fee3 100644 --- a/drivers/input/keyboard/nomadik-ske-keypad.c +++ b/drivers/input/keyboard/nomadik-ske-keypad.c @@ -67,6 +67,7 @@ struct ske_keypad { const struct ske_keypad_platform_data *board; unsigned short keymap[SKE_KPD_NUM_ROWS * SKE_KPD_NUM_COLS]; struct clk *clk; + struct clk *pclk; spinlock_t ske_keypad_lock; }; @@ -271,11 +272,18 @@ static int __init ske_keypad_probe(struct platform_device *pdev) goto err_free_mem_region; } + keypad->pclk = clk_get(&pdev->dev, "apb_pclk"); + if (IS_ERR(keypad->pclk)) { + dev_err(&pdev->dev, "failed to get pclk\n"); + error = PTR_ERR(keypad->pclk); + goto err_iounmap; + } + keypad->clk = clk_get(&pdev->dev, NULL); if (IS_ERR(keypad->clk)) { dev_err(&pdev->dev, "failed to get clk\n"); error = PTR_ERR(keypad->clk); - goto err_iounmap; + goto err_pclk; } input->id.bustype = BUS_HOST; @@ -287,14 +295,25 @@ static int __init ske_keypad_probe(struct platform_device *pdev) keypad->keymap, input); if (error) { dev_err(&pdev->dev, "Failed to build keymap\n"); - goto err_iounmap; + goto err_clk; } input_set_capability(input, EV_MSC, MSC_SCAN); if (!plat->no_autorepeat) __set_bit(EV_REP, input->evbit); - clk_enable(keypad->clk); + error = clk_prepare_enable(keypad->pclk); + if (error) { + dev_err(&pdev->dev, "Failed to prepare/enable pclk\n"); + goto err_clk; + } + + error = clk_prepare_enable(keypad->clk); + if (error) { + dev_err(&pdev->dev, "Failed to prepare/enable clk\n"); + goto err_pclk_disable; + } + /* go through board initialization helpers */ if (keypad->board->init) @@ -330,8 +349,13 @@ static int __init ske_keypad_probe(struct platform_device *pdev) err_free_irq: free_irq(keypad->irq, keypad); err_clk_disable: - clk_disable(keypad->clk); + clk_disable_unprepare(keypad->clk); +err_pclk_disable: + clk_disable_unprepare(keypad->pclk); +err_clk: clk_put(keypad->clk); +err_pclk: + clk_put(keypad->pclk); err_iounmap: iounmap(keypad->reg_base); err_free_mem_region: @@ -342,7 +366,7 @@ err_free_mem: return error; } -static int __devexit ske_keypad_remove(struct platform_device *pdev) +static int ske_keypad_remove(struct platform_device *pdev) { struct ske_keypad *keypad = platform_get_drvdata(pdev); struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0); @@ -351,7 +375,7 @@ static int __devexit ske_keypad_remove(struct platform_device *pdev) input_unregister_device(keypad->input); - clk_disable(keypad->clk); + clk_disable_unprepare(keypad->clk); clk_put(keypad->clk); if (keypad->board->exit) @@ -403,7 +427,7 @@ static struct platform_driver ske_keypad_driver = { .owner = THIS_MODULE, .pm = &ske_keypad_dev_pm_ops, }, - .remove = __devexit_p(ske_keypad_remove), + .remove = ske_keypad_remove, }; static int __init ske_keypad_init(void) diff --git a/drivers/input/keyboard/omap-keypad.c b/drivers/input/keyboard/omap-keypad.c index 4a5fcc8026f5..d0d5226d9cd4 100644 --- a/drivers/input/keyboard/omap-keypad.c +++ b/drivers/input/keyboard/omap-keypad.c @@ -244,7 +244,7 @@ static int omap_kp_resume(struct platform_device *dev) #define omap_kp_resume NULL #endif -static int __devinit omap_kp_probe(struct platform_device *pdev) +static int omap_kp_probe(struct platform_device *pdev) { struct omap_kp *omap_kp; struct input_dev *input_dev; @@ -357,7 +357,7 @@ err2: return -EINVAL; } -static int __devexit omap_kp_remove(struct platform_device *pdev) +static int omap_kp_remove(struct platform_device *pdev) { struct omap_kp *omap_kp = platform_get_drvdata(pdev); @@ -379,7 +379,7 @@ static int __devexit omap_kp_remove(struct platform_device *pdev) static struct platform_driver omap_kp_driver = { .probe = omap_kp_probe, - .remove = __devexit_p(omap_kp_remove), + .remove = omap_kp_remove, .suspend = omap_kp_suspend, .resume = omap_kp_resume, .driver = { diff --git a/drivers/input/keyboard/omap4-keypad.c b/drivers/input/keyboard/omap4-keypad.c index c05f98c41410..e25b022692cd 100644 --- a/drivers/input/keyboard/omap4-keypad.c +++ b/drivers/input/keyboard/omap4-keypad.c @@ -211,8 +211,8 @@ static void omap4_keypad_close(struct input_dev *input) } #ifdef CONFIG_OF -static int __devinit omap4_keypad_parse_dt(struct device *dev, - struct omap4_keypad *keypad_data) +static int omap4_keypad_parse_dt(struct device *dev, + struct omap4_keypad *keypad_data) { struct device_node *np = dev->of_node; @@ -241,7 +241,7 @@ static inline int omap4_keypad_parse_dt(struct device *dev, } #endif -static int __devinit omap4_keypad_probe(struct platform_device *pdev) +static int omap4_keypad_probe(struct platform_device *pdev) { const struct omap4_keypad_platform_data *pdata = dev_get_platdata(&pdev->dev); @@ -406,7 +406,7 @@ err_free_keypad: return error; } -static int __devexit omap4_keypad_remove(struct platform_device *pdev) +static int omap4_keypad_remove(struct platform_device *pdev) { struct omap4_keypad *keypad_data = platform_get_drvdata(pdev); struct resource *res; @@ -440,7 +440,7 @@ MODULE_DEVICE_TABLE(of, omap_keypad_dt_match); static struct platform_driver omap4_keypad_driver = { .probe = omap4_keypad_probe, - .remove = __devexit_p(omap4_keypad_remove), + .remove = omap4_keypad_remove, .driver = { .name = "omap4-keypad", .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/opencores-kbd.c b/drivers/input/keyboard/opencores-kbd.c index abe728c7b88e..7ac5f174c6f7 100644 --- a/drivers/input/keyboard/opencores-kbd.c +++ b/drivers/input/keyboard/opencores-kbd.c @@ -37,7 +37,7 @@ static irqreturn_t opencores_kbd_isr(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit opencores_kbd_probe(struct platform_device *pdev) +static int opencores_kbd_probe(struct platform_device *pdev) { struct input_dev *input; struct opencores_kbd *opencores_kbd; @@ -139,7 +139,7 @@ static int __devinit opencores_kbd_probe(struct platform_device *pdev) return error; } -static int __devexit opencores_kbd_remove(struct platform_device *pdev) +static int opencores_kbd_remove(struct platform_device *pdev) { struct opencores_kbd *opencores_kbd = platform_get_drvdata(pdev); @@ -158,7 +158,7 @@ static int __devexit opencores_kbd_remove(struct platform_device *pdev) static struct platform_driver opencores_kbd_device_driver = { .probe = opencores_kbd_probe, - .remove = __devexit_p(opencores_kbd_remove), + .remove = opencores_kbd_remove, .driver = { .name = "opencores-kbd", }, diff --git a/drivers/input/keyboard/pmic8xxx-keypad.c b/drivers/input/keyboard/pmic8xxx-keypad.c index 52c34657d301..74339e139d43 100644 --- a/drivers/input/keyboard/pmic8xxx-keypad.c +++ b/drivers/input/keyboard/pmic8xxx-keypad.c @@ -397,7 +397,7 @@ static irqreturn_t pmic8xxx_kp_irq(int irq, void *data) return IRQ_HANDLED; } -static int __devinit pmic8xxx_kpd_init(struct pmic8xxx_kp *kp) +static int pmic8xxx_kpd_init(struct pmic8xxx_kp *kp) { int bits, rc, cycles; u8 scan_val = 0, ctrl_val = 0; @@ -447,7 +447,7 @@ static int __devinit pmic8xxx_kpd_init(struct pmic8xxx_kp *kp) } -static int __devinit pmic8xxx_kp_config_gpio(int gpio_start, int num_gpios, +static int pmic8xxx_kp_config_gpio(int gpio_start, int num_gpios, struct pmic8xxx_kp *kp, struct pm_gpio *gpio_config) { int rc, i; @@ -518,7 +518,7 @@ static void pmic8xxx_kp_close(struct input_dev *dev) * - set irq edge type. * - enable the keypad controller. */ -static int __devinit pmic8xxx_kp_probe(struct platform_device *pdev) +static int pmic8xxx_kp_probe(struct platform_device *pdev) { const struct pm8xxx_keypad_platform_data *pdata = dev_get_platdata(&pdev->dev); @@ -712,7 +712,7 @@ err_alloc_device: return rc; } -static int __devexit pmic8xxx_kp_remove(struct platform_device *pdev) +static int pmic8xxx_kp_remove(struct platform_device *pdev) { struct pmic8xxx_kp *kp = platform_get_drvdata(pdev); @@ -773,7 +773,7 @@ static SIMPLE_DEV_PM_OPS(pm8xxx_kp_pm_ops, static struct platform_driver pmic8xxx_kp_driver = { .probe = pmic8xxx_kp_probe, - .remove = __devexit_p(pmic8xxx_kp_remove), + .remove = pmic8xxx_kp_remove, .driver = { .name = PM8XXX_KEYPAD_DEV_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/pxa27x_keypad.c b/drivers/input/keyboard/pxa27x_keypad.c index cad9d5dd5973..5330d8fbf6c0 100644 --- a/drivers/input/keyboard/pxa27x_keypad.c +++ b/drivers/input/keyboard/pxa27x_keypad.c @@ -482,7 +482,7 @@ static const struct dev_pm_ops pxa27x_keypad_pm_ops = { }; #endif -static int __devinit pxa27x_keypad_probe(struct platform_device *pdev) +static int pxa27x_keypad_probe(struct platform_device *pdev) { struct pxa27x_keypad_platform_data *pdata = pdev->dev.platform_data; struct pxa27x_keypad *keypad; @@ -595,7 +595,7 @@ failed_free: return error; } -static int __devexit pxa27x_keypad_remove(struct platform_device *pdev) +static int pxa27x_keypad_remove(struct platform_device *pdev) { struct pxa27x_keypad *keypad = platform_get_drvdata(pdev); struct resource *res; @@ -620,7 +620,7 @@ MODULE_ALIAS("platform:pxa27x-keypad"); static struct platform_driver pxa27x_keypad_driver = { .probe = pxa27x_keypad_probe, - .remove = __devexit_p(pxa27x_keypad_remove), + .remove = pxa27x_keypad_remove, .driver = { .name = "pxa27x-keypad", .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/pxa930_rotary.c b/drivers/input/keyboard/pxa930_rotary.c index 41488f9add20..bcad95be73aa 100644 --- a/drivers/input/keyboard/pxa930_rotary.c +++ b/drivers/input/keyboard/pxa930_rotary.c @@ -82,7 +82,7 @@ static void pxa930_rotary_close(struct input_dev *dev) clear_sbcr(r); } -static int __devinit pxa930_rotary_probe(struct platform_device *pdev) +static int pxa930_rotary_probe(struct platform_device *pdev) { struct pxa930_rotary_platform_data *pdata = pdev->dev.platform_data; struct pxa930_rotary *r; @@ -174,7 +174,7 @@ failed_free: return err; } -static int __devexit pxa930_rotary_remove(struct platform_device *pdev) +static int pxa930_rotary_remove(struct platform_device *pdev) { struct pxa930_rotary *r = platform_get_drvdata(pdev); @@ -193,7 +193,7 @@ static struct platform_driver pxa930_rotary_driver = { .owner = THIS_MODULE, }, .probe = pxa930_rotary_probe, - .remove = __devexit_p(pxa930_rotary_remove), + .remove = pxa930_rotary_remove, }; module_platform_driver(pxa930_rotary_driver); diff --git a/drivers/input/keyboard/qt1070.c b/drivers/input/keyboard/qt1070.c index ca68f2992d72..42b773b3125a 100644 --- a/drivers/input/keyboard/qt1070.c +++ b/drivers/input/keyboard/qt1070.c @@ -91,7 +91,7 @@ static int qt1070_write(struct i2c_client *client, u8 reg, u8 data) return ret; } -static bool __devinit qt1070_identify(struct i2c_client *client) +static bool qt1070_identify(struct i2c_client *client) { int id, ver; @@ -140,7 +140,7 @@ static irqreturn_t qt1070_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit qt1070_probe(struct i2c_client *client, +static int qt1070_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct qt1070_data *data; @@ -230,7 +230,7 @@ err_free_mem: return err; } -static int __devexit qt1070_remove(struct i2c_client *client) +static int qt1070_remove(struct i2c_client *client) { struct qt1070_data *data = i2c_get_clientdata(client); @@ -256,7 +256,7 @@ static struct i2c_driver qt1070_driver = { }, .id_table = qt1070_id, .probe = qt1070_probe, - .remove = __devexit_p(qt1070_remove), + .remove = qt1070_remove, }; module_i2c_driver(qt1070_driver); diff --git a/drivers/input/keyboard/qt2160.c b/drivers/input/keyboard/qt2160.c index 76b7d430d03a..3dc2b0f27b0c 100644 --- a/drivers/input/keyboard/qt2160.c +++ b/drivers/input/keyboard/qt2160.c @@ -183,7 +183,7 @@ static void qt2160_worker(struct work_struct *work) qt2160_schedule_read(qt2160); } -static int __devinit qt2160_read(struct i2c_client *client, u8 reg) +static int qt2160_read(struct i2c_client *client, u8 reg) { int ret; @@ -204,29 +204,20 @@ static int __devinit qt2160_read(struct i2c_client *client, u8 reg) return ret; } -static int __devinit qt2160_write(struct i2c_client *client, u8 reg, u8 data) +static int qt2160_write(struct i2c_client *client, u8 reg, u8 data) { - int error; - - error = i2c_smbus_write_byte(client, reg); - if (error) { - dev_err(&client->dev, - "couldn't send request. Returned %d\n", error); - return error; - } + int ret; - error = i2c_smbus_write_byte(client, data); - if (error) { + ret = i2c_smbus_write_byte_data(client, reg, data); + if (ret < 0) dev_err(&client->dev, - "couldn't write data. Returned %d\n", error); - return error; - } + "couldn't write data. Returned %d\n", ret); - return error; + return ret; } -static bool __devinit qt2160_identify(struct i2c_client *client) +static bool qt2160_identify(struct i2c_client *client) { int id, ver, rev; @@ -257,7 +248,7 @@ static bool __devinit qt2160_identify(struct i2c_client *client) return true; } -static int __devinit qt2160_probe(struct i2c_client *client, +static int qt2160_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct qt2160_data *qt2160; @@ -344,7 +335,7 @@ err_free_mem: return error; } -static int __devexit qt2160_remove(struct i2c_client *client) +static int qt2160_remove(struct i2c_client *client) { struct qt2160_data *qt2160 = i2c_get_clientdata(client); @@ -375,7 +366,7 @@ static struct i2c_driver qt2160_driver = { .id_table = qt2160_idtable, .probe = qt2160_probe, - .remove = __devexit_p(qt2160_remove), + .remove = qt2160_remove, }; module_i2c_driver(qt2160_driver); diff --git a/drivers/input/keyboard/samsung-keypad.c b/drivers/input/keyboard/samsung-keypad.c index 9d7a111486f7..22e357b51024 100644 --- a/drivers/input/keyboard/samsung-keypad.c +++ b/drivers/input/keyboard/samsung-keypad.c @@ -309,7 +309,7 @@ static void samsung_keypad_parse_dt_gpio(struct device *dev, struct samsung_keypad *keypad) { struct device_node *np = dev->of_node; - int gpio, ret, row, col; + int gpio, error, row, col; for (row = 0; row < keypad->rows; row++) { gpio = of_get_named_gpio(np, "row-gpios", row); @@ -320,10 +320,11 @@ static void samsung_keypad_parse_dt_gpio(struct device *dev, continue; } - ret = gpio_request(gpio, "keypad-row"); - if (ret) - dev_err(dev, "keypad row[%d] gpio request failed\n", - row); + error = devm_gpio_request(dev, gpio, "keypad-row"); + if (error) + dev_err(dev, + "keypad row[%d] gpio request failed: %d\n", + row, error); } for (col = 0; col < keypad->cols; col++) { @@ -335,38 +336,22 @@ static void samsung_keypad_parse_dt_gpio(struct device *dev, continue; } - ret = gpio_request(gpio, "keypad-col"); - if (ret) - dev_err(dev, "keypad column[%d] gpio request failed\n", - col); + error = devm_gpio_request(dev, gpio, "keypad-col"); + if (error) + dev_err(dev, + "keypad column[%d] gpio request failed: %d\n", + col, error); } } - -static void samsung_keypad_dt_gpio_free(struct samsung_keypad *keypad) -{ - int cnt; - - for (cnt = 0; cnt < keypad->rows; cnt++) - if (gpio_is_valid(keypad->row_gpios[cnt])) - gpio_free(keypad->row_gpios[cnt]); - - for (cnt = 0; cnt < keypad->cols; cnt++) - if (gpio_is_valid(keypad->col_gpios[cnt])) - gpio_free(keypad->col_gpios[cnt]); -} #else static struct samsung_keypad_platdata *samsung_keypad_parse_dt(struct device *dev) { return NULL; } - -static void samsung_keypad_dt_gpio_free(struct samsung_keypad *keypad) -{ -} #endif -static int __devinit samsung_keypad_probe(struct platform_device *pdev) +static int samsung_keypad_probe(struct platform_device *pdev) { const struct samsung_keypad_platdata *pdata; const struct matrix_keymap_data *keymap_data; @@ -405,36 +390,30 @@ static int __devinit samsung_keypad_probe(struct platform_device *pdev) row_shift = get_count_order(pdata->cols); keymap_size = (pdata->rows << row_shift) * sizeof(keypad->keycodes[0]); - keypad = kzalloc(sizeof(*keypad) + keymap_size, GFP_KERNEL); - input_dev = input_allocate_device(); - if (!keypad || !input_dev) { - error = -ENOMEM; - goto err_free_mem; - } + keypad = devm_kzalloc(&pdev->dev, sizeof(*keypad) + keymap_size, + GFP_KERNEL); + input_dev = devm_input_allocate_device(&pdev->dev); + if (!keypad || !input_dev) + return -ENOMEM; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!res) { - error = -ENODEV; - goto err_free_mem; - } + if (!res) + return -ENODEV; - keypad->base = ioremap(res->start, resource_size(res)); - if (!keypad->base) { - error = -EBUSY; - goto err_free_mem; - } + keypad->base = devm_ioremap(&pdev->dev, res->start, resource_size(res)); + if (!keypad->base) + return -EBUSY; - keypad->clk = clk_get(&pdev->dev, "keypad"); + keypad->clk = devm_clk_get(&pdev->dev, "keypad"); if (IS_ERR(keypad->clk)) { dev_err(&pdev->dev, "failed to get keypad clk\n"); - error = PTR_ERR(keypad->clk); - goto err_unmap_base; + return PTR_ERR(keypad->clk); } error = clk_prepare(keypad->clk); if (error) { dev_err(&pdev->dev, "keypad clock prepare failed\n"); - goto err_put_clk; + return error; } keypad->input_dev = input_dev; @@ -479,14 +458,15 @@ static int __devinit samsung_keypad_probe(struct platform_device *pdev) keypad->irq = platform_get_irq(pdev, 0); if (keypad->irq < 0) { error = keypad->irq; - goto err_put_clk; + goto err_unprepare_clk; } - error = request_threaded_irq(keypad->irq, NULL, samsung_keypad_irq, - IRQF_ONESHOT, dev_name(&pdev->dev), keypad); + error = devm_request_threaded_irq(&pdev->dev, keypad->irq, NULL, + samsung_keypad_irq, IRQF_ONESHOT, + dev_name(&pdev->dev), keypad); if (error) { dev_err(&pdev->dev, "failed to register keypad interrupt\n"); - goto err_put_clk; + goto err_unprepare_clk; } device_init_wakeup(&pdev->dev, pdata->wakeup); @@ -495,7 +475,7 @@ static int __devinit samsung_keypad_probe(struct platform_device *pdev) error = input_register_device(keypad->input_dev); if (error) - goto err_free_irq; + goto err_disable_runtime_pm; if (pdev->dev.of_node) { devm_kfree(&pdev->dev, (void *)pdata->keymap_data->keymap); @@ -504,26 +484,16 @@ static int __devinit samsung_keypad_probe(struct platform_device *pdev) } return 0; -err_free_irq: - free_irq(keypad->irq, keypad); +err_disable_runtime_pm: pm_runtime_disable(&pdev->dev); device_init_wakeup(&pdev->dev, 0); platform_set_drvdata(pdev, NULL); err_unprepare_clk: clk_unprepare(keypad->clk); -err_put_clk: - clk_put(keypad->clk); - samsung_keypad_dt_gpio_free(keypad); -err_unmap_base: - iounmap(keypad->base); -err_free_mem: - input_free_device(input_dev); - kfree(keypad); - return error; } -static int __devexit samsung_keypad_remove(struct platform_device *pdev) +static int samsung_keypad_remove(struct platform_device *pdev) { struct samsung_keypad *keypad = platform_get_drvdata(pdev); @@ -533,18 +503,7 @@ static int __devexit samsung_keypad_remove(struct platform_device *pdev) input_unregister_device(keypad->input_dev); - /* - * It is safe to free IRQ after unregistering device because - * samsung_keypad_close will shut off interrupts. - */ - free_irq(keypad->irq, keypad); - clk_unprepare(keypad->clk); - clk_put(keypad->clk); - samsung_keypad_dt_gpio_free(keypad); - - iounmap(keypad->base); - kfree(keypad); return 0; } @@ -685,7 +644,7 @@ MODULE_DEVICE_TABLE(platform, samsung_keypad_driver_ids); static struct platform_driver samsung_keypad_driver = { .probe = samsung_keypad_probe, - .remove = __devexit_p(samsung_keypad_remove), + .remove = samsung_keypad_remove, .driver = { .name = "samsung-keypad", .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/sh_keysc.c b/drivers/input/keyboard/sh_keysc.c index da54ad5db154..fdb9eb2df380 100644 --- a/drivers/input/keyboard/sh_keysc.c +++ b/drivers/input/keyboard/sh_keysc.c @@ -162,7 +162,7 @@ static irqreturn_t sh_keysc_isr(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit sh_keysc_probe(struct platform_device *pdev) +static int sh_keysc_probe(struct platform_device *pdev) { struct sh_keysc_priv *priv; struct sh_keysc_info *pdata; @@ -272,7 +272,7 @@ static int __devinit sh_keysc_probe(struct platform_device *pdev) return error; } -static int __devexit sh_keysc_remove(struct platform_device *pdev) +static int sh_keysc_remove(struct platform_device *pdev) { struct sh_keysc_priv *priv = platform_get_drvdata(pdev); @@ -331,7 +331,7 @@ static SIMPLE_DEV_PM_OPS(sh_keysc_dev_pm_ops, static struct platform_driver sh_keysc_device_driver = { .probe = sh_keysc_probe, - .remove = __devexit_p(sh_keysc_remove), + .remove = sh_keysc_remove, .driver = { .name = "sh_keysc", .pm = &sh_keysc_dev_pm_ops, diff --git a/drivers/input/keyboard/spear-keyboard.c b/drivers/input/keyboard/spear-keyboard.c index c7ca97f44bfb..695d237417d6 100644 --- a/drivers/input/keyboard/spear-keyboard.c +++ b/drivers/input/keyboard/spear-keyboard.c @@ -55,15 +55,15 @@ struct spear_kbd { struct input_dev *input; - struct resource *res; void __iomem *io_base; struct clk *clk; unsigned int irq; unsigned int mode; + unsigned int suspended_rate; unsigned short last_key; unsigned short keycodes[NUM_ROWS * NUM_COLS]; bool rep; - unsigned int suspended_rate; + bool irq_wake_enabled; u32 mode_ctl_reg; }; @@ -146,7 +146,7 @@ static void spear_kbd_close(struct input_dev *dev) } #ifdef CONFIG_OF -static int __devinit spear_kbd_parse_dt(struct platform_device *pdev, +static int spear_kbd_parse_dt(struct platform_device *pdev, struct spear_kbd *kbd) { struct device_node *np = pdev->dev.of_node; @@ -181,7 +181,7 @@ static inline int spear_kbd_parse_dt(struct platform_device *pdev, } #endif -static int __devinit spear_kbd_probe(struct platform_device *pdev) +static int spear_kbd_probe(struct platform_device *pdev) { struct kbd_platform_data *pdata = dev_get_platdata(&pdev->dev); const struct matrix_keymap_data *keymap = pdata ? pdata->keymap : NULL; @@ -203,12 +203,16 @@ static int __devinit spear_kbd_probe(struct platform_device *pdev) return irq; } - kbd = kzalloc(sizeof(*kbd), GFP_KERNEL); - input_dev = input_allocate_device(); - if (!kbd || !input_dev) { - dev_err(&pdev->dev, "out of memory\n"); - error = -ENOMEM; - goto err_free_mem; + kbd = devm_kzalloc(&pdev->dev, sizeof(*kbd), GFP_KERNEL); + if (!kbd) { + dev_err(&pdev->dev, "not enough memory for driver data\n"); + return -ENOMEM; + } + + input_dev = devm_input_allocate_device(&pdev->dev); + if (!input_dev) { + dev_err(&pdev->dev, "unable to allocate input device\n"); + return -ENOMEM; } kbd->input = input_dev; @@ -217,37 +221,25 @@ static int __devinit spear_kbd_probe(struct platform_device *pdev) if (!pdata) { error = spear_kbd_parse_dt(pdev, kbd); if (error) - goto err_free_mem; + return error; } else { kbd->mode = pdata->mode; kbd->rep = pdata->rep; kbd->suspended_rate = pdata->suspended_rate; } - kbd->res = request_mem_region(res->start, resource_size(res), - pdev->name); - if (!kbd->res) { - dev_err(&pdev->dev, "keyboard region already claimed\n"); - error = -EBUSY; - goto err_free_mem; - } - - kbd->io_base = ioremap(res->start, resource_size(res)); + kbd->io_base = devm_request_and_ioremap(&pdev->dev, res); if (!kbd->io_base) { - dev_err(&pdev->dev, "ioremap failed for kbd_region\n"); - error = -ENOMEM; - goto err_release_mem_region; + dev_err(&pdev->dev, "request-ioremap failed for kbd_region\n"); + return -ENOMEM; } - kbd->clk = clk_get(&pdev->dev, NULL); - if (IS_ERR(kbd->clk)) { - error = PTR_ERR(kbd->clk); - goto err_iounmap; - } + kbd->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(kbd->clk)) + return PTR_ERR(kbd->clk); input_dev->name = "Spear Keyboard"; input_dev->phys = "keyboard/input0"; - input_dev->dev.parent = &pdev->dev; input_dev->id.bustype = BUS_HOST; input_dev->id.vendor = 0x0001; input_dev->id.product = 0x0001; @@ -259,7 +251,7 @@ static int __devinit spear_kbd_probe(struct platform_device *pdev) kbd->keycodes, input_dev); if (error) { dev_err(&pdev->dev, "Failed to build keymap\n"); - goto err_put_clk; + return error; } if (kbd->rep) @@ -268,48 +260,36 @@ static int __devinit spear_kbd_probe(struct platform_device *pdev) input_set_drvdata(input_dev, kbd); - error = request_irq(irq, spear_kbd_interrupt, 0, "keyboard", kbd); + error = devm_request_irq(&pdev->dev, irq, spear_kbd_interrupt, 0, + "keyboard", kbd); if (error) { - dev_err(&pdev->dev, "request_irq fail\n"); - goto err_put_clk; + dev_err(&pdev->dev, "request_irq failed\n"); + return error; } + error = clk_prepare(kbd->clk); + if (error) + return error; + error = input_register_device(input_dev); if (error) { dev_err(&pdev->dev, "Unable to register keyboard device\n"); - goto err_free_irq; + clk_unprepare(kbd->clk); + return error; } device_init_wakeup(&pdev->dev, 1); platform_set_drvdata(pdev, kbd); return 0; - -err_free_irq: - free_irq(kbd->irq, kbd); -err_put_clk: - clk_put(kbd->clk); -err_iounmap: - iounmap(kbd->io_base); -err_release_mem_region: - release_mem_region(res->start, resource_size(res)); -err_free_mem: - input_free_device(input_dev); - kfree(kbd); - - return error; } -static int __devexit spear_kbd_remove(struct platform_device *pdev) +static int spear_kbd_remove(struct platform_device *pdev) { struct spear_kbd *kbd = platform_get_drvdata(pdev); - free_irq(kbd->irq, kbd); input_unregister_device(kbd->input); - clk_put(kbd->clk); - iounmap(kbd->io_base); - release_mem_region(kbd->res->start, resource_size(kbd->res)); - kfree(kbd); + clk_unprepare(kbd->clk); device_init_wakeup(&pdev->dev, 0); platform_set_drvdata(pdev, NULL); @@ -333,7 +313,8 @@ static int spear_kbd_suspend(struct device *dev) mode_ctl_reg = readl_relaxed(kbd->io_base + MODE_CTL_REG); if (device_may_wakeup(&pdev->dev)) { - enable_irq_wake(kbd->irq); + if (!enable_irq_wake(kbd->irq)) + kbd->irq_wake_enabled = true; /* * reprogram the keyboard operating frequency as on some @@ -379,7 +360,10 @@ static int spear_kbd_resume(struct device *dev) mutex_lock(&input_dev->mutex); if (device_may_wakeup(&pdev->dev)) { - disable_irq_wake(kbd->irq); + if (kbd->irq_wake_enabled) { + kbd->irq_wake_enabled = false; + disable_irq_wake(kbd->irq); + } } else { if (input_dev->users) clk_enable(kbd->clk); @@ -407,7 +391,7 @@ MODULE_DEVICE_TABLE(of, spear_kbd_id_table); static struct platform_driver spear_kbd_driver = { .probe = spear_kbd_probe, - .remove = __devexit_p(spear_kbd_remove), + .remove = spear_kbd_remove, .driver = { .name = "keyboard", .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/stmpe-keypad.c b/drivers/input/keyboard/stmpe-keypad.c index 470a8778dec1..5cbec56f7720 100644 --- a/drivers/input/keyboard/stmpe-keypad.c +++ b/drivers/input/keyboard/stmpe-keypad.c @@ -166,7 +166,7 @@ static irqreturn_t stmpe_keypad_irq(int irq, void *dev) return IRQ_HANDLED; } -static int __devinit stmpe_keypad_altfunc_init(struct stmpe_keypad *keypad) +static int stmpe_keypad_altfunc_init(struct stmpe_keypad *keypad) { const struct stmpe_keypad_variant *variant = keypad->variant; unsigned int col_gpios = variant->col_gpios; @@ -207,7 +207,7 @@ static int __devinit stmpe_keypad_altfunc_init(struct stmpe_keypad *keypad) return stmpe_set_altfunc(stmpe, pins, STMPE_BLOCK_KEYPAD); } -static int __devinit stmpe_keypad_chip_init(struct stmpe_keypad *keypad) +static int stmpe_keypad_chip_init(struct stmpe_keypad *keypad) { const struct stmpe_keypad_platform_data *plat = keypad->plat; const struct stmpe_keypad_variant *variant = keypad->variant; @@ -257,105 +257,131 @@ static int __devinit stmpe_keypad_chip_init(struct stmpe_keypad *keypad) (plat->debounce_ms << 1)); } -static int __devinit stmpe_keypad_probe(struct platform_device *pdev) +static void stmpe_keypad_fill_used_pins(struct stmpe_keypad *keypad) { - struct stmpe *stmpe = dev_get_drvdata(pdev->dev.parent); + int row, col; + + for (row = 0; row < STMPE_KEYPAD_MAX_ROWS; row++) { + for (col = 0; col < STMPE_KEYPAD_MAX_COLS; col++) { + int code = MATRIX_SCAN_CODE(row, col, + STMPE_KEYPAD_ROW_SHIFT); + if (keypad->keymap[code] != KEY_RESERVED) { + keypad->rows |= 1 << row; + keypad->cols |= 1 << col; + } + } + } +} + +#ifdef CONFIG_OF +static const struct stmpe_keypad_platform_data * +stmpe_keypad_of_probe(struct device *dev) +{ + struct device_node *np = dev->of_node; struct stmpe_keypad_platform_data *plat; + + if (!np) + return ERR_PTR(-ENODEV); + + plat = devm_kzalloc(dev, sizeof(*plat), GFP_KERNEL); + if (!plat) + return ERR_PTR(-ENOMEM); + + of_property_read_u32(np, "debounce-interval", &plat->debounce_ms); + of_property_read_u32(np, "st,scan-count", &plat->scan_count); + + plat->no_autorepeat = of_property_read_bool(np, "st,no-autorepeat"); + + return plat; +} +#else +static inline const struct stmpe_keypad_platform_data * +stmpe_keypad_of_probe(struct device *dev) +{ + return ERR_PTR(-EINVAL); +} +#endif + +static int stmpe_keypad_probe(struct platform_device *pdev) +{ + struct stmpe *stmpe = dev_get_drvdata(pdev->dev.parent); + const struct stmpe_keypad_platform_data *plat; struct stmpe_keypad *keypad; struct input_dev *input; - int ret; + int error; int irq; - int i; plat = stmpe->pdata->keypad; - if (!plat) - return -ENODEV; + if (!plat) { + plat = stmpe_keypad_of_probe(&pdev->dev); + if (IS_ERR(plat)) + return PTR_ERR(plat); + } irq = platform_get_irq(pdev, 0); if (irq < 0) return irq; - keypad = kzalloc(sizeof(struct stmpe_keypad), GFP_KERNEL); + keypad = devm_kzalloc(&pdev->dev, sizeof(struct stmpe_keypad), + GFP_KERNEL); if (!keypad) return -ENOMEM; - input = input_allocate_device(); - if (!input) { - ret = -ENOMEM; - goto out_freekeypad; - } + input = devm_input_allocate_device(&pdev->dev); + if (!input) + return -ENOMEM; input->name = "STMPE keypad"; input->id.bustype = BUS_I2C; input->dev.parent = &pdev->dev; - ret = matrix_keypad_build_keymap(plat->keymap_data, NULL, - STMPE_KEYPAD_MAX_ROWS, - STMPE_KEYPAD_MAX_COLS, - keypad->keymap, input); - if (ret) - goto out_freeinput; + error = matrix_keypad_build_keymap(plat->keymap_data, NULL, + STMPE_KEYPAD_MAX_ROWS, + STMPE_KEYPAD_MAX_COLS, + keypad->keymap, input); + if (error) + return error; input_set_capability(input, EV_MSC, MSC_SCAN); if (!plat->no_autorepeat) __set_bit(EV_REP, input->evbit); - for (i = 0; i < plat->keymap_data->keymap_size; i++) { - unsigned int key = plat->keymap_data->keymap[i]; - - keypad->cols |= 1 << KEY_COL(key); - keypad->rows |= 1 << KEY_ROW(key); - } + stmpe_keypad_fill_used_pins(keypad); keypad->stmpe = stmpe; keypad->plat = plat; keypad->input = input; keypad->variant = &stmpe_keypad_variants[stmpe->partnum]; - ret = stmpe_keypad_chip_init(keypad); - if (ret < 0) - goto out_freeinput; + error = stmpe_keypad_chip_init(keypad); + if (error < 0) + return error; - ret = input_register_device(input); - if (ret) { - dev_err(&pdev->dev, - "unable to register input device: %d\n", ret); - goto out_freeinput; + error = devm_request_threaded_irq(&pdev->dev, irq, + NULL, stmpe_keypad_irq, + IRQF_ONESHOT, "stmpe-keypad", keypad); + if (error) { + dev_err(&pdev->dev, "unable to get irq: %d\n", error); + return error; } - ret = request_threaded_irq(irq, NULL, stmpe_keypad_irq, IRQF_ONESHOT, - "stmpe-keypad", keypad); - if (ret) { - dev_err(&pdev->dev, "unable to get irq: %d\n", ret); - goto out_unregisterinput; + error = input_register_device(input); + if (error) { + dev_err(&pdev->dev, + "unable to register input device: %d\n", error); + return error; } platform_set_drvdata(pdev, keypad); return 0; - -out_unregisterinput: - input_unregister_device(input); - input = NULL; -out_freeinput: - input_free_device(input); -out_freekeypad: - kfree(keypad); - return ret; } -static int __devexit stmpe_keypad_remove(struct platform_device *pdev) +static int stmpe_keypad_remove(struct platform_device *pdev) { struct stmpe_keypad *keypad = platform_get_drvdata(pdev); - struct stmpe *stmpe = keypad->stmpe; - int irq = platform_get_irq(pdev, 0); - - stmpe_disable(stmpe, STMPE_BLOCK_KEYPAD); - free_irq(irq, keypad); - input_unregister_device(keypad->input); - platform_set_drvdata(pdev, NULL); - kfree(keypad); + stmpe_disable(keypad->stmpe, STMPE_BLOCK_KEYPAD); return 0; } @@ -364,7 +390,7 @@ static struct platform_driver stmpe_keypad_driver = { .driver.name = "stmpe-keypad", .driver.owner = THIS_MODULE, .probe = stmpe_keypad_probe, - .remove = __devexit_p(stmpe_keypad_remove), + .remove = stmpe_keypad_remove, }; module_platform_driver(stmpe_keypad_driver); diff --git a/drivers/input/keyboard/tc3589x-keypad.c b/drivers/input/keyboard/tc3589x-keypad.c index 7d498e698508..2fb0d76a04c4 100644 --- a/drivers/input/keyboard/tc3589x-keypad.c +++ b/drivers/input/keyboard/tc3589x-keypad.c @@ -299,7 +299,7 @@ static void tc3589x_keypad_close(struct input_dev *input) tc3589x_keypad_disable(keypad); } -static int __devinit tc3589x_keypad_probe(struct platform_device *pdev) +static int tc3589x_keypad_probe(struct platform_device *pdev) { struct tc3589x *tc3589x = dev_get_drvdata(pdev->dev.parent); struct tc_keypad *keypad; @@ -382,7 +382,7 @@ err_free_mem: return error; } -static int __devexit tc3589x_keypad_remove(struct platform_device *pdev) +static int tc3589x_keypad_remove(struct platform_device *pdev) { struct tc_keypad *keypad = platform_get_drvdata(pdev); int irq = platform_get_irq(pdev, 0); @@ -448,7 +448,7 @@ static struct platform_driver tc3589x_keypad_driver = { .pm = &tc3589x_keypad_dev_pm_ops, }, .probe = tc3589x_keypad_probe, - .remove = __devexit_p(tc3589x_keypad_remove), + .remove = tc3589x_keypad_remove, }; module_platform_driver(tc3589x_keypad_driver); diff --git a/drivers/input/keyboard/tca6416-keypad.c b/drivers/input/keyboard/tca6416-keypad.c index c355cdde8d22..bfc832c35a7c 100644 --- a/drivers/input/keyboard/tca6416-keypad.c +++ b/drivers/input/keyboard/tca6416-keypad.c @@ -166,7 +166,7 @@ static void tca6416_keys_close(struct input_dev *dev) disable_irq(chip->irqnum); } -static int __devinit tca6416_setup_registers(struct tca6416_keypad_chip *chip) +static int tca6416_setup_registers(struct tca6416_keypad_chip *chip) { int error; @@ -197,7 +197,7 @@ static int __devinit tca6416_setup_registers(struct tca6416_keypad_chip *chip) return 0; } -static int __devinit tca6416_keypad_probe(struct i2c_client *client, +static int tca6416_keypad_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct tca6416_keys_platform_data *pdata; @@ -313,7 +313,7 @@ fail1: return error; } -static int __devexit tca6416_keypad_remove(struct i2c_client *client) +static int tca6416_keypad_remove(struct i2c_client *client) { struct tca6416_keypad_chip *chip = i2c_get_clientdata(client); @@ -361,7 +361,7 @@ static struct i2c_driver tca6416_keypad_driver = { .pm = &tca6416_keypad_dev_pm_ops, }, .probe = tca6416_keypad_probe, - .remove = __devexit_p(tca6416_keypad_remove), + .remove = tca6416_keypad_remove, .id_table = tca6416_id, }; diff --git a/drivers/input/keyboard/tca8418_keypad.c b/drivers/input/keyboard/tca8418_keypad.c index 893869b29ed9..50e9c5e195e1 100644 --- a/drivers/input/keyboard/tca8418_keypad.c +++ b/drivers/input/keyboard/tca8418_keypad.c @@ -35,6 +35,7 @@ #include <linux/i2c.h> #include <linux/input.h> #include <linux/input/tca8418_keypad.h> +#include <linux/of.h> /* TCA8418 hardware limits */ #define TCA8418_MAX_ROWS 8 @@ -109,25 +110,11 @@ #define KEY_EVENT_CODE 0x7f #define KEY_EVENT_VALUE 0x80 - -static const struct i2c_device_id tca8418_id[] = { - { TCA8418_NAME, 8418, }, - { } -}; -MODULE_DEVICE_TABLE(i2c, tca8418_id); - struct tca8418_keypad { - unsigned int rows; - unsigned int cols; - unsigned int keypad_mask; /* Mask for keypad col/rol regs */ - unsigned int irq; - unsigned int row_shift; - struct i2c_client *client; struct input_dev *input; - /* Flexible array member, must be at end of struct */ - unsigned short keymap[]; + unsigned int row_shift; }; /* @@ -172,6 +159,8 @@ static int tca8418_read_byte(struct tca8418_keypad *keypad_data, static void tca8418_read_keypad(struct tca8418_keypad *keypad_data) { + struct input_dev *input = keypad_data->input; + unsigned short *keymap = input->keycode; int error, col, row; u8 reg, state, code; @@ -190,9 +179,8 @@ static void tca8418_read_keypad(struct tca8418_keypad *keypad_data) col = (col) ? col - 1 : TCA8418_MAX_COLS - 1; code = MATRIX_SCAN_CODE(row, col, keypad_data->row_shift); - input_event(keypad_data->input, EV_MSC, MSC_SCAN, code); - input_report_key(keypad_data->input, - keypad_data->keymap[code], state); + input_event(input, EV_MSC, MSC_SCAN, code); + input_report_key(input, keymap[code], state); /* Read for next loop */ error = tca8418_read_byte(keypad_data, REG_KEY_EVENT_A, ®); @@ -202,7 +190,7 @@ static void tca8418_read_keypad(struct tca8418_keypad *keypad_data) dev_err(&keypad_data->client->dev, "unable to read REG_KEY_EVENT_A\n"); - input_sync(keypad_data->input); + input_sync(input); } /* @@ -218,16 +206,18 @@ static irqreturn_t tca8418_irq_handler(int irq, void *dev_id) if (error) { dev_err(&keypad_data->client->dev, "unable to read REG_INT_STAT\n"); - goto exit; + return IRQ_NONE; } + if (!reg) + return IRQ_NONE; + if (reg & INT_STAT_OVR_FLOW_INT) dev_warn(&keypad_data->client->dev, "overflow occurred\n"); if (reg & INT_STAT_K_INT) tca8418_read_keypad(keypad_data); -exit: /* Clear all interrupts, even IRQs we didn't check (GPI, CAD, LCK) */ reg = 0xff; error = tca8418_write_byte(keypad_data, REG_INT_STAT, reg); @@ -241,7 +231,8 @@ exit: /* * Configure the TCA8418 for keypad operation */ -static int __devinit tca8418_configure(struct tca8418_keypad *keypad_data) +static int tca8418_configure(struct tca8418_keypad *keypad_data, + u32 rows, u32 cols) { int reg, error; @@ -253,9 +244,8 @@ static int __devinit tca8418_configure(struct tca8418_keypad *keypad_data) /* Assemble a mask for row and column registers */ - reg = ~(~0 << keypad_data->rows); - reg += (~(~0 << keypad_data->cols)) << 8; - keypad_data->keypad_mask = reg; + reg = ~(~0 << rows); + reg += (~(~0 << cols)) << 8; /* Set registers to keypad mode */ error |= tca8418_write_byte(keypad_data, REG_KP_GPIO1, reg); @@ -270,145 +260,144 @@ static int __devinit tca8418_configure(struct tca8418_keypad *keypad_data) return error; } -static int __devinit tca8418_keypad_probe(struct i2c_client *client, +static int tca8418_keypad_probe(struct i2c_client *client, const struct i2c_device_id *id) { + struct device *dev = &client->dev; const struct tca8418_keypad_platform_data *pdata = - client->dev.platform_data; + dev_get_platdata(dev); struct tca8418_keypad *keypad_data; struct input_dev *input; + const struct matrix_keymap_data *keymap_data = NULL; + u32 rows = 0, cols = 0; + bool rep = false; + bool irq_is_gpio = false; + int irq; int error, row_shift, max_keys; /* Copy the platform data */ - if (!pdata) { - dev_dbg(&client->dev, "no platform data\n"); - return -EINVAL; - } - - if (!pdata->keymap_data) { - dev_err(&client->dev, "no keymap data defined\n"); - return -EINVAL; + if (pdata) { + if (!pdata->keymap_data) { + dev_err(dev, "no keymap data defined\n"); + return -EINVAL; + } + keymap_data = pdata->keymap_data; + rows = pdata->rows; + cols = pdata->cols; + rep = pdata->rep; + irq_is_gpio = pdata->irq_is_gpio; + } else { + struct device_node *np = dev->of_node; + of_property_read_u32(np, "keypad,num-rows", &rows); + of_property_read_u32(np, "keypad,num-columns", &cols); + rep = of_property_read_bool(np, "keypad,autorepeat"); } - if (!pdata->rows || pdata->rows > TCA8418_MAX_ROWS) { - dev_err(&client->dev, "invalid rows\n"); + if (!rows || rows > TCA8418_MAX_ROWS) { + dev_err(dev, "invalid rows\n"); return -EINVAL; } - if (!pdata->cols || pdata->cols > TCA8418_MAX_COLS) { - dev_err(&client->dev, "invalid columns\n"); + if (!cols || cols > TCA8418_MAX_COLS) { + dev_err(dev, "invalid columns\n"); return -EINVAL; } /* Check i2c driver capabilities */ if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_BYTE)) { - dev_err(&client->dev, "%s adapter not supported\n", + dev_err(dev, "%s adapter not supported\n", dev_driver_string(&client->adapter->dev)); return -ENODEV; } - row_shift = get_count_order(pdata->cols); - max_keys = pdata->rows << row_shift; + row_shift = get_count_order(cols); + max_keys = rows << row_shift; - /* Allocate memory for keypad_data, keymap and input device */ - keypad_data = kzalloc(sizeof(*keypad_data) + - max_keys * sizeof(keypad_data->keymap[0]), GFP_KERNEL); + /* Allocate memory for keypad_data and input device */ + keypad_data = devm_kzalloc(dev, sizeof(*keypad_data), GFP_KERNEL); if (!keypad_data) return -ENOMEM; - keypad_data->rows = pdata->rows; - keypad_data->cols = pdata->cols; keypad_data->client = client; keypad_data->row_shift = row_shift; /* Initialize the chip or fail if chip isn't present */ - error = tca8418_configure(keypad_data); + error = tca8418_configure(keypad_data, rows, cols); if (error < 0) - goto fail1; + return error; /* Configure input device */ - input = input_allocate_device(); - if (!input) { - error = -ENOMEM; - goto fail1; - } + input = devm_input_allocate_device(dev); + if (!input) + return -ENOMEM; + keypad_data->input = input; input->name = client->name; - input->dev.parent = &client->dev; - input->id.bustype = BUS_I2C; input->id.vendor = 0x0001; input->id.product = 0x001; input->id.version = 0x0001; - error = matrix_keypad_build_keymap(pdata->keymap_data, NULL, - pdata->rows, pdata->cols, - keypad_data->keymap, input); + error = matrix_keypad_build_keymap(keymap_data, NULL, rows, cols, + NULL, input); if (error) { - dev_dbg(&client->dev, "Failed to build keymap\n"); - goto fail2; + dev_err(dev, "Failed to build keymap\n"); + return error; } - if (pdata->rep) + if (rep) __set_bit(EV_REP, input->evbit); input_set_capability(input, EV_MSC, MSC_SCAN); input_set_drvdata(input, keypad_data); - if (pdata->irq_is_gpio) - client->irq = gpio_to_irq(client->irq); + irq = client->irq; + if (irq_is_gpio) + irq = gpio_to_irq(irq); - error = request_threaded_irq(client->irq, NULL, tca8418_irq_handler, - IRQF_TRIGGER_FALLING | IRQF_ONESHOT, - client->name, keypad_data); + error = devm_request_threaded_irq(dev, irq, NULL, tca8418_irq_handler, + IRQF_TRIGGER_FALLING | + IRQF_SHARED | + IRQF_ONESHOT, + client->name, keypad_data); if (error) { - dev_dbg(&client->dev, - "Unable to claim irq %d; error %d\n", + dev_err(dev, "Unable to claim irq %d; error %d\n", client->irq, error); - goto fail2; + return error; } error = input_register_device(input); if (error) { - dev_dbg(&client->dev, - "Unable to register input device, error: %d\n", error); - goto fail3; + dev_err(dev, "Unable to register input device, error: %d\n", + error); + return error; } - i2c_set_clientdata(client, keypad_data); return 0; - -fail3: - free_irq(client->irq, keypad_data); -fail2: - input_free_device(input); -fail1: - kfree(keypad_data); - return error; } -static int __devexit tca8418_keypad_remove(struct i2c_client *client) -{ - struct tca8418_keypad *keypad_data = i2c_get_clientdata(client); - - free_irq(keypad_data->client->irq, keypad_data); - - input_unregister_device(keypad_data->input); - - kfree(keypad_data); - - return 0; -} +static const struct i2c_device_id tca8418_id[] = { + { TCA8418_NAME, 8418, }, + { } +}; +MODULE_DEVICE_TABLE(i2c, tca8418_id); +#ifdef CONFIG_OF +static const struct of_device_id tca8418_dt_ids[] __devinitconst = { + { .compatible = "ti,tca8418", }, + { } +}; +MODULE_DEVICE_TABLE(of, tca8418_dt_ids); +#endif static struct i2c_driver tca8418_keypad_driver = { .driver = { .name = TCA8418_NAME, .owner = THIS_MODULE, + .of_match_table = of_match_ptr(tca8418_dt_ids), }, .probe = tca8418_keypad_probe, - .remove = __devexit_p(tca8418_keypad_remove), .id_table = tca8418_id, }; diff --git a/drivers/input/keyboard/tegra-kbc.c b/drivers/input/keyboard/tegra-kbc.c index 5faaf2553e33..c76f96872d31 100644 --- a/drivers/input/keyboard/tegra-kbc.c +++ b/drivers/input/keyboard/tegra-kbc.c @@ -87,7 +87,7 @@ struct tegra_kbc { struct clk *clk; }; -static const u32 tegra_kbc_default_keymap[] __devinitdata = { +static const u32 tegra_kbc_default_keymap[] = { KEY(0, 2, KEY_W), KEY(0, 3, KEY_S), KEY(0, 4, KEY_A), @@ -223,7 +223,7 @@ static const u32 tegra_kbc_default_keymap[] __devinitdata = { }; static const -struct matrix_keymap_data tegra_kbc_default_keymap_data __devinitdata = { +struct matrix_keymap_data tegra_kbc_default_keymap_data = { .keymap = tegra_kbc_default_keymap, .keymap_size = ARRAY_SIZE(tegra_kbc_default_keymap), }; @@ -573,7 +573,7 @@ static void tegra_kbc_close(struct input_dev *dev) return tegra_kbc_stop(kbc); } -static bool __devinit +static bool tegra_kbc_check_pin_cfg(const struct tegra_kbc_platform_data *pdata, struct device *dev, unsigned int *num_rows) { @@ -619,7 +619,7 @@ tegra_kbc_check_pin_cfg(const struct tegra_kbc_platform_data *pdata, } #ifdef CONFIG_OF -static struct tegra_kbc_platform_data * __devinit tegra_kbc_dt_parse_pdata( +static struct tegra_kbc_platform_data *tegra_kbc_dt_parse_pdata( struct platform_device *pdev) { struct tegra_kbc_platform_data *pdata; @@ -670,7 +670,7 @@ static inline struct tegra_kbc_platform_data *tegra_kbc_dt_parse_pdata( } #endif -static int __devinit tegra_kbd_setup_keymap(struct tegra_kbc *kbc) +static int tegra_kbd_setup_keymap(struct tegra_kbc *kbc) { const struct tegra_kbc_platform_data *pdata = kbc->pdata; const struct matrix_keymap_data *keymap_data = pdata->keymap_data; @@ -697,7 +697,7 @@ static int __devinit tegra_kbd_setup_keymap(struct tegra_kbc *kbc) return retval; } -static int __devinit tegra_kbc_probe(struct platform_device *pdev) +static int tegra_kbc_probe(struct platform_device *pdev) { const struct tegra_kbc_platform_data *pdata = pdev->dev.platform_data; struct tegra_kbc *kbc; @@ -838,7 +838,7 @@ err_free_pdata: return err; } -static int __devexit tegra_kbc_remove(struct platform_device *pdev) +static int tegra_kbc_remove(struct platform_device *pdev) { struct tegra_kbc *kbc = platform_get_drvdata(pdev); struct resource *res; @@ -954,7 +954,7 @@ MODULE_DEVICE_TABLE(of, tegra_kbc_of_match); static struct platform_driver tegra_kbc_driver = { .probe = tegra_kbc_probe, - .remove = __devexit_p(tegra_kbc_remove), + .remove = tegra_kbc_remove, .driver = { .name = "tegra-kbc", .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/tnetv107x-keypad.c b/drivers/input/keyboard/tnetv107x-keypad.c index 4c34f21fbe2d..ee1635011292 100644 --- a/drivers/input/keyboard/tnetv107x-keypad.c +++ b/drivers/input/keyboard/tnetv107x-keypad.c @@ -153,7 +153,7 @@ static void keypad_stop(struct input_dev *dev) clk_disable(kp->clk); } -static int __devinit keypad_probe(struct platform_device *pdev) +static int keypad_probe(struct platform_device *pdev) { const struct matrix_keypad_platform_data *pdata; const struct matrix_keymap_data *keymap_data; @@ -301,7 +301,7 @@ error_res: return error; } -static int __devexit keypad_remove(struct platform_device *pdev) +static int keypad_remove(struct platform_device *pdev) { struct keypad_data *kp = platform_get_drvdata(pdev); @@ -319,7 +319,7 @@ static int __devexit keypad_remove(struct platform_device *pdev) static struct platform_driver keypad_driver = { .probe = keypad_probe, - .remove = __devexit_p(keypad_remove), + .remove = keypad_remove, .driver.name = "tnetv107x-keypad", .driver.owner = THIS_MODULE, }; diff --git a/drivers/input/keyboard/twl4030_keypad.c b/drivers/input/keyboard/twl4030_keypad.c index a2c6f79aa101..04f84fd57173 100644 --- a/drivers/input/keyboard/twl4030_keypad.c +++ b/drivers/input/keyboard/twl4030_keypad.c @@ -271,7 +271,7 @@ static irqreturn_t do_kp_irq(int irq, void *_kp) return IRQ_HANDLED; } -static int __devinit twl4030_kp_program(struct twl4030_keypad *kp) +static int twl4030_kp_program(struct twl4030_keypad *kp) { u8 reg; int i; @@ -328,7 +328,7 @@ static int __devinit twl4030_kp_program(struct twl4030_keypad *kp) * Registers keypad device with input subsystem * and configures TWL4030 keypad registers */ -static int __devinit twl4030_kp_probe(struct platform_device *pdev) +static int twl4030_kp_probe(struct platform_device *pdev) { struct twl4030_keypad_data *pdata = pdev->dev.platform_data; const struct matrix_keymap_data *keymap_data; @@ -432,7 +432,7 @@ err1: return error; } -static int __devexit twl4030_kp_remove(struct platform_device *pdev) +static int twl4030_kp_remove(struct platform_device *pdev) { struct twl4030_keypad *kp = platform_get_drvdata(pdev); @@ -452,7 +452,7 @@ static int __devexit twl4030_kp_remove(struct platform_device *pdev) static struct platform_driver twl4030_kp_driver = { .probe = twl4030_kp_probe, - .remove = __devexit_p(twl4030_kp_remove), + .remove = twl4030_kp_remove, .driver = { .name = "twl4030_keypad", .owner = THIS_MODULE, diff --git a/drivers/input/keyboard/w90p910_keypad.c b/drivers/input/keyboard/w90p910_keypad.c index e0f6cd1ad0fd..ee163bee8cce 100644 --- a/drivers/input/keyboard/w90p910_keypad.c +++ b/drivers/input/keyboard/w90p910_keypad.c @@ -118,7 +118,7 @@ static void w90p910_keypad_close(struct input_dev *dev) clk_disable(keypad->clk); } -static int __devinit w90p910_keypad_probe(struct platform_device *pdev) +static int w90p910_keypad_probe(struct platform_device *pdev) { const struct w90p910_keypad_platform_data *pdata = pdev->dev.platform_data; @@ -234,7 +234,7 @@ failed_free: return error; } -static int __devexit w90p910_keypad_remove(struct platform_device *pdev) +static int w90p910_keypad_remove(struct platform_device *pdev) { struct w90p910_keypad *keypad = platform_get_drvdata(pdev); struct resource *res; @@ -257,7 +257,7 @@ static int __devexit w90p910_keypad_remove(struct platform_device *pdev) static struct platform_driver w90p910_keypad_driver = { .probe = w90p910_keypad_probe, - .remove = __devexit_p(w90p910_keypad_remove), + .remove = w90p910_keypad_remove, .driver = { .name = "nuc900-kpi", .owner = THIS_MODULE, diff --git a/drivers/input/matrix-keymap.c b/drivers/input/matrix-keymap.c index d88d9be1d1b7..3ae496ea5fe6 100644 --- a/drivers/input/matrix-keymap.c +++ b/drivers/input/matrix-keymap.c @@ -18,6 +18,7 @@ */ #include <linux/device.h> +#include <linux/gfp.h> #include <linux/kernel.h> #include <linux/types.h> #include <linux/input.h> @@ -123,6 +124,11 @@ static int matrix_keypad_parse_of_keymap(const char *propname, * it will attempt load the keymap from property specified by @keymap_name * argument (or "linux,keymap" if @keymap_name is %NULL). * + * If @keymap is %NULL the function will automatically allocate managed + * block of memory to store the keymap. This memory will be associated with + * the parent device and automatically freed when device unbinds from the + * driver. + * * Callers are expected to set up input_dev->dev.parent before calling this * function. */ @@ -133,12 +139,27 @@ int matrix_keypad_build_keymap(const struct matrix_keymap_data *keymap_data, struct input_dev *input_dev) { unsigned int row_shift = get_count_order(cols); + size_t max_keys = rows << row_shift; int i; int error; + if (WARN_ON(!input_dev->dev.parent)) + return -EINVAL; + + if (!keymap) { + keymap = devm_kzalloc(input_dev->dev.parent, + max_keys * sizeof(*keymap), + GFP_KERNEL); + if (!keymap) { + dev_err(input_dev->dev.parent, + "Unable to allocate memory for keymap"); + return -ENOMEM; + } + } + input_dev->keycode = keymap; input_dev->keycodesize = sizeof(*keymap); - input_dev->keycodemax = rows << row_shift; + input_dev->keycodemax = max_keys; __set_bit(EV_KEY, input_dev->evbit); diff --git a/drivers/input/misc/88pm80x_onkey.c b/drivers/input/misc/88pm80x_onkey.c index 7f26e7b6c228..ee43e5b7c881 100644 --- a/drivers/input/misc/88pm80x_onkey.c +++ b/drivers/input/misc/88pm80x_onkey.c @@ -62,7 +62,7 @@ static irqreturn_t pm80x_onkey_handler(int irq, void *data) static SIMPLE_DEV_PM_OPS(pm80x_onkey_pm_ops, pm80x_dev_suspend, pm80x_dev_resume); -static int __devinit pm80x_onkey_probe(struct platform_device *pdev) +static int pm80x_onkey_probe(struct platform_device *pdev) { struct pm80x_chip *chip = dev_get_drvdata(pdev->dev.parent); @@ -139,7 +139,7 @@ out: return err; } -static int __devexit pm80x_onkey_remove(struct platform_device *pdev) +static int pm80x_onkey_remove(struct platform_device *pdev) { struct pm80x_onkey_info *info = platform_get_drvdata(pdev); @@ -157,7 +157,7 @@ static struct platform_driver pm80x_onkey_driver = { .pm = &pm80x_onkey_pm_ops, }, .probe = pm80x_onkey_probe, - .remove = __devexit_p(pm80x_onkey_remove), + .remove = pm80x_onkey_remove, }; module_platform_driver(pm80x_onkey_driver); diff --git a/drivers/input/misc/88pm860x_onkey.c b/drivers/input/misc/88pm860x_onkey.c index f9ce1835e4d7..abd8453e5212 100644 --- a/drivers/input/misc/88pm860x_onkey.c +++ b/drivers/input/misc/88pm860x_onkey.c @@ -56,7 +56,7 @@ static irqreturn_t pm860x_onkey_handler(int irq, void *data) return IRQ_HANDLED; } -static int __devinit pm860x_onkey_probe(struct platform_device *pdev) +static int pm860x_onkey_probe(struct platform_device *pdev) { struct pm860x_chip *chip = dev_get_drvdata(pdev->dev.parent); struct pm860x_onkey_info *info; @@ -121,7 +121,7 @@ out: return ret; } -static int __devexit pm860x_onkey_remove(struct platform_device *pdev) +static int pm860x_onkey_remove(struct platform_device *pdev) { struct pm860x_onkey_info *info = platform_get_drvdata(pdev); @@ -161,7 +161,7 @@ static struct platform_driver pm860x_onkey_driver = { .pm = &pm860x_onkey_pm_ops, }, .probe = pm860x_onkey_probe, - .remove = __devexit_p(pm860x_onkey_remove), + .remove = pm860x_onkey_remove, }; module_platform_driver(pm860x_onkey_driver); diff --git a/drivers/input/misc/Kconfig b/drivers/input/misc/Kconfig index 104a7c3153c0..259ef31abb18 100644 --- a/drivers/input/misc/Kconfig +++ b/drivers/input/misc/Kconfig @@ -300,8 +300,7 @@ config INPUT_ATI_REMOTE2 called ati_remote2. config INPUT_KEYSPAN_REMOTE - tristate "Keyspan DMR USB remote control (EXPERIMENTAL)" - depends on EXPERIMENTAL + tristate "Keyspan DMR USB remote control" depends on USB_ARCH_HAS_HCD select USB help @@ -350,7 +349,6 @@ config INPUT_POWERMATE config INPUT_YEALINK tristate "Yealink usb-p1k voip phone" - depends on EXPERIMENTAL depends on USB_ARCH_HAS_HCD select USB help @@ -366,7 +364,6 @@ config INPUT_YEALINK config INPUT_CM109 tristate "C-Media CM109 USB I/O Controller" - depends on EXPERIMENTAL depends on USB_ARCH_HAS_HCD select USB help @@ -377,6 +374,16 @@ config INPUT_CM109 To compile this driver as a module, choose M here: the module will be called cm109. +config INPUT_RETU_PWRBUTTON + tristate "Retu Power button Driver" + depends on MFD_RETU + help + Say Y here if you want to enable power key reporting via the + Retu chips found in Nokia Internet Tablets (770, N800, N810). + + To compile this driver as a module, choose M here. The module will + be called retu-pwrbutton. + config INPUT_TWL4030_PWRBUTTON tristate "TWL4030 Power button Driver" depends on TWL4030_CORE @@ -444,7 +451,7 @@ config INPUT_PCF50633_PMU config INPUT_PCF8574 tristate "PCF8574 Keypad input device" - depends on I2C && EXPERIMENTAL + depends on I2C help Say Y here if you want to support a keypad connected via I2C with a PCF8574. @@ -454,7 +461,7 @@ config INPUT_PCF8574 config INPUT_PWM_BEEPER tristate "PWM beeper support" - depends on HAVE_PWM + depends on HAVE_PWM || PWM help Say Y here to get support for PWM based beeper devices. @@ -496,6 +503,16 @@ config INPUT_DA9052_ONKEY To compile this driver as a module, choose M here: the module will be called da9052_onkey. +config INPUT_DA9055_ONKEY + tristate "Dialog Semiconductor DA9055 ONKEY" + depends on MFD_DA9055 + help + Support the ONKEY of DA9055 PMICs as an input device + reporting power button status. + + To compile this driver as a module, choose M here: the module + will be called da9055_onkey. + config INPUT_DM355EVM tristate "TI DaVinci DM355 EVM Keypad and IR Remote" depends on MFD_DM355EVM_MSP diff --git a/drivers/input/misc/Makefile b/drivers/input/misc/Makefile index 5ea769eda999..1f1e1b109d9d 100644 --- a/drivers/input/misc/Makefile +++ b/drivers/input/misc/Makefile @@ -24,6 +24,7 @@ obj-$(CONFIG_INPUT_CMA3000) += cma3000_d0x.o obj-$(CONFIG_INPUT_CMA3000_I2C) += cma3000_d0x_i2c.o obj-$(CONFIG_INPUT_COBALT_BTNS) += cobalt_btns.o obj-$(CONFIG_INPUT_DA9052_ONKEY) += da9052_onkey.o +obj-$(CONFIG_INPUT_DA9055_ONKEY) += da9055_onkey.o obj-$(CONFIG_INPUT_DM355EVM) += dm355evm_keys.o obj-$(CONFIG_INPUT_GP2A) += gp2ap002a00f.o obj-$(CONFIG_INPUT_GPIO_TILT_POLLED) += gpio_tilt_polled.o @@ -46,6 +47,7 @@ obj-$(CONFIG_INPUT_PMIC8XXX_PWRKEY) += pmic8xxx-pwrkey.o obj-$(CONFIG_INPUT_POWERMATE) += powermate.o obj-$(CONFIG_INPUT_PWM_BEEPER) += pwm-beeper.o obj-$(CONFIG_INPUT_RB532_BUTTON) += rb532_button.o +obj-$(CONFIG_INPUT_RETU_PWRBUTTON) += retu-pwrbutton.o obj-$(CONFIG_INPUT_GPIO_ROTARY_ENCODER) += rotary_encoder.o obj-$(CONFIG_INPUT_SGI_BTNS) += sgi_btns.o obj-$(CONFIG_INPUT_SPARCSPKR) += sparcspkr.o diff --git a/drivers/input/misc/ab8500-ponkey.c b/drivers/input/misc/ab8500-ponkey.c index 84ec691c05aa..2f090b46e716 100644 --- a/drivers/input/misc/ab8500-ponkey.c +++ b/drivers/input/misc/ab8500-ponkey.c @@ -45,7 +45,7 @@ static irqreturn_t ab8500_ponkey_handler(int irq, void *data) return IRQ_HANDLED; } -static int __devinit ab8500_ponkey_probe(struct platform_device *pdev) +static int ab8500_ponkey_probe(struct platform_device *pdev) { struct ab8500 *ab8500 = dev_get_drvdata(pdev->dev.parent); struct ab8500_ponkey *ponkey; @@ -118,7 +118,7 @@ err_free_mem: return error; } -static int __devexit ab8500_ponkey_remove(struct platform_device *pdev) +static int ab8500_ponkey_remove(struct platform_device *pdev) { struct ab8500_ponkey *ponkey = platform_get_drvdata(pdev); @@ -146,7 +146,7 @@ static struct platform_driver ab8500_ponkey_driver = { .of_match_table = of_match_ptr(ab8500_ponkey_match), }, .probe = ab8500_ponkey_probe, - .remove = __devexit_p(ab8500_ponkey_remove), + .remove = ab8500_ponkey_remove, }; module_platform_driver(ab8500_ponkey_driver); diff --git a/drivers/input/misc/ad714x-i2c.c b/drivers/input/misc/ad714x-i2c.c index c8a79015472a..29d2064c26f2 100644 --- a/drivers/input/misc/ad714x-i2c.c +++ b/drivers/input/misc/ad714x-i2c.c @@ -72,7 +72,7 @@ static int ad714x_i2c_read(struct ad714x_chip *chip, return 0; } -static int __devinit ad714x_i2c_probe(struct i2c_client *client, +static int ad714x_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct ad714x_chip *chip; @@ -87,7 +87,7 @@ static int __devinit ad714x_i2c_probe(struct i2c_client *client, return 0; } -static int __devexit ad714x_i2c_remove(struct i2c_client *client) +static int ad714x_i2c_remove(struct i2c_client *client) { struct ad714x_chip *chip = i2c_get_clientdata(client); @@ -112,7 +112,7 @@ static struct i2c_driver ad714x_i2c_driver = { .pm = &ad714x_i2c_pm, }, .probe = ad714x_i2c_probe, - .remove = __devexit_p(ad714x_i2c_remove), + .remove = ad714x_i2c_remove, .id_table = ad714x_id, }; diff --git a/drivers/input/misc/ad714x-spi.c b/drivers/input/misc/ad714x-spi.c index 75f6136d608e..bdccca42d138 100644 --- a/drivers/input/misc/ad714x-spi.c +++ b/drivers/input/misc/ad714x-spi.c @@ -83,7 +83,7 @@ static int ad714x_spi_write(struct ad714x_chip *chip, return 0; } -static int __devinit ad714x_spi_probe(struct spi_device *spi) +static int ad714x_spi_probe(struct spi_device *spi) { struct ad714x_chip *chip; int err; @@ -103,7 +103,7 @@ static int __devinit ad714x_spi_probe(struct spi_device *spi) return 0; } -static int __devexit ad714x_spi_remove(struct spi_device *spi) +static int ad714x_spi_remove(struct spi_device *spi) { struct ad714x_chip *chip = spi_get_drvdata(spi); @@ -120,7 +120,7 @@ static struct spi_driver ad714x_spi_driver = { .pm = &ad714x_spi_pm, }, .probe = ad714x_spi_probe, - .remove = __devexit_p(ad714x_spi_remove), + .remove = ad714x_spi_remove, }; module_spi_driver(ad714x_spi_driver); diff --git a/drivers/input/misc/adxl34x-i2c.c b/drivers/input/misc/adxl34x-i2c.c index dd1d1c145a7f..535dda48cace 100644 --- a/drivers/input/misc/adxl34x-i2c.c +++ b/drivers/input/misc/adxl34x-i2c.c @@ -73,7 +73,7 @@ static const struct adxl34x_bus_ops adxl34x_i2c_bops = { .read_block = adxl34x_i2c_read_block, }; -static int __devinit adxl34x_i2c_probe(struct i2c_client *client, +static int adxl34x_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct adxl34x *ac; @@ -98,7 +98,7 @@ static int __devinit adxl34x_i2c_probe(struct i2c_client *client, return 0; } -static int __devexit adxl34x_i2c_remove(struct i2c_client *client) +static int adxl34x_i2c_remove(struct i2c_client *client) { struct adxl34x *ac = i2c_get_clientdata(client); @@ -144,7 +144,7 @@ static struct i2c_driver adxl34x_driver = { .pm = &adxl34x_i2c_pm, }, .probe = adxl34x_i2c_probe, - .remove = __devexit_p(adxl34x_i2c_remove), + .remove = adxl34x_i2c_remove, .id_table = adxl34x_id, }; diff --git a/drivers/input/misc/adxl34x-spi.c b/drivers/input/misc/adxl34x-spi.c index 820a802a1e6e..ad5f40d37e48 100644 --- a/drivers/input/misc/adxl34x-spi.c +++ b/drivers/input/misc/adxl34x-spi.c @@ -65,7 +65,7 @@ static const struct adxl34x_bus_ops adxl34x_spi_bops = { .read_block = adxl34x_spi_read_block, }; -static int __devinit adxl34x_spi_probe(struct spi_device *spi) +static int adxl34x_spi_probe(struct spi_device *spi) { struct adxl34x *ac; @@ -87,7 +87,7 @@ static int __devinit adxl34x_spi_probe(struct spi_device *spi) return 0; } -static int __devexit adxl34x_spi_remove(struct spi_device *spi) +static int adxl34x_spi_remove(struct spi_device *spi) { struct adxl34x *ac = dev_get_drvdata(&spi->dev); @@ -126,7 +126,7 @@ static struct spi_driver adxl34x_driver = { .pm = &adxl34x_spi_pm, }, .probe = adxl34x_spi_probe, - .remove = __devexit_p(adxl34x_spi_remove), + .remove = adxl34x_spi_remove, }; module_spi_driver(adxl34x_driver); diff --git a/drivers/input/misc/bfin_rotary.c b/drivers/input/misc/bfin_rotary.c index 1c4146fccfdf..a6666e142a91 100644 --- a/drivers/input/misc/bfin_rotary.c +++ b/drivers/input/misc/bfin_rotary.c @@ -90,7 +90,7 @@ static irqreturn_t bfin_rotary_isr(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit bfin_rotary_probe(struct platform_device *pdev) +static int bfin_rotary_probe(struct platform_device *pdev) { struct bfin_rotary_platform_data *pdata = pdev->dev.platform_data; struct bfin_rot *rotary; @@ -196,7 +196,7 @@ out1: return error; } -static int __devexit bfin_rotary_remove(struct platform_device *pdev) +static int bfin_rotary_remove(struct platform_device *pdev) { struct bfin_rot *rotary = platform_get_drvdata(pdev); @@ -255,7 +255,7 @@ static const struct dev_pm_ops bfin_rotary_pm_ops = { static struct platform_driver bfin_rotary_device_driver = { .probe = bfin_rotary_probe, - .remove = __devexit_p(bfin_rotary_remove), + .remove = bfin_rotary_remove, .driver = { .name = "bfin-rotary", .owner = THIS_MODULE, diff --git a/drivers/input/misc/bma150.c b/drivers/input/misc/bma150.c index e2f1e9f952b1..08ffcabd7220 100644 --- a/drivers/input/misc/bma150.c +++ b/drivers/input/misc/bma150.c @@ -158,7 +158,7 @@ struct bma150_data { * are stated and verified by Bosch Sensortec where they are configured * to provide a generic sensitivity performance. */ -static struct bma150_cfg default_cfg __devinitdata = { +static struct bma150_cfg default_cfg = { .any_motion_int = 1, .hg_int = 1, .lg_int = 1, @@ -224,7 +224,7 @@ static int bma150_set_mode(struct bma150_data *bma150, u8 mode) return 0; } -static int __devinit bma150_soft_reset(struct bma150_data *bma150) +static int bma150_soft_reset(struct bma150_data *bma150) { int error; @@ -237,19 +237,19 @@ static int __devinit bma150_soft_reset(struct bma150_data *bma150) return 0; } -static int __devinit bma150_set_range(struct bma150_data *bma150, u8 range) +static int bma150_set_range(struct bma150_data *bma150, u8 range) { return bma150_set_reg_bits(bma150->client, range, BMA150_RANGE_POS, BMA150_RANGE_MSK, BMA150_RANGE_REG); } -static int __devinit bma150_set_bandwidth(struct bma150_data *bma150, u8 bw) +static int bma150_set_bandwidth(struct bma150_data *bma150, u8 bw) { return bma150_set_reg_bits(bma150->client, bw, BMA150_BANDWIDTH_POS, BMA150_BANDWIDTH_MSK, BMA150_BANDWIDTH_REG); } -static int __devinit bma150_set_low_g_interrupt(struct bma150_data *bma150, +static int bma150_set_low_g_interrupt(struct bma150_data *bma150, u8 enable, u8 hyst, u8 dur, u8 thres) { int error; @@ -273,7 +273,7 @@ static int __devinit bma150_set_low_g_interrupt(struct bma150_data *bma150, BMA150_LOW_G_EN_REG); } -static int __devinit bma150_set_high_g_interrupt(struct bma150_data *bma150, +static int bma150_set_high_g_interrupt(struct bma150_data *bma150, u8 enable, u8 hyst, u8 dur, u8 thres) { int error; @@ -300,7 +300,7 @@ static int __devinit bma150_set_high_g_interrupt(struct bma150_data *bma150, } -static int __devinit bma150_set_any_motion_interrupt(struct bma150_data *bma150, +static int bma150_set_any_motion_interrupt(struct bma150_data *bma150, u8 enable, u8 dur, u8 thres) { int error; @@ -424,7 +424,7 @@ static void bma150_poll_close(struct input_polled_dev *ipoll_dev) bma150_close(bma150); } -static int __devinit bma150_initialize(struct bma150_data *bma150, +static int bma150_initialize(struct bma150_data *bma150, const struct bma150_cfg *cfg) { int error; @@ -465,7 +465,7 @@ static int __devinit bma150_initialize(struct bma150_data *bma150, return bma150_set_mode(bma150, BMA150_MODE_SLEEP); } -static void __devinit bma150_init_input_device(struct bma150_data *bma150, +static void bma150_init_input_device(struct bma150_data *bma150, struct input_dev *idev) { idev->name = BMA150_DRIVER; @@ -479,7 +479,7 @@ static void __devinit bma150_init_input_device(struct bma150_data *bma150, input_set_abs_params(idev, ABS_Z, ABSMIN_ACC_VAL, ABSMAX_ACC_VAL, 0, 0); } -static int __devinit bma150_register_input_device(struct bma150_data *bma150) +static int bma150_register_input_device(struct bma150_data *bma150) { struct input_dev *idev; int error; @@ -504,7 +504,7 @@ static int __devinit bma150_register_input_device(struct bma150_data *bma150) return 0; } -static int __devinit bma150_register_polled_device(struct bma150_data *bma150) +static int bma150_register_polled_device(struct bma150_data *bma150) { struct input_polled_dev *ipoll_dev; int error; @@ -535,7 +535,7 @@ static int __devinit bma150_register_polled_device(struct bma150_data *bma150) return 0; } -static int __devinit bma150_probe(struct i2c_client *client, +static int bma150_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct bma150_platform_data *pdata = client->dev.platform_data; @@ -613,7 +613,7 @@ err_free_mem: return error; } -static int __devexit bma150_remove(struct i2c_client *client) +static int bma150_remove(struct i2c_client *client) { struct bma150_data *bma150 = i2c_get_clientdata(client); @@ -670,7 +670,7 @@ static struct i2c_driver bma150_driver = { .class = I2C_CLASS_HWMON, .id_table = bma150_id, .probe = bma150_probe, - .remove = __devexit_p(bma150_remove), + .remove = bma150_remove, }; module_i2c_driver(bma150_driver); diff --git a/drivers/input/misc/cma3000_d0x_i2c.c b/drivers/input/misc/cma3000_d0x_i2c.c index fe9b85f07792..4fdef98ceb56 100644 --- a/drivers/input/misc/cma3000_d0x_i2c.c +++ b/drivers/input/misc/cma3000_d0x_i2c.c @@ -55,7 +55,7 @@ static const struct cma3000_bus_ops cma3000_i2c_bops = { .write = cma3000_i2c_set, }; -static int __devinit cma3000_i2c_probe(struct i2c_client *client, +static int cma3000_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct cma3000_accl_data *data; @@ -69,7 +69,7 @@ static int __devinit cma3000_i2c_probe(struct i2c_client *client, return 0; } -static int __devexit cma3000_i2c_remove(struct i2c_client *client) +static int cma3000_i2c_remove(struct i2c_client *client) { struct cma3000_accl_data *data = i2c_get_clientdata(client); @@ -114,7 +114,7 @@ MODULE_DEVICE_TABLE(i2c, cma3000_i2c_id); static struct i2c_driver cma3000_i2c_driver = { .probe = cma3000_i2c_probe, - .remove = __devexit_p(cma3000_i2c_remove), + .remove = cma3000_i2c_remove, .id_table = cma3000_i2c_id, .driver = { .name = "cma3000_i2c_accl", diff --git a/drivers/input/misc/cobalt_btns.c b/drivers/input/misc/cobalt_btns.c index 53e43d295148..4f77f87847e8 100644 --- a/drivers/input/misc/cobalt_btns.c +++ b/drivers/input/misc/cobalt_btns.c @@ -73,7 +73,7 @@ static void handle_buttons(struct input_polled_dev *dev) } } -static int __devinit cobalt_buttons_probe(struct platform_device *pdev) +static int cobalt_buttons_probe(struct platform_device *pdev) { struct buttons_dev *bdev; struct input_polled_dev *poll_dev; @@ -135,7 +135,7 @@ static int __devinit cobalt_buttons_probe(struct platform_device *pdev) return error; } -static int __devexit cobalt_buttons_remove(struct platform_device *pdev) +static int cobalt_buttons_remove(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct buttons_dev *bdev = dev_get_drvdata(dev); @@ -157,7 +157,7 @@ MODULE_ALIAS("platform:Cobalt buttons"); static struct platform_driver cobalt_buttons_driver = { .probe = cobalt_buttons_probe, - .remove = __devexit_p(cobalt_buttons_remove), + .remove = cobalt_buttons_remove, .driver = { .name = "Cobalt buttons", .owner = THIS_MODULE, diff --git a/drivers/input/misc/da9052_onkey.c b/drivers/input/misc/da9052_onkey.c index 3be3acc3a6eb..020569a499f2 100644 --- a/drivers/input/misc/da9052_onkey.c +++ b/drivers/input/misc/da9052_onkey.c @@ -70,7 +70,7 @@ static irqreturn_t da9052_onkey_irq(int irq, void *data) return IRQ_HANDLED; } -static int __devinit da9052_onkey_probe(struct platform_device *pdev) +static int da9052_onkey_probe(struct platform_device *pdev) { struct da9052 *da9052 = dev_get_drvdata(pdev->dev.parent); struct da9052_onkey *onkey; @@ -129,7 +129,7 @@ err_free_mem: return error; } -static int __devexit da9052_onkey_remove(struct platform_device *pdev) +static int da9052_onkey_remove(struct platform_device *pdev) { struct da9052_onkey *onkey = platform_get_drvdata(pdev); @@ -144,7 +144,7 @@ static int __devexit da9052_onkey_remove(struct platform_device *pdev) static struct platform_driver da9052_onkey_driver = { .probe = da9052_onkey_probe, - .remove = __devexit_p(da9052_onkey_remove), + .remove = da9052_onkey_remove, .driver = { .name = "da9052-onkey", .owner = THIS_MODULE, diff --git a/drivers/input/misc/da9055_onkey.c b/drivers/input/misc/da9055_onkey.c new file mode 100644 index 000000000000..ee6ae3a00174 --- /dev/null +++ b/drivers/input/misc/da9055_onkey.c @@ -0,0 +1,171 @@ +/* + * ON pin driver for Dialog DA9055 PMICs + * + * Copyright(c) 2012 Dialog Semiconductor Ltd. + * + * Author: David Dajun Chen <dchen@diasemi.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#include <linux/init.h> +#include <linux/input.h> +#include <linux/module.h> +#include <linux/platform_device.h> + +#include <linux/mfd/da9055/core.h> +#include <linux/mfd/da9055/reg.h> + +struct da9055_onkey { + struct da9055 *da9055; + struct input_dev *input; + struct delayed_work work; +}; + +static void da9055_onkey_query(struct da9055_onkey *onkey) +{ + int key_stat; + + key_stat = da9055_reg_read(onkey->da9055, DA9055_REG_STATUS_A); + if (key_stat < 0) { + dev_err(onkey->da9055->dev, + "Failed to read onkey event %d\n", key_stat); + } else { + key_stat &= DA9055_NOKEY_STS; + /* + * Onkey status bit is cleared when onkey button is relased. + */ + if (!key_stat) { + input_report_key(onkey->input, KEY_POWER, 0); + input_sync(onkey->input); + } + } + + /* + * Interrupt is generated only when the ONKEY pin is asserted. + * Hence the deassertion of the pin is simulated through work queue. + */ + if (key_stat) + schedule_delayed_work(&onkey->work, msecs_to_jiffies(10)); + +} + +static void da9055_onkey_work(struct work_struct *work) +{ + struct da9055_onkey *onkey = container_of(work, struct da9055_onkey, + work.work); + + da9055_onkey_query(onkey); +} + +static irqreturn_t da9055_onkey_irq(int irq, void *data) +{ + struct da9055_onkey *onkey = data; + + input_report_key(onkey->input, KEY_POWER, 1); + input_sync(onkey->input); + + da9055_onkey_query(onkey); + + return IRQ_HANDLED; +} + +static int da9055_onkey_probe(struct platform_device *pdev) +{ + struct da9055 *da9055 = dev_get_drvdata(pdev->dev.parent); + struct da9055_onkey *onkey; + struct input_dev *input_dev; + int irq, err; + + irq = platform_get_irq_byname(pdev, "ONKEY"); + if (irq < 0) { + dev_err(&pdev->dev, + "Failed to get an IRQ for input device, %d\n", irq); + return -EINVAL; + } + + onkey = devm_kzalloc(&pdev->dev, sizeof(*onkey), GFP_KERNEL); + if (!onkey) { + dev_err(&pdev->dev, "Failed to allocate memory\n"); + return -ENOMEM; + } + + input_dev = input_allocate_device(); + if (!input_dev) { + dev_err(&pdev->dev, "Failed to allocate memory\n"); + return -ENOMEM; + } + + onkey->input = input_dev; + onkey->da9055 = da9055; + input_dev->name = "da9055-onkey"; + input_dev->phys = "da9055-onkey/input0"; + input_dev->dev.parent = &pdev->dev; + + input_dev->evbit[0] = BIT_MASK(EV_KEY); + __set_bit(KEY_POWER, input_dev->keybit); + + INIT_DELAYED_WORK(&onkey->work, da9055_onkey_work); + + irq = regmap_irq_get_virq(da9055->irq_data, irq); + err = request_threaded_irq(irq, NULL, da9055_onkey_irq, + IRQF_TRIGGER_HIGH | IRQF_ONESHOT, + "ONKEY", onkey); + if (err < 0) { + dev_err(&pdev->dev, + "Failed to register ONKEY IRQ %d, error = %d\n", + irq, err); + goto err_free_input; + } + + err = input_register_device(input_dev); + if (err) { + dev_err(&pdev->dev, "Unable to register input device, %d\n", + err); + goto err_free_irq; + } + + platform_set_drvdata(pdev, onkey); + + return 0; + +err_free_irq: + free_irq(irq, onkey); + cancel_delayed_work_sync(&onkey->work); +err_free_input: + input_free_device(input_dev); + + return err; +} + +static int da9055_onkey_remove(struct platform_device *pdev) +{ + struct da9055_onkey *onkey = platform_get_drvdata(pdev); + int irq = platform_get_irq_byname(pdev, "ONKEY"); + + irq = regmap_irq_get_virq(onkey->da9055->irq_data, irq); + free_irq(irq, onkey); + cancel_delayed_work_sync(&onkey->work); + input_unregister_device(onkey->input); + + return 0; +} + +static struct platform_driver da9055_onkey_driver = { + .probe = da9055_onkey_probe, + .remove = da9055_onkey_remove, + .driver = { + .name = "da9055-onkey", + .owner = THIS_MODULE, + }, +}; + +module_platform_driver(da9055_onkey_driver); + +MODULE_AUTHOR("David Dajun Chen <dchen@diasemi.com>"); +MODULE_DESCRIPTION("Onkey driver for DA9055"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("platform:da9055-onkey"); diff --git a/drivers/input/misc/dm355evm_keys.c b/drivers/input/misc/dm355evm_keys.c index c1313d8535c3..a309a5c0899e 100644 --- a/drivers/input/misc/dm355evm_keys.c +++ b/drivers/input/misc/dm355evm_keys.c @@ -173,7 +173,7 @@ static irqreturn_t dm355evm_keys_irq(int irq, void *_keys) /*----------------------------------------------------------------------*/ -static int __devinit dm355evm_keys_probe(struct platform_device *pdev) +static int dm355evm_keys_probe(struct platform_device *pdev) { struct dm355evm_keys *keys; struct input_dev *input; @@ -239,7 +239,7 @@ fail1: return status; } -static int __devexit dm355evm_keys_remove(struct platform_device *pdev) +static int dm355evm_keys_remove(struct platform_device *pdev) { struct dm355evm_keys *keys = platform_get_drvdata(pdev); @@ -262,7 +262,7 @@ static int __devexit dm355evm_keys_remove(struct platform_device *pdev) */ static struct platform_driver dm355evm_keys_driver = { .probe = dm355evm_keys_probe, - .remove = __devexit_p(dm355evm_keys_remove), + .remove = dm355evm_keys_remove, .driver = { .owner = THIS_MODULE, .name = "dm355evm_keys", diff --git a/drivers/input/misc/gp2ap002a00f.c b/drivers/input/misc/gp2ap002a00f.c index b6664cfa340a..fe30bd0fe4bd 100644 --- a/drivers/input/misc/gp2ap002a00f.c +++ b/drivers/input/misc/gp2ap002a00f.c @@ -98,7 +98,7 @@ static void gp2a_device_close(struct input_dev *dev) "unable to deactivate, err %d\n", error); } -static int __devinit gp2a_initialize(struct gp2a_data *dt) +static int gp2a_initialize(struct gp2a_data *dt) { int error; @@ -122,7 +122,7 @@ static int __devinit gp2a_initialize(struct gp2a_data *dt) return error; } -static int __devinit gp2a_probe(struct i2c_client *client, +static int gp2a_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct gp2a_platform_data *pdata = client->dev.platform_data; @@ -205,7 +205,7 @@ err_hw_shutdown: return error; } -static int __devexit gp2a_remove(struct i2c_client *client) +static int gp2a_remove(struct i2c_client *client) { struct gp2a_data *dt = i2c_get_clientdata(client); const struct gp2a_platform_data *pdata = dt->pdata; @@ -277,7 +277,7 @@ static struct i2c_driver gp2a_i2c_driver = { .pm = &gp2a_pm, }, .probe = gp2a_probe, - .remove = __devexit_p(gp2a_remove), + .remove = gp2a_remove, .id_table = gp2a_i2c_id, }; diff --git a/drivers/input/misc/gpio_tilt_polled.c b/drivers/input/misc/gpio_tilt_polled.c index 277a0574c199..da05cca8b562 100644 --- a/drivers/input/misc/gpio_tilt_polled.c +++ b/drivers/input/misc/gpio_tilt_polled.c @@ -96,7 +96,7 @@ static void gpio_tilt_polled_close(struct input_polled_dev *dev) pdata->disable(tdev->dev); } -static int __devinit gpio_tilt_polled_probe(struct platform_device *pdev) +static int gpio_tilt_polled_probe(struct platform_device *pdev) { const struct gpio_tilt_platform_data *pdata = pdev->dev.platform_data; struct device *dev = &pdev->dev; @@ -179,7 +179,7 @@ err_free_tdev: return error; } -static int __devexit gpio_tilt_polled_remove(struct platform_device *pdev) +static int gpio_tilt_polled_remove(struct platform_device *pdev) { struct gpio_tilt_polled_dev *tdev = platform_get_drvdata(pdev); const struct gpio_tilt_platform_data *pdata = tdev->pdata; @@ -198,7 +198,7 @@ static int __devexit gpio_tilt_polled_remove(struct platform_device *pdev) static struct platform_driver gpio_tilt_polled_driver = { .probe = gpio_tilt_polled_probe, - .remove = __devexit_p(gpio_tilt_polled_remove), + .remove = gpio_tilt_polled_remove, .driver = { .name = DRV_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/misc/ixp4xx-beeper.c b/drivers/input/misc/ixp4xx-beeper.c index 50e283068301..6ab3decc86e6 100644 --- a/drivers/input/misc/ixp4xx-beeper.c +++ b/drivers/input/misc/ixp4xx-beeper.c @@ -87,7 +87,7 @@ static irqreturn_t ixp4xx_spkr_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit ixp4xx_spkr_probe(struct platform_device *dev) +static int ixp4xx_spkr_probe(struct platform_device *dev) { struct input_dev *input_dev; int err; @@ -132,7 +132,7 @@ static int __devinit ixp4xx_spkr_probe(struct platform_device *dev) return err; } -static int __devexit ixp4xx_spkr_remove(struct platform_device *dev) +static int ixp4xx_spkr_remove(struct platform_device *dev) { struct input_dev *input_dev = platform_get_drvdata(dev); unsigned int pin = (unsigned int) input_get_drvdata(input_dev); @@ -165,7 +165,7 @@ static struct platform_driver ixp4xx_spkr_platform_driver = { .owner = THIS_MODULE, }, .probe = ixp4xx_spkr_probe, - .remove = __devexit_p(ixp4xx_spkr_remove), + .remove = ixp4xx_spkr_remove, .shutdown = ixp4xx_spkr_shutdown, }; module_platform_driver(ixp4xx_spkr_platform_driver); diff --git a/drivers/input/misc/kxtj9.c b/drivers/input/misc/kxtj9.c index f46139f19ff1..a993b67a8a5b 100644 --- a/drivers/input/misc/kxtj9.c +++ b/drivers/input/misc/kxtj9.c @@ -295,7 +295,7 @@ static void kxtj9_input_close(struct input_dev *dev) kxtj9_disable(tj9); } -static void __devinit kxtj9_init_input_device(struct kxtj9_data *tj9, +static void kxtj9_init_input_device(struct kxtj9_data *tj9, struct input_dev *input_dev) { __set_bit(EV_ABS, input_dev->evbit); @@ -308,7 +308,7 @@ static void __devinit kxtj9_init_input_device(struct kxtj9_data *tj9, input_dev->dev.parent = &tj9->client->dev; } -static int __devinit kxtj9_setup_input_device(struct kxtj9_data *tj9) +static int kxtj9_setup_input_device(struct kxtj9_data *tj9) { struct input_dev *input_dev; int err; @@ -433,7 +433,7 @@ static void kxtj9_polled_input_close(struct input_polled_dev *dev) kxtj9_disable(tj9); } -static int __devinit kxtj9_setup_polled_device(struct kxtj9_data *tj9) +static int kxtj9_setup_polled_device(struct kxtj9_data *tj9) { int err; struct input_polled_dev *poll_dev; @@ -466,7 +466,7 @@ static int __devinit kxtj9_setup_polled_device(struct kxtj9_data *tj9) return 0; } -static void __devexit kxtj9_teardown_polled_device(struct kxtj9_data *tj9) +static void kxtj9_teardown_polled_device(struct kxtj9_data *tj9) { input_unregister_polled_device(tj9->poll_dev); input_free_polled_device(tj9->poll_dev); @@ -485,7 +485,7 @@ static inline void kxtj9_teardown_polled_device(struct kxtj9_data *tj9) #endif -static int __devinit kxtj9_verify(struct kxtj9_data *tj9) +static int kxtj9_verify(struct kxtj9_data *tj9) { int retval; @@ -506,7 +506,7 @@ out: return retval; } -static int __devinit kxtj9_probe(struct i2c_client *client, +static int kxtj9_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct kxtj9_platform_data *pdata = client->dev.platform_data; @@ -594,7 +594,7 @@ err_free_mem: return err; } -static int __devexit kxtj9_remove(struct i2c_client *client) +static int kxtj9_remove(struct i2c_client *client) { struct kxtj9_data *tj9 = i2c_get_clientdata(client); @@ -663,7 +663,7 @@ static struct i2c_driver kxtj9_driver = { .pm = &kxtj9_pm_ops, }, .probe = kxtj9_probe, - .remove = __devexit_p(kxtj9_remove), + .remove = kxtj9_remove, .id_table = kxtj9_id, }; diff --git a/drivers/input/misc/m68kspkr.c b/drivers/input/misc/m68kspkr.c index 0c64d9bb718e..b40ee4b47f4f 100644 --- a/drivers/input/misc/m68kspkr.c +++ b/drivers/input/misc/m68kspkr.c @@ -48,7 +48,7 @@ static int m68kspkr_event(struct input_dev *dev, unsigned int type, unsigned int return 0; } -static int __devinit m68kspkr_probe(struct platform_device *dev) +static int m68kspkr_probe(struct platform_device *dev) { struct input_dev *input_dev; int err; @@ -80,7 +80,7 @@ static int __devinit m68kspkr_probe(struct platform_device *dev) return 0; } -static int __devexit m68kspkr_remove(struct platform_device *dev) +static int m68kspkr_remove(struct platform_device *dev) { struct input_dev *input_dev = platform_get_drvdata(dev); @@ -104,7 +104,7 @@ static struct platform_driver m68kspkr_platform_driver = { .owner = THIS_MODULE, }, .probe = m68kspkr_probe, - .remove = __devexit_p(m68kspkr_remove), + .remove = m68kspkr_remove, .shutdown = m68kspkr_shutdown, }; diff --git a/drivers/input/misc/max8925_onkey.c b/drivers/input/misc/max8925_onkey.c index 0a12b74140d3..369a39de4ff3 100644 --- a/drivers/input/misc/max8925_onkey.c +++ b/drivers/input/misc/max8925_onkey.c @@ -62,7 +62,7 @@ static irqreturn_t max8925_onkey_handler(int irq, void *data) return IRQ_HANDLED; } -static int __devinit max8925_onkey_probe(struct platform_device *pdev) +static int max8925_onkey_probe(struct platform_device *pdev) { struct max8925_chip *chip = dev_get_drvdata(pdev->dev.parent); struct max8925_onkey_info *info; @@ -141,7 +141,7 @@ err_free_mem: return error; } -static int __devexit max8925_onkey_remove(struct platform_device *pdev) +static int max8925_onkey_remove(struct platform_device *pdev) { struct max8925_onkey_info *info = platform_get_drvdata(pdev); struct max8925_chip *chip = dev_get_drvdata(pdev->dev.parent); @@ -195,7 +195,7 @@ static struct platform_driver max8925_onkey_driver = { .pm = &max8925_onkey_pm_ops, }, .probe = max8925_onkey_probe, - .remove = __devexit_p(max8925_onkey_remove), + .remove = max8925_onkey_remove, }; module_platform_driver(max8925_onkey_driver); diff --git a/drivers/input/misc/max8997_haptic.c b/drivers/input/misc/max8997_haptic.c index 05b7b8bfaf0a..e973133212a5 100644 --- a/drivers/input/misc/max8997_haptic.c +++ b/drivers/input/misc/max8997_haptic.c @@ -241,7 +241,7 @@ static void max8997_haptic_close(struct input_dev *dev) max8997_haptic_disable(chip); } -static int __devinit max8997_haptic_probe(struct platform_device *pdev) +static int max8997_haptic_probe(struct platform_device *pdev) { struct max8997_dev *iodev = dev_get_drvdata(pdev->dev.parent); const struct max8997_platform_data *pdata = @@ -354,7 +354,7 @@ err_free_mem: return error; } -static int __devexit max8997_haptic_remove(struct platform_device *pdev) +static int max8997_haptic_remove(struct platform_device *pdev) { struct max8997_haptic *chip = platform_get_drvdata(pdev); @@ -396,7 +396,7 @@ static struct platform_driver max8997_haptic_driver = { .pm = &max8997_haptic_pm_ops, }, .probe = max8997_haptic_probe, - .remove = __devexit_p(max8997_haptic_remove), + .remove = max8997_haptic_remove, .id_table = max8997_haptic_id, }; module_platform_driver(max8997_haptic_driver); diff --git a/drivers/input/misc/mc13783-pwrbutton.c b/drivers/input/misc/mc13783-pwrbutton.c index 8428f1e8e83e..0906ca593d5f 100644 --- a/drivers/input/misc/mc13783-pwrbutton.c +++ b/drivers/input/misc/mc13783-pwrbutton.c @@ -89,7 +89,7 @@ static irqreturn_t button_irq(int irq, void *_priv) return IRQ_HANDLED; } -static int __devinit mc13783_pwrbutton_probe(struct platform_device *pdev) +static int mc13783_pwrbutton_probe(struct platform_device *pdev) { const struct mc13xxx_buttons_platform_data *pdata; struct mc13xxx *mc13783 = dev_get_drvdata(pdev->dev.parent); @@ -230,7 +230,7 @@ free_input_dev: return err; } -static int __devexit mc13783_pwrbutton_remove(struct platform_device *pdev) +static int mc13783_pwrbutton_remove(struct platform_device *pdev) { struct mc13783_pwrb *priv = platform_get_drvdata(pdev); const struct mc13xxx_buttons_platform_data *pdata; @@ -257,7 +257,7 @@ static int __devexit mc13783_pwrbutton_remove(struct platform_device *pdev) static struct platform_driver mc13783_pwrbutton_driver = { .probe = mc13783_pwrbutton_probe, - .remove = __devexit_p(mc13783_pwrbutton_remove), + .remove = mc13783_pwrbutton_remove, .driver = { .name = "mc13783-pwrbutton", .owner = THIS_MODULE, diff --git a/drivers/input/misc/mma8450.c b/drivers/input/misc/mma8450.c index 873ebced544e..480557f14f23 100644 --- a/drivers/input/misc/mma8450.c +++ b/drivers/input/misc/mma8450.c @@ -167,7 +167,7 @@ static void mma8450_close(struct input_polled_dev *dev) /* * I2C init/probing/exit functions */ -static int __devinit mma8450_probe(struct i2c_client *c, +static int mma8450_probe(struct i2c_client *c, const struct i2c_device_id *id) { struct input_polled_dev *idev; @@ -212,7 +212,7 @@ err_free_mem: return err; } -static int __devexit mma8450_remove(struct i2c_client *c) +static int mma8450_remove(struct i2c_client *c) { struct mma8450 *m = i2c_get_clientdata(c); struct input_polled_dev *idev = m->idev; @@ -243,7 +243,7 @@ static struct i2c_driver mma8450_driver = { .of_match_table = mma8450_dt_ids, }, .probe = mma8450_probe, - .remove = __devexit_p(mma8450_remove), + .remove = mma8450_remove, .id_table = mma8450_id, }; diff --git a/drivers/input/misc/mpu3050.c b/drivers/input/misc/mpu3050.c index 306f84c2d8fb..dce0d95943c5 100644 --- a/drivers/input/misc/mpu3050.c +++ b/drivers/input/misc/mpu3050.c @@ -257,7 +257,7 @@ static irqreturn_t mpu3050_interrupt_thread(int irq, void *data) * * Called during device probe; configures the sampling method. */ -static int __devinit mpu3050_hw_init(struct mpu3050_sensor *sensor) +static int mpu3050_hw_init(struct mpu3050_sensor *sensor) { struct i2c_client *client = sensor->client; int ret; @@ -306,7 +306,7 @@ static int __devinit mpu3050_hw_init(struct mpu3050_sensor *sensor) * * If present install the relevant sysfs interfaces and input device. */ -static int __devinit mpu3050_probe(struct i2c_client *client, +static int mpu3050_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct mpu3050_sensor *sensor; @@ -402,7 +402,7 @@ err_free_mem: * * Our sensor is going away, clean up the resources. */ -static int __devexit mpu3050_remove(struct i2c_client *client) +static int mpu3050_remove(struct i2c_client *client) { struct mpu3050_sensor *sensor = i2c_get_clientdata(client); @@ -471,7 +471,7 @@ static struct i2c_driver mpu3050_i2c_driver = { .of_match_table = mpu3050_of_match, }, .probe = mpu3050_probe, - .remove = __devexit_p(mpu3050_remove), + .remove = mpu3050_remove, .id_table = mpu3050_ids, }; diff --git a/drivers/input/misc/pcap_keys.c b/drivers/input/misc/pcap_keys.c index e09b4fe81913..40ac9a5adf89 100644 --- a/drivers/input/misc/pcap_keys.c +++ b/drivers/input/misc/pcap_keys.c @@ -48,7 +48,7 @@ static irqreturn_t pcap_keys_handler(int irq, void *_pcap_keys) return IRQ_HANDLED; } -static int __devinit pcap_keys_probe(struct platform_device *pdev) +static int pcap_keys_probe(struct platform_device *pdev) { int err = -ENOMEM; struct pcap_keys *pcap_keys; @@ -104,7 +104,7 @@ fail: return err; } -static int __devexit pcap_keys_remove(struct platform_device *pdev) +static int pcap_keys_remove(struct platform_device *pdev) { struct pcap_keys *pcap_keys = platform_get_drvdata(pdev); @@ -119,7 +119,7 @@ static int __devexit pcap_keys_remove(struct platform_device *pdev) static struct platform_driver pcap_keys_device_driver = { .probe = pcap_keys_probe, - .remove = __devexit_p(pcap_keys_remove), + .remove = pcap_keys_remove, .driver = { .name = "pcap-keys", .owner = THIS_MODULE, diff --git a/drivers/input/misc/pcf50633-input.c b/drivers/input/misc/pcf50633-input.c index 53891de80b0e..73b13ebabe56 100644 --- a/drivers/input/misc/pcf50633-input.c +++ b/drivers/input/misc/pcf50633-input.c @@ -53,7 +53,7 @@ pcf50633_input_irq(int irq, void *data) input_sync(input->input_dev); } -static int __devinit pcf50633_input_probe(struct platform_device *pdev) +static int pcf50633_input_probe(struct platform_device *pdev) { struct pcf50633_input *input; struct input_dev *input_dev; @@ -93,7 +93,7 @@ static int __devinit pcf50633_input_probe(struct platform_device *pdev) return 0; } -static int __devexit pcf50633_input_remove(struct platform_device *pdev) +static int pcf50633_input_remove(struct platform_device *pdev) { struct pcf50633_input *input = platform_get_drvdata(pdev); @@ -111,7 +111,7 @@ static struct platform_driver pcf50633_input_driver = { .name = "pcf50633-input", }, .probe = pcf50633_input_probe, - .remove = __devexit_p(pcf50633_input_remove), + .remove = pcf50633_input_remove, }; module_platform_driver(pcf50633_input_driver); diff --git a/drivers/input/misc/pcf8574_keypad.c b/drivers/input/misc/pcf8574_keypad.c index 544c6635abe9..e37392976fdd 100644 --- a/drivers/input/misc/pcf8574_keypad.c +++ b/drivers/input/misc/pcf8574_keypad.c @@ -82,7 +82,7 @@ static irqreturn_t pcf8574_kp_irq_handler(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit pcf8574_kp_probe(struct i2c_client *client, const struct i2c_device_id *id) +static int pcf8574_kp_probe(struct i2c_client *client, const struct i2c_device_id *id) { int i, ret; struct input_dev *idev; @@ -156,7 +156,7 @@ static int __devinit pcf8574_kp_probe(struct i2c_client *client, const struct i2 return ret; } -static int __devexit pcf8574_kp_remove(struct i2c_client *client) +static int pcf8574_kp_remove(struct i2c_client *client) { struct kp_data *lp = i2c_get_clientdata(client); @@ -212,7 +212,7 @@ static struct i2c_driver pcf8574_kp_driver = { #endif }, .probe = pcf8574_kp_probe, - .remove = __devexit_p(pcf8574_kp_remove), + .remove = pcf8574_kp_remove, .id_table = pcf8574_kp_id, }; diff --git a/drivers/input/misc/pcspkr.c b/drivers/input/misc/pcspkr.c index b2484aa07f32..199db78acc4f 100644 --- a/drivers/input/misc/pcspkr.c +++ b/drivers/input/misc/pcspkr.c @@ -63,7 +63,7 @@ static int pcspkr_event(struct input_dev *dev, unsigned int type, unsigned int c return 0; } -static int __devinit pcspkr_probe(struct platform_device *dev) +static int pcspkr_probe(struct platform_device *dev) { struct input_dev *pcspkr_dev; int err; @@ -95,7 +95,7 @@ static int __devinit pcspkr_probe(struct platform_device *dev) return 0; } -static int __devexit pcspkr_remove(struct platform_device *dev) +static int pcspkr_remove(struct platform_device *dev) { struct input_dev *pcspkr_dev = platform_get_drvdata(dev); @@ -131,7 +131,7 @@ static struct platform_driver pcspkr_platform_driver = { .pm = &pcspkr_pm_ops, }, .probe = pcspkr_probe, - .remove = __devexit_p(pcspkr_remove), + .remove = pcspkr_remove, .shutdown = pcspkr_shutdown, }; module_platform_driver(pcspkr_platform_driver); diff --git a/drivers/input/misc/pm8xxx-vibrator.c b/drivers/input/misc/pm8xxx-vibrator.c index dfbfb463ea5d..a9da65e41c5b 100644 --- a/drivers/input/misc/pm8xxx-vibrator.c +++ b/drivers/input/misc/pm8xxx-vibrator.c @@ -178,7 +178,7 @@ static int pm8xxx_vib_play_effect(struct input_dev *dev, void *data, return 0; } -static int __devinit pm8xxx_vib_probe(struct platform_device *pdev) +static int pm8xxx_vib_probe(struct platform_device *pdev) { struct pm8xxx_vib *vib; @@ -242,7 +242,7 @@ err_free_mem: return error; } -static int __devexit pm8xxx_vib_remove(struct platform_device *pdev) +static int pm8xxx_vib_remove(struct platform_device *pdev) { struct pm8xxx_vib *vib = platform_get_drvdata(pdev); @@ -270,7 +270,7 @@ static SIMPLE_DEV_PM_OPS(pm8xxx_vib_pm_ops, pm8xxx_vib_suspend, NULL); static struct platform_driver pm8xxx_vib_driver = { .probe = pm8xxx_vib_probe, - .remove = __devexit_p(pm8xxx_vib_remove), + .remove = pm8xxx_vib_remove, .driver = { .name = "pm8xxx-vib", .owner = THIS_MODULE, diff --git a/drivers/input/misc/pmic8xxx-pwrkey.c b/drivers/input/misc/pmic8xxx-pwrkey.c index 0f83d0f1d015..4b811be73974 100644 --- a/drivers/input/misc/pmic8xxx-pwrkey.c +++ b/drivers/input/misc/pmic8xxx-pwrkey.c @@ -81,7 +81,7 @@ static int pmic8xxx_pwrkey_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(pm8xxx_pwr_key_pm_ops, pmic8xxx_pwrkey_suspend, pmic8xxx_pwrkey_resume); -static int __devinit pmic8xxx_pwrkey_probe(struct platform_device *pdev) +static int pmic8xxx_pwrkey_probe(struct platform_device *pdev) { struct input_dev *pwr; int key_release_irq = platform_get_irq(pdev, 0); @@ -187,7 +187,7 @@ free_pwrkey: return err; } -static int __devexit pmic8xxx_pwrkey_remove(struct platform_device *pdev) +static int pmic8xxx_pwrkey_remove(struct platform_device *pdev) { struct pmic8xxx_pwrkey *pwrkey = platform_get_drvdata(pdev); int key_release_irq = platform_get_irq(pdev, 0); @@ -206,7 +206,7 @@ static int __devexit pmic8xxx_pwrkey_remove(struct platform_device *pdev) static struct platform_driver pmic8xxx_pwrkey_driver = { .probe = pmic8xxx_pwrkey_probe, - .remove = __devexit_p(pmic8xxx_pwrkey_remove), + .remove = pmic8xxx_pwrkey_remove, .driver = { .name = PM8XXX_PWRKEY_DEV_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/misc/pwm-beeper.c b/drivers/input/misc/pwm-beeper.c index fc84c8a51147..0808868461de 100644 --- a/drivers/input/misc/pwm-beeper.c +++ b/drivers/input/misc/pwm-beeper.c @@ -65,7 +65,7 @@ static int pwm_beeper_event(struct input_dev *input, return 0; } -static int __devinit pwm_beeper_probe(struct platform_device *pdev) +static int pwm_beeper_probe(struct platform_device *pdev) { unsigned long pwm_id = (unsigned long)pdev->dev.platform_data; struct pwm_beeper *beeper; @@ -75,7 +75,11 @@ static int __devinit pwm_beeper_probe(struct platform_device *pdev) if (!beeper) return -ENOMEM; - beeper->pwm = pwm_request(pwm_id, "pwm beeper"); + beeper->pwm = pwm_get(&pdev->dev, NULL); + if (IS_ERR(beeper->pwm)) { + dev_dbg(&pdev->dev, "unable to request PWM, trying legacy API\n"); + beeper->pwm = pwm_request(pwm_id, "pwm beeper"); + } if (IS_ERR(beeper->pwm)) { error = PTR_ERR(beeper->pwm); @@ -125,7 +129,7 @@ err_free: return error; } -static int __devexit pwm_beeper_remove(struct platform_device *pdev) +static int pwm_beeper_remove(struct platform_device *pdev) { struct pwm_beeper *beeper = platform_get_drvdata(pdev); @@ -171,13 +175,21 @@ static SIMPLE_DEV_PM_OPS(pwm_beeper_pm_ops, #define PWM_BEEPER_PM_OPS NULL #endif +#ifdef CONFIG_OF +static const struct of_device_id pwm_beeper_match[] = { + { .compatible = "pwm-beeper", }, + { }, +}; +#endif + static struct platform_driver pwm_beeper_driver = { .probe = pwm_beeper_probe, - .remove = __devexit_p(pwm_beeper_remove), + .remove = pwm_beeper_remove, .driver = { .name = "pwm-beeper", .owner = THIS_MODULE, .pm = PWM_BEEPER_PM_OPS, + .of_match_table = of_match_ptr(pwm_beeper_match), }, }; module_platform_driver(pwm_beeper_driver); diff --git a/drivers/input/misc/rb532_button.c b/drivers/input/misc/rb532_button.c index aeb02bcf7233..fb4f8ac3343b 100644 --- a/drivers/input/misc/rb532_button.c +++ b/drivers/input/misc/rb532_button.c @@ -51,7 +51,7 @@ static void rb532_button_poll(struct input_polled_dev *poll_dev) input_sync(poll_dev->input); } -static int __devinit rb532_button_probe(struct platform_device *pdev) +static int rb532_button_probe(struct platform_device *pdev) { struct input_polled_dev *poll_dev; int error; @@ -81,7 +81,7 @@ static int __devinit rb532_button_probe(struct platform_device *pdev) return 0; } -static int __devexit rb532_button_remove(struct platform_device *pdev) +static int rb532_button_remove(struct platform_device *pdev) { struct input_polled_dev *poll_dev = dev_get_drvdata(&pdev->dev); @@ -94,7 +94,7 @@ static int __devexit rb532_button_remove(struct platform_device *pdev) static struct platform_driver rb532_button_driver = { .probe = rb532_button_probe, - .remove = __devexit_p(rb532_button_remove), + .remove = rb532_button_remove, .driver = { .name = DRV_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/misc/retu-pwrbutton.c b/drivers/input/misc/retu-pwrbutton.c new file mode 100644 index 000000000000..7ca09baa0016 --- /dev/null +++ b/drivers/input/misc/retu-pwrbutton.c @@ -0,0 +1,99 @@ +/* + * Retu power button driver. + * + * Copyright (C) 2004-2010 Nokia Corporation + * + * Original code written by Ari Saastamoinen, Juha Yrjölä and Felipe Balbi. + * Rewritten by Aaro Koskinen. + * + * This file is subject to the terms and conditions of the GNU General + * Public License. See the file "COPYING" in the main directory of this + * archive for more details. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/irq.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/errno.h> +#include <linux/input.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/mfd/retu.h> +#include <linux/interrupt.h> +#include <linux/platform_device.h> + +#define RETU_STATUS_PWRONX (1 << 5) + +static irqreturn_t retu_pwrbutton_irq(int irq, void *_pwr) +{ + struct input_dev *idev = _pwr; + struct retu_dev *rdev = input_get_drvdata(idev); + bool state; + + state = !(retu_read(rdev, RETU_REG_STATUS) & RETU_STATUS_PWRONX); + input_report_key(idev, KEY_POWER, state); + input_sync(idev); + + return IRQ_HANDLED; +} + +static int retu_pwrbutton_probe(struct platform_device *pdev) +{ + struct retu_dev *rdev = dev_get_drvdata(pdev->dev.parent); + struct input_dev *idev; + int irq; + int error; + + irq = platform_get_irq(pdev, 0); + if (irq < 0) + return irq; + + idev = devm_input_allocate_device(&pdev->dev); + if (!idev) + return -ENOMEM; + + idev->name = "retu-pwrbutton"; + idev->dev.parent = &pdev->dev; + + input_set_capability(idev, EV_KEY, KEY_POWER); + input_set_drvdata(idev, rdev); + + error = devm_request_threaded_irq(&pdev->dev, irq, + NULL, retu_pwrbutton_irq, 0, + "retu-pwrbutton", idev); + if (error) + return error; + + error = input_register_device(idev); + if (error) + return error; + + return 0; +} + +static int retu_pwrbutton_remove(struct platform_device *pdev) +{ + return 0; +} + +static struct platform_driver retu_pwrbutton_driver = { + .probe = retu_pwrbutton_probe, + .remove = retu_pwrbutton_remove, + .driver = { + .name = "retu-pwrbutton", + .owner = THIS_MODULE, + }, +}; +module_platform_driver(retu_pwrbutton_driver); + +MODULE_ALIAS("platform:retu-pwrbutton"); +MODULE_DESCRIPTION("Retu Power Button"); +MODULE_AUTHOR("Ari Saastamoinen"); +MODULE_AUTHOR("Felipe Balbi"); +MODULE_AUTHOR("Aaro Koskinen <aaro.koskinen@iki.fi>"); +MODULE_LICENSE("GPL"); diff --git a/drivers/input/misc/rotary_encoder.c b/drivers/input/misc/rotary_encoder.c index 99a49e4968d2..aff47b2c38ff 100644 --- a/drivers/input/misc/rotary_encoder.c +++ b/drivers/input/misc/rotary_encoder.c @@ -149,8 +149,7 @@ static struct of_device_id rotary_encoder_of_match[] = { }; MODULE_DEVICE_TABLE(of, rotary_encoder_of_match); -static struct rotary_encoder_platform_data * __devinit -rotary_encoder_parse_dt(struct device *dev) +static struct rotary_encoder_platform_data *rotary_encoder_parse_dt(struct device *dev) { const struct of_device_id *of_id = of_match_device(rotary_encoder_of_match, dev); @@ -192,7 +191,7 @@ rotary_encoder_parse_dt(struct device *dev) } #endif -static int __devinit rotary_encoder_probe(struct platform_device *pdev) +static int rotary_encoder_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; const struct rotary_encoder_platform_data *pdata = dev_get_platdata(dev); @@ -302,7 +301,7 @@ exit_free_mem: return err; } -static int __devexit rotary_encoder_remove(struct platform_device *pdev) +static int rotary_encoder_remove(struct platform_device *pdev) { struct rotary_encoder *encoder = platform_get_drvdata(pdev); const struct rotary_encoder_platform_data *pdata = encoder->pdata; @@ -325,7 +324,7 @@ static int __devexit rotary_encoder_remove(struct platform_device *pdev) static struct platform_driver rotary_encoder_driver = { .probe = rotary_encoder_probe, - .remove = __devexit_p(rotary_encoder_remove), + .remove = rotary_encoder_remove, .driver = { .name = DRV_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/misc/sgi_btns.c b/drivers/input/misc/sgi_btns.c index 5d9fd5571199..ad6415ceaf5f 100644 --- a/drivers/input/misc/sgi_btns.c +++ b/drivers/input/misc/sgi_btns.c @@ -91,7 +91,7 @@ static void handle_buttons(struct input_polled_dev *dev) } } -static int __devinit sgi_buttons_probe(struct platform_device *pdev) +static int sgi_buttons_probe(struct platform_device *pdev) { struct buttons_dev *bdev; struct input_polled_dev *poll_dev; @@ -143,7 +143,7 @@ static int __devinit sgi_buttons_probe(struct platform_device *pdev) return error; } -static int __devexit sgi_buttons_remove(struct platform_device *pdev) +static int sgi_buttons_remove(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct buttons_dev *bdev = dev_get_drvdata(dev); @@ -158,7 +158,7 @@ static int __devexit sgi_buttons_remove(struct platform_device *pdev) static struct platform_driver sgi_buttons_driver = { .probe = sgi_buttons_probe, - .remove = __devexit_p(sgi_buttons_remove), + .remove = sgi_buttons_remove, .driver = { .name = "sgibtns", .owner = THIS_MODULE, diff --git a/drivers/input/misc/sparcspkr.c b/drivers/input/misc/sparcspkr.c index 0122f5351577..a53586a7fbdb 100644 --- a/drivers/input/misc/sparcspkr.c +++ b/drivers/input/misc/sparcspkr.c @@ -139,7 +139,7 @@ static int grover_spkr_event(struct input_dev *dev, unsigned int type, unsigned return 0; } -static int __devinit sparcspkr_probe(struct device *dev) +static int sparcspkr_probe(struct device *dev) { struct sparcspkr_state *state = dev_get_drvdata(dev); struct input_dev *input_dev; @@ -182,7 +182,7 @@ static void sparcspkr_shutdown(struct platform_device *dev) state->event(input_dev, EV_SND, SND_BELL, 0); } -static int __devinit bbc_beep_probe(struct platform_device *op) +static int bbc_beep_probe(struct platform_device *op) { struct sparcspkr_state *state; struct bbc_beep_info *info; @@ -229,7 +229,7 @@ out_err: return err; } -static int __devexit bbc_remove(struct platform_device *op) +static int bbc_remove(struct platform_device *op) { struct sparcspkr_state *state = dev_get_drvdata(&op->dev); struct input_dev *input_dev = state->input_dev; @@ -263,11 +263,11 @@ static struct platform_driver bbc_beep_driver = { .of_match_table = bbc_beep_match, }, .probe = bbc_beep_probe, - .remove = __devexit_p(bbc_remove), + .remove = bbc_remove, .shutdown = sparcspkr_shutdown, }; -static int __devinit grover_beep_probe(struct platform_device *op) +static int grover_beep_probe(struct platform_device *op) { struct sparcspkr_state *state; struct grover_beep_info *info; @@ -310,7 +310,7 @@ out_err: return err; } -static int __devexit grover_remove(struct platform_device *op) +static int grover_remove(struct platform_device *op) { struct sparcspkr_state *state = dev_get_drvdata(&op->dev); struct grover_beep_info *info = &state->u.grover; @@ -345,7 +345,7 @@ static struct platform_driver grover_beep_driver = { .of_match_table = grover_beep_match, }, .probe = grover_beep_probe, - .remove = __devexit_p(grover_remove), + .remove = grover_remove, .shutdown = sparcspkr_shutdown, }; diff --git a/drivers/input/misc/twl4030-pwrbutton.c b/drivers/input/misc/twl4030-pwrbutton.c index b3dd96d6448b..27c2bc8aa890 100644 --- a/drivers/input/misc/twl4030-pwrbutton.c +++ b/drivers/input/misc/twl4030-pwrbutton.c @@ -39,8 +39,7 @@ static irqreturn_t powerbutton_irq(int irq, void *_pwr) int err; u8 value; - err = twl_i2c_read_u8(TWL4030_MODULE_PM_MASTER, &value, - STS_HW_CONDITIONS); + err = twl_i2c_read_u8(TWL_MODULE_PM_MASTER, &value, STS_HW_CONDITIONS); if (!err) { pm_wakeup_event(pwr->dev.parent, 0); input_report_key(pwr, KEY_POWER, value & PWR_PWRON_IRQ); diff --git a/drivers/input/misc/twl4030-vibra.c b/drivers/input/misc/twl4030-vibra.c index 2194a3c7236a..78eb6b30580a 100644 --- a/drivers/input/misc/twl4030-vibra.c +++ b/drivers/input/misc/twl4030-vibra.c @@ -207,7 +207,7 @@ static bool twl4030_vibra_check_coexist(struct twl4030_vibra_data *pdata, return false; } -static int __devinit twl4030_vibra_probe(struct platform_device *pdev) +static int twl4030_vibra_probe(struct platform_device *pdev) { struct twl4030_vibra_data *pdata = pdev->dev.platform_data; struct device_node *twl4030_core_node = pdev->dev.parent->of_node; @@ -269,7 +269,7 @@ err_kzalloc: return ret; } -static int __devexit twl4030_vibra_remove(struct platform_device *pdev) +static int twl4030_vibra_remove(struct platform_device *pdev) { struct vibra_info *info = platform_get_drvdata(pdev); @@ -283,7 +283,7 @@ static int __devexit twl4030_vibra_remove(struct platform_device *pdev) static struct platform_driver twl4030_vibra_driver = { .probe = twl4030_vibra_probe, - .remove = __devexit_p(twl4030_vibra_remove), + .remove = twl4030_vibra_remove, .driver = { .name = "twl4030-vibra", .owner = THIS_MODULE, diff --git a/drivers/input/misc/twl6040-vibra.c b/drivers/input/misc/twl6040-vibra.c index c8a288ae1d5b..71a28ee699f3 100644 --- a/drivers/input/misc/twl6040-vibra.c +++ b/drivers/input/misc/twl6040-vibra.c @@ -255,7 +255,7 @@ static int twl6040_vibra_suspend(struct device *dev) static SIMPLE_DEV_PM_OPS(twl6040_vibra_pm_ops, twl6040_vibra_suspend, NULL); -static int __devinit twl6040_vibra_probe(struct platform_device *pdev) +static int twl6040_vibra_probe(struct platform_device *pdev) { struct twl6040_vibra_data *pdata = pdev->dev.platform_data; struct device *twl6040_core_dev = pdev->dev.parent; @@ -418,7 +418,7 @@ err_kzalloc: return ret; } -static int __devexit twl6040_vibra_remove(struct platform_device *pdev) +static int twl6040_vibra_remove(struct platform_device *pdev) { struct vibra_info *info = platform_get_drvdata(pdev); @@ -433,7 +433,7 @@ static int __devexit twl6040_vibra_remove(struct platform_device *pdev) static struct platform_driver twl6040_vibra_driver = { .probe = twl6040_vibra_probe, - .remove = __devexit_p(twl6040_vibra_remove), + .remove = twl6040_vibra_remove, .driver = { .name = "twl6040-vibra", .owner = THIS_MODULE, diff --git a/drivers/input/misc/wistron_btns.c b/drivers/input/misc/wistron_btns.c index e2bdfd4bea70..56536f4b9572 100644 --- a/drivers/input/misc/wistron_btns.c +++ b/drivers/input/misc/wistron_btns.c @@ -170,7 +170,7 @@ static u16 bios_pop_queue(void) return regs.eax; } -static void __devinit bios_attach(void) +static void bios_attach(void) { struct regs regs; @@ -190,7 +190,7 @@ static void bios_detach(void) call_bios(®s); } -static u8 __devinit bios_get_cmos_address(void) +static u8 bios_get_cmos_address(void) { struct regs regs; @@ -202,7 +202,7 @@ static u8 __devinit bios_get_cmos_address(void) return regs.ecx; } -static u16 __devinit bios_get_default_setting(u8 subsys) +static u16 bios_get_default_setting(u8 subsys) { struct regs regs; @@ -1052,7 +1052,7 @@ static struct led_classdev wistron_wifi_led = { .brightness_set = wistron_wifi_led_set, }; -static void __devinit wistron_led_init(struct device *parent) +static void wistron_led_init(struct device *parent) { if (leds_present & FE_WIFI_LED) { u16 wifi = bios_get_default_setting(WIFI); @@ -1077,7 +1077,7 @@ static void __devinit wistron_led_init(struct device *parent) } } -static void __devexit wistron_led_remove(void) +static void wistron_led_remove(void) { if (leds_present & FE_MAIL_LED) led_classdev_unregister(&wistron_mail_led); @@ -1168,7 +1168,7 @@ static void wistron_poll(struct input_polled_dev *dev) dev->poll_interval = POLL_INTERVAL_DEFAULT; } -static int __devinit wistron_setup_keymap(struct input_dev *dev, +static int wistron_setup_keymap(struct input_dev *dev, struct key_entry *entry) { switch (entry->type) { @@ -1199,7 +1199,7 @@ static int __devinit wistron_setup_keymap(struct input_dev *dev, return 0; } -static int __devinit setup_input_dev(void) +static int setup_input_dev(void) { struct input_dev *input_dev; int error; @@ -1237,7 +1237,7 @@ static int __devinit setup_input_dev(void) /* Driver core */ -static int __devinit wistron_probe(struct platform_device *dev) +static int wistron_probe(struct platform_device *dev) { int err; @@ -1277,7 +1277,7 @@ static int __devinit wistron_probe(struct platform_device *dev) return 0; } -static int __devexit wistron_remove(struct platform_device *dev) +static int wistron_remove(struct platform_device *dev) { wistron_led_remove(); input_unregister_polled_device(wistron_idev); @@ -1334,7 +1334,7 @@ static struct platform_driver wistron_driver = { #endif }, .probe = wistron_probe, - .remove = __devexit_p(wistron_remove), + .remove = wistron_remove, }; static int __init wb_module_init(void) diff --git a/drivers/input/misc/wm831x-on.c b/drivers/input/misc/wm831x-on.c index 6790a812a1db..558767d8ebf4 100644 --- a/drivers/input/misc/wm831x-on.c +++ b/drivers/input/misc/wm831x-on.c @@ -69,14 +69,15 @@ static irqreturn_t wm831x_on_irq(int irq, void *data) return IRQ_HANDLED; } -static int __devinit wm831x_on_probe(struct platform_device *pdev) +static int wm831x_on_probe(struct platform_device *pdev) { struct wm831x *wm831x = dev_get_drvdata(pdev->dev.parent); struct wm831x_on *wm831x_on; int irq = wm831x_irq(wm831x, platform_get_irq(pdev, 0)); int ret; - wm831x_on = kzalloc(sizeof(struct wm831x_on), GFP_KERNEL); + wm831x_on = devm_kzalloc(&pdev->dev, sizeof(struct wm831x_on), + GFP_KERNEL); if (!wm831x_on) { dev_err(&pdev->dev, "Can't allocate data\n"); return -ENOMEM; @@ -120,11 +121,10 @@ err_irq: err_input_dev: input_free_device(wm831x_on->dev); err: - kfree(wm831x_on); return ret; } -static int __devexit wm831x_on_remove(struct platform_device *pdev) +static int wm831x_on_remove(struct platform_device *pdev) { struct wm831x_on *wm831x_on = platform_get_drvdata(pdev); int irq = platform_get_irq(pdev, 0); @@ -132,14 +132,13 @@ static int __devexit wm831x_on_remove(struct platform_device *pdev) free_irq(irq, wm831x_on); cancel_delayed_work_sync(&wm831x_on->work); input_unregister_device(wm831x_on->dev); - kfree(wm831x_on); return 0; } static struct platform_driver wm831x_on_driver = { .probe = wm831x_on_probe, - .remove = __devexit_p(wm831x_on_remove), + .remove = wm831x_on_remove, .driver = { .name = "wm831x-on", .owner = THIS_MODULE, diff --git a/drivers/input/misc/xen-kbdfront.c b/drivers/input/misc/xen-kbdfront.c index 6f7d99013031..e21c1816a8f9 100644 --- a/drivers/input/misc/xen-kbdfront.c +++ b/drivers/input/misc/xen-kbdfront.c @@ -104,7 +104,7 @@ static irqreturn_t input_handler(int rq, void *dev_id) return IRQ_HANDLED; } -static int __devinit xenkbd_probe(struct xenbus_device *dev, +static int xenkbd_probe(struct xenbus_device *dev, const struct xenbus_device_id *id) { int ret, i, abs; diff --git a/drivers/input/mouse/alps.c b/drivers/input/mouse/alps.c index cf5af1f495ec..e229fa3cad96 100644 --- a/drivers/input/mouse/alps.c +++ b/drivers/input/mouse/alps.c @@ -767,9 +767,8 @@ static psmouse_ret_t alps_handle_interleaved_ps2(struct psmouse *psmouse) psmouse->packet[5]) & 0x80) || (!alps_is_valid_first_byte(priv->i, psmouse->packet[6]))) { psmouse_dbg(psmouse, - "refusing packet %x %x %x %x (suspected interleaved ps/2)\n", - psmouse->packet[3], psmouse->packet[4], - psmouse->packet[5], psmouse->packet[6]); + "refusing packet %4ph (suspected interleaved ps/2)\n", + psmouse->packet + 3); return PSMOUSE_BAD_DATA; } @@ -831,9 +830,8 @@ static void alps_flush_packet(unsigned long data) psmouse->packet[4] | psmouse->packet[5]) & 0x80) { psmouse_dbg(psmouse, - "refusing packet %x %x %x (suspected interleaved ps/2)\n", - psmouse->packet[3], psmouse->packet[4], - psmouse->packet[5]); + "refusing packet %3ph (suspected interleaved ps/2)\n", + psmouse->packet + 3); } else { alps_process_packet(psmouse); } diff --git a/drivers/input/mouse/gpio_mouse.c b/drivers/input/mouse/gpio_mouse.c index 39fe9b737cae..532eaca4cc56 100644 --- a/drivers/input/mouse/gpio_mouse.c +++ b/drivers/input/mouse/gpio_mouse.c @@ -46,7 +46,7 @@ static void gpio_mouse_scan(struct input_polled_dev *dev) input_sync(input); } -static int __devinit gpio_mouse_probe(struct platform_device *pdev) +static int gpio_mouse_probe(struct platform_device *pdev) { struct gpio_mouse_platform_data *pdata = pdev->dev.platform_data; struct input_polled_dev *input_poll; @@ -150,7 +150,7 @@ static int __devinit gpio_mouse_probe(struct platform_device *pdev) return error; } -static int __devexit gpio_mouse_remove(struct platform_device *pdev) +static int gpio_mouse_remove(struct platform_device *pdev) { struct input_polled_dev *input = platform_get_drvdata(pdev); struct gpio_mouse_platform_data *pdata = input->private; @@ -172,7 +172,7 @@ static int __devexit gpio_mouse_remove(struct platform_device *pdev) static struct platform_driver gpio_mouse_device_driver = { .probe = gpio_mouse_probe, - .remove = __devexit_p(gpio_mouse_remove), + .remove = gpio_mouse_remove, .driver = { .name = "gpio_mouse", .owner = THIS_MODULE, diff --git a/drivers/input/mouse/maplemouse.c b/drivers/input/mouse/maplemouse.c index 5f278176eb9b..0a60717b91c6 100644 --- a/drivers/input/mouse/maplemouse.c +++ b/drivers/input/mouse/maplemouse.c @@ -64,7 +64,7 @@ static void dc_mouse_close(struct input_dev *dev) } /* allow the mouse to be used */ -static int __devinit probe_maple_mouse(struct device *dev) +static int probe_maple_mouse(struct device *dev) { struct maple_device *mdev = to_maple_dev(dev); struct maple_driver *mdrv = to_maple_driver(dev->driver); @@ -114,7 +114,7 @@ fail: return error; } -static int __devexit remove_maple_mouse(struct device *dev) +static int remove_maple_mouse(struct device *dev) { struct maple_device *mdev = to_maple_dev(dev); struct dc_mouse *mse = maple_get_drvdata(mdev); @@ -132,7 +132,7 @@ static struct maple_driver dc_mouse_driver = { .drv = { .name = "Dreamcast_mouse", .probe = probe_maple_mouse, - .remove = __devexit_p(remove_maple_mouse), + .remove = remove_maple_mouse, }, }; diff --git a/drivers/input/mouse/navpoint.c b/drivers/input/mouse/navpoint.c index c29ae7654d5e..8e1b98ea5648 100644 --- a/drivers/input/mouse/navpoint.c +++ b/drivers/input/mouse/navpoint.c @@ -206,7 +206,7 @@ static void navpoint_close(struct input_dev *input) navpoint_down(navpoint); } -static int __devinit navpoint_probe(struct platform_device *pdev) +static int navpoint_probe(struct platform_device *pdev) { const struct navpoint_platform_data *pdata = dev_get_platdata(&pdev->dev); @@ -299,7 +299,7 @@ err_free_gpio: return error; } -static int __devexit navpoint_remove(struct platform_device *pdev) +static int navpoint_remove(struct platform_device *pdev) { const struct navpoint_platform_data *pdata = dev_get_platdata(&pdev->dev); @@ -353,7 +353,7 @@ static SIMPLE_DEV_PM_OPS(navpoint_pm_ops, navpoint_suspend, navpoint_resume); static struct platform_driver navpoint_driver = { .probe = navpoint_probe, - .remove = __devexit_p(navpoint_remove), + .remove = navpoint_remove, .driver = { .name = "navpoint", .owner = THIS_MODULE, diff --git a/drivers/input/mouse/pxa930_trkball.c b/drivers/input/mouse/pxa930_trkball.c index 4fe055f2c536..0ecb9e7945eb 100644 --- a/drivers/input/mouse/pxa930_trkball.c +++ b/drivers/input/mouse/pxa930_trkball.c @@ -143,7 +143,7 @@ static void pxa930_trkball_close(struct input_dev *dev) pxa930_trkball_disable(trkball); } -static int __devinit pxa930_trkball_probe(struct platform_device *pdev) +static int pxa930_trkball_probe(struct platform_device *pdev) { struct pxa930_trkball *trkball; struct input_dev *input; @@ -230,7 +230,7 @@ failed: return error; } -static int __devexit pxa930_trkball_remove(struct platform_device *pdev) +static int pxa930_trkball_remove(struct platform_device *pdev) { struct pxa930_trkball *trkball = platform_get_drvdata(pdev); int irq = platform_get_irq(pdev, 0); @@ -248,7 +248,7 @@ static struct platform_driver pxa930_trkball_driver = { .name = "pxa930-trkball", }, .probe = pxa930_trkball_probe, - .remove = __devexit_p(pxa930_trkball_remove), + .remove = pxa930_trkball_remove, }; module_platform_driver(pxa930_trkball_driver); diff --git a/drivers/input/mouse/synaptics_i2c.c b/drivers/input/mouse/synaptics_i2c.c index 063a174d3a88..ad822608f6ee 100644 --- a/drivers/input/mouse/synaptics_i2c.c +++ b/drivers/input/mouse/synaptics_i2c.c @@ -535,7 +535,7 @@ static struct synaptics_i2c *synaptics_i2c_touch_create(struct i2c_client *clien return touch; } -static int __devinit synaptics_i2c_probe(struct i2c_client *client, +static int synaptics_i2c_probe(struct i2c_client *client, const struct i2c_device_id *dev_id) { int ret; @@ -601,7 +601,7 @@ err_mem_free: return ret; } -static int __devexit synaptics_i2c_remove(struct i2c_client *client) +static int synaptics_i2c_remove(struct i2c_client *client) { struct synaptics_i2c *touch = i2c_get_clientdata(client); @@ -662,7 +662,7 @@ static struct i2c_driver synaptics_i2c_driver = { }, .probe = synaptics_i2c_probe, - .remove = __devexit_p(synaptics_i2c_remove), + .remove = synaptics_i2c_remove, .id_table = synaptics_i2c_id_table, }; diff --git a/drivers/input/serio/Kconfig b/drivers/input/serio/Kconfig index 55f2c2293ec6..4a4e182c33e7 100644 --- a/drivers/input/serio/Kconfig +++ b/drivers/input/serio/Kconfig @@ -234,4 +234,13 @@ config SERIO_PS2MULT To compile this driver as a module, choose M here: the module will be called ps2mult. +config SERIO_ARC_PS2 + tristate "ARC PS/2 support" + help + Say Y here if you have an ARC FPGA platform with a PS/2 + controller in it. + + To compile this driver as a module, choose M here; the module + will be called arc_ps2. + endif diff --git a/drivers/input/serio/Makefile b/drivers/input/serio/Makefile index dbbe37616c92..4b0c8f84f1c1 100644 --- a/drivers/input/serio/Makefile +++ b/drivers/input/serio/Makefile @@ -25,3 +25,4 @@ obj-$(CONFIG_SERIO_RAW) += serio_raw.o obj-$(CONFIG_SERIO_AMS_DELTA) += ams_delta_serio.o obj-$(CONFIG_SERIO_XILINX_XPS_PS2) += xilinx_ps2.o obj-$(CONFIG_SERIO_ALTERA_PS2) += altera_ps2.o +obj-$(CONFIG_SERIO_ARC_PS2) += arc_ps2.o diff --git a/drivers/input/serio/altera_ps2.c b/drivers/input/serio/altera_ps2.c index cc11f4efe119..479ce5fe8955 100644 --- a/drivers/input/serio/altera_ps2.c +++ b/drivers/input/serio/altera_ps2.c @@ -81,7 +81,7 @@ static void altera_ps2_close(struct serio *io) /* * Add one device to this driver. */ -static int __devinit altera_ps2_probe(struct platform_device *pdev) +static int altera_ps2_probe(struct platform_device *pdev) { struct ps2if *ps2if; struct serio *serio; @@ -159,7 +159,7 @@ static int __devinit altera_ps2_probe(struct platform_device *pdev) /* * Remove one device from this driver. */ -static int __devexit altera_ps2_remove(struct platform_device *pdev) +static int altera_ps2_remove(struct platform_device *pdev) { struct ps2if *ps2if = platform_get_drvdata(pdev); @@ -187,7 +187,7 @@ MODULE_DEVICE_TABLE(of, altera_ps2_match); */ static struct platform_driver altera_ps2_driver = { .probe = altera_ps2_probe, - .remove = __devexit_p(altera_ps2_remove), + .remove = altera_ps2_remove, .driver = { .name = DRV_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/serio/ambakmi.c b/drivers/input/serio/ambakmi.c index 2e77246c2e5a..4e2fd44865e1 100644 --- a/drivers/input/serio/ambakmi.c +++ b/drivers/input/serio/ambakmi.c @@ -107,7 +107,7 @@ static void amba_kmi_close(struct serio *io) clk_disable_unprepare(kmi->clk); } -static int __devinit amba_kmi_probe(struct amba_device *dev, +static int amba_kmi_probe(struct amba_device *dev, const struct amba_id *id) { struct amba_kmi_port *kmi; @@ -163,7 +163,7 @@ static int __devinit amba_kmi_probe(struct amba_device *dev, return ret; } -static int __devexit amba_kmi_remove(struct amba_device *dev) +static int amba_kmi_remove(struct amba_device *dev) { struct amba_kmi_port *kmi = amba_get_drvdata(dev); @@ -204,7 +204,7 @@ static struct amba_driver ambakmi_driver = { }, .id_table = amba_kmi_idtable, .probe = amba_kmi_probe, - .remove = __devexit_p(amba_kmi_remove), + .remove = amba_kmi_remove, .resume = amba_kmi_resume, }; diff --git a/drivers/input/serio/arc_ps2.c b/drivers/input/serio/arc_ps2.c new file mode 100644 index 000000000000..b571eb3e4efc --- /dev/null +++ b/drivers/input/serio/arc_ps2.c @@ -0,0 +1,274 @@ +/* + * Copyright (C) 2004, 2007-2010, 2011-2012 Synopsys, Inc. (www.synopsys.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Driver is originally developed by Pavel Sokolov <psokolov@synopsys.com> + */ + +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/input.h> +#include <linux/serio.h> +#include <linux/platform_device.h> +#include <linux/io.h> +#include <linux/kernel.h> +#include <linux/slab.h> + +#define ARC_PS2_PORTS 2 + +#define ARC_ARC_PS2_ID 0x0001f609 + +#define STAT_TIMEOUT 128 + +#define PS2_STAT_RX_FRM_ERR (1) +#define PS2_STAT_RX_BUF_OVER (1 << 1) +#define PS2_STAT_RX_INT_EN (1 << 2) +#define PS2_STAT_RX_VAL (1 << 3) +#define PS2_STAT_TX_ISNOT_FUL (1 << 4) +#define PS2_STAT_TX_INT_EN (1 << 5) + +struct arc_ps2_port { + void __iomem *data_addr; + void __iomem *status_addr; + struct serio *io; +}; + +struct arc_ps2_data { + struct arc_ps2_port port[ARC_PS2_PORTS]; + void __iomem *addr; + unsigned int frame_error; + unsigned int buf_overflow; + unsigned int total_int; +}; + +static void arc_ps2_check_rx(struct arc_ps2_data *arc_ps2, + struct arc_ps2_port *port) +{ + unsigned int timeout = 1000; + unsigned int flag, status; + unsigned char data; + + do { + status = ioread32(port->status_addr); + if (!(status & PS2_STAT_RX_VAL)) + return; + + data = ioread32(port->data_addr) & 0xff; + + flag = 0; + arc_ps2->total_int++; + if (status & PS2_STAT_RX_FRM_ERR) { + arc_ps2->frame_error++; + flag |= SERIO_PARITY; + } else if (status & PS2_STAT_RX_BUF_OVER) { + arc_ps2->buf_overflow++; + flag |= SERIO_FRAME; + } + + serio_interrupt(port->io, data, flag); + } while (--timeout); + + dev_err(&port->io->dev, "PS/2 hardware stuck\n"); +} + +static irqreturn_t arc_ps2_interrupt(int irq, void *dev) +{ + struct arc_ps2_data *arc_ps2 = dev; + int i; + + for (i = 0; i < ARC_PS2_PORTS; i++) + arc_ps2_check_rx(arc_ps2, &arc_ps2->port[i]); + + return IRQ_HANDLED; +} + +static int arc_ps2_write(struct serio *io, unsigned char val) +{ + unsigned status; + struct arc_ps2_port *port = io->port_data; + int timeout = STAT_TIMEOUT; + + do { + status = ioread32(port->status_addr); + cpu_relax(); + + if (status & PS2_STAT_TX_ISNOT_FUL) { + iowrite32(val & 0xff, port->data_addr); + return 0; + } + + } while (--timeout); + + dev_err(&io->dev, "write timeout\n"); + return -ETIMEDOUT; +} + +static int arc_ps2_open(struct serio *io) +{ + struct arc_ps2_port *port = io->port_data; + + iowrite32(PS2_STAT_RX_INT_EN, port->status_addr); + + return 0; +} + +static void arc_ps2_close(struct serio *io) +{ + struct arc_ps2_port *port = io->port_data; + + iowrite32(ioread32(port->status_addr) & ~PS2_STAT_RX_INT_EN, + port->status_addr); +} + +static void __iomem *arc_ps2_calc_addr(struct arc_ps2_data *arc_ps2, + int index, bool status) +{ + void __iomem *addr; + + addr = arc_ps2->addr + 4 + 4 * index; + if (status) + addr += ARC_PS2_PORTS * 4; + + return addr; +} + +static void arc_ps2_inhibit_ports(struct arc_ps2_data *arc_ps2) +{ + void __iomem *addr; + u32 val; + int i; + + for (i = 0; i < ARC_PS2_PORTS; i++) { + addr = arc_ps2_calc_addr(arc_ps2, i, true); + val = ioread32(addr); + val &= ~(PS2_STAT_RX_INT_EN | PS2_STAT_TX_INT_EN); + iowrite32(val, addr); + } +} + +static int arc_ps2_create_port(struct platform_device *pdev, + struct arc_ps2_data *arc_ps2, + int index) +{ + struct arc_ps2_port *port = &arc_ps2->port[index]; + struct serio *io; + + io = kzalloc(sizeof(struct serio), GFP_KERNEL); + if (!io) + return -ENOMEM; + + io->id.type = SERIO_8042; + io->write = arc_ps2_write; + io->open = arc_ps2_open; + io->close = arc_ps2_close; + snprintf(io->name, sizeof(io->name), "ARC PS/2 port%d", index); + snprintf(io->phys, sizeof(io->phys), "arc/serio%d", index); + io->port_data = port; + + port->io = io; + + port->data_addr = arc_ps2_calc_addr(arc_ps2, index, false); + port->status_addr = arc_ps2_calc_addr(arc_ps2, index, true); + + dev_dbg(&pdev->dev, "port%d is allocated (data = 0x%p, status = 0x%p)\n", + index, port->data_addr, port->status_addr); + + serio_register_port(port->io); + return 0; +} + +static int arc_ps2_probe(struct platform_device *pdev) +{ + struct arc_ps2_data *arc_ps2; + struct resource *res; + int irq; + int error, id, i; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) { + dev_err(&pdev->dev, "no IO memory defined\n"); + return -EINVAL; + } + + irq = platform_get_irq_byname(pdev, "arc_ps2_irq"); + if (irq < 0) { + dev_err(&pdev->dev, "no IRQ defined\n"); + return -EINVAL; + } + + arc_ps2 = devm_kzalloc(&pdev->dev, sizeof(struct arc_ps2_data), + GFP_KERNEL); + if (!arc_ps2) { + dev_err(&pdev->dev, "out of memory\n"); + return -ENOMEM; + } + + arc_ps2->addr = devm_request_and_ioremap(&pdev->dev, res); + if (!arc_ps2->addr) + return -EBUSY; + + dev_info(&pdev->dev, "irq = %d, address = 0x%p, ports = %i\n", + irq, arc_ps2->addr, ARC_PS2_PORTS); + + id = ioread32(arc_ps2->addr); + if (id != ARC_ARC_PS2_ID) { + dev_err(&pdev->dev, "device id does not match\n"); + return -ENXIO; + } + + arc_ps2_inhibit_ports(arc_ps2); + + error = devm_request_irq(&pdev->dev, irq, arc_ps2_interrupt, + 0, "arc_ps2", arc_ps2); + if (error) { + dev_err(&pdev->dev, "Could not allocate IRQ\n"); + return error; + } + + for (i = 0; i < ARC_PS2_PORTS; i++) { + error = arc_ps2_create_port(pdev, arc_ps2, i); + if (error) { + while (--i >= 0) + serio_unregister_port(arc_ps2->port[i].io); + return error; + } + } + + platform_set_drvdata(pdev, arc_ps2); + + return 0; +} + +static int arc_ps2_remove(struct platform_device *pdev) +{ + struct arc_ps2_data *arc_ps2 = platform_get_drvdata(pdev); + int i; + + for (i = 0; i < ARC_PS2_PORTS; i++) + serio_unregister_port(arc_ps2->port[i].io); + + dev_dbg(&pdev->dev, "interrupt count = %i\n", arc_ps2->total_int); + dev_dbg(&pdev->dev, "frame error count = %i\n", arc_ps2->frame_error); + dev_dbg(&pdev->dev, "buffer overflow count = %i\n", + arc_ps2->buf_overflow); + + return 0; +} + +static struct platform_driver arc_ps2_driver = { + .driver = { + .name = "arc_ps2", + .owner = THIS_MODULE, + }, + .probe = arc_ps2_probe, + .remove = arc_ps2_remove, +}; + +module_platform_driver(arc_ps2_driver); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Pavel Sokolov <psokolov@synopsys.com>"); +MODULE_DESCRIPTION("ARC PS/2 Driver"); diff --git a/drivers/input/serio/ct82c710.c b/drivers/input/serio/ct82c710.c index 852816567241..cfe549d4eaa5 100644 --- a/drivers/input/serio/ct82c710.c +++ b/drivers/input/serio/ct82c710.c @@ -175,7 +175,7 @@ static int __init ct82c710_detect(void) return 0; } -static int __devinit ct82c710_probe(struct platform_device *dev) +static int ct82c710_probe(struct platform_device *dev) { ct82c710_port = kzalloc(sizeof(struct serio), GFP_KERNEL); if (!ct82c710_port) @@ -199,7 +199,7 @@ static int __devinit ct82c710_probe(struct platform_device *dev) return 0; } -static int __devexit ct82c710_remove(struct platform_device *dev) +static int ct82c710_remove(struct platform_device *dev) { serio_unregister_port(ct82c710_port); @@ -212,7 +212,7 @@ static struct platform_driver ct82c710_driver = { .owner = THIS_MODULE, }, .probe = ct82c710_probe, - .remove = __devexit_p(ct82c710_remove), + .remove = ct82c710_remove, }; diff --git a/drivers/input/serio/gscps2.c b/drivers/input/serio/gscps2.c index 4225f5d6b15f..8d9ba0c3827c 100644 --- a/drivers/input/serio/gscps2.c +++ b/drivers/input/serio/gscps2.c @@ -327,7 +327,7 @@ static void gscps2_close(struct serio *port) * @return: success/error report */ -static int __devinit gscps2_probe(struct parisc_device *dev) +static int gscps2_probe(struct parisc_device *dev) { struct gscps2port *ps2port; struct serio *serio; @@ -414,7 +414,7 @@ fail_nomem: * @return: success/error report */ -static int __devexit gscps2_remove(struct parisc_device *dev) +static int gscps2_remove(struct parisc_device *dev) { struct gscps2port *ps2port = dev_get_drvdata(&dev->dev); @@ -444,7 +444,7 @@ static struct parisc_driver parisc_ps2_driver = { .name = "gsc_ps2", .id_table = gscps2_device_tbl, .probe = gscps2_probe, - .remove = __devexit_p(gscps2_remove), + .remove = gscps2_remove, }; static int __init gscps2_init(void) diff --git a/drivers/input/serio/hil_mlc.c b/drivers/input/serio/hil_mlc.c index bfd3865d886b..65605e4ef3cf 100644 --- a/drivers/input/serio/hil_mlc.c +++ b/drivers/input/serio/hil_mlc.c @@ -686,13 +686,12 @@ static int hilse_donode(hil_mlc *mlc) write_lock_irqsave(&mlc->lock, flags); pack = node->object.packet; out: - if (mlc->istarted) - goto out2; - /* Prepare to receive input */ - if ((node + 1)->act & HILSE_IN) - hilse_setup_input(mlc, node + 1); + if (!mlc->istarted) { + /* Prepare to receive input */ + if ((node + 1)->act & HILSE_IN) + hilse_setup_input(mlc, node + 1); + } - out2: write_unlock_irqrestore(&mlc->lock, flags); if (down_trylock(&mlc->osem)) { @@ -1010,8 +1009,6 @@ static int __init hil_mlc_init(void) static void __exit hil_mlc_exit(void) { del_timer_sync(&hil_mlcs_kicker); - - tasklet_disable(&hil_mlcs_tasklet); tasklet_kill(&hil_mlcs_tasklet); } diff --git a/drivers/input/serio/i8042-io.h b/drivers/input/serio/i8042-io.h index 5d48bb66aa73..a5eed2ade53d 100644 --- a/drivers/input/serio/i8042-io.h +++ b/drivers/input/serio/i8042-io.h @@ -76,7 +76,7 @@ static inline int i8042_platform_init(void) if (check_legacy_ioport(I8042_DATA_REG)) return -ENODEV; #endif -#if !defined(__sh__) && !defined(__alpha__) && !defined(__mips__) +#if !defined(__sh__) && !defined(__alpha__) if (!request_region(I8042_DATA_REG, 16, "i8042")) return -EBUSY; #endif diff --git a/drivers/input/serio/i8042-sparcio.h b/drivers/input/serio/i8042-sparcio.h index 395a9af3adcd..d6aa4c67dbb6 100644 --- a/drivers/input/serio/i8042-sparcio.h +++ b/drivers/input/serio/i8042-sparcio.h @@ -49,7 +49,7 @@ static inline void i8042_write_command(int val) #define OBP_PS2MS_NAME1 "kdmouse" #define OBP_PS2MS_NAME2 "mouse" -static int __devinit sparc_i8042_probe(struct platform_device *op) +static int sparc_i8042_probe(struct platform_device *op) { struct device_node *dp = op->dev.of_node; @@ -80,7 +80,7 @@ static int __devinit sparc_i8042_probe(struct platform_device *op) return 0; } -static int __devexit sparc_i8042_remove(struct platform_device *op) +static int sparc_i8042_remove(struct platform_device *op) { of_iounmap(kbd_res, kbd_iobase, 8); @@ -102,7 +102,7 @@ static struct platform_driver sparc_i8042_driver = { .of_match_table = sparc_i8042_match, }, .probe = sparc_i8042_probe, - .remove = __devexit_p(sparc_i8042_remove), + .remove = sparc_i8042_remove, }; static int __init i8042_platform_init(void) diff --git a/drivers/input/serio/i8042-x86ia64io.h b/drivers/input/serio/i8042-x86ia64io.h index d6cc77a53c7e..5f306f79da0c 100644 --- a/drivers/input/serio/i8042-x86ia64io.h +++ b/drivers/input/serio/i8042-x86ia64io.h @@ -921,6 +921,7 @@ static int __init i8042_platform_init(void) int retval; #ifdef CONFIG_X86 + u8 a20_on = 0xdf; /* Just return if pre-detection shows no i8042 controller exist */ if (!x86_platform.i8042_detect()) return -ENODEV; @@ -960,6 +961,14 @@ static int __init i8042_platform_init(void) if (dmi_check_system(i8042_dmi_dritek_table)) i8042_dritek = true; + + /* + * A20 was already enabled during early kernel init. But some buggy + * BIOSes (in MSI Laptops) require A20 to be enabled using 8042 to + * resume from S3. So we do it here and hope that nothing breaks. + */ + i8042_command(&a20_on, 0x10d1); + i8042_command(NULL, 0x00ff); /* Null command for SMM firmware */ #endif /* CONFIG_X86 */ return retval; diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c index 86564414b75a..78e4de42efaa 100644 --- a/drivers/input/serio/i8042.c +++ b/drivers/input/serio/i8042.c @@ -1284,7 +1284,7 @@ static void __init i8042_register_ports(void) } } -static void __devexit i8042_unregister_ports(void) +static void i8042_unregister_ports(void) { int i; @@ -1437,7 +1437,7 @@ static int __init i8042_probe(struct platform_device *dev) return error; } -static int __devexit i8042_remove(struct platform_device *dev) +static int i8042_remove(struct platform_device *dev) { i8042_unregister_ports(); i8042_free_irqs(); @@ -1455,7 +1455,7 @@ static struct platform_driver i8042_driver = { .pm = &i8042_pm_ops, #endif }, - .remove = __devexit_p(i8042_remove), + .remove = i8042_remove, .shutdown = i8042_shutdown, }; diff --git a/drivers/input/serio/maceps2.c b/drivers/input/serio/maceps2.c index 61da763b1209..bc85e1cc66d8 100644 --- a/drivers/input/serio/maceps2.c +++ b/drivers/input/serio/maceps2.c @@ -116,7 +116,7 @@ static void maceps2_close(struct serio *dev) } -static struct serio * __devinit maceps2_allocate_port(int idx) +static struct serio *maceps2_allocate_port(int idx) { struct serio *serio; @@ -135,7 +135,7 @@ static struct serio * __devinit maceps2_allocate_port(int idx) return serio; } -static int __devinit maceps2_probe(struct platform_device *dev) +static int maceps2_probe(struct platform_device *dev) { maceps2_port[0] = maceps2_allocate_port(0); maceps2_port[1] = maceps2_allocate_port(1); @@ -151,7 +151,7 @@ static int __devinit maceps2_probe(struct platform_device *dev) return 0; } -static int __devexit maceps2_remove(struct platform_device *dev) +static int maceps2_remove(struct platform_device *dev) { serio_unregister_port(maceps2_port[0]); serio_unregister_port(maceps2_port[1]); @@ -165,7 +165,7 @@ static struct platform_driver maceps2_driver = { .owner = THIS_MODULE, }, .probe = maceps2_probe, - .remove = __devexit_p(maceps2_remove), + .remove = maceps2_remove, }; static int __init maceps2_init(void) diff --git a/drivers/input/serio/pcips2.c b/drivers/input/serio/pcips2.c index 0c42497aaaf4..76f83836fd5a 100644 --- a/drivers/input/serio/pcips2.c +++ b/drivers/input/serio/pcips2.c @@ -127,7 +127,7 @@ static void pcips2_close(struct serio *io) free_irq(ps2if->dev->irq, ps2if); } -static int __devinit pcips2_probe(struct pci_dev *dev, const struct pci_device_id *id) +static int pcips2_probe(struct pci_dev *dev, const struct pci_device_id *id) { struct pcips2_data *ps2if; struct serio *serio; @@ -176,7 +176,7 @@ static int __devinit pcips2_probe(struct pci_dev *dev, const struct pci_device_i return ret; } -static void __devexit pcips2_remove(struct pci_dev *dev) +static void pcips2_remove(struct pci_dev *dev) { struct pcips2_data *ps2if = pci_get_drvdata(dev); @@ -212,7 +212,7 @@ static struct pci_driver pcips2_driver = { .name = "pcips2", .id_table = pcips2_ids, .probe = pcips2_probe, - .remove = __devexit_p(pcips2_remove), + .remove = pcips2_remove, }; module_pci_driver(pcips2_driver); diff --git a/drivers/input/serio/q40kbd.c b/drivers/input/serio/q40kbd.c index 0c0df7f73802..70fe542839fb 100644 --- a/drivers/input/serio/q40kbd.c +++ b/drivers/input/serio/q40kbd.c @@ -122,7 +122,7 @@ static void q40kbd_close(struct serio *port) q40kbd_flush(q40kbd); } -static int __devinit q40kbd_probe(struct platform_device *pdev) +static int q40kbd_probe(struct platform_device *pdev) { struct q40kbd *q40kbd; struct serio *port; @@ -168,7 +168,7 @@ err_free_mem: return error; } -static int __devexit q40kbd_remove(struct platform_device *pdev) +static int q40kbd_remove(struct platform_device *pdev) { struct q40kbd *q40kbd = platform_get_drvdata(pdev); @@ -190,7 +190,7 @@ static struct platform_driver q40kbd_driver = { .name = "q40kbd", .owner = THIS_MODULE, }, - .remove = __devexit_p(q40kbd_remove), + .remove = q40kbd_remove, }; static int __init q40kbd_init(void) diff --git a/drivers/input/serio/rpckbd.c b/drivers/input/serio/rpckbd.c index 2af5df6a8fba..567566ae0dae 100644 --- a/drivers/input/serio/rpckbd.c +++ b/drivers/input/serio/rpckbd.c @@ -114,7 +114,7 @@ static void rpckbd_close(struct serio *port) * Allocate and initialize serio structure for subsequent registration * with serio core. */ -static int __devinit rpckbd_probe(struct platform_device *dev) +static int rpckbd_probe(struct platform_device *dev) { struct rpckbd_data *rpckbd; struct serio *serio; @@ -153,7 +153,7 @@ static int __devinit rpckbd_probe(struct platform_device *dev) return 0; } -static int __devexit rpckbd_remove(struct platform_device *dev) +static int rpckbd_remove(struct platform_device *dev) { struct serio *serio = platform_get_drvdata(dev); struct rpckbd_data *rpckbd = serio->port_data; @@ -166,7 +166,7 @@ static int __devexit rpckbd_remove(struct platform_device *dev) static struct platform_driver rpckbd_driver = { .probe = rpckbd_probe, - .remove = __devexit_p(rpckbd_remove), + .remove = rpckbd_remove, .driver = { .name = "kart", .owner = THIS_MODULE, diff --git a/drivers/input/serio/sa1111ps2.c b/drivers/input/serio/sa1111ps2.c index 389766707534..b3e688911fd9 100644 --- a/drivers/input/serio/sa1111ps2.c +++ b/drivers/input/serio/sa1111ps2.c @@ -193,7 +193,7 @@ static void ps2_close(struct serio *io) /* * Clear the input buffer. */ -static void __devinit ps2_clear_input(struct ps2if *ps2if) +static void ps2_clear_input(struct ps2if *ps2if) { int maxread = 100; @@ -203,7 +203,7 @@ static void __devinit ps2_clear_input(struct ps2if *ps2if) } } -static unsigned int __devinit ps2_test_one(struct ps2if *ps2if, +static unsigned int ps2_test_one(struct ps2if *ps2if, unsigned int mask) { unsigned int val; @@ -220,7 +220,7 @@ static unsigned int __devinit ps2_test_one(struct ps2if *ps2if, * Test the keyboard interface. We basically check to make sure that * we can drive each line to the keyboard independently of each other. */ -static int __devinit ps2_test(struct ps2if *ps2if) +static int ps2_test(struct ps2if *ps2if) { unsigned int stat; int ret = 0; @@ -251,7 +251,7 @@ static int __devinit ps2_test(struct ps2if *ps2if) /* * Add one device to this driver. */ -static int __devinit ps2_probe(struct sa1111_dev *dev) +static int ps2_probe(struct sa1111_dev *dev) { struct ps2if *ps2if; struct serio *serio; @@ -334,7 +334,7 @@ static int __devinit ps2_probe(struct sa1111_dev *dev) /* * Remove one device from this driver. */ -static int __devexit ps2_remove(struct sa1111_dev *dev) +static int ps2_remove(struct sa1111_dev *dev) { struct ps2if *ps2if = sa1111_get_drvdata(dev); @@ -357,7 +357,7 @@ static struct sa1111_driver ps2_driver = { }, .devid = SA1111_DEVID_PS2, .probe = ps2_probe, - .remove = __devexit_p(ps2_remove), + .remove = ps2_remove, }; static int __init ps2_init(void) diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c index d0f7533dbf88..25fc5971f426 100644 --- a/drivers/input/serio/serio.c +++ b/drivers/input/serio/serio.c @@ -891,8 +891,6 @@ static int serio_bus_match(struct device *dev, struct device_driver *drv) return serio_match_port(serio_drv->id_table, serio); } -#ifdef CONFIG_HOTPLUG - #define SERIO_ADD_UEVENT_VAR(fmt, val...) \ do { \ int err = add_uevent_var(env, fmt, val); \ @@ -920,15 +918,6 @@ static int serio_uevent(struct device *dev, struct kobj_uevent_env *env) } #undef SERIO_ADD_UEVENT_VAR -#else - -static int serio_uevent(struct device *dev, struct kobj_uevent_env *env) -{ - return -ENODEV; -} - -#endif /* CONFIG_HOTPLUG */ - #ifdef CONFIG_PM static int serio_suspend(struct device *dev) { diff --git a/drivers/input/serio/xilinx_ps2.c b/drivers/input/serio/xilinx_ps2.c index 1e983bec7d86..17be85948ffd 100644 --- a/drivers/input/serio/xilinx_ps2.c +++ b/drivers/input/serio/xilinx_ps2.c @@ -233,7 +233,7 @@ static void sxps2_close(struct serio *pserio) * It returns 0, if the driver is bound to the PS/2 device, or a negative * value if there is an error. */ -static int __devinit xps2_of_probe(struct platform_device *ofdev) +static int xps2_of_probe(struct platform_device *ofdev) { struct resource r_irq; /* Interrupt resources */ struct resource r_mem; /* IO mem resources */ @@ -333,7 +333,7 @@ failed1: * if the driver module is being unloaded. It frees any resources allocated to * the device. */ -static int __devexit xps2_of_remove(struct platform_device *of_dev) +static int xps2_of_remove(struct platform_device *of_dev) { struct xps2data *drvdata = platform_get_drvdata(of_dev); struct resource r_mem; /* IO mem resources */ @@ -355,7 +355,7 @@ static int __devexit xps2_of_remove(struct platform_device *of_dev) } /* Match table for of_platform binding */ -static const struct of_device_id xps2_of_match[] __devinitconst = { +static const struct of_device_id xps2_of_match[] = { { .compatible = "xlnx,xps-ps2-1.00.a", }, { /* end of list */ }, }; @@ -368,7 +368,7 @@ static struct platform_driver xps2_of_driver = { .of_match_table = xps2_of_match, }, .probe = xps2_of_probe, - .remove = __devexit_p(xps2_of_remove), + .remove = xps2_of_remove, }; module_platform_driver(xps2_of_driver); diff --git a/drivers/input/tablet/wacom_sys.c b/drivers/input/tablet/wacom_sys.c index 858ad446de91..f92d34f45a1c 100644 --- a/drivers/input/tablet/wacom_sys.c +++ b/drivers/input/tablet/wacom_sys.c @@ -386,23 +386,40 @@ static int wacom_parse_hid(struct usb_interface *intf, if (usage == WCM_DESKTOP) { if (finger) { features->device_type = BTN_TOOL_FINGER; - if (features->type == TABLETPC2FG) { - /* need to reset back */ + + switch (features->type) { + case TABLETPC2FG: features->pktlen = WACOM_PKGLEN_TPC2FG; - } + break; - if (features->type == MTSCREEN || features->type == WACOM_24HDT) + case MTSCREEN: + case WACOM_24HDT: features->pktlen = WACOM_PKGLEN_MTOUCH; + break; - if (features->type == BAMBOO_PT) { - /* need to reset back */ + case MTTPC: + features->pktlen = WACOM_PKGLEN_MTTPC; + break; + + case BAMBOO_PT: features->pktlen = WACOM_PKGLEN_BBTOUCH; + break; + + default: + features->pktlen = WACOM_PKGLEN_GRAPHIRE; + break; + } + + switch (features->type) { + case BAMBOO_PT: features->x_phy = get_unaligned_le16(&report[i + 5]); features->x_max = get_unaligned_le16(&report[i + 8]); i += 15; - } else if (features->type == WACOM_24HDT) { + break; + + case WACOM_24HDT: features->x_max = get_unaligned_le16(&report[i + 3]); features->x_phy = @@ -410,7 +427,9 @@ static int wacom_parse_hid(struct usb_interface *intf, features->unit = report[i - 1]; features->unitExpo = report[i - 3]; i += 12; - } else { + break; + + default: features->x_max = get_unaligned_le16(&report[i + 3]); features->x_phy = @@ -418,10 +437,11 @@ static int wacom_parse_hid(struct usb_interface *intf, features->unit = report[i + 9]; features->unitExpo = report[i + 11]; i += 12; + break; } } else if (pen) { /* penabled only accepts exact bytes of data */ - if (features->type == TABLETPC2FG) + if (features->type >= TABLETPC) features->pktlen = WACOM_PKGLEN_GRAPHIRE; features->device_type = BTN_TOOL_PEN; features->x_max = @@ -434,32 +454,40 @@ static int wacom_parse_hid(struct usb_interface *intf, case HID_USAGE_Y: if (usage == WCM_DESKTOP) { if (finger) { - int type = features->type; - - if (type == TABLETPC2FG || type == MTSCREEN) { + switch (features->type) { + case TABLETPC2FG: + case MTSCREEN: + case MTTPC: features->y_max = get_unaligned_le16(&report[i + 3]); features->y_phy = get_unaligned_le16(&report[i + 6]); i += 7; - } else if (type == WACOM_24HDT) { + break; + + case WACOM_24HDT: features->y_max = get_unaligned_le16(&report[i + 3]); features->y_phy = get_unaligned_le16(&report[i - 2]); i += 7; - } else if (type == BAMBOO_PT) { + break; + + case BAMBOO_PT: features->y_phy = get_unaligned_le16(&report[i + 3]); features->y_max = get_unaligned_le16(&report[i + 6]); i += 12; - } else { + break; + + default: features->y_max = features->x_max; features->y_phy = get_unaligned_le16(&report[i + 3]); i += 4; + break; } } else if (pen) { features->y_max = diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c index 0a67031ffc13..264138f3217e 100644 --- a/drivers/input/tablet/wacom_wac.c +++ b/drivers/input/tablet/wacom_wac.c @@ -467,9 +467,7 @@ static void wacom_intuos_general(struct wacom_wac *wacom) /* general pen packet */ if ((data[1] & 0xb8) == 0xa0) { t = (data[6] << 2) | ((data[7] >> 6) & 3); - if ((features->type >= INTUOS4S && features->type <= INTUOS4L) || - (features->type >= INTUOS5S && features->type <= INTUOS5L) || - (features->type >= WACOM_21UX2 && features->type <= WACOM_24HD)) { + if (features->type >= INTUOS4S && features->type <= WACOM_24HD) { t = (t << 1) | (data[1] & 1); } input_report_abs(input, ABS_PRESSURE, t); @@ -877,6 +875,11 @@ static int wacom_mt_touch(struct wacom_wac *wacom) int i; int current_num_contacts = data[2]; int contacts_to_send = 0; + int x_offset = 0; + + /* MTTPC does not support Height and Width */ + if (wacom->features.type == MTTPC) + x_offset = -4; /* * First packet resets the counter since only the first @@ -889,7 +892,7 @@ static int wacom_mt_touch(struct wacom_wac *wacom) contacts_to_send = min(5, wacom->num_contacts_left); for (i = 0; i < contacts_to_send; i++) { - int offset = (WACOM_BYTES_PER_MT_PACKET * i) + 3; + int offset = (WACOM_BYTES_PER_MT_PACKET + x_offset) * i + 3; bool touch = data[offset] & 0x1; int id = le16_to_cpup((__le16 *)&data[offset + 1]); int slot = find_slot_from_contactid(wacom, id); @@ -900,8 +903,8 @@ static int wacom_mt_touch(struct wacom_wac *wacom) input_mt_slot(input, slot); input_mt_report_slot_state(input, MT_TOOL_FINGER, touch); if (touch) { - int x = le16_to_cpup((__le16 *)&data[offset + 7]); - int y = le16_to_cpup((__le16 *)&data[offset + 9]); + int x = le16_to_cpup((__le16 *)&data[offset + x_offset + 7]); + int y = le16_to_cpup((__le16 *)&data[offset + x_offset + 9]); input_report_abs(input, ABS_MT_POSITION_X, x); input_report_abs(input, ABS_MT_POSITION_Y, y); } @@ -1336,6 +1339,7 @@ void wacom_wac_irq(struct wacom_wac *wacom_wac, size_t len) case TABLETPCE: case TABLETPC2FG: case MTSCREEN: + case MTTPC: sync = wacom_tpc_irq(wacom_wac, len); break; @@ -1657,6 +1661,7 @@ int wacom_setup_input_capabilities(struct input_dev *input_dev, /* fall through */ case MTSCREEN: + case MTTPC: if (features->device_type == BTN_TOOL_FINGER) { wacom_wac->slots = kmalloc(features->touch_max * sizeof(int), @@ -2018,6 +2023,15 @@ static const struct wacom_features wacom_features_0xED = static const struct wacom_features wacom_features_0xEF = { "Wacom ISDv4 EF", WACOM_PKGLEN_GRAPHIRE, 26202, 16325, 255, 0, TABLETPC, WACOM_INTUOS_RES, WACOM_INTUOS_RES }; +static const struct wacom_features wacom_features_0x100 = + { "Wacom ISDv4 100", WACOM_PKGLEN_MTTPC, 26202, 16325, 255, + 0, MTTPC, WACOM_INTUOS_RES, WACOM_INTUOS_RES }; +static const struct wacom_features wacom_features_0x101 = + { "Wacom ISDv4 101", WACOM_PKGLEN_MTTPC, 26202, 16325, 255, + 0, MTTPC, WACOM_INTUOS_RES, WACOM_INTUOS_RES }; +static const struct wacom_features wacom_features_0x4001 = + { "Wacom ISDv4 4001", WACOM_PKGLEN_MTTPC, 26202, 16325, 255, + 0, MTTPC, WACOM_INTUOS_RES, WACOM_INTUOS_RES }; static const struct wacom_features wacom_features_0x47 = { "Wacom Intuos2 6x8", WACOM_PKGLEN_INTUOS, 20320, 16240, 1023, 31, INTUOS, WACOM_INTUOS_RES, WACOM_INTUOS_RES }; @@ -2034,7 +2048,8 @@ static const struct wacom_features wacom_features_0xD1 = .touch_max = 2 }; static const struct wacom_features wacom_features_0xD2 = { "Wacom Bamboo Craft", WACOM_PKGLEN_BBFUN, 14720, 9200, 1023, - 31, BAMBOO_PT, WACOM_INTUOS_RES, WACOM_INTUOS_RES }; + 31, BAMBOO_PT, WACOM_INTUOS_RES, WACOM_INTUOS_RES, + .touch_max = 2 }; static const struct wacom_features wacom_features_0xD3 = { "Wacom Bamboo 2FG 6x8", WACOM_PKGLEN_BBFUN, 21648, 13700, 1023, 31, BAMBOO_PT, WACOM_INTUOS_RES, WACOM_INTUOS_RES, @@ -2194,6 +2209,9 @@ const struct usb_device_id wacom_ids[] = { { USB_DEVICE_WACOM(0xEC) }, { USB_DEVICE_WACOM(0xED) }, { USB_DEVICE_WACOM(0xEF) }, + { USB_DEVICE_WACOM(0x100) }, + { USB_DEVICE_WACOM(0x101) }, + { USB_DEVICE_WACOM(0x4001) }, { USB_DEVICE_WACOM(0x47) }, { USB_DEVICE_WACOM(0xF4) }, { USB_DEVICE_WACOM(0xF8) }, diff --git a/drivers/input/tablet/wacom_wac.h b/drivers/input/tablet/wacom_wac.h index 345f1e76975e..9396d7769f86 100644 --- a/drivers/input/tablet/wacom_wac.h +++ b/drivers/input/tablet/wacom_wac.h @@ -26,6 +26,7 @@ #define WACOM_PKGLEN_BBPEN 10 #define WACOM_PKGLEN_WIRELESS 32 #define WACOM_PKGLEN_MTOUCH 62 +#define WACOM_PKGLEN_MTTPC 40 /* wacom data size per MT contact */ #define WACOM_BYTES_PER_MT_PACKET 11 @@ -88,6 +89,7 @@ enum { TABLETPCE, TABLETPC2FG, MTSCREEN, + MTTPC, MAX_TYPE }; diff --git a/drivers/input/touchscreen/88pm860x-ts.c b/drivers/input/touchscreen/88pm860x-ts.c index 326218dbd6e6..c7068942ebe8 100644 --- a/drivers/input/touchscreen/88pm860x-ts.c +++ b/drivers/input/touchscreen/88pm860x-ts.c @@ -115,7 +115,7 @@ static void pm860x_touch_close(struct input_dev *dev) } #ifdef CONFIG_OF -static int __devinit pm860x_touch_dt_init(struct platform_device *pdev, +static int pm860x_touch_dt_init(struct platform_device *pdev, struct pm860x_chip *chip, int *res_x) { @@ -169,7 +169,7 @@ static int __devinit pm860x_touch_dt_init(struct platform_device *pdev, #define pm860x_touch_dt_init(x, y, z) (-1) #endif -static int __devinit pm860x_touch_probe(struct platform_device *pdev) +static int pm860x_touch_probe(struct platform_device *pdev) { struct pm860x_chip *chip = dev_get_drvdata(pdev->dev.parent); struct pm860x_touch_pdata *pdata = pdev->dev.platform_data; @@ -293,7 +293,7 @@ out: return ret; } -static int __devexit pm860x_touch_remove(struct platform_device *pdev) +static int pm860x_touch_remove(struct platform_device *pdev) { struct pm860x_touch *touch = platform_get_drvdata(pdev); @@ -310,7 +310,7 @@ static struct platform_driver pm860x_touch_driver = { .owner = THIS_MODULE, }, .probe = pm860x_touch_probe, - .remove = __devexit_p(pm860x_touch_remove), + .remove = pm860x_touch_remove, }; module_platform_driver(pm860x_touch_driver); diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig index 0c45caddd41c..515cfe790543 100644 --- a/drivers/input/touchscreen/Kconfig +++ b/drivers/input/touchscreen/Kconfig @@ -111,18 +111,6 @@ config TOUCHSCREEN_AUO_PIXCIR To compile this driver as a module, choose M here: the module will be called auo-pixcir-ts. -config TOUCHSCREEN_BITSY - tristate "Compaq iPAQ H3600 (Bitsy) touchscreen" - depends on SA1100_BITSY - select SERIO - help - Say Y here if you have the h3600 (Bitsy) touchscreen. - - If unsure, say N. - - To compile this driver as a module, choose M here: the - module will be called h3600_ts_input. - config TOUCHSCREEN_BU21013 tristate "BU21013 based touch panel controllers" depends on I2C diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile index 7c4c78ebe49e..6bfbeab67c9f 100644 --- a/drivers/input/touchscreen/Makefile +++ b/drivers/input/touchscreen/Makefile @@ -15,7 +15,6 @@ obj-$(CONFIG_TOUCHSCREEN_ADS7846) += ads7846.o obj-$(CONFIG_TOUCHSCREEN_ATMEL_MXT) += atmel_mxt_ts.o obj-$(CONFIG_TOUCHSCREEN_ATMEL_TSADCC) += atmel_tsadcc.o obj-$(CONFIG_TOUCHSCREEN_AUO_PIXCIR) += auo-pixcir-ts.o -obj-$(CONFIG_TOUCHSCREEN_BITSY) += h3600_ts_input.o obj-$(CONFIG_TOUCHSCREEN_BU21013) += bu21013_ts.o obj-$(CONFIG_TOUCHSCREEN_CY8CTMG110) += cy8ctmg110_ts.o obj-$(CONFIG_TOUCHSCREEN_CYTTSP_CORE) += cyttsp_core.o diff --git a/drivers/input/touchscreen/ad7877.c b/drivers/input/touchscreen/ad7877.c index 2c7692108e6c..23fa829b869d 100644 --- a/drivers/input/touchscreen/ad7877.c +++ b/drivers/input/touchscreen/ad7877.c @@ -682,7 +682,7 @@ static void ad7877_setup_ts_def_msg(struct spi_device *spi, struct ad7877 *ts) } } -static int __devinit ad7877_probe(struct spi_device *spi) +static int ad7877_probe(struct spi_device *spi) { struct ad7877 *ts; struct input_dev *input_dev; @@ -810,7 +810,7 @@ err_free_mem: return err; } -static int __devexit ad7877_remove(struct spi_device *spi) +static int ad7877_remove(struct spi_device *spi) { struct ad7877 *ts = dev_get_drvdata(&spi->dev); @@ -857,7 +857,7 @@ static struct spi_driver ad7877_driver = { .pm = &ad7877_pm, }, .probe = ad7877_probe, - .remove = __devexit_p(ad7877_remove), + .remove = ad7877_remove, }; module_spi_driver(ad7877_driver); diff --git a/drivers/input/touchscreen/ad7879-i2c.c b/drivers/input/touchscreen/ad7879-i2c.c index 3054354d0dd3..dcf390771549 100644 --- a/drivers/input/touchscreen/ad7879-i2c.c +++ b/drivers/input/touchscreen/ad7879-i2c.c @@ -54,7 +54,7 @@ static const struct ad7879_bus_ops ad7879_i2c_bus_ops = { .write = ad7879_i2c_write, }; -static int __devinit ad7879_i2c_probe(struct i2c_client *client, +static int ad7879_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct ad7879 *ts; @@ -75,7 +75,7 @@ static int __devinit ad7879_i2c_probe(struct i2c_client *client, return 0; } -static int __devexit ad7879_i2c_remove(struct i2c_client *client) +static int ad7879_i2c_remove(struct i2c_client *client) { struct ad7879 *ts = i2c_get_clientdata(client); @@ -98,7 +98,7 @@ static struct i2c_driver ad7879_i2c_driver = { .pm = &ad7879_pm_ops, }, .probe = ad7879_i2c_probe, - .remove = __devexit_p(ad7879_i2c_remove), + .remove = ad7879_i2c_remove, .id_table = ad7879_id, }; diff --git a/drivers/input/touchscreen/ad7879-spi.c b/drivers/input/touchscreen/ad7879-spi.c index db49abf056ba..606da5bd6115 100644 --- a/drivers/input/touchscreen/ad7879-spi.c +++ b/drivers/input/touchscreen/ad7879-spi.c @@ -110,7 +110,7 @@ static const struct ad7879_bus_ops ad7879_spi_bus_ops = { .write = ad7879_spi_write, }; -static int __devinit ad7879_spi_probe(struct spi_device *spi) +static int ad7879_spi_probe(struct spi_device *spi) { struct ad7879 *ts; int err; @@ -137,7 +137,7 @@ static int __devinit ad7879_spi_probe(struct spi_device *spi) return 0; } -static int __devexit ad7879_spi_remove(struct spi_device *spi) +static int ad7879_spi_remove(struct spi_device *spi) { struct ad7879 *ts = spi_get_drvdata(spi); @@ -154,7 +154,7 @@ static struct spi_driver ad7879_spi_driver = { .pm = &ad7879_pm_ops, }, .probe = ad7879_spi_probe, - .remove = __devexit_p(ad7879_spi_remove), + .remove = ad7879_spi_remove, }; module_spi_driver(ad7879_spi_driver); diff --git a/drivers/input/touchscreen/ads7846.c b/drivers/input/touchscreen/ads7846.c index 78e5d9ab0ba7..4f702b3ec1a3 100644 --- a/drivers/input/touchscreen/ads7846.c +++ b/drivers/input/touchscreen/ads7846.c @@ -955,7 +955,7 @@ static int ads7846_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(ads7846_pm, ads7846_suspend, ads7846_resume); -static int __devinit ads7846_setup_pendown(struct spi_device *spi, +static int ads7846_setup_pendown(struct spi_device *spi, struct ads7846 *ts) { struct ads7846_platform_data *pdata = spi->dev.platform_data; @@ -997,7 +997,7 @@ static int __devinit ads7846_setup_pendown(struct spi_device *spi, * Set up the transfers to read touchscreen state; this assumes we * use formula #2 for pressure, not #3. */ -static void __devinit ads7846_setup_spi_msg(struct ads7846 *ts, +static void ads7846_setup_spi_msg(struct ads7846 *ts, const struct ads7846_platform_data *pdata) { struct spi_message *m = &ts->msg[0]; @@ -1196,7 +1196,7 @@ static void __devinit ads7846_setup_spi_msg(struct ads7846 *ts, spi_message_add_tail(x, m); } -static int __devinit ads7846_probe(struct spi_device *spi) +static int ads7846_probe(struct spi_device *spi) { struct ads7846 *ts; struct ads7846_packet *packet; @@ -1390,7 +1390,7 @@ static int __devinit ads7846_probe(struct spi_device *spi) return err; } -static int __devexit ads7846_remove(struct spi_device *spi) +static int ads7846_remove(struct spi_device *spi) { struct ads7846 *ts = dev_get_drvdata(&spi->dev); @@ -1434,7 +1434,7 @@ static struct spi_driver ads7846_driver = { .pm = &ads7846_pm, }, .probe = ads7846_probe, - .remove = __devexit_p(ads7846_remove), + .remove = ads7846_remove, }; module_spi_driver(ads7846_driver); diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c index 1df2396af008..d04f810cb1dd 100644 --- a/drivers/input/touchscreen/atmel_mxt_ts.c +++ b/drivers/input/touchscreen/atmel_mxt_ts.c @@ -1095,7 +1095,7 @@ static void mxt_input_close(struct input_dev *dev) mxt_stop(data); } -static int __devinit mxt_probe(struct i2c_client *client, +static int mxt_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct mxt_platform_data *pdata = client->dev.platform_data; @@ -1200,7 +1200,7 @@ err_free_mem: return error; } -static int __devexit mxt_remove(struct i2c_client *client) +static int mxt_remove(struct i2c_client *client) { struct mxt_data *data = i2c_get_clientdata(client); @@ -1270,7 +1270,7 @@ static struct i2c_driver mxt_driver = { .pm = &mxt_pm_ops, }, .probe = mxt_probe, - .remove = __devexit_p(mxt_remove), + .remove = mxt_remove, .id_table = mxt_id, }; diff --git a/drivers/input/touchscreen/atmel_tsadcc.c b/drivers/input/touchscreen/atmel_tsadcc.c index ea392ee138ed..95f6785a94b0 100644 --- a/drivers/input/touchscreen/atmel_tsadcc.c +++ b/drivers/input/touchscreen/atmel_tsadcc.c @@ -177,7 +177,7 @@ static irqreturn_t atmel_tsadcc_interrupt(int irq, void *dev) * The functions for inserting/removing us as a module. */ -static int __devinit atmel_tsadcc_probe(struct platform_device *pdev) +static int atmel_tsadcc_probe(struct platform_device *pdev) { struct atmel_tsadcc *ts_dev; struct input_dev *input_dev; @@ -323,7 +323,7 @@ err_free_mem: return err; } -static int __devexit atmel_tsadcc_remove(struct platform_device *pdev) +static int atmel_tsadcc_remove(struct platform_device *pdev) { struct atmel_tsadcc *ts_dev = dev_get_drvdata(&pdev->dev); struct resource *res; @@ -346,7 +346,7 @@ static int __devexit atmel_tsadcc_remove(struct platform_device *pdev) static struct platform_driver atmel_tsadcc_driver = { .probe = atmel_tsadcc_probe, - .remove = __devexit_p(atmel_tsadcc_remove), + .remove = atmel_tsadcc_remove, .driver = { .name = "atmel_tsadcc", }, diff --git a/drivers/input/touchscreen/auo-pixcir-ts.c b/drivers/input/touchscreen/auo-pixcir-ts.c index c7047b6bb020..c6e19a96348e 100644 --- a/drivers/input/touchscreen/auo-pixcir-ts.c +++ b/drivers/input/touchscreen/auo-pixcir-ts.c @@ -286,7 +286,7 @@ static int auo_pixcir_power_mode(struct auo_pixcir_ts *ts, int mode) return 0; } -static __devinit int auo_pixcir_int_config(struct auo_pixcir_ts *ts, +static int auo_pixcir_int_config(struct auo_pixcir_ts *ts, int int_setting) { struct i2c_client *client = ts->client; @@ -482,7 +482,7 @@ unlock: static SIMPLE_DEV_PM_OPS(auo_pixcir_pm_ops, auo_pixcir_suspend, auo_pixcir_resume); -static int __devinit auo_pixcir_probe(struct i2c_client *client, +static int auo_pixcir_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct auo_pixcir_ts_platdata *pdata = client->dev.platform_data; @@ -599,7 +599,7 @@ err_gpio_int: return ret; } -static int __devexit auo_pixcir_remove(struct i2c_client *client) +static int auo_pixcir_remove(struct i2c_client *client) { struct auo_pixcir_ts *ts = i2c_get_clientdata(client); const struct auo_pixcir_ts_platdata *pdata = client->dev.platform_data; @@ -631,7 +631,7 @@ static struct i2c_driver auo_pixcir_driver = { .pm = &auo_pixcir_pm_ops, }, .probe = auo_pixcir_probe, - .remove = __devexit_p(auo_pixcir_remove), + .remove = auo_pixcir_remove, .id_table = auo_pixcir_idtable, }; diff --git a/drivers/input/touchscreen/bu21013_ts.c b/drivers/input/touchscreen/bu21013_ts.c index 5c487d23f11c..b9b5ddad6658 100644 --- a/drivers/input/touchscreen/bu21013_ts.c +++ b/drivers/input/touchscreen/bu21013_ts.c @@ -14,6 +14,9 @@ #include <linux/slab.h> #include <linux/regulator/consumer.h> #include <linux/module.h> +#include <linux/gpio.h> +#include <linux/of.h> +#include <linux/of_gpio.h> #define PEN_DOWN_INTR 0 #define MAX_FINGERS 2 @@ -148,11 +151,12 @@ struct bu21013_ts_data { struct i2c_client *client; wait_queue_head_t wait; - bool touch_stopped; const struct bu21013_platform_device *chip; struct input_dev *in_dev; - unsigned int intr_pin; struct regulator *regulator; + unsigned int irq; + unsigned int intr_pin; + bool touch_stopped; }; /** @@ -262,7 +266,7 @@ static irqreturn_t bu21013_gpio_irq(int irq, void *device_data) return IRQ_NONE; } - data->intr_pin = data->chip->irq_read_val(); + data->intr_pin = gpio_get_value(data->chip->touch_pin); if (data->intr_pin == PEN_DOWN_INTR) wait_event_timeout(data->wait, data->touch_stopped, msecs_to_jiffies(2)); @@ -418,8 +422,70 @@ static void bu21013_free_irq(struct bu21013_ts_data *bu21013_data) { bu21013_data->touch_stopped = true; wake_up(&bu21013_data->wait); - free_irq(bu21013_data->chip->irq, bu21013_data); + free_irq(bu21013_data->irq, bu21013_data); +} + +/** + * bu21013_cs_disable() - deconfigures the touch panel controller + * @bu21013_data: device structure pointer + * + * This function is used to deconfigure the chip selection + * for touch panel controller. + */ +static void bu21013_cs_disable(struct bu21013_ts_data *bu21013_data) +{ + int error; + + error = gpio_direction_output(bu21013_data->chip->cs_pin, 0); + if (error < 0) + dev_warn(&bu21013_data->client->dev, + "%s: gpio direction failed, error: %d\n", + __func__, error); + else + gpio_set_value(bu21013_data->chip->cs_pin, 0); + + gpio_free(bu21013_data->chip->cs_pin); +} + +#ifdef CONFIG_OF +static const struct bu21013_platform_device * +bu21013_parse_dt(struct device *dev) +{ + struct device_node *np = dev->of_node; + struct bu21013_platform_device *pdata; + + if (!np) { + dev_err(dev, "no device tree or platform data\n"); + return ERR_PTR(-EINVAL); + } + + pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL); + if (!pdata) + return ERR_PTR(-ENOMEM); + + pdata->y_flip = pdata->x_flip = false; + + pdata->x_flip = of_property_read_bool(np, "rohm,flip-x"); + pdata->y_flip = of_property_read_bool(np, "rohm,flip-y"); + + of_property_read_u32(np, "rohm,touch-max-x", &pdata->touch_x_max); + of_property_read_u32(np, "rohm,touch-max-y", &pdata->touch_y_max); + + pdata->touch_pin = of_get_named_gpio(np, "touch-gpio", 0); + pdata->cs_pin = of_get_named_gpio(np, "reset-gpio", 0); + + pdata->ext_clk = false; + + return pdata; } +#else +static inline const struct bu21013_platform_device * +bu21013_parse_dt(struct device *dev) +{ + dev_err(dev, "no platform data available\n"); + return ERR_PTR(-EINVAL); +} +#endif /** * bu21013_probe() - initializes the i2c-client touchscreen driver @@ -429,13 +495,13 @@ static void bu21013_free_irq(struct bu21013_ts_data *bu21013_data) * This function used to initializes the i2c-client touchscreen * driver and returns integer. */ -static int __devinit bu21013_probe(struct i2c_client *client, - const struct i2c_device_id *id) +static int bu21013_probe(struct i2c_client *client, + const struct i2c_device_id *id) { + const struct bu21013_platform_device *pdata = + dev_get_platdata(&client->dev); struct bu21013_ts_data *bu21013_data; struct input_dev *in_dev; - const struct bu21013_platform_device *pdata = - client->dev.platform_data; int error; if (!i2c_check_functionality(client->adapter, @@ -445,7 +511,13 @@ static int __devinit bu21013_probe(struct i2c_client *client, } if (!pdata) { - dev_err(&client->dev, "platform data not defined\n"); + pdata = bu21013_parse_dt(&client->dev); + if (IS_ERR(pdata)) + return PTR_ERR(pdata); + } + + if (!gpio_is_valid(pdata->touch_pin)) { + dev_err(&client->dev, "invalid touch_pin supplied\n"); return -EINVAL; } @@ -460,8 +532,9 @@ static int __devinit bu21013_probe(struct i2c_client *client, bu21013_data->in_dev = in_dev; bu21013_data->chip = pdata; bu21013_data->client = client; + bu21013_data->irq = gpio_to_irq(pdata->touch_pin); - bu21013_data->regulator = regulator_get(&client->dev, "V-TOUCH"); + bu21013_data->regulator = regulator_get(&client->dev, "avdd"); if (IS_ERR(bu21013_data->regulator)) { dev_err(&client->dev, "regulator_get failed\n"); error = PTR_ERR(bu21013_data->regulator); @@ -478,12 +551,11 @@ static int __devinit bu21013_probe(struct i2c_client *client, init_waitqueue_head(&bu21013_data->wait); /* configure the gpio pins */ - if (pdata->cs_en) { - error = pdata->cs_en(pdata->cs_pin); - if (error < 0) { - dev_err(&client->dev, "chip init failed\n"); - goto err_disable_regulator; - } + error = gpio_request_one(pdata->cs_pin, GPIOF_OUT_INIT_HIGH, + "touchp_reset"); + if (error < 0) { + dev_err(&client->dev, "Unable to request gpio reset_pin\n"); + goto err_disable_regulator; } /* configure the touch panel controller */ @@ -508,12 +580,13 @@ static int __devinit bu21013_probe(struct i2c_client *client, pdata->touch_y_max, 0, 0); input_set_drvdata(in_dev, bu21013_data); - error = request_threaded_irq(pdata->irq, NULL, bu21013_gpio_irq, + error = request_threaded_irq(bu21013_data->irq, NULL, bu21013_gpio_irq, IRQF_TRIGGER_FALLING | IRQF_SHARED | IRQF_ONESHOT, DRIVER_TP, bu21013_data); if (error) { - dev_err(&client->dev, "request irq %d failed\n", pdata->irq); + dev_err(&client->dev, "request irq %d failed\n", + bu21013_data->irq); goto err_cs_disable; } @@ -531,7 +604,7 @@ static int __devinit bu21013_probe(struct i2c_client *client, err_free_irq: bu21013_free_irq(bu21013_data); err_cs_disable: - pdata->cs_dis(pdata->cs_pin); + bu21013_cs_disable(bu21013_data); err_disable_regulator: regulator_disable(bu21013_data->regulator); err_put_regulator: @@ -549,13 +622,13 @@ err_free_mem: * This function uses to remove the i2c-client * touchscreen driver and returns integer. */ -static int __devexit bu21013_remove(struct i2c_client *client) +static int bu21013_remove(struct i2c_client *client) { struct bu21013_ts_data *bu21013_data = i2c_get_clientdata(client); bu21013_free_irq(bu21013_data); - bu21013_data->chip->cs_dis(bu21013_data->chip->cs_pin); + bu21013_cs_disable(bu21013_data); input_unregister_device(bu21013_data->in_dev); @@ -584,9 +657,9 @@ static int bu21013_suspend(struct device *dev) bu21013_data->touch_stopped = true; if (device_may_wakeup(&client->dev)) - enable_irq_wake(bu21013_data->chip->irq); + enable_irq_wake(bu21013_data->irq); else - disable_irq(bu21013_data->chip->irq); + disable_irq(bu21013_data->irq); regulator_disable(bu21013_data->regulator); @@ -621,9 +694,9 @@ static int bu21013_resume(struct device *dev) bu21013_data->touch_stopped = false; if (device_may_wakeup(&client->dev)) - disable_irq_wake(bu21013_data->chip->irq); + disable_irq_wake(bu21013_data->irq); else - enable_irq(bu21013_data->chip->irq); + enable_irq(bu21013_data->irq); return 0; } @@ -649,7 +722,7 @@ static struct i2c_driver bu21013_driver = { #endif }, .probe = bu21013_probe, - .remove = __devexit_p(bu21013_remove), + .remove = bu21013_remove, .id_table = bu21013_id, }; diff --git a/drivers/input/touchscreen/cy8ctmg110_ts.c b/drivers/input/touchscreen/cy8ctmg110_ts.c index 464f1bf4b61d..96e0eedcc7e5 100644 --- a/drivers/input/touchscreen/cy8ctmg110_ts.c +++ b/drivers/input/touchscreen/cy8ctmg110_ts.c @@ -99,9 +99,18 @@ static int cy8ctmg110_read_regs(struct cy8ctmg110 *tsc, int ret; struct i2c_msg msg[2] = { /* first write slave position to i2c devices */ - { client->addr, 0, 1, &cmd }, + { + .addr = client->addr, + .len = 1, + .buf = &cmd + }, /* Second read data from position */ - { client->addr, I2C_M_RD, len, data } + { + .addr = client->addr, + .flags = I2C_M_RD, + .len = len, + .buf = data + } }; ret = i2c_transfer(client->adapter, msg, 2); @@ -166,7 +175,7 @@ static irqreturn_t cy8ctmg110_irq_thread(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit cy8ctmg110_probe(struct i2c_client *client, +static int cy8ctmg110_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct cy8ctmg110_pdata *pdata = client->dev.platform_data; @@ -314,7 +323,7 @@ static int cy8ctmg110_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(cy8ctmg110_pm, cy8ctmg110_suspend, cy8ctmg110_resume); #endif -static int __devexit cy8ctmg110_remove(struct i2c_client *client) +static int cy8ctmg110_remove(struct i2c_client *client) { struct cy8ctmg110 *ts = i2c_get_clientdata(client); @@ -348,7 +357,7 @@ static struct i2c_driver cy8ctmg110_driver = { }, .id_table = cy8ctmg110_idtable, .probe = cy8ctmg110_probe, - .remove = __devexit_p(cy8ctmg110_remove), + .remove = cy8ctmg110_remove, }; module_i2c_driver(cy8ctmg110_driver); diff --git a/drivers/input/touchscreen/cyttsp_i2c.c b/drivers/input/touchscreen/cyttsp_i2c.c index 2af1d0c52bcd..4dbdf44b8fc5 100644 --- a/drivers/input/touchscreen/cyttsp_i2c.c +++ b/drivers/input/touchscreen/cyttsp_i2c.c @@ -81,7 +81,7 @@ static const struct cyttsp_bus_ops cyttsp_i2c_bus_ops = { .read = cyttsp_i2c_read_block_data, }; -static int __devinit cyttsp_i2c_probe(struct i2c_client *client, +static int cyttsp_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct cyttsp *ts; @@ -102,7 +102,7 @@ static int __devinit cyttsp_i2c_probe(struct i2c_client *client, return 0; } -static int __devexit cyttsp_i2c_remove(struct i2c_client *client) +static int cyttsp_i2c_remove(struct i2c_client *client) { struct cyttsp *ts = i2c_get_clientdata(client); @@ -124,7 +124,7 @@ static struct i2c_driver cyttsp_i2c_driver = { .pm = &cyttsp_pm_ops, }, .probe = cyttsp_i2c_probe, - .remove = __devexit_p(cyttsp_i2c_remove), + .remove = cyttsp_i2c_remove, .id_table = cyttsp_i2c_id, }; diff --git a/drivers/input/touchscreen/cyttsp_spi.c b/drivers/input/touchscreen/cyttsp_spi.c index 9f263410407b..638e20310f12 100644 --- a/drivers/input/touchscreen/cyttsp_spi.c +++ b/drivers/input/touchscreen/cyttsp_spi.c @@ -147,7 +147,7 @@ static const struct cyttsp_bus_ops cyttsp_spi_bus_ops = { .read = cyttsp_spi_read_block_data, }; -static int __devinit cyttsp_spi_probe(struct spi_device *spi) +static int cyttsp_spi_probe(struct spi_device *spi) { struct cyttsp *ts; int error; @@ -172,7 +172,7 @@ static int __devinit cyttsp_spi_probe(struct spi_device *spi) return 0; } -static int __devexit cyttsp_spi_remove(struct spi_device *spi) +static int cyttsp_spi_remove(struct spi_device *spi) { struct cyttsp *ts = spi_get_drvdata(spi); @@ -188,7 +188,7 @@ static struct spi_driver cyttsp_spi_driver = { .pm = &cyttsp_pm_ops, }, .probe = cyttsp_spi_probe, - .remove = __devexit_p(cyttsp_spi_remove), + .remove = cyttsp_spi_remove, }; module_spi_driver(cyttsp_spi_driver); diff --git a/drivers/input/touchscreen/da9034-ts.c b/drivers/input/touchscreen/da9034-ts.c index 36b65cf10d7f..34ad84105e6e 100644 --- a/drivers/input/touchscreen/da9034-ts.c +++ b/drivers/input/touchscreen/da9034-ts.c @@ -297,7 +297,7 @@ static void da9034_touch_close(struct input_dev *dev) } -static int __devinit da9034_touch_probe(struct platform_device *pdev) +static int da9034_touch_probe(struct platform_device *pdev) { struct da9034_touch_pdata *pdata = pdev->dev.platform_data; struct da9034_touch *touch; @@ -361,7 +361,7 @@ err_free_touch: return ret; } -static int __devexit da9034_touch_remove(struct platform_device *pdev) +static int da9034_touch_remove(struct platform_device *pdev) { struct da9034_touch *touch = platform_get_drvdata(pdev); @@ -377,7 +377,7 @@ static struct platform_driver da9034_touch_driver = { .owner = THIS_MODULE, }, .probe = da9034_touch_probe, - .remove = __devexit_p(da9034_touch_remove), + .remove = da9034_touch_remove, }; module_platform_driver(da9034_touch_driver); diff --git a/drivers/input/touchscreen/da9052_tsi.c b/drivers/input/touchscreen/da9052_tsi.c index 53133efe0418..8f561e22bdd4 100644 --- a/drivers/input/touchscreen/da9052_tsi.c +++ b/drivers/input/touchscreen/da9052_tsi.c @@ -141,7 +141,7 @@ static void da9052_ts_pen_work(struct work_struct *work) } } -static int __devinit da9052_ts_configure_gpio(struct da9052 *da9052) +static int da9052_ts_configure_gpio(struct da9052 *da9052) { int error; @@ -160,7 +160,7 @@ static int __devinit da9052_ts_configure_gpio(struct da9052 *da9052) return 0; } -static int __devinit da9052_configure_tsi(struct da9052_tsi *tsi) +static int da9052_configure_tsi(struct da9052_tsi *tsi) { int error; @@ -227,7 +227,7 @@ static void da9052_ts_input_close(struct input_dev *input_dev) da9052_reg_update(tsi->da9052, DA9052_TSI_CONT_A_REG, 1 << 1, 0); } -static int __devinit da9052_ts_probe(struct platform_device *pdev) +static int da9052_ts_probe(struct platform_device *pdev) { struct da9052 *da9052; struct da9052_tsi *tsi; @@ -317,7 +317,7 @@ err_free_mem: return error; } -static int __devexit da9052_ts_remove(struct platform_device *pdev) +static int da9052_ts_remove(struct platform_device *pdev) { struct da9052_tsi *tsi = platform_get_drvdata(pdev); @@ -336,7 +336,7 @@ static int __devexit da9052_ts_remove(struct platform_device *pdev) static struct platform_driver da9052_tsi_driver = { .probe = da9052_ts_probe, - .remove = __devexit_p(da9052_ts_remove), + .remove = da9052_ts_remove, .driver = { .name = "da9052-tsi", .owner = THIS_MODULE, diff --git a/drivers/input/touchscreen/edt-ft5x06.c b/drivers/input/touchscreen/edt-ft5x06.c index 099d144ab7c9..a9170157b442 100644 --- a/drivers/input/touchscreen/edt-ft5x06.c +++ b/drivers/input/touchscreen/edt-ft5x06.c @@ -491,14 +491,6 @@ static int edt_ft5x06_debugfs_mode_set(void *data, u64 mode) DEFINE_SIMPLE_ATTRIBUTE(debugfs_mode_fops, edt_ft5x06_debugfs_mode_get, edt_ft5x06_debugfs_mode_set, "%llu\n"); -static int edt_ft5x06_debugfs_raw_data_open(struct inode *inode, - struct file *file) -{ - file->private_data = inode->i_private; - - return 0; -} - static ssize_t edt_ft5x06_debugfs_raw_data_read(struct file *file, char __user *buf, size_t count, loff_t *off) { @@ -579,11 +571,11 @@ out: static const struct file_operations debugfs_raw_data_fops = { - .open = edt_ft5x06_debugfs_raw_data_open, + .open = simple_open, .read = edt_ft5x06_debugfs_raw_data_read, }; -static void __devinit +static void edt_ft5x06_ts_prepare_debugfs(struct edt_ft5x06_ts_data *tsdata, const char *debugfs_name) { @@ -600,7 +592,7 @@ edt_ft5x06_ts_prepare_debugfs(struct edt_ft5x06_ts_data *tsdata, tsdata->debug_dir, tsdata, &debugfs_raw_data_fops); } -static void __devexit +static void edt_ft5x06_ts_teardown_debugfs(struct edt_ft5x06_ts_data *tsdata) { if (tsdata->debug_dir) @@ -625,7 +617,7 @@ edt_ft5x06_ts_teardown_debugfs(struct edt_ft5x06_ts_data *tsdata) -static int __devinit edt_ft5x06_ts_reset(struct i2c_client *client, +static int edt_ft5x06_ts_reset(struct i2c_client *client, int reset_pin) { int error; @@ -649,7 +641,7 @@ static int __devinit edt_ft5x06_ts_reset(struct i2c_client *client, return 0; } -static int __devinit edt_ft5x06_ts_identify(struct i2c_client *client, +static int edt_ft5x06_ts_identify(struct i2c_client *client, char *model_name, char *fw_version) { @@ -683,7 +675,7 @@ static int __devinit edt_ft5x06_ts_identify(struct i2c_client *client, pdata->name <= edt_ft5x06_attr_##name.limit_high) \ edt_ft5x06_register_write(tsdata, reg, pdata->name) -static void __devinit +static void edt_ft5x06_ts_get_defaults(struct edt_ft5x06_ts_data *tsdata, const struct edt_ft5x06_platform_data *pdata) { @@ -697,7 +689,7 @@ edt_ft5x06_ts_get_defaults(struct edt_ft5x06_ts_data *tsdata, EDT_ATTR_CHECKSET(report_rate, WORK_REGISTER_REPORT_RATE); } -static void __devinit +static void edt_ft5x06_ts_get_parameters(struct edt_ft5x06_ts_data *tsdata) { tsdata->threshold = edt_ft5x06_register_read(tsdata, @@ -710,7 +702,7 @@ edt_ft5x06_ts_get_parameters(struct edt_ft5x06_ts_data *tsdata) tsdata->num_y = edt_ft5x06_register_read(tsdata, WORK_REGISTER_NUM_Y); } -static int __devinit edt_ft5x06_ts_probe(struct i2c_client *client, +static int edt_ft5x06_ts_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct edt_ft5x06_platform_data *pdata = @@ -830,7 +822,7 @@ err_free_mem: return error; } -static int __devexit edt_ft5x06_ts_remove(struct i2c_client *client) +static int edt_ft5x06_ts_remove(struct i2c_client *client) { const struct edt_ft5x06_platform_data *pdata = dev_get_platdata(&client->dev); @@ -891,7 +883,7 @@ static struct i2c_driver edt_ft5x06_ts_driver = { }, .id_table = edt_ft5x06_ts_id, .probe = edt_ft5x06_ts_probe, - .remove = __devexit_p(edt_ft5x06_ts_remove), + .remove = edt_ft5x06_ts_remove, }; module_i2c_driver(edt_ft5x06_ts_driver); diff --git a/drivers/input/touchscreen/eeti_ts.c b/drivers/input/touchscreen/eeti_ts.c index 908407efc672..55255a940072 100644 --- a/drivers/input/touchscreen/eeti_ts.c +++ b/drivers/input/touchscreen/eeti_ts.c @@ -154,7 +154,7 @@ static void eeti_ts_close(struct input_dev *dev) eeti_ts_stop(priv); } -static int __devinit eeti_ts_probe(struct i2c_client *client, +static int eeti_ts_probe(struct i2c_client *client, const struct i2c_device_id *idp) { struct eeti_ts_platform_data *pdata = client->dev.platform_data; @@ -248,7 +248,7 @@ err0: return err; } -static int __devexit eeti_ts_remove(struct i2c_client *client) +static int eeti_ts_remove(struct i2c_client *client) { struct eeti_ts_priv *priv = i2c_get_clientdata(client); @@ -321,7 +321,7 @@ static struct i2c_driver eeti_ts_driver = { #endif }, .probe = eeti_ts_probe, - .remove = __devexit_p(eeti_ts_remove), + .remove = eeti_ts_remove, .id_table = eeti_ts_id, }; diff --git a/drivers/input/touchscreen/egalax_ts.c b/drivers/input/touchscreen/egalax_ts.c index 13fa62fdfb0b..17c9097f3b5d 100644 --- a/drivers/input/touchscreen/egalax_ts.c +++ b/drivers/input/touchscreen/egalax_ts.c @@ -153,7 +153,7 @@ static int egalax_wake_up_device(struct i2c_client *client) return 0; } -static int __devinit egalax_firmware_version(struct i2c_client *client) +static int egalax_firmware_version(struct i2c_client *client) { static const u8 cmd[MAX_I2C_DATA_LEN] = { 0x03, 0x03, 0xa, 0x01, 0x41 }; int ret; @@ -165,7 +165,7 @@ static int __devinit egalax_firmware_version(struct i2c_client *client) return 0; } -static int __devinit egalax_ts_probe(struct i2c_client *client, +static int egalax_ts_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct egalax_ts *ts; @@ -246,7 +246,7 @@ err_free_ts: return error; } -static __devexit int egalax_ts_remove(struct i2c_client *client) +static int egalax_ts_remove(struct i2c_client *client) { struct egalax_ts *ts = i2c_get_clientdata(client); @@ -301,7 +301,7 @@ static struct i2c_driver egalax_ts_driver = { }, .id_table = egalax_ts_id, .probe = egalax_ts_probe, - .remove = __devexit_p(egalax_ts_remove), + .remove = egalax_ts_remove, }; module_i2c_driver(egalax_ts_driver); diff --git a/drivers/input/touchscreen/h3600_ts_input.c b/drivers/input/touchscreen/h3600_ts_input.c deleted file mode 100644 index b9e8686a6f1c..000000000000 --- a/drivers/input/touchscreen/h3600_ts_input.c +++ /dev/null @@ -1,479 +0,0 @@ -/* - * Copyright (c) 2001 "Crazy" James Simmons jsimmons@transvirtual.com - * - * Sponsored by Transvirtual Technology. - * - * Derived from the code in h3600_ts.[ch] by Charles Flynn - */ - -/* - * Driver for the h3600 Touch Screen and other Atmel controlled devices. - */ - -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Should you need to contact me, the author, you can do so by - * e-mail - mail your message to <jsimmons@transvirtual.com>. - */ - -#include <linux/errno.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/input.h> -#include <linux/serio.h> -#include <linux/init.h> -#include <linux/delay.h> - -/* SA1100 serial defines */ -#include <mach/hardware.h> -#include <mach/irqs.h> - -#define DRIVER_DESC "H3600 touchscreen driver" - -MODULE_AUTHOR("James Simmons <jsimmons@transvirtual.com>"); -MODULE_DESCRIPTION(DRIVER_DESC); -MODULE_LICENSE("GPL"); - -/* - * Definitions & global arrays. - */ - -/* The start and end of frame characters SOF and EOF */ -#define CHAR_SOF 0x02 -#define CHAR_EOF 0x03 -#define FRAME_OVERHEAD 3 /* CHAR_SOF,CHAR_EOF,LENGTH = 3 */ - -/* - Atmel events and response IDs contained in frame. - Programmer has no control over these numbers. - TODO there are holes - specifically 1,7,0x0a -*/ -#define VERSION_ID 0 /* Get Version (request/response) */ -#define KEYBD_ID 2 /* Keyboard (event) */ -#define TOUCHS_ID 3 /* Touch Screen (event)*/ -#define EEPROM_READ_ID 4 /* (request/response) */ -#define EEPROM_WRITE_ID 5 /* (request/response) */ -#define THERMAL_ID 6 /* (request/response) */ -#define NOTIFY_LED_ID 8 /* (request/response) */ -#define BATTERY_ID 9 /* (request/response) */ -#define SPI_READ_ID 0x0b /* ( request/response) */ -#define SPI_WRITE_ID 0x0c /* ( request/response) */ -#define FLITE_ID 0x0d /* backlight ( request/response) */ -#define STX_ID 0xa1 /* extension pack status (req/resp) */ - -#define MAX_ID 14 - -#define H3600_MAX_LENGTH 16 -#define H3600_KEY 0xf - -#define H3600_SCANCODE_RECORD 1 /* 1 -> record button */ -#define H3600_SCANCODE_CALENDAR 2 /* 2 -> calendar */ -#define H3600_SCANCODE_CONTACTS 3 /* 3 -> contact */ -#define H3600_SCANCODE_Q 4 /* 4 -> Q button */ -#define H3600_SCANCODE_START 5 /* 5 -> start menu */ -#define H3600_SCANCODE_UP 6 /* 6 -> up */ -#define H3600_SCANCODE_RIGHT 7 /* 7 -> right */ -#define H3600_SCANCODE_LEFT 8 /* 8 -> left */ -#define H3600_SCANCODE_DOWN 9 /* 9 -> down */ - -/* - * Per-touchscreen data. - */ -struct h3600_dev { - struct input_dev *dev; - struct serio *serio; - unsigned char event; /* event ID from packet */ - unsigned char chksum; - unsigned char len; - unsigned char idx; - unsigned char buf[H3600_MAX_LENGTH]; - char phys[32]; -}; - -static irqreturn_t action_button_handler(int irq, void *dev_id) -{ - int down = (GPLR & GPIO_BITSY_ACTION_BUTTON) ? 0 : 1; - struct input_dev *dev = dev_id; - - input_report_key(dev, KEY_ENTER, down); - input_sync(dev); - - return IRQ_HANDLED; -} - -static irqreturn_t npower_button_handler(int irq, void *dev_id) -{ - int down = (GPLR & GPIO_BITSY_NPOWER_BUTTON) ? 0 : 1; - struct input_dev *dev = dev_id; - - /* - * This interrupt is only called when we release the key. So we have - * to fake a key press. - */ - input_report_key(dev, KEY_SUSPEND, 1); - input_report_key(dev, KEY_SUSPEND, down); - input_sync(dev); - - return IRQ_HANDLED; -} - -#ifdef CONFIG_PM - -static int flite_brightness = 25; - -enum flite_pwr { - FLITE_PWR_OFF = 0, - FLITE_PWR_ON = 1 -}; - -/* - * h3600_flite_power: enables or disables power to frontlight, using last bright */ -unsigned int h3600_flite_power(struct input_dev *dev, enum flite_pwr pwr) -{ - unsigned char brightness = (pwr == FLITE_PWR_OFF) ? 0 : flite_brightness; - struct h3600_dev *ts = input_get_drvdata(dev); - - /* Must be in this order */ - serio_write(ts->serio, 1); - serio_write(ts->serio, pwr); - serio_write(ts->serio, brightness); - - return 0; -} - -#endif - -/* - * This function translates the native event packets to linux input event - * packets. Some packets coming from serial are not touchscreen related. In - * this case we send them off to be processed elsewhere. - */ -static void h3600ts_process_packet(struct h3600_dev *ts) -{ - struct input_dev *dev = ts->dev; - static int touched = 0; - int key, down = 0; - - switch (ts->event) { - /* - Buttons - returned as a single byte - 7 6 5 4 3 2 1 0 - S x x x N N N N - - S switch state ( 0=pressed 1=released) - x Unused. - NNNN switch number 0-15 - - Note: This is true for non interrupt generated key events. - */ - case KEYBD_ID: - down = (ts->buf[0] & 0x80) ? 0 : 1; - - switch (ts->buf[0] & 0x7f) { - case H3600_SCANCODE_RECORD: - key = KEY_RECORD; - break; - case H3600_SCANCODE_CALENDAR: - key = KEY_PROG1; - break; - case H3600_SCANCODE_CONTACTS: - key = KEY_PROG2; - break; - case H3600_SCANCODE_Q: - key = KEY_Q; - break; - case H3600_SCANCODE_START: - key = KEY_PROG3; - break; - case H3600_SCANCODE_UP: - key = KEY_UP; - break; - case H3600_SCANCODE_RIGHT: - key = KEY_RIGHT; - break; - case H3600_SCANCODE_LEFT: - key = KEY_LEFT; - break; - case H3600_SCANCODE_DOWN: - key = KEY_DOWN; - break; - default: - key = 0; - } - if (key) - input_report_key(dev, key, down); - break; - /* - * Native touchscreen event data is formatted as shown below:- - * - * +-------+-------+-------+-------+ - * | Xmsb | Xlsb | Ymsb | Ylsb | - * +-------+-------+-------+-------+ - * byte 0 1 2 3 - */ - case TOUCHS_ID: - if (!touched) { - input_report_key(dev, BTN_TOUCH, 1); - touched = 1; - } - - if (ts->len) { - unsigned short x, y; - - x = ts->buf[0]; x <<= 8; x += ts->buf[1]; - y = ts->buf[2]; y <<= 8; y += ts->buf[3]; - - input_report_abs(dev, ABS_X, x); - input_report_abs(dev, ABS_Y, y); - } else { - input_report_key(dev, BTN_TOUCH, 0); - touched = 0; - } - break; - default: - /* Send a non input event elsewhere */ - break; - } - - input_sync(dev); -} - -/* - * h3600ts_event() handles events from the input module. - */ -static int h3600ts_event(struct input_dev *dev, unsigned int type, - unsigned int code, int value) -{ -#if 0 - struct h3600_dev *ts = input_get_drvdata(dev); - - switch (type) { - case EV_LED: { - // serio_write(ts->serio, SOME_CMD); - return 0; - } - } - return -1; -#endif - return 0; -} - -/* - Frame format - byte 1 2 3 len + 4 - +-------+---------------+---------------+--=------------+ - |SOF |id |len | len bytes | Chksum | - +-------+---------------+---------------+--=------------+ - bit 0 7 8 11 12 15 16 - - +-------+---------------+-------+ - |SOF |id |0 |Chksum | - Note Chksum does not include SOF - +-------+---------------+-------+ - bit 0 7 8 11 12 15 16 - -*/ - -static int state; - -/* decode States */ -#define STATE_SOF 0 /* start of FRAME */ -#define STATE_ID 1 /* state where we decode the ID & len */ -#define STATE_DATA 2 /* state where we decode data */ -#define STATE_EOF 3 /* state where we decode checksum or EOF */ - -static irqreturn_t h3600ts_interrupt(struct serio *serio, unsigned char data, - unsigned int flags) -{ - struct h3600_dev *ts = serio_get_drvdata(serio); - - /* - * We have a new frame coming in. - */ - switch (state) { - case STATE_SOF: - if (data == CHAR_SOF) - state = STATE_ID; - break; - case STATE_ID: - ts->event = (data & 0xf0) >> 4; - ts->len = (data & 0xf); - ts->idx = 0; - if (ts->event >= MAX_ID) { - state = STATE_SOF; - break; - } - ts->chksum = data; - state = (ts->len > 0) ? STATE_DATA : STATE_EOF; - break; - case STATE_DATA: - ts->chksum += data; - ts->buf[ts->idx]= data; - if (++ts->idx == ts->len) - state = STATE_EOF; - break; - case STATE_EOF: - state = STATE_SOF; - if (data == CHAR_EOF || data == ts->chksum) - h3600ts_process_packet(ts); - break; - default: - printk("Error3\n"); - break; - } - - return IRQ_HANDLED; -} - -/* - * h3600ts_connect() is the routine that is called when someone adds a - * new serio device that supports H3600 protocol and registers it as - * an input device. - */ -static int h3600ts_connect(struct serio *serio, struct serio_driver *drv) -{ - struct h3600_dev *ts; - struct input_dev *input_dev; - int err; - - ts = kzalloc(sizeof(struct h3600_dev), GFP_KERNEL); - input_dev = input_allocate_device(); - if (!ts || !input_dev) { - err = -ENOMEM; - goto fail1; - } - - ts->serio = serio; - ts->dev = input_dev; - snprintf(ts->phys, sizeof(ts->phys), "%s/input0", serio->phys); - - input_dev->name = "H3600 TouchScreen"; - input_dev->phys = ts->phys; - input_dev->id.bustype = BUS_RS232; - input_dev->id.vendor = SERIO_H3600; - input_dev->id.product = 0x0666; /* FIXME !!! We can ask the hardware */ - input_dev->id.version = 0x0100; - input_dev->dev.parent = &serio->dev; - - input_set_drvdata(input_dev, ts); - - input_dev->event = h3600ts_event; - - input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS) | - BIT_MASK(EV_LED) | BIT_MASK(EV_PWR); - input_dev->ledbit[0] = BIT_MASK(LED_SLEEP); - input_set_abs_params(input_dev, ABS_X, 60, 985, 0, 0); - input_set_abs_params(input_dev, ABS_Y, 35, 1024, 0, 0); - - set_bit(KEY_RECORD, input_dev->keybit); - set_bit(KEY_Q, input_dev->keybit); - set_bit(KEY_PROG1, input_dev->keybit); - set_bit(KEY_PROG2, input_dev->keybit); - set_bit(KEY_PROG3, input_dev->keybit); - set_bit(KEY_UP, input_dev->keybit); - set_bit(KEY_RIGHT, input_dev->keybit); - set_bit(KEY_LEFT, input_dev->keybit); - set_bit(KEY_DOWN, input_dev->keybit); - set_bit(KEY_ENTER, input_dev->keybit); - set_bit(KEY_SUSPEND, input_dev->keybit); - set_bit(BTN_TOUCH, input_dev->keybit); - - /* Device specific stuff */ - set_GPIO_IRQ_edge(GPIO_BITSY_ACTION_BUTTON, GPIO_BOTH_EDGES); - set_GPIO_IRQ_edge(GPIO_BITSY_NPOWER_BUTTON, GPIO_RISING_EDGE); - - if (request_irq(IRQ_GPIO_BITSY_ACTION_BUTTON, action_button_handler, - IRQF_SHARED, "h3600_action", ts->dev)) { - printk(KERN_ERR "h3600ts.c: Could not allocate Action Button IRQ!\n"); - err = -EBUSY; - goto fail1; - } - - if (request_irq(IRQ_GPIO_BITSY_NPOWER_BUTTON, npower_button_handler, - IRQF_SHARED, "h3600_suspend", ts->dev)) { - printk(KERN_ERR "h3600ts.c: Could not allocate Power Button IRQ!\n"); - err = -EBUSY; - goto fail2; - } - - serio_set_drvdata(serio, ts); - - err = serio_open(serio, drv); - if (err) - goto fail3; - - //h3600_flite_control(1, 25); /* default brightness */ - err = input_register_device(ts->dev); - if (err) - goto fail4; - - return 0; - -fail4: serio_close(serio); -fail3: serio_set_drvdata(serio, NULL); - free_irq(IRQ_GPIO_BITSY_NPOWER_BUTTON, ts->dev); -fail2: free_irq(IRQ_GPIO_BITSY_ACTION_BUTTON, ts->dev); -fail1: input_free_device(input_dev); - kfree(ts); - return err; -} - -/* - * h3600ts_disconnect() is the opposite of h3600ts_connect() - */ - -static void h3600ts_disconnect(struct serio *serio) -{ - struct h3600_dev *ts = serio_get_drvdata(serio); - - free_irq(IRQ_GPIO_BITSY_ACTION_BUTTON, ts->dev); - free_irq(IRQ_GPIO_BITSY_NPOWER_BUTTON, ts->dev); - input_get_device(ts->dev); - input_unregister_device(ts->dev); - serio_close(serio); - serio_set_drvdata(serio, NULL); - input_put_device(ts->dev); - kfree(ts); -} - -/* - * The serio driver structure. - */ - -static struct serio_device_id h3600ts_serio_ids[] = { - { - .type = SERIO_RS232, - .proto = SERIO_H3600, - .id = SERIO_ANY, - .extra = SERIO_ANY, - }, - { 0 } -}; - -MODULE_DEVICE_TABLE(serio, h3600ts_serio_ids); - -static struct serio_driver h3600ts_drv = { - .driver = { - .name = "h3600ts", - }, - .description = DRIVER_DESC, - .id_table = h3600ts_serio_ids, - .interrupt = h3600ts_interrupt, - .connect = h3600ts_connect, - .disconnect = h3600ts_disconnect, -}; - -module_serio_driver(h3600ts_drv); diff --git a/drivers/input/touchscreen/htcpen.c b/drivers/input/touchscreen/htcpen.c index d13143b68b3e..6c4fb8436957 100644 --- a/drivers/input/touchscreen/htcpen.c +++ b/drivers/input/touchscreen/htcpen.c @@ -102,7 +102,7 @@ static void htcpen_close(struct input_dev *dev) synchronize_irq(HTCPEN_IRQ); } -static int __devinit htcpen_isa_probe(struct device *dev, unsigned int id) +static int htcpen_isa_probe(struct device *dev, unsigned int id) { struct input_dev *htcpen_dev; int err = -EBUSY; @@ -174,7 +174,7 @@ static int __devinit htcpen_isa_probe(struct device *dev, unsigned int id) return err; } -static int __devexit htcpen_isa_remove(struct device *dev, unsigned int id) +static int htcpen_isa_remove(struct device *dev, unsigned int id) { struct input_dev *htcpen_dev = dev_get_drvdata(dev); @@ -210,7 +210,7 @@ static int htcpen_isa_resume(struct device *dev, unsigned int n) static struct isa_driver htcpen_isa_driver = { .probe = htcpen_isa_probe, - .remove = __devexit_p(htcpen_isa_remove), + .remove = htcpen_isa_remove, #ifdef CONFIG_PM .suspend = htcpen_isa_suspend, .resume = htcpen_isa_resume, diff --git a/drivers/input/touchscreen/ili210x.c b/drivers/input/touchscreen/ili210x.c index 4ac69760ec08..1418bdda61bb 100644 --- a/drivers/input/touchscreen/ili210x.c +++ b/drivers/input/touchscreen/ili210x.c @@ -180,7 +180,7 @@ static const struct attribute_group ili210x_attr_group = { .attrs = ili210x_attributes, }; -static int __devinit ili210x_i2c_probe(struct i2c_client *client, +static int ili210x_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct device *dev = &client->dev; @@ -298,7 +298,7 @@ err_free_mem: return error; } -static int __devexit ili210x_i2c_remove(struct i2c_client *client) +static int ili210x_i2c_remove(struct i2c_client *client) { struct ili210x *priv = i2c_get_clientdata(client); @@ -350,7 +350,7 @@ static struct i2c_driver ili210x_ts_driver = { }, .id_table = ili210x_i2c_id, .probe = ili210x_i2c_probe, - .remove = __devexit_p(ili210x_i2c_remove), + .remove = ili210x_i2c_remove, }; module_i2c_driver(ili210x_ts_driver); diff --git a/drivers/input/touchscreen/intel-mid-touch.c b/drivers/input/touchscreen/intel-mid-touch.c index cf299377fc49..465db5dba8b4 100644 --- a/drivers/input/touchscreen/intel-mid-touch.c +++ b/drivers/input/touchscreen/intel-mid-touch.c @@ -427,7 +427,7 @@ out: } /* Utility to read PMIC ID */ -static int __devinit mrstouch_read_pmic_id(uint *vendor, uint *rev) +static int mrstouch_read_pmic_id(uint *vendor, uint *rev) { int err; u8 r; @@ -446,7 +446,7 @@ static int __devinit mrstouch_read_pmic_id(uint *vendor, uint *rev) * Parse ADC channels to find end of the channel configured by other ADC user * NEC and MAXIM requires 4 channels and FreeScale needs 18 channels */ -static int __devinit mrstouch_chan_parse(struct mrstouch_dev *tsdev) +static int mrstouch_chan_parse(struct mrstouch_dev *tsdev) { int found = 0; int err, i; @@ -478,7 +478,7 @@ static int __devinit mrstouch_chan_parse(struct mrstouch_dev *tsdev) /* * Writes touch screen channels to ADC address selection registers */ -static int __devinit mrstouch_ts_chan_set(uint offset) +static int mrstouch_ts_chan_set(uint offset) { u16 chan; @@ -494,7 +494,7 @@ static int __devinit mrstouch_ts_chan_set(uint offset) } /* Initialize ADC */ -static int __devinit mrstouch_adc_init(struct mrstouch_dev *tsdev) +static int mrstouch_adc_init(struct mrstouch_dev *tsdev) { int err, start; u8 ra, rm; @@ -568,7 +568,7 @@ static int __devinit mrstouch_adc_init(struct mrstouch_dev *tsdev) /* Probe function for touch screen driver */ -static int __devinit mrstouch_probe(struct platform_device *pdev) +static int mrstouch_probe(struct platform_device *pdev) { struct mrstouch_dev *tsdev; struct input_dev *input; @@ -643,7 +643,7 @@ err_free_mem: return err; } -static int __devexit mrstouch_remove(struct platform_device *pdev) +static int mrstouch_remove(struct platform_device *pdev) { struct mrstouch_dev *tsdev = platform_get_drvdata(pdev); @@ -662,7 +662,7 @@ static struct platform_driver mrstouch_driver = { .owner = THIS_MODULE, }, .probe = mrstouch_probe, - .remove = __devexit_p(mrstouch_remove), + .remove = mrstouch_remove, }; module_platform_driver(mrstouch_driver); diff --git a/drivers/input/touchscreen/jornada720_ts.c b/drivers/input/touchscreen/jornada720_ts.c index 7f03d1bd916e..282d7c7ad2fc 100644 --- a/drivers/input/touchscreen/jornada720_ts.c +++ b/drivers/input/touchscreen/jornada720_ts.c @@ -99,7 +99,7 @@ static irqreturn_t jornada720_ts_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit jornada720_ts_probe(struct platform_device *pdev) +static int jornada720_ts_probe(struct platform_device *pdev) { struct jornada_ts *jornada_ts; struct input_dev *input_dev; @@ -151,7 +151,7 @@ static int __devinit jornada720_ts_probe(struct platform_device *pdev) return error; } -static int __devexit jornada720_ts_remove(struct platform_device *pdev) +static int jornada720_ts_remove(struct platform_device *pdev) { struct jornada_ts *jornada_ts = platform_get_drvdata(pdev); @@ -168,7 +168,7 @@ MODULE_ALIAS("platform:jornada_ts"); static struct platform_driver jornada720_ts_driver = { .probe = jornada720_ts_probe, - .remove = __devexit_p(jornada720_ts_remove), + .remove = jornada720_ts_remove, .driver = { .name = "jornada_ts", .owner = THIS_MODULE, diff --git a/drivers/input/touchscreen/lpc32xx_ts.c b/drivers/input/touchscreen/lpc32xx_ts.c index 4c2b8ed3bf16..9101ee529c92 100644 --- a/drivers/input/touchscreen/lpc32xx_ts.c +++ b/drivers/input/touchscreen/lpc32xx_ts.c @@ -203,7 +203,7 @@ static void lpc32xx_ts_close(struct input_dev *dev) lpc32xx_stop_tsc(tsc); } -static int __devinit lpc32xx_ts_probe(struct platform_device *pdev) +static int lpc32xx_ts_probe(struct platform_device *pdev) { struct lpc32xx_tsc *tsc; struct input_dev *input; @@ -309,7 +309,7 @@ err_free_mem: return error; } -static int __devexit lpc32xx_ts_remove(struct platform_device *pdev) +static int lpc32xx_ts_remove(struct platform_device *pdev) { struct lpc32xx_tsc *tsc = platform_get_drvdata(pdev); struct resource *res; @@ -394,7 +394,7 @@ MODULE_DEVICE_TABLE(of, lpc32xx_tsc_of_match); static struct platform_driver lpc32xx_ts_driver = { .probe = lpc32xx_ts_probe, - .remove = __devexit_p(lpc32xx_ts_remove), + .remove = lpc32xx_ts_remove, .driver = { .name = MOD_NAME, .owner = THIS_MODULE, diff --git a/drivers/input/touchscreen/max11801_ts.c b/drivers/input/touchscreen/max11801_ts.c index 4eab50b856d7..00bc6caa27f5 100644 --- a/drivers/input/touchscreen/max11801_ts.c +++ b/drivers/input/touchscreen/max11801_ts.c @@ -156,7 +156,7 @@ out: return IRQ_HANDLED; } -static void __devinit max11801_ts_phy_init(struct max11801_data *data) +static void max11801_ts_phy_init(struct max11801_data *data) { struct i2c_client *client = data->client; @@ -174,7 +174,7 @@ static void __devinit max11801_ts_phy_init(struct max11801_data *data) max11801_write_reg(client, OP_MODE_CONF_REG, 0x36); } -static int __devinit max11801_ts_probe(struct i2c_client *client, +static int max11801_ts_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct max11801_data *data; @@ -228,7 +228,7 @@ err_free_mem: return error; } -static __devexit int max11801_ts_remove(struct i2c_client *client) +static int max11801_ts_remove(struct i2c_client *client) { struct max11801_data *data = i2c_get_clientdata(client); @@ -252,7 +252,7 @@ static struct i2c_driver max11801_ts_driver = { }, .id_table = max11801_ts_id, .probe = max11801_ts_probe, - .remove = __devexit_p(max11801_ts_remove), + .remove = max11801_ts_remove, }; module_i2c_driver(max11801_ts_driver); diff --git a/drivers/input/touchscreen/mc13783_ts.c b/drivers/input/touchscreen/mc13783_ts.c index 48dc5b0d26f1..02103b6abb39 100644 --- a/drivers/input/touchscreen/mc13783_ts.c +++ b/drivers/input/touchscreen/mc13783_ts.c @@ -229,7 +229,7 @@ err_free_mem: return ret; } -static int __devexit mc13783_ts_remove(struct platform_device *pdev) +static int mc13783_ts_remove(struct platform_device *pdev) { struct mc13783_ts_priv *priv = platform_get_drvdata(pdev); @@ -243,7 +243,7 @@ static int __devexit mc13783_ts_remove(struct platform_device *pdev) } static struct platform_driver mc13783_ts_driver = { - .remove = __devexit_p(mc13783_ts_remove), + .remove = mc13783_ts_remove, .driver = { .owner = THIS_MODULE, .name = MC13783_TS_NAME, diff --git a/drivers/input/touchscreen/mcs5000_ts.c b/drivers/input/touchscreen/mcs5000_ts.c index b528511861ce..f9f4e0c56eda 100644 --- a/drivers/input/touchscreen/mcs5000_ts.c +++ b/drivers/input/touchscreen/mcs5000_ts.c @@ -187,7 +187,7 @@ static void mcs5000_ts_phys_init(struct mcs5000_ts_data *data) OP_MODE_ACTIVE | REPORT_RATE_80); } -static int __devinit mcs5000_ts_probe(struct i2c_client *client, +static int mcs5000_ts_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct mcs5000_ts_data *data; @@ -249,7 +249,7 @@ err_free_mem: return ret; } -static int __devexit mcs5000_ts_remove(struct i2c_client *client) +static int mcs5000_ts_remove(struct i2c_client *client) { struct mcs5000_ts_data *data = i2c_get_clientdata(client); @@ -292,7 +292,7 @@ MODULE_DEVICE_TABLE(i2c, mcs5000_ts_id); static struct i2c_driver mcs5000_ts_driver = { .probe = mcs5000_ts_probe, - .remove = __devexit_p(mcs5000_ts_remove), + .remove = mcs5000_ts_remove, .driver = { .name = "mcs5000_ts", #ifdef CONFIG_PM diff --git a/drivers/input/touchscreen/mms114.c b/drivers/input/touchscreen/mms114.c index 560cf09d1c5a..98841d8aa635 100644 --- a/drivers/input/touchscreen/mms114.c +++ b/drivers/input/touchscreen/mms114.c @@ -10,6 +10,7 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/delay.h> +#include <linux/of.h> #include <linux/i2c.h> #include <linux/i2c/mms114.h> #include <linux/input/mt.h> @@ -360,14 +361,63 @@ static void mms114_input_close(struct input_dev *dev) mms114_stop(data); } -static int __devinit mms114_probe(struct i2c_client *client, +#ifdef CONFIG_OF +static struct mms114_platform_data *mms114_parse_dt(struct device *dev) +{ + struct mms114_platform_data *pdata; + struct device_node *np = dev->of_node; + + if (!np) + return NULL; + + pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL); + if (!pdata) { + dev_err(dev, "failed to allocate platform data\n"); + return NULL; + } + + if (of_property_read_u32(np, "x-size", &pdata->x_size)) { + dev_err(dev, "failed to get x-size property\n"); + return NULL; + }; + + if (of_property_read_u32(np, "y-size", &pdata->y_size)) { + dev_err(dev, "failed to get y-size property\n"); + return NULL; + }; + + of_property_read_u32(np, "contact-threshold", + &pdata->contact_threshold); + of_property_read_u32(np, "moving-threshold", + &pdata->moving_threshold); + + if (of_find_property(np, "x-invert", NULL)) + pdata->x_invert = true; + if (of_find_property(np, "y-invert", NULL)) + pdata->y_invert = true; + + return pdata; +} +#else +static inline struct mms114_platform_data *mms114_parse_dt(struct device *dev) +{ + return NULL; +} +#endif + +static int mms114_probe(struct i2c_client *client, const struct i2c_device_id *id) { + const struct mms114_platform_data *pdata; struct mms114_data *data; struct input_dev *input_dev; int error; - if (!client->dev.platform_data) { + pdata = dev_get_platdata(&client->dev); + if (!pdata) + pdata = mms114_parse_dt(&client->dev); + + if (!pdata) { dev_err(&client->dev, "Need platform data\n"); return -EINVAL; } @@ -389,7 +439,7 @@ static int __devinit mms114_probe(struct i2c_client *client, data->client = client; data->input_dev = input_dev; - data->pdata = client->dev.platform_data; + data->pdata = pdata; input_dev->name = "MELPAS MMS114 Touchscreen"; input_dev->id.bustype = BUS_I2C; @@ -458,7 +508,7 @@ err_free_mem: return error; } -static int __devexit mms114_remove(struct i2c_client *client) +static int mms114_remove(struct i2c_client *client) { struct mms114_data *data = i2c_get_clientdata(client); @@ -525,14 +575,22 @@ static const struct i2c_device_id mms114_id[] = { }; MODULE_DEVICE_TABLE(i2c, mms114_id); +#ifdef CONFIG_OF +static struct of_device_id mms114_dt_match[] = { + { .compatible = "melfas,mms114" }, + { } +}; +#endif + static struct i2c_driver mms114_driver = { .driver = { .name = "mms114", .owner = THIS_MODULE, .pm = &mms114_pm_ops, + .of_match_table = of_match_ptr(mms114_dt_match), }, .probe = mms114_probe, - .remove = __devexit_p(mms114_remove), + .remove = mms114_remove, .id_table = mms114_id, }; diff --git a/drivers/input/touchscreen/pcap_ts.c b/drivers/input/touchscreen/pcap_ts.c index f57aeb80f7e3..f22e04dd4e16 100644 --- a/drivers/input/touchscreen/pcap_ts.c +++ b/drivers/input/touchscreen/pcap_ts.c @@ -137,7 +137,7 @@ static void pcap_ts_close(struct input_dev *dev) pcap_ts->read_state << PCAP_ADC_TS_M_SHIFT); } -static int __devinit pcap_ts_probe(struct platform_device *pdev) +static int pcap_ts_probe(struct platform_device *pdev) { struct input_dev *input_dev; struct pcap_ts *pcap_ts; @@ -202,7 +202,7 @@ fail: return err; } -static int __devexit pcap_ts_remove(struct platform_device *pdev) +static int pcap_ts_remove(struct platform_device *pdev) { struct pcap_ts *pcap_ts = platform_get_drvdata(pdev); @@ -245,7 +245,7 @@ static const struct dev_pm_ops pcap_ts_pm_ops = { static struct platform_driver pcap_ts_driver = { .probe = pcap_ts_probe, - .remove = __devexit_p(pcap_ts_remove), + .remove = pcap_ts_remove, .driver = { .name = "pcap-ts", .owner = THIS_MODULE, diff --git a/drivers/input/touchscreen/pixcir_i2c_ts.c b/drivers/input/touchscreen/pixcir_i2c_ts.c index 953b4c105cad..6cc6b36663ff 100644 --- a/drivers/input/touchscreen/pixcir_i2c_ts.c +++ b/drivers/input/touchscreen/pixcir_i2c_ts.c @@ -125,7 +125,7 @@ static int pixcir_i2c_ts_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(pixcir_dev_pm_ops, pixcir_i2c_ts_suspend, pixcir_i2c_ts_resume); -static int __devinit pixcir_i2c_ts_probe(struct i2c_client *client, +static int pixcir_i2c_ts_probe(struct i2c_client *client, const struct i2c_device_id *id) { const struct pixcir_ts_platform_data *pdata = client->dev.platform_data; @@ -189,7 +189,7 @@ err_free_mem: return error; } -static int __devexit pixcir_i2c_ts_remove(struct i2c_client *client) +static int pixcir_i2c_ts_remove(struct i2c_client *client) { struct pixcir_i2c_ts_data *tsdata = i2c_get_clientdata(client); @@ -218,7 +218,7 @@ static struct i2c_driver pixcir_i2c_ts_driver = { .pm = &pixcir_dev_pm_ops, }, .probe = pixcir_i2c_ts_probe, - .remove = __devexit_p(pixcir_i2c_ts_remove), + .remove = pixcir_i2c_ts_remove, .id_table = pixcir_i2c_ts_id, }; diff --git a/drivers/input/touchscreen/s3c2410_ts.c b/drivers/input/touchscreen/s3c2410_ts.c index 549fa29548f8..b061af2c8376 100644 --- a/drivers/input/touchscreen/s3c2410_ts.c +++ b/drivers/input/touchscreen/s3c2410_ts.c @@ -238,7 +238,7 @@ static void s3c24xx_ts_select(struct s3c_adc_client *client, unsigned select) * Initialise, find and allocate any resources we need to run and then * register with the ADC and input systems. */ -static int __devinit s3c2410ts_probe(struct platform_device *pdev) +static int s3c2410ts_probe(struct platform_device *pdev) { struct s3c2410_ts_mach_info *info; struct device *dev = &pdev->dev; @@ -365,7 +365,7 @@ static int __devinit s3c2410ts_probe(struct platform_device *pdev) * * Free up our state ready to be removed. */ -static int __devexit s3c2410ts_remove(struct platform_device *pdev) +static int s3c2410ts_remove(struct platform_device *pdev) { free_irq(ts.irq_tc, ts.input); del_timer_sync(&touch_timer); @@ -430,7 +430,7 @@ static struct platform_driver s3c_ts_driver = { }, .id_table = s3cts_driver_ids, .probe = s3c2410ts_probe, - .remove = __devexit_p(s3c2410ts_remove), + .remove = s3c2410ts_remove, }; module_platform_driver(s3c_ts_driver); diff --git a/drivers/input/touchscreen/st1232.c b/drivers/input/touchscreen/st1232.c index 6cb68a1981bf..d9d05e222428 100644 --- a/drivers/input/touchscreen/st1232.c +++ b/drivers/input/touchscreen/st1232.c @@ -139,7 +139,7 @@ end: return IRQ_HANDLED; } -static int __devinit st1232_ts_probe(struct i2c_client *client, +static int st1232_ts_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct st1232_ts_data *ts; @@ -206,7 +206,7 @@ err_free_mem: return error; } -static int __devexit st1232_ts_remove(struct i2c_client *client) +static int st1232_ts_remove(struct i2c_client *client) { struct st1232_ts_data *ts = i2c_get_clientdata(client); @@ -255,7 +255,7 @@ static const struct i2c_device_id st1232_ts_id[] = { MODULE_DEVICE_TABLE(i2c, st1232_ts_id); #ifdef CONFIG_OF -static const struct of_device_id st1232_ts_dt_ids[] __devinitconst = { +static const struct of_device_id st1232_ts_dt_ids[] = { { .compatible = "sitronix,st1232", }, { } }; @@ -264,7 +264,7 @@ MODULE_DEVICE_TABLE(of, st1232_ts_dt_ids); static struct i2c_driver st1232_ts_driver = { .probe = st1232_ts_probe, - .remove = __devexit_p(st1232_ts_remove), + .remove = st1232_ts_remove, .id_table = st1232_ts_id, .driver = { .name = ST1232_TS_NAME, diff --git a/drivers/input/touchscreen/stmpe-ts.c b/drivers/input/touchscreen/stmpe-ts.c index 692b685720ce..84d884b4ec3e 100644 --- a/drivers/input/touchscreen/stmpe-ts.c +++ b/drivers/input/touchscreen/stmpe-ts.c @@ -1,4 +1,5 @@ -/* STMicroelectronics STMPE811 Touchscreen Driver +/* + * STMicroelectronics STMPE811 Touchscreen Driver * * (C) 2010 Luotao Fu <l.fu@pengutronix.de> * All rights reserved. @@ -16,6 +17,7 @@ #include <linux/interrupt.h> #include <linux/init.h> #include <linux/device.h> +#include <linux/of.h> #include <linux/platform_device.h> #include <linux/input.h> #include <linux/slab.h> @@ -166,7 +168,7 @@ static irqreturn_t stmpe_ts_handler(int irq, void *data) return IRQ_HANDLED; } -static int __devinit stmpe_init_hw(struct stmpe_touch *ts) +static int stmpe_init_hw(struct stmpe_touch *ts) { int ret; u8 adc_ctrl1, adc_ctrl1_mask, tsc_cfg, tsc_cfg_mask; @@ -261,41 +263,18 @@ static void stmpe_ts_close(struct input_dev *dev) STMPE_TSC_CTRL_TSC_EN, 0); } -static int __devinit stmpe_input_probe(struct platform_device *pdev) +static void stmpe_ts_get_platform_info(struct platform_device *pdev, + struct stmpe_touch *ts) { struct stmpe *stmpe = dev_get_drvdata(pdev->dev.parent); - struct stmpe_platform_data *pdata = stmpe->pdata; - struct stmpe_touch *ts; - struct input_dev *idev; + struct device_node *np = pdev->dev.of_node; struct stmpe_ts_platform_data *ts_pdata = NULL; - int ret; - int ts_irq; - - ts_irq = platform_get_irq_byname(pdev, "FIFO_TH"); - if (ts_irq < 0) - return ts_irq; - - ts = kzalloc(sizeof(*ts), GFP_KERNEL); - if (!ts) { - ret = -ENOMEM; - goto err_out; - } - idev = input_allocate_device(); - if (!idev) { - ret = -ENOMEM; - goto err_free_ts; - } - - platform_set_drvdata(pdev, ts); ts->stmpe = stmpe; - ts->idev = idev; - ts->dev = &pdev->dev; - if (pdata) - ts_pdata = pdata->ts; + if (stmpe->pdata && stmpe->pdata->ts) { + ts_pdata = stmpe->pdata->ts; - if (ts_pdata) { ts->sample_time = ts_pdata->sample_time; ts->mod_12b = ts_pdata->mod_12b; ts->ref_sel = ts_pdata->ref_sel; @@ -305,22 +284,71 @@ static int __devinit stmpe_input_probe(struct platform_device *pdev) ts->settling = ts_pdata->settling; ts->fraction_z = ts_pdata->fraction_z; ts->i_drive = ts_pdata->i_drive; + } else if (np) { + u32 val; + + if (!of_property_read_u32(np, "st,sample-time", &val)) + ts->sample_time = val; + if (!of_property_read_u32(np, "st,mod-12b", &val)) + ts->mod_12b = val; + if (!of_property_read_u32(np, "st,ref-sel", &val)) + ts->ref_sel = val; + if (!of_property_read_u32(np, "st,adc-freq", &val)) + ts->adc_freq = val; + if (!of_property_read_u32(np, "st,ave-ctrl", &val)) + ts->ave_ctrl = val; + if (!of_property_read_u32(np, "st,touch-det-delay", &val)) + ts->touch_det_delay = val; + if (!of_property_read_u32(np, "st,settling", &val)) + ts->settling = val; + if (!of_property_read_u32(np, "st,fraction-z", &val)) + ts->fraction_z = val; + if (!of_property_read_u32(np, "st,i-drive", &val)) + ts->i_drive = val; } +} + +static int stmpe_input_probe(struct platform_device *pdev) +{ + struct stmpe_touch *ts; + struct input_dev *idev; + int error; + int ts_irq; + + ts_irq = platform_get_irq_byname(pdev, "FIFO_TH"); + if (ts_irq < 0) + return ts_irq; + + ts = devm_kzalloc(&pdev->dev, sizeof(*ts), GFP_KERNEL); + if (!ts) + return -ENOMEM; + + idev = devm_input_allocate_device(&pdev->dev); + if (!idev) + return -ENOMEM; + + platform_set_drvdata(pdev, ts); + ts->idev = idev; + ts->dev = &pdev->dev; + + stmpe_ts_get_platform_info(pdev, ts); INIT_DELAYED_WORK(&ts->work, stmpe_work); - ret = request_threaded_irq(ts_irq, NULL, stmpe_ts_handler, - IRQF_ONESHOT, STMPE_TS_NAME, ts); - if (ret) { + error = devm_request_threaded_irq(&pdev->dev, ts_irq, + NULL, stmpe_ts_handler, + IRQF_ONESHOT, STMPE_TS_NAME, ts); + if (error) { dev_err(&pdev->dev, "Failed to request IRQ %d\n", ts_irq); - goto err_free_input; + return error; } - ret = stmpe_init_hw(ts); - if (ret) - goto err_free_irq; + error = stmpe_init_hw(ts); + if (error) + return error; idev->name = STMPE_TS_NAME; + idev->phys = STMPE_TS_NAME"/input0"; idev->id.bustype = BUS_I2C; idev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS); idev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH); @@ -334,40 +362,21 @@ static int __devinit stmpe_input_probe(struct platform_device *pdev) input_set_abs_params(idev, ABS_Y, 0, XY_MASK, 0, 0); input_set_abs_params(idev, ABS_PRESSURE, 0x0, 0xff, 0, 0); - ret = input_register_device(idev); - if (ret) { + error = input_register_device(idev); + if (error) { dev_err(&pdev->dev, "Could not register input device\n"); - goto err_free_irq; + return error; } - return ret; - -err_free_irq: - free_irq(ts_irq, ts); -err_free_input: - input_free_device(idev); - platform_set_drvdata(pdev, NULL); -err_free_ts: - kfree(ts); -err_out: - return ret; + return 0; } -static int __devexit stmpe_ts_remove(struct platform_device *pdev) +static int stmpe_ts_remove(struct platform_device *pdev) { struct stmpe_touch *ts = platform_get_drvdata(pdev); - unsigned int ts_irq = platform_get_irq_byname(pdev, "FIFO_TH"); stmpe_disable(ts->stmpe, STMPE_BLOCK_TOUCHSCREEN); - free_irq(ts_irq, ts); - - platform_set_drvdata(pdev, NULL); - - input_unregister_device(ts->idev); - - kfree(ts); - return 0; } @@ -377,7 +386,7 @@ static struct platform_driver stmpe_ts_driver = { .owner = THIS_MODULE, }, .probe = stmpe_input_probe, - .remove = __devexit_p(stmpe_ts_remove), + .remove = stmpe_ts_remove, }; module_platform_driver(stmpe_ts_driver); diff --git a/drivers/input/touchscreen/ti_am335x_tsc.c b/drivers/input/touchscreen/ti_am335x_tsc.c index 7a18a8a15228..51e7b87827a4 100644 --- a/drivers/input/touchscreen/ti_am335x_tsc.c +++ b/drivers/input/touchscreen/ti_am335x_tsc.c @@ -258,7 +258,7 @@ static irqreturn_t titsc_irq(int irq, void *dev) * The functions for inserting/removing driver as a module. */ -static int __devinit titsc_probe(struct platform_device *pdev) +static int titsc_probe(struct platform_device *pdev) { struct titsc *ts_dev; struct input_dev *input_dev; @@ -327,7 +327,7 @@ err_free_mem: return err; } -static int __devexit titsc_remove(struct platform_device *pdev) +static int titsc_remove(struct platform_device *pdev) { struct ti_tscadc_dev *tscadc_dev = pdev->dev.platform_data; struct titsc *ts_dev = tscadc_dev->tsc; @@ -384,7 +384,7 @@ static const struct dev_pm_ops titsc_pm_ops = { static struct platform_driver ti_tsc_driver = { .probe = titsc_probe, - .remove = __devexit_p(titsc_remove), + .remove = titsc_remove, .driver = { .name = "tsc", .owner = THIS_MODULE, diff --git a/drivers/input/touchscreen/tnetv107x-ts.c b/drivers/input/touchscreen/tnetv107x-ts.c index 368d2c6cf780..acfb87607b87 100644 --- a/drivers/input/touchscreen/tnetv107x-ts.c +++ b/drivers/input/touchscreen/tnetv107x-ts.c @@ -243,7 +243,7 @@ static void tsc_stop(struct input_dev *dev) clk_disable(ts->clk); } -static int __devinit tsc_probe(struct platform_device *pdev) +static int tsc_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct tsc_data *ts; @@ -357,7 +357,7 @@ error_res: return error; } -static int __devexit tsc_remove(struct platform_device *pdev) +static int tsc_remove(struct platform_device *pdev) { struct tsc_data *ts = platform_get_drvdata(pdev); @@ -374,7 +374,7 @@ static int __devexit tsc_remove(struct platform_device *pdev) static struct platform_driver tsc_driver = { .probe = tsc_probe, - .remove = __devexit_p(tsc_remove), + .remove = tsc_remove, .driver.name = "tnetv107x-ts", .driver.owner = THIS_MODULE, }; diff --git a/drivers/input/touchscreen/tps6507x-ts.c b/drivers/input/touchscreen/tps6507x-ts.c index f7eda3d00fad..820a066c3b8a 100644 --- a/drivers/input/touchscreen/tps6507x-ts.c +++ b/drivers/input/touchscreen/tps6507x-ts.c @@ -345,7 +345,7 @@ err0: return error; } -static int __devexit tps6507x_ts_remove(struct platform_device *pdev) +static int tps6507x_ts_remove(struct platform_device *pdev) { struct tps6507x_dev *tps6507x_dev = platform_get_drvdata(pdev); struct tps6507x_ts *tsc = tps6507x_dev->ts; @@ -367,7 +367,7 @@ static struct platform_driver tps6507x_ts_driver = { .owner = THIS_MODULE, }, .probe = tps6507x_ts_probe, - .remove = __devexit_p(tps6507x_ts_remove), + .remove = tps6507x_ts_remove, }; module_platform_driver(tps6507x_ts_driver); diff --git a/drivers/input/touchscreen/tsc2005.c b/drivers/input/touchscreen/tsc2005.c index 5ce3fa8ce646..9c0cdc7ea449 100644 --- a/drivers/input/touchscreen/tsc2005.c +++ b/drivers/input/touchscreen/tsc2005.c @@ -555,7 +555,7 @@ static void tsc2005_close(struct input_dev *input) mutex_unlock(&ts->mutex); } -static void __devinit tsc2005_setup_spi_xfer(struct tsc2005 *ts) +static void tsc2005_setup_spi_xfer(struct tsc2005 *ts) { tsc2005_setup_read(&ts->spi_x, TSC2005_REG_X, false); tsc2005_setup_read(&ts->spi_y, TSC2005_REG_Y, false); @@ -569,7 +569,7 @@ static void __devinit tsc2005_setup_spi_xfer(struct tsc2005 *ts) spi_message_add_tail(&ts->spi_z2.spi_xfer, &ts->spi_read_msg); } -static int __devinit tsc2005_probe(struct spi_device *spi) +static int tsc2005_probe(struct spi_device *spi) { const struct tsc2005_platform_data *pdata = spi->dev.platform_data; struct tsc2005 *ts; @@ -686,7 +686,7 @@ err_free_mem: return error; } -static int __devexit tsc2005_remove(struct spi_device *spi) +static int tsc2005_remove(struct spi_device *spi) { struct tsc2005 *ts = spi_get_drvdata(spi); @@ -745,7 +745,7 @@ static struct spi_driver tsc2005_driver = { .pm = &tsc2005_pm_ops, }, .probe = tsc2005_probe, - .remove = __devexit_p(tsc2005_remove), + .remove = tsc2005_remove, }; module_spi_driver(tsc2005_driver); diff --git a/drivers/input/touchscreen/tsc2007.c b/drivers/input/touchscreen/tsc2007.c index 1473d2382afd..0b67ba476b4c 100644 --- a/drivers/input/touchscreen/tsc2007.c +++ b/drivers/input/touchscreen/tsc2007.c @@ -273,7 +273,7 @@ static void tsc2007_close(struct input_dev *input_dev) tsc2007_stop(ts); } -static int __devinit tsc2007_probe(struct i2c_client *client, +static int tsc2007_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct tsc2007 *ts; @@ -366,7 +366,7 @@ static int __devinit tsc2007_probe(struct i2c_client *client, return err; } -static int __devexit tsc2007_remove(struct i2c_client *client) +static int tsc2007_remove(struct i2c_client *client) { struct tsc2007 *ts = i2c_get_clientdata(client); struct tsc2007_platform_data *pdata = client->dev.platform_data; @@ -396,7 +396,7 @@ static struct i2c_driver tsc2007_driver = { }, .id_table = tsc2007_idtable, .probe = tsc2007_probe, - .remove = __devexit_p(tsc2007_remove), + .remove = tsc2007_remove, }; module_i2c_driver(tsc2007_driver); diff --git a/drivers/input/touchscreen/ucb1400_ts.c b/drivers/input/touchscreen/ucb1400_ts.c index 46e83ad53f43..1271f97b4079 100644 --- a/drivers/input/touchscreen/ucb1400_ts.c +++ b/drivers/input/touchscreen/ucb1400_ts.c @@ -274,7 +274,7 @@ static void ucb1400_ts_close(struct input_dev *idev) * Try to probe our interrupt, rather than relying on lots of * hard-coded machine dependencies. */ -static int __devinit ucb1400_ts_detect_irq(struct ucb1400_ts *ucb, +static int ucb1400_ts_detect_irq(struct ucb1400_ts *ucb, struct platform_device *pdev) { unsigned long mask, timeout; @@ -318,7 +318,7 @@ static int __devinit ucb1400_ts_detect_irq(struct ucb1400_ts *ucb, return 0; } -static int __devinit ucb1400_ts_probe(struct platform_device *pdev) +static int ucb1400_ts_probe(struct platform_device *pdev) { struct ucb1400_ts *ucb = pdev->dev.platform_data; int error, x_res, y_res; @@ -397,7 +397,7 @@ err: return error; } -static int __devexit ucb1400_ts_remove(struct platform_device *pdev) +static int ucb1400_ts_remove(struct platform_device *pdev) { struct ucb1400_ts *ucb = pdev->dev.platform_data; @@ -442,7 +442,7 @@ static SIMPLE_DEV_PM_OPS(ucb1400_ts_pm_ops, static struct platform_driver ucb1400_ts_driver = { .probe = ucb1400_ts_probe, - .remove = __devexit_p(ucb1400_ts_remove), + .remove = ucb1400_ts_remove, .driver = { .name = "ucb1400_ts", .owner = THIS_MODULE, diff --git a/drivers/input/touchscreen/w90p910_ts.c b/drivers/input/touchscreen/w90p910_ts.c index 9396b21d0e8f..d2ef8f05c66e 100644 --- a/drivers/input/touchscreen/w90p910_ts.c +++ b/drivers/input/touchscreen/w90p910_ts.c @@ -215,7 +215,7 @@ static void w90p910_close(struct input_dev *dev) clk_disable(w90p910_ts->clk); } -static int __devinit w90x900ts_probe(struct platform_device *pdev) +static int w90x900ts_probe(struct platform_device *pdev) { struct w90p910_ts *w90p910_ts; struct input_dev *input_dev; @@ -301,7 +301,7 @@ fail1: input_free_device(input_dev); return err; } -static int __devexit w90x900ts_remove(struct platform_device *pdev) +static int w90x900ts_remove(struct platform_device *pdev) { struct w90p910_ts *w90p910_ts = platform_get_drvdata(pdev); struct resource *res; @@ -325,7 +325,7 @@ static int __devexit w90x900ts_remove(struct platform_device *pdev) static struct platform_driver w90x900ts_driver = { .probe = w90x900ts_probe, - .remove = __devexit_p(w90x900ts_remove), + .remove = w90x900ts_remove, .driver = { .name = "nuc900-ts", .owner = THIS_MODULE, diff --git a/drivers/input/touchscreen/wacom_i2c.c b/drivers/input/touchscreen/wacom_i2c.c index 0c01657132fd..bf0d07620bac 100644 --- a/drivers/input/touchscreen/wacom_i2c.c +++ b/drivers/input/touchscreen/wacom_i2c.c @@ -144,7 +144,7 @@ static void wacom_i2c_close(struct input_dev *dev) disable_irq(client->irq); } -static int __devinit wacom_i2c_probe(struct i2c_client *client, +static int wacom_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct wacom_i2c *wac_i2c; @@ -225,7 +225,7 @@ err_free_mem: return error; } -static int __devexit wacom_i2c_remove(struct i2c_client *client) +static int wacom_i2c_remove(struct i2c_client *client) { struct wacom_i2c *wac_i2c = i2c_get_clientdata(client); @@ -272,7 +272,7 @@ static struct i2c_driver wacom_i2c_driver = { }, .probe = wacom_i2c_probe, - .remove = __devexit_p(wacom_i2c_remove), + .remove = wacom_i2c_remove, .id_table = wacom_i2c_id, }; module_i2c_driver(wacom_i2c_driver); diff --git a/drivers/input/touchscreen/wm831x-ts.c b/drivers/input/touchscreen/wm831x-ts.c index 52abb98a8ae5..f88fab56178c 100644 --- a/drivers/input/touchscreen/wm831x-ts.c +++ b/drivers/input/touchscreen/wm831x-ts.c @@ -233,7 +233,7 @@ static void wm831x_ts_input_close(struct input_dev *idev) } } -static __devinit int wm831x_ts_probe(struct platform_device *pdev) +static int wm831x_ts_probe(struct platform_device *pdev) { struct wm831x_ts *wm831x_ts; struct wm831x *wm831x = dev_get_drvdata(pdev->dev.parent); @@ -245,7 +245,8 @@ static __devinit int wm831x_ts_probe(struct platform_device *pdev) if (core_pdata) pdata = core_pdata->touch; - wm831x_ts = kzalloc(sizeof(struct wm831x_ts), GFP_KERNEL); + wm831x_ts = devm_kzalloc(&pdev->dev, sizeof(struct wm831x_ts), + GFP_KERNEL); input_dev = input_allocate_device(); if (!wm831x_ts || !input_dev) { error = -ENOMEM; @@ -376,21 +377,18 @@ err_data_irq: free_irq(wm831x_ts->data_irq, wm831x_ts); err_alloc: input_free_device(input_dev); - kfree(wm831x_ts); return error; } -static __devexit int wm831x_ts_remove(struct platform_device *pdev) +static int wm831x_ts_remove(struct platform_device *pdev) { struct wm831x_ts *wm831x_ts = platform_get_drvdata(pdev); free_irq(wm831x_ts->pd_irq, wm831x_ts); free_irq(wm831x_ts->data_irq, wm831x_ts); input_unregister_device(wm831x_ts->input_dev); - kfree(wm831x_ts); - platform_set_drvdata(pdev, NULL); return 0; } @@ -400,7 +398,7 @@ static struct platform_driver wm831x_ts_driver = { .owner = THIS_MODULE, }, .probe = wm831x_ts_probe, - .remove = __devexit_p(wm831x_ts_remove), + .remove = wm831x_ts_remove, }; module_platform_driver(wm831x_ts_driver); diff --git a/drivers/isdn/mISDN/dsp_core.c b/drivers/isdn/mISDN/dsp_core.c index 28c99c623bcd..22b720ec80cb 100644 --- a/drivers/isdn/mISDN/dsp_core.c +++ b/drivers/isdn/mISDN/dsp_core.c @@ -1217,8 +1217,7 @@ static void __exit dsp_cleanup(void) { mISDN_unregister_Bprotocol(&DSP); - if (timer_pending(&dsp_spl_tl)) - del_timer(&dsp_spl_tl); + del_timer_sync(&dsp_spl_tl); if (!list_empty(&dsp_ilist)) { printk(KERN_ERR "mISDN_dsp: Audio DSP object inst list not " diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c index 196368009001..9c6b96414862 100644 --- a/drivers/macintosh/smu.c +++ b/drivers/macintosh/smu.c @@ -997,7 +997,7 @@ static struct smu_sdbp_header *smu_create_sdb_partition(int id) "%02x !\n", id, hdr->id); goto failure; } - if (prom_add_property(smu->of_node, prop)) { + if (of_add_property(smu->of_node, prop)) { printk(KERN_DEBUG "SMU: Failed creating sdb-partition-%02x " "property !\n", id); goto failure; diff --git a/drivers/macintosh/windfarm_fcu_controls.c b/drivers/macintosh/windfarm_fcu_controls.c index b3411edb324b..fd6ed15a979d 100644 --- a/drivers/macintosh/windfarm_fcu_controls.c +++ b/drivers/macintosh/windfarm_fcu_controls.c @@ -593,19 +593,7 @@ static struct i2c_driver wf_fcu_driver = { .id_table = wf_fcu_id, }; -static int __init wf_fcu_init(void) -{ - return i2c_add_driver(&wf_fcu_driver); -} - -static void __exit wf_fcu_exit(void) -{ - i2c_del_driver(&wf_fcu_driver); -} - - -module_init(wf_fcu_init); -module_exit(wf_fcu_exit); +module_i2c_driver(wf_fcu_driver); MODULE_AUTHOR("Benjamin Herrenschmidt <benh@kernel.crashing.org>"); MODULE_DESCRIPTION("FCU control objects for PowerMacs thermal control"); diff --git a/drivers/macintosh/windfarm_lm75_sensor.c b/drivers/macintosh/windfarm_lm75_sensor.c index b0c2d3695b34..9ef32b3df91f 100644 --- a/drivers/macintosh/windfarm_lm75_sensor.c +++ b/drivers/macintosh/windfarm_lm75_sensor.c @@ -174,19 +174,7 @@ static struct i2c_driver wf_lm75_driver = { .id_table = wf_lm75_id, }; -static int __init wf_lm75_sensor_init(void) -{ - return i2c_add_driver(&wf_lm75_driver); -} - -static void __exit wf_lm75_sensor_exit(void) -{ - i2c_del_driver(&wf_lm75_driver); -} - - -module_init(wf_lm75_sensor_init); -module_exit(wf_lm75_sensor_exit); +module_i2c_driver(wf_lm75_driver); MODULE_AUTHOR("Benjamin Herrenschmidt <benh@kernel.crashing.org>"); MODULE_DESCRIPTION("LM75 sensor objects for PowerMacs thermal control"); diff --git a/drivers/macintosh/windfarm_max6690_sensor.c b/drivers/macintosh/windfarm_max6690_sensor.c index 371b058d2f7d..945a25b2f31e 100644 --- a/drivers/macintosh/windfarm_max6690_sensor.c +++ b/drivers/macintosh/windfarm_max6690_sensor.c @@ -130,18 +130,7 @@ static struct i2c_driver wf_max6690_driver = { .id_table = wf_max6690_id, }; -static int __init wf_max6690_sensor_init(void) -{ - return i2c_add_driver(&wf_max6690_driver); -} - -static void __exit wf_max6690_sensor_exit(void) -{ - i2c_del_driver(&wf_max6690_driver); -} - -module_init(wf_max6690_sensor_init); -module_exit(wf_max6690_sensor_exit); +module_i2c_driver(wf_max6690_driver); MODULE_AUTHOR("Paul Mackerras <paulus@samba.org>"); MODULE_DESCRIPTION("MAX6690 sensor objects for PowerMac thermal control"); diff --git a/drivers/macintosh/windfarm_smu_sat.c b/drivers/macintosh/windfarm_smu_sat.c index 426e810233d7..d87f5ee04ca9 100644 --- a/drivers/macintosh/windfarm_smu_sat.c +++ b/drivers/macintosh/windfarm_smu_sat.c @@ -364,18 +364,7 @@ static struct i2c_driver wf_sat_driver = { .id_table = wf_sat_id, }; -static int __init sat_sensors_init(void) -{ - return i2c_add_driver(&wf_sat_driver); -} - -static void __exit sat_sensors_exit(void) -{ - i2c_del_driver(&wf_sat_driver); -} - -module_init(sat_sensors_init); -module_exit(sat_sensors_exit); +module_i2c_driver(wf_sat_driver); MODULE_AUTHOR("Paul Mackerras <paulus@samba.org>"); MODULE_DESCRIPTION("SMU satellite sensors for PowerMac thermal control"); diff --git a/drivers/md/md.c b/drivers/md/md.c index bd8bf0953fe3..3db3d1b271f7 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -452,7 +452,7 @@ void md_flush_request(struct mddev *mddev, struct bio *bio) spin_lock_irq(&mddev->write_lock); wait_event_lock_irq(mddev->sb_wait, !mddev->flush_bio, - mddev->write_lock, /*nothing*/); + mddev->write_lock); mddev->flush_bio = bio; spin_unlock_irq(&mddev->write_lock); @@ -1414,12 +1414,11 @@ static __le32 calc_sb_1_csum(struct mdp_superblock_1 * sb) unsigned long long newcsum; int size = 256 + le32_to_cpu(sb->max_dev)*2; __le32 *isuper = (__le32*)sb; - int i; disk_csum = sb->sb_csum; sb->sb_csum = 0; newcsum = 0; - for (i=0; size>=4; size -= 4 ) + for (; size >= 4; size -= 4) newcsum += le32_to_cpu(*isuper++); if (size == 2) @@ -4753,6 +4752,8 @@ md_attr_store(struct kobject *kobj, struct attribute *attr, } mddev_get(mddev); spin_unlock(&all_mddevs_lock); + if (entry->store == new_dev_store) + flush_workqueue(md_misc_wq); rv = mddev_lock(mddev); if (!rv) { rv = entry->store(mddev, page, length); @@ -6346,24 +6347,23 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, * Commands dealing with the RAID driver but not any * particular array: */ - switch (cmd) - { - case RAID_VERSION: - err = get_version(argp); - goto done; + switch (cmd) { + case RAID_VERSION: + err = get_version(argp); + goto done; - case PRINT_RAID_DEBUG: - err = 0; - md_print_devices(); - goto done; + case PRINT_RAID_DEBUG: + err = 0; + md_print_devices(); + goto done; #ifndef MODULE - case RAID_AUTORUN: - err = 0; - autostart_arrays(arg); - goto done; + case RAID_AUTORUN: + err = 0; + autostart_arrays(arg); + goto done; #endif - default:; + default:; } /* @@ -6398,6 +6398,10 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, goto abort; } + if (cmd == ADD_NEW_DISK) + /* need to ensure md_delayed_delete() has completed */ + flush_workqueue(md_misc_wq); + err = mddev_lock(mddev); if (err) { printk(KERN_INFO @@ -6406,50 +6410,44 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, goto abort; } - switch (cmd) - { - case SET_ARRAY_INFO: - { - mdu_array_info_t info; - if (!arg) - memset(&info, 0, sizeof(info)); - else if (copy_from_user(&info, argp, sizeof(info))) { - err = -EFAULT; - goto abort_unlock; - } - if (mddev->pers) { - err = update_array_info(mddev, &info); - if (err) { - printk(KERN_WARNING "md: couldn't update" - " array info. %d\n", err); - goto abort_unlock; - } - goto done_unlock; - } - if (!list_empty(&mddev->disks)) { - printk(KERN_WARNING - "md: array %s already has disks!\n", - mdname(mddev)); - err = -EBUSY; - goto abort_unlock; - } - if (mddev->raid_disks) { - printk(KERN_WARNING - "md: array %s already initialised!\n", - mdname(mddev)); - err = -EBUSY; - goto abort_unlock; - } - err = set_array_info(mddev, &info); - if (err) { - printk(KERN_WARNING "md: couldn't set" - " array info. %d\n", err); - goto abort_unlock; - } + if (cmd == SET_ARRAY_INFO) { + mdu_array_info_t info; + if (!arg) + memset(&info, 0, sizeof(info)); + else if (copy_from_user(&info, argp, sizeof(info))) { + err = -EFAULT; + goto abort_unlock; + } + if (mddev->pers) { + err = update_array_info(mddev, &info); + if (err) { + printk(KERN_WARNING "md: couldn't update" + " array info. %d\n", err); + goto abort_unlock; } goto done_unlock; - - default:; + } + if (!list_empty(&mddev->disks)) { + printk(KERN_WARNING + "md: array %s already has disks!\n", + mdname(mddev)); + err = -EBUSY; + goto abort_unlock; + } + if (mddev->raid_disks) { + printk(KERN_WARNING + "md: array %s already initialised!\n", + mdname(mddev)); + err = -EBUSY; + goto abort_unlock; + } + err = set_array_info(mddev, &info); + if (err) { + printk(KERN_WARNING "md: couldn't set" + " array info. %d\n", err); + goto abort_unlock; + } + goto done_unlock; } /* @@ -6468,52 +6466,51 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, /* * Commands even a read-only array can execute: */ - switch (cmd) - { - case GET_BITMAP_FILE: - err = get_bitmap_file(mddev, argp); - goto done_unlock; + switch (cmd) { + case GET_BITMAP_FILE: + err = get_bitmap_file(mddev, argp); + goto done_unlock; - case RESTART_ARRAY_RW: - err = restart_array(mddev); - goto done_unlock; + case RESTART_ARRAY_RW: + err = restart_array(mddev); + goto done_unlock; - case STOP_ARRAY: - err = do_md_stop(mddev, 0, bdev); - goto done_unlock; + case STOP_ARRAY: + err = do_md_stop(mddev, 0, bdev); + goto done_unlock; - case STOP_ARRAY_RO: - err = md_set_readonly(mddev, bdev); - goto done_unlock; + case STOP_ARRAY_RO: + err = md_set_readonly(mddev, bdev); + goto done_unlock; - case BLKROSET: - if (get_user(ro, (int __user *)(arg))) { - err = -EFAULT; - goto done_unlock; - } - err = -EINVAL; + case BLKROSET: + if (get_user(ro, (int __user *)(arg))) { + err = -EFAULT; + goto done_unlock; + } + err = -EINVAL; - /* if the bdev is going readonly the value of mddev->ro - * does not matter, no writes are coming - */ - if (ro) - goto done_unlock; + /* if the bdev is going readonly the value of mddev->ro + * does not matter, no writes are coming + */ + if (ro) + goto done_unlock; - /* are we are already prepared for writes? */ - if (mddev->ro != 1) - goto done_unlock; + /* are we are already prepared for writes? */ + if (mddev->ro != 1) + goto done_unlock; - /* transitioning to readauto need only happen for - * arrays that call md_write_start - */ - if (mddev->pers) { - err = restart_array(mddev); - if (err == 0) { - mddev->ro = 2; - set_disk_ro(mddev->gendisk, 0); - } + /* transitioning to readauto need only happen for + * arrays that call md_write_start + */ + if (mddev->pers) { + err = restart_array(mddev); + if (err == 0) { + mddev->ro = 2; + set_disk_ro(mddev->gendisk, 0); } - goto done_unlock; + } + goto done_unlock; } /* @@ -6535,37 +6532,36 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, } } - switch (cmd) + switch (cmd) { + case ADD_NEW_DISK: { - case ADD_NEW_DISK: - { - mdu_disk_info_t info; - if (copy_from_user(&info, argp, sizeof(info))) - err = -EFAULT; - else - err = add_new_disk(mddev, &info); - goto done_unlock; - } + mdu_disk_info_t info; + if (copy_from_user(&info, argp, sizeof(info))) + err = -EFAULT; + else + err = add_new_disk(mddev, &info); + goto done_unlock; + } - case HOT_REMOVE_DISK: - err = hot_remove_disk(mddev, new_decode_dev(arg)); - goto done_unlock; + case HOT_REMOVE_DISK: + err = hot_remove_disk(mddev, new_decode_dev(arg)); + goto done_unlock; - case HOT_ADD_DISK: - err = hot_add_disk(mddev, new_decode_dev(arg)); - goto done_unlock; + case HOT_ADD_DISK: + err = hot_add_disk(mddev, new_decode_dev(arg)); + goto done_unlock; - case RUN_ARRAY: - err = do_md_run(mddev); - goto done_unlock; + case RUN_ARRAY: + err = do_md_run(mddev); + goto done_unlock; - case SET_BITMAP_FILE: - err = set_bitmap_file(mddev, (int)arg); - goto done_unlock; + case SET_BITMAP_FILE: + err = set_bitmap_file(mddev, (int)arg); + goto done_unlock; - default: - err = -EINVAL; - goto abort_unlock; + default: + err = -EINVAL; + goto abort_unlock; } done_unlock: @@ -7184,6 +7180,7 @@ void md_done_sync(struct mddev *mddev, int blocks, int ok) wake_up(&mddev->recovery_wait); if (!ok) { set_bit(MD_RECOVERY_INTR, &mddev->recovery); + set_bit(MD_RECOVERY_ERROR, &mddev->recovery); md_wakeup_thread(mddev->thread); // stop recovery, signal do_sync .... } @@ -7281,6 +7278,7 @@ EXPORT_SYMBOL_GPL(md_allow_write); #define SYNC_MARKS 10 #define SYNC_MARK_STEP (3*HZ) +#define UPDATE_FREQUENCY (5*60*HZ) void md_do_sync(struct md_thread *thread) { struct mddev *mddev = thread->mddev; @@ -7289,6 +7287,7 @@ void md_do_sync(struct md_thread *thread) window; sector_t max_sectors,j, io_sectors; unsigned long mark[SYNC_MARKS]; + unsigned long update_time; sector_t mark_cnt[SYNC_MARKS]; int last_mark,m; struct list_head *tmp; @@ -7448,6 +7447,7 @@ void md_do_sync(struct md_thread *thread) mddev->curr_resync_completed = j; sysfs_notify(&mddev->kobj, NULL, "sync_completed"); md_new_event(mddev); + update_time = jiffies; blk_start_plug(&plug); while (j < max_sectors) { @@ -7459,6 +7459,7 @@ void md_do_sync(struct md_thread *thread) ((mddev->curr_resync > mddev->curr_resync_completed && (mddev->curr_resync - mddev->curr_resync_completed) > (max_sectors >> 4)) || + time_after_eq(jiffies, update_time + UPDATE_FREQUENCY) || (j - mddev->curr_resync_completed)*2 >= mddev->resync_max - mddev->curr_resync_completed )) { @@ -7466,6 +7467,10 @@ void md_do_sync(struct md_thread *thread) wait_event(mddev->recovery_wait, atomic_read(&mddev->recovery_active) == 0); mddev->curr_resync_completed = j; + if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) && + j > mddev->recovery_cp) + mddev->recovery_cp = j; + update_time = jiffies; set_bit(MD_CHANGE_CLEAN, &mddev->flags); sysfs_notify(&mddev->kobj, NULL, "sync_completed"); } @@ -7570,8 +7575,13 @@ void md_do_sync(struct md_thread *thread) printk(KERN_INFO "md: checkpointing %s of %s.\n", desc, mdname(mddev)); - mddev->recovery_cp = - mddev->curr_resync_completed; + if (test_bit(MD_RECOVERY_ERROR, + &mddev->recovery)) + mddev->recovery_cp = + mddev->curr_resync_completed; + else + mddev->recovery_cp = + mddev->curr_resync; } } else mddev->recovery_cp = MaxSector; diff --git a/drivers/md/md.h b/drivers/md/md.h index af443ab868db..eca59c3074ef 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -307,6 +307,7 @@ struct mddev { * REQUEST: user-space has requested a sync (used with SYNC) * CHECK: user-space request for check-only, no repair * RESHAPE: A reshape is happening + * ERROR: sync-action interrupted because io-error * * If neither SYNC or RESHAPE are set, then it is a recovery. */ @@ -320,6 +321,7 @@ struct mddev { #define MD_RECOVERY_CHECK 7 #define MD_RECOVERY_RESHAPE 8 #define MD_RECOVERY_FROZEN 9 +#define MD_RECOVERY_ERROR 10 unsigned long recovery; /* If a RAID personality determines that recovery (of a particular @@ -551,32 +553,6 @@ struct md_thread { #define THREAD_WAKEUP 0 -#define __wait_event_lock_irq(wq, condition, lock, cmd) \ -do { \ - wait_queue_t __wait; \ - init_waitqueue_entry(&__wait, current); \ - \ - add_wait_queue(&wq, &__wait); \ - for (;;) { \ - set_current_state(TASK_UNINTERRUPTIBLE); \ - if (condition) \ - break; \ - spin_unlock_irq(&lock); \ - cmd; \ - schedule(); \ - spin_lock_irq(&lock); \ - } \ - current->state = TASK_RUNNING; \ - remove_wait_queue(&wq, &__wait); \ -} while (0) - -#define wait_event_lock_irq(wq, condition, lock, cmd) \ -do { \ - if (condition) \ - break; \ - __wait_event_lock_irq(wq, condition, lock, cmd); \ -} while (0) - static inline void safe_put_page(struct page *p) { if (p) put_page(p); diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index a0f73092176e..d5bddfc4010e 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -822,7 +822,7 @@ static void raise_barrier(struct r1conf *conf) /* Wait until no block IO is waiting */ wait_event_lock_irq(conf->wait_barrier, !conf->nr_waiting, - conf->resync_lock, ); + conf->resync_lock); /* block any new IO from starting */ conf->barrier++; @@ -830,7 +830,7 @@ static void raise_barrier(struct r1conf *conf) /* Now wait for all pending IO to complete */ wait_event_lock_irq(conf->wait_barrier, !conf->nr_pending && conf->barrier < RESYNC_DEPTH, - conf->resync_lock, ); + conf->resync_lock); spin_unlock_irq(&conf->resync_lock); } @@ -864,8 +864,7 @@ static void wait_barrier(struct r1conf *conf) (conf->nr_pending && current->bio_list && !bio_list_empty(current->bio_list)), - conf->resync_lock, - ); + conf->resync_lock); conf->nr_waiting--; } conf->nr_pending++; @@ -898,10 +897,10 @@ static void freeze_array(struct r1conf *conf) spin_lock_irq(&conf->resync_lock); conf->barrier++; conf->nr_waiting++; - wait_event_lock_irq(conf->wait_barrier, - conf->nr_pending == conf->nr_queued+1, - conf->resync_lock, - flush_pending_writes(conf)); + wait_event_lock_irq_cmd(conf->wait_barrier, + conf->nr_pending == conf->nr_queued+1, + conf->resync_lock, + flush_pending_writes(conf)); spin_unlock_irq(&conf->resync_lock); } static void unfreeze_array(struct r1conf *conf) diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index c9acbd717131..64d48249c03b 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -952,7 +952,7 @@ static void raise_barrier(struct r10conf *conf, int force) /* Wait until no block IO is waiting (unless 'force') */ wait_event_lock_irq(conf->wait_barrier, force || !conf->nr_waiting, - conf->resync_lock, ); + conf->resync_lock); /* block any new IO from starting */ conf->barrier++; @@ -960,7 +960,7 @@ static void raise_barrier(struct r10conf *conf, int force) /* Now wait for all pending IO to complete */ wait_event_lock_irq(conf->wait_barrier, !conf->nr_pending && conf->barrier < RESYNC_DEPTH, - conf->resync_lock, ); + conf->resync_lock); spin_unlock_irq(&conf->resync_lock); } @@ -993,8 +993,7 @@ static void wait_barrier(struct r10conf *conf) (conf->nr_pending && current->bio_list && !bio_list_empty(current->bio_list)), - conf->resync_lock, - ); + conf->resync_lock); conf->nr_waiting--; } conf->nr_pending++; @@ -1027,10 +1026,10 @@ static void freeze_array(struct r10conf *conf) spin_lock_irq(&conf->resync_lock); conf->barrier++; conf->nr_waiting++; - wait_event_lock_irq(conf->wait_barrier, - conf->nr_pending == conf->nr_queued+1, - conf->resync_lock, - flush_pending_writes(conf)); + wait_event_lock_irq_cmd(conf->wait_barrier, + conf->nr_pending == conf->nr_queued+1, + conf->resync_lock, + flush_pending_writes(conf)); spin_unlock_irq(&conf->resync_lock); } diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 3380372c0393..19d77a026639 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -53,6 +53,8 @@ #include <linux/cpu.h> #include <linux/slab.h> #include <linux/ratelimit.h> +#include <trace/events/block.h> + #include "md.h" #include "raid5.h" #include "raid0.h" @@ -182,6 +184,8 @@ static void return_io(struct bio *return_bi) return_bi = bi->bi_next; bi->bi_next = NULL; bi->bi_size = 0; + trace_block_bio_complete(bdev_get_queue(bi->bi_bdev), + bi, 0); bio_endio(bi, 0); bi = return_bi; } @@ -466,7 +470,7 @@ get_active_stripe(struct r5conf *conf, sector_t sector, do { wait_event_lock_irq(conf->wait_for_stripe, conf->quiesce == 0 || noquiesce, - conf->device_lock, /* nothing */); + conf->device_lock); sh = __find_stripe(conf, sector, conf->generation - previous); if (!sh) { if (!conf->inactive_blocked) @@ -480,8 +484,7 @@ get_active_stripe(struct r5conf *conf, sector_t sector, (atomic_read(&conf->active_stripes) < (conf->max_nr_stripes *3/4) || !conf->inactive_blocked), - conf->device_lock, - ); + conf->device_lock); conf->inactive_blocked = 0; } else init_stripe(sh, sector, previous); @@ -671,6 +674,9 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s) bi->bi_next = NULL; if (rrdev) set_bit(R5_DOUBLE_LOCKED, &sh->dev[i].flags); + trace_block_bio_remap(bdev_get_queue(bi->bi_bdev), + bi, disk_devt(conf->mddev->gendisk), + sh->dev[i].sector); generic_make_request(bi); } if (rrdev) { @@ -698,6 +704,9 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s) rbi->bi_io_vec[0].bv_offset = 0; rbi->bi_size = STRIPE_SIZE; rbi->bi_next = NULL; + trace_block_bio_remap(bdev_get_queue(rbi->bi_bdev), + rbi, disk_devt(conf->mddev->gendisk), + sh->dev[i].sector); generic_make_request(rbi); } if (!rdev && !rrdev) { @@ -1646,8 +1655,7 @@ static int resize_stripes(struct r5conf *conf, int newsize) spin_lock_irq(&conf->device_lock); wait_event_lock_irq(conf->wait_for_stripe, !list_empty(&conf->inactive_list), - conf->device_lock, - ); + conf->device_lock); osh = get_free_stripe(conf); spin_unlock_irq(&conf->device_lock); atomic_set(&nsh->count, 1); @@ -2855,8 +2863,10 @@ static void handle_stripe_dirtying(struct r5conf *conf, pr_debug("for sector %llu, rmw=%d rcw=%d\n", (unsigned long long)sh->sector, rmw, rcw); set_bit(STRIPE_HANDLE, &sh->state); - if (rmw < rcw && rmw > 0) + if (rmw < rcw && rmw > 0) { /* prefer read-modify-write, but need to get some data */ + blk_add_trace_msg(conf->mddev->queue, "raid5 rmw %llu %d", + (unsigned long long)sh->sector, rmw); for (i = disks; i--; ) { struct r5dev *dev = &sh->dev[i]; if ((dev->towrite || i == sh->pd_idx) && @@ -2867,7 +2877,7 @@ static void handle_stripe_dirtying(struct r5conf *conf, if ( test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) { pr_debug("Read_old block " - "%d for r-m-w\n", i); + "%d for r-m-w\n", i); set_bit(R5_LOCKED, &dev->flags); set_bit(R5_Wantread, &dev->flags); s->locked++; @@ -2877,8 +2887,10 @@ static void handle_stripe_dirtying(struct r5conf *conf, } } } + } if (rcw <= rmw && rcw > 0) { /* want reconstruct write, but need to get some data */ + int qread =0; rcw = 0; for (i = disks; i--; ) { struct r5dev *dev = &sh->dev[i]; @@ -2897,12 +2909,17 @@ static void handle_stripe_dirtying(struct r5conf *conf, set_bit(R5_LOCKED, &dev->flags); set_bit(R5_Wantread, &dev->flags); s->locked++; + qread++; } else { set_bit(STRIPE_DELAYED, &sh->state); set_bit(STRIPE_HANDLE, &sh->state); } } } + if (rcw) + blk_add_trace_msg(conf->mddev->queue, "raid5 rcw %llu %d %d %d", + (unsigned long long)sh->sector, + rcw, qread, test_bit(STRIPE_DELAYED, &sh->state)); } /* now if nothing is locked, and if we have enough data, * we can start a write request @@ -3224,10 +3241,7 @@ static void handle_stripe_expansion(struct r5conf *conf, struct stripe_head *sh) } /* done submitting copies, wait for them to complete */ - if (tx) { - async_tx_ack(tx); - dma_wait_for_async_tx(tx); - } + async_tx_quiesce(&tx); } /* @@ -3903,6 +3917,8 @@ static void raid5_align_endio(struct bio *bi, int error) rdev_dec_pending(rdev, conf->mddev); if (!error && uptodate) { + trace_block_bio_complete(bdev_get_queue(raid_bi->bi_bdev), + raid_bi, 0); bio_endio(raid_bi, 0); if (atomic_dec_and_test(&conf->active_aligned_reads)) wake_up(&conf->wait_for_stripe); @@ -4003,10 +4019,13 @@ static int chunk_aligned_read(struct mddev *mddev, struct bio * raid_bio) spin_lock_irq(&conf->device_lock); wait_event_lock_irq(conf->wait_for_stripe, conf->quiesce == 0, - conf->device_lock, /* nothing */); + conf->device_lock); atomic_inc(&conf->active_aligned_reads); spin_unlock_irq(&conf->device_lock); + trace_block_bio_remap(bdev_get_queue(align_bi->bi_bdev), + align_bi, disk_devt(mddev->gendisk), + raid_bio->bi_sector); generic_make_request(align_bi); return 1; } else { @@ -4081,6 +4100,7 @@ static void raid5_unplug(struct blk_plug_cb *blk_cb, bool from_schedule) struct stripe_head *sh; struct mddev *mddev = cb->cb.data; struct r5conf *conf = mddev->private; + int cnt = 0; if (cb->list.next && !list_empty(&cb->list)) { spin_lock_irq(&conf->device_lock); @@ -4095,9 +4115,11 @@ static void raid5_unplug(struct blk_plug_cb *blk_cb, bool from_schedule) smp_mb__before_clear_bit(); clear_bit(STRIPE_ON_UNPLUG_LIST, &sh->state); __release_stripe(conf, sh); + cnt++; } spin_unlock_irq(&conf->device_lock); } + trace_block_unplug(mddev->queue, cnt, !from_schedule); kfree(cb); } @@ -4355,6 +4377,8 @@ static void make_request(struct mddev *mddev, struct bio * bi) if ( rw == WRITE ) md_write_end(mddev); + trace_block_bio_complete(bdev_get_queue(bi->bi_bdev), + bi, 0); bio_endio(bi, 0); } } @@ -4731,8 +4755,11 @@ static int retry_aligned_read(struct r5conf *conf, struct bio *raid_bio) handled++; } remaining = raid5_dec_bi_active_stripes(raid_bio); - if (remaining == 0) + if (remaining == 0) { + trace_block_bio_complete(bdev_get_queue(raid_bio->bi_bdev), + raid_bio, 0); bio_endio(raid_bio, 0); + } if (atomic_dec_and_test(&conf->active_aligned_reads)) wake_up(&conf->wait_for_stripe); return handled; @@ -6095,7 +6122,7 @@ static void raid5_quiesce(struct mddev *mddev, int state) wait_event_lock_irq(conf->wait_for_stripe, atomic_read(&conf->active_stripes) == 0 && atomic_read(&conf->active_aligned_reads) == 0, - conf->device_lock, /* nothing */); + conf->device_lock); conf->quiesce = 1; spin_unlock_irq(&conf->device_lock); /* allow reshape to continue */ diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c index 0c3ced70707b..164afa71bba7 100644 --- a/drivers/message/fusion/mptscsih.c +++ b/drivers/message/fusion/mptscsih.c @@ -792,6 +792,7 @@ mptscsih_io_done(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *mr) * than an unsolicited DID_ABORT. */ sc->result = DID_RESET << 16; + break; case MPI_IOCSTATUS_SCSI_EXT_TERMINATED: /* 0x004C */ if (ioc->bus_type == FC) diff --git a/drivers/mfd/stmpe.c b/drivers/mfd/stmpe.c index 19636199d7a5..5e8e6927cfcd 100644 --- a/drivers/mfd/stmpe.c +++ b/drivers/mfd/stmpe.c @@ -326,6 +326,7 @@ static struct resource stmpe_keypad_resources[] = { static struct mfd_cell stmpe_keypad_cell = { .name = "stmpe-keypad", + .of_compatible = "st,stmpe-keypad", .resources = stmpe_keypad_resources, .num_resources = ARRAY_SIZE(stmpe_keypad_resources), }; @@ -409,6 +410,7 @@ static struct resource stmpe_ts_resources[] = { static struct mfd_cell stmpe_ts_cell = { .name = "stmpe-ts", + .of_compatible = "st,stmpe-ts", .resources = stmpe_ts_resources, .num_resources = ARRAY_SIZE(stmpe_ts_resources), }; diff --git a/drivers/mtd/ar7part.c b/drivers/mtd/ar7part.c index 945393129952..7c057a05adb6 100644 --- a/drivers/mtd/ar7part.c +++ b/drivers/mtd/ar7part.c @@ -26,19 +26,16 @@ #include <linux/mtd/mtd.h> #include <linux/mtd/partitions.h> #include <linux/bootmem.h> -#include <linux/magic.h> #include <linux/module.h> +#include <uapi/linux/magic.h> + #define AR7_PARTS 4 #define ROOT_OFFSET 0xe0000 #define LOADER_MAGIC1 le32_to_cpu(0xfeedfa42) #define LOADER_MAGIC2 le32_to_cpu(0xfeed1281) -#ifndef SQUASHFS_MAGIC -#define SQUASHFS_MAGIC 0x73717368 -#endif - struct ar7_bin_rec { unsigned int checksum; unsigned int length; diff --git a/drivers/mtd/bcm63xxpart.c b/drivers/mtd/bcm63xxpart.c index 63d2a64331f7..6eeb84c81bc2 100644 --- a/drivers/mtd/bcm63xxpart.c +++ b/drivers/mtd/bcm63xxpart.c @@ -37,8 +37,7 @@ #define BCM63XX_EXTENDED_SIZE 0xBFC00000 /* Extended flash address */ -#define BCM63XX_MIN_CFE_SIZE 0x10000 /* always at least 64KiB */ -#define BCM63XX_MIN_NVRAM_SIZE 0x10000 /* always at least 64KiB */ +#define BCM63XX_CFE_BLOCK_SIZE 0x10000 /* always at least 64KiB */ #define BCM63XX_CFE_MAGIC_OFFSET 0x4e0 @@ -79,7 +78,7 @@ static int bcm63xx_parse_cfe_partitions(struct mtd_info *master, unsigned int rootfsaddr, kerneladdr, spareaddr; unsigned int rootfslen, kernellen, sparelen, totallen; unsigned int cfelen, nvramlen; - int namelen = 0; + unsigned int cfe_erasesize; int i; u32 computed_crc; bool rootfs_first = false; @@ -87,8 +86,11 @@ static int bcm63xx_parse_cfe_partitions(struct mtd_info *master, if (bcm63xx_detect_cfe(master)) return -EINVAL; - cfelen = max_t(uint32_t, master->erasesize, BCM63XX_MIN_CFE_SIZE); - nvramlen = max_t(uint32_t, master->erasesize, BCM63XX_MIN_NVRAM_SIZE); + cfe_erasesize = max_t(uint32_t, master->erasesize, + BCM63XX_CFE_BLOCK_SIZE); + + cfelen = cfe_erasesize; + nvramlen = cfe_erasesize; /* Allocate memory for buffer */ buf = vmalloc(sizeof(struct bcm_tag)); @@ -121,7 +123,6 @@ static int bcm63xx_parse_cfe_partitions(struct mtd_info *master, kerneladdr = kerneladdr - BCM63XX_EXTENDED_SIZE; rootfsaddr = rootfsaddr - BCM63XX_EXTENDED_SIZE; spareaddr = roundup(totallen, master->erasesize) + cfelen; - sparelen = master->size - spareaddr - nvramlen; if (rootfsaddr < kerneladdr) { /* default Broadcom layout */ @@ -139,19 +140,15 @@ static int bcm63xx_parse_cfe_partitions(struct mtd_info *master, rootfslen = 0; rootfsaddr = 0; spareaddr = cfelen; - sparelen = master->size - cfelen - nvramlen; } + sparelen = master->size - spareaddr - nvramlen; /* Determine number of partitions */ - namelen = 8; - if (rootfslen > 0) { + if (rootfslen > 0) nrparts++; - namelen += 6; - } - if (kernellen > 0) { + + if (kernellen > 0) nrparts++; - namelen += 6; - } /* Ask kernel for more memory */ parts = kzalloc(sizeof(*parts) * nrparts + 10 * nrparts, GFP_KERNEL); @@ -193,17 +190,16 @@ static int bcm63xx_parse_cfe_partitions(struct mtd_info *master, parts[curpart].name = "nvram"; parts[curpart].offset = master->size - nvramlen; parts[curpart].size = nvramlen; + curpart++; /* Global partition "linux" to make easy firmware upgrade */ - curpart++; parts[curpart].name = "linux"; parts[curpart].offset = cfelen; parts[curpart].size = master->size - cfelen - nvramlen; for (i = 0; i < nrparts; i++) - pr_info("Partition %d is %s offset %lx and length %lx\n", i, - parts[i].name, (long unsigned int)(parts[i].offset), - (long unsigned int)(parts[i].size)); + pr_info("Partition %d is %s offset %llx and length %llx\n", i, + parts[i].name, parts[i].offset, parts[i].size); pr_info("Spare partition is offset %x and length %x\n", spareaddr, sparelen); diff --git a/drivers/mtd/chips/cfi_cmdset_0002.c b/drivers/mtd/chips/cfi_cmdset_0002.c index 5ff5c4a16943..b86197286f24 100644 --- a/drivers/mtd/chips/cfi_cmdset_0002.c +++ b/drivers/mtd/chips/cfi_cmdset_0002.c @@ -1536,8 +1536,20 @@ static int __xipram do_write_buffer(struct map_info *map, struct flchip *chip, UDELAY(map, chip, adr, 1); } - /* reset on all failures. */ - map_write( map, CMD(0xF0), chip->start ); + /* + * Recovery from write-buffer programming failures requires + * the write-to-buffer-reset sequence. Since the last part + * of the sequence also works as a normal reset, we can run + * the same commands regardless of why we are here. + * See e.g. + * http://www.spansion.com/Support/Application%20Notes/MirrorBit_Write_Buffer_Prog_Page_Buffer_Read_AN.pdf + */ + cfi_send_gen_cmd(0xAA, cfi->addr_unlock1, chip->start, map, cfi, + cfi->device_type, NULL); + cfi_send_gen_cmd(0x55, cfi->addr_unlock2, chip->start, map, cfi, + cfi->device_type, NULL); + cfi_send_gen_cmd(0xF0, cfi->addr_unlock1, chip->start, map, cfi, + cfi->device_type, NULL); xip_enable(map, chip, adr); /* FIXME - should have reset delay before continuing */ diff --git a/drivers/mtd/cmdlinepart.c b/drivers/mtd/cmdlinepart.c index aed1b8a63c9f..c533f27d863f 100644 --- a/drivers/mtd/cmdlinepart.c +++ b/drivers/mtd/cmdlinepart.c @@ -56,8 +56,8 @@ /* special size referring to all the remaining space in a partition */ -#define SIZE_REMAINING UINT_MAX -#define OFFSET_CONTINUOUS UINT_MAX +#define SIZE_REMAINING ULLONG_MAX +#define OFFSET_CONTINUOUS ULLONG_MAX struct cmdline_mtd_partition { struct cmdline_mtd_partition *next; @@ -89,7 +89,7 @@ static struct mtd_partition * newpart(char *s, int extra_mem_size) { struct mtd_partition *parts; - unsigned long size, offset = OFFSET_CONTINUOUS; + unsigned long long size, offset = OFFSET_CONTINUOUS; char *name; int name_len; unsigned char *extra_mem; @@ -104,7 +104,8 @@ static struct mtd_partition * newpart(char *s, } else { size = memparse(s, &s); if (size < PAGE_SIZE) { - printk(KERN_ERR ERRP "partition size too small (%lx)\n", size); + printk(KERN_ERR ERRP "partition size too small (%llx)\n", + size); return ERR_PTR(-EINVAL); } } @@ -296,7 +297,7 @@ static int parse_cmdline_partitions(struct mtd_info *master, struct mtd_partition **pparts, struct mtd_part_parser_data *data) { - unsigned long offset; + unsigned long long offset; int i, err; struct cmdline_mtd_partition *part; const char *mtd_id = master->name; @@ -308,48 +309,52 @@ static int parse_cmdline_partitions(struct mtd_info *master, return err; } + /* + * Search for the partition definition matching master->name. + * If master->name is not set, stop at first partition definition. + */ for (part = partitions; part; part = part->next) { - if ((!mtd_id) || (!strcmp(part->mtd_id, mtd_id))) { - for (i = 0, offset = 0; i < part->num_parts; i++) { - if (part->parts[i].offset == OFFSET_CONTINUOUS) - part->parts[i].offset = offset; - else - offset = part->parts[i].offset; - - if (part->parts[i].size == SIZE_REMAINING) - part->parts[i].size = master->size - offset; - - if (part->parts[i].size == 0) { - printk(KERN_WARNING ERRP - "%s: skipping zero sized partition\n", - part->mtd_id); - part->num_parts--; - memmove(&part->parts[i], - &part->parts[i + 1], - sizeof(*part->parts) * (part->num_parts - i)); - continue; - } - - if (offset + part->parts[i].size > master->size) { - printk(KERN_WARNING ERRP - "%s: partitioning exceeds flash size, truncating\n", - part->mtd_id); - part->parts[i].size = master->size - offset; - } - offset += part->parts[i].size; - } - - *pparts = kmemdup(part->parts, - sizeof(*part->parts) * part->num_parts, - GFP_KERNEL); - if (!*pparts) - return -ENOMEM; - - return part->num_parts; + if ((!mtd_id) || (!strcmp(part->mtd_id, mtd_id))) + break; + } + + if (!part) + return 0; + + for (i = 0, offset = 0; i < part->num_parts; i++) { + if (part->parts[i].offset == OFFSET_CONTINUOUS) + part->parts[i].offset = offset; + else + offset = part->parts[i].offset; + + if (part->parts[i].size == SIZE_REMAINING) + part->parts[i].size = master->size - offset; + + if (part->parts[i].size == 0) { + printk(KERN_WARNING ERRP + "%s: skipping zero sized partition\n", + part->mtd_id); + part->num_parts--; + memmove(&part->parts[i], &part->parts[i + 1], + sizeof(*part->parts) * (part->num_parts - i)); + continue; } + + if (offset + part->parts[i].size > master->size) { + printk(KERN_WARNING ERRP + "%s: partitioning exceeds flash size, truncating\n", + part->mtd_id); + part->parts[i].size = master->size - offset; + } + offset += part->parts[i].size; } - return 0; + *pparts = kmemdup(part->parts, sizeof(*part->parts) * part->num_parts, + GFP_KERNEL); + if (!*pparts) + return -ENOMEM; + + return part->num_parts; } diff --git a/drivers/mtd/devices/bcm47xxsflash.c b/drivers/mtd/devices/bcm47xxsflash.c index 2dc5a6f3fd57..4714584aa993 100644 --- a/drivers/mtd/devices/bcm47xxsflash.c +++ b/drivers/mtd/devices/bcm47xxsflash.c @@ -66,7 +66,7 @@ out: return err; } -static int __devexit bcm47xxsflash_remove(struct platform_device *pdev) +static int bcm47xxsflash_remove(struct platform_device *pdev) { struct bcma_sflash *sflash = dev_get_platdata(&pdev->dev); @@ -77,7 +77,7 @@ static int __devexit bcm47xxsflash_remove(struct platform_device *pdev) } static struct platform_driver bcma_sflash_driver = { - .remove = __devexit_p(bcm47xxsflash_remove), + .remove = bcm47xxsflash_remove, .driver = { .name = "bcma_sflash", .owner = THIS_MODULE, diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c index 681e2ee0f2d6..e081bfeaaf7d 100644 --- a/drivers/mtd/devices/block2mtd.c +++ b/drivers/mtd/devices/block2mtd.c @@ -62,6 +62,7 @@ static int _block2mtd_erase(struct block2mtd_dev *dev, loff_t to, size_t len) memset(page_address(page), 0xff, PAGE_SIZE); set_page_dirty(page); unlock_page(page); + balance_dirty_pages_ratelimited(mapping); break; } @@ -152,6 +153,7 @@ static int _block2mtd_write(struct block2mtd_dev *dev, const u_char *buf, memcpy(page_address(page) + offset, buf, cpylen); set_page_dirty(page); unlock_page(page); + balance_dirty_pages_ratelimited(mapping); } page_cache_release(page); @@ -433,7 +435,7 @@ static int __init block2mtd_init(void) } -static void __devexit block2mtd_exit(void) +static void block2mtd_exit(void) { struct list_head *pos, *next; diff --git a/drivers/mtd/devices/docg3.c b/drivers/mtd/devices/docg3.c index d34d83b8f9c2..8510ccb9c6f0 100644 --- a/drivers/mtd/devices/docg3.c +++ b/drivers/mtd/devices/docg3.c @@ -1440,7 +1440,7 @@ static int doc_write_oob(struct mtd_info *mtd, loff_t ofs, oobdelta = mtd->ecclayout->oobavail; break; default: - oobdelta = 0; + return -EINVAL; } if ((len % DOC_LAYOUT_PAGE_SIZE) || (ooblen % oobdelta) || (ofs % DOC_LAYOUT_PAGE_SIZE)) diff --git a/drivers/mtd/devices/docprobe.c b/drivers/mtd/devices/docprobe.c index 706b847b46b3..88b3fd3e18a7 100644 --- a/drivers/mtd/devices/docprobe.c +++ b/drivers/mtd/devices/docprobe.c @@ -70,8 +70,6 @@ static unsigned long __initdata doc_locations[] = { 0xe0000, 0xe2000, 0xe4000, 0xe6000, 0xe8000, 0xea000, 0xec000, 0xee000, #endif /* CONFIG_MTD_DOCPROBE_HIGH */ -#else -#warning Unknown architecture for DiskOnChip. No default probe locations defined #endif 0xffffffff }; diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c index 03838bab1f59..4eeeb2d7f6ea 100644 --- a/drivers/mtd/devices/m25p80.c +++ b/drivers/mtd/devices/m25p80.c @@ -73,14 +73,6 @@ #define MAX_READY_WAIT_JIFFIES (40 * HZ) /* M25P16 specs 40s max chip erase */ #define MAX_CMD_SIZE 5 -#ifdef CONFIG_M25PXX_USE_FAST_READ -#define OPCODE_READ OPCODE_FAST_READ -#define FAST_READ_DUMMY_BYTE 1 -#else -#define OPCODE_READ OPCODE_NORM_READ -#define FAST_READ_DUMMY_BYTE 0 -#endif - #define JEDEC_MFR(_jedec_id) ((_jedec_id) >> 16) /****************************************************************************/ @@ -93,6 +85,7 @@ struct m25p { u16 addr_width; u8 erase_opcode; u8 *command; + bool fast_read; }; static inline struct m25p *mtd_to_m25p(struct mtd_info *mtd) @@ -168,6 +161,7 @@ static inline int set_4byte(struct m25p *flash, u32 jedec_id, int enable) { switch (JEDEC_MFR(jedec_id)) { case CFI_MFR_MACRONIX: + case 0xEF /* winbond */: flash->command[0] = enable ? OPCODE_EN4B : OPCODE_EX4B; return spi_write(flash->spi, flash->command, 1); default: @@ -342,6 +336,7 @@ static int m25p80_read(struct mtd_info *mtd, loff_t from, size_t len, struct m25p *flash = mtd_to_m25p(mtd); struct spi_transfer t[2]; struct spi_message m; + uint8_t opcode; pr_debug("%s: %s from 0x%08x, len %zd\n", dev_name(&flash->spi->dev), __func__, (u32)from, len); @@ -354,7 +349,7 @@ static int m25p80_read(struct mtd_info *mtd, loff_t from, size_t len, * Should add 1 byte DUMMY_BYTE. */ t[0].tx_buf = flash->command; - t[0].len = m25p_cmdsz(flash) + FAST_READ_DUMMY_BYTE; + t[0].len = m25p_cmdsz(flash) + (flash->fast_read ? 1 : 0); spi_message_add_tail(&t[0], &m); t[1].rx_buf = buf; @@ -376,12 +371,14 @@ static int m25p80_read(struct mtd_info *mtd, loff_t from, size_t len, */ /* Set up the write data buffer. */ - flash->command[0] = OPCODE_READ; + opcode = flash->fast_read ? OPCODE_FAST_READ : OPCODE_NORM_READ; + flash->command[0] = opcode; m25p_addr2cmd(flash, from, flash->command); spi_sync(flash->spi, &m); - *retlen = m.actual_length - m25p_cmdsz(flash) - FAST_READ_DUMMY_BYTE; + *retlen = m.actual_length - m25p_cmdsz(flash) - + (flash->fast_read ? 1 : 0); mutex_unlock(&flash->lock); @@ -664,7 +661,8 @@ static const struct spi_device_id m25p_ids[] = { { "mx25l25655e", INFO(0xc22619, 0, 64 * 1024, 512, 0) }, /* Micron */ - { "n25q128", INFO(0x20ba18, 0, 64 * 1024, 256, 0) }, + { "n25q128a11", INFO(0x20bb18, 0, 64 * 1024, 256, 0) }, + { "n25q128a13", INFO(0x20ba18, 0, 64 * 1024, 256, 0) }, { "n25q256a", INFO(0x20ba19, 0, 64 * 1024, 512, SECT_4K) }, /* Spansion -- single (large) sector size only, at least @@ -745,6 +743,8 @@ static const struct spi_device_id m25p_ids[] = { { "w25x64", INFO(0xef3017, 0, 64 * 1024, 128, SECT_4K) }, { "w25q64", INFO(0xef4017, 0, 64 * 1024, 128, SECT_4K) }, { "w25q80", INFO(0xef5014, 0, 64 * 1024, 16, SECT_4K) }, + { "w25q80bl", INFO(0xef4014, 0, 64 * 1024, 16, SECT_4K) }, + { "w25q256", INFO(0xef4019, 0, 64 * 1024, 512, SECT_4K) }, /* Catalyst / On Semiconductor -- non-JEDEC */ { "cat25c11", CAT25_INFO( 16, 8, 16, 1) }, @@ -756,7 +756,7 @@ static const struct spi_device_id m25p_ids[] = { }; MODULE_DEVICE_TABLE(spi, m25p_ids); -static const struct spi_device_id *__devinit jedec_probe(struct spi_device *spi) +static const struct spi_device_id *jedec_probe(struct spi_device *spi) { int tmp; u8 code = OPCODE_RDID; @@ -801,7 +801,7 @@ static const struct spi_device_id *__devinit jedec_probe(struct spi_device *spi) * matches what the READ command supports, at least until this driver * understands FAST_READ (for clocks over 25 MHz). */ -static int __devinit m25p_probe(struct spi_device *spi) +static int m25p_probe(struct spi_device *spi) { const struct spi_device_id *id = spi_get_device_id(spi); struct flash_platform_data *data; @@ -809,9 +809,10 @@ static int __devinit m25p_probe(struct spi_device *spi) struct flash_info *info; unsigned i; struct mtd_part_parser_data ppdata; + struct device_node __maybe_unused *np = spi->dev.of_node; #ifdef CONFIG_MTD_OF_PARTS - if (!of_device_is_available(spi->dev.of_node)) + if (!of_device_is_available(np)) return -ENODEV; #endif @@ -863,7 +864,8 @@ static int __devinit m25p_probe(struct spi_device *spi) flash = kzalloc(sizeof *flash, GFP_KERNEL); if (!flash) return -ENOMEM; - flash->command = kmalloc(MAX_CMD_SIZE + FAST_READ_DUMMY_BYTE, GFP_KERNEL); + flash->command = kmalloc(MAX_CMD_SIZE + (flash->fast_read ? 1 : 0), + GFP_KERNEL); if (!flash->command) { kfree(flash); return -ENOMEM; @@ -920,6 +922,16 @@ static int __devinit m25p_probe(struct spi_device *spi) flash->page_size = info->page_size; flash->mtd.writebufsize = flash->page_size; + flash->fast_read = false; +#ifdef CONFIG_OF + if (np && of_property_read_bool(np, "m25p,fast-read")) + flash->fast_read = true; +#endif + +#ifdef CONFIG_M25PXX_USE_FAST_READ + flash->fast_read = true; +#endif + if (info->addr_width) flash->addr_width = info->addr_width; else { @@ -961,7 +973,7 @@ static int __devinit m25p_probe(struct spi_device *spi) } -static int __devexit m25p_remove(struct spi_device *spi) +static int m25p_remove(struct spi_device *spi) { struct m25p *flash = dev_get_drvdata(&spi->dev); int status; @@ -983,7 +995,7 @@ static struct spi_driver m25p80_driver = { }, .id_table = m25p_ids, .probe = m25p_probe, - .remove = __devexit_p(m25p_remove), + .remove = m25p_remove, /* REVISIT: many of these chips have deep power-down modes, which * should clearly be entered on suspend() to minimize power use. diff --git a/drivers/mtd/devices/mtd_dataflash.c b/drivers/mtd/devices/mtd_dataflash.c index 928fb0e6d73a..ea7ea7b595d8 100644 --- a/drivers/mtd/devices/mtd_dataflash.c +++ b/drivers/mtd/devices/mtd_dataflash.c @@ -618,7 +618,7 @@ static char *otp_setup(struct mtd_info *device, char revision) /* * Register DataFlash device with MTD subsystem. */ -static int __devinit +static int add_dataflash_otp(struct spi_device *spi, char *name, int nr_pages, int pagesize, int pageoffset, char revision) { @@ -679,7 +679,7 @@ add_dataflash_otp(struct spi_device *spi, char *name, return err; } -static inline int __devinit +static inline int add_dataflash(struct spi_device *spi, char *name, int nr_pages, int pagesize, int pageoffset) { @@ -705,7 +705,7 @@ struct flash_info { #define IS_POW2PS 0x0001 /* uses 2^N byte pages */ }; -static struct flash_info __devinitdata dataflash_data [] = { +static struct flash_info dataflash_data[] = { /* * NOTE: chips with SUP_POW2PS (rev D and up) need two entries, @@ -740,7 +740,7 @@ static struct flash_info __devinitdata dataflash_data [] = { { "at45db642d", 0x1f2800, 8192, 1024, 10, SUP_POW2PS | IS_POW2PS}, }; -static struct flash_info *__devinit jedec_probe(struct spi_device *spi) +static struct flash_info *jedec_probe(struct spi_device *spi) { int tmp; uint8_t code = OP_READ_ID; @@ -823,7 +823,7 @@ static struct flash_info *__devinit jedec_probe(struct spi_device *spi) * AT45DB0642 64Mbit (8M) xx111xxx (0x3c) 8192 1056 11 * AT45DB1282 128Mbit (16M) xx0100xx (0x10) 16384 1056 11 */ -static int __devinit dataflash_probe(struct spi_device *spi) +static int dataflash_probe(struct spi_device *spi) { int status; struct flash_info *info; @@ -897,7 +897,7 @@ static int __devinit dataflash_probe(struct spi_device *spi) return status; } -static int __devexit dataflash_remove(struct spi_device *spi) +static int dataflash_remove(struct spi_device *spi) { struct dataflash *flash = dev_get_drvdata(&spi->dev); int status; @@ -920,7 +920,7 @@ static struct spi_driver dataflash_driver = { }, .probe = dataflash_probe, - .remove = __devexit_p(dataflash_remove), + .remove = dataflash_remove, /* FIXME: investigate suspend and resume... */ }; diff --git a/drivers/mtd/devices/spear_smi.c b/drivers/mtd/devices/spear_smi.c index dcc3c9511530..2d2c2a5d4d2a 100644 --- a/drivers/mtd/devices/spear_smi.c +++ b/drivers/mtd/devices/spear_smi.c @@ -756,7 +756,7 @@ err_probe: #ifdef CONFIG_OF -static int __devinit spear_smi_probe_config_dt(struct platform_device *pdev, +static int spear_smi_probe_config_dt(struct platform_device *pdev, struct device_node *np) { struct spear_smi_plat_data *pdata = dev_get_platdata(&pdev->dev); @@ -799,7 +799,7 @@ static int __devinit spear_smi_probe_config_dt(struct platform_device *pdev, return 0; } #else -static int __devinit spear_smi_probe_config_dt(struct platform_device *pdev, +static int spear_smi_probe_config_dt(struct platform_device *pdev, struct device_node *np) { return -ENOSYS; @@ -901,7 +901,7 @@ static int spear_smi_setup_banks(struct platform_device *pdev, * and do proper init for any found one. * Returns 0 on success, non zero otherwise */ -static int __devinit spear_smi_probe(struct platform_device *pdev) +static int spear_smi_probe(struct platform_device *pdev) { struct device_node *np = pdev->dev.of_node; struct spear_smi_plat_data *pdata = NULL; @@ -1016,7 +1016,7 @@ err: * * free all allocations and delete the partitions. */ -static int __devexit spear_smi_remove(struct platform_device *pdev) +static int spear_smi_remove(struct platform_device *pdev) { struct spear_smi *dev; struct spear_snor_flash *flash; @@ -1092,20 +1092,9 @@ static struct platform_driver spear_smi_driver = { #endif }, .probe = spear_smi_probe, - .remove = __devexit_p(spear_smi_remove), + .remove = spear_smi_remove, }; - -static int spear_smi_init(void) -{ - return platform_driver_register(&spear_smi_driver); -} -module_init(spear_smi_init); - -static void spear_smi_exit(void) -{ - platform_driver_unregister(&spear_smi_driver); -} -module_exit(spear_smi_exit); +module_platform_driver(spear_smi_driver); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Ashish Priyadarshi, Shiraz Hashim <shiraz.hashim@st.com>"); diff --git a/drivers/mtd/devices/sst25l.c b/drivers/mtd/devices/sst25l.c index ab8a2f4c8d60..8091b0163694 100644 --- a/drivers/mtd/devices/sst25l.c +++ b/drivers/mtd/devices/sst25l.c @@ -64,7 +64,7 @@ struct flash_info { #define to_sst25l_flash(x) container_of(x, struct sst25l_flash, mtd) -static struct flash_info __devinitdata sst25l_flash_info[] = { +static struct flash_info sst25l_flash_info[] = { {"sst25lf020a", 0xbf43, 256, 1024, 4096}, {"sst25lf040a", 0xbf44, 256, 2048, 4096}, }; @@ -313,7 +313,7 @@ out: return ret; } -static struct flash_info *__devinit sst25l_match_device(struct spi_device *spi) +static struct flash_info *sst25l_match_device(struct spi_device *spi) { struct flash_info *flash_info = NULL; struct spi_message m; @@ -353,7 +353,7 @@ static struct flash_info *__devinit sst25l_match_device(struct spi_device *spi) return flash_info; } -static int __devinit sst25l_probe(struct spi_device *spi) +static int sst25l_probe(struct spi_device *spi) { struct flash_info *flash_info; struct sst25l_flash *flash; @@ -411,7 +411,7 @@ static int __devinit sst25l_probe(struct spi_device *spi) return 0; } -static int __devexit sst25l_remove(struct spi_device *spi) +static int sst25l_remove(struct spi_device *spi) { struct sst25l_flash *flash = dev_get_drvdata(&spi->dev); int ret; @@ -428,7 +428,7 @@ static struct spi_driver sst25l_driver = { .owner = THIS_MODULE, }, .probe = sst25l_probe, - .remove = __devexit_p(sst25l_remove), + .remove = sst25l_remove, }; module_spi_driver(sst25l_driver); diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig index df304868bebb..62ba82c396c2 100644 --- a/drivers/mtd/maps/Kconfig +++ b/drivers/mtd/maps/Kconfig @@ -358,13 +358,6 @@ config MTD_IXP2000 IXP2000 based board and would like to use the flash chips on it, say 'Y'. -config MTD_FORTUNET - tristate "CFI Flash device mapped on the FortuNet board" - depends on MTD_CFI && SA1100_FORTUNET - help - This enables access to the Flash on the FortuNet board. If you - have such a board, say 'Y'. - config MTD_AUTCPU12 bool "NV-RAM mapping AUTCPU12 board" depends on ARCH_AUTCPU12 diff --git a/drivers/mtd/maps/Makefile b/drivers/mtd/maps/Makefile index a0240edd1961..4ded28711bc1 100644 --- a/drivers/mtd/maps/Makefile +++ b/drivers/mtd/maps/Makefile @@ -39,7 +39,6 @@ obj-$(CONFIG_MTD_SOLUTIONENGINE)+= solutionengine.o obj-$(CONFIG_MTD_PCI) += pci.o obj-$(CONFIG_MTD_AUTCPU12) += autcpu12-nvram.o obj-$(CONFIG_MTD_IMPA7) += impa7.o -obj-$(CONFIG_MTD_FORTUNET) += fortunet.o obj-$(CONFIG_MTD_UCLINUX) += uclinux.o obj-$(CONFIG_MTD_NETtel) += nettel.o obj-$(CONFIG_MTD_SCB2_FLASH) += scb2_flash.o diff --git a/drivers/mtd/maps/amd76xrom.c b/drivers/mtd/maps/amd76xrom.c index e2875d6fe129..f7207b0a76dc 100644 --- a/drivers/mtd/maps/amd76xrom.c +++ b/drivers/mtd/maps/amd76xrom.c @@ -100,8 +100,8 @@ static void amd76xrom_cleanup(struct amd76xrom_window *window) } -static int __devinit amd76xrom_init_one (struct pci_dev *pdev, - const struct pci_device_id *ent) +static int amd76xrom_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) { static char *rom_probe_types[] = { "cfi_probe", "jedec_probe", NULL }; u8 byte; @@ -289,7 +289,7 @@ static int __devinit amd76xrom_init_one (struct pci_dev *pdev, } -static void __devexit amd76xrom_remove_one (struct pci_dev *pdev) +static void amd76xrom_remove_one(struct pci_dev *pdev) { struct amd76xrom_window *window = &amd76xrom_window; @@ -347,4 +347,3 @@ module_exit(cleanup_amd76xrom); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Eric Biederman <ebiederman@lnxi.com>"); MODULE_DESCRIPTION("MTD map driver for BIOS chips on the AMD76X southbridge"); - diff --git a/drivers/mtd/maps/autcpu12-nvram.c b/drivers/mtd/maps/autcpu12-nvram.c index 76fb594bb1d9..a2dc2ae4b24e 100644 --- a/drivers/mtd/maps/autcpu12-nvram.c +++ b/drivers/mtd/maps/autcpu12-nvram.c @@ -33,7 +33,7 @@ struct autcpu12_nvram_priv { struct map_info map; }; -static int __devinit autcpu12_nvram_probe(struct platform_device *pdev) +static int autcpu12_nvram_probe(struct platform_device *pdev) { map_word tmp, save0, save1; struct resource *res; @@ -105,7 +105,7 @@ static int __devinit autcpu12_nvram_probe(struct platform_device *pdev) return -ENOMEM; } -static int __devexit autcpu12_nvram_remove(struct platform_device *pdev) +static int autcpu12_nvram_remove(struct platform_device *pdev) { struct autcpu12_nvram_priv *priv = platform_get_drvdata(pdev); @@ -121,7 +121,7 @@ static struct platform_driver autcpu12_nvram_driver = { .owner = THIS_MODULE, }, .probe = autcpu12_nvram_probe, - .remove = __devexit_p(autcpu12_nvram_remove), + .remove = autcpu12_nvram_remove, }; module_platform_driver(autcpu12_nvram_driver); diff --git a/drivers/mtd/maps/bfin-async-flash.c b/drivers/mtd/maps/bfin-async-flash.c index ef5cde84a8b3..f833edfaab79 100644 --- a/drivers/mtd/maps/bfin-async-flash.c +++ b/drivers/mtd/maps/bfin-async-flash.c @@ -30,7 +30,8 @@ #include <linux/io.h> #include <asm/unaligned.h> -#define pr_devinit(fmt, args...) ({ static const __devinitconst char __fmt[] = fmt; printk(__fmt, ## args); }) +#define pr_devinit(fmt, args...) \ + ({ static const char __fmt[] = fmt; printk(__fmt, ## args); }) #define DRIVER_NAME "bfin-async-flash" @@ -123,7 +124,7 @@ static void bfin_flash_copy_to(struct map_info *map, unsigned long to, const voi static const char *part_probe_types[] = { "cmdlinepart", "RedBoot", NULL }; -static int __devinit bfin_flash_probe(struct platform_device *pdev) +static int bfin_flash_probe(struct platform_device *pdev) { int ret; struct physmap_flash_data *pdata = pdev->dev.platform_data; @@ -172,7 +173,7 @@ static int __devinit bfin_flash_probe(struct platform_device *pdev) return 0; } -static int __devexit bfin_flash_remove(struct platform_device *pdev) +static int bfin_flash_remove(struct platform_device *pdev) { struct async_state *state = platform_get_drvdata(pdev); gpio_free(state->enet_flash_pin); @@ -184,7 +185,7 @@ static int __devexit bfin_flash_remove(struct platform_device *pdev) static struct platform_driver bfin_flash_driver = { .probe = bfin_flash_probe, - .remove = __devexit_p(bfin_flash_remove), + .remove = bfin_flash_remove, .driver = { .name = DRIVER_NAME, }, diff --git a/drivers/mtd/maps/ck804xrom.c b/drivers/mtd/maps/ck804xrom.c index 3d0e762fa5f2..586a1c77e48a 100644 --- a/drivers/mtd/maps/ck804xrom.c +++ b/drivers/mtd/maps/ck804xrom.c @@ -112,8 +112,8 @@ static void ck804xrom_cleanup(struct ck804xrom_window *window) } -static int __devinit ck804xrom_init_one (struct pci_dev *pdev, - const struct pci_device_id *ent) +static int ck804xrom_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) { static char *rom_probe_types[] = { "cfi_probe", "jedec_probe", NULL }; u8 byte; @@ -320,7 +320,7 @@ static int __devinit ck804xrom_init_one (struct pci_dev *pdev, } -static void __devexit ck804xrom_remove_one (struct pci_dev *pdev) +static void ck804xrom_remove_one(struct pci_dev *pdev) { struct ck804xrom_window *window = &ck804xrom_window; diff --git a/drivers/mtd/maps/esb2rom.c b/drivers/mtd/maps/esb2rom.c index 08322b1c3e81..ff8681a25831 100644 --- a/drivers/mtd/maps/esb2rom.c +++ b/drivers/mtd/maps/esb2rom.c @@ -144,7 +144,7 @@ static void esb2rom_cleanup(struct esb2rom_window *window) pci_dev_put(window->pdev); } -static int __devinit esb2rom_init_one(struct pci_dev *pdev, +static int esb2rom_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { static char *rom_probe_types[] = { "cfi_probe", "jedec_probe", NULL }; @@ -378,13 +378,13 @@ static int __devinit esb2rom_init_one(struct pci_dev *pdev, return 0; } -static void __devexit esb2rom_remove_one (struct pci_dev *pdev) +static void esb2rom_remove_one(struct pci_dev *pdev) { struct esb2rom_window *window = &esb2rom_window; esb2rom_cleanup(window); } -static struct pci_device_id esb2rom_pci_tbl[] __devinitdata = { +static struct pci_device_id esb2rom_pci_tbl[] = { { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801BA_0, PCI_ANY_ID, PCI_ANY_ID, }, { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_0, diff --git a/drivers/mtd/maps/fortunet.c b/drivers/mtd/maps/fortunet.c deleted file mode 100644 index 956e2e4f30ea..000000000000 --- a/drivers/mtd/maps/fortunet.c +++ /dev/null @@ -1,277 +0,0 @@ -/* fortunet.c memory map - * - */ - -#include <linux/module.h> -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/string.h> - -#include <linux/mtd/mtd.h> -#include <linux/mtd/map.h> -#include <linux/mtd/partitions.h> - -#include <asm/io.h> - -#define MAX_NUM_REGIONS 4 -#define MAX_NUM_PARTITIONS 8 - -#define DEF_WINDOW_ADDR_PHY 0x00000000 -#define DEF_WINDOW_SIZE 0x00800000 // 8 Mega Bytes - -#define MTD_FORTUNET_PK "MTD FortuNet: " - -#define MAX_NAME_SIZE 128 - -struct map_region -{ - int window_addr_physical; - int altbankwidth; - struct map_info map_info; - struct mtd_info *mymtd; - struct mtd_partition parts[MAX_NUM_PARTITIONS]; - char map_name[MAX_NAME_SIZE]; - char parts_name[MAX_NUM_PARTITIONS][MAX_NAME_SIZE]; -}; - -static struct map_region map_regions[MAX_NUM_REGIONS]; -static int map_regions_set[MAX_NUM_REGIONS] = {0,0,0,0}; -static int map_regions_parts[MAX_NUM_REGIONS] = {0,0,0,0}; - - - -struct map_info default_map = { - .size = DEF_WINDOW_SIZE, - .bankwidth = 4, -}; - -static char * __init get_string_option(char *dest,int dest_size,char *sor) -{ - if(!dest_size) - return sor; - dest_size--; - while(*sor) - { - if(*sor==',') - { - sor++; - break; - } - else if(*sor=='\"') - { - sor++; - while(*sor) - { - if(*sor=='\"') - { - sor++; - break; - } - *dest = *sor; - dest++; - sor++; - dest_size--; - if(!dest_size) - { - *dest = 0; - return sor; - } - } - } - else - { - *dest = *sor; - dest++; - sor++; - dest_size--; - if(!dest_size) - { - *dest = 0; - return sor; - } - } - } - *dest = 0; - return sor; -} - -static int __init MTD_New_Region(char *line) -{ - char string[MAX_NAME_SIZE]; - int params[6]; - get_options (get_string_option(string,sizeof(string),line),6,params); - if(params[0]<1) - { - printk(MTD_FORTUNET_PK "Bad parameters for MTD Region " - " name,region-number[,base,size,bankwidth,altbankwidth]\n"); - return 1; - } - if((params[1]<0)||(params[1]>=MAX_NUM_REGIONS)) - { - printk(MTD_FORTUNET_PK "Bad region index of %d only have 0..%u regions\n", - params[1],MAX_NUM_REGIONS-1); - return 1; - } - memset(&map_regions[params[1]],0,sizeof(map_regions[params[1]])); - memcpy(&map_regions[params[1]].map_info, - &default_map,sizeof(map_regions[params[1]].map_info)); - map_regions_set[params[1]] = 1; - map_regions[params[1]].window_addr_physical = DEF_WINDOW_ADDR_PHY; - map_regions[params[1]].altbankwidth = 2; - map_regions[params[1]].mymtd = NULL; - map_regions[params[1]].map_info.name = map_regions[params[1]].map_name; - strcpy(map_regions[params[1]].map_info.name,string); - if(params[0]>1) - { - map_regions[params[1]].window_addr_physical = params[2]; - } - if(params[0]>2) - { - map_regions[params[1]].map_info.size = params[3]; - } - if(params[0]>3) - { - map_regions[params[1]].map_info.bankwidth = params[4]; - } - if(params[0]>4) - { - map_regions[params[1]].altbankwidth = params[5]; - } - return 1; -} - -static int __init MTD_New_Partition(char *line) -{ - char string[MAX_NAME_SIZE]; - int params[4]; - get_options (get_string_option(string,sizeof(string),line),4,params); - if(params[0]<3) - { - printk(MTD_FORTUNET_PK "Bad parameters for MTD Partition " - " name,region-number,size,offset\n"); - return 1; - } - if((params[1]<0)||(params[1]>=MAX_NUM_REGIONS)) - { - printk(MTD_FORTUNET_PK "Bad region index of %d only have 0..%u regions\n", - params[1],MAX_NUM_REGIONS-1); - return 1; - } - if(map_regions_parts[params[1]]>=MAX_NUM_PARTITIONS) - { - printk(MTD_FORTUNET_PK "Out of space for partition in this region\n"); - return 1; - } - map_regions[params[1]].parts[map_regions_parts[params[1]]].name = - map_regions[params[1]]. parts_name[map_regions_parts[params[1]]]; - strcpy(map_regions[params[1]].parts[map_regions_parts[params[1]]].name,string); - map_regions[params[1]].parts[map_regions_parts[params[1]]].size = - params[2]; - map_regions[params[1]].parts[map_regions_parts[params[1]]].offset = - params[3]; - map_regions[params[1]].parts[map_regions_parts[params[1]]].mask_flags = 0; - map_regions_parts[params[1]]++; - return 1; -} - -__setup("MTD_Region=", MTD_New_Region); -__setup("MTD_Partition=", MTD_New_Partition); - -/* Backwards-spelling-compatibility */ -__setup("MTD_Partion=", MTD_New_Partition); - -static int __init init_fortunet(void) -{ - int ix,iy; - for(iy=ix=0;ix<MAX_NUM_REGIONS;ix++) - { - if(map_regions_parts[ix]&&(!map_regions_set[ix])) - { - printk(MTD_FORTUNET_PK "Region %d is not setup (Setting to default)\n", - ix); - memset(&map_regions[ix],0,sizeof(map_regions[ix])); - memcpy(&map_regions[ix].map_info,&default_map, - sizeof(map_regions[ix].map_info)); - map_regions_set[ix] = 1; - map_regions[ix].window_addr_physical = DEF_WINDOW_ADDR_PHY; - map_regions[ix].altbankwidth = 2; - map_regions[ix].mymtd = NULL; - map_regions[ix].map_info.name = map_regions[ix].map_name; - strcpy(map_regions[ix].map_info.name,"FORTUNET"); - } - if(map_regions_set[ix]) - { - iy++; - printk(KERN_NOTICE MTD_FORTUNET_PK "%s flash device at physically " - " address %x size %x\n", - map_regions[ix].map_info.name, - map_regions[ix].window_addr_physical, - map_regions[ix].map_info.size); - - map_regions[ix].map_info.phys = map_regions[ix].window_addr_physical, - - map_regions[ix].map_info.virt = - ioremap_nocache( - map_regions[ix].window_addr_physical, - map_regions[ix].map_info.size); - if(!map_regions[ix].map_info.virt) - { - int j = 0; - printk(MTD_FORTUNET_PK "%s flash failed to ioremap!\n", - map_regions[ix].map_info.name); - for (j = 0 ; j < ix; j++) - iounmap(map_regions[j].map_info.virt); - return -ENXIO; - } - simple_map_init(&map_regions[ix].map_info); - - printk(KERN_NOTICE MTD_FORTUNET_PK "%s flash is virtually at: %x\n", - map_regions[ix].map_info.name, - map_regions[ix].map_info.virt); - map_regions[ix].mymtd = do_map_probe("cfi_probe", - &map_regions[ix].map_info); - if((!map_regions[ix].mymtd)&&( - map_regions[ix].altbankwidth!=map_regions[ix].map_info.bankwidth)) - { - printk(KERN_NOTICE MTD_FORTUNET_PK "Trying alternate bankwidth " - "for %s flash.\n", - map_regions[ix].map_info.name); - map_regions[ix].map_info.bankwidth = - map_regions[ix].altbankwidth; - map_regions[ix].mymtd = do_map_probe("cfi_probe", - &map_regions[ix].map_info); - } - map_regions[ix].mymtd->owner = THIS_MODULE; - mtd_device_register(map_regions[ix].mymtd, - map_regions[ix].parts, - map_regions_parts[ix]); - } - } - if(iy) - return 0; - return -ENXIO; -} - -static void __exit cleanup_fortunet(void) -{ - int ix; - for(ix=0;ix<MAX_NUM_REGIONS;ix++) - { - if(map_regions_set[ix]) - { - if( map_regions[ix].mymtd ) - { - mtd_device_unregister(map_regions[ix].mymtd); - map_destroy( map_regions[ix].mymtd ); - } - iounmap((void *)map_regions[ix].map_info.virt); - } - } -} - -module_init(init_fortunet); -module_exit(cleanup_fortunet); - -MODULE_AUTHOR("FortuNet, Inc."); -MODULE_DESCRIPTION("MTD map driver for FortuNet boards"); diff --git a/drivers/mtd/maps/gpio-addr-flash.c b/drivers/mtd/maps/gpio-addr-flash.c index e4de96ba52b3..7b643de2500b 100644 --- a/drivers/mtd/maps/gpio-addr-flash.c +++ b/drivers/mtd/maps/gpio-addr-flash.c @@ -26,7 +26,8 @@ #include <linux/slab.h> #include <linux/types.h> -#define pr_devinit(fmt, args...) ({ static const __devinitconst char __fmt[] = fmt; printk(__fmt, ## args); }) +#define pr_devinit(fmt, args...) \ + ({ static const char __fmt[] = fmt; printk(__fmt, ## args); }) #define DRIVER_NAME "gpio-addr-flash" #define PFX DRIVER_NAME ": " @@ -142,7 +143,8 @@ static void gf_write(struct map_info *map, map_word d1, unsigned long ofs) * * See gf_copy_from() caveat. */ -static void gf_copy_to(struct map_info *map, unsigned long to, const void *from, ssize_t len) +static void gf_copy_to(struct map_info *map, unsigned long to, + const void *from, ssize_t len) { struct async_state *state = gf_map_info_to_state(map); @@ -185,7 +187,7 @@ static const char *part_probe_types[] = { "cmdlinepart", "RedBoot", NULL }; * ... * }; */ -static int __devinit gpio_flash_probe(struct platform_device *pdev) +static int gpio_flash_probe(struct platform_device *pdev) { size_t i, arr_size; struct physmap_flash_data *pdata; @@ -258,7 +260,7 @@ static int __devinit gpio_flash_probe(struct platform_device *pdev) return 0; } -static int __devexit gpio_flash_remove(struct platform_device *pdev) +static int gpio_flash_remove(struct platform_device *pdev) { struct async_state *state = platform_get_drvdata(pdev); size_t i = 0; @@ -273,7 +275,7 @@ static int __devexit gpio_flash_remove(struct platform_device *pdev) static struct platform_driver gpio_flash_driver = { .probe = gpio_flash_probe, - .remove = __devexit_p(gpio_flash_remove), + .remove = gpio_flash_remove, .driver = { .name = DRIVER_NAME, }, diff --git a/drivers/mtd/maps/ichxrom.c b/drivers/mtd/maps/ichxrom.c index 6689dcb3124d..c7478e18f485 100644 --- a/drivers/mtd/maps/ichxrom.c +++ b/drivers/mtd/maps/ichxrom.c @@ -84,8 +84,8 @@ static void ichxrom_cleanup(struct ichxrom_window *window) } -static int __devinit ichxrom_init_one (struct pci_dev *pdev, - const struct pci_device_id *ent) +static int ichxrom_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) { static char *rom_probe_types[] = { "cfi_probe", "jedec_probe", NULL }; struct ichxrom_window *window = &ichxrom_window; @@ -315,13 +315,13 @@ static int __devinit ichxrom_init_one (struct pci_dev *pdev, } -static void __devexit ichxrom_remove_one (struct pci_dev *pdev) +static void ichxrom_remove_one(struct pci_dev *pdev) { struct ichxrom_window *window = &ichxrom_window; ichxrom_cleanup(window); } -static struct pci_device_id ichxrom_pci_tbl[] __devinitdata = { +static struct pci_device_id ichxrom_pci_tbl[] = { { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801BA_0, PCI_ANY_ID, PCI_ANY_ID, }, { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_0, diff --git a/drivers/mtd/maps/intel_vr_nor.c b/drivers/mtd/maps/intel_vr_nor.c index 93f03175c82d..3ee2ad1dcbe7 100644 --- a/drivers/mtd/maps/intel_vr_nor.c +++ b/drivers/mtd/maps/intel_vr_nor.c @@ -63,24 +63,24 @@ struct vr_nor_mtd { #define TIMING_BYTE_EN (1 << 0) /* 8-bit vs 16-bit bus */ #define TIMING_MASK 0x3FFF0000 -static void __devexit vr_nor_destroy_partitions(struct vr_nor_mtd *p) +static void vr_nor_destroy_partitions(struct vr_nor_mtd *p) { mtd_device_unregister(p->info); } -static int __devinit vr_nor_init_partitions(struct vr_nor_mtd *p) +static int vr_nor_init_partitions(struct vr_nor_mtd *p) { /* register the flash bank */ /* partition the flash bank */ return mtd_device_parse_register(p->info, NULL, NULL, NULL, 0); } -static void __devexit vr_nor_destroy_mtd_setup(struct vr_nor_mtd *p) +static void vr_nor_destroy_mtd_setup(struct vr_nor_mtd *p) { map_destroy(p->info); } -static int __devinit vr_nor_mtd_setup(struct vr_nor_mtd *p) +static int vr_nor_mtd_setup(struct vr_nor_mtd *p) { static const char *probe_types[] = { "cfi_probe", "jedec_probe", NULL }; @@ -96,7 +96,7 @@ static int __devinit vr_nor_mtd_setup(struct vr_nor_mtd *p) return 0; } -static void __devexit vr_nor_destroy_maps(struct vr_nor_mtd *p) +static void vr_nor_destroy_maps(struct vr_nor_mtd *p) { unsigned int exp_timing_cs0; @@ -116,7 +116,7 @@ static void __devexit vr_nor_destroy_maps(struct vr_nor_mtd *p) * Initialize the map_info structure and map the flash. * Returns 0 on success, nonzero otherwise. */ -static int __devinit vr_nor_init_maps(struct vr_nor_mtd *p) +static int vr_nor_init_maps(struct vr_nor_mtd *p) { unsigned long csr_phys, csr_len; unsigned long win_phys, win_len; @@ -176,7 +176,7 @@ static struct pci_device_id vr_nor_pci_ids[] = { {0,} }; -static void __devexit vr_nor_pci_remove(struct pci_dev *dev) +static void vr_nor_pci_remove(struct pci_dev *dev) { struct vr_nor_mtd *p = pci_get_drvdata(dev); @@ -189,7 +189,7 @@ static void __devexit vr_nor_pci_remove(struct pci_dev *dev) pci_disable_device(dev); } -static int __devinit +static int vr_nor_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) { struct vr_nor_mtd *p = NULL; @@ -256,7 +256,7 @@ vr_nor_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) static struct pci_driver vr_nor_pci_driver = { .name = DRV_NAME, .probe = vr_nor_pci_probe, - .remove = __devexit_p(vr_nor_pci_remove), + .remove = vr_nor_pci_remove, .id_table = vr_nor_pci_ids, }; diff --git a/drivers/mtd/maps/lantiq-flash.c b/drivers/mtd/maps/lantiq-flash.c index c03456f17004..3c3c791eb96a 100644 --- a/drivers/mtd/maps/lantiq-flash.c +++ b/drivers/mtd/maps/lantiq-flash.c @@ -45,7 +45,7 @@ struct ltq_mtd { }; static const char ltq_map_name[] = "ltq_nor"; -static const char *ltq_probe_types[] __devinitconst = { +static const char *ltq_probe_types[] = { "cmdlinepart", "ofpart", NULL }; static map_word @@ -109,7 +109,7 @@ ltq_copy_to(struct map_info *map, unsigned long to, spin_unlock_irqrestore(&ebu_lock, flags); } -static int __devinit +static int ltq_mtd_probe(struct platform_device *pdev) { struct mtd_part_parser_data ppdata; @@ -185,7 +185,7 @@ err_out: return err; } -static int __devexit +static int ltq_mtd_remove(struct platform_device *pdev) { struct ltq_mtd *ltq_mtd = platform_get_drvdata(pdev); @@ -209,7 +209,7 @@ MODULE_DEVICE_TABLE(of, ltq_mtd_match); static struct platform_driver ltq_mtd_driver = { .probe = ltq_mtd_probe, - .remove = __devexit_p(ltq_mtd_remove), + .remove = ltq_mtd_remove, .driver = { .name = "ltq-nor", .owner = THIS_MODULE, diff --git a/drivers/mtd/maps/latch-addr-flash.c b/drivers/mtd/maps/latch-addr-flash.c index 3c7ad17fca78..ab0fead56b83 100644 --- a/drivers/mtd/maps/latch-addr-flash.c +++ b/drivers/mtd/maps/latch-addr-flash.c @@ -125,7 +125,7 @@ static int latch_addr_flash_remove(struct platform_device *dev) return 0; } -static int __devinit latch_addr_flash_probe(struct platform_device *dev) +static int latch_addr_flash_probe(struct platform_device *dev) { struct latch_addr_flash_data *latch_addr_data; struct latch_addr_flash_info *info; @@ -218,7 +218,7 @@ done: static struct platform_driver latch_addr_flash_driver = { .probe = latch_addr_flash_probe, - .remove = __devexit_p(latch_addr_flash_remove), + .remove = latch_addr_flash_remove, .driver = { .name = DRIVER_NAME, }, diff --git a/drivers/mtd/maps/pci.c b/drivers/mtd/maps/pci.c index 1c30c1a307f4..ed82914966f5 100644 --- a/drivers/mtd/maps/pci.c +++ b/drivers/mtd/maps/pci.c @@ -253,7 +253,7 @@ static struct pci_device_id mtd_pci_ids[] = { * Generic code follows. */ -static int __devinit +static int mtd_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) { struct mtd_pci_info *info = (struct mtd_pci_info *)id->driver_data; @@ -308,7 +308,7 @@ out: return err; } -static void __devexit +static void mtd_pci_remove(struct pci_dev *dev) { struct mtd_info *mtd = pci_get_drvdata(dev); @@ -326,7 +326,7 @@ mtd_pci_remove(struct pci_dev *dev) static struct pci_driver mtd_pci_driver = { .name = "MTD PCI", .probe = mtd_pci_probe, - .remove = __devexit_p(mtd_pci_remove), + .remove = mtd_pci_remove, .id_table = mtd_pci_ids, }; diff --git a/drivers/mtd/maps/physmap_of.c b/drivers/mtd/maps/physmap_of.c index 6f19acadb06c..37cdc201652f 100644 --- a/drivers/mtd/maps/physmap_of.c +++ b/drivers/mtd/maps/physmap_of.c @@ -77,7 +77,7 @@ static int of_flash_remove(struct platform_device *dev) /* Helper function to handle probing of the obsolete "direct-mapped" * compatible binding, which has an extra "probe-type" property * describing the type of flash probe necessary. */ -static struct mtd_info * __devinit obsolete_probe(struct platform_device *dev, +static struct mtd_info *obsolete_probe(struct platform_device *dev, struct map_info *map) { struct device_node *dp = dev->dev.of_node; @@ -116,7 +116,7 @@ static struct mtd_info * __devinit obsolete_probe(struct platform_device *dev, information. */ static const char *part_probe_types_def[] = { "cmdlinepart", "RedBoot", "ofpart", "ofoldpart", NULL }; -static const char ** __devinit of_get_probes(struct device_node *dp) +static const char **of_get_probes(struct device_node *dp) { const char *cp; int cplen; @@ -145,14 +145,14 @@ static const char ** __devinit of_get_probes(struct device_node *dp) return res; } -static void __devinit of_free_probes(const char **probes) +static void of_free_probes(const char **probes) { if (probes != part_probe_types_def) kfree(probes); } static struct of_device_id of_flash_match[]; -static int __devinit of_flash_probe(struct platform_device *dev) +static int of_flash_probe(struct platform_device *dev) { const char **part_probe_types; const struct of_device_id *match; @@ -170,6 +170,7 @@ static int __devinit of_flash_probe(struct platform_device *dev) resource_size_t res_size; struct mtd_part_parser_data ppdata; bool map_indirect; + const char *mtd_name; match = of_match_device(of_flash_match, &dev->dev); if (!match) @@ -178,6 +179,8 @@ static int __devinit of_flash_probe(struct platform_device *dev) reg_tuple_size = (of_n_addr_cells(dp) + of_n_size_cells(dp)) * sizeof(u32); + of_property_read_string(dp, "linux,mtd-name", &mtd_name); + /* * Get number of "reg" tuples. Scan for MTD devices on area's * described by each "reg" region. This makes it possible (including @@ -234,7 +237,7 @@ static int __devinit of_flash_probe(struct platform_device *dev) goto err_out; } - info->list[i].map.name = dev_name(&dev->dev); + info->list[i].map.name = mtd_name ?: dev_name(&dev->dev); info->list[i].map.phys = res.start; info->list[i].map.size = res_size; info->list[i].map.bankwidth = be32_to_cpup(width); @@ -282,6 +285,7 @@ static int __devinit of_flash_probe(struct platform_device *dev) } err = 0; + info->cmtd = NULL; if (info->list_size == 1) { info->cmtd = info->list[0].mtd; } else if (info->list_size > 1) { @@ -290,9 +294,10 @@ static int __devinit of_flash_probe(struct platform_device *dev) */ info->cmtd = mtd_concat_create(mtd_list, info->list_size, dev_name(&dev->dev)); - if (info->cmtd == NULL) - err = -ENXIO; } + if (info->cmtd == NULL) + err = -ENXIO; + if (err) goto err_out; diff --git a/drivers/mtd/maps/pismo.c b/drivers/mtd/maps/pismo.c index 65bd1cd4d627..afea93b515d5 100644 --- a/drivers/mtd/maps/pismo.c +++ b/drivers/mtd/maps/pismo.c @@ -58,7 +58,7 @@ static void pismo_set_vpp(struct platform_device *pdev, int on) pismo->vpp(pismo->vpp_data, on); } -static unsigned int __devinit pismo_width_to_bytes(unsigned int width) +static unsigned int pismo_width_to_bytes(unsigned int width) { width &= 15; if (width > 2) @@ -66,7 +66,7 @@ static unsigned int __devinit pismo_width_to_bytes(unsigned int width) return 1 << width; } -static int __devinit pismo_eeprom_read(struct i2c_client *client, void *buf, +static int pismo_eeprom_read(struct i2c_client *client, void *buf, u8 addr, size_t size) { int ret; @@ -88,7 +88,7 @@ static int __devinit pismo_eeprom_read(struct i2c_client *client, void *buf, return ret == ARRAY_SIZE(msg) ? size : -EIO; } -static int __devinit pismo_add_device(struct pismo_data *pismo, int i, +static int pismo_add_device(struct pismo_data *pismo, int i, struct pismo_mem *region, const char *name, void *pdata, size_t psize) { struct platform_device *dev; @@ -129,7 +129,7 @@ static int __devinit pismo_add_device(struct pismo_data *pismo, int i, return ret; } -static int __devinit pismo_add_nor(struct pismo_data *pismo, int i, +static int pismo_add_nor(struct pismo_data *pismo, int i, struct pismo_mem *region) { struct physmap_flash_data data = { @@ -143,7 +143,7 @@ static int __devinit pismo_add_nor(struct pismo_data *pismo, int i, &data, sizeof(data)); } -static int __devinit pismo_add_sram(struct pismo_data *pismo, int i, +static int pismo_add_sram(struct pismo_data *pismo, int i, struct pismo_mem *region) { struct platdata_mtd_ram data = { @@ -154,7 +154,7 @@ static int __devinit pismo_add_sram(struct pismo_data *pismo, int i, &data, sizeof(data)); } -static void __devinit pismo_add_one(struct pismo_data *pismo, int i, +static void pismo_add_one(struct pismo_data *pismo, int i, const struct pismo_cs_block *cs, phys_addr_t base) { struct device *dev = &pismo->client->dev; @@ -197,7 +197,7 @@ static void __devinit pismo_add_one(struct pismo_data *pismo, int i, } } -static int __devexit pismo_remove(struct i2c_client *client) +static int pismo_remove(struct i2c_client *client) { struct pismo_data *pismo = i2c_get_clientdata(client); int i; @@ -210,7 +210,7 @@ static int __devexit pismo_remove(struct i2c_client *client) return 0; } -static int __devinit pismo_probe(struct i2c_client *client, +static int pismo_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct i2c_adapter *adapter = to_i2c_adapter(client->dev.parent); @@ -267,7 +267,7 @@ static struct i2c_driver pismo_driver = { .owner = THIS_MODULE, }, .probe = pismo_probe, - .remove = __devexit_p(pismo_remove), + .remove = pismo_remove, .id_table = pismo_id, }; diff --git a/drivers/mtd/maps/pxa2xx-flash.c b/drivers/mtd/maps/pxa2xx-flash.c index 81884c277405..43e3dbb976d9 100644 --- a/drivers/mtd/maps/pxa2xx-flash.c +++ b/drivers/mtd/maps/pxa2xx-flash.c @@ -49,7 +49,7 @@ struct pxa2xx_flash_info { static const char *probes[] = { "RedBoot", "cmdlinepart", NULL }; -static int __devinit pxa2xx_flash_probe(struct platform_device *pdev) +static int pxa2xx_flash_probe(struct platform_device *pdev) { struct flash_platform_data *flash = pdev->dev.platform_data; struct pxa2xx_flash_info *info; @@ -105,7 +105,7 @@ static int __devinit pxa2xx_flash_probe(struct platform_device *pdev) return 0; } -static int __devexit pxa2xx_flash_remove(struct platform_device *dev) +static int pxa2xx_flash_remove(struct platform_device *dev) { struct pxa2xx_flash_info *info = platform_get_drvdata(dev); @@ -139,7 +139,7 @@ static struct platform_driver pxa2xx_flash_driver = { .owner = THIS_MODULE, }, .probe = pxa2xx_flash_probe, - .remove = __devexit_p(pxa2xx_flash_remove), + .remove = pxa2xx_flash_remove, .shutdown = pxa2xx_flash_shutdown, }; diff --git a/drivers/mtd/maps/sa1100-flash.c b/drivers/mtd/maps/sa1100-flash.c index a675bdbcb0fe..f694417cf7e6 100644 --- a/drivers/mtd/maps/sa1100-flash.c +++ b/drivers/mtd/maps/sa1100-flash.c @@ -149,8 +149,8 @@ static void sa1100_destroy(struct sa_info *info, struct flash_platform_data *pla plat->exit(); } -static struct sa_info *__devinit -sa1100_setup_mtd(struct platform_device *pdev, struct flash_platform_data *plat) +static struct sa_info *sa1100_setup_mtd(struct platform_device *pdev, + struct flash_platform_data *plat) { struct sa_info *info; int nr, size, i, ret = 0; @@ -246,7 +246,7 @@ sa1100_setup_mtd(struct platform_device *pdev, struct flash_platform_data *plat) static const char *part_probes[] = { "cmdlinepart", "RedBoot", NULL }; -static int __devinit sa1100_mtd_probe(struct platform_device *pdev) +static int sa1100_mtd_probe(struct platform_device *pdev) { struct flash_platform_data *plat = pdev->dev.platform_data; struct sa_info *info; diff --git a/drivers/mtd/maps/scb2_flash.c b/drivers/mtd/maps/scb2_flash.c index 9dcbc684abdb..71796137e97b 100644 --- a/drivers/mtd/maps/scb2_flash.c +++ b/drivers/mtd/maps/scb2_flash.c @@ -69,7 +69,7 @@ static struct map_info scb2_map = { }; static int region_fail; -static int __devinit +static int scb2_fixup_mtd(struct mtd_info *mtd) { int i; @@ -133,7 +133,7 @@ scb2_fixup_mtd(struct mtd_info *mtd) /* CSB5's 'Function Control Register' has bits for decoding @ >= 0xffc00000 */ #define CSB5_FCR 0x41 #define CSB5_FCR_DECODE_ALL 0x0e -static int __devinit +static int scb2_flash_probe(struct pci_dev *dev, const struct pci_device_id *ent) { u8 reg; @@ -197,7 +197,7 @@ scb2_flash_probe(struct pci_dev *dev, const struct pci_device_id *ent) return 0; } -static void __devexit +static void scb2_flash_remove(struct pci_dev *dev) { if (!scb2_mtd) @@ -231,7 +231,7 @@ static struct pci_driver scb2_flash_driver = { .name = "Intel SCB2 BIOS Flash", .id_table = scb2_flash_pci_ids, .probe = scb2_flash_probe, - .remove = __devexit_p(scb2_flash_remove), + .remove = scb2_flash_remove, }; module_pci_driver(scb2_flash_driver); diff --git a/drivers/mtd/maps/sun_uflash.c b/drivers/mtd/maps/sun_uflash.c index 175e537b444f..d467f3b11c96 100644 --- a/drivers/mtd/maps/sun_uflash.c +++ b/drivers/mtd/maps/sun_uflash.c @@ -108,7 +108,7 @@ int uflash_devinit(struct platform_device *op, struct device_node *dp) return 0; } -static int __devinit uflash_probe(struct platform_device *op) +static int uflash_probe(struct platform_device *op) { struct device_node *dp = op->dev.of_node; @@ -121,7 +121,7 @@ static int __devinit uflash_probe(struct platform_device *op) return uflash_devinit(op, dp); } -static int __devexit uflash_remove(struct platform_device *op) +static int uflash_remove(struct platform_device *op) { struct uflash_dev *up = dev_get_drvdata(&op->dev); @@ -155,7 +155,7 @@ static struct platform_driver uflash_driver = { .of_match_table = uflash_match, }, .probe = uflash_probe, - .remove = __devexit_p(uflash_remove), + .remove = uflash_remove, }; module_platform_driver(uflash_driver); diff --git a/drivers/mtd/maps/vmu-flash.c b/drivers/mtd/maps/vmu-flash.c index 2e2b0945edc7..6b223cfe92b7 100644 --- a/drivers/mtd/maps/vmu-flash.c +++ b/drivers/mtd/maps/vmu-flash.c @@ -596,7 +596,7 @@ fail_name: } /* Handles very basic info about the flash, queries for details */ -static int __devinit vmu_connect(struct maple_device *mdev) +static int vmu_connect(struct maple_device *mdev) { unsigned long test_flash_data, basic_flash_data; int c, error; @@ -690,7 +690,7 @@ fail_nomem: return error; } -static void __devexit vmu_disconnect(struct maple_device *mdev) +static void vmu_disconnect(struct maple_device *mdev) { struct memcard *card; struct mdev_part *mpart; @@ -772,7 +772,7 @@ static void vmu_file_error(struct maple_device *mdev, void *recvbuf) } -static int __devinit probe_maple_vmu(struct device *dev) +static int probe_maple_vmu(struct device *dev) { int error; struct maple_device *mdev = to_maple_dev(dev); @@ -789,7 +789,7 @@ static int __devinit probe_maple_vmu(struct device *dev) return 0; } -static int __devexit remove_maple_vmu(struct device *dev) +static int remove_maple_vmu(struct device *dev) { struct maple_device *mdev = to_maple_dev(dev); @@ -802,7 +802,7 @@ static struct maple_driver vmu_flash_driver = { .drv = { .name = "Dreamcast_visual_memory", .probe = probe_maple_vmu, - .remove = __devexit_p(remove_maple_vmu), + .remove = remove_maple_vmu, }, }; diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c index f1f06715d4e0..5ad39bb5ab4c 100644 --- a/drivers/mtd/mtd_blkdevs.c +++ b/drivers/mtd/mtd_blkdevs.c @@ -32,7 +32,6 @@ #include <linux/hdreg.h> #include <linux/init.h> #include <linux/mutex.h> -#include <linux/kthread.h> #include <asm/uaccess.h> #include "mtdcore.h" @@ -121,16 +120,14 @@ static int do_blktrans_request(struct mtd_blktrans_ops *tr, int mtd_blktrans_cease_background(struct mtd_blktrans_dev *dev) { - if (kthread_should_stop()) - return 1; - return dev->bg_stop; } EXPORT_SYMBOL_GPL(mtd_blktrans_cease_background); -static int mtd_blktrans_thread(void *arg) +static void mtd_blktrans_work(struct work_struct *work) { - struct mtd_blktrans_dev *dev = arg; + struct mtd_blktrans_dev *dev = + container_of(work, struct mtd_blktrans_dev, work); struct mtd_blktrans_ops *tr = dev->tr; struct request_queue *rq = dev->rq; struct request *req = NULL; @@ -138,7 +135,7 @@ static int mtd_blktrans_thread(void *arg) spin_lock_irq(rq->queue_lock); - while (!kthread_should_stop()) { + while (1) { int res; dev->bg_stop = false; @@ -156,15 +153,7 @@ static int mtd_blktrans_thread(void *arg) background_done = !dev->bg_stop; continue; } - set_current_state(TASK_INTERRUPTIBLE); - - if (kthread_should_stop()) - set_current_state(TASK_RUNNING); - - spin_unlock_irq(rq->queue_lock); - schedule(); - spin_lock_irq(rq->queue_lock); - continue; + break; } spin_unlock_irq(rq->queue_lock); @@ -185,8 +174,6 @@ static int mtd_blktrans_thread(void *arg) __blk_end_request_all(req, -EIO); spin_unlock_irq(rq->queue_lock); - - return 0; } static void mtd_blktrans_request(struct request_queue *rq) @@ -199,10 +186,8 @@ static void mtd_blktrans_request(struct request_queue *rq) if (!dev) while ((req = blk_fetch_request(rq)) != NULL) __blk_end_request_all(req, -ENODEV); - else { - dev->bg_stop = true; - wake_up_process(dev->thread); - } + else + queue_work(dev->wq, &dev->work); } static int blktrans_open(struct block_device *bdev, fmode_t mode) @@ -325,7 +310,7 @@ unlock: return ret; } -static const struct block_device_operations mtd_blktrans_ops = { +static const struct block_device_operations mtd_block_ops = { .owner = THIS_MODULE, .open = blktrans_open, .release = blktrans_release, @@ -401,7 +386,7 @@ int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new) gd->private_data = new; gd->major = tr->major; gd->first_minor = (new->devnum) << tr->part_bits; - gd->fops = &mtd_blktrans_ops; + gd->fops = &mtd_block_ops; if (tr->part_bits) if (new->devnum < 26) @@ -437,14 +422,13 @@ int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new) gd->queue = new->rq; - /* Create processing thread */ - /* TODO: workqueue ? */ - new->thread = kthread_run(mtd_blktrans_thread, new, - "%s%d", tr->name, new->mtd->index); - if (IS_ERR(new->thread)) { - ret = PTR_ERR(new->thread); + /* Create processing workqueue */ + new->wq = alloc_workqueue("%s%d", 0, 0, + tr->name, new->mtd->index); + if (!new->wq) goto error4; - } + INIT_WORK(&new->work, mtd_blktrans_work); + gd->driverfs_dev = &new->mtd->dev; if (new->readonly) @@ -484,9 +468,8 @@ int del_mtd_blktrans_dev(struct mtd_blktrans_dev *old) /* Stop new requests to arrive */ del_gendisk(old->disk); - - /* Stop the thread */ - kthread_stop(old->thread); + /* Stop workqueue. This will perform any pending request. */ + destroy_workqueue(old->wq); /* Kill current requests */ spin_lock_irqsave(&old->queue_lock, flags); diff --git a/drivers/mtd/mtdoops.c b/drivers/mtd/mtdoops.c index f5b3f91fa1cc..97bb8f6304d4 100644 --- a/drivers/mtd/mtdoops.c +++ b/drivers/mtd/mtdoops.c @@ -271,7 +271,7 @@ static void find_next_position(struct mtdoops_context *cxt) if (count[0] == 0xffffffff && count[1] == 0xffffffff) mark_page_unused(cxt, page); - if (count[0] == 0xffffffff) + if (count[0] == 0xffffffff || count[1] != MTDOOPS_KERNMSG_MAGIC) continue; if (maxcount == 0xffffffff) { maxcount = count[0]; @@ -289,14 +289,13 @@ static void find_next_position(struct mtdoops_context *cxt) } } if (maxcount == 0xffffffff) { - cxt->nextpage = 0; - cxt->nextcount = 1; - schedule_work(&cxt->work_erase); - return; + cxt->nextpage = cxt->oops_pages - 1; + cxt->nextcount = 0; + } + else { + cxt->nextpage = maxpos; + cxt->nextcount = maxcount; } - - cxt->nextpage = maxpos; - cxt->nextcount = maxcount; mtdoops_inc_counter(cxt); } diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig index dae191b3c081..5819eb575210 100644 --- a/drivers/mtd/nand/Kconfig +++ b/drivers/mtd/nand/Kconfig @@ -50,16 +50,30 @@ config MTD_NAND_MUSEUM_IDS of these chips were reused by later, larger chips. config MTD_NAND_DENALI - depends on PCI + tristate "Support Denali NAND controller" + help + Enable support for the Denali NAND controller. This should be + combined with either the PCI or platform drivers to provide device + registration. + +config MTD_NAND_DENALI_PCI tristate "Support Denali NAND controller on Intel Moorestown" + depends on PCI && MTD_NAND_DENALI help Enable the driver for NAND flash on Intel Moorestown, using the Denali NAND controller core. - + +config MTD_NAND_DENALI_DT + tristate "Support Denali NAND controller as a DT device" + depends on HAVE_CLK && MTD_NAND_DENALI + help + Enable the driver for NAND flash on platforms using a Denali NAND + controller as a DT device. + config MTD_NAND_DENALI_SCRATCH_REG_ADDR hex "Denali NAND size scratch register address" default "0xFF108018" - depends on MTD_NAND_DENALI + depends on MTD_NAND_DENALI_PCI help Some platforms place the NAND chip size in a scratch register because (some versions of) the driver aren't able to automatically @@ -433,6 +447,14 @@ config MTD_NAND_GPMI_NAND block, such as SD card. So pay attention to it when you enable the GPMI. +config MTD_NAND_BCM47XXNFLASH + tristate "Support for NAND flash on BCM4706 BCMA bus" + depends on BCMA_NFLASH + help + BCMA bus can have various flash memories attached, they are + registered by bcma as platform devices. This enables driver for + NAND flash memories. For now only BCM4706 is supported. + config MTD_NAND_PLATFORM tristate "Support for generic platform NAND driver" depends on HAS_IOMEM @@ -499,12 +521,6 @@ config MTD_NAND_MXC This enables the driver for the NAND flash controller on the MXC processors. -config MTD_NAND_NOMADIK - tristate "ST Nomadik 8815 NAND support" - depends on ARCH_NOMADIK - help - Driver for the NAND flash controller on the Nomadik, with ECC. - config MTD_NAND_SH_FLCTL tristate "Support for NAND on Renesas SuperH FLCTL" depends on SUPERH || ARCH_SHMOBILE diff --git a/drivers/mtd/nand/Makefile b/drivers/mtd/nand/Makefile index 6c7f2b3ca8ae..d76d91205691 100644 --- a/drivers/mtd/nand/Makefile +++ b/drivers/mtd/nand/Makefile @@ -11,6 +11,8 @@ obj-$(CONFIG_MTD_SM_COMMON) += sm_common.o obj-$(CONFIG_MTD_NAND_CAFE) += cafe_nand.o obj-$(CONFIG_MTD_NAND_AMS_DELTA) += ams-delta.o obj-$(CONFIG_MTD_NAND_DENALI) += denali.o +obj-$(CONFIG_MTD_NAND_DENALI_PCI) += denali_pci.o +obj-$(CONFIG_MTD_NAND_DENALI_DT) += denali_dt.o obj-$(CONFIG_MTD_NAND_AU1550) += au1550nd.o obj-$(CONFIG_MTD_NAND_BF5XX) += bf5xx_nand.o obj-$(CONFIG_MTD_NAND_PPCHAMELEONEVB) += ppchameleonevb.o @@ -45,11 +47,11 @@ obj-$(CONFIG_MTD_NAND_MXC) += mxc_nand.o obj-$(CONFIG_MTD_NAND_SOCRATES) += socrates_nand.o obj-$(CONFIG_MTD_NAND_TXX9NDFMC) += txx9ndfmc.o obj-$(CONFIG_MTD_NAND_NUC900) += nuc900_nand.o -obj-$(CONFIG_MTD_NAND_NOMADIK) += nomadik_nand.o obj-$(CONFIG_MTD_NAND_MPC5121_NFC) += mpc5121_nfc.o obj-$(CONFIG_MTD_NAND_RICOH) += r852.o obj-$(CONFIG_MTD_NAND_JZ4740) += jz4740_nand.o obj-$(CONFIG_MTD_NAND_GPMI_NAND) += gpmi-nand/ obj-$(CONFIG_MTD_NAND_XWAY) += xway_nand.o +obj-$(CONFIG_MTD_NAND_BCM47XXNFLASH) += bcm47xxnflash/ nand-objs := nand_base.o nand_bbt.o diff --git a/drivers/mtd/nand/ams-delta.c b/drivers/mtd/nand/ams-delta.c index 9e7723aa7acc..f1d71cdc8aac 100644 --- a/drivers/mtd/nand/ams-delta.c +++ b/drivers/mtd/nand/ams-delta.c @@ -173,7 +173,7 @@ static const struct gpio _mandatory_gpio[] = { /* * Main initialization routine */ -static int __devinit ams_delta_init(struct platform_device *pdev) +static int ams_delta_init(struct platform_device *pdev) { struct nand_chip *this; struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0); @@ -270,7 +270,7 @@ out_free: /* * Clean up routine */ -static int __devexit ams_delta_cleanup(struct platform_device *pdev) +static int ams_delta_cleanup(struct platform_device *pdev) { void __iomem *io_base = platform_get_drvdata(pdev); @@ -289,7 +289,7 @@ static int __devexit ams_delta_cleanup(struct platform_device *pdev) static struct platform_driver ams_delta_nand_driver = { .probe = ams_delta_init, - .remove = __devexit_p(ams_delta_cleanup), + .remove = ams_delta_cleanup, .driver = { .name = "ams-delta-nand", .owner = THIS_MODULE, diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c index 92623ac2015a..90bdca61c797 100644 --- a/drivers/mtd/nand/atmel_nand.c +++ b/drivers/mtd/nand/atmel_nand.c @@ -331,13 +331,13 @@ static void atmel_write_buf(struct mtd_info *mtd, const u8 *buf, int len) * 12-bits 20-bytes 21-bytes * 24-bits 39-bytes 42-bytes */ -static int __devinit pmecc_get_ecc_bytes(int cap, int sector_size) +static int pmecc_get_ecc_bytes(int cap, int sector_size) { int m = 12 + sector_size / 512; return (m * cap + 7) / 8; } -static void __devinit pmecc_config_ecc_layout(struct nand_ecclayout *layout, +static void pmecc_config_ecc_layout(struct nand_ecclayout *layout, int oobsize, int ecc_len) { int i; @@ -353,7 +353,7 @@ static void __devinit pmecc_config_ecc_layout(struct nand_ecclayout *layout, oobsize - ecc_len - layout->oobfree[0].offset; } -static void __devinit __iomem *pmecc_get_alpha_to(struct atmel_nand_host *host) +static void __iomem *pmecc_get_alpha_to(struct atmel_nand_host *host) { int table_size; @@ -375,7 +375,7 @@ static void pmecc_data_free(struct atmel_nand_host *host) kfree(host->pmecc_delta); } -static int __devinit pmecc_data_alloc(struct atmel_nand_host *host) +static int pmecc_data_alloc(struct atmel_nand_host *host) { const int cap = host->pmecc_corr_cap; @@ -724,6 +724,7 @@ static int pmecc_correction(struct mtd_info *mtd, u32 pmecc_stat, uint8_t *buf, struct atmel_nand_host *host = nand_chip->priv; int i, err_nbr, eccbytes; uint8_t *buf_pos; + int total_err = 0; eccbytes = nand_chip->ecc.bytes; for (i = 0; i < eccbytes; i++) @@ -751,12 +752,13 @@ normal_check: pmecc_correct_data(mtd, buf_pos, ecc, i, host->pmecc_bytes_per_sector, err_nbr); mtd->ecc_stats.corrected += err_nbr; + total_err += err_nbr; } } pmecc_stat >>= 1; } - return 0; + return total_err; } static int atmel_nand_pmecc_read_page(struct mtd_info *mtd, @@ -768,6 +770,7 @@ static int atmel_nand_pmecc_read_page(struct mtd_info *mtd, uint32_t *eccpos = chip->ecc.layout->eccpos; uint32_t stat; unsigned long end_time; + int bitflips = 0; pmecc_writel(host->ecc, CTRL, PMECC_CTRL_RST); pmecc_writel(host->ecc, CTRL, PMECC_CTRL_DISABLE); @@ -790,11 +793,14 @@ static int atmel_nand_pmecc_read_page(struct mtd_info *mtd, } stat = pmecc_readl_relaxed(host->ecc, ISR); - if (stat != 0) - if (pmecc_correction(mtd, stat, buf, &oob[eccpos[0]]) != 0) - return -EIO; + if (stat != 0) { + bitflips = pmecc_correction(mtd, stat, buf, &oob[eccpos[0]]); + if (bitflips < 0) + /* uncorrectable errors */ + return 0; + } - return 0; + return bitflips; } static int atmel_nand_pmecc_write_page(struct mtd_info *mtd, @@ -1206,7 +1212,7 @@ static void atmel_nand_hwctl(struct mtd_info *mtd, int mode) } #if defined(CONFIG_OF) -static int __devinit atmel_of_init_port(struct atmel_nand_host *host, +static int atmel_of_init_port(struct atmel_nand_host *host, struct device_node *np) { u32 val, table_offset; @@ -1293,7 +1299,7 @@ static int __devinit atmel_of_init_port(struct atmel_nand_host *host, return 0; } #else -static int __devinit atmel_of_init_port(struct atmel_nand_host *host, +static int atmel_of_init_port(struct atmel_nand_host *host, struct device_node *np) { return -EINVAL; diff --git a/drivers/mtd/nand/au1550nd.c b/drivers/mtd/nand/au1550nd.c index 5c47b200045a..217459d02b2f 100644 --- a/drivers/mtd/nand/au1550nd.c +++ b/drivers/mtd/nand/au1550nd.c @@ -382,7 +382,7 @@ static void au1550_command(struct mtd_info *mtd, unsigned command, int column, i while(!this->dev_ready(mtd)); } -static int __devinit find_nand_cs(unsigned long nand_base) +static int find_nand_cs(unsigned long nand_base) { void __iomem *base = (void __iomem *)KSEG1ADDR(AU1000_STATIC_MEM_PHYS_ADDR); @@ -403,7 +403,7 @@ static int __devinit find_nand_cs(unsigned long nand_base) return -ENODEV; } -static int __devinit au1550nd_probe(struct platform_device *pdev) +static int au1550nd_probe(struct platform_device *pdev) { struct au1550nd_platdata *pd; struct au1550nd_ctx *ctx; @@ -491,7 +491,7 @@ out1: return ret; } -static int __devexit au1550nd_remove(struct platform_device *pdev) +static int au1550nd_remove(struct platform_device *pdev) { struct au1550nd_ctx *ctx = platform_get_drvdata(pdev); struct resource *r = platform_get_resource(pdev, IORESOURCE_MEM, 0); @@ -509,7 +509,7 @@ static struct platform_driver au1550nd_driver = { .owner = THIS_MODULE, }, .probe = au1550nd_probe, - .remove = __devexit_p(au1550nd_remove), + .remove = au1550nd_remove, }; module_platform_driver(au1550nd_driver); diff --git a/drivers/mtd/nand/bcm47xxnflash/Makefile b/drivers/mtd/nand/bcm47xxnflash/Makefile new file mode 100644 index 000000000000..f05b119e134b --- /dev/null +++ b/drivers/mtd/nand/bcm47xxnflash/Makefile @@ -0,0 +1,4 @@ +bcm47xxnflash-y += main.o +bcm47xxnflash-y += ops_bcm4706.o + +obj-$(CONFIG_MTD_NAND_BCM47XXNFLASH) += bcm47xxnflash.o diff --git a/drivers/mtd/nand/bcm47xxnflash/bcm47xxnflash.h b/drivers/mtd/nand/bcm47xxnflash/bcm47xxnflash.h new file mode 100644 index 000000000000..0bdb2ce4da75 --- /dev/null +++ b/drivers/mtd/nand/bcm47xxnflash/bcm47xxnflash.h @@ -0,0 +1,22 @@ +#ifndef __BCM47XXNFLASH_H +#define __BCM47XXNFLASH_H + +#include <linux/mtd/mtd.h> +#include <linux/mtd/nand.h> + +struct bcm47xxnflash { + struct bcma_drv_cc *cc; + + struct nand_chip nand_chip; + struct mtd_info mtd; + + unsigned curr_command; + int curr_page_addr; + int curr_column; + + u8 id_data[8]; +}; + +int bcm47xxnflash_ops_bcm4706_init(struct bcm47xxnflash *b47n); + +#endif /* BCM47XXNFLASH */ diff --git a/drivers/mtd/nand/bcm47xxnflash/main.c b/drivers/mtd/nand/bcm47xxnflash/main.c new file mode 100644 index 000000000000..2b8b05bec3dd --- /dev/null +++ b/drivers/mtd/nand/bcm47xxnflash/main.c @@ -0,0 +1,108 @@ +/* + * BCM47XX NAND flash driver + * + * Copyright (C) 2012 RafaÅ‚ MiÅ‚ecki <zajec5@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/platform_device.h> +#include <linux/bcma/bcma.h> + +#include "bcm47xxnflash.h" + +MODULE_DESCRIPTION("NAND flash driver for BCMA bus"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("RafaÅ‚ MiÅ‚ecki"); + +static const char *probes[] = { "bcm47xxpart", NULL }; + +static int bcm47xxnflash_probe(struct platform_device *pdev) +{ + struct bcma_nflash *nflash = dev_get_platdata(&pdev->dev); + struct bcm47xxnflash *b47n; + int err = 0; + + b47n = kzalloc(sizeof(*b47n), GFP_KERNEL); + if (!b47n) { + err = -ENOMEM; + goto out; + } + + b47n->nand_chip.priv = b47n; + b47n->mtd.owner = THIS_MODULE; + b47n->mtd.priv = &b47n->nand_chip; /* Required */ + b47n->cc = container_of(nflash, struct bcma_drv_cc, nflash); + + if (b47n->cc->core->bus->chipinfo.id == BCMA_CHIP_ID_BCM4706) { + err = bcm47xxnflash_ops_bcm4706_init(b47n); + } else { + pr_err("Device not supported\n"); + err = -ENOTSUPP; + } + if (err) { + pr_err("Initialization failed: %d\n", err); + goto err_init; + } + + err = mtd_device_parse_register(&b47n->mtd, probes, NULL, NULL, 0); + if (err) { + pr_err("Failed to register MTD device: %d\n", err); + goto err_dev_reg; + } + + return 0; + +err_dev_reg: +err_init: + kfree(b47n); +out: + return err; +} + +static int __devexit bcm47xxnflash_remove(struct platform_device *pdev) +{ + struct bcma_nflash *nflash = dev_get_platdata(&pdev->dev); + + if (nflash->mtd) + mtd_device_unregister(nflash->mtd); + + return 0; +} + +static struct platform_driver bcm47xxnflash_driver = { + .remove = __devexit_p(bcm47xxnflash_remove), + .driver = { + .name = "bcma_nflash", + .owner = THIS_MODULE, + }, +}; + +static int __init bcm47xxnflash_init(void) +{ + int err; + + /* + * Platform device "bcma_nflash" exists on SoCs and is registered very + * early, it won't be added during runtime (use platform_driver_probe). + */ + err = platform_driver_probe(&bcm47xxnflash_driver, bcm47xxnflash_probe); + if (err) + pr_err("Failed to register serial flash driver: %d\n", err); + + return err; +} + +static void __exit bcm47xxnflash_exit(void) +{ + platform_driver_unregister(&bcm47xxnflash_driver); +} + +module_init(bcm47xxnflash_init); +module_exit(bcm47xxnflash_exit); diff --git a/drivers/mtd/nand/bcm47xxnflash/ops_bcm4706.c b/drivers/mtd/nand/bcm47xxnflash/ops_bcm4706.c new file mode 100644 index 000000000000..86c9a79b89b3 --- /dev/null +++ b/drivers/mtd/nand/bcm47xxnflash/ops_bcm4706.c @@ -0,0 +1,413 @@ +/* + * BCM47XX NAND flash driver + * + * Copyright (C) 2012 RafaÅ‚ MiÅ‚ecki <zajec5@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/bcma/bcma.h> + +#include "bcm47xxnflash.h" + +/* Broadcom uses 1'000'000 but it seems to be too many. Tests on WNDR4500 has + * shown 164 retries as maxiumum. */ +#define NFLASH_READY_RETRIES 1000 + +#define NFLASH_SECTOR_SIZE 512 + +#define NCTL_CMD0 0x00010000 +#define NCTL_CMD1W 0x00080000 +#define NCTL_READ 0x00100000 +#define NCTL_WRITE 0x00200000 +#define NCTL_SPECADDR 0x01000000 +#define NCTL_READY 0x04000000 +#define NCTL_ERR 0x08000000 +#define NCTL_CSA 0x40000000 +#define NCTL_START 0x80000000 + +/************************************************** + * Various helpers + **************************************************/ + +static inline u8 bcm47xxnflash_ops_bcm4706_ns_to_cycle(u16 ns, u16 clock) +{ + return ((ns * 1000 * clock) / 1000000) + 1; +} + +static int bcm47xxnflash_ops_bcm4706_ctl_cmd(struct bcma_drv_cc *cc, u32 code) +{ + int i = 0; + + bcma_cc_write32(cc, BCMA_CC_NFLASH_CTL, NCTL_START | code); + for (i = 0; i < NFLASH_READY_RETRIES; i++) { + if (!(bcma_cc_read32(cc, BCMA_CC_NFLASH_CTL) & NCTL_START)) { + i = 0; + break; + } + } + if (i) { + pr_err("NFLASH control command not ready!\n"); + return -EBUSY; + } + return 0; +} + +static int bcm47xxnflash_ops_bcm4706_poll(struct bcma_drv_cc *cc) +{ + int i; + + for (i = 0; i < NFLASH_READY_RETRIES; i++) { + if (bcma_cc_read32(cc, BCMA_CC_NFLASH_CTL) & NCTL_READY) { + if (bcma_cc_read32(cc, BCMA_CC_NFLASH_CTL) & + BCMA_CC_NFLASH_CTL_ERR) { + pr_err("Error on polling\n"); + return -EBUSY; + } else { + return 0; + } + } + } + + pr_err("Polling timeout!\n"); + return -EBUSY; +} + +/************************************************** + * R/W + **************************************************/ + +static void bcm47xxnflash_ops_bcm4706_read(struct mtd_info *mtd, uint8_t *buf, + int len) +{ + struct nand_chip *nand_chip = (struct nand_chip *)mtd->priv; + struct bcm47xxnflash *b47n = (struct bcm47xxnflash *)nand_chip->priv; + + u32 ctlcode; + u32 *dest = (u32 *)buf; + int i; + int toread; + + BUG_ON(b47n->curr_page_addr & ~nand_chip->pagemask); + /* Don't validate column using nand_chip->page_shift, it may be bigger + * when accessing OOB */ + + while (len) { + /* We can read maximum of 0x200 bytes at once */ + toread = min(len, 0x200); + + /* Set page and column */ + bcma_cc_write32(b47n->cc, BCMA_CC_NFLASH_COL_ADDR, + b47n->curr_column); + bcma_cc_write32(b47n->cc, BCMA_CC_NFLASH_ROW_ADDR, + b47n->curr_page_addr); + + /* Prepare to read */ + ctlcode = NCTL_CSA | NCTL_CMD1W | 0x00040000 | 0x00020000 | + NCTL_CMD0; + ctlcode |= NAND_CMD_READSTART << 8; + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(b47n->cc, ctlcode)) + return; + if (bcm47xxnflash_ops_bcm4706_poll(b47n->cc)) + return; + + /* Eventually read some data :) */ + for (i = 0; i < toread; i += 4, dest++) { + ctlcode = NCTL_CSA | 0x30000000 | NCTL_READ; + if (i == toread - 4) /* Last read goes without that */ + ctlcode &= ~NCTL_CSA; + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(b47n->cc, + ctlcode)) + return; + *dest = bcma_cc_read32(b47n->cc, BCMA_CC_NFLASH_DATA); + } + + b47n->curr_column += toread; + len -= toread; + } +} + +static void bcm47xxnflash_ops_bcm4706_write(struct mtd_info *mtd, + const uint8_t *buf, int len) +{ + struct nand_chip *nand_chip = (struct nand_chip *)mtd->priv; + struct bcm47xxnflash *b47n = (struct bcm47xxnflash *)nand_chip->priv; + struct bcma_drv_cc *cc = b47n->cc; + + u32 ctlcode; + const u32 *data = (u32 *)buf; + int i; + + BUG_ON(b47n->curr_page_addr & ~nand_chip->pagemask); + /* Don't validate column using nand_chip->page_shift, it may be bigger + * when accessing OOB */ + + for (i = 0; i < len; i += 4, data++) { + bcma_cc_write32(cc, BCMA_CC_NFLASH_DATA, *data); + + ctlcode = NCTL_CSA | 0x30000000 | NCTL_WRITE; + if (i == len - 4) /* Last read goes without that */ + ctlcode &= ~NCTL_CSA; + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(cc, ctlcode)) { + pr_err("%s ctl_cmd didn't work!\n", __func__); + return; + } + } + + b47n->curr_column += len; +} + +/************************************************** + * NAND chip ops + **************************************************/ + +/* Default nand_select_chip calls cmd_ctrl, which is not used in BCM4706 */ +static void bcm47xxnflash_ops_bcm4706_select_chip(struct mtd_info *mtd, + int chip) +{ + return; +} + +/* + * Default nand_command and nand_command_lp don't match BCM4706 hardware layout. + * For example, reading chip id is performed in a non-standard way. + * Setting column and page is also handled differently, we use a special + * registers of ChipCommon core. Hacking cmd_ctrl to understand and convert + * standard commands would be much more complicated. + */ +static void bcm47xxnflash_ops_bcm4706_cmdfunc(struct mtd_info *mtd, + unsigned command, int column, + int page_addr) +{ + struct nand_chip *nand_chip = (struct nand_chip *)mtd->priv; + struct bcm47xxnflash *b47n = (struct bcm47xxnflash *)nand_chip->priv; + struct bcma_drv_cc *cc = b47n->cc; + u32 ctlcode; + int i; + + if (column != -1) + b47n->curr_column = column; + if (page_addr != -1) + b47n->curr_page_addr = page_addr; + + switch (command) { + case NAND_CMD_RESET: + pr_warn("Chip reset not implemented yet\n"); + break; + case NAND_CMD_READID: + ctlcode = NCTL_CSA | 0x01000000 | NCTL_CMD1W | NCTL_CMD0; + ctlcode |= NAND_CMD_READID; + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(b47n->cc, ctlcode)) { + pr_err("READID error\n"); + break; + } + + /* + * Reading is specific, last one has to go without NCTL_CSA + * bit. We don't know how many reads NAND subsystem is going + * to perform, so cache everything. + */ + for (i = 0; i < ARRAY_SIZE(b47n->id_data); i++) { + ctlcode = NCTL_CSA | NCTL_READ; + if (i == ARRAY_SIZE(b47n->id_data) - 1) + ctlcode &= ~NCTL_CSA; + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(b47n->cc, + ctlcode)) { + pr_err("READID error\n"); + break; + } + b47n->id_data[i] = + bcma_cc_read32(b47n->cc, BCMA_CC_NFLASH_DATA) + & 0xFF; + } + + break; + case NAND_CMD_STATUS: + ctlcode = NCTL_CSA | NCTL_CMD0 | NAND_CMD_STATUS; + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(cc, ctlcode)) + pr_err("STATUS command error\n"); + break; + case NAND_CMD_READ0: + break; + case NAND_CMD_READOOB: + if (page_addr != -1) + b47n->curr_column += mtd->writesize; + break; + case NAND_CMD_ERASE1: + bcma_cc_write32(cc, BCMA_CC_NFLASH_ROW_ADDR, + b47n->curr_page_addr); + ctlcode = 0x00040000 | NCTL_CMD1W | NCTL_CMD0 | + NAND_CMD_ERASE1 | (NAND_CMD_ERASE2 << 8); + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(cc, ctlcode)) + pr_err("ERASE1 failed\n"); + break; + case NAND_CMD_ERASE2: + break; + case NAND_CMD_SEQIN: + /* Set page and column */ + bcma_cc_write32(cc, BCMA_CC_NFLASH_COL_ADDR, + b47n->curr_column); + bcma_cc_write32(cc, BCMA_CC_NFLASH_ROW_ADDR, + b47n->curr_page_addr); + + /* Prepare to write */ + ctlcode = 0x40000000 | 0x00040000 | 0x00020000 | 0x00010000; + ctlcode |= NAND_CMD_SEQIN; + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(cc, ctlcode)) + pr_err("SEQIN failed\n"); + break; + case NAND_CMD_PAGEPROG: + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(cc, 0x00010000 | + NAND_CMD_PAGEPROG)) + pr_err("PAGEPROG failed\n"); + if (bcm47xxnflash_ops_bcm4706_poll(cc)) + pr_err("PAGEPROG not ready\n"); + break; + default: + pr_err("Command 0x%X unsupported\n", command); + break; + } + b47n->curr_command = command; +} + +static u8 bcm47xxnflash_ops_bcm4706_read_byte(struct mtd_info *mtd) +{ + struct nand_chip *nand_chip = (struct nand_chip *)mtd->priv; + struct bcm47xxnflash *b47n = (struct bcm47xxnflash *)nand_chip->priv; + struct bcma_drv_cc *cc = b47n->cc; + u32 tmp = 0; + + switch (b47n->curr_command) { + case NAND_CMD_READID: + if (b47n->curr_column >= ARRAY_SIZE(b47n->id_data)) { + pr_err("Requested invalid id_data: %d\n", + b47n->curr_column); + return 0; + } + return b47n->id_data[b47n->curr_column++]; + case NAND_CMD_STATUS: + if (bcm47xxnflash_ops_bcm4706_ctl_cmd(cc, NCTL_READ)) + return 0; + return bcma_cc_read32(cc, BCMA_CC_NFLASH_DATA) & 0xff; + case NAND_CMD_READOOB: + bcm47xxnflash_ops_bcm4706_read(mtd, (u8 *)&tmp, 4); + return tmp & 0xFF; + } + + pr_err("Invalid command for byte read: 0x%X\n", b47n->curr_command); + return 0; +} + +static void bcm47xxnflash_ops_bcm4706_read_buf(struct mtd_info *mtd, + uint8_t *buf, int len) +{ + struct nand_chip *nand_chip = (struct nand_chip *)mtd->priv; + struct bcm47xxnflash *b47n = (struct bcm47xxnflash *)nand_chip->priv; + + switch (b47n->curr_command) { + case NAND_CMD_READ0: + case NAND_CMD_READOOB: + bcm47xxnflash_ops_bcm4706_read(mtd, buf, len); + return; + } + + pr_err("Invalid command for buf read: 0x%X\n", b47n->curr_command); +} + +static void bcm47xxnflash_ops_bcm4706_write_buf(struct mtd_info *mtd, + const uint8_t *buf, int len) +{ + struct nand_chip *nand_chip = (struct nand_chip *)mtd->priv; + struct bcm47xxnflash *b47n = (struct bcm47xxnflash *)nand_chip->priv; + + switch (b47n->curr_command) { + case NAND_CMD_SEQIN: + bcm47xxnflash_ops_bcm4706_write(mtd, buf, len); + return; + } + + pr_err("Invalid command for buf write: 0x%X\n", b47n->curr_command); +} + +/************************************************** + * Init + **************************************************/ + +int bcm47xxnflash_ops_bcm4706_init(struct bcm47xxnflash *b47n) +{ + int err; + u32 freq; + u16 clock; + u8 w0, w1, w2, w3, w4; + + unsigned long chipsize; /* MiB */ + u8 tbits, col_bits, col_size, row_bits, row_bsize; + u32 val; + + b47n->nand_chip.select_chip = bcm47xxnflash_ops_bcm4706_select_chip; + b47n->nand_chip.cmdfunc = bcm47xxnflash_ops_bcm4706_cmdfunc; + b47n->nand_chip.read_byte = bcm47xxnflash_ops_bcm4706_read_byte; + b47n->nand_chip.read_buf = bcm47xxnflash_ops_bcm4706_read_buf; + b47n->nand_chip.write_buf = bcm47xxnflash_ops_bcm4706_write_buf; + b47n->nand_chip.bbt_options = NAND_BBT_USE_FLASH; + b47n->nand_chip.ecc.mode = NAND_ECC_NONE; /* TODO: implement ECC */ + + /* Enable NAND flash access */ + bcma_cc_set32(b47n->cc, BCMA_CC_4706_FLASHSCFG, + BCMA_CC_4706_FLASHSCFG_NF1); + + /* Configure wait counters */ + if (b47n->cc->status & BCMA_CC_CHIPST_4706_PKG_OPTION) { + freq = 100000000; + } else { + freq = bcma_chipco_pll_read(b47n->cc, 4); + freq = (freq * 0xFFF) >> 3; + freq = (freq * 25000000) >> 3; + } + clock = freq / 1000000; + w0 = bcm47xxnflash_ops_bcm4706_ns_to_cycle(15, clock); + w1 = bcm47xxnflash_ops_bcm4706_ns_to_cycle(20, clock); + w2 = bcm47xxnflash_ops_bcm4706_ns_to_cycle(10, clock); + w3 = bcm47xxnflash_ops_bcm4706_ns_to_cycle(10, clock); + w4 = bcm47xxnflash_ops_bcm4706_ns_to_cycle(100, clock); + bcma_cc_write32(b47n->cc, BCMA_CC_NFLASH_WAITCNT0, + (w4 << 24 | w3 << 18 | w2 << 12 | w1 << 6 | w0)); + + /* Scan NAND */ + err = nand_scan(&b47n->mtd, 1); + if (err) { + pr_err("Could not scan NAND flash: %d\n", err); + goto exit; + } + + /* Configure FLASH */ + chipsize = b47n->nand_chip.chipsize >> 20; + tbits = ffs(chipsize); /* find first bit set */ + if (!tbits || tbits != fls(chipsize)) { + pr_err("Invalid flash size: 0x%lX\n", chipsize); + err = -ENOTSUPP; + goto exit; + } + tbits += 19; /* Broadcom increases *index* by 20, we increase *pos* */ + + col_bits = b47n->nand_chip.page_shift + 1; + col_size = (col_bits + 7) / 8; + + row_bits = tbits - col_bits + 1; + row_bsize = (row_bits + 7) / 8; + + val = ((row_bsize - 1) << 6) | ((col_size - 1) << 4) | 2; + bcma_cc_write32(b47n->cc, BCMA_CC_NFLASH_CONF, val); + +exit: + if (err) + bcma_cc_mask32(b47n->cc, BCMA_CC_4706_FLASHSCFG, + ~BCMA_CC_4706_FLASHSCFG_NF1); + return err; +} diff --git a/drivers/mtd/nand/bf5xx_nand.c b/drivers/mtd/nand/bf5xx_nand.c index ab0caa74eb43..4271e948d1e2 100644 --- a/drivers/mtd/nand/bf5xx_nand.c +++ b/drivers/mtd/nand/bf5xx_nand.c @@ -658,7 +658,7 @@ static int bf5xx_nand_hw_init(struct bf5xx_nand_info *info) /* * Device management interface */ -static int __devinit bf5xx_nand_add_partition(struct bf5xx_nand_info *info) +static int bf5xx_nand_add_partition(struct bf5xx_nand_info *info) { struct mtd_info *mtd = &info->mtd; struct mtd_partition *parts = info->platform->partitions; @@ -667,7 +667,7 @@ static int __devinit bf5xx_nand_add_partition(struct bf5xx_nand_info *info) return mtd_device_register(mtd, parts, nr); } -static int __devexit bf5xx_nand_remove(struct platform_device *pdev) +static int bf5xx_nand_remove(struct platform_device *pdev) { struct bf5xx_nand_info *info = to_nand_info(pdev); @@ -725,7 +725,7 @@ static int bf5xx_nand_scan(struct mtd_info *mtd) * it can allocate all necessary resources then calls the * nand layer to look for devices */ -static int __devinit bf5xx_nand_probe(struct platform_device *pdev) +static int bf5xx_nand_probe(struct platform_device *pdev) { struct bf5xx_nand_platform *plat = to_nand_plat(pdev); struct bf5xx_nand_info *info = NULL; @@ -865,7 +865,7 @@ static int bf5xx_nand_resume(struct platform_device *dev) /* driver device registration */ static struct platform_driver bf5xx_nand_driver = { .probe = bf5xx_nand_probe, - .remove = __devexit_p(bf5xx_nand_remove), + .remove = bf5xx_nand_remove, .suspend = bf5xx_nand_suspend, .resume = bf5xx_nand_resume, .driver = { diff --git a/drivers/mtd/nand/cafe_nand.c b/drivers/mtd/nand/cafe_nand.c index 2bb7170502c2..010d61266536 100644 --- a/drivers/mtd/nand/cafe_nand.c +++ b/drivers/mtd/nand/cafe_nand.c @@ -585,7 +585,7 @@ static int cafe_nand_block_bad(struct mtd_info *mtd, loff_t ofs, int getchip) } /* F_2[X]/(X**6+X+1) */ -static unsigned short __devinit gf64_mul(u8 a, u8 b) +static unsigned short gf64_mul(u8 a, u8 b) { u8 c; unsigned int i; @@ -604,7 +604,7 @@ static unsigned short __devinit gf64_mul(u8 a, u8 b) } /* F_64[X]/(X**2+X+A**-1) with A the generator of F_64[X] */ -static u16 __devinit gf4096_mul(u16 a, u16 b) +static u16 gf4096_mul(u16 a, u16 b) { u8 ah, al, bh, bl, ch, cl; @@ -619,14 +619,14 @@ static u16 __devinit gf4096_mul(u16 a, u16 b) return (ch << 6) ^ cl; } -static int __devinit cafe_mul(int x) +static int cafe_mul(int x) { if (x == 0) return 1; return gf4096_mul(x, 0xe01); } -static int __devinit cafe_nand_probe(struct pci_dev *pdev, +static int cafe_nand_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { struct mtd_info *mtd; @@ -821,7 +821,7 @@ static int __devinit cafe_nand_probe(struct pci_dev *pdev, return err; } -static void __devexit cafe_nand_remove(struct pci_dev *pdev) +static void cafe_nand_remove(struct pci_dev *pdev) { struct mtd_info *mtd = pci_get_drvdata(pdev); struct cafe_priv *cafe = mtd->priv; @@ -887,7 +887,7 @@ static struct pci_driver cafe_nand_pci_driver = { .name = "CAFÉ NAND", .id_table = cafe_nand_tbl, .probe = cafe_nand_probe, - .remove = __devexit_p(cafe_nand_remove), + .remove = cafe_nand_remove, .resume = cafe_nand_resume, }; diff --git a/drivers/mtd/nand/cs553x_nand.c b/drivers/mtd/nand/cs553x_nand.c index adb6c3ef37fb..2cdeab8bebc4 100644 --- a/drivers/mtd/nand/cs553x_nand.c +++ b/drivers/mtd/nand/cs553x_nand.c @@ -237,6 +237,7 @@ static int __init cs553x_init_one(int cs, int mmio, unsigned long adr) this->ecc.hwctl = cs_enable_hwecc; this->ecc.calculate = cs_calculate_ecc; this->ecc.correct = nand_correct_data; + this->ecc.strength = 1; /* Enable the following for a flash based bad block table */ this->bbt_options = NAND_BBT_USE_FLASH; @@ -247,8 +248,6 @@ static int __init cs553x_init_one(int cs, int mmio, unsigned long adr) goto out_ior; } - this->ecc.strength = 1; - new_mtd->name = kasprintf(GFP_KERNEL, "cs553x_nand_cs%d", cs); cs553x_mtd[cs] = new_mtd; diff --git a/drivers/mtd/nand/davinci_nand.c b/drivers/mtd/nand/davinci_nand.c index 945047ad0952..3502606f6480 100644 --- a/drivers/mtd/nand/davinci_nand.c +++ b/drivers/mtd/nand/davinci_nand.c @@ -821,9 +821,16 @@ syndrome_done: if (ret < 0) goto err_scan; - ret = mtd_device_parse_register(&info->mtd, NULL, NULL, pdata->parts, - pdata->nr_parts); - + if (pdata->parts) + ret = mtd_device_parse_register(&info->mtd, NULL, NULL, + pdata->parts, pdata->nr_parts); + else { + struct mtd_part_parser_data ppdata; + + ppdata.of_node = pdev->dev.of_node; + ret = mtd_device_parse_register(&info->mtd, NULL, &ppdata, + NULL, 0); + } if (ret < 0) goto err_scan; diff --git a/drivers/mtd/nand/denali.c b/drivers/mtd/nand/denali.c index e706a237170f..0c8bb6bf8424 100644 --- a/drivers/mtd/nand/denali.c +++ b/drivers/mtd/nand/denali.c @@ -16,14 +16,12 @@ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. * */ - #include <linux/interrupt.h> #include <linux/delay.h> #include <linux/dma-mapping.h> #include <linux/wait.h> #include <linux/mutex.h> #include <linux/slab.h> -#include <linux/pci.h> #include <linux/mtd/mtd.h> #include <linux/module.h> @@ -89,13 +87,6 @@ MODULE_PARM_DESC(onfi_timing_mode, "Overrides default ONFI setting." * format the bank into the proper bits for the controller */ #define BANK(x) ((x) << 24) -/* List of platforms this NAND controller has be integrated into */ -static const struct pci_device_id denali_pci_ids[] = { - { PCI_VDEVICE(INTEL, 0x0701), INTEL_CE4100 }, - { PCI_VDEVICE(INTEL, 0x0809), INTEL_MRST }, - { /* end: all zeroes */ } -}; - /* forward declarations */ static void clear_interrupts(struct denali_nand_info *denali); static uint32_t wait_for_irq(struct denali_nand_info *denali, @@ -699,7 +690,7 @@ static uint32_t wait_for_irq(struct denali_nand_info *denali, uint32_t irq_mask) if (comp_res == 0) { /* timeout */ - printk(KERN_ERR "timeout occurred, status = 0x%x, mask = 0x%x\n", + pr_err("timeout occurred, status = 0x%x, mask = 0x%x\n", intr_status, irq_mask); intr_status = 0; @@ -1305,8 +1296,7 @@ static void denali_cmdfunc(struct mtd_info *mtd, unsigned int cmd, int col, /* TODO: Read OOB data */ break; default: - printk(KERN_ERR ": unsupported command" - " received 0x%x\n", cmd); + pr_err(": unsupported command received 0x%x\n", cmd); break; } } @@ -1425,107 +1415,48 @@ void denali_drv_init(struct denali_nand_info *denali) denali->irq_status = 0; } -/* driver entry point */ -static int denali_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) +int denali_init(struct denali_nand_info *denali) { - int ret = -ENODEV; - resource_size_t csr_base, mem_base; - unsigned long csr_len, mem_len; - struct denali_nand_info *denali; - - denali = kzalloc(sizeof(*denali), GFP_KERNEL); - if (!denali) - return -ENOMEM; + int ret; - ret = pci_enable_device(dev); - if (ret) { - printk(KERN_ERR "Spectra: pci_enable_device failed.\n"); - goto failed_alloc_memery; - } - - if (id->driver_data == INTEL_CE4100) { + if (denali->platform == INTEL_CE4100) { /* Due to a silicon limitation, we can only support * ONFI timing mode 1 and below. */ if (onfi_timing_mode < -1 || onfi_timing_mode > 1) { - printk(KERN_ERR "Intel CE4100 only supports" - " ONFI timing mode 1 or below\n"); - ret = -EINVAL; - goto failed_enable_dev; - } - denali->platform = INTEL_CE4100; - mem_base = pci_resource_start(dev, 0); - mem_len = pci_resource_len(dev, 1); - csr_base = pci_resource_start(dev, 1); - csr_len = pci_resource_len(dev, 1); - } else { - denali->platform = INTEL_MRST; - csr_base = pci_resource_start(dev, 0); - csr_len = pci_resource_len(dev, 0); - mem_base = pci_resource_start(dev, 1); - mem_len = pci_resource_len(dev, 1); - if (!mem_len) { - mem_base = csr_base + csr_len; - mem_len = csr_len; + pr_err("Intel CE4100 only supports ONFI timing mode 1 or below\n"); + return -EINVAL; } } /* Is 32-bit DMA supported? */ - ret = dma_set_mask(&dev->dev, DMA_BIT_MASK(32)); + ret = dma_set_mask(denali->dev, DMA_BIT_MASK(32)); if (ret) { - printk(KERN_ERR "Spectra: no usable DMA configuration\n"); - goto failed_enable_dev; + pr_err("Spectra: no usable DMA configuration\n"); + return ret; } - denali->buf.dma_buf = dma_map_single(&dev->dev, denali->buf.buf, + denali->buf.dma_buf = dma_map_single(denali->dev, denali->buf.buf, DENALI_BUF_SIZE, DMA_BIDIRECTIONAL); - if (dma_mapping_error(&dev->dev, denali->buf.dma_buf)) { - dev_err(&dev->dev, "Spectra: failed to map DMA buffer\n"); - goto failed_enable_dev; - } - - pci_set_master(dev); - denali->dev = &dev->dev; - denali->mtd.dev.parent = &dev->dev; - - ret = pci_request_regions(dev, DENALI_NAND_NAME); - if (ret) { - printk(KERN_ERR "Spectra: Unable to request memory regions\n"); - goto failed_dma_map; - } - - denali->flash_reg = ioremap_nocache(csr_base, csr_len); - if (!denali->flash_reg) { - printk(KERN_ERR "Spectra: Unable to remap memory region\n"); - ret = -ENOMEM; - goto failed_req_regions; - } - - denali->flash_mem = ioremap_nocache(mem_base, mem_len); - if (!denali->flash_mem) { - printk(KERN_ERR "Spectra: ioremap_nocache failed!"); - ret = -ENOMEM; - goto failed_remap_reg; + if (dma_mapping_error(denali->dev, denali->buf.dma_buf)) { + dev_err(denali->dev, "Spectra: failed to map DMA buffer\n"); + return -EIO; } - + denali->mtd.dev.parent = denali->dev; denali_hw_init(denali); denali_drv_init(denali); /* denali_isr register is done after all the hardware * initilization is finished*/ - if (request_irq(dev->irq, denali_isr, IRQF_SHARED, + if (request_irq(denali->irq, denali_isr, IRQF_SHARED, DENALI_NAND_NAME, denali)) { - printk(KERN_ERR "Spectra: Unable to allocate IRQ\n"); - ret = -ENODEV; - goto failed_remap_mem; + pr_err("Spectra: Unable to allocate IRQ\n"); + return -ENODEV; } /* now that our ISR is registered, we can enable interrupts */ denali_set_intr_modes(denali, true); - - pci_set_drvdata(dev, denali); - denali->mtd.name = "denali-nand"; denali->mtd.owner = THIS_MODULE; denali->mtd.priv = &denali->nand; @@ -1549,8 +1480,7 @@ static int denali_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) */ if (denali->mtd.writesize > NAND_MAX_PAGESIZE + NAND_MAX_OOBSIZE) { ret = -ENODEV; - printk(KERN_ERR "Spectra: device size not supported by this " - "version of MTD."); + pr_err("Spectra: device size not supported by this version of MTD."); goto failed_req_irq; } @@ -1602,8 +1532,8 @@ static int denali_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) } else if (denali->mtd.oobsize < (denali->bbtskipbytes + ECC_8BITS * (denali->mtd.writesize / ECC_SECTOR_SIZE))) { - printk(KERN_ERR "Your NAND chip OOB is not large enough to" - " contain 8bit ECC correction codes"); + pr_err("Your NAND chip OOB is not large enough to \ + contain 8bit ECC correction codes"); goto failed_req_irq; } else { denali->nand.ecc.strength = 8; @@ -1655,56 +1585,24 @@ static int denali_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) ret = mtd_device_register(&denali->mtd, NULL, 0); if (ret) { - dev_err(&dev->dev, "Spectra: Failed to register MTD: %d\n", + dev_err(denali->dev, "Spectra: Failed to register MTD: %d\n", ret); goto failed_req_irq; } return 0; failed_req_irq: - denali_irq_cleanup(dev->irq, denali); -failed_remap_mem: - iounmap(denali->flash_mem); -failed_remap_reg: - iounmap(denali->flash_reg); -failed_req_regions: - pci_release_regions(dev); -failed_dma_map: - dma_unmap_single(&dev->dev, denali->buf.dma_buf, DENALI_BUF_SIZE, - DMA_BIDIRECTIONAL); -failed_enable_dev: - pci_disable_device(dev); -failed_alloc_memery: - kfree(denali); + denali_irq_cleanup(denali->irq, denali); + return ret; } +EXPORT_SYMBOL(denali_init); /* driver exit point */ -static void denali_pci_remove(struct pci_dev *dev) +void denali_remove(struct denali_nand_info *denali) { - struct denali_nand_info *denali = pci_get_drvdata(dev); - - nand_release(&denali->mtd); - - denali_irq_cleanup(dev->irq, denali); - - iounmap(denali->flash_reg); - iounmap(denali->flash_mem); - pci_release_regions(dev); - pci_disable_device(dev); - dma_unmap_single(&dev->dev, denali->buf.dma_buf, DENALI_BUF_SIZE, - DMA_BIDIRECTIONAL); - pci_set_drvdata(dev, NULL); - kfree(denali); + denali_irq_cleanup(denali->irq, denali); + dma_unmap_single(denali->dev, denali->buf.dma_buf, DENALI_BUF_SIZE, + DMA_BIDIRECTIONAL); } - -MODULE_DEVICE_TABLE(pci, denali_pci_ids); - -static struct pci_driver denali_pci_driver = { - .name = DENALI_NAND_NAME, - .id_table = denali_pci_ids, - .probe = denali_pci_probe, - .remove = denali_pci_remove, -}; - -module_pci_driver(denali_pci_driver); +EXPORT_SYMBOL(denali_remove); diff --git a/drivers/mtd/nand/denali.h b/drivers/mtd/nand/denali.h index fabb9d56b39e..cec5712862c9 100644 --- a/drivers/mtd/nand/denali.h +++ b/drivers/mtd/nand/denali.h @@ -466,6 +466,7 @@ struct nand_buf { #define INTEL_CE4100 1 #define INTEL_MRST 2 +#define DT 3 struct denali_nand_info { struct mtd_info mtd; @@ -487,6 +488,7 @@ struct denali_nand_info { uint32_t irq_status; int irq_debug_array[32]; int idx; + int irq; uint32_t devnum; /* represent how many nands connected */ uint32_t fwblks; /* represent how many blocks FW used */ @@ -496,4 +498,7 @@ struct denali_nand_info { uint32_t max_banks; }; +extern int denali_init(struct denali_nand_info *denali); +extern void denali_remove(struct denali_nand_info *denali); + #endif /*_LLD_NAND_*/ diff --git a/drivers/mtd/nand/denali_dt.c b/drivers/mtd/nand/denali_dt.c new file mode 100644 index 000000000000..546f8cb5688d --- /dev/null +++ b/drivers/mtd/nand/denali_dt.c @@ -0,0 +1,167 @@ +/* + * NAND Flash Controller Device Driver for DT + * + * Copyright © 2011, Picochip. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ +#include <linux/clk.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/ioport.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/slab.h> + +#include "denali.h" + +struct denali_dt { + struct denali_nand_info denali; + struct clk *clk; +}; + +static void __iomem *request_and_map(struct device *dev, + const struct resource *res) +{ + void __iomem *ptr; + + if (!devm_request_mem_region(dev, res->start, resource_size(res), + "denali-dt")) { + dev_err(dev, "unable to request %s\n", res->name); + return NULL; + } + + ptr = devm_ioremap_nocache(dev, res->start, resource_size(res)); + if (!res) + dev_err(dev, "ioremap_nocache of %s failed!", res->name); + + return ptr; +} + +static const struct of_device_id denali_nand_dt_ids[] = { + { .compatible = "denali,denali-nand-dt" }, + { /* sentinel */ } + }; + +MODULE_DEVICE_TABLE(of, denali_nand_dt_ids); + +static u64 denali_dma_mask; + +static int denali_dt_probe(struct platform_device *ofdev) +{ + struct resource *denali_reg, *nand_data; + struct denali_dt *dt; + struct denali_nand_info *denali; + int ret; + const struct of_device_id *of_id; + + of_id = of_match_device(denali_nand_dt_ids, &ofdev->dev); + if (of_id) { + ofdev->id_entry = of_id->data; + } else { + pr_err("Failed to find the right device id.\n"); + return -ENOMEM; + } + + dt = devm_kzalloc(&ofdev->dev, sizeof(*dt), GFP_KERNEL); + if (!dt) + return -ENOMEM; + denali = &dt->denali; + + denali_reg = platform_get_resource_byname(ofdev, IORESOURCE_MEM, "denali_reg"); + nand_data = platform_get_resource_byname(ofdev, IORESOURCE_MEM, "nand_data"); + if (!denali_reg || !nand_data) { + dev_err(&ofdev->dev, "resources not completely defined\n"); + return -EINVAL; + } + + denali->platform = DT; + denali->dev = &ofdev->dev; + denali->irq = platform_get_irq(ofdev, 0); + if (denali->irq < 0) { + dev_err(&ofdev->dev, "no irq defined\n"); + return -ENXIO; + } + + denali->flash_reg = request_and_map(&ofdev->dev, denali_reg); + if (!denali->flash_reg) + return -ENOMEM; + + denali->flash_mem = request_and_map(&ofdev->dev, nand_data); + if (!denali->flash_mem) + return -ENOMEM; + + if (!of_property_read_u32(ofdev->dev.of_node, + "dma-mask", (u32 *)&denali_dma_mask)) { + denali->dev->dma_mask = &denali_dma_mask; + } else { + denali->dev->dma_mask = NULL; + } + + dt->clk = clk_get(&ofdev->dev, NULL); + if (IS_ERR(dt->clk)) { + dev_err(&ofdev->dev, "no clk available\n"); + return PTR_ERR(dt->clk); + } + clk_prepare_enable(dt->clk); + + ret = denali_init(denali); + if (ret) + goto out_disable_clk; + + platform_set_drvdata(ofdev, dt); + return 0; + +out_disable_clk: + clk_disable_unprepare(dt->clk); + clk_put(dt->clk); + + return ret; +} + +static int denali_dt_remove(struct platform_device *ofdev) +{ + struct denali_dt *dt = platform_get_drvdata(ofdev); + + denali_remove(&dt->denali); + clk_disable(dt->clk); + clk_put(dt->clk); + + return 0; +} + +static struct platform_driver denali_dt_driver = { + .probe = denali_dt_probe, + .remove = denali_dt_remove, + .driver = { + .name = "denali-nand-dt", + .owner = THIS_MODULE, + .of_match_table = of_match_ptr(denali_nand_dt_ids), + }, +}; + +static int __init denali_init_dt(void) +{ + return platform_driver_register(&denali_dt_driver); +} +module_init(denali_init_dt); + +static void __exit denali_exit_dt(void) +{ + platform_driver_unregister(&denali_dt_driver); +} +module_exit(denali_exit_dt); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jamie Iles"); +MODULE_DESCRIPTION("DT driver for Denali NAND controller"); diff --git a/drivers/mtd/nand/denali_pci.c b/drivers/mtd/nand/denali_pci.c new file mode 100644 index 000000000000..e3e46623b2b4 --- /dev/null +++ b/drivers/mtd/nand/denali_pci.c @@ -0,0 +1,144 @@ +/* + * NAND Flash Controller Device Driver + * Copyright © 2009-2010, Intel Corporation and its suppliers. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/slab.h> + +#include "denali.h" + +#define DENALI_NAND_NAME "denali-nand-pci" + +/* List of platforms this NAND controller has be integrated into */ +static DEFINE_PCI_DEVICE_TABLE(denali_pci_ids) = { + { PCI_VDEVICE(INTEL, 0x0701), INTEL_CE4100 }, + { PCI_VDEVICE(INTEL, 0x0809), INTEL_MRST }, + { /* end: all zeroes */ } +}; +MODULE_DEVICE_TABLE(pci, denali_pci_ids); + +static int denali_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) +{ + int ret = -ENODEV; + resource_size_t csr_base, mem_base; + unsigned long csr_len, mem_len; + struct denali_nand_info *denali; + + denali = kzalloc(sizeof(*denali), GFP_KERNEL); + if (!denali) + return -ENOMEM; + + ret = pci_enable_device(dev); + if (ret) { + pr_err("Spectra: pci_enable_device failed.\n"); + goto failed_alloc_memery; + } + + if (id->driver_data == INTEL_CE4100) { + denali->platform = INTEL_CE4100; + mem_base = pci_resource_start(dev, 0); + mem_len = pci_resource_len(dev, 1); + csr_base = pci_resource_start(dev, 1); + csr_len = pci_resource_len(dev, 1); + } else { + denali->platform = INTEL_MRST; + csr_base = pci_resource_start(dev, 0); + csr_len = pci_resource_len(dev, 0); + mem_base = pci_resource_start(dev, 1); + mem_len = pci_resource_len(dev, 1); + if (!mem_len) { + mem_base = csr_base + csr_len; + mem_len = csr_len; + } + } + + pci_set_master(dev); + denali->dev = &dev->dev; + denali->irq = dev->irq; + + ret = pci_request_regions(dev, DENALI_NAND_NAME); + if (ret) { + pr_err("Spectra: Unable to request memory regions\n"); + goto failed_enable_dev; + } + + denali->flash_reg = ioremap_nocache(csr_base, csr_len); + if (!denali->flash_reg) { + pr_err("Spectra: Unable to remap memory region\n"); + ret = -ENOMEM; + goto failed_req_regions; + } + + denali->flash_mem = ioremap_nocache(mem_base, mem_len); + if (!denali->flash_mem) { + pr_err("Spectra: ioremap_nocache failed!"); + ret = -ENOMEM; + goto failed_remap_reg; + } + + ret = denali_init(denali); + if (ret) + goto failed_remap_mem; + + pci_set_drvdata(dev, denali); + + return 0; + +failed_remap_mem: + iounmap(denali->flash_mem); +failed_remap_reg: + iounmap(denali->flash_reg); +failed_req_regions: + pci_release_regions(dev); +failed_enable_dev: + pci_disable_device(dev); +failed_alloc_memery: + kfree(denali); + + return ret; +} + +/* driver exit point */ +static void denali_pci_remove(struct pci_dev *dev) +{ + struct denali_nand_info *denali = pci_get_drvdata(dev); + + denali_remove(denali); + iounmap(denali->flash_reg); + iounmap(denali->flash_mem); + pci_release_regions(dev); + pci_disable_device(dev); + pci_set_drvdata(dev, NULL); + kfree(denali); +} + +static struct pci_driver denali_pci_driver = { + .name = DENALI_NAND_NAME, + .id_table = denali_pci_ids, + .probe = denali_pci_probe, + .remove = denali_pci_remove, +}; + +static int denali_init_pci(void) +{ + pr_info("Spectra MTD driver built on %s @ %s\n", __DATE__, __TIME__); + return pci_register_driver(&denali_pci_driver); +} +module_init(denali_init_pci); + +static void denali_exit_pci(void) +{ + pci_unregister_driver(&denali_pci_driver); +} +module_exit(denali_exit_pci); diff --git a/drivers/mtd/nand/diskonchip.c b/drivers/mtd/nand/diskonchip.c index 256eb30f6180..81fa5784f98b 100644 --- a/drivers/mtd/nand/diskonchip.c +++ b/drivers/mtd/nand/diskonchip.c @@ -53,8 +53,6 @@ static unsigned long __initdata doc_locations[] = { 0xe0000, 0xe2000, 0xe4000, 0xe6000, 0xe8000, 0xea000, 0xec000, 0xee000, #endif /* CONFIG_MTD_DOCPROBE_HIGH */ -#else -#warning Unknown architecture for DiskOnChip. No default probe locations defined #endif 0xffffffff }; diff --git a/drivers/mtd/nand/docg4.c b/drivers/mtd/nand/docg4.c index 799da5d1c857..18fa4489e52e 100644 --- a/drivers/mtd/nand/docg4.c +++ b/drivers/mtd/nand/docg4.c @@ -46,6 +46,25 @@ #include <linux/bitrev.h> /* + * In "reliable mode" consecutive 2k pages are used in parallel (in some + * fashion) to store the same data. The data can be read back from the + * even-numbered pages in the normal manner; odd-numbered pages will appear to + * contain junk. Systems that boot from the docg4 typically write the secondary + * program loader (SPL) code in this mode. The SPL is loaded by the initial + * program loader (IPL, stored in the docg4's 2k NOR-like region that is mapped + * to the reset vector address). This module parameter enables you to use this + * driver to write the SPL. When in this mode, no more than 2k of data can be + * written at a time, because the addresses do not increment in the normal + * manner, and the starting offset must be within an even-numbered 2k region; + * i.e., invalid starting offsets are 0x800, 0xa00, 0xc00, 0xe00, 0x1800, + * 0x1a00, ... Reliable mode is a special case and should not be used unless + * you know what you're doing. + */ +static bool reliable_mode; +module_param(reliable_mode, bool, 0); +MODULE_PARM_DESC(reliable_mode, "pages are programmed in reliable mode"); + +/* * You'll want to ignore badblocks if you're reading a partition that contains * data written by the TrueFFS library (i.e., by PalmOS, Windows, etc), since * it does not use mtd nand's method for marking bad blocks (using oob area). @@ -113,6 +132,7 @@ struct docg4_priv { #define DOCG4_SEQ_PAGEWRITE 0x16 #define DOCG4_SEQ_PAGEPROG 0x1e #define DOCG4_SEQ_BLOCKERASE 0x24 +#define DOCG4_SEQ_SETMODE 0x45 /* DOC_FLASHCOMMAND register commands */ #define DOCG4_CMD_PAGE_READ 0x00 @@ -122,6 +142,8 @@ struct docg4_priv { #define DOC_CMD_PROG_BLOCK_ADDR 0x60 #define DOCG4_CMD_PAGEWRITE 0x80 #define DOC_CMD_PROG_CYCLE2 0x10 +#define DOCG4_CMD_FAST_MODE 0xa3 /* functionality guessed */ +#define DOC_CMD_RELIABLE_MODE 0x22 #define DOC_CMD_RESET 0xff /* DOC_POWERMODE register bits */ @@ -190,17 +212,20 @@ struct docg4_priv { #define DOCG4_T 4 /* BCH alg corrects up to 4 bit errors */ #define DOCG4_FACTORY_BBT_PAGE 16 /* page where read-only factory bbt lives */ +#define DOCG4_REDUNDANT_BBT_PAGE 24 /* page where redundant factory bbt lives */ /* - * Oob bytes 0 - 6 are available to the user. - * Byte 7 is hamming ecc for first 7 bytes. Bytes 8 - 14 are hw-generated ecc. + * Bytes 0, 1 are used as badblock marker. + * Bytes 2 - 6 are available to the user. + * Byte 7 is hamming ecc for first 7 oob bytes only. + * Bytes 8 - 14 are hw-generated ecc covering entire page + oob bytes 0 - 14. * Byte 15 (the last) is used by the driver as a "page written" flag. */ static struct nand_ecclayout docg4_oobinfo = { .eccbytes = 9, .eccpos = {7, 8, 9, 10, 11, 12, 13, 14, 15}, - .oobavail = 7, - .oobfree = { {0, 7} } + .oobavail = 5, + .oobfree = { {.offset = 2, .length = 5} } }; /* @@ -611,6 +636,14 @@ static void write_page_prologue(struct mtd_info *mtd, uint32_t docg4_addr) dev_dbg(doc->dev, "docg4: %s: g4 addr: %x\n", __func__, docg4_addr); sequence_reset(mtd); + + if (unlikely(reliable_mode)) { + writew(DOCG4_SEQ_SETMODE, docptr + DOC_FLASHSEQUENCE); + writew(DOCG4_CMD_FAST_MODE, docptr + DOC_FLASHCOMMAND); + writew(DOC_CMD_RELIABLE_MODE, docptr + DOC_FLASHCOMMAND); + write_nop(docptr); + } + writew(DOCG4_SEQ_PAGEWRITE, docptr + DOC_FLASHSEQUENCE); writew(DOCG4_CMD_PAGEWRITE, docptr + DOC_FLASHCOMMAND); write_nop(docptr); @@ -691,6 +724,15 @@ static void docg4_command(struct mtd_info *mtd, unsigned command, int column, break; case NAND_CMD_SEQIN: + if (unlikely(reliable_mode)) { + uint16_t g4_page = g4_addr >> 16; + + /* writes to odd-numbered 2k pages are invalid */ + if (g4_page & 0x01) + dev_warn(doc->dev, + "invalid reliable mode address\n"); + } + write_page_prologue(mtd, g4_addr); /* hack for deferred write of oob bytes */ @@ -979,16 +1021,15 @@ static int __init read_factory_bbt(struct mtd_info *mtd) struct docg4_priv *doc = nand->priv; uint32_t g4_addr = mtd_to_docg4_address(DOCG4_FACTORY_BBT_PAGE, 0); uint8_t *buf; - int i, block, status; + int i, block; + __u32 eccfailed_stats = mtd->ecc_stats.failed; buf = kzalloc(DOCG4_PAGE_SIZE, GFP_KERNEL); if (buf == NULL) return -ENOMEM; read_page_prologue(mtd, g4_addr); - status = docg4_read_page(mtd, nand, buf, 0, DOCG4_FACTORY_BBT_PAGE); - if (status) - goto exit; + docg4_read_page(mtd, nand, buf, 0, DOCG4_FACTORY_BBT_PAGE); /* * If no memory-based bbt was created, exit. This will happen if module @@ -1000,6 +1041,20 @@ static int __init read_factory_bbt(struct mtd_info *mtd) if (nand->bbt == NULL) /* no memory-based bbt */ goto exit; + if (mtd->ecc_stats.failed > eccfailed_stats) { + /* + * Whoops, an ecc failure ocurred reading the factory bbt. + * It is stored redundantly, so we get another chance. + */ + eccfailed_stats = mtd->ecc_stats.failed; + docg4_read_page(mtd, nand, buf, 0, DOCG4_REDUNDANT_BBT_PAGE); + if (mtd->ecc_stats.failed > eccfailed_stats) { + dev_warn(doc->dev, + "The factory bbt could not be read!\n"); + goto exit; + } + } + /* * Parse factory bbt and update memory-based bbt. Factory bbt format is * simple: one bit per block, block numbers increase left to right (msb @@ -1019,7 +1074,7 @@ static int __init read_factory_bbt(struct mtd_info *mtd) } exit: kfree(buf); - return status; + return 0; } static int docg4_block_markbad(struct mtd_info *mtd, loff_t ofs) diff --git a/drivers/mtd/nand/fsl_elbc_nand.c b/drivers/mtd/nand/fsl_elbc_nand.c index cc1480a5e4c1..20657209a472 100644 --- a/drivers/mtd/nand/fsl_elbc_nand.c +++ b/drivers/mtd/nand/fsl_elbc_nand.c @@ -109,20 +109,6 @@ static struct nand_ecclayout fsl_elbc_oob_lp_eccm1 = { }; /* - * fsl_elbc_oob_lp_eccm* specify that LP NAND's OOB free area starts at offset - * 1, so we have to adjust bad block pattern. This pattern should be used for - * x8 chips only. So far hardware does not support x16 chips anyway. - */ -static u8 scan_ff_pattern[] = { 0xff, }; - -static struct nand_bbt_descr largepage_memorybased = { - .options = 0, - .offs = 0, - .len = 1, - .pattern = scan_ff_pattern, -}; - -/* * ELBC may use HW ECC, so that OOB offsets, that NAND core uses for bbt, * interfere with ECC positions, that's why we implement our own descriptors. * OOB {11, 5}, works for both SP and LP chips, with ECCM = 1 and ECCM = 0. @@ -699,7 +685,6 @@ static int fsl_elbc_chip_init_tail(struct mtd_info *mtd) chip->ecc.layout = (priv->fmr & FMR_ECCM) ? &fsl_elbc_oob_lp_eccm1 : &fsl_elbc_oob_lp_eccm0; - chip->badblock_pattern = &largepage_memorybased; } } else { dev_err(priv->dev, @@ -814,7 +799,7 @@ static int fsl_elbc_chip_remove(struct fsl_elbc_mtd *priv) static DEFINE_MUTEX(fsl_elbc_nand_mutex); -static int __devinit fsl_elbc_nand_probe(struct platform_device *pdev) +static int fsl_elbc_nand_probe(struct platform_device *pdev) { struct fsl_lbc_regs __iomem *lbc; struct fsl_elbc_mtd *priv; diff --git a/drivers/mtd/nand/fsl_ifc_nand.c b/drivers/mtd/nand/fsl_ifc_nand.c index 3551a99076ba..ad6222627fed 100644 --- a/drivers/mtd/nand/fsl_ifc_nand.c +++ b/drivers/mtd/nand/fsl_ifc_nand.c @@ -389,7 +389,7 @@ static void fsl_ifc_cmdfunc(struct mtd_info *mtd, unsigned int command, timing = IFC_FIR_OP_RBCD; out_be32(&ifc->ifc_nand.nand_fir0, - (IFC_FIR_OP_CMD0 << IFC_NAND_FIR0_OP0_SHIFT) | + (IFC_FIR_OP_CW0 << IFC_NAND_FIR0_OP0_SHIFT) | (IFC_FIR_OP_UA << IFC_NAND_FIR0_OP1_SHIFT) | (timing << IFC_NAND_FIR0_OP2_SHIFT)); out_be32(&ifc->ifc_nand.nand_fcr0, @@ -754,7 +754,7 @@ static void fsl_ifc_sram_init(struct fsl_ifc_mtd *priv) /* READID */ out_be32(&ifc->ifc_nand.nand_fir0, - (IFC_FIR_OP_CMD0 << IFC_NAND_FIR0_OP0_SHIFT) | + (IFC_FIR_OP_CW0 << IFC_NAND_FIR0_OP0_SHIFT) | (IFC_FIR_OP_UA << IFC_NAND_FIR0_OP1_SHIFT) | (IFC_FIR_OP_RB << IFC_NAND_FIR0_OP2_SHIFT)); out_be32(&ifc->ifc_nand.nand_fcr0, @@ -922,7 +922,7 @@ static int match_bank(struct fsl_ifc_regs __iomem *ifc, int bank, static DEFINE_MUTEX(fsl_ifc_nand_mutex); -static int __devinit fsl_ifc_nand_probe(struct platform_device *dev) +static int fsl_ifc_nand_probe(struct platform_device *dev) { struct fsl_ifc_regs __iomem *ifc; struct fsl_ifc_mtd *priv; diff --git a/drivers/mtd/nand/fsl_upm.c b/drivers/mtd/nand/fsl_upm.c index 45df542b9c61..5a8f5c4ce512 100644 --- a/drivers/mtd/nand/fsl_upm.c +++ b/drivers/mtd/nand/fsl_upm.c @@ -152,7 +152,7 @@ static void fun_write_buf(struct mtd_info *mtd, const uint8_t *buf, int len) fun_wait_rnb(fun); } -static int __devinit fun_chip_init(struct fsl_upm_nand *fun, +static int fun_chip_init(struct fsl_upm_nand *fun, const struct device_node *upm_np, const struct resource *io_res) { @@ -201,7 +201,7 @@ err: return ret; } -static int __devinit fun_probe(struct platform_device *ofdev) +static int fun_probe(struct platform_device *ofdev) { struct fsl_upm_nand *fun; struct resource io_res; @@ -318,7 +318,7 @@ err1: return ret; } -static int __devexit fun_remove(struct platform_device *ofdev) +static int fun_remove(struct platform_device *ofdev) { struct fsl_upm_nand *fun = dev_get_drvdata(&ofdev->dev); int i; @@ -350,7 +350,7 @@ static struct platform_driver of_fun_driver = { .of_match_table = of_fun_match, }, .probe = fun_probe, - .remove = __devexit_p(fun_remove), + .remove = fun_remove, }; module_platform_driver(of_fun_driver); diff --git a/drivers/mtd/nand/fsmc_nand.c b/drivers/mtd/nand/fsmc_nand.c index 38d26240d8b1..1d7446434b0e 100644 --- a/drivers/mtd/nand/fsmc_nand.c +++ b/drivers/mtd/nand/fsmc_nand.c @@ -361,7 +361,7 @@ static void fsmc_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl) struct nand_chip *this = mtd->priv; struct fsmc_nand_data *host = container_of(mtd, struct fsmc_nand_data, mtd); - void *__iomem *regs = host->regs_va; + void __iomem *regs = host->regs_va; unsigned int bank = host->bank; if (ctrl & NAND_CTRL_CHANGE) { @@ -383,13 +383,13 @@ static void fsmc_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl) pc |= FSMC_ENABLE; else pc &= ~FSMC_ENABLE; - writel(pc, FSMC_NAND_REG(regs, bank, PC)); + writel_relaxed(pc, FSMC_NAND_REG(regs, bank, PC)); } mb(); if (cmd != NAND_CMD_NONE) - writeb(cmd, this->IO_ADDR_W); + writeb_relaxed(cmd, this->IO_ADDR_W); } /* @@ -426,14 +426,18 @@ static void fsmc_nand_setup(void __iomem *regs, uint32_t bank, tset = (tims->tset & FSMC_TSET_MASK) << FSMC_TSET_SHIFT; if (busw) - writel(value | FSMC_DEVWID_16, FSMC_NAND_REG(regs, bank, PC)); + writel_relaxed(value | FSMC_DEVWID_16, + FSMC_NAND_REG(regs, bank, PC)); else - writel(value | FSMC_DEVWID_8, FSMC_NAND_REG(regs, bank, PC)); + writel_relaxed(value | FSMC_DEVWID_8, + FSMC_NAND_REG(regs, bank, PC)); - writel(readl(FSMC_NAND_REG(regs, bank, PC)) | tclr | tar, + writel_relaxed(readl(FSMC_NAND_REG(regs, bank, PC)) | tclr | tar, FSMC_NAND_REG(regs, bank, PC)); - writel(thiz | thold | twait | tset, FSMC_NAND_REG(regs, bank, COMM)); - writel(thiz | thold | twait | tset, FSMC_NAND_REG(regs, bank, ATTRIB)); + writel_relaxed(thiz | thold | twait | tset, + FSMC_NAND_REG(regs, bank, COMM)); + writel_relaxed(thiz | thold | twait | tset, + FSMC_NAND_REG(regs, bank, ATTRIB)); } /* @@ -446,11 +450,11 @@ static void fsmc_enable_hwecc(struct mtd_info *mtd, int mode) void __iomem *regs = host->regs_va; uint32_t bank = host->bank; - writel(readl(FSMC_NAND_REG(regs, bank, PC)) & ~FSMC_ECCPLEN_256, + writel_relaxed(readl(FSMC_NAND_REG(regs, bank, PC)) & ~FSMC_ECCPLEN_256, FSMC_NAND_REG(regs, bank, PC)); - writel(readl(FSMC_NAND_REG(regs, bank, PC)) & ~FSMC_ECCEN, + writel_relaxed(readl(FSMC_NAND_REG(regs, bank, PC)) & ~FSMC_ECCEN, FSMC_NAND_REG(regs, bank, PC)); - writel(readl(FSMC_NAND_REG(regs, bank, PC)) | FSMC_ECCEN, + writel_relaxed(readl(FSMC_NAND_REG(regs, bank, PC)) | FSMC_ECCEN, FSMC_NAND_REG(regs, bank, PC)); } @@ -470,7 +474,7 @@ static int fsmc_read_hwecc_ecc4(struct mtd_info *mtd, const uint8_t *data, unsigned long deadline = jiffies + FSMC_BUSY_WAIT_TIMEOUT; do { - if (readl(FSMC_NAND_REG(regs, bank, STS)) & FSMC_CODE_RDY) + if (readl_relaxed(FSMC_NAND_REG(regs, bank, STS)) & FSMC_CODE_RDY) break; else cond_resched(); @@ -481,25 +485,25 @@ static int fsmc_read_hwecc_ecc4(struct mtd_info *mtd, const uint8_t *data, return -ETIMEDOUT; } - ecc_tmp = readl(FSMC_NAND_REG(regs, bank, ECC1)); + ecc_tmp = readl_relaxed(FSMC_NAND_REG(regs, bank, ECC1)); ecc[0] = (uint8_t) (ecc_tmp >> 0); ecc[1] = (uint8_t) (ecc_tmp >> 8); ecc[2] = (uint8_t) (ecc_tmp >> 16); ecc[3] = (uint8_t) (ecc_tmp >> 24); - ecc_tmp = readl(FSMC_NAND_REG(regs, bank, ECC2)); + ecc_tmp = readl_relaxed(FSMC_NAND_REG(regs, bank, ECC2)); ecc[4] = (uint8_t) (ecc_tmp >> 0); ecc[5] = (uint8_t) (ecc_tmp >> 8); ecc[6] = (uint8_t) (ecc_tmp >> 16); ecc[7] = (uint8_t) (ecc_tmp >> 24); - ecc_tmp = readl(FSMC_NAND_REG(regs, bank, ECC3)); + ecc_tmp = readl_relaxed(FSMC_NAND_REG(regs, bank, ECC3)); ecc[8] = (uint8_t) (ecc_tmp >> 0); ecc[9] = (uint8_t) (ecc_tmp >> 8); ecc[10] = (uint8_t) (ecc_tmp >> 16); ecc[11] = (uint8_t) (ecc_tmp >> 24); - ecc_tmp = readl(FSMC_NAND_REG(regs, bank, STS)); + ecc_tmp = readl_relaxed(FSMC_NAND_REG(regs, bank, STS)); ecc[12] = (uint8_t) (ecc_tmp >> 16); return 0; @@ -519,7 +523,7 @@ static int fsmc_read_hwecc_ecc1(struct mtd_info *mtd, const uint8_t *data, uint32_t bank = host->bank; uint32_t ecc_tmp; - ecc_tmp = readl(FSMC_NAND_REG(regs, bank, ECC1)); + ecc_tmp = readl_relaxed(FSMC_NAND_REG(regs, bank, ECC1)); ecc[0] = (uint8_t) (ecc_tmp >> 0); ecc[1] = (uint8_t) (ecc_tmp >> 8); ecc[2] = (uint8_t) (ecc_tmp >> 16); @@ -601,7 +605,7 @@ static int dma_xfer(struct fsmc_nand_data *host, void *buffer, int len, dma_async_issue_pending(chan); ret = - wait_for_completion_interruptible_timeout(&host->dma_access_complete, + wait_for_completion_timeout(&host->dma_access_complete, msecs_to_jiffies(3000)); if (ret <= 0) { chan->device->device_control(chan, DMA_TERMINATE_ALL, 0); @@ -628,10 +632,10 @@ static void fsmc_write_buf(struct mtd_info *mtd, const uint8_t *buf, int len) uint32_t *p = (uint32_t *)buf; len = len >> 2; for (i = 0; i < len; i++) - writel(p[i], chip->IO_ADDR_W); + writel_relaxed(p[i], chip->IO_ADDR_W); } else { for (i = 0; i < len; i++) - writeb(buf[i], chip->IO_ADDR_W); + writeb_relaxed(buf[i], chip->IO_ADDR_W); } } @@ -651,10 +655,10 @@ static void fsmc_read_buf(struct mtd_info *mtd, uint8_t *buf, int len) uint32_t *p = (uint32_t *)buf; len = len >> 2; for (i = 0; i < len; i++) - p[i] = readl(chip->IO_ADDR_R); + p[i] = readl_relaxed(chip->IO_ADDR_R); } else { for (i = 0; i < len; i++) - buf[i] = readb(chip->IO_ADDR_R); + buf[i] = readb_relaxed(chip->IO_ADDR_R); } } @@ -783,7 +787,7 @@ static int fsmc_bch8_correct_data(struct mtd_info *mtd, uint8_t *dat, uint32_t num_err, i; uint32_t ecc1, ecc2, ecc3, ecc4; - num_err = (readl(FSMC_NAND_REG(regs, bank, STS)) >> 10) & 0xF; + num_err = (readl_relaxed(FSMC_NAND_REG(regs, bank, STS)) >> 10) & 0xF; /* no bit flipping */ if (likely(num_err == 0)) @@ -826,10 +830,10 @@ static int fsmc_bch8_correct_data(struct mtd_info *mtd, uint8_t *dat, * uint64_t array and error offset indexes are populated in err_idx * array */ - ecc1 = readl(FSMC_NAND_REG(regs, bank, ECC1)); - ecc2 = readl(FSMC_NAND_REG(regs, bank, ECC2)); - ecc3 = readl(FSMC_NAND_REG(regs, bank, ECC3)); - ecc4 = readl(FSMC_NAND_REG(regs, bank, STS)); + ecc1 = readl_relaxed(FSMC_NAND_REG(regs, bank, ECC1)); + ecc2 = readl_relaxed(FSMC_NAND_REG(regs, bank, ECC2)); + ecc3 = readl_relaxed(FSMC_NAND_REG(regs, bank, ECC3)); + ecc4 = readl_relaxed(FSMC_NAND_REG(regs, bank, STS)); err_idx[0] = (ecc1 >> 0) & 0x1FFF; err_idx[1] = (ecc1 >> 13) & 0x1FFF; @@ -860,7 +864,7 @@ static bool filter(struct dma_chan *chan, void *slave) } #ifdef CONFIG_OF -static int __devinit fsmc_nand_probe_config_dt(struct platform_device *pdev, +static int fsmc_nand_probe_config_dt(struct platform_device *pdev, struct device_node *np) { struct fsmc_nand_platform_data *pdata = dev_get_platdata(&pdev->dev); @@ -876,15 +880,13 @@ static int __devinit fsmc_nand_probe_config_dt(struct platform_device *pdev, return -EINVAL; } } - of_property_read_u32(np, "st,ale-off", &pdata->ale_off); - of_property_read_u32(np, "st,cle-off", &pdata->cle_off); if (of_get_property(np, "nand-skip-bbtscan", NULL)) pdata->options = NAND_SKIP_BBTSCAN; return 0; } #else -static int __devinit fsmc_nand_probe_config_dt(struct platform_device *pdev, +static int fsmc_nand_probe_config_dt(struct platform_device *pdev, struct device_node *np) { return -ENOSYS; @@ -935,41 +937,28 @@ static int __init fsmc_nand_probe(struct platform_device *pdev) if (!res) return -EINVAL; - if (!devm_request_mem_region(&pdev->dev, res->start, resource_size(res), - pdev->name)) { - dev_err(&pdev->dev, "Failed to get memory data resourse\n"); - return -ENOENT; - } - - host->data_pa = (dma_addr_t)res->start; - host->data_va = devm_ioremap(&pdev->dev, res->start, - resource_size(res)); + host->data_va = devm_request_and_ioremap(&pdev->dev, res); if (!host->data_va) { dev_err(&pdev->dev, "data ioremap failed\n"); return -ENOMEM; } + host->data_pa = (dma_addr_t)res->start; - if (!devm_request_mem_region(&pdev->dev, res->start + pdata->ale_off, - resource_size(res), pdev->name)) { - dev_err(&pdev->dev, "Failed to get memory ale resourse\n"); - return -ENOENT; - } + res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "nand_addr"); + if (!res) + return -EINVAL; - host->addr_va = devm_ioremap(&pdev->dev, res->start + pdata->ale_off, - resource_size(res)); + host->addr_va = devm_request_and_ioremap(&pdev->dev, res); if (!host->addr_va) { dev_err(&pdev->dev, "ale ioremap failed\n"); return -ENOMEM; } - if (!devm_request_mem_region(&pdev->dev, res->start + pdata->cle_off, - resource_size(res), pdev->name)) { - dev_err(&pdev->dev, "Failed to get memory cle resourse\n"); - return -ENOENT; - } + res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "nand_cmd"); + if (!res) + return -EINVAL; - host->cmd_va = devm_ioremap(&pdev->dev, res->start + pdata->cle_off, - resource_size(res)); + host->cmd_va = devm_request_and_ioremap(&pdev->dev, res); if (!host->cmd_va) { dev_err(&pdev->dev, "ale ioremap failed\n"); return -ENOMEM; @@ -979,14 +968,7 @@ static int __init fsmc_nand_probe(struct platform_device *pdev) if (!res) return -EINVAL; - if (!devm_request_mem_region(&pdev->dev, res->start, resource_size(res), - pdev->name)) { - dev_err(&pdev->dev, "Failed to get memory regs resourse\n"); - return -ENOENT; - } - - host->regs_va = devm_ioremap(&pdev->dev, res->start, - resource_size(res)); + host->regs_va = devm_request_and_ioremap(&pdev->dev, res); if (!host->regs_va) { dev_err(&pdev->dev, "regs ioremap failed\n"); return -ENOMEM; diff --git a/drivers/mtd/nand/gpio.c b/drivers/mtd/nand/gpio.c index bc73bc5f2713..e789e3f51710 100644 --- a/drivers/mtd/nand/gpio.c +++ b/drivers/mtd/nand/gpio.c @@ -90,14 +90,14 @@ static void gpio_nand_writebuf(struct mtd_info *mtd, const u_char *buf, int len) { struct nand_chip *this = mtd->priv; - writesb(this->IO_ADDR_W, buf, len); + iowrite8_rep(this->IO_ADDR_W, buf, len); } static void gpio_nand_readbuf(struct mtd_info *mtd, u_char *buf, int len) { struct nand_chip *this = mtd->priv; - readsb(this->IO_ADDR_R, buf, len); + ioread8_rep(this->IO_ADDR_R, buf, len); } static void gpio_nand_writebuf16(struct mtd_info *mtd, const u_char *buf, @@ -106,7 +106,7 @@ static void gpio_nand_writebuf16(struct mtd_info *mtd, const u_char *buf, struct nand_chip *this = mtd->priv; if (IS_ALIGNED((unsigned long)buf, 2)) { - writesw(this->IO_ADDR_W, buf, len>>1); + iowrite16_rep(this->IO_ADDR_W, buf, len>>1); } else { int i; unsigned short *ptr = (unsigned short *)buf; @@ -121,7 +121,7 @@ static void gpio_nand_readbuf16(struct mtd_info *mtd, u_char *buf, int len) struct nand_chip *this = mtd->priv; if (IS_ALIGNED((unsigned long)buf, 2)) { - readsw(this->IO_ADDR_R, buf, len>>1); + ioread16_rep(this->IO_ADDR_R, buf, len>>1); } else { int i; unsigned short *ptr = (unsigned short *)buf; @@ -134,7 +134,11 @@ static void gpio_nand_readbuf16(struct mtd_info *mtd, u_char *buf, int len) static int gpio_nand_devready(struct mtd_info *mtd) { struct gpiomtd *gpiomtd = gpio_nand_getpriv(mtd); - return gpio_get_value(gpiomtd->plat.gpio_rdy); + + if (gpio_is_valid(gpiomtd->plat.gpio_rdy)) + return gpio_get_value(gpiomtd->plat.gpio_rdy); + + return 1; } #ifdef CONFIG_OF @@ -227,7 +231,7 @@ gpio_nand_get_io_sync(struct platform_device *pdev) return platform_get_resource(pdev, IORESOURCE_MEM, 1); } -static int __devexit gpio_nand_remove(struct platform_device *dev) +static int gpio_nand_remove(struct platform_device *dev) { struct gpiomtd *gpiomtd = platform_get_drvdata(dev); struct resource *res; @@ -252,7 +256,8 @@ static int __devexit gpio_nand_remove(struct platform_device *dev) gpio_free(gpiomtd->plat.gpio_nce); if (gpio_is_valid(gpiomtd->plat.gpio_nwp)) gpio_free(gpiomtd->plat.gpio_nwp); - gpio_free(gpiomtd->plat.gpio_rdy); + if (gpio_is_valid(gpiomtd->plat.gpio_rdy)) + gpio_free(gpiomtd->plat.gpio_rdy); kfree(gpiomtd); @@ -277,7 +282,7 @@ static void __iomem *request_and_remap(struct resource *res, size_t size, return ptr; } -static int __devinit gpio_nand_probe(struct platform_device *dev) +static int gpio_nand_probe(struct platform_device *dev) { struct gpiomtd *gpiomtd; struct nand_chip *this; @@ -336,10 +341,12 @@ static int __devinit gpio_nand_probe(struct platform_device *dev) if (ret) goto err_cle; gpio_direction_output(gpiomtd->plat.gpio_cle, 0); - ret = gpio_request(gpiomtd->plat.gpio_rdy, "NAND RDY"); - if (ret) - goto err_rdy; - gpio_direction_input(gpiomtd->plat.gpio_rdy); + if (gpio_is_valid(gpiomtd->plat.gpio_rdy)) { + ret = gpio_request(gpiomtd->plat.gpio_rdy, "NAND RDY"); + if (ret) + goto err_rdy; + gpio_direction_input(gpiomtd->plat.gpio_rdy); + } this->IO_ADDR_W = this->IO_ADDR_R; @@ -386,7 +393,8 @@ static int __devinit gpio_nand_probe(struct platform_device *dev) err_wp: if (gpio_is_valid(gpiomtd->plat.gpio_nwp)) gpio_set_value(gpiomtd->plat.gpio_nwp, 0); - gpio_free(gpiomtd->plat.gpio_rdy); + if (gpio_is_valid(gpiomtd->plat.gpio_rdy)) + gpio_free(gpiomtd->plat.gpio_rdy); err_rdy: gpio_free(gpiomtd->plat.gpio_cle); err_cle: diff --git a/drivers/mtd/nand/gpmi-nand/gpmi-lib.c b/drivers/mtd/nand/gpmi-nand/gpmi-lib.c index 3502accd4bc3..d84699c7968e 100644 --- a/drivers/mtd/nand/gpmi-nand/gpmi-lib.c +++ b/drivers/mtd/nand/gpmi-nand/gpmi-lib.c @@ -18,7 +18,6 @@ * with this program; if not, write to the Free Software Foundation, Inc., * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. */ -#include <linux/mtd/gpmi-nand.h> #include <linux/delay.h> #include <linux/clk.h> @@ -166,6 +165,15 @@ int gpmi_init(struct gpmi_nand_data *this) if (ret) goto err_out; + /* + * Reset BCH here, too. We got failures otherwise :( + * See later BCH reset for explanation of MX23 handling + */ + ret = gpmi_reset_block(r->bch_regs, GPMI_IS_MX23(this)); + if (ret) + goto err_out; + + /* Choose NAND mode. */ writel(BM_GPMI_CTRL1_GPMI_MODE, r->gpmi_regs + HW_GPMI_CTRL1_CLR); diff --git a/drivers/mtd/nand/gpmi-nand/gpmi-nand.c b/drivers/mtd/nand/gpmi-nand/gpmi-nand.c index d79696b2f19b..5cd141f7bfc2 100644 --- a/drivers/mtd/nand/gpmi-nand/gpmi-nand.c +++ b/drivers/mtd/nand/gpmi-nand/gpmi-nand.c @@ -25,7 +25,6 @@ #include <linux/slab.h> #include <linux/interrupt.h> #include <linux/module.h> -#include <linux/mtd/gpmi-nand.h> #include <linux/mtd/partitions.h> #include <linux/pinctrl/consumer.h> #include <linux/of.h> @@ -33,6 +32,12 @@ #include <linux/of_mtd.h> #include "gpmi-nand.h" +/* Resource names for the GPMI NAND driver. */ +#define GPMI_NAND_GPMI_REGS_ADDR_RES_NAME "gpmi-nand" +#define GPMI_NAND_BCH_REGS_ADDR_RES_NAME "bch" +#define GPMI_NAND_BCH_INTERRUPT_RES_NAME "bch" +#define GPMI_NAND_DMA_INTERRUPT_RES_NAME "gpmi-dma" + /* add our owner bbt descriptor */ static uint8_t scan_ff_pattern[] = { 0xff }; static struct nand_bbt_descr gpmi_bbt_descr = { @@ -222,7 +227,7 @@ void prepare_data_dma(struct gpmi_nand_data *this, enum dma_data_direction dr) ret = dma_map_sg(this->dev, sgl, 1, dr); if (ret == 0) - pr_err("map failed.\n"); + pr_err("DMA mapping failed.\n"); this->direct_dma_map_ok = false; } @@ -314,7 +319,7 @@ int start_dma_with_bch_irq(struct gpmi_nand_data *this, return 0; } -static int __devinit +static int acquire_register_block(struct gpmi_nand_data *this, const char *res_name) { struct platform_device *pdev = this->pdev; @@ -355,7 +360,7 @@ static void release_register_block(struct gpmi_nand_data *this) res->bch_regs = NULL; } -static int __devinit +static int acquire_bch_irq(struct gpmi_nand_data *this, irq_handler_t irq_h) { struct platform_device *pdev = this->pdev; @@ -422,7 +427,7 @@ static void release_dma_channels(struct gpmi_nand_data *this) } } -static int __devinit acquire_dma_channels(struct gpmi_nand_data *this) +static int acquire_dma_channels(struct gpmi_nand_data *this) { struct platform_device *pdev = this->pdev; struct resource *r_dma; @@ -456,7 +461,7 @@ static int __devinit acquire_dma_channels(struct gpmi_nand_data *this) dma_chan = dma_request_channel(mask, gpmi_dma_filter, this); if (!dma_chan) { - pr_err("dma_request_channel failed.\n"); + pr_err("Failed to request DMA channel.\n"); goto acquire_err; } @@ -487,7 +492,7 @@ static char *extra_clks_for_mx6q[GPMI_CLK_MAX] = { "gpmi_apb", "gpmi_bch", "gpmi_bch_apb", "per1_bch", }; -static int __devinit gpmi_get_clks(struct gpmi_nand_data *this) +static int gpmi_get_clks(struct gpmi_nand_data *this) { struct resources *r = &this->resources; char **extra_clks = NULL; @@ -533,7 +538,7 @@ err_clock: return -ENOMEM; } -static int __devinit acquire_resources(struct gpmi_nand_data *this) +static int acquire_resources(struct gpmi_nand_data *this) { struct pinctrl *pinctrl; int ret; @@ -583,7 +588,7 @@ static void release_resources(struct gpmi_nand_data *this) release_dma_channels(this); } -static int __devinit init_hardware(struct gpmi_nand_data *this) +static int init_hardware(struct gpmi_nand_data *this) { int ret; @@ -625,7 +630,8 @@ static int read_page_prepare(struct gpmi_nand_data *this, length, DMA_FROM_DEVICE); if (dma_mapping_error(dev, dest_phys)) { if (alt_size < length) { - pr_err("Alternate buffer is too small\n"); + pr_err("%s, Alternate buffer is too small\n", + __func__); return -ENOMEM; } goto map_failed; @@ -675,7 +681,8 @@ static int send_page_prepare(struct gpmi_nand_data *this, DMA_TO_DEVICE); if (dma_mapping_error(dev, source_phys)) { if (alt_size < length) { - pr_err("Alternate buffer is too small\n"); + pr_err("%s, Alternate buffer is too small\n", + __func__); return -ENOMEM; } goto map_failed; @@ -763,7 +770,7 @@ static int gpmi_alloc_dma_buffer(struct gpmi_nand_data *this) error_alloc: gpmi_free_dma_buffer(this); - pr_err("allocate DMA buffer ret!!\n"); + pr_err("Error allocating DMA buffers!\n"); return -ENOMEM; } @@ -1474,7 +1481,7 @@ static int gpmi_set_geometry(struct gpmi_nand_data *this) /* Set up the NFC geometry which is used by BCH. */ ret = bch_set_geometry(this); if (ret) { - pr_err("set geometry ret : %d\n", ret); + pr_err("Error setting BCH geometry : %d\n", ret); return ret; } @@ -1535,7 +1542,7 @@ static void gpmi_nfc_exit(struct gpmi_nand_data *this) gpmi_free_dma_buffer(this); } -static int __devinit gpmi_nfc_init(struct gpmi_nand_data *this) +static int gpmi_nfc_init(struct gpmi_nand_data *this) { struct mtd_info *mtd = &this->mtd; struct nand_chip *chip = &this->nand; @@ -1618,7 +1625,7 @@ static const struct of_device_id gpmi_nand_id_table[] = { }; MODULE_DEVICE_TABLE(of, gpmi_nand_id_table); -static int __devinit gpmi_nand_probe(struct platform_device *pdev) +static int gpmi_nand_probe(struct platform_device *pdev) { struct gpmi_nand_data *this; const struct of_device_id *of_id; @@ -1668,7 +1675,7 @@ exit_acquire_resources: return ret; } -static int __devexit gpmi_nand_remove(struct platform_device *pdev) +static int gpmi_nand_remove(struct platform_device *pdev) { struct gpmi_nand_data *this = platform_get_drvdata(pdev); @@ -1685,7 +1692,7 @@ static struct platform_driver gpmi_nand_driver = { .of_match_table = gpmi_nand_id_table, }, .probe = gpmi_nand_probe, - .remove = __devexit_p(gpmi_nand_remove), + .remove = gpmi_nand_remove, .id_table = gpmi_ids, }; module_platform_driver(gpmi_nand_driver); diff --git a/drivers/mtd/nand/gpmi-nand/gpmi-nand.h b/drivers/mtd/nand/gpmi-nand/gpmi-nand.h index 7ac25c1e58f9..3d93a5e39090 100644 --- a/drivers/mtd/nand/gpmi-nand/gpmi-nand.h +++ b/drivers/mtd/nand/gpmi-nand/gpmi-nand.h @@ -130,7 +130,6 @@ struct gpmi_nand_data { /* System Interface */ struct device *dev; struct platform_device *pdev; - struct gpmi_nand_platform_data *pdata; /* Resources */ struct resources resources; diff --git a/drivers/mtd/nand/jz4740_nand.c b/drivers/mtd/nand/jz4740_nand.c index 100b6775e175..8d415f014e1d 100644 --- a/drivers/mtd/nand/jz4740_nand.c +++ b/drivers/mtd/nand/jz4740_nand.c @@ -316,13 +316,17 @@ err: return ret; } -static inline void jz_nand_iounmap_resource(struct resource *res, void __iomem *base) +static inline void jz_nand_iounmap_resource(struct resource *res, + void __iomem *base) { iounmap(base); release_mem_region(res->start, resource_size(res)); } -static int __devinit jz_nand_detect_bank(struct platform_device *pdev, struct jz_nand *nand, unsigned char bank, size_t chipnr, uint8_t *nand_maf_id, uint8_t *nand_dev_id) { +static int jz_nand_detect_bank(struct platform_device *pdev, + struct jz_nand *nand, unsigned char bank, + size_t chipnr, uint8_t *nand_maf_id, + uint8_t *nand_dev_id) { int ret; int gpio; char gpio_name[9]; @@ -400,7 +404,7 @@ notfound_gpio: return ret; } -static int __devinit jz_nand_probe(struct platform_device *pdev) +static int jz_nand_probe(struct platform_device *pdev) { int ret; struct jz_nand *nand; @@ -541,7 +545,7 @@ err_free: return ret; } -static int __devexit jz_nand_remove(struct platform_device *pdev) +static int jz_nand_remove(struct platform_device *pdev) { struct jz_nand *nand = platform_get_drvdata(pdev); struct jz_nand_platform_data *pdata = pdev->dev.platform_data; @@ -573,7 +577,7 @@ static int __devexit jz_nand_remove(struct platform_device *pdev) static struct platform_driver jz_nand_driver = { .probe = jz_nand_probe, - .remove = __devexit_p(jz_nand_remove), + .remove = jz_nand_remove, .driver = { .name = "jz4740-nand", .owner = THIS_MODULE, diff --git a/drivers/mtd/nand/lpc32xx_mlc.c b/drivers/mtd/nand/lpc32xx_mlc.c index c29b7ac1f6af..f182befa7360 100644 --- a/drivers/mtd/nand/lpc32xx_mlc.c +++ b/drivers/mtd/nand/lpc32xx_mlc.c @@ -655,7 +655,7 @@ static struct lpc32xx_nand_cfg_mlc *lpc32xx_parse_dt(struct device *dev) /* * Probe for NAND controller */ -static int __devinit lpc32xx_nand_probe(struct platform_device *pdev) +static int lpc32xx_nand_probe(struct platform_device *pdev) { struct lpc32xx_nand_host *host; struct mtd_info *mtd; @@ -845,7 +845,7 @@ err_exit1: /* * Remove NAND device */ -static int __devexit lpc32xx_nand_remove(struct platform_device *pdev) +static int lpc32xx_nand_remove(struct platform_device *pdev) { struct lpc32xx_nand_host *host = platform_get_drvdata(pdev); struct mtd_info *mtd = &host->mtd; @@ -907,7 +907,7 @@ MODULE_DEVICE_TABLE(of, lpc32xx_nand_match); static struct platform_driver lpc32xx_nand_driver = { .probe = lpc32xx_nand_probe, - .remove = __devexit_p(lpc32xx_nand_remove), + .remove = lpc32xx_nand_remove, .resume = lpc32xx_nand_resume, .suspend = lpc32xx_nand_suspend, .driver = { diff --git a/drivers/mtd/nand/lpc32xx_slc.c b/drivers/mtd/nand/lpc32xx_slc.c index 32409c45d479..030b78c62895 100644 --- a/drivers/mtd/nand/lpc32xx_slc.c +++ b/drivers/mtd/nand/lpc32xx_slc.c @@ -755,7 +755,7 @@ static struct lpc32xx_nand_cfg_slc *lpc32xx_parse_dt(struct device *dev) /* * Probe for NAND controller */ -static int __devinit lpc32xx_nand_probe(struct platform_device *pdev) +static int lpc32xx_nand_probe(struct platform_device *pdev) { struct lpc32xx_nand_host *host; struct mtd_info *mtd; @@ -949,7 +949,7 @@ err_exit1: /* * Remove NAND device. */ -static int __devexit lpc32xx_nand_remove(struct platform_device *pdev) +static int lpc32xx_nand_remove(struct platform_device *pdev) { uint32_t tmp; struct lpc32xx_nand_host *host = platform_get_drvdata(pdev); @@ -1021,7 +1021,7 @@ MODULE_DEVICE_TABLE(of, lpc32xx_nand_match); static struct platform_driver lpc32xx_nand_driver = { .probe = lpc32xx_nand_probe, - .remove = __devexit_p(lpc32xx_nand_remove), + .remove = lpc32xx_nand_remove, .resume = lpc32xx_nand_resume, .suspend = lpc32xx_nand_suspend, .driver = { diff --git a/drivers/mtd/nand/mpc5121_nfc.c b/drivers/mtd/nand/mpc5121_nfc.c index f776c8577b8c..3c9cdcbc4cba 100644 --- a/drivers/mtd/nand/mpc5121_nfc.c +++ b/drivers/mtd/nand/mpc5121_nfc.c @@ -626,7 +626,7 @@ static void mpc5121_nfc_free(struct device *dev, struct mtd_info *mtd) iounmap(prv->csreg); } -static int __devinit mpc5121_nfc_probe(struct platform_device *op) +static int mpc5121_nfc_probe(struct platform_device *op) { struct device_node *rootnode, *dn = op->dev.of_node; struct device *dev = &op->dev; @@ -827,7 +827,7 @@ error: return retval; } -static int __devexit mpc5121_nfc_remove(struct platform_device *op) +static int mpc5121_nfc_remove(struct platform_device *op) { struct device *dev = &op->dev; struct mtd_info *mtd = dev_get_drvdata(dev); @@ -841,14 +841,14 @@ static int __devexit mpc5121_nfc_remove(struct platform_device *op) return 0; } -static struct of_device_id mpc5121_nfc_match[] __devinitdata = { +static struct of_device_id mpc5121_nfc_match[] = { { .compatible = "fsl,mpc5121-nfc", }, {}, }; static struct platform_driver mpc5121_nfc_driver = { .probe = mpc5121_nfc_probe, - .remove = __devexit_p(mpc5121_nfc_remove), + .remove = mpc5121_nfc_remove, .driver = { .name = DRV_NAME, .owner = THIS_MODULE, diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c index 022dcdc256fb..45204e41a028 100644 --- a/drivers/mtd/nand/mxc_nand.c +++ b/drivers/mtd/nand/mxc_nand.c @@ -266,7 +266,8 @@ static struct nand_ecclayout nandv2_hw_eccoob_4k = { } }; -static const char *part_probes[] = { "RedBoot", "cmdlinepart", "ofpart", NULL }; +static const char const *part_probes[] = { + "cmdlinepart", "RedBoot", "ofpart", NULL }; static void memcpy32_fromio(void *trg, const void __iomem *src, size_t size) { @@ -1378,7 +1379,7 @@ static int __init mxcnd_probe_dt(struct mxc_nand_host *host) } #endif -static int __devinit mxcnd_probe(struct platform_device *pdev) +static int mxcnd_probe(struct platform_device *pdev) { struct nand_chip *this; struct mtd_info *mtd; @@ -1556,12 +1557,13 @@ static int __devinit mxcnd_probe(struct platform_device *pdev) return 0; escan: - clk_disable_unprepare(host->clk); + if (host->clk_act) + clk_disable_unprepare(host->clk); return err; } -static int __devexit mxcnd_remove(struct platform_device *pdev) +static int mxcnd_remove(struct platform_device *pdev) { struct mxc_nand_host *host = platform_get_drvdata(pdev); @@ -1580,7 +1582,7 @@ static struct platform_driver mxcnd_driver = { }, .id_table = mxcnd_devtype, .probe = mxcnd_probe, - .remove = __devexit_p(mxcnd_remove), + .remove = mxcnd_remove, }; module_platform_driver(mxcnd_driver); diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c index 1a03b7f673ce..8323ac991ad1 100644 --- a/drivers/mtd/nand/nand_base.c +++ b/drivers/mtd/nand/nand_base.c @@ -93,8 +93,7 @@ static struct nand_ecclayout nand_oob_128 = { .length = 78} } }; -static int nand_get_device(struct nand_chip *chip, struct mtd_info *mtd, - int new_state); +static int nand_get_device(struct mtd_info *mtd, int new_state); static int nand_do_write_oob(struct mtd_info *mtd, loff_t to, struct mtd_oob_ops *ops); @@ -130,15 +129,12 @@ static int check_offs_len(struct mtd_info *mtd, * nand_release_device - [GENERIC] release chip * @mtd: MTD device structure * - * Deselect, release chip lock and wake up anyone waiting on the device. + * Release chip lock and wake up anyone waiting on the device. */ static void nand_release_device(struct mtd_info *mtd) { struct nand_chip *chip = mtd->priv; - /* De-select the NAND device */ - chip->select_chip(mtd, -1); - /* Release the controller and the chip */ spin_lock(&chip->controller->lock); chip->controller->active = NULL; @@ -160,7 +156,7 @@ static uint8_t nand_read_byte(struct mtd_info *mtd) } /** - * nand_read_byte16 - [DEFAULT] read one byte endianess aware from the chip + * nand_read_byte16 - [DEFAULT] read one byte endianness aware from the chip * nand_read_byte16 - [DEFAULT] read one byte endianness aware from the chip * @mtd: MTD device structure * @@ -303,7 +299,7 @@ static int nand_block_bad(struct mtd_info *mtd, loff_t ofs, int getchip) if (getchip) { chipnr = (int)(ofs >> chip->chip_shift); - nand_get_device(chip, mtd, FL_READING); + nand_get_device(mtd, FL_READING); /* Select the NAND device */ chip->select_chip(mtd, chipnr); @@ -333,8 +329,10 @@ static int nand_block_bad(struct mtd_info *mtd, loff_t ofs, int getchip) i++; } while (!res && i < 2 && (chip->bbt_options & NAND_BBT_SCAN2NDPAGE)); - if (getchip) + if (getchip) { + chip->select_chip(mtd, -1); nand_release_device(mtd); + } return res; } @@ -383,7 +381,7 @@ static int nand_default_block_markbad(struct mtd_info *mtd, loff_t ofs) struct mtd_oob_ops ops; loff_t wr_ofs = ofs; - nand_get_device(chip, mtd, FL_WRITING); + nand_get_device(mtd, FL_WRITING); ops.datbuf = NULL; ops.oobbuf = buf; @@ -492,7 +490,7 @@ static void panic_nand_wait_ready(struct mtd_info *mtd, unsigned long timeo) void nand_wait_ready(struct mtd_info *mtd) { struct nand_chip *chip = mtd->priv; - unsigned long timeo = jiffies + 2; + unsigned long timeo = jiffies + msecs_to_jiffies(20); /* 400ms timeout */ if (in_interrupt() || oops_in_progress) @@ -750,15 +748,15 @@ static void panic_nand_get_device(struct nand_chip *chip, /** * nand_get_device - [GENERIC] Get chip for selected access - * @chip: the nand chip descriptor * @mtd: MTD device structure * @new_state: the state which is requested * * Get the device and lock it for exclusive access */ static int -nand_get_device(struct nand_chip *chip, struct mtd_info *mtd, int new_state) +nand_get_device(struct mtd_info *mtd, int new_state) { + struct nand_chip *chip = mtd->priv; spinlock_t *lock = &chip->controller->lock; wait_queue_head_t *wq = &chip->controller->wq; DECLARE_WAITQUEUE(wait, current); @@ -865,6 +863,8 @@ static int nand_wait(struct mtd_info *mtd, struct nand_chip *chip) led_trigger_event(nand_led_trigger, LED_OFF); status = (int)chip->read_byte(mtd); + /* This can happen if in case of timeout or buggy dev_ready */ + WARN_ON(!(status & NAND_STATUS_READY)); return status; } @@ -899,7 +899,7 @@ static int __nand_unlock(struct mtd_info *mtd, loff_t ofs, /* Call wait ready function */ status = chip->waitfunc(mtd, chip); /* See if device thinks it succeeded */ - if (status & 0x01) { + if (status & NAND_STATUS_FAIL) { pr_debug("%s: error status = 0x%08x\n", __func__, status); ret = -EIO; @@ -932,7 +932,7 @@ int nand_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) if (ofs + len == mtd->size) len -= mtd->erasesize; - nand_get_device(chip, mtd, FL_UNLOCKING); + nand_get_device(mtd, FL_UNLOCKING); /* Shift to get chip number */ chipnr = ofs >> chip->chip_shift; @@ -950,6 +950,7 @@ int nand_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) ret = __nand_unlock(mtd, ofs, len, 0); out: + chip->select_chip(mtd, -1); nand_release_device(mtd); return ret; @@ -981,7 +982,7 @@ int nand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) if (check_offs_len(mtd, ofs, len)) ret = -EINVAL; - nand_get_device(chip, mtd, FL_LOCKING); + nand_get_device(mtd, FL_LOCKING); /* Shift to get chip number */ chipnr = ofs >> chip->chip_shift; @@ -1004,7 +1005,7 @@ int nand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) /* Call wait ready function */ status = chip->waitfunc(mtd, chip); /* See if device thinks it succeeded */ - if (status & 0x01) { + if (status & NAND_STATUS_FAIL) { pr_debug("%s: error status = 0x%08x\n", __func__, status); ret = -EIO; @@ -1014,6 +1015,7 @@ int nand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) ret = __nand_unlock(mtd, ofs, len, 0x1); out: + chip->select_chip(mtd, -1); nand_release_device(mtd); return ret; @@ -1550,6 +1552,7 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from, chip->select_chip(mtd, chipnr); } } + chip->select_chip(mtd, -1); ops->retlen = ops->len - (size_t) readlen; if (oob) @@ -1577,11 +1580,10 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from, static int nand_read(struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, uint8_t *buf) { - struct nand_chip *chip = mtd->priv; struct mtd_oob_ops ops; int ret; - nand_get_device(chip, mtd, FL_READING); + nand_get_device(mtd, FL_READING); ops.len = len; ops.datbuf = buf; ops.oobbuf = NULL; @@ -1804,6 +1806,7 @@ static int nand_do_read_oob(struct mtd_info *mtd, loff_t from, chip->select_chip(mtd, chipnr); } } + chip->select_chip(mtd, -1); ops->oobretlen = ops->ooblen - readlen; @@ -1827,7 +1830,6 @@ static int nand_do_read_oob(struct mtd_info *mtd, loff_t from, static int nand_read_oob(struct mtd_info *mtd, loff_t from, struct mtd_oob_ops *ops) { - struct nand_chip *chip = mtd->priv; int ret = -ENOTSUPP; ops->retlen = 0; @@ -1839,7 +1841,7 @@ static int nand_read_oob(struct mtd_info *mtd, loff_t from, return -EINVAL; } - nand_get_device(chip, mtd, FL_READING); + nand_get_device(mtd, FL_READING); switch (ops->mode) { case MTD_OPS_PLACE_OOB: @@ -2186,8 +2188,10 @@ static int nand_do_write_ops(struct mtd_info *mtd, loff_t to, chip->select_chip(mtd, chipnr); /* Check, if it is write protected */ - if (nand_check_wp(mtd)) - return -EIO; + if (nand_check_wp(mtd)) { + ret = -EIO; + goto err_out; + } realpage = (int)(to >> chip->page_shift); page = realpage & chip->pagemask; @@ -2199,8 +2203,10 @@ static int nand_do_write_ops(struct mtd_info *mtd, loff_t to, chip->pagebuf = -1; /* Don't allow multipage oob writes with offset */ - if (oob && ops->ooboffs && (ops->ooboffs + ops->ooblen > oobmaxlen)) - return -EINVAL; + if (oob && ops->ooboffs && (ops->ooboffs + ops->ooblen > oobmaxlen)) { + ret = -EINVAL; + goto err_out; + } while (1) { int bytes = mtd->writesize; @@ -2251,6 +2257,9 @@ static int nand_do_write_ops(struct mtd_info *mtd, loff_t to, ops->retlen = ops->len - writelen; if (unlikely(oob)) ops->oobretlen = ops->ooblen; + +err_out: + chip->select_chip(mtd, -1); return ret; } @@ -2302,11 +2311,10 @@ static int panic_nand_write(struct mtd_info *mtd, loff_t to, size_t len, static int nand_write(struct mtd_info *mtd, loff_t to, size_t len, size_t *retlen, const uint8_t *buf) { - struct nand_chip *chip = mtd->priv; struct mtd_oob_ops ops; int ret; - nand_get_device(chip, mtd, FL_WRITING); + nand_get_device(mtd, FL_WRITING); ops.len = len; ops.datbuf = (uint8_t *)buf; ops.oobbuf = NULL; @@ -2377,8 +2385,10 @@ static int nand_do_write_oob(struct mtd_info *mtd, loff_t to, chip->cmdfunc(mtd, NAND_CMD_RESET, -1, -1); /* Check, if it is write protected */ - if (nand_check_wp(mtd)) + if (nand_check_wp(mtd)) { + chip->select_chip(mtd, -1); return -EROFS; + } /* Invalidate the page cache, if we write to the cached page */ if (page == chip->pagebuf) @@ -2391,6 +2401,8 @@ static int nand_do_write_oob(struct mtd_info *mtd, loff_t to, else status = chip->ecc.write_oob(mtd, chip, page & chip->pagemask); + chip->select_chip(mtd, -1); + if (status) return status; @@ -2408,7 +2420,6 @@ static int nand_do_write_oob(struct mtd_info *mtd, loff_t to, static int nand_write_oob(struct mtd_info *mtd, loff_t to, struct mtd_oob_ops *ops) { - struct nand_chip *chip = mtd->priv; int ret = -ENOTSUPP; ops->retlen = 0; @@ -2420,7 +2431,7 @@ static int nand_write_oob(struct mtd_info *mtd, loff_t to, return -EINVAL; } - nand_get_device(chip, mtd, FL_WRITING); + nand_get_device(mtd, FL_WRITING); switch (ops->mode) { case MTD_OPS_PLACE_OOB: @@ -2513,7 +2524,7 @@ int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr, return -EINVAL; /* Grab the lock and see if the device is available */ - nand_get_device(chip, mtd, FL_ERASING); + nand_get_device(mtd, FL_ERASING); /* Shift to get first page */ page = (int)(instr->addr >> chip->page_shift); @@ -2623,6 +2634,7 @@ erase_exit: ret = instr->state == MTD_ERASE_DONE ? 0 : -EIO; /* Deselect and wake up anyone waiting on the device */ + chip->select_chip(mtd, -1); nand_release_device(mtd); /* Do call back function */ @@ -2658,12 +2670,10 @@ erase_exit: */ static void nand_sync(struct mtd_info *mtd) { - struct nand_chip *chip = mtd->priv; - pr_debug("%s: called\n", __func__); /* Grab the lock and see if the device is available */ - nand_get_device(chip, mtd, FL_SYNCING); + nand_get_device(mtd, FL_SYNCING); /* Release it and go back */ nand_release_device(mtd); } @@ -2749,9 +2759,7 @@ static int nand_onfi_get_features(struct mtd_info *mtd, struct nand_chip *chip, */ static int nand_suspend(struct mtd_info *mtd) { - struct nand_chip *chip = mtd->priv; - - return nand_get_device(chip, mtd, FL_PM_SUSPENDED); + return nand_get_device(mtd, FL_PM_SUSPENDED); } /** @@ -2849,6 +2857,8 @@ static int nand_flash_detect_onfi(struct mtd_info *mtd, struct nand_chip *chip, int i; int val; + /* ONFI need to be probed in 8 bits mode */ + WARN_ON(chip->options & NAND_BUSWIDTH_16); /* Try ONFI for unknown chip or LP */ chip->cmdfunc(mtd, NAND_CMD_READID, 0x20, -1); if (chip->read_byte(mtd) != 'O' || chip->read_byte(mtd) != 'N' || @@ -2913,7 +2923,7 @@ static int nand_flash_detect_onfi(struct mtd_info *mtd, struct nand_chip *chip, * * Check if an ID string is repeated within a given sequence of bytes at * specific repetition interval period (e.g., {0x20,0x01,0x7F,0x20} has a - * period of 2). This is a helper function for nand_id_len(). Returns non-zero + * period of 3). This is a helper function for nand_id_len(). Returns non-zero * if the repetition has a period of @period; otherwise, returns zero. */ static int nand_id_has_period(u8 *id_data, int arrlen, int period) @@ -3242,11 +3252,15 @@ ident_done: break; } - /* - * Check, if buswidth is correct. Hardware drivers should set - * chip correct! - */ - if (busw != (chip->options & NAND_BUSWIDTH_16)) { + if (chip->options & NAND_BUSWIDTH_AUTO) { + WARN_ON(chip->options & NAND_BUSWIDTH_16); + chip->options |= busw; + nand_set_defaults(chip, busw); + } else if (busw != (chip->options & NAND_BUSWIDTH_16)) { + /* + * Check, if buswidth is correct. Hardware drivers should set + * chip correct! + */ pr_info("NAND device: Manufacturer ID:" " 0x%02x, Chip ID: 0x%02x (%s %s)\n", *maf_id, *dev_id, nand_manuf_ids[maf_idx].name, mtd->name); @@ -3285,10 +3299,10 @@ ident_done: chip->cmdfunc = nand_command_lp; pr_info("NAND device: Manufacturer ID: 0x%02x, Chip ID: 0x%02x (%s %s)," - " page size: %d, OOB size: %d\n", + " %dMiB, page size: %d, OOB size: %d\n", *maf_id, *dev_id, nand_manuf_ids[maf_idx].name, chip->onfi_version ? chip->onfi_params.model : type->name, - mtd->writesize, mtd->oobsize); + (int)(chip->chipsize >> 20), mtd->writesize, mtd->oobsize); return type; } @@ -3327,6 +3341,8 @@ int nand_scan_ident(struct mtd_info *mtd, int maxchips, return PTR_ERR(type); } + chip->select_chip(mtd, -1); + /* Check for a chip array */ for (i = 1; i < maxchips; i++) { chip->select_chip(mtd, i); @@ -3336,8 +3352,11 @@ int nand_scan_ident(struct mtd_info *mtd, int maxchips, chip->cmdfunc(mtd, NAND_CMD_READID, 0x00, -1); /* Read manufacturer and device IDs */ if (nand_maf_id != chip->read_byte(mtd) || - nand_dev_id != chip->read_byte(mtd)) + nand_dev_id != chip->read_byte(mtd)) { + chip->select_chip(mtd, -1); break; + } + chip->select_chip(mtd, -1); } if (i > 1) pr_info("%d NAND chips detected\n", i); @@ -3596,9 +3615,6 @@ int nand_scan_tail(struct mtd_info *mtd) /* Initialize state */ chip->state = FL_READY; - /* De-select the device */ - chip->select_chip(mtd, -1); - /* Invalidate the pagebuffer reference */ chip->pagebuf = -1; diff --git a/drivers/mtd/nand/nandsim.c b/drivers/mtd/nand/nandsim.c index a932c485eb04..818b65c85d12 100644 --- a/drivers/mtd/nand/nandsim.c +++ b/drivers/mtd/nand/nandsim.c @@ -42,6 +42,8 @@ #include <linux/sched.h> #include <linux/fs.h> #include <linux/pagemap.h> +#include <linux/seq_file.h> +#include <linux/debugfs.h> /* Default simulator parameters values */ #if !defined(CONFIG_NANDSIM_FIRST_ID_BYTE) || \ @@ -105,7 +107,6 @@ static char *weakblocks = NULL; static char *weakpages = NULL; static unsigned int bitflips = 0; static char *gravepages = NULL; -static unsigned int rptwear = 0; static unsigned int overridesize = 0; static char *cache_file = NULL; static unsigned int bbt; @@ -130,7 +131,6 @@ module_param(weakblocks, charp, 0400); module_param(weakpages, charp, 0400); module_param(bitflips, uint, 0400); module_param(gravepages, charp, 0400); -module_param(rptwear, uint, 0400); module_param(overridesize, uint, 0400); module_param(cache_file, charp, 0400); module_param(bbt, uint, 0400); @@ -162,7 +162,6 @@ MODULE_PARM_DESC(bitflips, "Maximum number of random bit flips per page (z MODULE_PARM_DESC(gravepages, "Pages that lose data [: maximum reads (defaults to 3)]" " separated by commas e.g. 1401:2 means page 1401" " can be read only twice before failing"); -MODULE_PARM_DESC(rptwear, "Number of erases between reporting wear, if not zero"); MODULE_PARM_DESC(overridesize, "Specifies the NAND Flash size overriding the ID bytes. " "The size is specified in erase blocks and as the exponent of a power of two" " e.g. 5 means a size of 32 erase blocks"); @@ -286,6 +285,11 @@ MODULE_PARM_DESC(bch, "Enable BCH ecc and set how many bits should " /* Maximum page cache pages needed to read or write a NAND page to the cache_file */ #define NS_MAX_HELD_PAGES 16 +struct nandsim_debug_info { + struct dentry *dfs_root; + struct dentry *dfs_wear_report; +}; + /* * A union to represent flash memory contents and flash buffer. */ @@ -365,6 +369,8 @@ struct nandsim { void *file_buf; struct page *held_pages[NS_MAX_HELD_PAGES]; int held_cnt; + + struct nandsim_debug_info dbg; }; /* @@ -442,11 +448,123 @@ static LIST_HEAD(grave_pages); static unsigned long *erase_block_wear = NULL; static unsigned int wear_eb_count = 0; static unsigned long total_wear = 0; -static unsigned int rptwear_cnt = 0; /* MTD structure for NAND controller */ static struct mtd_info *nsmtd; +static int nandsim_debugfs_show(struct seq_file *m, void *private) +{ + unsigned long wmin = -1, wmax = 0, avg; + unsigned long deciles[10], decile_max[10], tot = 0; + unsigned int i; + + /* Calc wear stats */ + for (i = 0; i < wear_eb_count; ++i) { + unsigned long wear = erase_block_wear[i]; + if (wear < wmin) + wmin = wear; + if (wear > wmax) + wmax = wear; + tot += wear; + } + + for (i = 0; i < 9; ++i) { + deciles[i] = 0; + decile_max[i] = (wmax * (i + 1) + 5) / 10; + } + deciles[9] = 0; + decile_max[9] = wmax; + for (i = 0; i < wear_eb_count; ++i) { + int d; + unsigned long wear = erase_block_wear[i]; + for (d = 0; d < 10; ++d) + if (wear <= decile_max[d]) { + deciles[d] += 1; + break; + } + } + avg = tot / wear_eb_count; + + /* Output wear report */ + seq_printf(m, "Total numbers of erases: %lu\n", tot); + seq_printf(m, "Number of erase blocks: %u\n", wear_eb_count); + seq_printf(m, "Average number of erases: %lu\n", avg); + seq_printf(m, "Maximum number of erases: %lu\n", wmax); + seq_printf(m, "Minimum number of erases: %lu\n", wmin); + for (i = 0; i < 10; ++i) { + unsigned long from = (i ? decile_max[i - 1] + 1 : 0); + if (from > decile_max[i]) + continue; + seq_printf(m, "Number of ebs with erase counts from %lu to %lu : %lu\n", + from, + decile_max[i], + deciles[i]); + } + + return 0; +} + +static int nandsim_debugfs_open(struct inode *inode, struct file *file) +{ + return single_open(file, nandsim_debugfs_show, inode->i_private); +} + +static const struct file_operations dfs_fops = { + .open = nandsim_debugfs_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +/** + * nandsim_debugfs_create - initialize debugfs + * @dev: nandsim device description object + * + * This function creates all debugfs files for UBI device @ubi. Returns zero in + * case of success and a negative error code in case of failure. + */ +static int nandsim_debugfs_create(struct nandsim *dev) +{ + struct nandsim_debug_info *dbg = &dev->dbg; + struct dentry *dent; + int err; + + if (!IS_ENABLED(CONFIG_DEBUG_FS)) + return 0; + + dent = debugfs_create_dir("nandsim", NULL); + if (IS_ERR_OR_NULL(dent)) { + int err = dent ? -ENODEV : PTR_ERR(dent); + + NS_ERR("cannot create \"nandsim\" debugfs directory, err %d\n", + err); + return err; + } + dbg->dfs_root = dent; + + dent = debugfs_create_file("wear_report", S_IRUSR, + dbg->dfs_root, dev, &dfs_fops); + if (IS_ERR_OR_NULL(dent)) + goto out_remove; + dbg->dfs_wear_report = dent; + + return 0; + +out_remove: + debugfs_remove_recursive(dbg->dfs_root); + err = dent ? PTR_ERR(dent) : -ENODEV; + return err; +} + +/** + * nandsim_debugfs_remove - destroy all debugfs files + */ +static void nandsim_debugfs_remove(struct nandsim *ns) +{ + if (IS_ENABLED(CONFIG_DEBUG_FS)) + debugfs_remove_recursive(ns->dbg.dfs_root); +} + /* * Allocate array of page pointers, create slab allocation for an array * and initialize the array by NULL pointers. @@ -911,8 +1029,6 @@ static int setup_wear_reporting(struct mtd_info *mtd) { size_t mem; - if (!rptwear) - return 0; wear_eb_count = div_u64(mtd->size, mtd->erasesize); mem = wear_eb_count * sizeof(unsigned long); if (mem / sizeof(unsigned long) != wear_eb_count) { @@ -929,64 +1045,18 @@ static int setup_wear_reporting(struct mtd_info *mtd) static void update_wear(unsigned int erase_block_no) { - unsigned long wmin = -1, wmax = 0, avg; - unsigned long deciles[10], decile_max[10], tot = 0; - unsigned int i; - if (!erase_block_wear) return; total_wear += 1; + /* + * TODO: Notify this through a debugfs entry, + * instead of showing an error message. + */ if (total_wear == 0) NS_ERR("Erase counter total overflow\n"); erase_block_wear[erase_block_no] += 1; if (erase_block_wear[erase_block_no] == 0) NS_ERR("Erase counter overflow for erase block %u\n", erase_block_no); - rptwear_cnt += 1; - if (rptwear_cnt < rptwear) - return; - rptwear_cnt = 0; - /* Calc wear stats */ - for (i = 0; i < wear_eb_count; ++i) { - unsigned long wear = erase_block_wear[i]; - if (wear < wmin) - wmin = wear; - if (wear > wmax) - wmax = wear; - tot += wear; - } - for (i = 0; i < 9; ++i) { - deciles[i] = 0; - decile_max[i] = (wmax * (i + 1) + 5) / 10; - } - deciles[9] = 0; - decile_max[9] = wmax; - for (i = 0; i < wear_eb_count; ++i) { - int d; - unsigned long wear = erase_block_wear[i]; - for (d = 0; d < 10; ++d) - if (wear <= decile_max[d]) { - deciles[d] += 1; - break; - } - } - avg = tot / wear_eb_count; - /* Output wear report */ - NS_INFO("*** Wear Report ***\n"); - NS_INFO("Total numbers of erases: %lu\n", tot); - NS_INFO("Number of erase blocks: %u\n", wear_eb_count); - NS_INFO("Average number of erases: %lu\n", avg); - NS_INFO("Maximum number of erases: %lu\n", wmax); - NS_INFO("Minimum number of erases: %lu\n", wmin); - for (i = 0; i < 10; ++i) { - unsigned long from = (i ? decile_max[i - 1] + 1 : 0); - if (from > decile_max[i]) - continue; - NS_INFO("Number of ebs with erase counts from %lu to %lu : %lu\n", - from, - decile_max[i], - deciles[i]); - } - NS_INFO("*** End of Wear Report ***\n"); } /* @@ -1397,10 +1467,7 @@ int do_read_error(struct nandsim *ns, int num) unsigned int page_no = ns->regs.row; if (read_error(page_no)) { - int i; - memset(ns->buf.byte, 0xFF, num); - for (i = 0; i < num; ++i) - ns->buf.byte[i] = random32(); + prandom_bytes(ns->buf.byte, num); NS_WARN("simulating read error in page %u\n", page_no); return 1; } @@ -2330,6 +2397,9 @@ static int __init ns_init_module(void) if ((retval = setup_wear_reporting(nsmtd)) != 0) goto err_exit; + if ((retval = nandsim_debugfs_create(nand)) != 0) + goto err_exit; + if ((retval = init_nandsim(nsmtd)) != 0) goto err_exit; @@ -2369,6 +2439,7 @@ static void __exit ns_cleanup_module(void) struct nandsim *ns = ((struct nand_chip *)nsmtd->priv)->priv; int i; + nandsim_debugfs_remove(ns); free_nandsim(ns); /* Free nandsim private resources */ nand_release(nsmtd); /* Unregister driver */ for (i = 0;i < ARRAY_SIZE(ns->partitions); ++i) diff --git a/drivers/mtd/nand/ndfc.c b/drivers/mtd/nand/ndfc.c index 5fd3f010e3ae..8e148f1478fd 100644 --- a/drivers/mtd/nand/ndfc.c +++ b/drivers/mtd/nand/ndfc.c @@ -197,7 +197,7 @@ err: return ret; } -static int __devinit ndfc_probe(struct platform_device *ofdev) +static int ndfc_probe(struct platform_device *ofdev) { struct ndfc_controller *ndfc; const __be32 *reg; @@ -256,7 +256,7 @@ static int __devinit ndfc_probe(struct platform_device *ofdev) return 0; } -static int __devexit ndfc_remove(struct platform_device *ofdev) +static int ndfc_remove(struct platform_device *ofdev) { struct ndfc_controller *ndfc = dev_get_drvdata(&ofdev->dev); @@ -279,7 +279,7 @@ static struct platform_driver ndfc_driver = { .of_match_table = ndfc_match, }, .probe = ndfc_probe, - .remove = __devexit_p(ndfc_remove), + .remove = ndfc_remove, }; module_platform_driver(ndfc_driver); diff --git a/drivers/mtd/nand/nomadik_nand.c b/drivers/mtd/nand/nomadik_nand.c deleted file mode 100644 index 9ee0c4edfacf..000000000000 --- a/drivers/mtd/nand/nomadik_nand.c +++ /dev/null @@ -1,235 +0,0 @@ -/* - * drivers/mtd/nand/nomadik_nand.c - * - * Overview: - * Driver for on-board NAND flash on Nomadik Platforms - * - * Copyright © 2007 STMicroelectronics Pvt. Ltd. - * Author: Sachin Verma <sachin.verma@st.com> - * - * Copyright © 2009 Alessandro Rubini - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - */ - -#include <linux/init.h> -#include <linux/module.h> -#include <linux/types.h> -#include <linux/mtd/mtd.h> -#include <linux/mtd/nand.h> -#include <linux/mtd/nand_ecc.h> -#include <linux/platform_device.h> -#include <linux/mtd/partitions.h> -#include <linux/io.h> -#include <linux/slab.h> -#include <linux/platform_data/mtd-nomadik-nand.h> -#include <mach/fsmc.h> - -#include <mtd/mtd-abi.h> - -struct nomadik_nand_host { - struct mtd_info mtd; - struct nand_chip nand; - void __iomem *data_va; - void __iomem *cmd_va; - void __iomem *addr_va; - struct nand_bbt_descr *bbt_desc; -}; - -static struct nand_ecclayout nomadik_ecc_layout = { - .eccbytes = 3 * 4, - .eccpos = { /* each subpage has 16 bytes: pos 2,3,4 hosts ECC */ - 0x02, 0x03, 0x04, - 0x12, 0x13, 0x14, - 0x22, 0x23, 0x24, - 0x32, 0x33, 0x34}, - /* let's keep bytes 5,6,7 for us, just in case we change ECC algo */ - .oobfree = { {0x08, 0x08}, {0x18, 0x08}, {0x28, 0x08}, {0x38, 0x08} }, -}; - -static void nomadik_ecc_control(struct mtd_info *mtd, int mode) -{ - /* No need to enable hw ecc, it's on by default */ -} - -static void nomadik_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl) -{ - struct nand_chip *nand = mtd->priv; - struct nomadik_nand_host *host = nand->priv; - - if (cmd == NAND_CMD_NONE) - return; - - if (ctrl & NAND_CLE) - writeb(cmd, host->cmd_va); - else - writeb(cmd, host->addr_va); -} - -static int nomadik_nand_probe(struct platform_device *pdev) -{ - struct nomadik_nand_platform_data *pdata = pdev->dev.platform_data; - struct nomadik_nand_host *host; - struct mtd_info *mtd; - struct nand_chip *nand; - struct resource *res; - int ret = 0; - - /* Allocate memory for the device structure (and zero it) */ - host = kzalloc(sizeof(struct nomadik_nand_host), GFP_KERNEL); - if (!host) { - dev_err(&pdev->dev, "Failed to allocate device structure.\n"); - return -ENOMEM; - } - - /* Call the client's init function, if any */ - if (pdata->init) - ret = pdata->init(); - if (ret < 0) { - dev_err(&pdev->dev, "Init function failed\n"); - goto err; - } - - /* ioremap three regions */ - res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "nand_addr"); - if (!res) { - ret = -EIO; - goto err_unmap; - } - host->addr_va = ioremap(res->start, resource_size(res)); - - res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "nand_data"); - if (!res) { - ret = -EIO; - goto err_unmap; - } - host->data_va = ioremap(res->start, resource_size(res)); - - res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "nand_cmd"); - if (!res) { - ret = -EIO; - goto err_unmap; - } - host->cmd_va = ioremap(res->start, resource_size(res)); - - if (!host->addr_va || !host->data_va || !host->cmd_va) { - ret = -ENOMEM; - goto err_unmap; - } - - /* Link all private pointers */ - mtd = &host->mtd; - nand = &host->nand; - mtd->priv = nand; - nand->priv = host; - - host->mtd.owner = THIS_MODULE; - nand->IO_ADDR_R = host->data_va; - nand->IO_ADDR_W = host->data_va; - nand->cmd_ctrl = nomadik_cmd_ctrl; - - /* - * This stanza declares ECC_HW but uses soft routines. It's because - * HW claims to make the calculation but not the correction. However, - * I haven't managed to get the desired data out of it until now. - */ - nand->ecc.mode = NAND_ECC_SOFT; - nand->ecc.layout = &nomadik_ecc_layout; - nand->ecc.hwctl = nomadik_ecc_control; - nand->ecc.size = 512; - nand->ecc.bytes = 3; - - nand->options = pdata->options; - - /* - * Scan to find existence of the device - */ - if (nand_scan(&host->mtd, 1)) { - ret = -ENXIO; - goto err_unmap; - } - - mtd_device_register(&host->mtd, pdata->parts, pdata->nparts); - - platform_set_drvdata(pdev, host); - return 0; - - err_unmap: - if (host->cmd_va) - iounmap(host->cmd_va); - if (host->data_va) - iounmap(host->data_va); - if (host->addr_va) - iounmap(host->addr_va); - err: - kfree(host); - return ret; -} - -/* - * Clean up routine - */ -static int nomadik_nand_remove(struct platform_device *pdev) -{ - struct nomadik_nand_host *host = platform_get_drvdata(pdev); - struct nomadik_nand_platform_data *pdata = pdev->dev.platform_data; - - if (pdata->exit) - pdata->exit(); - - if (host) { - nand_release(&host->mtd); - iounmap(host->cmd_va); - iounmap(host->data_va); - iounmap(host->addr_va); - kfree(host); - } - return 0; -} - -static int nomadik_nand_suspend(struct device *dev) -{ - struct nomadik_nand_host *host = dev_get_drvdata(dev); - int ret = 0; - if (host) - ret = mtd_suspend(&host->mtd); - return ret; -} - -static int nomadik_nand_resume(struct device *dev) -{ - struct nomadik_nand_host *host = dev_get_drvdata(dev); - if (host) - mtd_resume(&host->mtd); - return 0; -} - -static const struct dev_pm_ops nomadik_nand_pm_ops = { - .suspend = nomadik_nand_suspend, - .resume = nomadik_nand_resume, -}; - -static struct platform_driver nomadik_nand_driver = { - .probe = nomadik_nand_probe, - .remove = nomadik_nand_remove, - .driver = { - .owner = THIS_MODULE, - .name = "nomadik_nand", - .pm = &nomadik_nand_pm_ops, - }, -}; - -module_platform_driver(nomadik_nand_driver); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("ST Microelectronics (sachin.verma@st.com)"); -MODULE_DESCRIPTION("NAND driver for Nomadik Platform"); diff --git a/drivers/mtd/nand/nuc900_nand.c b/drivers/mtd/nand/nuc900_nand.c index 94dc46bc118c..a6191198d259 100644 --- a/drivers/mtd/nand/nuc900_nand.c +++ b/drivers/mtd/nand/nuc900_nand.c @@ -246,7 +246,7 @@ static void nuc900_nand_enable(struct nuc900_nand *nand) spin_unlock(&nand->lock); } -static int __devinit nuc900_nand_probe(struct platform_device *pdev) +static int nuc900_nand_probe(struct platform_device *pdev) { struct nuc900_nand *nuc900_nand; struct nand_chip *chip; @@ -317,7 +317,7 @@ fail1: kfree(nuc900_nand); return retval; } -static int __devexit nuc900_nand_remove(struct platform_device *pdev) +static int nuc900_nand_remove(struct platform_device *pdev) { struct nuc900_nand *nuc900_nand = platform_get_drvdata(pdev); struct resource *res; @@ -340,7 +340,7 @@ static int __devexit nuc900_nand_remove(struct platform_device *pdev) static struct platform_driver nuc900_nand_driver = { .probe = nuc900_nand_probe, - .remove = __devexit_p(nuc900_nand_remove), + .remove = nuc900_nand_remove, .driver = { .name = "nuc900-fmi", .owner = THIS_MODULE, diff --git a/drivers/mtd/nand/omap2.c b/drivers/mtd/nand/omap2.c index 1f34ba104ef4..0002d5e94f0d 100644 --- a/drivers/mtd/nand/omap2.c +++ b/drivers/mtd/nand/omap2.c @@ -1323,7 +1323,7 @@ static void omap3_free_bch(struct mtd_info *mtd) } #endif /* CONFIG_MTD_NAND_OMAP_BCH */ -static int __devinit omap_nand_probe(struct platform_device *pdev) +static int omap_nand_probe(struct platform_device *pdev) { struct omap_nand_info *info; struct omap_nand_platform_data *pdata; diff --git a/drivers/mtd/nand/orion_nand.c b/drivers/mtd/nand/orion_nand.c index aefaf8cd31ef..cd72b9299f6b 100644 --- a/drivers/mtd/nand/orion_nand.c +++ b/drivers/mtd/nand/orion_nand.c @@ -194,7 +194,7 @@ no_res: return ret; } -static int __devexit orion_nand_remove(struct platform_device *pdev) +static int orion_nand_remove(struct platform_device *pdev) { struct mtd_info *mtd = platform_get_drvdata(pdev); struct nand_chip *nc = mtd->priv; @@ -223,7 +223,7 @@ static struct of_device_id orion_nand_of_match_table[] = { #endif static struct platform_driver orion_nand_driver = { - .remove = __devexit_p(orion_nand_remove), + .remove = orion_nand_remove, .driver = { .name = "orion_nand", .owner = THIS_MODULE, diff --git a/drivers/mtd/nand/pasemi_nand.c b/drivers/mtd/nand/pasemi_nand.c index 1440e51cedcc..5a67082c07ee 100644 --- a/drivers/mtd/nand/pasemi_nand.c +++ b/drivers/mtd/nand/pasemi_nand.c @@ -89,7 +89,7 @@ int pasemi_device_ready(struct mtd_info *mtd) return !!(inl(lpcctl) & LBICTRL_LPCCTL_NR); } -static int __devinit pasemi_nand_probe(struct platform_device *ofdev) +static int pasemi_nand_probe(struct platform_device *ofdev) { struct pci_dev *pdev; struct device_node *np = ofdev->dev.of_node; @@ -184,7 +184,7 @@ static int __devinit pasemi_nand_probe(struct platform_device *ofdev) return err; } -static int __devexit pasemi_nand_remove(struct platform_device *ofdev) +static int pasemi_nand_remove(struct platform_device *ofdev) { struct nand_chip *chip; diff --git a/drivers/mtd/nand/plat_nand.c b/drivers/mtd/nand/plat_nand.c index a47ee68a0cfa..c004566a9ad2 100644 --- a/drivers/mtd/nand/plat_nand.c +++ b/drivers/mtd/nand/plat_nand.c @@ -28,7 +28,7 @@ static const char *part_probe_types[] = { "cmdlinepart", NULL }; /* * Probe for the NAND device. */ -static int __devinit plat_nand_probe(struct platform_device *pdev) +static int plat_nand_probe(struct platform_device *pdev) { struct platform_nand_data *pdata = pdev->dev.platform_data; struct mtd_part_parser_data ppdata; @@ -134,7 +134,7 @@ out_free: /* * Remove a NAND device. */ -static int __devexit plat_nand_remove(struct platform_device *pdev) +static int plat_nand_remove(struct platform_device *pdev) { struct plat_nand_data *data = platform_get_drvdata(pdev); struct platform_nand_data *pdata = pdev->dev.platform_data; @@ -160,7 +160,7 @@ MODULE_DEVICE_TABLE(of, plat_nand_match); static struct platform_driver plat_nand_driver = { .probe = plat_nand_probe, - .remove = __devexit_p(plat_nand_remove), + .remove = plat_nand_remove, .driver = { .name = "gen_nand", .owner = THIS_MODULE, diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c index 79ded48e7427..df954b4dcba2 100644 --- a/drivers/mtd/nand/s3c2410.c +++ b/drivers/mtd/nand/s3c2410.c @@ -730,11 +730,14 @@ static int s3c2410_nand_add_partition(struct s3c2410_nand_info *info, struct s3c2410_nand_mtd *mtd, struct s3c2410_nand_set *set) { - if (set) + if (set) { mtd->mtd.name = set->name; - return mtd_device_parse_register(&mtd->mtd, NULL, NULL, + return mtd_device_parse_register(&mtd->mtd, NULL, NULL, set->partitions, set->nr_partitions); + } + + return -ENODEV; } /** diff --git a/drivers/mtd/nand/sh_flctl.c b/drivers/mtd/nand/sh_flctl.c index f48ac5d80bbf..57b3971c9c0a 100644 --- a/drivers/mtd/nand/sh_flctl.c +++ b/drivers/mtd/nand/sh_flctl.c @@ -23,11 +23,18 @@ #include <linux/module.h> #include <linux/kernel.h> +#include <linux/completion.h> #include <linux/delay.h> +#include <linux/dmaengine.h> +#include <linux/dma-mapping.h> #include <linux/interrupt.h> #include <linux/io.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/of_mtd.h> #include <linux/platform_device.h> #include <linux/pm_runtime.h> +#include <linux/sh_dma.h> #include <linux/slab.h> #include <linux/string.h> @@ -106,6 +113,84 @@ static void wait_completion(struct sh_flctl *flctl) writeb(0x0, FLTRCR(flctl)); } +static void flctl_dma_complete(void *param) +{ + struct sh_flctl *flctl = param; + + complete(&flctl->dma_complete); +} + +static void flctl_release_dma(struct sh_flctl *flctl) +{ + if (flctl->chan_fifo0_rx) { + dma_release_channel(flctl->chan_fifo0_rx); + flctl->chan_fifo0_rx = NULL; + } + if (flctl->chan_fifo0_tx) { + dma_release_channel(flctl->chan_fifo0_tx); + flctl->chan_fifo0_tx = NULL; + } +} + +static void flctl_setup_dma(struct sh_flctl *flctl) +{ + dma_cap_mask_t mask; + struct dma_slave_config cfg; + struct platform_device *pdev = flctl->pdev; + struct sh_flctl_platform_data *pdata = pdev->dev.platform_data; + int ret; + + if (!pdata) + return; + + if (pdata->slave_id_fifo0_tx <= 0 || pdata->slave_id_fifo0_rx <= 0) + return; + + /* We can only either use DMA for both Tx and Rx or not use it at all */ + dma_cap_zero(mask); + dma_cap_set(DMA_SLAVE, mask); + + flctl->chan_fifo0_tx = dma_request_channel(mask, shdma_chan_filter, + (void *)pdata->slave_id_fifo0_tx); + dev_dbg(&pdev->dev, "%s: TX: got channel %p\n", __func__, + flctl->chan_fifo0_tx); + + if (!flctl->chan_fifo0_tx) + return; + + memset(&cfg, 0, sizeof(cfg)); + cfg.slave_id = pdata->slave_id_fifo0_tx; + cfg.direction = DMA_MEM_TO_DEV; + cfg.dst_addr = (dma_addr_t)FLDTFIFO(flctl); + cfg.src_addr = 0; + ret = dmaengine_slave_config(flctl->chan_fifo0_tx, &cfg); + if (ret < 0) + goto err; + + flctl->chan_fifo0_rx = dma_request_channel(mask, shdma_chan_filter, + (void *)pdata->slave_id_fifo0_rx); + dev_dbg(&pdev->dev, "%s: RX: got channel %p\n", __func__, + flctl->chan_fifo0_rx); + + if (!flctl->chan_fifo0_rx) + goto err; + + cfg.slave_id = pdata->slave_id_fifo0_rx; + cfg.direction = DMA_DEV_TO_MEM; + cfg.dst_addr = 0; + cfg.src_addr = (dma_addr_t)FLDTFIFO(flctl); + ret = dmaengine_slave_config(flctl->chan_fifo0_rx, &cfg); + if (ret < 0) + goto err; + + init_completion(&flctl->dma_complete); + + return; + +err: + flctl_release_dma(flctl); +} + static void set_addr(struct mtd_info *mtd, int column, int page_addr) { struct sh_flctl *flctl = mtd_to_flctl(mtd); @@ -225,7 +310,7 @@ static enum flctl_ecc_res_t wait_recfifo_ready for (i = 0; i < 3; i++) { uint8_t org; - int index; + unsigned int index; data = readl(ecc_reg[i]); @@ -261,6 +346,70 @@ static void wait_wecfifo_ready(struct sh_flctl *flctl) timeout_error(flctl, __func__); } +static int flctl_dma_fifo0_transfer(struct sh_flctl *flctl, unsigned long *buf, + int len, enum dma_data_direction dir) +{ + struct dma_async_tx_descriptor *desc = NULL; + struct dma_chan *chan; + enum dma_transfer_direction tr_dir; + dma_addr_t dma_addr; + dma_cookie_t cookie = -EINVAL; + uint32_t reg; + int ret; + + if (dir == DMA_FROM_DEVICE) { + chan = flctl->chan_fifo0_rx; + tr_dir = DMA_DEV_TO_MEM; + } else { + chan = flctl->chan_fifo0_tx; + tr_dir = DMA_MEM_TO_DEV; + } + + dma_addr = dma_map_single(chan->device->dev, buf, len, dir); + + if (dma_addr) + desc = dmaengine_prep_slave_single(chan, dma_addr, len, + tr_dir, DMA_PREP_INTERRUPT | DMA_CTRL_ACK); + + if (desc) { + reg = readl(FLINTDMACR(flctl)); + reg |= DREQ0EN; + writel(reg, FLINTDMACR(flctl)); + + desc->callback = flctl_dma_complete; + desc->callback_param = flctl; + cookie = dmaengine_submit(desc); + + dma_async_issue_pending(chan); + } else { + /* DMA failed, fall back to PIO */ + flctl_release_dma(flctl); + dev_warn(&flctl->pdev->dev, + "DMA failed, falling back to PIO\n"); + ret = -EIO; + goto out; + } + + ret = + wait_for_completion_timeout(&flctl->dma_complete, + msecs_to_jiffies(3000)); + + if (ret <= 0) { + chan->device->device_control(chan, DMA_TERMINATE_ALL, 0); + dev_err(&flctl->pdev->dev, "wait_for_completion_timeout\n"); + } + +out: + reg = readl(FLINTDMACR(flctl)); + reg &= ~DREQ0EN; + writel(reg, FLINTDMACR(flctl)); + + dma_unmap_single(chan->device->dev, dma_addr, len, dir); + + /* ret > 0 is success */ + return ret; +} + static void read_datareg(struct sh_flctl *flctl, int offset) { unsigned long data; @@ -279,11 +428,20 @@ static void read_fiforeg(struct sh_flctl *flctl, int rlen, int offset) len_4align = (rlen + 3) / 4; + /* initiate DMA transfer */ + if (flctl->chan_fifo0_rx && rlen >= 32 && + flctl_dma_fifo0_transfer(flctl, buf, rlen, DMA_DEV_TO_MEM) > 0) + goto convert; /* DMA success */ + + /* do polling transfer */ for (i = 0; i < len_4align; i++) { wait_rfifo_ready(flctl); buf[i] = readl(FLDTFIFO(flctl)); - buf[i] = be32_to_cpu(buf[i]); } + +convert: + for (i = 0; i < len_4align; i++) + buf[i] = be32_to_cpu(buf[i]); } static enum flctl_ecc_res_t read_ecfiforeg @@ -305,28 +463,39 @@ static enum flctl_ecc_res_t read_ecfiforeg return res; } -static void write_fiforeg(struct sh_flctl *flctl, int rlen, int offset) +static void write_fiforeg(struct sh_flctl *flctl, int rlen, + unsigned int offset) { int i, len_4align; - unsigned long *data = (unsigned long *)&flctl->done_buff[offset]; - void *fifo_addr = (void *)FLDTFIFO(flctl); + unsigned long *buf = (unsigned long *)&flctl->done_buff[offset]; len_4align = (rlen + 3) / 4; for (i = 0; i < len_4align; i++) { wait_wfifo_ready(flctl); - writel(cpu_to_be32(data[i]), fifo_addr); + writel(cpu_to_be32(buf[i]), FLDTFIFO(flctl)); } } -static void write_ec_fiforeg(struct sh_flctl *flctl, int rlen, int offset) +static void write_ec_fiforeg(struct sh_flctl *flctl, int rlen, + unsigned int offset) { int i, len_4align; - unsigned long *data = (unsigned long *)&flctl->done_buff[offset]; + unsigned long *buf = (unsigned long *)&flctl->done_buff[offset]; len_4align = (rlen + 3) / 4; + + for (i = 0; i < len_4align; i++) + buf[i] = cpu_to_be32(buf[i]); + + /* initiate DMA transfer */ + if (flctl->chan_fifo0_tx && rlen >= 32 && + flctl_dma_fifo0_transfer(flctl, buf, rlen, DMA_MEM_TO_DEV) > 0) + return; /* DMA success */ + + /* do polling transfer */ for (i = 0; i < len_4align; i++) { wait_wecfifo_ready(flctl); - writel(cpu_to_be32(data[i]), FLECFIFO(flctl)); + writel(buf[i], FLECFIFO(flctl)); } } @@ -750,41 +919,35 @@ static void flctl_select_chip(struct mtd_info *mtd, int chipnr) static void flctl_write_buf(struct mtd_info *mtd, const uint8_t *buf, int len) { struct sh_flctl *flctl = mtd_to_flctl(mtd); - int index = flctl->index; - memcpy(&flctl->done_buff[index], buf, len); + memcpy(&flctl->done_buff[flctl->index], buf, len); flctl->index += len; } static uint8_t flctl_read_byte(struct mtd_info *mtd) { struct sh_flctl *flctl = mtd_to_flctl(mtd); - int index = flctl->index; uint8_t data; - data = flctl->done_buff[index]; + data = flctl->done_buff[flctl->index]; flctl->index++; return data; } static uint16_t flctl_read_word(struct mtd_info *mtd) { - struct sh_flctl *flctl = mtd_to_flctl(mtd); - int index = flctl->index; - uint16_t data; - uint16_t *buf = (uint16_t *)&flctl->done_buff[index]; + struct sh_flctl *flctl = mtd_to_flctl(mtd); + uint16_t *buf = (uint16_t *)&flctl->done_buff[flctl->index]; - data = *buf; - flctl->index += 2; - return data; + flctl->index += 2; + return *buf; } static void flctl_read_buf(struct mtd_info *mtd, uint8_t *buf, int len) { struct sh_flctl *flctl = mtd_to_flctl(mtd); - int index = flctl->index; - memcpy(buf, &flctl->done_buff[index], len); + memcpy(buf, &flctl->done_buff[flctl->index], len); flctl->index += len; } @@ -858,7 +1021,74 @@ static irqreturn_t flctl_handle_flste(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit flctl_probe(struct platform_device *pdev) +#ifdef CONFIG_OF +struct flctl_soc_config { + unsigned long flcmncr_val; + unsigned has_hwecc:1; + unsigned use_holden:1; +}; + +static struct flctl_soc_config flctl_sh7372_config = { + .flcmncr_val = CLK_16B_12L_4H | TYPESEL_SET | SHBUSSEL, + .has_hwecc = 1, + .use_holden = 1, +}; + +static const struct of_device_id of_flctl_match[] = { + { .compatible = "renesas,shmobile-flctl-sh7372", + .data = &flctl_sh7372_config }, + {}, +}; +MODULE_DEVICE_TABLE(of, of_flctl_match); + +static struct sh_flctl_platform_data *flctl_parse_dt(struct device *dev) +{ + const struct of_device_id *match; + struct flctl_soc_config *config; + struct sh_flctl_platform_data *pdata; + struct device_node *dn = dev->of_node; + int ret; + + match = of_match_device(of_flctl_match, dev); + if (match) + config = (struct flctl_soc_config *)match->data; + else { + dev_err(dev, "%s: no OF configuration attached\n", __func__); + return NULL; + } + + pdata = devm_kzalloc(dev, sizeof(struct sh_flctl_platform_data), + GFP_KERNEL); + if (!pdata) { + dev_err(dev, "%s: failed to allocate config data\n", __func__); + return NULL; + } + + /* set SoC specific options */ + pdata->flcmncr_val = config->flcmncr_val; + pdata->has_hwecc = config->has_hwecc; + pdata->use_holden = config->use_holden; + + /* parse user defined options */ + ret = of_get_nand_bus_width(dn); + if (ret == 16) + pdata->flcmncr_val |= SEL_16BIT; + else if (ret != 8) { + dev_err(dev, "%s: invalid bus width\n", __func__); + return NULL; + } + + return pdata; +} +#else /* CONFIG_OF */ +#define of_flctl_match NULL +static struct sh_flctl_platform_data *flctl_parse_dt(struct device *dev) +{ + return NULL; +} +#endif /* CONFIG_OF */ + +static int flctl_probe(struct platform_device *pdev) { struct resource *res; struct sh_flctl *flctl; @@ -867,12 +1097,7 @@ static int __devinit flctl_probe(struct platform_device *pdev) struct sh_flctl_platform_data *pdata; int ret = -ENXIO; int irq; - - pdata = pdev->dev.platform_data; - if (pdata == NULL) { - dev_err(&pdev->dev, "no platform data defined\n"); - return -EINVAL; - } + struct mtd_part_parser_data ppdata = {}; flctl = kzalloc(sizeof(struct sh_flctl), GFP_KERNEL); if (!flctl) { @@ -904,6 +1129,17 @@ static int __devinit flctl_probe(struct platform_device *pdev) goto err_flste; } + if (pdev->dev.of_node) + pdata = flctl_parse_dt(&pdev->dev); + else + pdata = pdev->dev.platform_data; + + if (!pdata) { + dev_err(&pdev->dev, "no setup data defined\n"); + ret = -EINVAL; + goto err_pdata; + } + platform_set_drvdata(pdev, flctl); flctl_mtd = &flctl->mtd; nand = &flctl->chip; @@ -932,6 +1168,8 @@ static int __devinit flctl_probe(struct platform_device *pdev) pm_runtime_enable(&pdev->dev); pm_runtime_resume(&pdev->dev); + flctl_setup_dma(flctl); + ret = nand_scan_ident(flctl_mtd, 1, NULL); if (ret) goto err_chip; @@ -944,12 +1182,16 @@ static int __devinit flctl_probe(struct platform_device *pdev) if (ret) goto err_chip; - mtd_device_register(flctl_mtd, pdata->parts, pdata->nr_parts); + ppdata.of_node = pdev->dev.of_node; + ret = mtd_device_parse_register(flctl_mtd, NULL, &ppdata, pdata->parts, + pdata->nr_parts); return 0; err_chip: + flctl_release_dma(flctl); pm_runtime_disable(&pdev->dev); +err_pdata: free_irq(irq, flctl); err_flste: iounmap(flctl->reg); @@ -958,10 +1200,11 @@ err_iomap: return ret; } -static int __devexit flctl_remove(struct platform_device *pdev) +static int flctl_remove(struct platform_device *pdev) { struct sh_flctl *flctl = platform_get_drvdata(pdev); + flctl_release_dma(flctl); nand_release(&flctl->mtd); pm_runtime_disable(&pdev->dev); free_irq(platform_get_irq(pdev, 0), flctl); @@ -976,6 +1219,7 @@ static struct platform_driver flctl_driver = { .driver = { .name = "sh_flctl", .owner = THIS_MODULE, + .of_match_table = of_flctl_match, }, }; diff --git a/drivers/mtd/nand/sharpsl.c b/drivers/mtd/nand/sharpsl.c index 3421e3762a5a..127bc4271821 100644 --- a/drivers/mtd/nand/sharpsl.c +++ b/drivers/mtd/nand/sharpsl.c @@ -106,7 +106,7 @@ static int sharpsl_nand_calculate_ecc(struct mtd_info *mtd, const u_char * dat, /* * Main initialization routine */ -static int __devinit sharpsl_nand_probe(struct platform_device *pdev) +static int sharpsl_nand_probe(struct platform_device *pdev) { struct nand_chip *this; struct resource *r; @@ -205,7 +205,7 @@ err_get_res: /* * Clean up routine */ -static int __devexit sharpsl_nand_remove(struct platform_device *pdev) +static int sharpsl_nand_remove(struct platform_device *pdev) { struct sharpsl_nand *sharpsl = platform_get_drvdata(pdev); @@ -228,7 +228,7 @@ static struct platform_driver sharpsl_nand_driver = { .owner = THIS_MODULE, }, .probe = sharpsl_nand_probe, - .remove = __devexit_p(sharpsl_nand_remove), + .remove = sharpsl_nand_remove, }; module_platform_driver(sharpsl_nand_driver); diff --git a/drivers/mtd/nand/socrates_nand.c b/drivers/mtd/nand/socrates_nand.c index f3f28fafbf7a..09dde7d27178 100644 --- a/drivers/mtd/nand/socrates_nand.c +++ b/drivers/mtd/nand/socrates_nand.c @@ -140,7 +140,7 @@ static int socrates_nand_device_ready(struct mtd_info *mtd) /* * Probe for the NAND device. */ -static int __devinit socrates_nand_probe(struct platform_device *ofdev) +static int socrates_nand_probe(struct platform_device *ofdev) { struct socrates_nand_host *host; struct mtd_info *mtd; @@ -220,7 +220,7 @@ out: /* * Remove a NAND device. */ -static int __devexit socrates_nand_remove(struct platform_device *ofdev) +static int socrates_nand_remove(struct platform_device *ofdev) { struct socrates_nand_host *host = dev_get_drvdata(&ofdev->dev); struct mtd_info *mtd = &host->mtd; @@ -251,7 +251,7 @@ static struct platform_driver socrates_nand_driver = { .of_match_table = socrates_nand_match, }, .probe = socrates_nand_probe, - .remove = __devexit_p(socrates_nand_remove), + .remove = socrates_nand_remove, }; module_platform_driver(socrates_nand_driver); diff --git a/drivers/mtd/ofpart.c b/drivers/mtd/ofpart.c index d9127e2ed808..dbd3aa574eaf 100644 --- a/drivers/mtd/ofpart.c +++ b/drivers/mtd/ofpart.c @@ -71,7 +71,10 @@ static int parse_ofpart_partitions(struct mtd_info *master, (*pparts)[i].name = (char *)partname; if (of_get_property(pp, "read-only", &len)) - (*pparts)[i].mask_flags = MTD_WRITEABLE; + (*pparts)[i].mask_flags |= MTD_WRITEABLE; + + if (of_get_property(pp, "lock", &len)) + (*pparts)[i].mask_flags |= MTD_POWERUP_LOCK; i++; } diff --git a/drivers/mtd/onenand/generic.c b/drivers/mtd/onenand/generic.c index 1c4f97c63e62..9f11562f849d 100644 --- a/drivers/mtd/onenand/generic.c +++ b/drivers/mtd/onenand/generic.c @@ -35,7 +35,7 @@ struct onenand_info { struct onenand_chip onenand; }; -static int __devinit generic_onenand_probe(struct platform_device *pdev) +static int generic_onenand_probe(struct platform_device *pdev) { struct onenand_info *info; struct onenand_platform_data *pdata = pdev->dev.platform_data; @@ -88,7 +88,7 @@ out_free_info: return err; } -static int __devexit generic_onenand_remove(struct platform_device *pdev) +static int generic_onenand_remove(struct platform_device *pdev) { struct onenand_info *info = platform_get_drvdata(pdev); struct resource *res = pdev->resource; @@ -112,7 +112,7 @@ static struct platform_driver generic_onenand_driver = { .owner = THIS_MODULE, }, .probe = generic_onenand_probe, - .remove = __devexit_p(generic_onenand_remove), + .remove = generic_onenand_remove, }; module_platform_driver(generic_onenand_driver); diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c index 00cd3da29435..065f3fe02a2f 100644 --- a/drivers/mtd/onenand/omap2.c +++ b/drivers/mtd/onenand/omap2.c @@ -630,7 +630,7 @@ static int omap2_onenand_disable(struct mtd_info *mtd) return ret; } -static int __devinit omap2_onenand_probe(struct platform_device *pdev) +static int omap2_onenand_probe(struct platform_device *pdev) { struct omap_onenand_platform_data *pdata; struct omap2_onenand *c; @@ -799,7 +799,7 @@ err_kfree: return r; } -static int __devexit omap2_onenand_remove(struct platform_device *pdev) +static int omap2_onenand_remove(struct platform_device *pdev) { struct omap2_onenand *c = dev_get_drvdata(&pdev->dev); @@ -822,7 +822,7 @@ static int __devexit omap2_onenand_remove(struct platform_device *pdev) static struct platform_driver omap2_onenand_driver = { .probe = omap2_onenand_probe, - .remove = __devexit_p(omap2_onenand_remove), + .remove = omap2_onenand_remove, .shutdown = omap2_onenand_shutdown, .driver = { .name = DRIVER_NAME, diff --git a/drivers/mtd/onenand/samsung.c b/drivers/mtd/onenand/samsung.c index 8e4b3f2742ba..33f2a8fb8df9 100644 --- a/drivers/mtd/onenand/samsung.c +++ b/drivers/mtd/onenand/samsung.c @@ -1053,7 +1053,7 @@ onenand_fail: return err; } -static int __devexit s3c_onenand_remove(struct platform_device *pdev) +static int s3c_onenand_remove(struct platform_device *pdev) { struct mtd_info *mtd = platform_get_drvdata(pdev); @@ -1130,7 +1130,7 @@ static struct platform_driver s3c_onenand_driver = { }, .id_table = s3c_onenand_driver_ids, .probe = s3c_onenand_probe, - .remove = __devexit_p(s3c_onenand_remove), + .remove = s3c_onenand_remove, }; module_platform_driver(s3c_onenand_driver); diff --git a/drivers/mtd/tests/mtd_nandbiterrs.c b/drivers/mtd/tests/mtd_nandbiterrs.c index cc8d62cb280c..207bf9a9972f 100644 --- a/drivers/mtd/tests/mtd_nandbiterrs.c +++ b/drivers/mtd/tests/mtd_nandbiterrs.c @@ -39,6 +39,9 @@ * this program; see the file COPYING. If not, write to the Free Software * Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/init.h> #include <linux/module.h> #include <linux/moduleparam.h> @@ -47,8 +50,6 @@ #include <linux/mtd/nand.h> #include <linux/slab.h> -#define msg(FMT, VA...) pr_info("mtd_nandbiterrs: "FMT, ##VA) - static int dev; module_param(dev, int, S_IRUGO); MODULE_PARM_DESC(dev, "MTD device number to use"); @@ -103,7 +104,7 @@ static int erase_block(void) struct erase_info ei; loff_t addr = eraseblock * mtd->erasesize; - msg("erase_block\n"); + pr_info("erase_block\n"); memset(&ei, 0, sizeof(struct erase_info)); ei.mtd = mtd; @@ -112,7 +113,7 @@ static int erase_block(void) err = mtd_erase(mtd, &ei); if (err || ei.state == MTD_ERASE_FAILED) { - msg("error %d while erasing\n", err); + pr_err("error %d while erasing\n", err); if (!err) err = -EIO; return err; @@ -128,11 +129,11 @@ static int write_page(int log) size_t written; if (log) - msg("write_page\n"); + pr_info("write_page\n"); err = mtd_write(mtd, offset, mtd->writesize, &written, wbuffer); if (err || written != mtd->writesize) { - msg("error: write failed at %#llx\n", (long long)offset); + pr_err("error: write failed at %#llx\n", (long long)offset); if (!err) err = -EIO; } @@ -147,7 +148,7 @@ static int rewrite_page(int log) struct mtd_oob_ops ops; if (log) - msg("rewrite page\n"); + pr_info("rewrite page\n"); ops.mode = MTD_OPS_RAW; /* No ECC */ ops.len = mtd->writesize; @@ -160,7 +161,7 @@ static int rewrite_page(int log) err = mtd_write_oob(mtd, offset, &ops); if (err || ops.retlen != mtd->writesize) { - msg("error: write_oob failed (%d)\n", err); + pr_err("error: write_oob failed (%d)\n", err); if (!err) err = -EIO; } @@ -177,7 +178,7 @@ static int read_page(int log) struct mtd_ecc_stats oldstats; if (log) - msg("read_page\n"); + pr_info("read_page\n"); /* Saving last mtd stats */ memcpy(&oldstats, &mtd->ecc_stats, sizeof(oldstats)); @@ -187,7 +188,7 @@ static int read_page(int log) err = mtd->ecc_stats.corrected - oldstats.corrected; if (err < 0 || read != mtd->writesize) { - msg("error: read failed at %#llx\n", (long long)offset); + pr_err("error: read failed at %#llx\n", (long long)offset); if (err >= 0) err = -EIO; } @@ -201,11 +202,11 @@ static int verify_page(int log) unsigned i, errs = 0; if (log) - msg("verify_page\n"); + pr_info("verify_page\n"); for (i = 0; i < mtd->writesize; i++) { if (rbuffer[i] != hash(i+seed)) { - msg("Error: page offset %u, expected %02x, got %02x\n", + pr_err("Error: page offset %u, expected %02x, got %02x\n", i, hash(i+seed), rbuffer[i]); errs++; } @@ -230,13 +231,13 @@ static int insert_biterror(unsigned byte) for (bit = 7; bit >= 0; bit--) { if (CBIT(wbuffer[byte], bit)) { BCLR(wbuffer[byte], bit); - msg("Inserted biterror @ %u/%u\n", byte, bit); + pr_info("Inserted biterror @ %u/%u\n", byte, bit); return 0; } } byte++; } - msg("biterror: Failed to find a '1' bit\n"); + pr_err("biterror: Failed to find a '1' bit\n"); return -EIO; } @@ -248,7 +249,7 @@ static int incremental_errors_test(void) unsigned i; unsigned errs_per_subpage = 0; - msg("incremental biterrors test\n"); + pr_info("incremental biterrors test\n"); for (i = 0; i < mtd->writesize; i++) wbuffer[i] = hash(i+seed); @@ -265,9 +266,9 @@ static int incremental_errors_test(void) err = read_page(1); if (err > 0) - msg("Read reported %d corrected bit errors\n", err); + pr_info("Read reported %d corrected bit errors\n", err); if (err < 0) { - msg("After %d biterrors per subpage, read reported error %d\n", + pr_err("After %d biterrors per subpage, read reported error %d\n", errs_per_subpage, err); err = 0; goto exit; @@ -275,11 +276,11 @@ static int incremental_errors_test(void) err = verify_page(1); if (err) { - msg("ECC failure, read data is incorrect despite read success\n"); + pr_err("ECC failure, read data is incorrect despite read success\n"); goto exit; } - msg("Successfully corrected %d bit errors per subpage\n", + pr_info("Successfully corrected %d bit errors per subpage\n", errs_per_subpage); for (i = 0; i < subcount; i++) { @@ -311,7 +312,7 @@ static int overwrite_test(void) memset(bitstats, 0, sizeof(bitstats)); - msg("overwrite biterrors test\n"); + pr_info("overwrite biterrors test\n"); for (i = 0; i < mtd->writesize; i++) wbuffer[i] = hash(i+seed); @@ -329,18 +330,18 @@ static int overwrite_test(void) err = read_page(0); if (err >= 0) { if (err >= MAXBITS) { - msg("Implausible number of bit errors corrected\n"); + pr_info("Implausible number of bit errors corrected\n"); err = -EIO; break; } bitstats[err]++; if (err > max_corrected) { max_corrected = err; - msg("Read reported %d corrected bit errors\n", + pr_info("Read reported %d corrected bit errors\n", err); } } else { /* err < 0 */ - msg("Read reported error %d\n", err); + pr_info("Read reported error %d\n", err); err = 0; break; } @@ -348,7 +349,7 @@ static int overwrite_test(void) err = verify_page(0); if (err) { bitstats[max_corrected] = opno; - msg("ECC failure, read data is incorrect despite read success\n"); + pr_info("ECC failure, read data is incorrect despite read success\n"); break; } @@ -357,9 +358,9 @@ static int overwrite_test(void) /* At this point bitstats[0] contains the number of ops with no bit * errors, bitstats[1] the number of ops with 1 bit error, etc. */ - msg("Bit error histogram (%d operations total):\n", opno); + pr_info("Bit error histogram (%d operations total):\n", opno); for (i = 0; i < max_corrected; i++) - msg("Page reads with %3d corrected bit errors: %d\n", + pr_info("Page reads with %3d corrected bit errors: %d\n", i, bitstats[i]); exit: @@ -370,36 +371,36 @@ static int __init mtd_nandbiterrs_init(void) { int err = 0; - msg("\n"); - msg("==================================================\n"); - msg("MTD device: %d\n", dev); + printk("\n"); + printk(KERN_INFO "==================================================\n"); + pr_info("MTD device: %d\n", dev); mtd = get_mtd_device(NULL, dev); if (IS_ERR(mtd)) { err = PTR_ERR(mtd); - msg("error: cannot get MTD device\n"); + pr_err("error: cannot get MTD device\n"); goto exit_mtddev; } if (mtd->type != MTD_NANDFLASH) { - msg("this test requires NAND flash\n"); + pr_info("this test requires NAND flash\n"); err = -ENODEV; goto exit_nand; } - msg("MTD device size %llu, eraseblock=%u, page=%u, oob=%u\n", + pr_info("MTD device size %llu, eraseblock=%u, page=%u, oob=%u\n", (unsigned long long)mtd->size, mtd->erasesize, mtd->writesize, mtd->oobsize); subsize = mtd->writesize >> mtd->subpage_sft; subcount = mtd->writesize / subsize; - msg("Device uses %d subpages of %d bytes\n", subcount, subsize); + pr_info("Device uses %d subpages of %d bytes\n", subcount, subsize); offset = page_offset * mtd->writesize; eraseblock = mtd_div_by_eb(offset, mtd); - msg("Using page=%u, offset=%llu, eraseblock=%u\n", + pr_info("Using page=%u, offset=%llu, eraseblock=%u\n", page_offset, offset, eraseblock); wbuffer = kmalloc(mtd->writesize, GFP_KERNEL); @@ -432,8 +433,8 @@ static int __init mtd_nandbiterrs_init(void) goto exit_error; err = -EIO; - msg("finished successfully.\n"); - msg("==================================================\n"); + pr_info("finished successfully.\n"); + printk(KERN_INFO "==================================================\n"); exit_error: kfree(rbuffer); diff --git a/drivers/mtd/tests/mtd_nandecctest.c b/drivers/mtd/tests/mtd_nandecctest.c index b437fa425077..1eee264509a8 100644 --- a/drivers/mtd/tests/mtd_nandecctest.c +++ b/drivers/mtd/tests/mtd_nandecctest.c @@ -1,3 +1,5 @@ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/kernel.h> #include <linux/module.h> #include <linux/list.h> @@ -264,13 +266,13 @@ static int nand_ecc_test_run(const size_t size) correct_data, size); if (err) { - pr_err("mtd_nandecctest: not ok - %s-%zd\n", + pr_err("not ok - %s-%zd\n", nand_ecc_test[i].name, size); dump_data_ecc(error_data, error_ecc, correct_data, correct_ecc, size); break; } - pr_info("mtd_nandecctest: ok - %s-%zd\n", + pr_info("ok - %s-%zd\n", nand_ecc_test[i].name, size); } error: diff --git a/drivers/mtd/tests/mtd_oobtest.c b/drivers/mtd/tests/mtd_oobtest.c index ed9b62827f1b..e827fa8cd844 100644 --- a/drivers/mtd/tests/mtd_oobtest.c +++ b/drivers/mtd/tests/mtd_oobtest.c @@ -19,6 +19,8 @@ * Author: Adrian Hunter <ext-adrian.hunter@nokia.com> */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <asm/div64.h> #include <linux/init.h> #include <linux/module.h> @@ -28,8 +30,6 @@ #include <linux/slab.h> #include <linux/sched.h> -#define PRINT_PREF KERN_INFO "mtd_oobtest: " - static int dev = -EINVAL; module_param(dev, int, S_IRUGO); MODULE_PARM_DESC(dev, "MTD device number to use"); @@ -80,13 +80,12 @@ static int erase_eraseblock(int ebnum) err = mtd_erase(mtd, &ei); if (err) { - printk(PRINT_PREF "error %d while erasing EB %d\n", err, ebnum); + pr_err("error %d while erasing EB %d\n", err, ebnum); return err; } if (ei.state == MTD_ERASE_FAILED) { - printk(PRINT_PREF "some erase error occurred at EB %d\n", - ebnum); + pr_err("some erase error occurred at EB %d\n", ebnum); return -EIO; } @@ -98,7 +97,7 @@ static int erase_whole_device(void) int err; unsigned int i; - printk(PRINT_PREF "erasing whole device\n"); + pr_info("erasing whole device\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -107,7 +106,7 @@ static int erase_whole_device(void) return err; cond_resched(); } - printk(PRINT_PREF "erased %u eraseblocks\n", i); + pr_info("erased %u eraseblocks\n", i); return 0; } @@ -141,9 +140,9 @@ static int write_eraseblock(int ebnum) ops.oobbuf = writebuf; err = mtd_write_oob(mtd, addr, &ops); if (err || ops.oobretlen != use_len) { - printk(PRINT_PREF "error: writeoob failed at %#llx\n", + pr_err("error: writeoob failed at %#llx\n", (long long)addr); - printk(PRINT_PREF "error: use_len %d, use_offset %d\n", + pr_err("error: use_len %d, use_offset %d\n", use_len, use_offset); errcnt += 1; return err ? err : -1; @@ -160,7 +159,7 @@ static int write_whole_device(void) int err; unsigned int i; - printk(PRINT_PREF "writing OOBs of whole device\n"); + pr_info("writing OOBs of whole device\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -168,10 +167,10 @@ static int write_whole_device(void) if (err) return err; if (i % 256 == 0) - printk(PRINT_PREF "written up to eraseblock %u\n", i); + pr_info("written up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "written %u eraseblocks\n", i); + pr_info("written %u eraseblocks\n", i); return 0; } @@ -194,17 +193,17 @@ static int verify_eraseblock(int ebnum) ops.oobbuf = readbuf; err = mtd_read_oob(mtd, addr, &ops); if (err || ops.oobretlen != use_len) { - printk(PRINT_PREF "error: readoob failed at %#llx\n", + pr_err("error: readoob failed at %#llx\n", (long long)addr); errcnt += 1; return err ? err : -1; } if (memcmp(readbuf, writebuf, use_len)) { - printk(PRINT_PREF "error: verify failed at %#llx\n", + pr_err("error: verify failed at %#llx\n", (long long)addr); errcnt += 1; if (errcnt > 1000) { - printk(PRINT_PREF "error: too many errors\n"); + pr_err("error: too many errors\n"); return -1; } } @@ -221,29 +220,28 @@ static int verify_eraseblock(int ebnum) ops.oobbuf = readbuf; err = mtd_read_oob(mtd, addr, &ops); if (err || ops.oobretlen != mtd->ecclayout->oobavail) { - printk(PRINT_PREF "error: readoob failed at " - "%#llx\n", (long long)addr); + pr_err("error: readoob failed at %#llx\n", + (long long)addr); errcnt += 1; return err ? err : -1; } if (memcmp(readbuf + use_offset, writebuf, use_len)) { - printk(PRINT_PREF "error: verify failed at " - "%#llx\n", (long long)addr); + pr_err("error: verify failed at %#llx\n", + (long long)addr); errcnt += 1; if (errcnt > 1000) { - printk(PRINT_PREF "error: too many " - "errors\n"); + pr_err("error: too many errors\n"); return -1; } } for (k = 0; k < use_offset; ++k) if (readbuf[k] != 0xff) { - printk(PRINT_PREF "error: verify 0xff " + pr_err("error: verify 0xff " "failed at %#llx\n", (long long)addr); errcnt += 1; if (errcnt > 1000) { - printk(PRINT_PREF "error: too " + pr_err("error: too " "many errors\n"); return -1; } @@ -251,12 +249,12 @@ static int verify_eraseblock(int ebnum) for (k = use_offset + use_len; k < mtd->ecclayout->oobavail; ++k) if (readbuf[k] != 0xff) { - printk(PRINT_PREF "error: verify 0xff " + pr_err("error: verify 0xff " "failed at %#llx\n", (long long)addr); errcnt += 1; if (errcnt > 1000) { - printk(PRINT_PREF "error: too " + pr_err("error: too " "many errors\n"); return -1; } @@ -286,17 +284,17 @@ static int verify_eraseblock_in_one_go(int ebnum) ops.oobbuf = readbuf; err = mtd_read_oob(mtd, addr, &ops); if (err || ops.oobretlen != len) { - printk(PRINT_PREF "error: readoob failed at %#llx\n", + pr_err("error: readoob failed at %#llx\n", (long long)addr); errcnt += 1; return err ? err : -1; } if (memcmp(readbuf, writebuf, len)) { - printk(PRINT_PREF "error: verify failed at %#llx\n", + pr_err("error: verify failed at %#llx\n", (long long)addr); errcnt += 1; if (errcnt > 1000) { - printk(PRINT_PREF "error: too many errors\n"); + pr_err("error: too many errors\n"); return -1; } } @@ -309,7 +307,7 @@ static int verify_all_eraseblocks(void) int err; unsigned int i; - printk(PRINT_PREF "verifying all eraseblocks\n"); + pr_info("verifying all eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -317,10 +315,10 @@ static int verify_all_eraseblocks(void) if (err) return err; if (i % 256 == 0) - printk(PRINT_PREF "verified up to eraseblock %u\n", i); + pr_info("verified up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "verified %u eraseblocks\n", i); + pr_info("verified %u eraseblocks\n", i); return 0; } @@ -331,7 +329,7 @@ static int is_block_bad(int ebnum) ret = mtd_block_isbad(mtd, addr); if (ret) - printk(PRINT_PREF "block %d is bad\n", ebnum); + pr_info("block %d is bad\n", ebnum); return ret; } @@ -341,18 +339,18 @@ static int scan_for_bad_eraseblocks(void) bbt = kmalloc(ebcnt, GFP_KERNEL); if (!bbt) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); return -ENOMEM; } - printk(PRINT_PREF "scanning for bad eraseblocks\n"); + pr_info("scanning for bad eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { bbt[i] = is_block_bad(i) ? 1 : 0; if (bbt[i]) bad += 1; cond_resched(); } - printk(PRINT_PREF "scanned %d eraseblocks, %d are bad\n", i, bad); + pr_info("scanned %d eraseblocks, %d are bad\n", i, bad); return 0; } @@ -368,22 +366,22 @@ static int __init mtd_oobtest_init(void) printk(KERN_INFO "=================================================\n"); if (dev < 0) { - printk(PRINT_PREF "Please specify a valid mtd-device via module paramter\n"); - printk(KERN_CRIT "CAREFUL: This test wipes all data on the specified MTD device!\n"); + pr_info("Please specify a valid mtd-device via module parameter\n"); + pr_crit("CAREFUL: This test wipes all data on the specified MTD device!\n"); return -EINVAL; } - printk(PRINT_PREF "MTD device: %d\n", dev); + pr_info("MTD device: %d\n", dev); mtd = get_mtd_device(NULL, dev); if (IS_ERR(mtd)) { err = PTR_ERR(mtd); - printk(PRINT_PREF "error: cannot get MTD device\n"); + pr_err("error: cannot get MTD device\n"); return err; } if (mtd->type != MTD_NANDFLASH) { - printk(PRINT_PREF "this test requires NAND flash\n"); + pr_info("this test requires NAND flash\n"); goto out; } @@ -392,7 +390,7 @@ static int __init mtd_oobtest_init(void) ebcnt = tmp; pgcnt = mtd->erasesize / mtd->writesize; - printk(PRINT_PREF "MTD device size %llu, eraseblock size %u, " + pr_info("MTD device size %llu, eraseblock size %u, " "page size %u, count of eraseblocks %u, pages per " "eraseblock %u, OOB size %u\n", (unsigned long long)mtd->size, mtd->erasesize, @@ -401,12 +399,12 @@ static int __init mtd_oobtest_init(void) err = -ENOMEM; readbuf = kmalloc(mtd->erasesize, GFP_KERNEL); if (!readbuf) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } writebuf = kmalloc(mtd->erasesize, GFP_KERNEL); if (!writebuf) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } @@ -420,7 +418,7 @@ static int __init mtd_oobtest_init(void) vary_offset = 0; /* First test: write all OOB, read it back and verify */ - printk(PRINT_PREF "test 1 of 5\n"); + pr_info("test 1 of 5\n"); err = erase_whole_device(); if (err) @@ -440,7 +438,7 @@ static int __init mtd_oobtest_init(void) * Second test: write all OOB, a block at a time, read it back and * verify. */ - printk(PRINT_PREF "test 2 of 5\n"); + pr_info("test 2 of 5\n"); err = erase_whole_device(); if (err) @@ -453,7 +451,7 @@ static int __init mtd_oobtest_init(void) /* Check all eraseblocks */ simple_srand(3); - printk(PRINT_PREF "verifying all eraseblocks\n"); + pr_info("verifying all eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -461,16 +459,16 @@ static int __init mtd_oobtest_init(void) if (err) goto out; if (i % 256 == 0) - printk(PRINT_PREF "verified up to eraseblock %u\n", i); + pr_info("verified up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "verified %u eraseblocks\n", i); + pr_info("verified %u eraseblocks\n", i); /* * Third test: write OOB at varying offsets and lengths, read it back * and verify. */ - printk(PRINT_PREF "test 3 of 5\n"); + pr_info("test 3 of 5\n"); err = erase_whole_device(); if (err) @@ -503,7 +501,7 @@ static int __init mtd_oobtest_init(void) vary_offset = 0; /* Fourth test: try to write off end of device */ - printk(PRINT_PREF "test 4 of 5\n"); + pr_info("test 4 of 5\n"); err = erase_whole_device(); if (err) @@ -522,14 +520,14 @@ static int __init mtd_oobtest_init(void) ops.ooboffs = mtd->ecclayout->oobavail; ops.datbuf = NULL; ops.oobbuf = writebuf; - printk(PRINT_PREF "attempting to start write past end of OOB\n"); - printk(PRINT_PREF "an error is expected...\n"); + pr_info("attempting to start write past end of OOB\n"); + pr_info("an error is expected...\n"); err = mtd_write_oob(mtd, addr0, &ops); if (err) { - printk(PRINT_PREF "error occurred as expected\n"); + pr_info("error occurred as expected\n"); err = 0; } else { - printk(PRINT_PREF "error: can write past end of OOB\n"); + pr_err("error: can write past end of OOB\n"); errcnt += 1; } @@ -542,19 +540,19 @@ static int __init mtd_oobtest_init(void) ops.ooboffs = mtd->ecclayout->oobavail; ops.datbuf = NULL; ops.oobbuf = readbuf; - printk(PRINT_PREF "attempting to start read past end of OOB\n"); - printk(PRINT_PREF "an error is expected...\n"); + pr_info("attempting to start read past end of OOB\n"); + pr_info("an error is expected...\n"); err = mtd_read_oob(mtd, addr0, &ops); if (err) { - printk(PRINT_PREF "error occurred as expected\n"); + pr_info("error occurred as expected\n"); err = 0; } else { - printk(PRINT_PREF "error: can read past end of OOB\n"); + pr_err("error: can read past end of OOB\n"); errcnt += 1; } if (bbt[ebcnt - 1]) - printk(PRINT_PREF "skipping end of device tests because last " + pr_info("skipping end of device tests because last " "block is bad\n"); else { /* Attempt to write off end of device */ @@ -566,14 +564,14 @@ static int __init mtd_oobtest_init(void) ops.ooboffs = 0; ops.datbuf = NULL; ops.oobbuf = writebuf; - printk(PRINT_PREF "attempting to write past end of device\n"); - printk(PRINT_PREF "an error is expected...\n"); + pr_info("attempting to write past end of device\n"); + pr_info("an error is expected...\n"); err = mtd_write_oob(mtd, mtd->size - mtd->writesize, &ops); if (err) { - printk(PRINT_PREF "error occurred as expected\n"); + pr_info("error occurred as expected\n"); err = 0; } else { - printk(PRINT_PREF "error: wrote past end of device\n"); + pr_err("error: wrote past end of device\n"); errcnt += 1; } @@ -586,14 +584,14 @@ static int __init mtd_oobtest_init(void) ops.ooboffs = 0; ops.datbuf = NULL; ops.oobbuf = readbuf; - printk(PRINT_PREF "attempting to read past end of device\n"); - printk(PRINT_PREF "an error is expected...\n"); + pr_info("attempting to read past end of device\n"); + pr_info("an error is expected...\n"); err = mtd_read_oob(mtd, mtd->size - mtd->writesize, &ops); if (err) { - printk(PRINT_PREF "error occurred as expected\n"); + pr_info("error occurred as expected\n"); err = 0; } else { - printk(PRINT_PREF "error: read past end of device\n"); + pr_err("error: read past end of device\n"); errcnt += 1; } @@ -610,14 +608,14 @@ static int __init mtd_oobtest_init(void) ops.ooboffs = 1; ops.datbuf = NULL; ops.oobbuf = writebuf; - printk(PRINT_PREF "attempting to write past end of device\n"); - printk(PRINT_PREF "an error is expected...\n"); + pr_info("attempting to write past end of device\n"); + pr_info("an error is expected...\n"); err = mtd_write_oob(mtd, mtd->size - mtd->writesize, &ops); if (err) { - printk(PRINT_PREF "error occurred as expected\n"); + pr_info("error occurred as expected\n"); err = 0; } else { - printk(PRINT_PREF "error: wrote past end of device\n"); + pr_err("error: wrote past end of device\n"); errcnt += 1; } @@ -630,20 +628,20 @@ static int __init mtd_oobtest_init(void) ops.ooboffs = 1; ops.datbuf = NULL; ops.oobbuf = readbuf; - printk(PRINT_PREF "attempting to read past end of device\n"); - printk(PRINT_PREF "an error is expected...\n"); + pr_info("attempting to read past end of device\n"); + pr_info("an error is expected...\n"); err = mtd_read_oob(mtd, mtd->size - mtd->writesize, &ops); if (err) { - printk(PRINT_PREF "error occurred as expected\n"); + pr_info("error occurred as expected\n"); err = 0; } else { - printk(PRINT_PREF "error: read past end of device\n"); + pr_err("error: read past end of device\n"); errcnt += 1; } } /* Fifth test: write / read across block boundaries */ - printk(PRINT_PREF "test 5 of 5\n"); + pr_info("test 5 of 5\n"); /* Erase all eraseblocks */ err = erase_whole_device(); @@ -652,7 +650,7 @@ static int __init mtd_oobtest_init(void) /* Write all eraseblocks */ simple_srand(11); - printk(PRINT_PREF "writing OOBs of whole device\n"); + pr_info("writing OOBs of whole device\n"); for (i = 0; i < ebcnt - 1; ++i) { int cnt = 2; int pg; @@ -674,17 +672,16 @@ static int __init mtd_oobtest_init(void) if (err) goto out; if (i % 256 == 0) - printk(PRINT_PREF "written up to eraseblock " - "%u\n", i); + pr_info("written up to eraseblock %u\n", i); cond_resched(); addr += mtd->writesize; } } - printk(PRINT_PREF "written %u eraseblocks\n", i); + pr_info("written %u eraseblocks\n", i); /* Check all eraseblocks */ simple_srand(11); - printk(PRINT_PREF "verifying all eraseblocks\n"); + pr_info("verifying all eraseblocks\n"); for (i = 0; i < ebcnt - 1; ++i) { if (bbt[i] || bbt[i + 1]) continue; @@ -702,28 +699,28 @@ static int __init mtd_oobtest_init(void) if (err) goto out; if (memcmp(readbuf, writebuf, mtd->ecclayout->oobavail * 2)) { - printk(PRINT_PREF "error: verify failed at %#llx\n", + pr_err("error: verify failed at %#llx\n", (long long)addr); errcnt += 1; if (errcnt > 1000) { - printk(PRINT_PREF "error: too many errors\n"); + pr_err("error: too many errors\n"); goto out; } } if (i % 256 == 0) - printk(PRINT_PREF "verified up to eraseblock %u\n", i); + pr_info("verified up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "verified %u eraseblocks\n", i); + pr_info("verified %u eraseblocks\n", i); - printk(PRINT_PREF "finished with %d errors\n", errcnt); + pr_info("finished with %d errors\n", errcnt); out: kfree(bbt); kfree(writebuf); kfree(readbuf); put_mtd_device(mtd); if (err) - printk(PRINT_PREF "error %d occurred\n", err); + pr_info("error %d occurred\n", err); printk(KERN_INFO "=================================================\n"); return err; } diff --git a/drivers/mtd/tests/mtd_pagetest.c b/drivers/mtd/tests/mtd_pagetest.c index 252ddb092fb2..f93a76f88113 100644 --- a/drivers/mtd/tests/mtd_pagetest.c +++ b/drivers/mtd/tests/mtd_pagetest.c @@ -19,6 +19,8 @@ * Author: Adrian Hunter <ext-adrian.hunter@nokia.com> */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <asm/div64.h> #include <linux/init.h> #include <linux/module.h> @@ -28,8 +30,6 @@ #include <linux/slab.h> #include <linux/sched.h> -#define PRINT_PREF KERN_INFO "mtd_pagetest: " - static int dev = -EINVAL; module_param(dev, int, S_IRUGO); MODULE_PARM_DESC(dev, "MTD device number to use"); @@ -79,12 +79,12 @@ static int erase_eraseblock(int ebnum) err = mtd_erase(mtd, &ei); if (err) { - printk(PRINT_PREF "error %d while erasing EB %d\n", err, ebnum); + pr_err("error %d while erasing EB %d\n", err, ebnum); return err; } if (ei.state == MTD_ERASE_FAILED) { - printk(PRINT_PREF "some erase error occurred at EB %d\n", + pr_err("some erase error occurred at EB %d\n", ebnum); return -EIO; } @@ -102,7 +102,7 @@ static int write_eraseblock(int ebnum) cond_resched(); err = mtd_write(mtd, addr, mtd->erasesize, &written, writebuf); if (err || written != mtd->erasesize) - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", (long long)addr); return err; @@ -131,7 +131,7 @@ static int verify_eraseblock(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != bufsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr0); return err; } @@ -139,7 +139,7 @@ static int verify_eraseblock(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != bufsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)(addrn - bufsize)); return err; } @@ -148,12 +148,12 @@ static int verify_eraseblock(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != bufsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); break; } if (memcmp(twopages, writebuf + (j * pgsize), bufsize)) { - printk(PRINT_PREF "error: verify failed at %#llx\n", + pr_err("error: verify failed at %#llx\n", (long long)addr); errcnt += 1; } @@ -166,7 +166,7 @@ static int verify_eraseblock(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != bufsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr0); return err; } @@ -174,7 +174,7 @@ static int verify_eraseblock(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != bufsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)(addrn - bufsize)); return err; } @@ -183,14 +183,14 @@ static int verify_eraseblock(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != bufsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); return err; } memcpy(boundary, writebuf + mtd->erasesize - pgsize, pgsize); set_random_data(boundary + pgsize, pgsize); if (memcmp(twopages, boundary, bufsize)) { - printk(PRINT_PREF "error: verify failed at %#llx\n", + pr_err("error: verify failed at %#llx\n", (long long)addr); errcnt += 1; } @@ -206,10 +206,10 @@ static int crosstest(void) loff_t addr, addr0, addrn; unsigned char *pp1, *pp2, *pp3, *pp4; - printk(PRINT_PREF "crosstest\n"); + pr_info("crosstest\n"); pp1 = kmalloc(pgsize * 4, GFP_KERNEL); if (!pp1) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); return -ENOMEM; } pp2 = pp1 + pgsize; @@ -231,7 +231,7 @@ static int crosstest(void) if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); kfree(pp1); return err; @@ -243,7 +243,7 @@ static int crosstest(void) if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); kfree(pp1); return err; @@ -251,12 +251,12 @@ static int crosstest(void) /* Read first page to pp2 */ addr = addr0; - printk(PRINT_PREF "reading page at %#llx\n", (long long)addr); + pr_info("reading page at %#llx\n", (long long)addr); err = mtd_read(mtd, addr, pgsize, &read, pp2); if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); kfree(pp1); return err; @@ -264,12 +264,12 @@ static int crosstest(void) /* Read last page to pp3 */ addr = addrn - pgsize; - printk(PRINT_PREF "reading page at %#llx\n", (long long)addr); + pr_info("reading page at %#llx\n", (long long)addr); err = mtd_read(mtd, addr, pgsize, &read, pp3); if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); kfree(pp1); return err; @@ -277,25 +277,25 @@ static int crosstest(void) /* Read first page again to pp4 */ addr = addr0; - printk(PRINT_PREF "reading page at %#llx\n", (long long)addr); + pr_info("reading page at %#llx\n", (long long)addr); err = mtd_read(mtd, addr, pgsize, &read, pp4); if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); kfree(pp1); return err; } /* pp2 and pp4 should be the same */ - printk(PRINT_PREF "verifying pages read at %#llx match\n", + pr_info("verifying pages read at %#llx match\n", (long long)addr0); if (memcmp(pp2, pp4, pgsize)) { - printk(PRINT_PREF "verify failed!\n"); + pr_err("verify failed!\n"); errcnt += 1; } else if (!err) - printk(PRINT_PREF "crosstest ok\n"); + pr_info("crosstest ok\n"); kfree(pp1); return err; } @@ -307,7 +307,7 @@ static int erasecrosstest(void) loff_t addr0; char *readbuf = twopages; - printk(PRINT_PREF "erasecrosstest\n"); + pr_info("erasecrosstest\n"); ebnum = 0; addr0 = 0; @@ -320,79 +320,79 @@ static int erasecrosstest(void) while (ebnum2 && bbt[ebnum2]) ebnum2 -= 1; - printk(PRINT_PREF "erasing block %d\n", ebnum); + pr_info("erasing block %d\n", ebnum); err = erase_eraseblock(ebnum); if (err) return err; - printk(PRINT_PREF "writing 1st page of block %d\n", ebnum); + pr_info("writing 1st page of block %d\n", ebnum); set_random_data(writebuf, pgsize); strcpy(writebuf, "There is no data like this!"); err = mtd_write(mtd, addr0, pgsize, &written, writebuf); if (err || written != pgsize) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_info("error: write failed at %#llx\n", (long long)addr0); return err ? err : -1; } - printk(PRINT_PREF "reading 1st page of block %d\n", ebnum); + pr_info("reading 1st page of block %d\n", ebnum); memset(readbuf, 0, pgsize); err = mtd_read(mtd, addr0, pgsize, &read, readbuf); if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr0); return err ? err : -1; } - printk(PRINT_PREF "verifying 1st page of block %d\n", ebnum); + pr_info("verifying 1st page of block %d\n", ebnum); if (memcmp(writebuf, readbuf, pgsize)) { - printk(PRINT_PREF "verify failed!\n"); + pr_err("verify failed!\n"); errcnt += 1; return -1; } - printk(PRINT_PREF "erasing block %d\n", ebnum); + pr_info("erasing block %d\n", ebnum); err = erase_eraseblock(ebnum); if (err) return err; - printk(PRINT_PREF "writing 1st page of block %d\n", ebnum); + pr_info("writing 1st page of block %d\n", ebnum); set_random_data(writebuf, pgsize); strcpy(writebuf, "There is no data like this!"); err = mtd_write(mtd, addr0, pgsize, &written, writebuf); if (err || written != pgsize) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", (long long)addr0); return err ? err : -1; } - printk(PRINT_PREF "erasing block %d\n", ebnum2); + pr_info("erasing block %d\n", ebnum2); err = erase_eraseblock(ebnum2); if (err) return err; - printk(PRINT_PREF "reading 1st page of block %d\n", ebnum); + pr_info("reading 1st page of block %d\n", ebnum); memset(readbuf, 0, pgsize); err = mtd_read(mtd, addr0, pgsize, &read, readbuf); if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr0); return err ? err : -1; } - printk(PRINT_PREF "verifying 1st page of block %d\n", ebnum); + pr_info("verifying 1st page of block %d\n", ebnum); if (memcmp(writebuf, readbuf, pgsize)) { - printk(PRINT_PREF "verify failed!\n"); + pr_err("verify failed!\n"); errcnt += 1; return -1; } if (!err) - printk(PRINT_PREF "erasecrosstest ok\n"); + pr_info("erasecrosstest ok\n"); return err; } @@ -402,7 +402,7 @@ static int erasetest(void) int err = 0, i, ebnum, ok = 1; loff_t addr0; - printk(PRINT_PREF "erasetest\n"); + pr_info("erasetest\n"); ebnum = 0; addr0 = 0; @@ -411,40 +411,40 @@ static int erasetest(void) ebnum += 1; } - printk(PRINT_PREF "erasing block %d\n", ebnum); + pr_info("erasing block %d\n", ebnum); err = erase_eraseblock(ebnum); if (err) return err; - printk(PRINT_PREF "writing 1st page of block %d\n", ebnum); + pr_info("writing 1st page of block %d\n", ebnum); set_random_data(writebuf, pgsize); err = mtd_write(mtd, addr0, pgsize, &written, writebuf); if (err || written != pgsize) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", (long long)addr0); return err ? err : -1; } - printk(PRINT_PREF "erasing block %d\n", ebnum); + pr_info("erasing block %d\n", ebnum); err = erase_eraseblock(ebnum); if (err) return err; - printk(PRINT_PREF "reading 1st page of block %d\n", ebnum); + pr_info("reading 1st page of block %d\n", ebnum); err = mtd_read(mtd, addr0, pgsize, &read, twopages); if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr0); return err ? err : -1; } - printk(PRINT_PREF "verifying 1st page of block %d is all 0xff\n", + pr_info("verifying 1st page of block %d is all 0xff\n", ebnum); for (i = 0; i < pgsize; ++i) if (twopages[i] != 0xff) { - printk(PRINT_PREF "verifying all 0xff failed at %d\n", + pr_err("verifying all 0xff failed at %d\n", i); errcnt += 1; ok = 0; @@ -452,7 +452,7 @@ static int erasetest(void) } if (ok && !err) - printk(PRINT_PREF "erasetest ok\n"); + pr_info("erasetest ok\n"); return err; } @@ -464,7 +464,7 @@ static int is_block_bad(int ebnum) ret = mtd_block_isbad(mtd, addr); if (ret) - printk(PRINT_PREF "block %d is bad\n", ebnum); + pr_info("block %d is bad\n", ebnum); return ret; } @@ -474,18 +474,18 @@ static int scan_for_bad_eraseblocks(void) bbt = kzalloc(ebcnt, GFP_KERNEL); if (!bbt) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); return -ENOMEM; } - printk(PRINT_PREF "scanning for bad eraseblocks\n"); + pr_info("scanning for bad eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { bbt[i] = is_block_bad(i) ? 1 : 0; if (bbt[i]) bad += 1; cond_resched(); } - printk(PRINT_PREF "scanned %d eraseblocks, %d are bad\n", i, bad); + pr_info("scanned %d eraseblocks, %d are bad\n", i, bad); return 0; } @@ -499,22 +499,22 @@ static int __init mtd_pagetest_init(void) printk(KERN_INFO "=================================================\n"); if (dev < 0) { - printk(PRINT_PREF "Please specify a valid mtd-device via module paramter\n"); - printk(KERN_CRIT "CAREFUL: This test wipes all data on the specified MTD device!\n"); + pr_info("Please specify a valid mtd-device via module parameter\n"); + pr_crit("CAREFUL: This test wipes all data on the specified MTD device!\n"); return -EINVAL; } - printk(PRINT_PREF "MTD device: %d\n", dev); + pr_info("MTD device: %d\n", dev); mtd = get_mtd_device(NULL, dev); if (IS_ERR(mtd)) { err = PTR_ERR(mtd); - printk(PRINT_PREF "error: cannot get MTD device\n"); + pr_err("error: cannot get MTD device\n"); return err; } if (mtd->type != MTD_NANDFLASH) { - printk(PRINT_PREF "this test requires NAND flash\n"); + pr_info("this test requires NAND flash\n"); goto out; } @@ -524,7 +524,7 @@ static int __init mtd_pagetest_init(void) pgcnt = mtd->erasesize / mtd->writesize; pgsize = mtd->writesize; - printk(PRINT_PREF "MTD device size %llu, eraseblock size %u, " + pr_info("MTD device size %llu, eraseblock size %u, " "page size %u, count of eraseblocks %u, pages per " "eraseblock %u, OOB size %u\n", (unsigned long long)mtd->size, mtd->erasesize, @@ -534,17 +534,17 @@ static int __init mtd_pagetest_init(void) bufsize = pgsize * 2; writebuf = kmalloc(mtd->erasesize, GFP_KERNEL); if (!writebuf) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } twopages = kmalloc(bufsize, GFP_KERNEL); if (!twopages) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } boundary = kmalloc(bufsize, GFP_KERNEL); if (!boundary) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } @@ -553,7 +553,7 @@ static int __init mtd_pagetest_init(void) goto out; /* Erase all eraseblocks */ - printk(PRINT_PREF "erasing whole device\n"); + pr_info("erasing whole device\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -562,11 +562,11 @@ static int __init mtd_pagetest_init(void) goto out; cond_resched(); } - printk(PRINT_PREF "erased %u eraseblocks\n", i); + pr_info("erased %u eraseblocks\n", i); /* Write all eraseblocks */ simple_srand(1); - printk(PRINT_PREF "writing whole device\n"); + pr_info("writing whole device\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -574,14 +574,14 @@ static int __init mtd_pagetest_init(void) if (err) goto out; if (i % 256 == 0) - printk(PRINT_PREF "written up to eraseblock %u\n", i); + pr_info("written up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "written %u eraseblocks\n", i); + pr_info("written %u eraseblocks\n", i); /* Check all eraseblocks */ simple_srand(1); - printk(PRINT_PREF "verifying all eraseblocks\n"); + pr_info("verifying all eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -589,10 +589,10 @@ static int __init mtd_pagetest_init(void) if (err) goto out; if (i % 256 == 0) - printk(PRINT_PREF "verified up to eraseblock %u\n", i); + pr_info("verified up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "verified %u eraseblocks\n", i); + pr_info("verified %u eraseblocks\n", i); err = crosstest(); if (err) @@ -606,7 +606,7 @@ static int __init mtd_pagetest_init(void) if (err) goto out; - printk(PRINT_PREF "finished with %d errors\n", errcnt); + pr_info("finished with %d errors\n", errcnt); out: kfree(bbt); @@ -615,7 +615,7 @@ out: kfree(writebuf); put_mtd_device(mtd); if (err) - printk(PRINT_PREF "error %d occurred\n", err); + pr_info("error %d occurred\n", err); printk(KERN_INFO "=================================================\n"); return err; } diff --git a/drivers/mtd/tests/mtd_readtest.c b/drivers/mtd/tests/mtd_readtest.c index 121aba189cec..266de04b6d29 100644 --- a/drivers/mtd/tests/mtd_readtest.c +++ b/drivers/mtd/tests/mtd_readtest.c @@ -19,6 +19,8 @@ * Author: Adrian Hunter <ext-adrian.hunter@nokia.com> */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/init.h> #include <linux/module.h> #include <linux/moduleparam.h> @@ -27,8 +29,6 @@ #include <linux/slab.h> #include <linux/sched.h> -#define PRINT_PREF KERN_INFO "mtd_readtest: " - static int dev = -EINVAL; module_param(dev, int, S_IRUGO); MODULE_PARM_DESC(dev, "MTD device number to use"); @@ -51,12 +51,12 @@ static int read_eraseblock_by_page(int ebnum) void *oobbuf = iobuf1; for (i = 0; i < pgcnt; i++) { - memset(buf, 0 , pgcnt); + memset(buf, 0 , pgsize); ret = mtd_read(mtd, addr, pgsize, &read, buf); if (ret == -EUCLEAN) ret = 0; if (ret || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); if (!err) err = ret; @@ -77,7 +77,7 @@ static int read_eraseblock_by_page(int ebnum) ret = mtd_read_oob(mtd, addr, &ops); if ((ret && !mtd_is_bitflip(ret)) || ops.oobretlen != mtd->oobsize) { - printk(PRINT_PREF "error: read oob failed at " + pr_err("error: read oob failed at " "%#llx\n", (long long)addr); if (!err) err = ret; @@ -99,7 +99,7 @@ static void dump_eraseblock(int ebnum) char line[128]; int pg, oob; - printk(PRINT_PREF "dumping eraseblock %d\n", ebnum); + pr_info("dumping eraseblock %d\n", ebnum); n = mtd->erasesize; for (i = 0; i < n;) { char *p = line; @@ -112,7 +112,7 @@ static void dump_eraseblock(int ebnum) } if (!mtd->oobsize) return; - printk(PRINT_PREF "dumping oob from eraseblock %d\n", ebnum); + pr_info("dumping oob from eraseblock %d\n", ebnum); n = mtd->oobsize; for (pg = 0, i = 0; pg < pgcnt; pg++) for (oob = 0; oob < n;) { @@ -134,7 +134,7 @@ static int is_block_bad(int ebnum) ret = mtd_block_isbad(mtd, addr); if (ret) - printk(PRINT_PREF "block %d is bad\n", ebnum); + pr_info("block %d is bad\n", ebnum); return ret; } @@ -144,21 +144,21 @@ static int scan_for_bad_eraseblocks(void) bbt = kzalloc(ebcnt, GFP_KERNEL); if (!bbt) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); return -ENOMEM; } if (!mtd_can_have_bb(mtd)) return 0; - printk(PRINT_PREF "scanning for bad eraseblocks\n"); + pr_info("scanning for bad eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { bbt[i] = is_block_bad(i) ? 1 : 0; if (bbt[i]) bad += 1; cond_resched(); } - printk(PRINT_PREF "scanned %d eraseblocks, %d are bad\n", i, bad); + pr_info("scanned %d eraseblocks, %d are bad\n", i, bad); return 0; } @@ -171,21 +171,21 @@ static int __init mtd_readtest_init(void) printk(KERN_INFO "=================================================\n"); if (dev < 0) { - printk(PRINT_PREF "Please specify a valid mtd-device via module paramter\n"); + pr_info("Please specify a valid mtd-device via module parameter\n"); return -EINVAL; } - printk(PRINT_PREF "MTD device: %d\n", dev); + pr_info("MTD device: %d\n", dev); mtd = get_mtd_device(NULL, dev); if (IS_ERR(mtd)) { err = PTR_ERR(mtd); - printk(PRINT_PREF "error: Cannot get MTD device\n"); + pr_err("error: Cannot get MTD device\n"); return err; } if (mtd->writesize == 1) { - printk(PRINT_PREF "not NAND flash, assume page size is 512 " + pr_info("not NAND flash, assume page size is 512 " "bytes.\n"); pgsize = 512; } else @@ -196,7 +196,7 @@ static int __init mtd_readtest_init(void) ebcnt = tmp; pgcnt = mtd->erasesize / pgsize; - printk(PRINT_PREF "MTD device size %llu, eraseblock size %u, " + pr_info("MTD device size %llu, eraseblock size %u, " "page size %u, count of eraseblocks %u, pages per " "eraseblock %u, OOB size %u\n", (unsigned long long)mtd->size, mtd->erasesize, @@ -205,12 +205,12 @@ static int __init mtd_readtest_init(void) err = -ENOMEM; iobuf = kmalloc(mtd->erasesize, GFP_KERNEL); if (!iobuf) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } iobuf1 = kmalloc(mtd->erasesize, GFP_KERNEL); if (!iobuf1) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } @@ -219,7 +219,7 @@ static int __init mtd_readtest_init(void) goto out; /* Read all eraseblocks 1 page at a time */ - printk(PRINT_PREF "testing page read\n"); + pr_info("testing page read\n"); for (i = 0; i < ebcnt; ++i) { int ret; @@ -235,9 +235,9 @@ static int __init mtd_readtest_init(void) } if (err) - printk(PRINT_PREF "finished with errors\n"); + pr_info("finished with errors\n"); else - printk(PRINT_PREF "finished\n"); + pr_info("finished\n"); out: @@ -246,7 +246,7 @@ out: kfree(bbt); put_mtd_device(mtd); if (err) - printk(PRINT_PREF "error %d occurred\n", err); + pr_info("error %d occurred\n", err); printk(KERN_INFO "=================================================\n"); return err; } diff --git a/drivers/mtd/tests/mtd_speedtest.c b/drivers/mtd/tests/mtd_speedtest.c index 42b0f7456fc4..596cbea8df4c 100644 --- a/drivers/mtd/tests/mtd_speedtest.c +++ b/drivers/mtd/tests/mtd_speedtest.c @@ -19,6 +19,8 @@ * Author: Adrian Hunter <adrian.hunter@nokia.com> */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/init.h> #include <linux/module.h> #include <linux/moduleparam.h> @@ -28,8 +30,6 @@ #include <linux/sched.h> #include <linux/random.h> -#define PRINT_PREF KERN_INFO "mtd_speedtest: " - static int dev = -EINVAL; module_param(dev, int, S_IRUGO); MODULE_PARM_DESC(dev, "MTD device number to use"); @@ -70,12 +70,12 @@ static int erase_eraseblock(int ebnum) err = mtd_erase(mtd, &ei); if (err) { - printk(PRINT_PREF "error %d while erasing EB %d\n", err, ebnum); + pr_err("error %d while erasing EB %d\n", err, ebnum); return err; } if (ei.state == MTD_ERASE_FAILED) { - printk(PRINT_PREF "some erase error occurred at EB %d\n", + pr_err("some erase error occurred at EB %d\n", ebnum); return -EIO; } @@ -96,13 +96,13 @@ static int multiblock_erase(int ebnum, int blocks) err = mtd_erase(mtd, &ei); if (err) { - printk(PRINT_PREF "error %d while erasing EB %d, blocks %d\n", + pr_err("error %d while erasing EB %d, blocks %d\n", err, ebnum, blocks); return err; } if (ei.state == MTD_ERASE_FAILED) { - printk(PRINT_PREF "some erase error occurred at EB %d," + pr_err("some erase error occurred at EB %d," "blocks %d\n", ebnum, blocks); return -EIO; } @@ -134,7 +134,7 @@ static int write_eraseblock(int ebnum) err = mtd_write(mtd, addr, mtd->erasesize, &written, iobuf); if (err || written != mtd->erasesize) { - printk(PRINT_PREF "error: write failed at %#llx\n", addr); + pr_err("error: write failed at %#llx\n", addr); if (!err) err = -EINVAL; } @@ -152,7 +152,7 @@ static int write_eraseblock_by_page(int ebnum) for (i = 0; i < pgcnt; i++) { err = mtd_write(mtd, addr, pgsize, &written, buf); if (err || written != pgsize) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", addr); if (!err) err = -EINVAL; @@ -175,7 +175,7 @@ static int write_eraseblock_by_2pages(int ebnum) for (i = 0; i < n; i++) { err = mtd_write(mtd, addr, sz, &written, buf); if (err || written != sz) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", addr); if (!err) err = -EINVAL; @@ -187,7 +187,7 @@ static int write_eraseblock_by_2pages(int ebnum) if (pgcnt % 2) { err = mtd_write(mtd, addr, pgsize, &written, buf); if (err || written != pgsize) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", addr); if (!err) err = -EINVAL; @@ -208,7 +208,7 @@ static int read_eraseblock(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != mtd->erasesize) { - printk(PRINT_PREF "error: read failed at %#llx\n", addr); + pr_err("error: read failed at %#llx\n", addr); if (!err) err = -EINVAL; } @@ -229,7 +229,7 @@ static int read_eraseblock_by_page(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", addr); if (!err) err = -EINVAL; @@ -255,7 +255,7 @@ static int read_eraseblock_by_2pages(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != sz) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", addr); if (!err) err = -EINVAL; @@ -270,7 +270,7 @@ static int read_eraseblock_by_2pages(int ebnum) if (mtd_is_bitflip(err)) err = 0; if (err || read != pgsize) { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", addr); if (!err) err = -EINVAL; @@ -287,7 +287,7 @@ static int is_block_bad(int ebnum) ret = mtd_block_isbad(mtd, addr); if (ret) - printk(PRINT_PREF "block %d is bad\n", ebnum); + pr_info("block %d is bad\n", ebnum); return ret; } @@ -321,21 +321,21 @@ static int scan_for_bad_eraseblocks(void) bbt = kzalloc(ebcnt, GFP_KERNEL); if (!bbt) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); return -ENOMEM; } if (!mtd_can_have_bb(mtd)) goto out; - printk(PRINT_PREF "scanning for bad eraseblocks\n"); + pr_info("scanning for bad eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { bbt[i] = is_block_bad(i) ? 1 : 0; if (bbt[i]) bad += 1; cond_resched(); } - printk(PRINT_PREF "scanned %d eraseblocks, %d are bad\n", i, bad); + pr_info("scanned %d eraseblocks, %d are bad\n", i, bad); out: goodebcnt = ebcnt - bad; return 0; @@ -351,25 +351,25 @@ static int __init mtd_speedtest_init(void) printk(KERN_INFO "=================================================\n"); if (dev < 0) { - printk(PRINT_PREF "Please specify a valid mtd-device via module paramter\n"); - printk(KERN_CRIT "CAREFUL: This test wipes all data on the specified MTD device!\n"); + pr_info("Please specify a valid mtd-device via module parameter\n"); + pr_crit("CAREFUL: This test wipes all data on the specified MTD device!\n"); return -EINVAL; } if (count) - printk(PRINT_PREF "MTD device: %d count: %d\n", dev, count); + pr_info("MTD device: %d count: %d\n", dev, count); else - printk(PRINT_PREF "MTD device: %d\n", dev); + pr_info("MTD device: %d\n", dev); mtd = get_mtd_device(NULL, dev); if (IS_ERR(mtd)) { err = PTR_ERR(mtd); - printk(PRINT_PREF "error: cannot get MTD device\n"); + pr_err("error: cannot get MTD device\n"); return err; } if (mtd->writesize == 1) { - printk(PRINT_PREF "not NAND flash, assume page size is 512 " + pr_info("not NAND flash, assume page size is 512 " "bytes.\n"); pgsize = 512; } else @@ -380,7 +380,7 @@ static int __init mtd_speedtest_init(void) ebcnt = tmp; pgcnt = mtd->erasesize / pgsize; - printk(PRINT_PREF "MTD device size %llu, eraseblock size %u, " + pr_info("MTD device size %llu, eraseblock size %u, " "page size %u, count of eraseblocks %u, pages per " "eraseblock %u, OOB size %u\n", (unsigned long long)mtd->size, mtd->erasesize, @@ -392,7 +392,7 @@ static int __init mtd_speedtest_init(void) err = -ENOMEM; iobuf = kmalloc(mtd->erasesize, GFP_KERNEL); if (!iobuf) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } @@ -407,7 +407,7 @@ static int __init mtd_speedtest_init(void) goto out; /* Write all eraseblocks, 1 eraseblock at a time */ - printk(PRINT_PREF "testing eraseblock write speed\n"); + pr_info("testing eraseblock write speed\n"); start_timing(); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) @@ -419,10 +419,10 @@ static int __init mtd_speedtest_init(void) } stop_timing(); speed = calc_speed(); - printk(PRINT_PREF "eraseblock write speed is %ld KiB/s\n", speed); + pr_info("eraseblock write speed is %ld KiB/s\n", speed); /* Read all eraseblocks, 1 eraseblock at a time */ - printk(PRINT_PREF "testing eraseblock read speed\n"); + pr_info("testing eraseblock read speed\n"); start_timing(); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) @@ -434,14 +434,14 @@ static int __init mtd_speedtest_init(void) } stop_timing(); speed = calc_speed(); - printk(PRINT_PREF "eraseblock read speed is %ld KiB/s\n", speed); + pr_info("eraseblock read speed is %ld KiB/s\n", speed); err = erase_whole_device(); if (err) goto out; /* Write all eraseblocks, 1 page at a time */ - printk(PRINT_PREF "testing page write speed\n"); + pr_info("testing page write speed\n"); start_timing(); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) @@ -453,10 +453,10 @@ static int __init mtd_speedtest_init(void) } stop_timing(); speed = calc_speed(); - printk(PRINT_PREF "page write speed is %ld KiB/s\n", speed); + pr_info("page write speed is %ld KiB/s\n", speed); /* Read all eraseblocks, 1 page at a time */ - printk(PRINT_PREF "testing page read speed\n"); + pr_info("testing page read speed\n"); start_timing(); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) @@ -468,14 +468,14 @@ static int __init mtd_speedtest_init(void) } stop_timing(); speed = calc_speed(); - printk(PRINT_PREF "page read speed is %ld KiB/s\n", speed); + pr_info("page read speed is %ld KiB/s\n", speed); err = erase_whole_device(); if (err) goto out; /* Write all eraseblocks, 2 pages at a time */ - printk(PRINT_PREF "testing 2 page write speed\n"); + pr_info("testing 2 page write speed\n"); start_timing(); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) @@ -487,10 +487,10 @@ static int __init mtd_speedtest_init(void) } stop_timing(); speed = calc_speed(); - printk(PRINT_PREF "2 page write speed is %ld KiB/s\n", speed); + pr_info("2 page write speed is %ld KiB/s\n", speed); /* Read all eraseblocks, 2 pages at a time */ - printk(PRINT_PREF "testing 2 page read speed\n"); + pr_info("testing 2 page read speed\n"); start_timing(); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) @@ -502,10 +502,10 @@ static int __init mtd_speedtest_init(void) } stop_timing(); speed = calc_speed(); - printk(PRINT_PREF "2 page read speed is %ld KiB/s\n", speed); + pr_info("2 page read speed is %ld KiB/s\n", speed); /* Erase all eraseblocks */ - printk(PRINT_PREF "Testing erase speed\n"); + pr_info("Testing erase speed\n"); start_timing(); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) @@ -517,12 +517,12 @@ static int __init mtd_speedtest_init(void) } stop_timing(); speed = calc_speed(); - printk(PRINT_PREF "erase speed is %ld KiB/s\n", speed); + pr_info("erase speed is %ld KiB/s\n", speed); /* Multi-block erase all eraseblocks */ for (k = 1; k < 7; k++) { blocks = 1 << k; - printk(PRINT_PREF "Testing %dx multi-block erase speed\n", + pr_info("Testing %dx multi-block erase speed\n", blocks); start_timing(); for (i = 0; i < ebcnt; ) { @@ -541,16 +541,16 @@ static int __init mtd_speedtest_init(void) } stop_timing(); speed = calc_speed(); - printk(PRINT_PREF "%dx multi-block erase speed is %ld KiB/s\n", + pr_info("%dx multi-block erase speed is %ld KiB/s\n", blocks, speed); } - printk(PRINT_PREF "finished\n"); + pr_info("finished\n"); out: kfree(iobuf); kfree(bbt); put_mtd_device(mtd); if (err) - printk(PRINT_PREF "error %d occurred\n", err); + pr_info("error %d occurred\n", err); printk(KERN_INFO "=================================================\n"); return err; } diff --git a/drivers/mtd/tests/mtd_stresstest.c b/drivers/mtd/tests/mtd_stresstest.c index cb268cebf01a..3729f679ae5d 100644 --- a/drivers/mtd/tests/mtd_stresstest.c +++ b/drivers/mtd/tests/mtd_stresstest.c @@ -19,6 +19,8 @@ * Author: Adrian Hunter <ext-adrian.hunter@nokia.com> */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/init.h> #include <linux/module.h> #include <linux/moduleparam.h> @@ -29,8 +31,6 @@ #include <linux/vmalloc.h> #include <linux/random.h> -#define PRINT_PREF KERN_INFO "mtd_stresstest: " - static int dev = -EINVAL; module_param(dev, int, S_IRUGO); MODULE_PARM_DESC(dev, "MTD device number to use"); @@ -94,12 +94,12 @@ static int erase_eraseblock(int ebnum) err = mtd_erase(mtd, &ei); if (unlikely(err)) { - printk(PRINT_PREF "error %d while erasing EB %d\n", err, ebnum); + pr_err("error %d while erasing EB %d\n", err, ebnum); return err; } if (unlikely(ei.state == MTD_ERASE_FAILED)) { - printk(PRINT_PREF "some erase error occurred at EB %d\n", + pr_err("some erase error occurred at EB %d\n", ebnum); return -EIO; } @@ -114,7 +114,7 @@ static int is_block_bad(int ebnum) ret = mtd_block_isbad(mtd, addr); if (ret) - printk(PRINT_PREF "block %d is bad\n", ebnum); + pr_info("block %d is bad\n", ebnum); return ret; } @@ -137,7 +137,7 @@ static int do_read(void) if (mtd_is_bitflip(err)) err = 0; if (unlikely(err || read != len)) { - printk(PRINT_PREF "error: read failed at 0x%llx\n", + pr_err("error: read failed at 0x%llx\n", (long long)addr); if (!err) err = -EINVAL; @@ -174,7 +174,7 @@ static int do_write(void) addr = eb * mtd->erasesize + offs; err = mtd_write(mtd, addr, len, &written, writebuf); if (unlikely(err || written != len)) { - printk(PRINT_PREF "error: write failed at 0x%llx\n", + pr_err("error: write failed at 0x%llx\n", (long long)addr); if (!err) err = -EINVAL; @@ -203,21 +203,21 @@ static int scan_for_bad_eraseblocks(void) bbt = kzalloc(ebcnt, GFP_KERNEL); if (!bbt) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); return -ENOMEM; } if (!mtd_can_have_bb(mtd)) return 0; - printk(PRINT_PREF "scanning for bad eraseblocks\n"); + pr_info("scanning for bad eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { bbt[i] = is_block_bad(i) ? 1 : 0; if (bbt[i]) bad += 1; cond_resched(); } - printk(PRINT_PREF "scanned %d eraseblocks, %d are bad\n", i, bad); + pr_info("scanned %d eraseblocks, %d are bad\n", i, bad); return 0; } @@ -231,22 +231,22 @@ static int __init mtd_stresstest_init(void) printk(KERN_INFO "=================================================\n"); if (dev < 0) { - printk(PRINT_PREF "Please specify a valid mtd-device via module paramter\n"); - printk(KERN_CRIT "CAREFUL: This test wipes all data on the specified MTD device!\n"); + pr_info("Please specify a valid mtd-device via module parameter\n"); + pr_crit("CAREFUL: This test wipes all data on the specified MTD device!\n"); return -EINVAL; } - printk(PRINT_PREF "MTD device: %d\n", dev); + pr_info("MTD device: %d\n", dev); mtd = get_mtd_device(NULL, dev); if (IS_ERR(mtd)) { err = PTR_ERR(mtd); - printk(PRINT_PREF "error: cannot get MTD device\n"); + pr_err("error: cannot get MTD device\n"); return err; } if (mtd->writesize == 1) { - printk(PRINT_PREF "not NAND flash, assume page size is 512 " + pr_info("not NAND flash, assume page size is 512 " "bytes.\n"); pgsize = 512; } else @@ -257,14 +257,14 @@ static int __init mtd_stresstest_init(void) ebcnt = tmp; pgcnt = mtd->erasesize / pgsize; - printk(PRINT_PREF "MTD device size %llu, eraseblock size %u, " + pr_info("MTD device size %llu, eraseblock size %u, " "page size %u, count of eraseblocks %u, pages per " "eraseblock %u, OOB size %u\n", (unsigned long long)mtd->size, mtd->erasesize, pgsize, ebcnt, pgcnt, mtd->oobsize); if (ebcnt < 2) { - printk(PRINT_PREF "error: need at least 2 eraseblocks\n"); + pr_err("error: need at least 2 eraseblocks\n"); err = -ENOSPC; goto out_put_mtd; } @@ -277,7 +277,7 @@ static int __init mtd_stresstest_init(void) writebuf = vmalloc(bufsize); offsets = kmalloc(ebcnt * sizeof(int), GFP_KERNEL); if (!readbuf || !writebuf || !offsets) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out; } for (i = 0; i < ebcnt; i++) @@ -290,16 +290,16 @@ static int __init mtd_stresstest_init(void) goto out; /* Do operations */ - printk(PRINT_PREF "doing operations\n"); + pr_info("doing operations\n"); for (op = 0; op < count; op++) { if ((op & 1023) == 0) - printk(PRINT_PREF "%d operations done\n", op); + pr_info("%d operations done\n", op); err = do_operation(); if (err) goto out; cond_resched(); } - printk(PRINT_PREF "finished, %d operations done\n", op); + pr_info("finished, %d operations done\n", op); out: kfree(offsets); @@ -309,7 +309,7 @@ out: out_put_mtd: put_mtd_device(mtd); if (err) - printk(PRINT_PREF "error %d occurred\n", err); + pr_info("error %d occurred\n", err); printk(KERN_INFO "=================================================\n"); return err; } diff --git a/drivers/mtd/tests/mtd_subpagetest.c b/drivers/mtd/tests/mtd_subpagetest.c index 9667bf535282..c880c2229c59 100644 --- a/drivers/mtd/tests/mtd_subpagetest.c +++ b/drivers/mtd/tests/mtd_subpagetest.c @@ -19,6 +19,8 @@ * */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/init.h> #include <linux/module.h> #include <linux/moduleparam.h> @@ -27,8 +29,6 @@ #include <linux/slab.h> #include <linux/sched.h> -#define PRINT_PREF KERN_INFO "mtd_subpagetest: " - static int dev = -EINVAL; module_param(dev, int, S_IRUGO); MODULE_PARM_DESC(dev, "MTD device number to use"); @@ -82,12 +82,12 @@ static int erase_eraseblock(int ebnum) err = mtd_erase(mtd, &ei); if (err) { - printk(PRINT_PREF "error %d while erasing EB %d\n", err, ebnum); + pr_err("error %d while erasing EB %d\n", err, ebnum); return err; } if (ei.state == MTD_ERASE_FAILED) { - printk(PRINT_PREF "some erase error occurred at EB %d\n", + pr_err("some erase error occurred at EB %d\n", ebnum); return -EIO; } @@ -100,7 +100,7 @@ static int erase_whole_device(void) int err; unsigned int i; - printk(PRINT_PREF "erasing whole device\n"); + pr_info("erasing whole device\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -109,7 +109,7 @@ static int erase_whole_device(void) return err; cond_resched(); } - printk(PRINT_PREF "erased %u eraseblocks\n", i); + pr_info("erased %u eraseblocks\n", i); return 0; } @@ -122,11 +122,11 @@ static int write_eraseblock(int ebnum) set_random_data(writebuf, subpgsize); err = mtd_write(mtd, addr, subpgsize, &written, writebuf); if (unlikely(err || written != subpgsize)) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", (long long)addr); if (written != subpgsize) { - printk(PRINT_PREF " write size: %#x\n", subpgsize); - printk(PRINT_PREF " written: %#zx\n", written); + pr_err(" write size: %#x\n", subpgsize); + pr_err(" written: %#zx\n", written); } return err ? err : -1; } @@ -136,11 +136,11 @@ static int write_eraseblock(int ebnum) set_random_data(writebuf, subpgsize); err = mtd_write(mtd, addr, subpgsize, &written, writebuf); if (unlikely(err || written != subpgsize)) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", (long long)addr); if (written != subpgsize) { - printk(PRINT_PREF " write size: %#x\n", subpgsize); - printk(PRINT_PREF " written: %#zx\n", written); + pr_err(" write size: %#x\n", subpgsize); + pr_err(" written: %#zx\n", written); } return err ? err : -1; } @@ -160,12 +160,12 @@ static int write_eraseblock2(int ebnum) set_random_data(writebuf, subpgsize * k); err = mtd_write(mtd, addr, subpgsize * k, &written, writebuf); if (unlikely(err || written != subpgsize * k)) { - printk(PRINT_PREF "error: write failed at %#llx\n", + pr_err("error: write failed at %#llx\n", (long long)addr); if (written != subpgsize) { - printk(PRINT_PREF " write size: %#x\n", + pr_err(" write size: %#x\n", subpgsize * k); - printk(PRINT_PREF " written: %#08zx\n", + pr_err(" written: %#08zx\n", written); } return err ? err : -1; @@ -198,23 +198,23 @@ static int verify_eraseblock(int ebnum) err = mtd_read(mtd, addr, subpgsize, &read, readbuf); if (unlikely(err || read != subpgsize)) { if (mtd_is_bitflip(err) && read == subpgsize) { - printk(PRINT_PREF "ECC correction at %#llx\n", + pr_info("ECC correction at %#llx\n", (long long)addr); err = 0; } else { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); return err ? err : -1; } } if (unlikely(memcmp(readbuf, writebuf, subpgsize))) { - printk(PRINT_PREF "error: verify failed at %#llx\n", + pr_err("error: verify failed at %#llx\n", (long long)addr); - printk(PRINT_PREF "------------- written----------------\n"); + pr_info("------------- written----------------\n"); print_subpage(writebuf); - printk(PRINT_PREF "------------- read ------------------\n"); + pr_info("------------- read ------------------\n"); print_subpage(readbuf); - printk(PRINT_PREF "-------------------------------------\n"); + pr_info("-------------------------------------\n"); errcnt += 1; } @@ -225,23 +225,23 @@ static int verify_eraseblock(int ebnum) err = mtd_read(mtd, addr, subpgsize, &read, readbuf); if (unlikely(err || read != subpgsize)) { if (mtd_is_bitflip(err) && read == subpgsize) { - printk(PRINT_PREF "ECC correction at %#llx\n", + pr_info("ECC correction at %#llx\n", (long long)addr); err = 0; } else { - printk(PRINT_PREF "error: read failed at %#llx\n", + pr_err("error: read failed at %#llx\n", (long long)addr); return err ? err : -1; } } if (unlikely(memcmp(readbuf, writebuf, subpgsize))) { - printk(PRINT_PREF "error: verify failed at %#llx\n", + pr_info("error: verify failed at %#llx\n", (long long)addr); - printk(PRINT_PREF "------------- written----------------\n"); + pr_info("------------- written----------------\n"); print_subpage(writebuf); - printk(PRINT_PREF "------------- read ------------------\n"); + pr_info("------------- read ------------------\n"); print_subpage(readbuf); - printk(PRINT_PREF "-------------------------------------\n"); + pr_info("-------------------------------------\n"); errcnt += 1; } @@ -262,17 +262,17 @@ static int verify_eraseblock2(int ebnum) err = mtd_read(mtd, addr, subpgsize * k, &read, readbuf); if (unlikely(err || read != subpgsize * k)) { if (mtd_is_bitflip(err) && read == subpgsize * k) { - printk(PRINT_PREF "ECC correction at %#llx\n", + pr_info("ECC correction at %#llx\n", (long long)addr); err = 0; } else { - printk(PRINT_PREF "error: read failed at " + pr_err("error: read failed at " "%#llx\n", (long long)addr); return err ? err : -1; } } if (unlikely(memcmp(readbuf, writebuf, subpgsize * k))) { - printk(PRINT_PREF "error: verify failed at %#llx\n", + pr_err("error: verify failed at %#llx\n", (long long)addr); errcnt += 1; } @@ -295,17 +295,17 @@ static int verify_eraseblock_ff(int ebnum) err = mtd_read(mtd, addr, subpgsize, &read, readbuf); if (unlikely(err || read != subpgsize)) { if (mtd_is_bitflip(err) && read == subpgsize) { - printk(PRINT_PREF "ECC correction at %#llx\n", + pr_info("ECC correction at %#llx\n", (long long)addr); err = 0; } else { - printk(PRINT_PREF "error: read failed at " + pr_err("error: read failed at " "%#llx\n", (long long)addr); return err ? err : -1; } } if (unlikely(memcmp(readbuf, writebuf, subpgsize))) { - printk(PRINT_PREF "error: verify 0xff failed at " + pr_err("error: verify 0xff failed at " "%#llx\n", (long long)addr); errcnt += 1; } @@ -320,7 +320,7 @@ static int verify_all_eraseblocks_ff(void) int err; unsigned int i; - printk(PRINT_PREF "verifying all eraseblocks for 0xff\n"); + pr_info("verifying all eraseblocks for 0xff\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -328,10 +328,10 @@ static int verify_all_eraseblocks_ff(void) if (err) return err; if (i % 256 == 0) - printk(PRINT_PREF "verified up to eraseblock %u\n", i); + pr_info("verified up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "verified %u eraseblocks\n", i); + pr_info("verified %u eraseblocks\n", i); return 0; } @@ -342,7 +342,7 @@ static int is_block_bad(int ebnum) ret = mtd_block_isbad(mtd, addr); if (ret) - printk(PRINT_PREF "block %d is bad\n", ebnum); + pr_info("block %d is bad\n", ebnum); return ret; } @@ -352,18 +352,18 @@ static int scan_for_bad_eraseblocks(void) bbt = kzalloc(ebcnt, GFP_KERNEL); if (!bbt) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); return -ENOMEM; } - printk(PRINT_PREF "scanning for bad eraseblocks\n"); + pr_info("scanning for bad eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { bbt[i] = is_block_bad(i) ? 1 : 0; if (bbt[i]) bad += 1; cond_resched(); } - printk(PRINT_PREF "scanned %d eraseblocks, %d are bad\n", i, bad); + pr_info("scanned %d eraseblocks, %d are bad\n", i, bad); return 0; } @@ -377,22 +377,22 @@ static int __init mtd_subpagetest_init(void) printk(KERN_INFO "=================================================\n"); if (dev < 0) { - printk(PRINT_PREF "Please specify a valid mtd-device via module paramter\n"); - printk(KERN_CRIT "CAREFUL: This test wipes all data on the specified MTD device!\n"); + pr_info("Please specify a valid mtd-device via module parameter\n"); + pr_crit("CAREFUL: This test wipes all data on the specified MTD device!\n"); return -EINVAL; } - printk(PRINT_PREF "MTD device: %d\n", dev); + pr_info("MTD device: %d\n", dev); mtd = get_mtd_device(NULL, dev); if (IS_ERR(mtd)) { err = PTR_ERR(mtd); - printk(PRINT_PREF "error: cannot get MTD device\n"); + pr_err("error: cannot get MTD device\n"); return err; } if (mtd->type != MTD_NANDFLASH) { - printk(PRINT_PREF "this test requires NAND flash\n"); + pr_info("this test requires NAND flash\n"); goto out; } @@ -402,7 +402,7 @@ static int __init mtd_subpagetest_init(void) ebcnt = tmp; pgcnt = mtd->erasesize / mtd->writesize; - printk(PRINT_PREF "MTD device size %llu, eraseblock size %u, " + pr_info("MTD device size %llu, eraseblock size %u, " "page size %u, subpage size %u, count of eraseblocks %u, " "pages per eraseblock %u, OOB size %u\n", (unsigned long long)mtd->size, mtd->erasesize, @@ -412,12 +412,12 @@ static int __init mtd_subpagetest_init(void) bufsize = subpgsize * 32; writebuf = kmalloc(bufsize, GFP_KERNEL); if (!writebuf) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_info("error: cannot allocate memory\n"); goto out; } readbuf = kmalloc(bufsize, GFP_KERNEL); if (!readbuf) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_info("error: cannot allocate memory\n"); goto out; } @@ -429,7 +429,7 @@ static int __init mtd_subpagetest_init(void) if (err) goto out; - printk(PRINT_PREF "writing whole device\n"); + pr_info("writing whole device\n"); simple_srand(1); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) @@ -438,13 +438,13 @@ static int __init mtd_subpagetest_init(void) if (unlikely(err)) goto out; if (i % 256 == 0) - printk(PRINT_PREF "written up to eraseblock %u\n", i); + pr_info("written up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "written %u eraseblocks\n", i); + pr_info("written %u eraseblocks\n", i); simple_srand(1); - printk(PRINT_PREF "verifying all eraseblocks\n"); + pr_info("verifying all eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -452,10 +452,10 @@ static int __init mtd_subpagetest_init(void) if (unlikely(err)) goto out; if (i % 256 == 0) - printk(PRINT_PREF "verified up to eraseblock %u\n", i); + pr_info("verified up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "verified %u eraseblocks\n", i); + pr_info("verified %u eraseblocks\n", i); err = erase_whole_device(); if (err) @@ -467,7 +467,7 @@ static int __init mtd_subpagetest_init(void) /* Write all eraseblocks */ simple_srand(3); - printk(PRINT_PREF "writing whole device\n"); + pr_info("writing whole device\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -475,14 +475,14 @@ static int __init mtd_subpagetest_init(void) if (unlikely(err)) goto out; if (i % 256 == 0) - printk(PRINT_PREF "written up to eraseblock %u\n", i); + pr_info("written up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "written %u eraseblocks\n", i); + pr_info("written %u eraseblocks\n", i); /* Check all eraseblocks */ simple_srand(3); - printk(PRINT_PREF "verifying all eraseblocks\n"); + pr_info("verifying all eraseblocks\n"); for (i = 0; i < ebcnt; ++i) { if (bbt[i]) continue; @@ -490,10 +490,10 @@ static int __init mtd_subpagetest_init(void) if (unlikely(err)) goto out; if (i % 256 == 0) - printk(PRINT_PREF "verified up to eraseblock %u\n", i); + pr_info("verified up to eraseblock %u\n", i); cond_resched(); } - printk(PRINT_PREF "verified %u eraseblocks\n", i); + pr_info("verified %u eraseblocks\n", i); err = erase_whole_device(); if (err) @@ -503,7 +503,7 @@ static int __init mtd_subpagetest_init(void) if (err) goto out; - printk(PRINT_PREF "finished with %d errors\n", errcnt); + pr_info("finished with %d errors\n", errcnt); out: kfree(bbt); @@ -511,7 +511,7 @@ out: kfree(writebuf); put_mtd_device(mtd); if (err) - printk(PRINT_PREF "error %d occurred\n", err); + pr_info("error %d occurred\n", err); printk(KERN_INFO "=================================================\n"); return err; } diff --git a/drivers/mtd/tests/mtd_torturetest.c b/drivers/mtd/tests/mtd_torturetest.c index b65861bc7b8e..c4cde1e9eddb 100644 --- a/drivers/mtd/tests/mtd_torturetest.c +++ b/drivers/mtd/tests/mtd_torturetest.c @@ -23,6 +23,8 @@ * damage caused by this program. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/init.h> #include <linux/module.h> #include <linux/moduleparam.h> @@ -31,7 +33,6 @@ #include <linux/slab.h> #include <linux/sched.h> -#define PRINT_PREF KERN_INFO "mtd_torturetest: " #define RETRIES 3 static int eb = 8; @@ -107,12 +108,12 @@ static inline int erase_eraseblock(int ebnum) err = mtd_erase(mtd, &ei); if (err) { - printk(PRINT_PREF "error %d while erasing EB %d\n", err, ebnum); + pr_err("error %d while erasing EB %d\n", err, ebnum); return err; } if (ei.state == MTD_ERASE_FAILED) { - printk(PRINT_PREF "some erase error occurred at EB %d\n", + pr_err("some erase error occurred at EB %d\n", ebnum); return -EIO; } @@ -139,40 +140,40 @@ static inline int check_eraseblock(int ebnum, unsigned char *buf) retry: err = mtd_read(mtd, addr, len, &read, check_buf); if (mtd_is_bitflip(err)) - printk(PRINT_PREF "single bit flip occurred at EB %d " + pr_err("single bit flip occurred at EB %d " "MTD reported that it was fixed.\n", ebnum); else if (err) { - printk(PRINT_PREF "error %d while reading EB %d, " + pr_err("error %d while reading EB %d, " "read %zd\n", err, ebnum, read); return err; } if (read != len) { - printk(PRINT_PREF "failed to read %zd bytes from EB %d, " + pr_err("failed to read %zd bytes from EB %d, " "read only %zd, but no error reported\n", len, ebnum, read); return -EIO; } if (memcmp(buf, check_buf, len)) { - printk(PRINT_PREF "read wrong data from EB %d\n", ebnum); + pr_err("read wrong data from EB %d\n", ebnum); report_corrupt(check_buf, buf); if (retries++ < RETRIES) { /* Try read again */ yield(); - printk(PRINT_PREF "re-try reading data from EB %d\n", + pr_info("re-try reading data from EB %d\n", ebnum); goto retry; } else { - printk(PRINT_PREF "retried %d times, still errors, " + pr_info("retried %d times, still errors, " "give-up\n", RETRIES); return -EINVAL; } } if (retries != 0) - printk(PRINT_PREF "only attempt number %d was OK (!!!)\n", + pr_info("only attempt number %d was OK (!!!)\n", retries); return 0; @@ -191,12 +192,12 @@ static inline int write_pattern(int ebnum, void *buf) } err = mtd_write(mtd, addr, len, &written, buf); if (err) { - printk(PRINT_PREF "error %d while writing EB %d, written %zd" + pr_err("error %d while writing EB %d, written %zd" " bytes\n", err, ebnum, written); return err; } if (written != len) { - printk(PRINT_PREF "written only %zd bytes of %zd, but no error" + pr_info("written only %zd bytes of %zd, but no error" " reported\n", written, len); return -EIO; } @@ -211,64 +212,64 @@ static int __init tort_init(void) printk(KERN_INFO "\n"); printk(KERN_INFO "=================================================\n"); - printk(PRINT_PREF "Warning: this program is trying to wear out your " + pr_info("Warning: this program is trying to wear out your " "flash, stop it if this is not wanted.\n"); if (dev < 0) { - printk(PRINT_PREF "Please specify a valid mtd-device via module paramter\n"); - printk(KERN_CRIT "CAREFUL: This test wipes all data on the specified MTD device!\n"); + pr_info("Please specify a valid mtd-device via module parameter\n"); + pr_crit("CAREFUL: This test wipes all data on the specified MTD device!\n"); return -EINVAL; } - printk(PRINT_PREF "MTD device: %d\n", dev); - printk(PRINT_PREF "torture %d eraseblocks (%d-%d) of mtd%d\n", + pr_info("MTD device: %d\n", dev); + pr_info("torture %d eraseblocks (%d-%d) of mtd%d\n", ebcnt, eb, eb + ebcnt - 1, dev); if (pgcnt) - printk(PRINT_PREF "torturing just %d pages per eraseblock\n", + pr_info("torturing just %d pages per eraseblock\n", pgcnt); - printk(PRINT_PREF "write verify %s\n", check ? "enabled" : "disabled"); + pr_info("write verify %s\n", check ? "enabled" : "disabled"); mtd = get_mtd_device(NULL, dev); if (IS_ERR(mtd)) { err = PTR_ERR(mtd); - printk(PRINT_PREF "error: cannot get MTD device\n"); + pr_err("error: cannot get MTD device\n"); return err; } if (mtd->writesize == 1) { - printk(PRINT_PREF "not NAND flash, assume page size is 512 " + pr_info("not NAND flash, assume page size is 512 " "bytes.\n"); pgsize = 512; } else pgsize = mtd->writesize; if (pgcnt && (pgcnt > mtd->erasesize / pgsize || pgcnt < 0)) { - printk(PRINT_PREF "error: invalid pgcnt value %d\n", pgcnt); + pr_err("error: invalid pgcnt value %d\n", pgcnt); goto out_mtd; } err = -ENOMEM; patt_5A5 = kmalloc(mtd->erasesize, GFP_KERNEL); if (!patt_5A5) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out_mtd; } patt_A5A = kmalloc(mtd->erasesize, GFP_KERNEL); if (!patt_A5A) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out_patt_5A5; } patt_FF = kmalloc(mtd->erasesize, GFP_KERNEL); if (!patt_FF) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out_patt_A5A; } check_buf = kmalloc(mtd->erasesize, GFP_KERNEL); if (!check_buf) { - printk(PRINT_PREF "error: cannot allocate memory\n"); + pr_err("error: cannot allocate memory\n"); goto out_patt_FF; } @@ -295,13 +296,13 @@ static int __init tort_init(void) err = mtd_block_isbad(mtd, (loff_t)i * mtd->erasesize); if (err < 0) { - printk(PRINT_PREF "block_isbad() returned %d " + pr_info("block_isbad() returned %d " "for EB %d\n", err, i); goto out; } if (err) { - printk("EB %d is bad. Skip it.\n", i); + pr_err("EB %d is bad. Skip it.\n", i); bad_ebs[i - eb] = 1; } } @@ -329,7 +330,7 @@ static int __init tort_init(void) continue; err = check_eraseblock(i, patt_FF); if (err) { - printk(PRINT_PREF "verify failed" + pr_info("verify failed" " for 0xFF... pattern\n"); goto out; } @@ -362,7 +363,7 @@ static int __init tort_init(void) patt = patt_A5A; err = check_eraseblock(i, patt); if (err) { - printk(PRINT_PREF "verify failed for %s" + pr_info("verify failed for %s" " pattern\n", ((eb + erase_cycles) & 1) ? "0x55AA55..." : "0xAA55AA..."); @@ -380,7 +381,7 @@ static int __init tort_init(void) stop_timing(); ms = (finish.tv_sec - start.tv_sec) * 1000 + (finish.tv_usec - start.tv_usec) / 1000; - printk(PRINT_PREF "%08u erase cycles done, took %lu " + pr_info("%08u erase cycles done, took %lu " "milliseconds (%lu seconds)\n", erase_cycles, ms, ms / 1000); start_timing(); @@ -391,7 +392,7 @@ static int __init tort_init(void) } out: - printk(PRINT_PREF "finished after %u erase cycles\n", + pr_info("finished after %u erase cycles\n", erase_cycles); kfree(check_buf); out_patt_FF: @@ -403,7 +404,7 @@ out_patt_5A5: out_mtd: put_mtd_device(mtd); if (err) - printk(PRINT_PREF "error %d occurred during torturing\n", err); + pr_info("error %d occurred during torturing\n", err); printk(KERN_INFO "=================================================\n"); return err; } @@ -441,9 +442,9 @@ static void report_corrupt(unsigned char *read, unsigned char *written) &bits) >= 0) pages++; - printk(PRINT_PREF "verify fails on %d pages, %d bytes/%d bits\n", + pr_info("verify fails on %d pages, %d bytes/%d bits\n", pages, bytes, bits); - printk(PRINT_PREF "The following is a list of all differences between" + pr_info("The following is a list of all differences between" " what was read from flash and what was expected\n"); for (i = 0; i < check_len; i += pgsize) { @@ -457,7 +458,7 @@ static void report_corrupt(unsigned char *read, unsigned char *written) printk("-------------------------------------------------------" "----------------------------------\n"); - printk(PRINT_PREF "Page %zd has %d bytes/%d bits failing verify," + pr_info("Page %zd has %d bytes/%d bits failing verify," " starting at offset 0x%x\n", (mtd->erasesize - check_len + i) / pgsize, bytes, bits, first); diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index ef2cb2418535..b7d45f367d4a 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4431,8 +4431,6 @@ static void bond_uninit(struct net_device *bond_dev) list_del(&bond->bond_list); - bond_work_cancel_all(bond); - bond_debug_unregister(bond); __hw_addr_flush(&bond->mc_list); diff --git a/drivers/net/can/sja1000/sja1000_of_platform.c b/drivers/net/can/sja1000/sja1000_of_platform.c index 0f5917000aa2..6433b81256cd 100644 --- a/drivers/net/can/sja1000/sja1000_of_platform.c +++ b/drivers/net/can/sja1000/sja1000_of_platform.c @@ -121,7 +121,7 @@ static int sja1000_ofp_probe(struct platform_device *ofdev) } irq = irq_of_parse_and_map(np, 0); - if (irq == NO_IRQ) { + if (irq == 0) { dev_err(&ofdev->dev, "no irq found\n"); err = -ENODEV; goto exit_unmap_mem; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index a2998bea5d4b..01588b66a38c 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -1832,7 +1832,6 @@ int bnx2x_config_rss_pf(struct bnx2x *bp, struct bnx2x_rss_config_obj *rss_obj, bool config_hash) { struct bnx2x_config_rss_params params = {NULL}; - int i; /* Although RSS is meaningless when there is a single HW queue we * still need it enabled in order to have HW Rx hash generated. @@ -1864,9 +1863,7 @@ int bnx2x_config_rss_pf(struct bnx2x *bp, struct bnx2x_rss_config_obj *rss_obj, if (config_hash) { /* RSS keys */ - for (i = 0; i < sizeof(params.rss_key) / 4; i++) - params.rss_key[i] = random32(); - + prandom_bytes(params.rss_key, sizeof(params.rss_key)); __set_bit(BNX2X_RSS_SET_SRCH, ¶ms.rss_flags); } diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index abf26c7c1d19..3bc1912afba9 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -616,7 +616,7 @@ static inline bool be_error(struct be_adapter *adapter) return adapter->eeh_error || adapter->hw_error || adapter->fw_timeout; } -static inline bool be_crit_error(struct be_adapter *adapter) +static inline bool be_hw_error(struct be_adapter *adapter) { return adapter->eeh_error || adapter->hw_error; } diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index f2875aa47661..8a250c38fb82 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -298,7 +298,12 @@ void be_async_mcc_enable(struct be_adapter *adapter) void be_async_mcc_disable(struct be_adapter *adapter) { + spin_lock_bh(&adapter->mcc_cq_lock); + adapter->mcc_obj.rearm_cq = false; + be_cq_notify(adapter, adapter->mcc_obj.cq.id, false, 0); + + spin_unlock_bh(&adapter->mcc_cq_lock); } int be_process_mcc(struct be_adapter *adapter) diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index f95612b907ae..9dca22be8125 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -1689,15 +1689,41 @@ static void be_rx_cq_clean(struct be_rx_obj *rxo) struct be_queue_info *rxq = &rxo->q; struct be_queue_info *rx_cq = &rxo->cq; struct be_rx_compl_info *rxcp; + struct be_adapter *adapter = rxo->adapter; + int flush_wait = 0; u16 tail; - /* First cleanup pending rx completions */ - while ((rxcp = be_rx_compl_get(rxo)) != NULL) { - be_rx_compl_discard(rxo, rxcp); - be_cq_notify(rxo->adapter, rx_cq->id, false, 1); + /* Consume pending rx completions. + * Wait for the flush completion (identified by zero num_rcvd) + * to arrive. Notify CQ even when there are no more CQ entries + * for HW to flush partially coalesced CQ entries. + * In Lancer, there is no need to wait for flush compl. + */ + for (;;) { + rxcp = be_rx_compl_get(rxo); + if (rxcp == NULL) { + if (lancer_chip(adapter)) + break; + + if (flush_wait++ > 10 || be_hw_error(adapter)) { + dev_warn(&adapter->pdev->dev, + "did not receive flush compl\n"); + break; + } + be_cq_notify(adapter, rx_cq->id, true, 0); + mdelay(1); + } else { + be_rx_compl_discard(rxo, rxcp); + be_cq_notify(adapter, rx_cq->id, true, 1); + if (rxcp->num_rcvd == 0) + break; + } } - /* Then free posted rx buffer that were not used */ + /* After cleanup, leave the CQ in unarmed state */ + be_cq_notify(adapter, rx_cq->id, false, 0); + + /* Then free posted rx buffers that were not used */ tail = (rxq->head + rxq->len - atomic_read(&rxq->used)) % rxq->len; for (; atomic_read(&rxq->used) > 0; index_inc(&tail, rxq->len)) { page_info = get_rx_page_info(rxo, tail); @@ -2157,7 +2183,7 @@ void be_detect_error(struct be_adapter *adapter) u32 sliport_status = 0, sliport_err1 = 0, sliport_err2 = 0; u32 i; - if (be_crit_error(adapter)) + if (be_hw_error(adapter)) return; if (lancer_chip(adapter)) { @@ -2398,13 +2424,22 @@ static int be_close(struct net_device *netdev) be_roce_dev_close(adapter); - be_async_mcc_disable(adapter); - if (!lancer_chip(adapter)) be_intr_set(adapter, false); - for_all_evt_queues(adapter, eqo, i) { + for_all_evt_queues(adapter, eqo, i) napi_disable(&eqo->napi); + + be_async_mcc_disable(adapter); + + /* Wait for all pending tx completions to arrive so that + * all tx skbs are freed. + */ + be_tx_compl_clean(adapter); + + be_rx_qs_destroy(adapter); + + for_all_evt_queues(adapter, eqo, i) { if (msix_enabled(adapter)) synchronize_irq(be_msix_vec_get(adapter, eqo)); else @@ -2414,12 +2449,6 @@ static int be_close(struct net_device *netdev) be_irq_unregister(adapter); - /* Wait for all pending tx completions to arrive so that - * all tx skbs are freed. - */ - be_tx_compl_clean(adapter); - - be_rx_qs_destroy(adapter); return 0; } diff --git a/drivers/net/ethernet/freescale/Kconfig b/drivers/net/ethernet/freescale/Kconfig index 5ba6e1cbd346..ec490d741fc0 100644 --- a/drivers/net/ethernet/freescale/Kconfig +++ b/drivers/net/ethernet/freescale/Kconfig @@ -94,9 +94,8 @@ config GIANFAR config FEC_PTP bool "PTP Hardware Clock (PHC)" - depends on FEC && ARCH_MXC + depends on FEC && ARCH_MXC && !SOC_IMX25 && !SOC_IMX27 && !SOC_IMX35 && !SOC_IMX5 select PTP_1588_CLOCK - default y if SOC_IMX6Q --help--- Say Y here if you want to use PTP Hardware Clock (PHC) in the driver. Only the basic clock operations have been implemented. diff --git a/drivers/net/ethernet/ibm/ehea/ehea_phyp.h b/drivers/net/ethernet/ibm/ehea/ehea_phyp.h index 8364815c32ff..99b6c2a38dbf 100644 --- a/drivers/net/ethernet/ibm/ehea/ehea_phyp.h +++ b/drivers/net/ethernet/ibm/ehea/ehea_phyp.h @@ -39,26 +39,6 @@ * hcp_* - structures, variables and functions releated to Hypervisor Calls */ -static inline u32 get_longbusy_msecs(int long_busy_ret_code) -{ - switch (long_busy_ret_code) { - case H_LONG_BUSY_ORDER_1_MSEC: - return 1; - case H_LONG_BUSY_ORDER_10_MSEC: - return 10; - case H_LONG_BUSY_ORDER_100_MSEC: - return 100; - case H_LONG_BUSY_ORDER_1_SEC: - return 1000; - case H_LONG_BUSY_ORDER_10_SEC: - return 10000; - case H_LONG_BUSY_ORDER_100_SEC: - return 100000; - default: - return 1; - } -} - /* Number of pages which can be registered at once by H_REGISTER_HEA_RPAGES */ #define EHEA_MAX_RPAGE 512 diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c index 83f0ea929d3d..8ebc352bcbe6 100644 --- a/drivers/net/ethernet/micrel/ksz884x.c +++ b/drivers/net/ethernet/micrel/ksz884x.c @@ -4761,7 +4761,7 @@ static void transmit_cleanup(struct dev_info *hw_priv, int normal) struct ksz_dma_buf *dma_buf; struct net_device *dev = NULL; - spin_lock(&hw_priv->hwlock); + spin_lock_irq(&hw_priv->hwlock); last = info->last; while (info->avail < info->alloc) { @@ -4795,7 +4795,7 @@ static void transmit_cleanup(struct dev_info *hw_priv, int normal) info->avail++; } info->last = last; - spin_unlock(&hw_priv->hwlock); + spin_unlock_irq(&hw_priv->hwlock); /* Notify the network subsystem that the packet has been sent. */ if (dev) @@ -5259,11 +5259,15 @@ static irqreturn_t netdev_intr(int irq, void *dev_id) struct dev_info *hw_priv = priv->adapter; struct ksz_hw *hw = &hw_priv->hw; + spin_lock(&hw_priv->hwlock); + hw_read_intr(hw, &int_enable); /* Not our interrupt! */ - if (!int_enable) + if (!int_enable) { + spin_unlock(&hw_priv->hwlock); return IRQ_NONE; + } do { hw_ack_intr(hw, int_enable); @@ -5310,6 +5314,8 @@ static irqreturn_t netdev_intr(int irq, void *dev_id) hw_ena_intr(hw); + spin_unlock(&hw_priv->hwlock); + return IRQ_HANDLED; } diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h index 537902479689..bc7ec64e9c7a 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h @@ -36,8 +36,8 @@ #define _QLCNIC_LINUX_MAJOR 5 #define _QLCNIC_LINUX_MINOR 0 -#define _QLCNIC_LINUX_SUBVERSION 29 -#define QLCNIC_LINUX_VERSIONID "5.0.29" +#define _QLCNIC_LINUX_SUBVERSION 30 +#define QLCNIC_LINUX_VERSIONID "5.0.30" #define QLCNIC_DRV_IDC_VER 0x01 #define QLCNIC_DRIVER_VERSION ((_QLCNIC_LINUX_MAJOR << 16) |\ (_QLCNIC_LINUX_MINOR << 8) | (_QLCNIC_LINUX_SUBVERSION)) diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c index 58f094ca052e..b14b8f0787ea 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c @@ -134,7 +134,7 @@ int qlcnic_fw_cmd_get_minidump_temp(struct qlcnic_adapter *adapter) __le32 *tmp_buf; struct qlcnic_cmd_args cmd; struct qlcnic_hardware_context *ahw; - struct qlcnic_dump_template_hdr *tmpl_hdr, *tmp_tmpl; + struct qlcnic_dump_template_hdr *tmpl_hdr; dma_addr_t tmp_addr_t = 0; ahw = adapter->ahw; @@ -150,6 +150,8 @@ int qlcnic_fw_cmd_get_minidump_temp(struct qlcnic_adapter *adapter) } temp_size = cmd.rsp.arg2; version = cmd.rsp.arg3; + dev_info(&adapter->pdev->dev, + "minidump template version = 0x%x", version); if (!temp_size) return -EIO; @@ -174,7 +176,6 @@ int qlcnic_fw_cmd_get_minidump_temp(struct qlcnic_adapter *adapter) err = -EIO; goto error; } - tmp_tmpl = tmp_addr; ahw->fw_dump.tmpl_hdr = vzalloc(temp_size); if (!ahw->fw_dump.tmpl_hdr) { err = -EIO; diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c index fc48e000f35f..7a6d5ebe4e0f 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c @@ -365,7 +365,7 @@ static int qlcnic_send_cmd_descs(struct qlcnic_adapter *adapter, struct cmd_desc_type0 *cmd_desc_arr, int nr_desc) { - u32 i, producer, consumer; + u32 i, producer; struct qlcnic_cmd_buffer *pbuf; struct cmd_desc_type0 *cmd_desc; struct qlcnic_host_tx_ring *tx_ring; @@ -379,7 +379,6 @@ qlcnic_send_cmd_descs(struct qlcnic_adapter *adapter, __netif_tx_lock_bh(tx_ring->txq); producer = tx_ring->producer; - consumer = tx_ring->sw_consumer; if (nr_desc >= qlcnic_tx_avail(tx_ring)) { netif_tx_stop_queue(tx_ring->txq); @@ -402,7 +401,7 @@ qlcnic_send_cmd_descs(struct qlcnic_adapter *adapter, pbuf->frag_count = 0; memcpy(&tx_ring->desc_head[producer], - &cmd_desc_arr[i], sizeof(struct cmd_desc_type0)); + cmd_desc, sizeof(struct cmd_desc_type0)); producer = get_next_index(producer, tx_ring->num_desc); i++; diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c index a7554d9aab0c..d833f5927891 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c @@ -445,13 +445,10 @@ static int qlcnic_set_function_modes(struct qlcnic_adapter *adapter) { u8 id; - u32 ref_count; int i, ret = 1; u32 data = QLCNIC_MGMT_FUNC; struct qlcnic_hardware_context *ahw = adapter->ahw; - /* If other drivers are not in use set their privilege level */ - ref_count = QLCRD32(adapter, QLCNIC_CRB_DRV_ACTIVE); ret = qlcnic_api_lock(adapter); if (ret) goto err_lock; @@ -531,11 +528,9 @@ static int qlcnic_setup_pci_map(struct pci_dev *pdev, { u32 offset; void __iomem *mem_ptr0 = NULL; - resource_size_t mem_base; unsigned long mem_len, pci_len0 = 0, bar0_len; /* remap phys address */ - mem_base = pci_resource_start(pdev, 0); /* 0 is for BAR 0 */ mem_len = pci_resource_len(pdev, 0); qlcnic_get_bar_length(pdev->device, &bar0_len); diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c index 12ff29270745..0b8d8625834c 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c @@ -197,7 +197,7 @@ static u32 qlcnic_dump_ctrl(struct qlcnic_adapter *adapter, int i, k, timeout = 0; void __iomem *base = adapter->ahw->pci_base0; u32 addr, data; - u8 opcode, no_ops; + u8 no_ops; struct __ctrl *ctr = &entry->region.ctrl; struct qlcnic_dump_template_hdr *t_hdr = adapter->ahw->fw_dump.tmpl_hdr; @@ -206,7 +206,6 @@ static u32 qlcnic_dump_ctrl(struct qlcnic_adapter *adapter, for (i = 0; i < no_ops; i++) { k = 0; - opcode = 0; for (k = 0; k < 8; k++) { if (!(ctr->opcode & (1 << k))) continue; diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index cb6fc5a743ca..5ac93323a40c 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -577,28 +577,30 @@ static irqreturn_t cp_interrupt (int irq, void *dev_instance) { struct net_device *dev = dev_instance; struct cp_private *cp; + int handled = 0; u16 status; if (unlikely(dev == NULL)) return IRQ_NONE; cp = netdev_priv(dev); + spin_lock(&cp->lock); + status = cpr16(IntrStatus); if (!status || (status == 0xFFFF)) - return IRQ_NONE; + goto out_unlock; + + handled = 1; netif_dbg(cp, intr, dev, "intr, status %04x cmd %02x cpcmd %04x\n", status, cpr8(Cmd), cpr16(CpCmd)); cpw16(IntrStatus, status & ~cp_rx_intr_mask); - spin_lock(&cp->lock); - /* close possible race's with dev_close */ if (unlikely(!netif_running(dev))) { cpw16(IntrMask, 0); - spin_unlock(&cp->lock); - return IRQ_HANDLED; + goto out_unlock; } if (status & (RxOK | RxErr | RxEmpty | RxFIFOOvr)) @@ -612,7 +614,6 @@ static irqreturn_t cp_interrupt (int irq, void *dev_instance) if (status & LinkChg) mii_check_media(&cp->mii_if, netif_msg_link(cp), false); - spin_unlock(&cp->lock); if (status & PciErr) { u16 pci_status; @@ -625,7 +626,10 @@ static irqreturn_t cp_interrupt (int irq, void *dev_instance) /* TODO: reset hardware */ } - return IRQ_HANDLED; +out_unlock: + spin_unlock(&cp->lock); + + return IRQ_RETVAL(handled); } #ifdef CONFIG_NET_POLL_CONTROLLER diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c index 022b45bc14ff..a670d23d9340 100644 --- a/drivers/net/ethernet/smsc/smc91x.c +++ b/drivers/net/ethernet/smsc/smc91x.c @@ -2386,8 +2386,6 @@ static const struct of_device_id smc91x_match[] = { {}, }; MODULE_DEVICE_TABLE(of, smc91x_match); -#else -#define smc91x_match NULL #endif static struct dev_pm_ops smc_drv_pm_ops = { @@ -2402,7 +2400,7 @@ static struct platform_driver smc_driver = { .name = CARDNAME, .owner = THIS_MODULE, .pm = &smc_drv_pm_ops, - .of_match_table = smc91x_match, + .of_match_table = of_match_ptr(smc91x_match), }, }; diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 4616bf27d515..e112877d15d3 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -2575,11 +2575,13 @@ static const struct dev_pm_ops smsc911x_pm_ops = { #define SMSC911X_PM_OPS NULL #endif +#ifdef CONFIG_OF static const struct of_device_id smsc911x_dt_ids[] = { { .compatible = "smsc,lan9115", }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, smsc911x_dt_ids); +#endif static struct platform_driver smsc911x_driver = { .probe = smsc911x_drv_probe, @@ -2588,7 +2590,7 @@ static struct platform_driver smsc911x_driver = { .name = SMSC_CHIPNAME, .owner = THIS_MODULE, .pm = SMSC911X_PM_OPS, - .of_match_table = smsc911x_dt_ids, + .of_match_table = of_match_ptr(smsc911x_dt_ids), }, }; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 023a4fb4efa5..b05df8983be5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -127,14 +127,14 @@ static inline int stmmac_register_platform(void) } static inline void stmmac_unregister_platform(void) { - platform_driver_register(&stmmac_pltfr_driver); + platform_driver_unregister(&stmmac_pltfr_driver); } #else static inline int stmmac_register_platform(void) { pr_debug("stmmac: do not register the platf driver\n"); - return -EINVAL; + return 0; } static inline void stmmac_unregister_platform(void) { @@ -162,7 +162,7 @@ static inline int stmmac_register_pci(void) { pr_debug("stmmac: do not register the PCI driver\n"); - return -EINVAL; + return 0; } static inline void stmmac_unregister_pci(void) { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 542edbcd92c7..f07c0612abf6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -2194,18 +2194,20 @@ int stmmac_restore(struct net_device *ndev) */ static int __init stmmac_init(void) { - int err_plt = 0; - int err_pci = 0; - - err_plt = stmmac_register_platform(); - err_pci = stmmac_register_pci(); - - if ((err_pci) && (err_plt)) { - pr_err("stmmac: driver registration failed\n"); - return -EINVAL; - } + int ret; + ret = stmmac_register_platform(); + if (ret) + goto err; + ret = stmmac_register_pci(); + if (ret) + goto err_pci; return 0; +err_pci: + stmmac_unregister_platform(); +err: + pr_err("stmmac: driver registration failed\n"); + return ret; } static void __exit stmmac_exit(void) diff --git a/drivers/net/ethernet/ti/cpts.c b/drivers/net/ethernet/ti/cpts.c index 337766738eca..5e62c1aeeffb 100644 --- a/drivers/net/ethernet/ti/cpts.c +++ b/drivers/net/ethernet/ti/cpts.c @@ -27,8 +27,6 @@ #include <linux/uaccess.h> #include <linux/workqueue.h> -#include <plat/clock.h> - #include "cpts.h" #ifdef CONFIG_TI_CPTS diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 40b426edc9e6..504f7f1cad94 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -138,6 +138,8 @@ struct tun_file { /* only used for fasnyc */ unsigned int flags; u16 queue_index; + struct list_head next; + struct tun_struct *detached; }; struct tun_flow_entry { @@ -182,6 +184,8 @@ struct tun_struct { struct hlist_head flows[TUN_NUM_FLOW_ENTRIES]; struct timer_list flow_gc_timer; unsigned long ageing_time; + unsigned int numdisabled; + struct list_head disabled; }; static inline u32 tun_hashfn(u32 rxhash) @@ -385,6 +389,23 @@ static void tun_set_real_num_queues(struct tun_struct *tun) netif_set_real_num_rx_queues(tun->dev, tun->numqueues); } +static void tun_disable_queue(struct tun_struct *tun, struct tun_file *tfile) +{ + tfile->detached = tun; + list_add_tail(&tfile->next, &tun->disabled); + ++tun->numdisabled; +} + +static struct tun_struct *tun_enable_queue(struct tun_file *tfile) +{ + struct tun_struct *tun = tfile->detached; + + tfile->detached = NULL; + list_del_init(&tfile->next); + --tun->numdisabled; + return tun; +} + static void __tun_detach(struct tun_file *tfile, bool clean) { struct tun_file *ntfile; @@ -406,20 +427,25 @@ static void __tun_detach(struct tun_file *tfile, bool clean) ntfile->queue_index = index; --tun->numqueues; - sock_put(&tfile->sk); + if (clean) + sock_put(&tfile->sk); + else + tun_disable_queue(tun, tfile); synchronize_net(); tun_flow_delete_by_queue(tun, tun->numqueues + 1); /* Drop read queue */ skb_queue_purge(&tfile->sk.sk_receive_queue); tun_set_real_num_queues(tun); - - if (tun->numqueues == 0 && !(tun->flags & TUN_PERSIST)) - if (dev->reg_state == NETREG_REGISTERED) - unregister_netdevice(dev); - } + } else if (tfile->detached && clean) + tun = tun_enable_queue(tfile); if (clean) { + if (tun && tun->numqueues == 0 && tun->numdisabled == 0 && + !(tun->flags & TUN_PERSIST)) + if (tun->dev->reg_state == NETREG_REGISTERED) + unregister_netdevice(tun->dev); + BUG_ON(!test_bit(SOCK_EXTERNALLY_ALLOCATED, &tfile->socket.flags)); sk_release_kernel(&tfile->sk); @@ -436,7 +462,7 @@ static void tun_detach(struct tun_file *tfile, bool clean) static void tun_detach_all(struct net_device *dev) { struct tun_struct *tun = netdev_priv(dev); - struct tun_file *tfile; + struct tun_file *tfile, *tmp; int i, n = tun->numqueues; for (i = 0; i < n; i++) { @@ -457,6 +483,12 @@ static void tun_detach_all(struct net_device *dev) skb_queue_purge(&tfile->sk.sk_receive_queue); sock_put(&tfile->sk); } + list_for_each_entry_safe(tfile, tmp, &tun->disabled, next) { + tun_enable_queue(tfile); + skb_queue_purge(&tfile->sk.sk_receive_queue); + sock_put(&tfile->sk); + } + BUG_ON(tun->numdisabled != 0); } static int tun_attach(struct tun_struct *tun, struct file *file) @@ -473,7 +505,8 @@ static int tun_attach(struct tun_struct *tun, struct file *file) goto out; err = -E2BIG; - if (tun->numqueues == MAX_TAP_QUEUES) + if (!tfile->detached && + tun->numqueues + tun->numdisabled == MAX_TAP_QUEUES) goto out; err = 0; @@ -487,9 +520,13 @@ static int tun_attach(struct tun_struct *tun, struct file *file) tfile->queue_index = tun->numqueues; rcu_assign_pointer(tfile->tun, tun); rcu_assign_pointer(tun->tfiles[tun->numqueues], tfile); - sock_hold(&tfile->sk); tun->numqueues++; + if (tfile->detached) + tun_enable_queue(tfile); + else + sock_hold(&tfile->sk); + tun_set_real_num_queues(tun); /* device is allowed to go away first, so no need to hold extra @@ -1162,6 +1199,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY; } + skb_reset_network_header(skb); rxhash = skb_get_rxhash(skb); netif_rx_ni(skb); @@ -1349,6 +1387,7 @@ static void tun_free_netdev(struct net_device *dev) { struct tun_struct *tun = netdev_priv(dev); + BUG_ON(!(list_empty(&tun->disabled))); tun_flow_uninit(tun); free_netdev(dev); } @@ -1543,6 +1582,10 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr) err = tun_attach(tun, file); if (err < 0) return err; + + if (tun->flags & TUN_TAP_MQ && + (tun->numqueues + tun->numdisabled > 1)) + return err; } else { char *name; @@ -1601,6 +1644,7 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr) TUN_USER_FEATURES; dev->features = dev->hw_features; + INIT_LIST_HEAD(&tun->disabled); err = tun_attach(tun, file); if (err < 0) goto err_free_dev; @@ -1755,32 +1799,28 @@ static int tun_set_queue(struct file *file, struct ifreq *ifr) { struct tun_file *tfile = file->private_data; struct tun_struct *tun; - struct net_device *dev; int ret = 0; rtnl_lock(); if (ifr->ifr_flags & IFF_ATTACH_QUEUE) { - dev = __dev_get_by_name(tfile->net, ifr->ifr_name); - if (!dev) { - ret = -EINVAL; - goto unlock; - } - - tun = netdev_priv(dev); - if (dev->netdev_ops != &tap_netdev_ops && - dev->netdev_ops != &tun_netdev_ops) + tun = tfile->detached; + if (!tun) ret = -EINVAL; else if (tun_not_capable(tun)) ret = -EPERM; else ret = tun_attach(tun, file); - } else if (ifr->ifr_flags & IFF_DETACH_QUEUE) - __tun_detach(tfile, false); - else + } else if (ifr->ifr_flags & IFF_DETACH_QUEUE) { + tun = rcu_dereference_protected(tfile->tun, + lockdep_rtnl_is_held()); + if (!tun || !(tun->flags & TUN_TAP_MQ)) + ret = -EINVAL; + else + __tun_detach(tfile, false); + } else ret = -EINVAL; -unlock: rtnl_unlock(); return ret; } @@ -2092,6 +2132,7 @@ static int tun_chr_open(struct inode *inode, struct file * file) file->private_data = tfile; set_bit(SOCK_EXTERNALLY_ALLOCATED, &tfile->socket.flags); + INIT_LIST_HEAD(&tfile->next); return 0; } diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c index d0129827602b..3f3d12d766e7 100644 --- a/drivers/net/usb/cdc_ether.c +++ b/drivers/net/usb/cdc_ether.c @@ -457,12 +457,6 @@ int usbnet_cdc_bind(struct usbnet *dev, struct usb_interface *intf) } EXPORT_SYMBOL_GPL(usbnet_cdc_bind); -static int cdc_manage_power(struct usbnet *dev, int on) -{ - dev->intf->needs_remote_wakeup = on; - return 0; -} - static const struct driver_info cdc_info = { .description = "CDC Ethernet Device", .flags = FLAG_ETHER | FLAG_POINTTOPOINT, @@ -470,7 +464,7 @@ static const struct driver_info cdc_info = { .bind = usbnet_cdc_bind, .unbind = usbnet_cdc_unbind, .status = usbnet_cdc_status, - .manage_power = cdc_manage_power, + .manage_power = usbnet_manage_power, }; static const struct driver_info wwan_info = { @@ -479,7 +473,7 @@ static const struct driver_info wwan_info = { .bind = usbnet_cdc_bind, .unbind = usbnet_cdc_unbind, .status = usbnet_cdc_status, - .manage_power = cdc_manage_power, + .manage_power = usbnet_manage_power, }; /*-------------------------------------------------------------------------*/ @@ -487,6 +481,7 @@ static const struct driver_info wwan_info = { #define HUAWEI_VENDOR_ID 0x12D1 #define NOVATEL_VENDOR_ID 0x1410 #define ZTE_VENDOR_ID 0x19D2 +#define DELL_VENDOR_ID 0x413C static const struct usb_device_id products [] = { /* @@ -594,27 +589,29 @@ static const struct usb_device_id products [] = { /* Novatel USB551L and MC551 - handled by qmi_wwan */ { - .match_flags = USB_DEVICE_ID_MATCH_VENDOR - | USB_DEVICE_ID_MATCH_PRODUCT - | USB_DEVICE_ID_MATCH_INT_INFO, - .idVendor = NOVATEL_VENDOR_ID, - .idProduct = 0xB001, - .bInterfaceClass = USB_CLASS_COMM, - .bInterfaceSubClass = USB_CDC_SUBCLASS_ETHERNET, - .bInterfaceProtocol = USB_CDC_PROTO_NONE, + USB_DEVICE_AND_INTERFACE_INFO(NOVATEL_VENDOR_ID, 0xB001, USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE), .driver_info = 0, }, /* Novatel E362 - handled by qmi_wwan */ { - .match_flags = USB_DEVICE_ID_MATCH_VENDOR - | USB_DEVICE_ID_MATCH_PRODUCT - | USB_DEVICE_ID_MATCH_INT_INFO, - .idVendor = NOVATEL_VENDOR_ID, - .idProduct = 0x9010, - .bInterfaceClass = USB_CLASS_COMM, - .bInterfaceSubClass = USB_CDC_SUBCLASS_ETHERNET, - .bInterfaceProtocol = USB_CDC_PROTO_NONE, + USB_DEVICE_AND_INTERFACE_INFO(NOVATEL_VENDOR_ID, 0x9010, USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE), + .driver_info = 0, +}, + +/* Dell Wireless 5800 (Novatel E362) - handled by qmi_wwan */ +{ + USB_DEVICE_AND_INTERFACE_INFO(DELL_VENDOR_ID, 0x8195, USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE), + .driver_info = 0, +}, + +/* Dell Wireless 5800 (Novatel E362) - handled by qmi_wwan */ +{ + USB_DEVICE_AND_INTERFACE_INFO(DELL_VENDOR_ID, 0x8196, USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE), .driver_info = 0, }, diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c index d38bc20a60e2..71b6e92b8e9b 100644 --- a/drivers/net/usb/cdc_ncm.c +++ b/drivers/net/usb/cdc_ncm.c @@ -1129,19 +1129,13 @@ static void cdc_ncm_disconnect(struct usb_interface *intf) usbnet_disconnect(intf); } -static int cdc_ncm_manage_power(struct usbnet *dev, int status) -{ - dev->intf->needs_remote_wakeup = status; - return 0; -} - static const struct driver_info cdc_ncm_info = { .description = "CDC NCM", .flags = FLAG_POINTTOPOINT | FLAG_NO_SETINT | FLAG_MULTI_PACKET, .bind = cdc_ncm_bind, .unbind = cdc_ncm_unbind, .check_connect = cdc_ncm_check_connect, - .manage_power = cdc_ncm_manage_power, + .manage_power = usbnet_manage_power, .status = cdc_ncm_status, .rx_fixup = cdc_ncm_rx_fixup, .tx_fixup = cdc_ncm_tx_fixup, @@ -1155,7 +1149,7 @@ static const struct driver_info wwan_info = { .bind = cdc_ncm_bind, .unbind = cdc_ncm_unbind, .check_connect = cdc_ncm_check_connect, - .manage_power = cdc_ncm_manage_power, + .manage_power = usbnet_manage_power, .status = cdc_ncm_status, .rx_fixup = cdc_ncm_rx_fixup, .tx_fixup = cdc_ncm_tx_fixup, diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index 1ea91f4237f0..91d7cb9728eb 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -383,6 +383,20 @@ static const struct usb_device_id products[] = { USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&qmi_wwan_info, }, + { /* Dell Wireless 5800 (Novatel E362) */ + USB_DEVICE_AND_INTERFACE_INFO(0x413C, 0x8195, + USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, + USB_CDC_PROTO_NONE), + .driver_info = (unsigned long)&qmi_wwan_info, + }, + { /* Dell Wireless 5800 V2 (Novatel E362) */ + USB_DEVICE_AND_INTERFACE_INFO(0x413C, 0x8196, + USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, + USB_CDC_PROTO_NONE), + .driver_info = (unsigned long)&qmi_wwan_info, + }, /* 3. Combined interface devices matching on interface number */ {QMI_FIXED_INTF(0x12d1, 0x140c, 1)}, /* Huawei E173 */ @@ -419,6 +433,7 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x19d2, 0x0199, 1)}, /* ZTE MF820S */ {QMI_FIXED_INTF(0x19d2, 0x0200, 1)}, {QMI_FIXED_INTF(0x19d2, 0x0257, 3)}, /* ZTE MF821 */ + {QMI_FIXED_INTF(0x19d2, 0x0284, 4)}, /* ZTE MF880 */ {QMI_FIXED_INTF(0x19d2, 0x0326, 4)}, /* ZTE MF821D */ {QMI_FIXED_INTF(0x19d2, 0x1008, 4)}, /* ZTE (Vodafone) K3570-Z */ {QMI_FIXED_INTF(0x19d2, 0x1010, 4)}, /* ZTE (Vodafone) K3571-Z */ diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index c04110ba677f..3d4bf01641b4 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -719,7 +719,8 @@ int usbnet_stop (struct net_device *net) dev->flags = 0; del_timer_sync (&dev->delay); tasklet_kill (&dev->bh); - if (info->manage_power) + if (info->manage_power && + !test_and_clear_bit(EVENT_NO_RUNTIME_PM, &dev->flags)) info->manage_power(dev, 0); else usb_autopm_put_interface(dev->intf); @@ -794,14 +795,14 @@ int usbnet_open (struct net_device *net) tasklet_schedule (&dev->bh); if (info->manage_power) { retval = info->manage_power(dev, 1); - if (retval < 0) - goto done_manage_power_error; - usb_autopm_put_interface(dev->intf); + if (retval < 0) { + retval = 0; + set_bit(EVENT_NO_RUNTIME_PM, &dev->flags); + } else { + usb_autopm_put_interface(dev->intf); + } } return retval; - -done_manage_power_error: - clear_bit(EVENT_DEV_OPEN, &dev->flags); done: usb_autopm_put_interface(dev->intf); done_nopm: @@ -1615,6 +1616,16 @@ void usbnet_device_suggests_idle(struct usbnet *dev) } EXPORT_SYMBOL(usbnet_device_suggests_idle); +/* + * For devices that can do without special commands + */ +int usbnet_manage_power(struct usbnet *dev, int on) +{ + dev->intf->needs_remote_wakeup = on; + return 0; +} +EXPORT_SYMBOL(usbnet_manage_power); + /*-------------------------------------------------------------------------*/ static int __usbnet_read_cmd(struct usbnet *dev, u8 cmd, u8 reqtype, u16 value, u16 index, void *data, u16 size) diff --git a/drivers/net/wimax/i2400m/i2400m-usb.h b/drivers/net/wimax/i2400m/i2400m-usb.h index 6650fde99e1d..9f1e947f3557 100644 --- a/drivers/net/wimax/i2400m/i2400m-usb.h +++ b/drivers/net/wimax/i2400m/i2400m-usb.h @@ -152,6 +152,9 @@ enum { /* Device IDs */ USB_DEVICE_ID_I6050 = 0x0186, USB_DEVICE_ID_I6050_2 = 0x0188, + USB_DEVICE_ID_I6150 = 0x07d6, + USB_DEVICE_ID_I6150_2 = 0x07d7, + USB_DEVICE_ID_I6150_3 = 0x07d9, USB_DEVICE_ID_I6250 = 0x0187, }; diff --git a/drivers/net/wimax/i2400m/usb.c b/drivers/net/wimax/i2400m/usb.c index 713d033891e6..080f36303a4f 100644 --- a/drivers/net/wimax/i2400m/usb.c +++ b/drivers/net/wimax/i2400m/usb.c @@ -510,6 +510,9 @@ int i2400mu_probe(struct usb_interface *iface, switch (id->idProduct) { case USB_DEVICE_ID_I6050: case USB_DEVICE_ID_I6050_2: + case USB_DEVICE_ID_I6150: + case USB_DEVICE_ID_I6150_2: + case USB_DEVICE_ID_I6150_3: case USB_DEVICE_ID_I6250: i2400mu->i6050 = 1; break; @@ -759,6 +762,9 @@ static struct usb_device_id i2400mu_id_table[] = { { USB_DEVICE(0x8086, USB_DEVICE_ID_I6050) }, { USB_DEVICE(0x8086, USB_DEVICE_ID_I6050_2) }, + { USB_DEVICE(0x8087, USB_DEVICE_ID_I6150) }, + { USB_DEVICE(0x8087, USB_DEVICE_ID_I6150_2) }, + { USB_DEVICE(0x8087, USB_DEVICE_ID_I6150_3) }, { USB_DEVICE(0x8086, USB_DEVICE_ID_I6250) }, { USB_DEVICE(0x8086, 0x0181) }, { USB_DEVICE(0x8086, 0x1403) }, diff --git a/drivers/net/wireless/Kconfig b/drivers/net/wireless/Kconfig index 6deaae18db57..28aa05f60c26 100644 --- a/drivers/net/wireless/Kconfig +++ b/drivers/net/wireless/Kconfig @@ -156,11 +156,7 @@ config PRISM54 ---help--- This enables support for FullMAC PCI/Cardbus prism54 devices. This driver is now deprecated in favor for the SoftMAC driver, p54pci. - p54pci supports FullMAC PCI/Cardbus devices as well. For details on - the scheduled removal of this driver on the kernel see the feature - removal schedule: - - Documentation/feature-removal-schedule.txt + p54pci supports FullMAC PCI/Cardbus devices as well. For more information refer to the p54 wiki: diff --git a/drivers/net/wireless/Makefile b/drivers/net/wireless/Makefile index 062dfdff6364..67156efe14c4 100644 --- a/drivers/net/wireless/Makefile +++ b/drivers/net/wireless/Makefile @@ -47,7 +47,7 @@ obj-$(CONFIG_RT2X00) += rt2x00/ obj-$(CONFIG_P54_COMMON) += p54/ -obj-$(CONFIG_ATH_COMMON) += ath/ +obj-$(CONFIG_ATH_CARDS) += ath/ obj-$(CONFIG_MAC80211_HWSIM) += mac80211_hwsim.o diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c index 4ffb6a584cd0..44f8b3f3cbed 100644 --- a/drivers/net/wireless/rt2x00/rt2x00dev.c +++ b/drivers/net/wireless/rt2x00/rt2x00dev.c @@ -685,6 +685,14 @@ void rt2x00lib_rxdone(struct queue_entry *entry, gfp_t gfp) * to mac80211. */ rx_status = IEEE80211_SKB_RXCB(entry->skb); + + /* Ensure that all fields of rx_status are initialized + * properly. The skb->cb array was used for driver + * specific informations, so rx_status might contain + * garbage. + */ + memset(rx_status, 0, sizeof(*rx_status)); + rx_status->mactime = rxdesc.timestamp; rx_status->band = rt2x00dev->curr_band; rx_status->freq = rt2x00dev->curr_freq; diff --git a/drivers/of/base.c b/drivers/of/base.c index be846408dbc1..2390ddb22d60 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -629,7 +629,7 @@ struct device_node *of_find_matching_node_and_match(struct device_node *from, read_unlock(&devtree_lock); return np; } -EXPORT_SYMBOL(of_find_matching_node); +EXPORT_SYMBOL(of_find_matching_node_and_match); /** * of_modalias_node - Lookup appropriate modalias for a device node @@ -1114,13 +1114,36 @@ int of_parse_phandle_with_args(const struct device_node *np, const char *list_na } EXPORT_SYMBOL(of_parse_phandle_with_args); +#if defined(CONFIG_OF_DYNAMIC) +static int of_property_notify(int action, struct device_node *np, + struct property *prop) +{ + struct of_prop_reconfig pr; + + pr.dn = np; + pr.prop = prop; + return of_reconfig_notify(action, &pr); +} +#else +static int of_property_notify(int action, struct device_node *np, + struct property *prop) +{ + return 0; +} +#endif + /** - * prom_add_property - Add a property to a node + * of_add_property - Add a property to a node */ -int prom_add_property(struct device_node *np, struct property *prop) +int of_add_property(struct device_node *np, struct property *prop) { struct property **next; unsigned long flags; + int rc; + + rc = of_property_notify(OF_RECONFIG_ADD_PROPERTY, np, prop); + if (rc) + return rc; prop->next = NULL; write_lock_irqsave(&devtree_lock, flags); @@ -1146,18 +1169,23 @@ int prom_add_property(struct device_node *np, struct property *prop) } /** - * prom_remove_property - Remove a property from a node. + * of_remove_property - Remove a property from a node. * * Note that we don't actually remove it, since we have given out * who-knows-how-many pointers to the data using get-property. * Instead we just move the property to the "dead properties" * list, so it won't be found any more. */ -int prom_remove_property(struct device_node *np, struct property *prop) +int of_remove_property(struct device_node *np, struct property *prop) { struct property **next; unsigned long flags; int found = 0; + int rc; + + rc = of_property_notify(OF_RECONFIG_REMOVE_PROPERTY, np, prop); + if (rc) + return rc; write_lock_irqsave(&devtree_lock, flags); next = &np->properties; @@ -1187,7 +1215,7 @@ int prom_remove_property(struct device_node *np, struct property *prop) } /* - * prom_update_property - Update a property in a node, if the property does + * of_update_property - Update a property in a node, if the property does * not exist, add it. * * Note that we don't actually remove it, since we have given out @@ -1195,19 +1223,22 @@ int prom_remove_property(struct device_node *np, struct property *prop) * Instead we just move the property to the "dead properties" list, * and add the new property to the property list */ -int prom_update_property(struct device_node *np, - struct property *newprop) +int of_update_property(struct device_node *np, struct property *newprop) { struct property **next, *oldprop; unsigned long flags; - int found = 0; + int rc, found = 0; + + rc = of_property_notify(OF_RECONFIG_UPDATE_PROPERTY, np, newprop); + if (rc) + return rc; if (!newprop->name) return -EINVAL; oldprop = of_find_property(np, newprop->name, NULL); if (!oldprop) - return prom_add_property(np, newprop); + return of_add_property(np, newprop); write_lock_irqsave(&devtree_lock, flags); next = &np->properties; @@ -1246,12 +1277,55 @@ int prom_update_property(struct device_node *np, * device tree nodes. */ +static BLOCKING_NOTIFIER_HEAD(of_reconfig_chain); + +int of_reconfig_notifier_register(struct notifier_block *nb) +{ + return blocking_notifier_chain_register(&of_reconfig_chain, nb); +} +EXPORT_SYMBOL_GPL(of_reconfig_notifier_register); + +int of_reconfig_notifier_unregister(struct notifier_block *nb) +{ + return blocking_notifier_chain_unregister(&of_reconfig_chain, nb); +} +EXPORT_SYMBOL_GPL(of_reconfig_notifier_unregister); + +int of_reconfig_notify(unsigned long action, void *p) +{ + int rc; + + rc = blocking_notifier_call_chain(&of_reconfig_chain, action, p); + return notifier_to_errno(rc); +} + +#ifdef CONFIG_PROC_DEVICETREE +static void of_add_proc_dt_entry(struct device_node *dn) +{ + struct proc_dir_entry *ent; + + ent = proc_mkdir(strrchr(dn->full_name, '/') + 1, dn->parent->pde); + if (ent) + proc_device_tree_add_node(dn, ent); +} +#else +static void of_add_proc_dt_entry(struct device_node *dn) +{ + return; +} +#endif + /** * of_attach_node - Plug a device node into the tree and global list. */ -void of_attach_node(struct device_node *np) +int of_attach_node(struct device_node *np) { unsigned long flags; + int rc; + + rc = of_reconfig_notify(OF_RECONFIG_ATTACH_NODE, np); + if (rc) + return rc; write_lock_irqsave(&devtree_lock, flags); np->sibling = np->parent->child; @@ -1259,24 +1333,61 @@ void of_attach_node(struct device_node *np) np->parent->child = np; of_allnodes = np; write_unlock_irqrestore(&devtree_lock, flags); + + of_add_proc_dt_entry(np); + return 0; } +#ifdef CONFIG_PROC_DEVICETREE +static void of_remove_proc_dt_entry(struct device_node *dn) +{ + struct device_node *parent = dn->parent; + struct property *prop = dn->properties; + + while (prop) { + remove_proc_entry(prop->name, dn->pde); + prop = prop->next; + } + + if (dn->pde) + remove_proc_entry(dn->pde->name, parent->pde); +} +#else +static void of_remove_proc_dt_entry(struct device_node *dn) +{ + return; +} +#endif + /** * of_detach_node - "Unplug" a node from the device tree. * * The caller must hold a reference to the node. The memory associated with * the node is not freed until its refcount goes to zero. */ -void of_detach_node(struct device_node *np) +int of_detach_node(struct device_node *np) { struct device_node *parent; unsigned long flags; + int rc = 0; + + rc = of_reconfig_notify(OF_RECONFIG_DETACH_NODE, np); + if (rc) + return rc; write_lock_irqsave(&devtree_lock, flags); + if (of_node_check_flag(np, OF_DETACHED)) { + /* someone already detached it */ + write_unlock_irqrestore(&devtree_lock, flags); + return rc; + } + parent = np->parent; - if (!parent) - goto out_unlock; + if (!parent) { + write_unlock_irqrestore(&devtree_lock, flags); + return rc; + } if (of_allnodes == np) of_allnodes = np->allnext; @@ -1301,9 +1412,10 @@ void of_detach_node(struct device_node *np) } of_node_set_flag(np, OF_DETACHED); - -out_unlock: write_unlock_irqrestore(&devtree_lock, flags); + + of_remove_proc_dt_entry(np); + return rc; } #endif /* defined(CONFIG_OF_DYNAMIC) */ diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c index a65c39c473bf..808be06bb67e 100644 --- a/drivers/of/fdt.c +++ b/drivers/of/fdt.c @@ -488,14 +488,8 @@ int __init of_scan_flat_dt(int (*it)(unsigned long node, depth++; pathp = (char *)p; p = ALIGN(p + strlen(pathp) + 1, 4); - if ((*pathp) == '/') { - const char *lp, *np; - for (lp = NULL, np = pathp; *np; np++) - if ((*np) == '/') - lp = np+1; - if (lp != NULL) - pathp = lp; - } + if (*pathp == '/') + pathp = kbasename(pathp); rc = it(p, pathp, depth, data); if (rc != 0) break; diff --git a/drivers/platform/x86/asus-nb-wmi.c b/drivers/platform/x86/asus-nb-wmi.c index 6b0ebdeae916..be790402e0f1 100644 --- a/drivers/platform/x86/asus-nb-wmi.c +++ b/drivers/platform/x86/asus-nb-wmi.c @@ -32,7 +32,7 @@ #define ASUS_NB_WMI_FILE "asus-nb-wmi" -MODULE_AUTHOR("Corentin Chary <corentincj@iksaif.net>"); +MODULE_AUTHOR("Corentin Chary <corentin.chary@gmail.com>"); MODULE_DESCRIPTION("Asus Notebooks WMI Hotkey Driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index c0e9ff489b24..f80ae4d10f68 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -51,7 +51,7 @@ #include "asus-wmi.h" -MODULE_AUTHOR("Corentin Chary <corentincj@iksaif.net>, " +MODULE_AUTHOR("Corentin Chary <corentin.chary@gmail.com>, " "Yong Wang <yong.y.wang@intel.com>"); MODULE_DESCRIPTION("Asus Generic WMI Driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/platform/x86/eeepc-wmi.c b/drivers/platform/x86/eeepc-wmi.c index 5838332ea5bd..60cb76a5b513 100644 --- a/drivers/platform/x86/eeepc-wmi.c +++ b/drivers/platform/x86/eeepc-wmi.c @@ -39,7 +39,7 @@ #define EEEPC_WMI_FILE "eeepc-wmi" -MODULE_AUTHOR("Corentin Chary <corentincj@iksaif.net>"); +MODULE_AUTHOR("Corentin Chary <corentin.chary@gmail.com>"); MODULE_DESCRIPTION("Eee PC WMI Hotkey Driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/power/charger-manager.c b/drivers/power/charger-manager.c index adb3a4b59cb3..6ba047f5ac2c 100644 --- a/drivers/power/charger-manager.c +++ b/drivers/power/charger-manager.c @@ -239,44 +239,37 @@ static bool is_full_charged(struct charger_manager *cm) int uV; /* If there is no battery, it cannot be charged */ - if (!is_batt_present(cm)) { - val.intval = 0; - goto out; - } + if (!is_batt_present(cm)) + return false; if (cm->fuel_gauge && desc->fullbatt_full_capacity > 0) { + val.intval = 0; + /* Not full if capacity of fuel gauge isn't full */ ret = cm->fuel_gauge->get_property(cm->fuel_gauge, POWER_SUPPLY_PROP_CHARGE_FULL, &val); - if (!ret && val.intval > desc->fullbatt_full_capacity) { - val.intval = 1; - goto out; - } + if (!ret && val.intval > desc->fullbatt_full_capacity) + return true; } /* Full, if it's over the fullbatt voltage */ if (desc->fullbatt_uV > 0) { ret = get_batt_uV(cm, &uV); - if (!ret && uV >= desc->fullbatt_uV) { - val.intval = 1; - goto out; - } + if (!ret && uV >= desc->fullbatt_uV) + return true; } /* Full, if the capacity is more than fullbatt_soc */ if (cm->fuel_gauge && desc->fullbatt_soc > 0) { + val.intval = 0; + ret = cm->fuel_gauge->get_property(cm->fuel_gauge, POWER_SUPPLY_PROP_CAPACITY, &val); - if (!ret && val.intval >= desc->fullbatt_soc) { - val.intval = 1; - goto out; - } + if (!ret && val.intval >= desc->fullbatt_soc) + return true; } - val.intval = 0; - -out: - return val.intval ? true : false; + return false; } /** @@ -489,8 +482,9 @@ static void fullbatt_vchk(struct work_struct *work) return; } - diff = desc->fullbatt_uV; - diff -= batt_uV; + diff = desc->fullbatt_uV - batt_uV; + if (diff < 0) + return; dev_info(cm->dev, "VBATT dropped %duV after full-batt.\n", diff); diff --git a/drivers/pwm/Kconfig b/drivers/pwm/Kconfig index ed81720e7b2b..e513cd998170 100644 --- a/drivers/pwm/Kconfig +++ b/drivers/pwm/Kconfig @@ -112,6 +112,17 @@ config PWM_SAMSUNG To compile this driver as a module, choose M here: the module will be called pwm-samsung. +config PWM_SPEAR + tristate "STMicroelectronics SPEAr PWM support" + depends on PLAT_SPEAR + depends on OF + help + Generic PWM framework driver for the PWM controller on ST + SPEAr SoCs. + + To compile this driver as a module, choose M here: the module + will be called pwm-spear. + config PWM_TEGRA tristate "NVIDIA Tegra PWM support" depends on ARCH_TEGRA @@ -125,6 +136,7 @@ config PWM_TEGRA config PWM_TIECAP tristate "ECAP PWM support" depends on SOC_AM33XX + select PWM_TIPWMSS help PWM driver support for the ECAP APWM controller found on AM33XX TI SOC @@ -135,6 +147,7 @@ config PWM_TIECAP config PWM_TIEHRPWM tristate "EHRPWM PWM support" depends on SOC_AM33XX + select PWM_TIPWMSS help PWM driver support for the EHRPWM controller found on AM33XX TI SOC @@ -142,14 +155,32 @@ config PWM_TIEHRPWM To compile this driver as a module, choose M here: the module will be called pwm-tiehrpwm. -config PWM_TWL6030 - tristate "TWL6030 PWM support" +config PWM_TIPWMSS + bool + depends on SOC_AM33XX && (PWM_TIEHRPWM || PWM_TIECAP) + help + PWM Subsystem driver support for AM33xx SOC. + + PWM submodules require PWM config space access from submodule + drivers and require common parent driver support. + +config PWM_TWL + tristate "TWL4030/6030 PWM support" + depends on TWL4030_CORE + help + Generic PWM framework driver for TWL4030/6030. + + To compile this driver as a module, choose M here: the module + will be called pwm-twl. + +config PWM_TWL_LED + tristate "TWL4030/6030 PWM support for LED drivers" depends on TWL4030_CORE help - Generic PWM framework driver for TWL6030. + Generic PWM framework driver for TWL4030/6030 LED terminals. To compile this driver as a module, choose M here: the module - will be called pwm-twl6030. + will be called pwm-twl-led. config PWM_VT8500 tristate "vt8500 pwm support" diff --git a/drivers/pwm/Makefile b/drivers/pwm/Makefile index acfe4821c58b..62a2963cfe58 100644 --- a/drivers/pwm/Makefile +++ b/drivers/pwm/Makefile @@ -8,8 +8,11 @@ obj-$(CONFIG_PWM_MXS) += pwm-mxs.o obj-$(CONFIG_PWM_PUV3) += pwm-puv3.o obj-$(CONFIG_PWM_PXA) += pwm-pxa.o obj-$(CONFIG_PWM_SAMSUNG) += pwm-samsung.o +obj-$(CONFIG_PWM_SPEAR) += pwm-spear.o obj-$(CONFIG_PWM_TEGRA) += pwm-tegra.o obj-$(CONFIG_PWM_TIECAP) += pwm-tiecap.o obj-$(CONFIG_PWM_TIEHRPWM) += pwm-tiehrpwm.o -obj-$(CONFIG_PWM_TWL6030) += pwm-twl6030.o +obj-$(CONFIG_PWM_TIPWMSS) += pwm-tipwmss.o +obj-$(CONFIG_PWM_TWL) += pwm-twl.o +obj-$(CONFIG_PWM_TWL_LED) += pwm-twl-led.o obj-$(CONFIG_PWM_VT8500) += pwm-vt8500.o diff --git a/drivers/pwm/core.c b/drivers/pwm/core.c index f5acdaa52707..903138b18842 100644 --- a/drivers/pwm/core.c +++ b/drivers/pwm/core.c @@ -32,6 +32,9 @@ #define MAX_PWMS 1024 +/* flags in the third cell of the DT PWM specifier */ +#define PWM_SPEC_POLARITY (1 << 0) + static DEFINE_MUTEX(pwm_lookup_lock); static LIST_HEAD(pwm_lookup_list); static DEFINE_MUTEX(pwm_lock); @@ -129,6 +132,32 @@ static int pwm_device_request(struct pwm_device *pwm, const char *label) return 0; } +struct pwm_device * +of_pwm_xlate_with_flags(struct pwm_chip *pc, const struct of_phandle_args *args) +{ + struct pwm_device *pwm; + + if (pc->of_pwm_n_cells < 3) + return ERR_PTR(-EINVAL); + + if (args->args[0] >= pc->npwm) + return ERR_PTR(-EINVAL); + + pwm = pwm_request_from_chip(pc, args->args[0], NULL); + if (IS_ERR(pwm)) + return pwm; + + pwm_set_period(pwm, args->args[1]); + + if (args->args[2] & PWM_SPEC_POLARITY) + pwm_set_polarity(pwm, PWM_POLARITY_INVERSED); + else + pwm_set_polarity(pwm, PWM_POLARITY_NORMAL); + + return pwm; +} +EXPORT_SYMBOL_GPL(of_pwm_xlate_with_flags); + static struct pwm_device * of_pwm_simple_xlate(struct pwm_chip *pc, const struct of_phandle_args *args) { diff --git a/drivers/pwm/pwm-imx.c b/drivers/pwm/pwm-imx.c index 8f26e9fcea97..65a86bdeabed 100644 --- a/drivers/pwm/pwm-imx.c +++ b/drivers/pwm/pwm-imx.c @@ -235,7 +235,7 @@ static int imx_pwm_probe(struct platform_device *pdev) { const struct of_device_id *of_id = of_match_device(imx_pwm_dt_ids, &pdev->dev); - struct imx_pwm_data *data; + const struct imx_pwm_data *data; struct imx_chip *imx; struct resource *r; int ret = 0; diff --git a/drivers/pwm/pwm-lpc32xx.c b/drivers/pwm/pwm-lpc32xx.c index 015a82235620..14106440294f 100644 --- a/drivers/pwm/pwm-lpc32xx.c +++ b/drivers/pwm/pwm-lpc32xx.c @@ -49,9 +49,24 @@ static int lpc32xx_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm, c = 0; /* 0 set division by 256 */ period_cycles = c; + /* The duty-cycle value is as follows: + * + * DUTY-CYCLE HIGH LEVEL + * 1 99.9% + * 25 90.0% + * 128 50.0% + * 220 10.0% + * 255 0.1% + * 0 0.0% + * + * In other words, the register value is duty-cycle % 256 with + * duty-cycle in the range 1-256. + */ c = 256 * duty_ns; do_div(c, period_ns); - duty_cycles = c; + if (c > 255) + c = 255; + duty_cycles = 256 - c; writel(PWM_ENABLE | PWM_RELOADV(period_cycles) | PWM_DUTY(duty_cycles), lpc32xx->base + (pwm->hwpwm << 2)); @@ -106,6 +121,7 @@ static int lpc32xx_pwm_probe(struct platform_device *pdev) lpc32xx->chip.dev = &pdev->dev; lpc32xx->chip.ops = &lpc32xx_pwm_ops; lpc32xx->chip.npwm = 2; + lpc32xx->chip.base = -1; ret = pwmchip_add(&lpc32xx->chip); if (ret < 0) { @@ -121,8 +137,11 @@ static int lpc32xx_pwm_probe(struct platform_device *pdev) static int lpc32xx_pwm_remove(struct platform_device *pdev) { struct lpc32xx_pwm_chip *lpc32xx = platform_get_drvdata(pdev); + unsigned int i; + + for (i = 0; i < lpc32xx->chip.npwm; i++) + pwm_disable(&lpc32xx->chip.pwms[i]); - clk_disable(lpc32xx->clk); return pwmchip_remove(&lpc32xx->chip); } diff --git a/drivers/pwm/pwm-samsung.c b/drivers/pwm/pwm-samsung.c index e9b15d099c03..5207e6cd8648 100644 --- a/drivers/pwm/pwm-samsung.c +++ b/drivers/pwm/pwm-samsung.c @@ -222,6 +222,7 @@ static int s3c_pwm_probe(struct platform_device *pdev) /* calculate base of control bits in TCON */ s3c->tcon_base = id == 0 ? 0 : (id * 4) + 4; + s3c->pwm_id = id; s3c->chip.dev = &pdev->dev; s3c->chip.ops = &s3c_pwm_ops; s3c->chip.base = -1; diff --git a/drivers/pwm/pwm-spear.c b/drivers/pwm/pwm-spear.c new file mode 100644 index 000000000000..83b21d9d5cf9 --- /dev/null +++ b/drivers/pwm/pwm-spear.c @@ -0,0 +1,276 @@ +/* + * ST Microelectronics SPEAr Pulse Width Modulator driver + * + * Copyright (C) 2012 ST Microelectronics + * Shiraz Hashim <shiraz.hashim@st.com> + * + * This file is licensed under the terms of the GNU General Public + * License version 2. This program is licensed "as is" without any + * warranty of any kind, whether express or implied. + */ + +#include <linux/clk.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/ioport.h> +#include <linux/kernel.h> +#include <linux/math64.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/platform_device.h> +#include <linux/pwm.h> +#include <linux/slab.h> +#include <linux/types.h> + +#define NUM_PWM 4 + +/* PWM registers and bits definitions */ +#define PWMCR 0x00 /* Control Register */ +#define PWMCR_PWM_ENABLE 0x1 +#define PWMCR_PRESCALE_SHIFT 2 +#define PWMCR_MIN_PRESCALE 0x00 +#define PWMCR_MAX_PRESCALE 0x3FFF + +#define PWMDCR 0x04 /* Duty Cycle Register */ +#define PWMDCR_MIN_DUTY 0x0001 +#define PWMDCR_MAX_DUTY 0xFFFF + +#define PWMPCR 0x08 /* Period Register */ +#define PWMPCR_MIN_PERIOD 0x0001 +#define PWMPCR_MAX_PERIOD 0xFFFF + +/* Following only available on 13xx SoCs */ +#define PWMMCR 0x3C /* Master Control Register */ +#define PWMMCR_PWM_ENABLE 0x1 + +/** + * struct spear_pwm_chip - struct representing pwm chip + * + * @mmio_base: base address of pwm chip + * @clk: pointer to clk structure of pwm chip + * @chip: linux pwm chip representation + * @dev: pointer to device structure of pwm chip + */ +struct spear_pwm_chip { + void __iomem *mmio_base; + struct clk *clk; + struct pwm_chip chip; + struct device *dev; +}; + +static inline struct spear_pwm_chip *to_spear_pwm_chip(struct pwm_chip *chip) +{ + return container_of(chip, struct spear_pwm_chip, chip); +} + +static inline u32 spear_pwm_readl(struct spear_pwm_chip *chip, unsigned int num, + unsigned long offset) +{ + return readl_relaxed(chip->mmio_base + (num << 4) + offset); +} + +static inline void spear_pwm_writel(struct spear_pwm_chip *chip, + unsigned int num, unsigned long offset, + unsigned long val) +{ + writel_relaxed(val, chip->mmio_base + (num << 4) + offset); +} + +static int spear_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm, + int duty_ns, int period_ns) +{ + struct spear_pwm_chip *pc = to_spear_pwm_chip(chip); + u64 val, div, clk_rate; + unsigned long prescale = PWMCR_MIN_PRESCALE, pv, dc; + int ret; + + /* + * Find pv, dc and prescale to suit duty_ns and period_ns. This is done + * according to formulas described below: + * + * period_ns = 10^9 * (PRESCALE + 1) * PV / PWM_CLK_RATE + * duty_ns = 10^9 * (PRESCALE + 1) * DC / PWM_CLK_RATE + * + * PV = (PWM_CLK_RATE * period_ns) / (10^9 * (PRESCALE + 1)) + * DC = (PWM_CLK_RATE * duty_ns) / (10^9 * (PRESCALE + 1)) + */ + clk_rate = clk_get_rate(pc->clk); + while (1) { + div = 1000000000; + div *= 1 + prescale; + val = clk_rate * period_ns; + pv = div64_u64(val, div); + val = clk_rate * duty_ns; + dc = div64_u64(val, div); + + /* if duty_ns and period_ns are not achievable then return */ + if (pv < PWMPCR_MIN_PERIOD || dc < PWMDCR_MIN_DUTY) + return -EINVAL; + + /* + * if pv and dc have crossed their upper limit, then increase + * prescale and recalculate pv and dc. + */ + if (pv > PWMPCR_MAX_PERIOD || dc > PWMDCR_MAX_DUTY) { + if (++prescale > PWMCR_MAX_PRESCALE) + return -EINVAL; + continue; + } + break; + } + + /* + * NOTE: the clock to PWM has to be enabled first before writing to the + * registers. + */ + ret = clk_enable(pc->clk); + if (ret) + return ret; + + spear_pwm_writel(pc, pwm->hwpwm, PWMCR, + prescale << PWMCR_PRESCALE_SHIFT); + spear_pwm_writel(pc, pwm->hwpwm, PWMDCR, dc); + spear_pwm_writel(pc, pwm->hwpwm, PWMPCR, pv); + clk_disable(pc->clk); + + return 0; +} + +static int spear_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct spear_pwm_chip *pc = to_spear_pwm_chip(chip); + int rc = 0; + u32 val; + + rc = clk_enable(pc->clk); + if (!rc) + return rc; + + val = spear_pwm_readl(pc, pwm->hwpwm, PWMCR); + val |= PWMCR_PWM_ENABLE; + spear_pwm_writel(pc, pwm->hwpwm, PWMCR, val); + + return 0; +} + +static void spear_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct spear_pwm_chip *pc = to_spear_pwm_chip(chip); + u32 val; + + val = spear_pwm_readl(pc, pwm->hwpwm, PWMCR); + val &= ~PWMCR_PWM_ENABLE; + spear_pwm_writel(pc, pwm->hwpwm, PWMCR, val); + + clk_disable(pc->clk); +} + +static const struct pwm_ops spear_pwm_ops = { + .config = spear_pwm_config, + .enable = spear_pwm_enable, + .disable = spear_pwm_disable, + .owner = THIS_MODULE, +}; + +static int spear_pwm_probe(struct platform_device *pdev) +{ + struct device_node *np = pdev->dev.of_node; + struct spear_pwm_chip *pc; + struct resource *r; + int ret; + u32 val; + + r = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!r) { + dev_err(&pdev->dev, "no memory resources defined\n"); + return -ENODEV; + } + + pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL); + if (!pc) { + dev_err(&pdev->dev, "failed to allocate memory\n"); + return -ENOMEM; + } + + pc->mmio_base = devm_request_and_ioremap(&pdev->dev, r); + if (!pc->mmio_base) + return -EADDRNOTAVAIL; + + pc->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(pc->clk)) + return PTR_ERR(pc->clk); + + pc->dev = &pdev->dev; + platform_set_drvdata(pdev, pc); + + pc->chip.dev = &pdev->dev; + pc->chip.ops = &spear_pwm_ops; + pc->chip.base = -1; + pc->chip.npwm = NUM_PWM; + + ret = clk_prepare(pc->clk); + if (!ret) + return ret; + + if (of_device_is_compatible(np, "st,spear1340-pwm")) { + ret = clk_enable(pc->clk); + if (!ret) { + clk_unprepare(pc->clk); + return ret; + } + /* + * Following enables PWM chip, channels would still be + * enabled individually through their control register + */ + val = readl_relaxed(pc->mmio_base + PWMMCR); + val |= PWMMCR_PWM_ENABLE; + writel_relaxed(val, pc->mmio_base + PWMMCR); + + clk_disable(pc->clk); + } + + ret = pwmchip_add(&pc->chip); + if (!ret) { + clk_unprepare(pc->clk); + dev_err(&pdev->dev, "pwmchip_add() failed: %d\n", ret); + } + + return ret; +} + +static int spear_pwm_remove(struct platform_device *pdev) +{ + struct spear_pwm_chip *pc = platform_get_drvdata(pdev); + int i; + + for (i = 0; i < NUM_PWM; i++) + pwm_disable(&pc->chip.pwms[i]); + + /* clk was prepared in probe, hence unprepare it here */ + clk_unprepare(pc->clk); + return pwmchip_remove(&pc->chip); +} + +static struct of_device_id spear_pwm_of_match[] = { + { .compatible = "st,spear320-pwm" }, + { .compatible = "st,spear1340-pwm" }, + { } +}; + +MODULE_DEVICE_TABLE(of, spear_pwm_of_match); + +static struct platform_driver spear_pwm_driver = { + .driver = { + .name = "spear-pwm", + .of_match_table = spear_pwm_of_match, + }, + .probe = spear_pwm_probe, + .remove = spear_pwm_remove, +}; + +module_platform_driver(spear_pwm_driver); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Shiraz Hashim <shiraz.hashim@st.com>"); +MODULE_AUTHOR("Viresh Kumar <viresh.kumar@linaro.com>"); +MODULE_ALIAS("platform:spear-pwm"); diff --git a/drivers/pwm/pwm-tiecap.c b/drivers/pwm/pwm-tiecap.c index 87c091b245cc..5cf016dd9822 100644 --- a/drivers/pwm/pwm-tiecap.c +++ b/drivers/pwm/pwm-tiecap.c @@ -25,6 +25,10 @@ #include <linux/clk.h> #include <linux/pm_runtime.h> #include <linux/pwm.h> +#include <linux/of_device.h> +#include <linux/pinctrl/consumer.h> + +#include "pwm-tipwmss.h" /* ECAP registers and bits definitions */ #define CAP1 0x08 @@ -184,12 +188,24 @@ static const struct pwm_ops ecap_pwm_ops = { .owner = THIS_MODULE, }; +static const struct of_device_id ecap_of_match[] = { + { .compatible = "ti,am33xx-ecap" }, + {}, +}; +MODULE_DEVICE_TABLE(of, ecap_of_match); + static int ecap_pwm_probe(struct platform_device *pdev) { int ret; struct resource *r; struct clk *clk; struct ecap_pwm_chip *pc; + u16 status; + struct pinctrl *pinctrl; + + pinctrl = devm_pinctrl_get_select_default(&pdev->dev); + if (IS_ERR(pinctrl)) + dev_warn(&pdev->dev, "unable to select pin group\n"); pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL); if (!pc) { @@ -211,6 +227,8 @@ static int ecap_pwm_probe(struct platform_device *pdev) pc->chip.dev = &pdev->dev; pc->chip.ops = &ecap_pwm_ops; + pc->chip.of_xlate = of_pwm_xlate_with_flags; + pc->chip.of_pwm_n_cells = 3; pc->chip.base = -1; pc->chip.npwm = 1; @@ -231,14 +249,40 @@ static int ecap_pwm_probe(struct platform_device *pdev) } pm_runtime_enable(&pdev->dev); + pm_runtime_get_sync(&pdev->dev); + + status = pwmss_submodule_state_change(pdev->dev.parent, + PWMSS_ECAPCLK_EN); + if (!(status & PWMSS_ECAPCLK_EN_ACK)) { + dev_err(&pdev->dev, "PWMSS config space clock enable failed\n"); + ret = -EINVAL; + goto pwmss_clk_failure; + } + + pm_runtime_put_sync(&pdev->dev); + platform_set_drvdata(pdev, pc); return 0; + +pwmss_clk_failure: + pm_runtime_put_sync(&pdev->dev); + pm_runtime_disable(&pdev->dev); + pwmchip_remove(&pc->chip); + return ret; } static int ecap_pwm_remove(struct platform_device *pdev) { struct ecap_pwm_chip *pc = platform_get_drvdata(pdev); + pm_runtime_get_sync(&pdev->dev); + /* + * Due to hardware misbehaviour, acknowledge of the stop_req + * is missing. Hence checking of the status bit skipped. + */ + pwmss_submodule_state_change(pdev->dev.parent, PWMSS_ECAPCLK_STOP_REQ); + pm_runtime_put_sync(&pdev->dev); + pm_runtime_put_sync(&pdev->dev); pm_runtime_disable(&pdev->dev); return pwmchip_remove(&pc->chip); @@ -246,7 +290,9 @@ static int ecap_pwm_remove(struct platform_device *pdev) static struct platform_driver ecap_pwm_driver = { .driver = { - .name = "ecap", + .name = "ecap", + .owner = THIS_MODULE, + .of_match_table = ecap_of_match, }, .probe = ecap_pwm_probe, .remove = ecap_pwm_remove, diff --git a/drivers/pwm/pwm-tiehrpwm.c b/drivers/pwm/pwm-tiehrpwm.c index 9ffd389d0c8b..72a6dd40c9ec 100644 --- a/drivers/pwm/pwm-tiehrpwm.c +++ b/drivers/pwm/pwm-tiehrpwm.c @@ -25,6 +25,10 @@ #include <linux/err.h> #include <linux/clk.h> #include <linux/pm_runtime.h> +#include <linux/of_device.h> +#include <linux/pinctrl/consumer.h> + +#include "pwm-tipwmss.h" /* EHRPWM registers and bits definitions */ @@ -115,6 +119,7 @@ struct ehrpwm_pwm_chip { void __iomem *mmio_base; unsigned long period_cycles[NUM_PWM_CHANNEL]; enum pwm_polarity polarity[NUM_PWM_CHANNEL]; + struct clk *tbclk; }; static inline struct ehrpwm_pwm_chip *to_ehrpwm_pwm_chip(struct pwm_chip *chip) @@ -335,6 +340,9 @@ static int ehrpwm_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm) /* Channels polarity can be configured from action qualifier module */ configure_polarity(pc, pwm->hwpwm); + /* Enable TBCLK before enabling PWM device */ + clk_enable(pc->tbclk); + /* Enable time counter for free_run */ ehrpwm_modify(pc->mmio_base, TBCTL, TBCTL_RUN_MASK, TBCTL_FREE_RUN); return 0; @@ -363,6 +371,9 @@ static void ehrpwm_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm) ehrpwm_modify(pc->mmio_base, AQCSFRC, aqcsfrc_mask, aqcsfrc_val); + /* Disabling TBCLK on PWM disable */ + clk_disable(pc->tbclk); + /* Stop Time base counter */ ehrpwm_modify(pc->mmio_base, TBCTL, TBCTL_RUN_MASK, TBCTL_STOP_NEXT); @@ -392,12 +403,24 @@ static const struct pwm_ops ehrpwm_pwm_ops = { .owner = THIS_MODULE, }; +static const struct of_device_id ehrpwm_of_match[] = { + { .compatible = "ti,am33xx-ehrpwm" }, + {}, +}; +MODULE_DEVICE_TABLE(of, ehrpwm_of_match); + static int ehrpwm_pwm_probe(struct platform_device *pdev) { int ret; struct resource *r; struct clk *clk; struct ehrpwm_pwm_chip *pc; + u16 status; + struct pinctrl *pinctrl; + + pinctrl = devm_pinctrl_get_select_default(&pdev->dev); + if (IS_ERR(pinctrl)) + dev_warn(&pdev->dev, "unable to select pin group\n"); pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL); if (!pc) { @@ -419,6 +442,8 @@ static int ehrpwm_pwm_probe(struct platform_device *pdev) pc->chip.dev = &pdev->dev; pc->chip.ops = &ehrpwm_pwm_ops; + pc->chip.of_xlate = of_pwm_xlate_with_flags; + pc->chip.of_pwm_n_cells = 3; pc->chip.base = -1; pc->chip.npwm = NUM_PWM_CHANNEL; @@ -432,6 +457,13 @@ static int ehrpwm_pwm_probe(struct platform_device *pdev) if (!pc->mmio_base) return -EADDRNOTAVAIL; + /* Acquire tbclk for Time Base EHRPWM submodule */ + pc->tbclk = devm_clk_get(&pdev->dev, "tbclk"); + if (IS_ERR(pc->tbclk)) { + dev_err(&pdev->dev, "Failed to get tbclk\n"); + return PTR_ERR(pc->tbclk); + } + ret = pwmchip_add(&pc->chip); if (ret < 0) { dev_err(&pdev->dev, "pwmchip_add() failed: %d\n", ret); @@ -439,14 +471,40 @@ static int ehrpwm_pwm_probe(struct platform_device *pdev) } pm_runtime_enable(&pdev->dev); + pm_runtime_get_sync(&pdev->dev); + + status = pwmss_submodule_state_change(pdev->dev.parent, + PWMSS_EPWMCLK_EN); + if (!(status & PWMSS_EPWMCLK_EN_ACK)) { + dev_err(&pdev->dev, "PWMSS config space clock enable failed\n"); + ret = -EINVAL; + goto pwmss_clk_failure; + } + + pm_runtime_put_sync(&pdev->dev); + platform_set_drvdata(pdev, pc); return 0; + +pwmss_clk_failure: + pm_runtime_put_sync(&pdev->dev); + pm_runtime_disable(&pdev->dev); + pwmchip_remove(&pc->chip); + return ret; } static int ehrpwm_pwm_remove(struct platform_device *pdev) { struct ehrpwm_pwm_chip *pc = platform_get_drvdata(pdev); + pm_runtime_get_sync(&pdev->dev); + /* + * Due to hardware misbehaviour, acknowledge of the stop_req + * is missing. Hence checking of the status bit skipped. + */ + pwmss_submodule_state_change(pdev->dev.parent, PWMSS_EPWMCLK_STOP_REQ); + pm_runtime_put_sync(&pdev->dev); + pm_runtime_put_sync(&pdev->dev); pm_runtime_disable(&pdev->dev); return pwmchip_remove(&pc->chip); @@ -454,7 +512,9 @@ static int ehrpwm_pwm_remove(struct platform_device *pdev) static struct platform_driver ehrpwm_pwm_driver = { .driver = { - .name = "ehrpwm", + .name = "ehrpwm", + .owner = THIS_MODULE, + .of_match_table = ehrpwm_of_match, }, .probe = ehrpwm_pwm_probe, .remove = ehrpwm_pwm_remove, diff --git a/drivers/pwm/pwm-tipwmss.c b/drivers/pwm/pwm-tipwmss.c new file mode 100644 index 000000000000..3448a1c88590 --- /dev/null +++ b/drivers/pwm/pwm-tipwmss.c @@ -0,0 +1,139 @@ +/* + * TI PWM Subsystem driver + * + * Copyright (C) 2012 Texas Instruments Incorporated - http://www.ti.com/ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/io.h> +#include <linux/err.h> +#include <linux/pm_runtime.h> +#include <linux/of_device.h> + +#include "pwm-tipwmss.h" + +#define PWMSS_CLKCONFIG 0x8 /* Clock gating reg */ +#define PWMSS_CLKSTATUS 0xc /* Clock gating status reg */ + +struct pwmss_info { + void __iomem *mmio_base; + struct mutex pwmss_lock; + u16 pwmss_clkconfig; +}; + +u16 pwmss_submodule_state_change(struct device *dev, int set) +{ + struct pwmss_info *info = dev_get_drvdata(dev); + u16 val; + + mutex_lock(&info->pwmss_lock); + val = readw(info->mmio_base + PWMSS_CLKCONFIG); + val |= set; + writew(val , info->mmio_base + PWMSS_CLKCONFIG); + mutex_unlock(&info->pwmss_lock); + + return readw(info->mmio_base + PWMSS_CLKSTATUS); +} +EXPORT_SYMBOL(pwmss_submodule_state_change); + +static const struct of_device_id pwmss_of_match[] = { + { .compatible = "ti,am33xx-pwmss" }, + {}, +}; +MODULE_DEVICE_TABLE(of, pwmss_of_match); + +static int pwmss_probe(struct platform_device *pdev) +{ + int ret; + struct resource *r; + struct pwmss_info *info; + struct device_node *node = pdev->dev.of_node; + + info = devm_kzalloc(&pdev->dev, sizeof(*info), GFP_KERNEL); + if (!info) { + dev_err(&pdev->dev, "failed to allocate memory\n"); + return -ENOMEM; + } + + mutex_init(&info->pwmss_lock); + + r = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!r) { + dev_err(&pdev->dev, "no memory resource defined\n"); + return -ENODEV; + } + + info->mmio_base = devm_request_and_ioremap(&pdev->dev, r); + if (!info->mmio_base) + return -EADDRNOTAVAIL; + + pm_runtime_enable(&pdev->dev); + pm_runtime_get_sync(&pdev->dev); + platform_set_drvdata(pdev, info); + + /* Populate all the child nodes here... */ + ret = of_platform_populate(node, NULL, NULL, &pdev->dev); + if (ret) + dev_err(&pdev->dev, "no child node found\n"); + + return ret; +} + +static int pwmss_remove(struct platform_device *pdev) +{ + struct pwmss_info *info = platform_get_drvdata(pdev); + + pm_runtime_put_sync(&pdev->dev); + pm_runtime_disable(&pdev->dev); + mutex_destroy(&info->pwmss_lock); + return 0; +} + +static int pwmss_suspend(struct device *dev) +{ + struct pwmss_info *info = dev_get_drvdata(dev); + + info->pwmss_clkconfig = readw(info->mmio_base + PWMSS_CLKCONFIG); + pm_runtime_put_sync(dev); + return 0; +} + +static int pwmss_resume(struct device *dev) +{ + struct pwmss_info *info = dev_get_drvdata(dev); + + pm_runtime_get_sync(dev); + writew(info->pwmss_clkconfig, info->mmio_base + PWMSS_CLKCONFIG); + return 0; +} + +static SIMPLE_DEV_PM_OPS(pwmss_pm_ops, pwmss_suspend, pwmss_resume); + +static struct platform_driver pwmss_driver = { + .driver = { + .name = "pwmss", + .owner = THIS_MODULE, + .pm = &pwmss_pm_ops, + .of_match_table = pwmss_of_match, + }, + .probe = pwmss_probe, + .remove = pwmss_remove, +}; + +module_platform_driver(pwmss_driver); + +MODULE_DESCRIPTION("PWM Subsystem driver"); +MODULE_AUTHOR("Texas Instruments"); +MODULE_LICENSE("GPL"); diff --git a/drivers/pwm/pwm-tipwmss.h b/drivers/pwm/pwm-tipwmss.h new file mode 100644 index 000000000000..11f76a1e266b --- /dev/null +++ b/drivers/pwm/pwm-tipwmss.h @@ -0,0 +1,39 @@ +/* + * TI PWM Subsystem driver + * + * Copyright (C) 2012 Texas Instruments Incorporated - http://www.ti.com/ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#ifndef __TIPWMSS_H +#define __TIPWMSS_H + +#ifdef CONFIG_PWM_TIPWMSS +/* PWM substem clock gating */ +#define PWMSS_ECAPCLK_EN BIT(0) +#define PWMSS_ECAPCLK_STOP_REQ BIT(1) +#define PWMSS_EPWMCLK_EN BIT(8) +#define PWMSS_EPWMCLK_STOP_REQ BIT(9) + +#define PWMSS_ECAPCLK_EN_ACK BIT(0) +#define PWMSS_EPWMCLK_EN_ACK BIT(8) + +extern u16 pwmss_submodule_state_change(struct device *dev, int set); +#else +static inline u16 pwmss_submodule_state_change(struct device *dev, int set) +{ + /* return success status value */ + return 0xFFFF; +} +#endif +#endif /* __TIPWMSS_H */ diff --git a/drivers/pwm/pwm-twl-led.c b/drivers/pwm/pwm-twl-led.c new file mode 100644 index 000000000000..9dfa0f3eca30 --- /dev/null +++ b/drivers/pwm/pwm-twl-led.c @@ -0,0 +1,344 @@ +/* + * Driver for TWL4030/6030 Pulse Width Modulator used as LED driver + * + * Copyright (C) 2012 Texas Instruments + * Author: Peter Ujfalusi <peter.ujfalusi@ti.com> + * + * This driver is a complete rewrite of the former pwm-twl6030.c authorded by: + * Hemanth V <hemanthv@ti.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/pwm.h> +#include <linux/i2c/twl.h> +#include <linux/slab.h> + +/* + * This driver handles the PWM driven LED terminals of TWL4030 and TWL6030. + * To generate the signal on TWL4030: + * - LEDA uses PWMA + * - LEDB uses PWMB + * TWL6030 has one LED pin with dedicated LEDPWM + */ + +#define TWL4030_LED_MAX 0x7f +#define TWL6030_LED_MAX 0xff + +/* Registers, bits and macro for TWL4030 */ +#define TWL4030_LEDEN_REG 0x00 +#define TWL4030_PWMA_REG 0x01 + +#define TWL4030_LEDXON (1 << 0) +#define TWL4030_LEDXPWM (1 << 4) +#define TWL4030_LED_PINS (TWL4030_LEDXON | TWL4030_LEDXPWM) +#define TWL4030_LED_TOGGLE(led, x) ((x) << (led)) + +/* Register, bits and macro for TWL6030 */ +#define TWL6030_LED_PWM_CTRL1 0xf4 +#define TWL6030_LED_PWM_CTRL2 0xf5 + +#define TWL6040_LED_MODE_HW 0x00 +#define TWL6040_LED_MODE_ON 0x01 +#define TWL6040_LED_MODE_OFF 0x02 +#define TWL6040_LED_MODE_MASK 0x03 + +struct twl_pwmled_chip { + struct pwm_chip chip; + struct mutex mutex; +}; + +static inline struct twl_pwmled_chip *to_twl(struct pwm_chip *chip) +{ + return container_of(chip, struct twl_pwmled_chip, chip); +} + +static int twl4030_pwmled_config(struct pwm_chip *chip, struct pwm_device *pwm, + int duty_ns, int period_ns) +{ + int duty_cycle = DIV_ROUND_UP(duty_ns * TWL4030_LED_MAX, period_ns) + 1; + u8 pwm_config[2] = { 1, 0 }; + int base, ret; + + /* + * To configure the duty period: + * On-cycle is set to 1 (the minimum allowed value) + * The off time of 0 is not configurable, so the mapping is: + * 0 -> off cycle = 2, + * 1 -> off cycle = 2, + * 2 -> off cycle = 3, + * 126 - > off cycle 127, + * 127 - > off cycle 1 + * When on cycle == off cycle the PWM will be always on + */ + if (duty_cycle == 1) + duty_cycle = 2; + else if (duty_cycle > TWL4030_LED_MAX) + duty_cycle = 1; + + base = pwm->hwpwm * 2 + TWL4030_PWMA_REG; + + pwm_config[1] = duty_cycle; + + ret = twl_i2c_write(TWL4030_MODULE_LED, pwm_config, base, 2); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to configure PWM\n", pwm->label); + + return ret; +} + +static int twl4030_pwmled_enable(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwmled_chip *twl = to_twl(chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL4030_MODULE_LED, &val, TWL4030_LEDEN_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read LEDEN\n", pwm->label); + goto out; + } + + val |= TWL4030_LED_TOGGLE(pwm->hwpwm, TWL4030_LED_PINS); + + ret = twl_i2c_write_u8(TWL4030_MODULE_LED, val, TWL4030_LEDEN_REG); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to enable PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); + return ret; +} + +static void twl4030_pwmled_disable(struct pwm_chip *chip, + struct pwm_device *pwm) +{ + struct twl_pwmled_chip *twl = to_twl(chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL4030_MODULE_LED, &val, TWL4030_LEDEN_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read LEDEN\n", pwm->label); + goto out; + } + + val &= ~TWL4030_LED_TOGGLE(pwm->hwpwm, TWL4030_LED_PINS); + + ret = twl_i2c_write_u8(TWL4030_MODULE_LED, val, TWL4030_LEDEN_REG); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to disable PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); +} + +static int twl6030_pwmled_config(struct pwm_chip *chip, struct pwm_device *pwm, + int duty_ns, int period_ns) +{ + int duty_cycle = (duty_ns * TWL6030_LED_MAX) / period_ns; + u8 on_time; + int ret; + + on_time = duty_cycle & 0xff; + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, on_time, + TWL6030_LED_PWM_CTRL1); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to configure PWM\n", pwm->label); + + return ret; +} + +static int twl6030_pwmled_enable(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwmled_chip *twl = to_twl(chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL6030_MODULE_ID1, &val, TWL6030_LED_PWM_CTRL2); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read PWM_CTRL2\n", + pwm->label); + goto out; + } + + val &= ~TWL6040_LED_MODE_MASK; + val |= TWL6040_LED_MODE_ON; + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_LED_PWM_CTRL2); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to enable PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); + return ret; +} + +static void twl6030_pwmled_disable(struct pwm_chip *chip, + struct pwm_device *pwm) +{ + struct twl_pwmled_chip *twl = to_twl(chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL6030_MODULE_ID1, &val, TWL6030_LED_PWM_CTRL2); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read PWM_CTRL2\n", + pwm->label); + goto out; + } + + val &= ~TWL6040_LED_MODE_MASK; + val |= TWL6040_LED_MODE_OFF; + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_LED_PWM_CTRL2); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to disable PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); +} + +static int twl6030_pwmled_request(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwmled_chip *twl = to_twl(chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL6030_MODULE_ID1, &val, TWL6030_LED_PWM_CTRL2); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read PWM_CTRL2\n", + pwm->label); + goto out; + } + + val &= ~TWL6040_LED_MODE_MASK; + val |= TWL6040_LED_MODE_OFF; + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_LED_PWM_CTRL2); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to request PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); + return ret; +} + +static void twl6030_pwmled_free(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwmled_chip *twl = to_twl(chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL6030_MODULE_ID1, &val, TWL6030_LED_PWM_CTRL2); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read PWM_CTRL2\n", + pwm->label); + goto out; + } + + val &= ~TWL6040_LED_MODE_MASK; + val |= TWL6040_LED_MODE_HW; + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_LED_PWM_CTRL2); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to free PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); +} + +static const struct pwm_ops twl4030_pwmled_ops = { + .enable = twl4030_pwmled_enable, + .disable = twl4030_pwmled_disable, + .config = twl4030_pwmled_config, +}; + +static const struct pwm_ops twl6030_pwmled_ops = { + .enable = twl6030_pwmled_enable, + .disable = twl6030_pwmled_disable, + .config = twl6030_pwmled_config, + .request = twl6030_pwmled_request, + .free = twl6030_pwmled_free, +}; + +static int twl_pwmled_probe(struct platform_device *pdev) +{ + struct twl_pwmled_chip *twl; + int ret; + + twl = devm_kzalloc(&pdev->dev, sizeof(*twl), GFP_KERNEL); + if (!twl) + return -ENOMEM; + + if (twl_class_is_4030()) { + twl->chip.ops = &twl4030_pwmled_ops; + twl->chip.npwm = 2; + } else { + twl->chip.ops = &twl6030_pwmled_ops; + twl->chip.npwm = 1; + } + + twl->chip.dev = &pdev->dev; + twl->chip.base = -1; + + mutex_init(&twl->mutex); + + ret = pwmchip_add(&twl->chip); + if (ret < 0) + return ret; + + platform_set_drvdata(pdev, twl); + + return 0; +} + +static int twl_pwmled_remove(struct platform_device *pdev) +{ + struct twl_pwmled_chip *twl = platform_get_drvdata(pdev); + + return pwmchip_remove(&twl->chip); +} + +#ifdef CONFIG_OF +static struct of_device_id twl_pwmled_of_match[] = { + { .compatible = "ti,twl4030-pwmled" }, + { .compatible = "ti,twl6030-pwmled" }, + { }, +}; +MODULE_DEVICE_TABLE(of, twl_pwmled_of_match); +#endif + +static struct platform_driver twl_pwmled_driver = { + .driver = { + .name = "twl-pwmled", + .of_match_table = of_match_ptr(twl_pwmled_of_match), + }, + .probe = twl_pwmled_probe, + .remove = twl_pwmled_remove, +}; +module_platform_driver(twl_pwmled_driver); + +MODULE_AUTHOR("Peter Ujfalusi <peter.ujfalusi@ti.com>"); +MODULE_DESCRIPTION("PWM driver for TWL4030 and TWL6030 LED outputs"); +MODULE_ALIAS("platform:twl-pwmled"); +MODULE_LICENSE("GPL"); diff --git a/drivers/pwm/pwm-twl.c b/drivers/pwm/pwm-twl.c new file mode 100644 index 000000000000..e65db95d5e59 --- /dev/null +++ b/drivers/pwm/pwm-twl.c @@ -0,0 +1,359 @@ +/* + * Driver for TWL4030/6030 Generic Pulse Width Modulator + * + * Copyright (C) 2012 Texas Instruments + * Author: Peter Ujfalusi <peter.ujfalusi@ti.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/pwm.h> +#include <linux/i2c/twl.h> +#include <linux/slab.h> + +/* + * This driver handles the PWMs of TWL4030 and TWL6030. + * The TRM names for the PWMs on TWL4030 are: PWM0, PWM1 + * TWL6030 also have two PWMs named in the TRM as PWM1, PWM2 + */ + +#define TWL_PWM_MAX 0x7f + +/* Registers, bits and macro for TWL4030 */ +#define TWL4030_GPBR1_REG 0x0c +#define TWL4030_PMBR1_REG 0x0d + +/* GPBR1 register bits */ +#define TWL4030_PWMXCLK_ENABLE (1 << 0) +#define TWL4030_PWMX_ENABLE (1 << 2) +#define TWL4030_PWMX_BITS (TWL4030_PWMX_ENABLE | TWL4030_PWMXCLK_ENABLE) +#define TWL4030_PWM_TOGGLE(pwm, x) ((x) << (pwm)) + +/* PMBR1 register bits */ +#define TWL4030_GPIO6_PWM0_MUTE_MASK (0x03 << 2) +#define TWL4030_GPIO6_PWM0_MUTE_PWM0 (0x01 << 2) +#define TWL4030_GPIO7_VIBRASYNC_PWM1_MASK (0x03 << 4) +#define TWL4030_GPIO7_VIBRASYNC_PWM1_PWM1 (0x03 << 4) + +/* Register, bits and macro for TWL6030 */ +#define TWL6030_TOGGLE3_REG 0x92 + +#define TWL6030_PWMXR (1 << 0) +#define TWL6030_PWMXS (1 << 1) +#define TWL6030_PWMXEN (1 << 2) +#define TWL6030_PWM_TOGGLE(pwm, x) ((x) << (pwm * 3)) + +struct twl_pwm_chip { + struct pwm_chip chip; + struct mutex mutex; + u8 twl6030_toggle3; + u8 twl4030_pwm_mux; +}; + +static inline struct twl_pwm_chip *to_twl(struct pwm_chip *chip) +{ + return container_of(chip, struct twl_pwm_chip, chip); +} + +static int twl_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm, + int duty_ns, int period_ns) +{ + int duty_cycle = DIV_ROUND_UP(duty_ns * TWL_PWM_MAX, period_ns) + 1; + u8 pwm_config[2] = { 1, 0 }; + int base, ret; + + /* + * To configure the duty period: + * On-cycle is set to 1 (the minimum allowed value) + * The off time of 0 is not configurable, so the mapping is: + * 0 -> off cycle = 2, + * 1 -> off cycle = 2, + * 2 -> off cycle = 3, + * 126 - > off cycle 127, + * 127 - > off cycle 1 + * When on cycle == off cycle the PWM will be always on + */ + if (duty_cycle == 1) + duty_cycle = 2; + else if (duty_cycle > TWL_PWM_MAX) + duty_cycle = 1; + + base = pwm->hwpwm * 3; + + pwm_config[1] = duty_cycle; + + ret = twl_i2c_write(TWL_MODULE_PWM, pwm_config, base, 2); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to configure PWM\n", pwm->label); + + return ret; +} + +static int twl4030_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwm_chip *twl = to_twl(chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL4030_MODULE_INTBR, &val, TWL4030_GPBR1_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read GPBR1\n", pwm->label); + goto out; + } + + val |= TWL4030_PWM_TOGGLE(pwm->hwpwm, TWL4030_PWMXCLK_ENABLE); + + ret = twl_i2c_write_u8(TWL4030_MODULE_INTBR, val, TWL4030_GPBR1_REG); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to enable PWM\n", pwm->label); + + val |= TWL4030_PWM_TOGGLE(pwm->hwpwm, TWL4030_PWMX_ENABLE); + + ret = twl_i2c_write_u8(TWL4030_MODULE_INTBR, val, TWL4030_GPBR1_REG); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to enable PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); + return ret; +} + +static void twl4030_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwm_chip *twl = to_twl(chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL4030_MODULE_INTBR, &val, TWL4030_GPBR1_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read GPBR1\n", pwm->label); + goto out; + } + + val &= ~TWL4030_PWM_TOGGLE(pwm->hwpwm, TWL4030_PWMX_ENABLE); + + ret = twl_i2c_write_u8(TWL4030_MODULE_INTBR, val, TWL4030_GPBR1_REG); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to disable PWM\n", pwm->label); + + val &= ~TWL4030_PWM_TOGGLE(pwm->hwpwm, TWL4030_PWMXCLK_ENABLE); + + ret = twl_i2c_write_u8(TWL4030_MODULE_INTBR, val, TWL4030_GPBR1_REG); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to disable PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); +} + +static int twl4030_pwm_request(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwm_chip *twl = to_twl(chip); + int ret; + u8 val, mask, bits; + + if (pwm->hwpwm == 1) { + mask = TWL4030_GPIO7_VIBRASYNC_PWM1_MASK; + bits = TWL4030_GPIO7_VIBRASYNC_PWM1_PWM1; + } else { + mask = TWL4030_GPIO6_PWM0_MUTE_MASK; + bits = TWL4030_GPIO6_PWM0_MUTE_PWM0; + } + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL4030_MODULE_INTBR, &val, TWL4030_PMBR1_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read PMBR1\n", pwm->label); + goto out; + } + + /* Save the current MUX configuration for the PWM */ + twl->twl4030_pwm_mux &= ~mask; + twl->twl4030_pwm_mux |= (val & mask); + + /* Select PWM functionality */ + val &= ~mask; + val |= bits; + + ret = twl_i2c_write_u8(TWL4030_MODULE_INTBR, val, TWL4030_PMBR1_REG); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to request PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); + return ret; +} + +static void twl4030_pwm_free(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwm_chip *twl = container_of(chip, struct twl_pwm_chip, + chip); + int ret; + u8 val, mask; + + if (pwm->hwpwm == 1) + mask = TWL4030_GPIO7_VIBRASYNC_PWM1_MASK; + else + mask = TWL4030_GPIO6_PWM0_MUTE_MASK; + + mutex_lock(&twl->mutex); + ret = twl_i2c_read_u8(TWL4030_MODULE_INTBR, &val, TWL4030_PMBR1_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read PMBR1\n", pwm->label); + goto out; + } + + /* Restore the MUX configuration for the PWM */ + val &= ~mask; + val |= (twl->twl4030_pwm_mux & mask); + + ret = twl_i2c_write_u8(TWL4030_MODULE_INTBR, val, TWL4030_PMBR1_REG); + if (ret < 0) + dev_err(chip->dev, "%s: Failed to free PWM\n", pwm->label); + +out: + mutex_unlock(&twl->mutex); +} + +static int twl6030_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwm_chip *twl = container_of(chip, struct twl_pwm_chip, + chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + val = twl->twl6030_toggle3; + val |= TWL6030_PWM_TOGGLE(pwm->hwpwm, TWL6030_PWMXS | TWL6030_PWMXEN); + val &= ~TWL6030_PWM_TOGGLE(pwm->hwpwm, TWL6030_PWMXR); + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_TOGGLE3_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to enable PWM\n", pwm->label); + goto out; + } + + twl->twl6030_toggle3 = val; +out: + mutex_unlock(&twl->mutex); + return 0; +} + +static void twl6030_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm) +{ + struct twl_pwm_chip *twl = container_of(chip, struct twl_pwm_chip, + chip); + int ret; + u8 val; + + mutex_lock(&twl->mutex); + val = twl->twl6030_toggle3; + val |= TWL6030_PWM_TOGGLE(pwm->hwpwm, TWL6030_PWMXR); + val &= ~TWL6030_PWM_TOGGLE(pwm->hwpwm, TWL6030_PWMXS | TWL6030_PWMXEN); + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_TOGGLE3_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to read TOGGLE3\n", pwm->label); + goto out; + } + + val |= TWL6030_PWM_TOGGLE(pwm->hwpwm, TWL6030_PWMXS | TWL6030_PWMXEN); + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, TWL6030_TOGGLE3_REG); + if (ret < 0) { + dev_err(chip->dev, "%s: Failed to disable PWM\n", pwm->label); + goto out; + } + + twl->twl6030_toggle3 = val; +out: + mutex_unlock(&twl->mutex); +} + +static const struct pwm_ops twl4030_pwm_ops = { + .config = twl_pwm_config, + .enable = twl4030_pwm_enable, + .disable = twl4030_pwm_disable, + .request = twl4030_pwm_request, + .free = twl4030_pwm_free, +}; + +static const struct pwm_ops twl6030_pwm_ops = { + .config = twl_pwm_config, + .enable = twl6030_pwm_enable, + .disable = twl6030_pwm_disable, +}; + +static int twl_pwm_probe(struct platform_device *pdev) +{ + struct twl_pwm_chip *twl; + int ret; + + twl = devm_kzalloc(&pdev->dev, sizeof(*twl), GFP_KERNEL); + if (!twl) + return -ENOMEM; + + if (twl_class_is_4030()) + twl->chip.ops = &twl4030_pwm_ops; + else + twl->chip.ops = &twl6030_pwm_ops; + + twl->chip.dev = &pdev->dev; + twl->chip.base = -1; + twl->chip.npwm = 2; + + mutex_init(&twl->mutex); + + ret = pwmchip_add(&twl->chip); + if (ret < 0) + return ret; + + platform_set_drvdata(pdev, twl); + + return 0; +} + +static int twl_pwm_remove(struct platform_device *pdev) +{ + struct twl_pwm_chip *twl = platform_get_drvdata(pdev); + + return pwmchip_remove(&twl->chip); +} + +#ifdef CONFIG_OF +static struct of_device_id twl_pwm_of_match[] = { + { .compatible = "ti,twl4030-pwm" }, + { .compatible = "ti,twl6030-pwm" }, + { }, +}; +MODULE_DEVICE_TABLE(of, twl_pwm_of_match); +#endif + +static struct platform_driver twl_pwm_driver = { + .driver = { + .name = "twl-pwm", + .of_match_table = of_match_ptr(twl_pwm_of_match), + }, + .probe = twl_pwm_probe, + .remove = twl_pwm_remove, +}; +module_platform_driver(twl_pwm_driver); + +MODULE_AUTHOR("Peter Ujfalusi <peter.ujfalusi@ti.com>"); +MODULE_DESCRIPTION("PWM driver for TWL4030 and TWL6030"); +MODULE_ALIAS("platform:twl-pwm"); +MODULE_LICENSE("GPL"); diff --git a/drivers/pwm/pwm-twl6030.c b/drivers/pwm/pwm-twl6030.c deleted file mode 100644 index 378a7e286366..000000000000 --- a/drivers/pwm/pwm-twl6030.c +++ /dev/null @@ -1,184 +0,0 @@ -/* - * twl6030_pwm.c - * Driver for PHOENIX (TWL6030) Pulse Width Modulator - * - * Copyright (C) 2010 Texas Instruments - * Author: Hemanth V <hemanthv@ti.com> - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License version 2 as published by - * the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program. If not, see <http://www.gnu.org/licenses/>. - */ - -#include <linux/module.h> -#include <linux/platform_device.h> -#include <linux/pwm.h> -#include <linux/i2c/twl.h> -#include <linux/slab.h> - -#define LED_PWM_CTRL1 0xF4 -#define LED_PWM_CTRL2 0xF5 - -/* Max value for CTRL1 register */ -#define PWM_CTRL1_MAX 255 - -/* Pull down disable */ -#define PWM_CTRL2_DIS_PD (1 << 6) - -/* Current control 2.5 milli Amps */ -#define PWM_CTRL2_CURR_02 (2 << 4) - -/* LED supply source */ -#define PWM_CTRL2_SRC_VAC (1 << 2) - -/* LED modes */ -#define PWM_CTRL2_MODE_HW (0 << 0) -#define PWM_CTRL2_MODE_SW (1 << 0) -#define PWM_CTRL2_MODE_DIS (2 << 0) - -#define PWM_CTRL2_MODE_MASK 0x3 - -struct twl6030_pwm_chip { - struct pwm_chip chip; -}; - -static int twl6030_pwm_request(struct pwm_chip *chip, struct pwm_device *pwm) -{ - int ret; - u8 val; - - /* Configure PWM */ - val = PWM_CTRL2_DIS_PD | PWM_CTRL2_CURR_02 | PWM_CTRL2_SRC_VAC | - PWM_CTRL2_MODE_HW; - - ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, LED_PWM_CTRL2); - if (ret < 0) { - dev_err(chip->dev, "%s: Failed to configure PWM, Error %d\n", - pwm->label, ret); - return ret; - } - - return 0; -} - -static int twl6030_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm, - int duty_ns, int period_ns) -{ - u8 duty_cycle = (duty_ns * PWM_CTRL1_MAX) / period_ns; - int ret; - - ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, duty_cycle, LED_PWM_CTRL1); - if (ret < 0) { - pr_err("%s: Failed to configure PWM, Error %d\n", - pwm->label, ret); - return ret; - } - - return 0; -} - -static int twl6030_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm) -{ - int ret; - u8 val; - - ret = twl_i2c_read_u8(TWL6030_MODULE_ID1, &val, LED_PWM_CTRL2); - if (ret < 0) { - dev_err(chip->dev, "%s: Failed to enable PWM, Error %d\n", - pwm->label, ret); - return ret; - } - - /* Change mode to software control */ - val &= ~PWM_CTRL2_MODE_MASK; - val |= PWM_CTRL2_MODE_SW; - - ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, LED_PWM_CTRL2); - if (ret < 0) { - dev_err(chip->dev, "%s: Failed to enable PWM, Error %d\n", - pwm->label, ret); - return ret; - } - - twl_i2c_read_u8(TWL6030_MODULE_ID1, &val, LED_PWM_CTRL2); - return 0; -} - -static void twl6030_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm) -{ - int ret; - u8 val; - - ret = twl_i2c_read_u8(TWL6030_MODULE_ID1, &val, LED_PWM_CTRL2); - if (ret < 0) { - dev_err(chip->dev, "%s: Failed to disable PWM, Error %d\n", - pwm->label, ret); - return; - } - - val &= ~PWM_CTRL2_MODE_MASK; - val |= PWM_CTRL2_MODE_HW; - - ret = twl_i2c_write_u8(TWL6030_MODULE_ID1, val, LED_PWM_CTRL2); - if (ret < 0) { - dev_err(chip->dev, "%s: Failed to disable PWM, Error %d\n", - pwm->label, ret); - } -} - -static const struct pwm_ops twl6030_pwm_ops = { - .request = twl6030_pwm_request, - .config = twl6030_pwm_config, - .enable = twl6030_pwm_enable, - .disable = twl6030_pwm_disable, -}; - -static int twl6030_pwm_probe(struct platform_device *pdev) -{ - struct twl6030_pwm_chip *twl6030; - int ret; - - twl6030 = devm_kzalloc(&pdev->dev, sizeof(*twl6030), GFP_KERNEL); - if (!twl6030) - return -ENOMEM; - - twl6030->chip.dev = &pdev->dev; - twl6030->chip.ops = &twl6030_pwm_ops; - twl6030->chip.base = -1; - twl6030->chip.npwm = 1; - - ret = pwmchip_add(&twl6030->chip); - if (ret < 0) - return ret; - - platform_set_drvdata(pdev, twl6030); - - return 0; -} - -static int twl6030_pwm_remove(struct platform_device *pdev) -{ - struct twl6030_pwm_chip *twl6030 = platform_get_drvdata(pdev); - - return pwmchip_remove(&twl6030->chip); -} - -static struct platform_driver twl6030_pwm_driver = { - .driver = { - .name = "twl6030-pwm", - }, - .probe = twl6030_pwm_probe, - .remove = twl6030_pwm_remove, -}; -module_platform_driver(twl6030_pwm_driver); - -MODULE_ALIAS("platform:twl6030-pwm"); -MODULE_LICENSE("GPL"); diff --git a/drivers/pwm/pwm-vt8500.c b/drivers/pwm/pwm-vt8500.c index ad14389b7144..b0ba2d403439 100644 --- a/drivers/pwm/pwm-vt8500.c +++ b/drivers/pwm/pwm-vt8500.c @@ -1,7 +1,8 @@ /* * drivers/pwm/pwm-vt8500.c * - * Copyright (C) 2010 Alexey Charkov <alchark@gmail.com> + * Copyright (C) 2012 Tony Prisk <linux@prisktech.co.nz> + * Copyright (C) 2010 Alexey Charkov <alchark@gmail.com> * * This software is licensed under the terms of the GNU General Public * License version 2, as published by the Free Software Foundation, and @@ -21,14 +22,24 @@ #include <linux/io.h> #include <linux/pwm.h> #include <linux/delay.h> +#include <linux/clk.h> #include <asm/div64.h> -#define VT8500_NR_PWMS 4 +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/of_address.h> + +/* + * SoC architecture allocates register space for 4 PWMs but only + * 2 are currently implemented. + */ +#define VT8500_NR_PWMS 2 struct vt8500_chip { struct pwm_chip chip; void __iomem *base; + struct clk *clk; }; #define to_vt8500_chip(chip) container_of(chip, struct vt8500_chip, chip) @@ -51,8 +62,15 @@ static int vt8500_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm, struct vt8500_chip *vt8500 = to_vt8500_chip(chip); unsigned long long c; unsigned long period_cycles, prescale, pv, dc; + int err; - c = 25000000/2; /* wild guess --- need to implement clocks */ + err = clk_enable(vt8500->clk); + if (err < 0) { + dev_err(chip->dev, "failed to enable clock\n"); + return err; + } + + c = clk_get_rate(vt8500->clk); c = c * period_ns; do_div(c, 1000000000); period_cycles = c; @@ -64,8 +82,10 @@ static int vt8500_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm, if (pv > 4095) pv = 4095; - if (prescale > 1023) + if (prescale > 1023) { + clk_disable(vt8500->clk); return -EINVAL; + } c = (unsigned long long)pv * duty_ns; do_div(c, period_ns); @@ -80,13 +100,21 @@ static int vt8500_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm, pwm_busy_wait(vt8500->base + 0x40 + pwm->hwpwm, (1 << 3)); writel(dc, vt8500->base + 0xc + (pwm->hwpwm << 4)); + clk_disable(vt8500->clk); return 0; } static int vt8500_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm) { + int err; struct vt8500_chip *vt8500 = to_vt8500_chip(chip); + err = clk_enable(vt8500->clk); + if (err < 0) { + dev_err(chip->dev, "failed to enable clock\n"); + return err; + } + pwm_busy_wait(vt8500->base + 0x40 + pwm->hwpwm, (1 << 0)); writel(5, vt8500->base + (pwm->hwpwm << 4)); return 0; @@ -98,6 +126,8 @@ static void vt8500_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm) pwm_busy_wait(vt8500->base + 0x40 + pwm->hwpwm, (1 << 0)); writel(0, vt8500->base + (pwm->hwpwm << 4)); + + clk_disable(vt8500->clk); } static struct pwm_ops vt8500_pwm_ops = { @@ -107,12 +137,24 @@ static struct pwm_ops vt8500_pwm_ops = { .owner = THIS_MODULE, }; -static int __devinit pwm_probe(struct platform_device *pdev) +static const struct of_device_id vt8500_pwm_dt_ids[] = { + { .compatible = "via,vt8500-pwm", }, + { /* Sentinel */ } +}; +MODULE_DEVICE_TABLE(of, vt8500_pwm_dt_ids); + +static int vt8500_pwm_probe(struct platform_device *pdev) { struct vt8500_chip *chip; struct resource *r; + struct device_node *np = pdev->dev.of_node; int ret; + if (!np) { + dev_err(&pdev->dev, "invalid devicetree node\n"); + return -EINVAL; + } + chip = devm_kzalloc(&pdev->dev, sizeof(*chip), GFP_KERNEL); if (chip == NULL) { dev_err(&pdev->dev, "failed to allocate memory\n"); @@ -124,6 +166,12 @@ static int __devinit pwm_probe(struct platform_device *pdev) chip->chip.base = -1; chip->chip.npwm = VT8500_NR_PWMS; + chip->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(chip->clk)) { + dev_err(&pdev->dev, "clock source not specified\n"); + return PTR_ERR(chip->clk); + } + r = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (r == NULL) { dev_err(&pdev->dev, "no memory resource defined\n"); @@ -131,18 +179,26 @@ static int __devinit pwm_probe(struct platform_device *pdev) } chip->base = devm_request_and_ioremap(&pdev->dev, r); - if (chip->base == NULL) + if (!chip->base) return -EADDRNOTAVAIL; + ret = clk_prepare(chip->clk); + if (ret < 0) { + dev_err(&pdev->dev, "failed to prepare clock\n"); + return ret; + } + ret = pwmchip_add(&chip->chip); - if (ret < 0) + if (ret < 0) { + dev_err(&pdev->dev, "failed to add PWM chip\n"); return ret; + } platform_set_drvdata(pdev, chip); return ret; } -static int __devexit pwm_remove(struct platform_device *pdev) +static int vt8500_pwm_remove(struct platform_device *pdev) { struct vt8500_chip *chip; @@ -150,28 +206,22 @@ static int __devexit pwm_remove(struct platform_device *pdev) if (chip == NULL) return -ENODEV; + clk_unprepare(chip->clk); + return pwmchip_remove(&chip->chip); } -static struct platform_driver pwm_driver = { +static struct platform_driver vt8500_pwm_driver = { + .probe = vt8500_pwm_probe, + .remove = vt8500_pwm_remove, .driver = { .name = "vt8500-pwm", .owner = THIS_MODULE, + .of_match_table = vt8500_pwm_dt_ids, }, - .probe = pwm_probe, - .remove = __devexit_p(pwm_remove), }; +module_platform_driver(vt8500_pwm_driver); -static int __init pwm_init(void) -{ - return platform_driver_register(&pwm_driver); -} -arch_initcall(pwm_init); - -static void __exit pwm_exit(void) -{ - platform_driver_unregister(&pwm_driver); -} -module_exit(pwm_exit); - -MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("VT8500 PWM Driver"); +MODULE_AUTHOR("Tony Prisk <linux@prisktech.co.nz>"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index 19c03ab2bdcb..d0cea02b5dfc 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -269,6 +269,15 @@ config RTC_DRV_X1205 This driver can also be built as a module. If so, the module will be called rtc-x1205. +config RTC_DRV_PCF8523 + tristate "NXP PCF8523" + help + If you say yes here you get support for the NXP PCF8523 RTC + chips. + + This driver can also be built as a module. If so, the module + will be called rtc-pcf8523. + config RTC_DRV_PCF8563 tristate "Philips PCF8563/Epson RTC8564" help @@ -600,6 +609,16 @@ config RTC_DRV_DA9052 Say y here to support the RTC driver for Dialog Semiconductor DA9052-BC and DA9053-AA/Bx PMICs. +config RTC_DRV_DA9055 + tristate "Dialog Semiconductor DA9055 RTC" + depends on MFD_DA9055 + help + If you say yes here you will get support for the + RTC of the Dialog DA9055 PMIC. + + This driver can also be built as a module. If so, the module + will be called rtc-da9055 + config RTC_DRV_EFI tristate "EFI RTC" depends on IA64 @@ -768,7 +787,7 @@ config RTC_DRV_DAVINCI config RTC_DRV_IMXDI tristate "Freescale IMX DryIce Real Time Clock" - depends on SOC_IMX25 + depends on ARCH_MXC help Support for Freescale IMX DryIce RTC @@ -777,11 +796,13 @@ config RTC_DRV_IMXDI config RTC_DRV_OMAP tristate "TI OMAP1" - depends on ARCH_OMAP15XX || ARCH_OMAP16XX || ARCH_OMAP730 || ARCH_DAVINCI_DA8XX + depends on ARCH_OMAP15XX || ARCH_OMAP16XX || ARCH_OMAP730 || ARCH_DAVINCI_DA8XX || SOC_AM33XX help - Say "yes" here to support the real time clock on TI OMAP1 and - DA8xx/OMAP-L13x chips. This driver can also be built as a - module called rtc-omap. + Say "yes" here to support the on chip real time clock + present on TI OMAP1, AM33xx and DA8xx/OMAP-L13x. + + This driver can also be built as a module, if so, module + will be called rtc-omap. config HAVE_S3C_RTC bool diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile index 56297f0fd388..c3f62c80dc06 100644 --- a/drivers/rtc/Makefile +++ b/drivers/rtc/Makefile @@ -29,6 +29,7 @@ obj-$(CONFIG_RTC_DRV_BQ4802) += rtc-bq4802.o obj-$(CONFIG_RTC_DRV_CMOS) += rtc-cmos.o obj-$(CONFIG_RTC_DRV_COH901331) += rtc-coh901331.o obj-$(CONFIG_RTC_DRV_DA9052) += rtc-da9052.o +obj-$(CONFIG_RTC_DRV_DA9055) += rtc-da9055.o obj-$(CONFIG_RTC_DRV_DAVINCI) += rtc-davinci.o obj-$(CONFIG_RTC_DRV_DM355EVM) += rtc-dm355evm.o obj-$(CONFIG_RTC_DRV_VRTC) += rtc-mrst.o @@ -76,6 +77,7 @@ obj-$(CONFIG_RTC_DRV_MV) += rtc-mv.o obj-$(CONFIG_RTC_DRV_NUC900) += rtc-nuc900.o obj-$(CONFIG_RTC_DRV_OMAP) += rtc-omap.o obj-$(CONFIG_RTC_DRV_PCAP) += rtc-pcap.o +obj-$(CONFIG_RTC_DRV_PCF8523) += rtc-pcf8523.o obj-$(CONFIG_RTC_DRV_PCF8563) += rtc-pcf8563.o obj-$(CONFIG_RTC_DRV_PCF8583) += rtc-pcf8583.o obj-$(CONFIG_RTC_DRV_PCF2123) += rtc-pcf2123.o diff --git a/drivers/rtc/rtc-da9055.c b/drivers/rtc/rtc-da9055.c new file mode 100644 index 000000000000..96bafc5c3bf8 --- /dev/null +++ b/drivers/rtc/rtc-da9055.c @@ -0,0 +1,413 @@ +/* + * Real time clock driver for DA9055 + * + * Copyright(c) 2012 Dialog Semiconductor Ltd. + * + * Author: Dajun Dajun Chen <dajun.chen@diasemi.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + */ + +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/rtc.h> + +#include <linux/mfd/da9055/core.h> +#include <linux/mfd/da9055/reg.h> +#include <linux/mfd/da9055/pdata.h> + +struct da9055_rtc { + struct rtc_device *rtc; + struct da9055 *da9055; + int alarm_enable; +}; + +static int da9055_rtc_enable_alarm(struct da9055_rtc *rtc, bool enable) +{ + int ret; + if (enable) { + ret = da9055_reg_update(rtc->da9055, DA9055_REG_ALARM_Y, + DA9055_RTC_ALM_EN, + DA9055_RTC_ALM_EN); + if (ret != 0) + dev_err(rtc->da9055->dev, "Failed to enable ALM: %d\n", + ret); + rtc->alarm_enable = 1; + } else { + ret = da9055_reg_update(rtc->da9055, DA9055_REG_ALARM_Y, + DA9055_RTC_ALM_EN, 0); + if (ret != 0) + dev_err(rtc->da9055->dev, + "Failed to disable ALM: %d\n", ret); + rtc->alarm_enable = 0; + } + return ret; +} + +static irqreturn_t da9055_rtc_alm_irq(int irq, void *data) +{ + struct da9055_rtc *rtc = data; + + da9055_rtc_enable_alarm(rtc, 0); + rtc_update_irq(rtc->rtc, 1, RTC_IRQF | RTC_AF); + + return IRQ_HANDLED; +} + +static int da9055_read_alarm(struct da9055 *da9055, struct rtc_time *rtc_tm) +{ + int ret; + uint8_t v[5]; + + ret = da9055_group_read(da9055, DA9055_REG_ALARM_MI, 5, v); + if (ret != 0) { + dev_err(da9055->dev, "Failed to group read ALM: %d\n", ret); + return ret; + } + + rtc_tm->tm_year = (v[4] & DA9055_RTC_ALM_YEAR) + 100; + rtc_tm->tm_mon = (v[3] & DA9055_RTC_ALM_MONTH) - 1; + rtc_tm->tm_mday = v[2] & DA9055_RTC_ALM_DAY; + rtc_tm->tm_hour = v[1] & DA9055_RTC_ALM_HOUR; + rtc_tm->tm_min = v[0] & DA9055_RTC_ALM_MIN; + + return rtc_valid_tm(rtc_tm); +} + +static int da9055_set_alarm(struct da9055 *da9055, struct rtc_time *rtc_tm) +{ + int ret; + uint8_t v[2]; + + rtc_tm->tm_year -= 100; + rtc_tm->tm_mon += 1; + + ret = da9055_reg_update(da9055, DA9055_REG_ALARM_MI, + DA9055_RTC_ALM_MIN, rtc_tm->tm_min); + if (ret != 0) { + dev_err(da9055->dev, "Failed to write ALRM MIN: %d\n", ret); + return ret; + } + + v[0] = rtc_tm->tm_hour; + v[1] = rtc_tm->tm_mday; + + ret = da9055_group_write(da9055, DA9055_REG_ALARM_H, 2, v); + if (ret < 0) + return ret; + + ret = da9055_reg_update(da9055, DA9055_REG_ALARM_MO, + DA9055_RTC_ALM_MONTH, rtc_tm->tm_mon); + if (ret < 0) + dev_err(da9055->dev, "Failed to write ALM Month:%d\n", ret); + + ret = da9055_reg_update(da9055, DA9055_REG_ALARM_Y, + DA9055_RTC_ALM_YEAR, rtc_tm->tm_year); + if (ret < 0) + dev_err(da9055->dev, "Failed to write ALM Year:%d\n", ret); + + return ret; +} + +static int da9055_rtc_get_alarm_status(struct da9055 *da9055) +{ + int ret; + + ret = da9055_reg_read(da9055, DA9055_REG_ALARM_Y); + if (ret < 0) { + dev_err(da9055->dev, "Failed to read ALM: %d\n", ret); + return ret; + } + ret &= DA9055_RTC_ALM_EN; + return (ret > 0) ? 1 : 0; +} + +static int da9055_rtc_read_time(struct device *dev, struct rtc_time *rtc_tm) +{ + struct da9055_rtc *rtc = dev_get_drvdata(dev); + uint8_t v[6]; + int ret; + + ret = da9055_reg_read(rtc->da9055, DA9055_REG_COUNT_S); + if (ret < 0) + return ret; + + /* + * Registers are only valid when RTC_READ + * status bit is asserted + */ + if (!(ret & DA9055_RTC_READ)) + return -EBUSY; + + ret = da9055_group_read(rtc->da9055, DA9055_REG_COUNT_S, 6, v); + if (ret < 0) { + dev_err(rtc->da9055->dev, "Failed to read RTC time : %d\n", + ret); + return ret; + } + + rtc_tm->tm_year = (v[5] & DA9055_RTC_YEAR) + 100; + rtc_tm->tm_mon = (v[4] & DA9055_RTC_MONTH) - 1; + rtc_tm->tm_mday = v[3] & DA9055_RTC_DAY; + rtc_tm->tm_hour = v[2] & DA9055_RTC_HOUR; + rtc_tm->tm_min = v[1] & DA9055_RTC_MIN; + rtc_tm->tm_sec = v[0] & DA9055_RTC_SEC; + + return rtc_valid_tm(rtc_tm); +} + +static int da9055_rtc_set_time(struct device *dev, struct rtc_time *tm) +{ + struct da9055_rtc *rtc; + uint8_t v[6]; + + rtc = dev_get_drvdata(dev); + + v[0] = tm->tm_sec; + v[1] = tm->tm_min; + v[2] = tm->tm_hour; + v[3] = tm->tm_mday; + v[4] = tm->tm_mon + 1; + v[5] = tm->tm_year - 100; + + return da9055_group_write(rtc->da9055, DA9055_REG_COUNT_S, 6, v); +} + +static int da9055_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) +{ + int ret; + struct rtc_time *tm = &alrm->time; + struct da9055_rtc *rtc = dev_get_drvdata(dev); + + ret = da9055_read_alarm(rtc->da9055, tm); + + if (ret) + return ret; + + alrm->enabled = da9055_rtc_get_alarm_status(rtc->da9055); + + return 0; +} + +static int da9055_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) +{ + int ret; + struct rtc_time *tm = &alrm->time; + struct da9055_rtc *rtc = dev_get_drvdata(dev); + + ret = da9055_rtc_enable_alarm(rtc, 0); + if (ret < 0) + return ret; + + ret = da9055_set_alarm(rtc->da9055, tm); + if (ret) + return ret; + + ret = da9055_rtc_enable_alarm(rtc, 1); + + return ret; +} + +static int da9055_rtc_alarm_irq_enable(struct device *dev, unsigned int enabled) +{ + struct da9055_rtc *rtc = dev_get_drvdata(dev); + + return da9055_rtc_enable_alarm(rtc, enabled); +} + +static const struct rtc_class_ops da9055_rtc_ops = { + .read_time = da9055_rtc_read_time, + .set_time = da9055_rtc_set_time, + .read_alarm = da9055_rtc_read_alarm, + .set_alarm = da9055_rtc_set_alarm, + .alarm_irq_enable = da9055_rtc_alarm_irq_enable, +}; + +static int __init da9055_rtc_device_init(struct da9055 *da9055, + struct da9055_pdata *pdata) +{ + int ret; + + /* Enable RTC and the internal Crystal */ + ret = da9055_reg_update(da9055, DA9055_REG_CONTROL_B, + DA9055_RTC_EN, DA9055_RTC_EN); + if (ret < 0) + return ret; + ret = da9055_reg_update(da9055, DA9055_REG_EN_32K, + DA9055_CRYSTAL_EN, DA9055_CRYSTAL_EN); + if (ret < 0) + return ret; + + /* Enable RTC in Power Down mode */ + ret = da9055_reg_update(da9055, DA9055_REG_CONTROL_B, + DA9055_RTC_MODE_PD, DA9055_RTC_MODE_PD); + if (ret < 0) + return ret; + + /* Enable RTC in Reset mode */ + if (pdata && pdata->reset_enable) { + ret = da9055_reg_update(da9055, DA9055_REG_CONTROL_B, + DA9055_RTC_MODE_SD, + DA9055_RTC_MODE_SD << + DA9055_RTC_MODE_SD_SHIFT); + if (ret < 0) + return ret; + } + + /* Disable the RTC TICK ALM */ + ret = da9055_reg_update(da9055, DA9055_REG_ALARM_MO, + DA9055_RTC_TICK_WAKE_MASK, 0); + if (ret < 0) + return ret; + + return 0; +} + +static int da9055_rtc_probe(struct platform_device *pdev) +{ + struct da9055_rtc *rtc; + struct da9055_pdata *pdata = NULL; + int ret, alm_irq; + + rtc = devm_kzalloc(&pdev->dev, sizeof(struct da9055_rtc), GFP_KERNEL); + if (!rtc) + return -ENOMEM; + + rtc->da9055 = dev_get_drvdata(pdev->dev.parent); + pdata = rtc->da9055->dev->platform_data; + platform_set_drvdata(pdev, rtc); + + ret = da9055_rtc_device_init(rtc->da9055, pdata); + if (ret < 0) + goto err_rtc; + + ret = da9055_reg_read(rtc->da9055, DA9055_REG_ALARM_Y); + if (ret < 0) + goto err_rtc; + + if (ret & DA9055_RTC_ALM_EN) + rtc->alarm_enable = 1; + + device_init_wakeup(&pdev->dev, 1); + + rtc->rtc = rtc_device_register(pdev->name, &pdev->dev, + &da9055_rtc_ops, THIS_MODULE); + if (IS_ERR(rtc->rtc)) { + ret = PTR_ERR(rtc->rtc); + goto err_rtc; + } + + alm_irq = platform_get_irq_byname(pdev, "ALM"); + alm_irq = regmap_irq_get_virq(rtc->da9055->irq_data, alm_irq); + ret = devm_request_threaded_irq(&pdev->dev, alm_irq, NULL, + da9055_rtc_alm_irq, + IRQF_TRIGGER_HIGH | IRQF_ONESHOT, + "ALM", rtc); + if (ret != 0) + dev_err(rtc->da9055->dev, "irq registration failed: %d\n", ret); + +err_rtc: + return ret; + +} + +static int da9055_rtc_remove(struct platform_device *pdev) +{ + struct da9055_rtc *rtc = pdev->dev.platform_data; + + rtc_device_unregister(rtc->rtc); + platform_set_drvdata(pdev, NULL); + + return 0; +} + +#ifdef CONFIG_PM +/* Turn off the alarm if it should not be a wake source. */ +static int da9055_rtc_suspend(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct da9055_rtc *rtc = dev_get_drvdata(&pdev->dev); + int ret; + + if (!device_may_wakeup(&pdev->dev)) { + /* Disable the ALM IRQ */ + ret = da9055_rtc_enable_alarm(rtc, 0); + if (ret < 0) + dev_err(&pdev->dev, "Failed to disable RTC ALM\n"); + } + + return 0; +} + +/* Enable the alarm if it should be enabled (in case it was disabled to + * prevent use as a wake source). + */ +static int da9055_rtc_resume(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct da9055_rtc *rtc = dev_get_drvdata(&pdev->dev); + int ret; + + if (!device_may_wakeup(&pdev->dev)) { + if (rtc->alarm_enable) { + ret = da9055_rtc_enable_alarm(rtc, 1); + if (ret < 0) + dev_err(&pdev->dev, + "Failed to restart RTC ALM\n"); + } + } + + return 0; +} + +/* Unconditionally disable the alarm */ +static int da9055_rtc_freeze(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct da9055_rtc *rtc = dev_get_drvdata(&pdev->dev); + int ret; + + ret = da9055_rtc_enable_alarm(rtc, 0); + if (ret < 0) + dev_err(&pdev->dev, "Failed to freeze RTC ALMs\n"); + + return 0; + +} +#else +#define da9055_rtc_suspend NULL +#define da9055_rtc_resume NULL +#define da9055_rtc_freeze NULL +#endif + +static const struct dev_pm_ops da9055_rtc_pm_ops = { + .suspend = da9055_rtc_suspend, + .resume = da9055_rtc_resume, + + .freeze = da9055_rtc_freeze, + .thaw = da9055_rtc_resume, + .restore = da9055_rtc_resume, + + .poweroff = da9055_rtc_suspend, +}; + +static struct platform_driver da9055_rtc_driver = { + .probe = da9055_rtc_probe, + .remove = da9055_rtc_remove, + .driver = { + .name = "da9055-rtc", + .owner = THIS_MODULE, + .pm = &da9055_rtc_pm_ops, + }, +}; + +module_platform_driver(da9055_rtc_driver); + +MODULE_AUTHOR("David Dajun Chen <dchen@diasemi.com>"); +MODULE_DESCRIPTION("RTC driver for Dialog DA9055 PMIC"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("platform:da9055-rtc"); diff --git a/drivers/rtc/rtc-davinci.c b/drivers/rtc/rtc-davinci.c index 14c2109dbaa3..07cd03eae606 100644 --- a/drivers/rtc/rtc-davinci.c +++ b/drivers/rtc/rtc-davinci.c @@ -485,7 +485,7 @@ static int __init davinci_rtc_probe(struct platform_device *pdev) struct resource *res, *mem; int ret = 0; - davinci_rtc = kzalloc(sizeof(struct davinci_rtc), GFP_KERNEL); + davinci_rtc = devm_kzalloc(&pdev->dev, sizeof(struct davinci_rtc), GFP_KERNEL); if (!davinci_rtc) { dev_dbg(dev, "could not allocate memory for private data\n"); return -ENOMEM; @@ -494,15 +494,13 @@ static int __init davinci_rtc_probe(struct platform_device *pdev) davinci_rtc->irq = platform_get_irq(pdev, 0); if (davinci_rtc->irq < 0) { dev_err(dev, "no RTC irq\n"); - ret = davinci_rtc->irq; - goto fail1; + return davinci_rtc->irq; } res = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!res) { dev_err(dev, "no mem resource\n"); - ret = -EINVAL; - goto fail1; + return -EINVAL; } davinci_rtc->pbase = res->start; @@ -513,8 +511,7 @@ static int __init davinci_rtc_probe(struct platform_device *pdev) if (!mem) { dev_err(dev, "RTC registers at %08x are not free\n", davinci_rtc->pbase); - ret = -EBUSY; - goto fail1; + return -EBUSY; } davinci_rtc->base = ioremap(davinci_rtc->pbase, davinci_rtc->base_size); @@ -529,8 +526,9 @@ static int __init davinci_rtc_probe(struct platform_device *pdev) davinci_rtc->rtc = rtc_device_register(pdev->name, &pdev->dev, &davinci_rtc_ops, THIS_MODULE); if (IS_ERR(davinci_rtc->rtc)) { - dev_err(dev, "unable to register RTC device, err %ld\n", - PTR_ERR(davinci_rtc->rtc)); + ret = PTR_ERR(davinci_rtc->rtc); + dev_err(dev, "unable to register RTC device, err %d\n", + ret); goto fail3; } @@ -566,9 +564,6 @@ fail3: iounmap(davinci_rtc->base); fail2: release_mem_region(davinci_rtc->pbase, davinci_rtc->base_size); -fail1: - kfree(davinci_rtc); - return ret; } @@ -589,8 +584,6 @@ static int __devexit davinci_rtc_remove(struct platform_device *pdev) platform_set_drvdata(pdev, NULL); - kfree(davinci_rtc); - return 0; } diff --git a/drivers/rtc/rtc-dev.c b/drivers/rtc/rtc-dev.c index cace6d3aed9a..9a86b4bd8699 100644 --- a/drivers/rtc/rtc-dev.c +++ b/drivers/rtc/rtc-dev.c @@ -379,25 +379,6 @@ static long rtc_dev_ioctl(struct file *file, err = put_user(rtc->irq_freq, (unsigned long __user *)uarg); break; -#if 0 - case RTC_EPOCH_SET: -#ifndef rtc_epoch - /* - * There were no RTC clocks before 1900. - */ - if (arg < 1900) { - err = -EINVAL; - break; - } - rtc_epoch = arg; - err = 0; -#endif - break; - - case RTC_EPOCH_READ: - err = put_user(rtc_epoch, (unsigned long __user *)uarg); - break; -#endif case RTC_WKALM_SET: mutex_unlock(&rtc->ops_lock); if (copy_from_user(&alarm, uarg, sizeof(alarm))) diff --git a/drivers/rtc/rtc-imxdi.c b/drivers/rtc/rtc-imxdi.c index 4eed51044c5d..18a4f0dd78a3 100644 --- a/drivers/rtc/rtc-imxdi.c +++ b/drivers/rtc/rtc-imxdi.c @@ -37,6 +37,7 @@ #include <linux/rtc.h> #include <linux/sched.h> #include <linux/workqueue.h> +#include <linux/of.h> /* DryIce Register Definitions */ @@ -495,10 +496,20 @@ static int __devexit dryice_rtc_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_OF +static const struct of_device_id dryice_dt_ids[] = { + { .compatible = "fsl,imx25-rtc" }, + { /* sentinel */ } +}; + +MODULE_DEVICE_TABLE(of, dryice_dt_ids); +#endif + static struct platform_driver dryice_rtc_driver = { .driver = { .name = "imxdi_rtc", .owner = THIS_MODULE, + .of_match_table = of_match_ptr(dryice_dt_ids), }, .remove = __devexit_p(dryice_rtc_remove), }; diff --git a/drivers/rtc/rtc-omap.c b/drivers/rtc/rtc-omap.c index 0b614e32653d..600971407aac 100644 --- a/drivers/rtc/rtc-omap.c +++ b/drivers/rtc/rtc-omap.c @@ -20,6 +20,9 @@ #include <linux/rtc.h> #include <linux/bcd.h> #include <linux/platform_device.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/pm_runtime.h> #include <asm/io.h> @@ -38,6 +41,8 @@ * the SoC). See the BOARD-SPECIFIC CUSTOMIZATION comment. */ +#define DRIVER_NAME "omap_rtc" + #define OMAP_RTC_BASE 0xfffb4800 /* RTC registers */ @@ -64,6 +69,9 @@ #define OMAP_RTC_COMP_MSB_REG 0x50 #define OMAP_RTC_OSC_REG 0x54 +#define OMAP_RTC_KICK0_REG 0x6c +#define OMAP_RTC_KICK1_REG 0x70 + /* OMAP_RTC_CTRL_REG bit fields: */ #define OMAP_RTC_CTRL_SPLIT (1<<7) #define OMAP_RTC_CTRL_DISABLE (1<<6) @@ -88,10 +96,18 @@ #define OMAP_RTC_INTERRUPTS_IT_ALARM (1<<3) #define OMAP_RTC_INTERRUPTS_IT_TIMER (1<<2) +/* OMAP_RTC_KICKER values */ +#define KICK0_VALUE 0x83e70b13 +#define KICK1_VALUE 0x95a4f1e0 + +#define OMAP_RTC_HAS_KICKER 0x1 + static void __iomem *rtc_base; -#define rtc_read(addr) __raw_readb(rtc_base + (addr)) -#define rtc_write(val, addr) __raw_writeb(val, rtc_base + (addr)) +#define rtc_read(addr) readb(rtc_base + (addr)) +#define rtc_write(val, addr) writeb(val, rtc_base + (addr)) + +#define rtc_writel(val, addr) writel(val, rtc_base + (addr)) /* we rely on the rtc framework to handle locking (rtc->ops_lock), @@ -285,11 +301,38 @@ static struct rtc_class_ops omap_rtc_ops = { static int omap_rtc_alarm; static int omap_rtc_timer; +#define OMAP_RTC_DATA_DA830_IDX 1 + +static struct platform_device_id omap_rtc_devtype[] = { + { + .name = DRIVER_NAME, + }, { + .name = "da830-rtc", + .driver_data = OMAP_RTC_HAS_KICKER, + }, + {}, +}; +MODULE_DEVICE_TABLE(platform, omap_rtc_devtype); + +static const struct of_device_id omap_rtc_of_match[] = { + { .compatible = "ti,da830-rtc", + .data = &omap_rtc_devtype[OMAP_RTC_DATA_DA830_IDX], + }, + {}, +}; +MODULE_DEVICE_TABLE(of, omap_rtc_of_match); + static int __init omap_rtc_probe(struct platform_device *pdev) { struct resource *res, *mem; struct rtc_device *rtc; u8 reg, new_ctrl; + const struct platform_device_id *id_entry; + const struct of_device_id *of_id; + + of_id = of_match_device(omap_rtc_of_match, &pdev->dev); + if (of_id) + pdev->id_entry = of_id->data; omap_rtc_timer = platform_get_irq(pdev, 0); if (omap_rtc_timer <= 0) { @@ -322,6 +365,16 @@ static int __init omap_rtc_probe(struct platform_device *pdev) goto fail; } + /* Enable the clock/module so that we can access the registers */ + pm_runtime_enable(&pdev->dev); + pm_runtime_get_sync(&pdev->dev); + + id_entry = platform_get_device_id(pdev); + if (id_entry && (id_entry->driver_data & OMAP_RTC_HAS_KICKER)) { + rtc_writel(KICK0_VALUE, OMAP_RTC_KICK0_REG); + rtc_writel(KICK1_VALUE, OMAP_RTC_KICK1_REG); + } + rtc = rtc_device_register(pdev->name, &pdev->dev, &omap_rtc_ops, THIS_MODULE); if (IS_ERR(rtc)) { @@ -398,6 +451,10 @@ fail2: fail1: rtc_device_unregister(rtc); fail0: + if (id_entry && (id_entry->driver_data & OMAP_RTC_HAS_KICKER)) + rtc_writel(0, OMAP_RTC_KICK0_REG); + pm_runtime_put_sync(&pdev->dev); + pm_runtime_disable(&pdev->dev); iounmap(rtc_base); fail: release_mem_region(mem->start, resource_size(mem)); @@ -408,6 +465,8 @@ static int __exit omap_rtc_remove(struct platform_device *pdev) { struct rtc_device *rtc = platform_get_drvdata(pdev); struct resource *mem = dev_get_drvdata(&rtc->dev); + const struct platform_device_id *id_entry = + platform_get_device_id(pdev); device_init_wakeup(&pdev->dev, 0); @@ -420,6 +479,13 @@ static int __exit omap_rtc_remove(struct platform_device *pdev) free_irq(omap_rtc_alarm, rtc); rtc_device_unregister(rtc); + if (id_entry && (id_entry->driver_data & OMAP_RTC_HAS_KICKER)) + rtc_writel(0, OMAP_RTC_KICK0_REG); + + /* Disable the clock/module */ + pm_runtime_put_sync(&pdev->dev); + pm_runtime_disable(&pdev->dev); + iounmap(rtc_base); release_mem_region(mem->start, resource_size(mem)); return 0; @@ -442,11 +508,17 @@ static int omap_rtc_suspend(struct platform_device *pdev, pm_message_t state) else rtc_write(0, OMAP_RTC_INTERRUPTS_REG); + /* Disable the clock/module */ + pm_runtime_put_sync(&pdev->dev); + return 0; } static int omap_rtc_resume(struct platform_device *pdev) { + /* Enable the clock/module so that we can access the registers */ + pm_runtime_get_sync(&pdev->dev); + if (device_may_wakeup(&pdev->dev)) disable_irq_wake(omap_rtc_alarm); else @@ -471,9 +543,11 @@ static struct platform_driver omap_rtc_driver = { .resume = omap_rtc_resume, .shutdown = omap_rtc_shutdown, .driver = { - .name = "omap_rtc", + .name = DRIVER_NAME, .owner = THIS_MODULE, + .of_match_table = of_match_ptr(omap_rtc_of_match), }, + .id_table = omap_rtc_devtype, }; static int __init rtc_init(void) diff --git a/drivers/rtc/rtc-pcf8523.c b/drivers/rtc/rtc-pcf8523.c new file mode 100644 index 000000000000..be05a645f99e --- /dev/null +++ b/drivers/rtc/rtc-pcf8523.c @@ -0,0 +1,326 @@ +/* + * Copyright (C) 2012 Avionic Design GmbH + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/bcd.h> +#include <linux/i2c.h> +#include <linux/module.h> +#include <linux/rtc.h> +#include <linux/of.h> + +#define DRIVER_NAME "rtc-pcf8523" + +#define REG_CONTROL1 0x00 +#define REG_CONTROL1_CAP_SEL (1 << 7) +#define REG_CONTROL1_STOP (1 << 5) + +#define REG_CONTROL3 0x02 +#define REG_CONTROL3_PM_BLD (1 << 7) /* battery low detection disabled */ +#define REG_CONTROL3_PM_VDD (1 << 6) /* switch-over disabled */ +#define REG_CONTROL3_PM_DSM (1 << 5) /* direct switching mode */ +#define REG_CONTROL3_PM_MASK 0xe0 + +#define REG_SECONDS 0x03 +#define REG_SECONDS_OS (1 << 7) + +#define REG_MINUTES 0x04 +#define REG_HOURS 0x05 +#define REG_DAYS 0x06 +#define REG_WEEKDAYS 0x07 +#define REG_MONTHS 0x08 +#define REG_YEARS 0x09 + +struct pcf8523 { + struct rtc_device *rtc; +}; + +static int pcf8523_read(struct i2c_client *client, u8 reg, u8 *valuep) +{ + struct i2c_msg msgs[2]; + u8 value = 0; + int err; + + msgs[0].addr = client->addr; + msgs[0].flags = 0; + msgs[0].len = sizeof(reg); + msgs[0].buf = ® + + msgs[1].addr = client->addr; + msgs[1].flags = I2C_M_RD; + msgs[1].len = sizeof(value); + msgs[1].buf = &value; + + err = i2c_transfer(client->adapter, msgs, ARRAY_SIZE(msgs)); + if (err < 0) + return err; + + *valuep = value; + + return 0; +} + +static int pcf8523_write(struct i2c_client *client, u8 reg, u8 value) +{ + u8 buffer[2] = { reg, value }; + struct i2c_msg msg; + int err; + + msg.addr = client->addr; + msg.flags = 0; + msg.len = sizeof(buffer); + msg.buf = buffer; + + err = i2c_transfer(client->adapter, &msg, 1); + if (err < 0) + return err; + + return 0; +} + +static int pcf8523_select_capacitance(struct i2c_client *client, bool high) +{ + u8 value; + int err; + + err = pcf8523_read(client, REG_CONTROL1, &value); + if (err < 0) + return err; + + if (!high) + value &= ~REG_CONTROL1_CAP_SEL; + else + value |= REG_CONTROL1_CAP_SEL; + + err = pcf8523_write(client, REG_CONTROL1, value); + if (err < 0) + return err; + + return err; +} + +static int pcf8523_set_pm(struct i2c_client *client, u8 pm) +{ + u8 value; + int err; + + err = pcf8523_read(client, REG_CONTROL3, &value); + if (err < 0) + return err; + + value = (value & ~REG_CONTROL3_PM_MASK) | pm; + + err = pcf8523_write(client, REG_CONTROL3, value); + if (err < 0) + return err; + + return 0; +} + +static int pcf8523_stop_rtc(struct i2c_client *client) +{ + u8 value; + int err; + + err = pcf8523_read(client, REG_CONTROL1, &value); + if (err < 0) + return err; + + value |= REG_CONTROL1_STOP; + + err = pcf8523_write(client, REG_CONTROL1, value); + if (err < 0) + return err; + + return 0; +} + +static int pcf8523_start_rtc(struct i2c_client *client) +{ + u8 value; + int err; + + err = pcf8523_read(client, REG_CONTROL1, &value); + if (err < 0) + return err; + + value &= ~REG_CONTROL1_STOP; + + err = pcf8523_write(client, REG_CONTROL1, value); + if (err < 0) + return err; + + return 0; +} + +static int pcf8523_rtc_read_time(struct device *dev, struct rtc_time *tm) +{ + struct i2c_client *client = to_i2c_client(dev); + u8 start = REG_SECONDS, regs[7]; + struct i2c_msg msgs[2]; + int err; + + msgs[0].addr = client->addr; + msgs[0].flags = 0; + msgs[0].len = 1; + msgs[0].buf = &start; + + msgs[1].addr = client->addr; + msgs[1].flags = I2C_M_RD; + msgs[1].len = sizeof(regs); + msgs[1].buf = regs; + + err = i2c_transfer(client->adapter, msgs, ARRAY_SIZE(msgs)); + if (err < 0) + return err; + + if (regs[0] & REG_SECONDS_OS) { + /* + * If the oscillator was stopped, try to clear the flag. Upon + * power-up the flag is always set, but if we cannot clear it + * the oscillator isn't running properly for some reason. The + * sensible thing therefore is to return an error, signalling + * that the clock cannot be assumed to be correct. + */ + + regs[0] &= ~REG_SECONDS_OS; + + err = pcf8523_write(client, REG_SECONDS, regs[0]); + if (err < 0) + return err; + + err = pcf8523_read(client, REG_SECONDS, ®s[0]); + if (err < 0) + return err; + + if (regs[0] & REG_SECONDS_OS) + return -EAGAIN; + } + + tm->tm_sec = bcd2bin(regs[0] & 0x7f); + tm->tm_min = bcd2bin(regs[1] & 0x7f); + tm->tm_hour = bcd2bin(regs[2] & 0x3f); + tm->tm_mday = bcd2bin(regs[3] & 0x3f); + tm->tm_wday = regs[4] & 0x7; + tm->tm_mon = bcd2bin(regs[5] & 0x1f); + tm->tm_year = bcd2bin(regs[6]) + 100; + + return rtc_valid_tm(tm); +} + +static int pcf8523_rtc_set_time(struct device *dev, struct rtc_time *tm) +{ + struct i2c_client *client = to_i2c_client(dev); + struct i2c_msg msg; + u8 regs[8]; + int err; + + err = pcf8523_stop_rtc(client); + if (err < 0) + return err; + + regs[0] = REG_SECONDS; + regs[1] = bin2bcd(tm->tm_sec); + regs[2] = bin2bcd(tm->tm_min); + regs[3] = bin2bcd(tm->tm_hour); + regs[4] = bin2bcd(tm->tm_mday); + regs[5] = tm->tm_wday; + regs[6] = bin2bcd(tm->tm_mon); + regs[7] = bin2bcd(tm->tm_year - 100); + + msg.addr = client->addr; + msg.flags = 0; + msg.len = sizeof(regs); + msg.buf = regs; + + err = i2c_transfer(client->adapter, &msg, 1); + if (err < 0) { + /* + * If the time cannot be set, restart the RTC anyway. Note + * that errors are ignored if the RTC cannot be started so + * that we have a chance to propagate the original error. + */ + pcf8523_start_rtc(client); + return err; + } + + return pcf8523_start_rtc(client); +} + +static const struct rtc_class_ops pcf8523_rtc_ops = { + .read_time = pcf8523_rtc_read_time, + .set_time = pcf8523_rtc_set_time, +}; + +static int pcf8523_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct pcf8523 *pcf; + int err; + + if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) + return -ENODEV; + + pcf = devm_kzalloc(&client->dev, sizeof(*pcf), GFP_KERNEL); + if (!pcf) + return -ENOMEM; + + err = pcf8523_select_capacitance(client, true); + if (err < 0) + return err; + + err = pcf8523_set_pm(client, 0); + if (err < 0) + return err; + + pcf->rtc = rtc_device_register(DRIVER_NAME, &client->dev, + &pcf8523_rtc_ops, THIS_MODULE); + if (IS_ERR(pcf->rtc)) + return PTR_ERR(pcf->rtc); + + i2c_set_clientdata(client, pcf); + + return 0; +} + +static int pcf8523_remove(struct i2c_client *client) +{ + struct pcf8523 *pcf = i2c_get_clientdata(client); + + rtc_device_unregister(pcf->rtc); + + return 0; +} + +static const struct i2c_device_id pcf8523_id[] = { + { "pcf8523", 0 }, + { } +}; +MODULE_DEVICE_TABLE(i2c, pcf8523_id); + +#ifdef CONFIG_OF +static const struct of_device_id pcf8523_of_match[] = { + { .compatible = "nxp,pcf8523" }, + { } +}; +MODULE_DEVICE_TABLE(of, pcf8523_of_match); +#endif + +static struct i2c_driver pcf8523_driver = { + .driver = { + .name = DRIVER_NAME, + .owner = THIS_MODULE, + .of_match_table = of_match_ptr(pcf8523_of_match), + }, + .probe = pcf8523_probe, + .remove = pcf8523_remove, + .id_table = pcf8523_id, +}; +module_i2c_driver(pcf8523_driver); + +MODULE_AUTHOR("Thierry Reding <thierry.reding@avionic-design.de>"); +MODULE_DESCRIPTION("NXP PCF8523 RTC driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c index a7a2a998fa91..4bd9414aee65 100644 --- a/drivers/rtc/rtc-s3c.c +++ b/drivers/rtc/rtc-s3c.c @@ -47,8 +47,6 @@ struct s3c_rtc_drv_data { /* I have yet to find an S3C implementation with more than one * of these rtc blocks in */ -static struct resource *s3c_rtc_mem; - static struct clk *rtc_clk; static void __iomem *s3c_rtc_base; static int s3c_rtc_alarmno = NO_IRQ; @@ -427,21 +425,13 @@ static int __devexit s3c_rtc_remove(struct platform_device *dev) { struct rtc_device *rtc = platform_get_drvdata(dev); - free_irq(s3c_rtc_alarmno, rtc); - free_irq(s3c_rtc_tickno, rtc); - platform_set_drvdata(dev, NULL); rtc_device_unregister(rtc); s3c_rtc_setaie(&dev->dev, 0); - clk_put(rtc_clk); rtc_clk = NULL; - iounmap(s3c_rtc_base); - release_resource(s3c_rtc_mem); - kfree(s3c_rtc_mem); - return 0; } @@ -496,28 +486,18 @@ static int __devinit s3c_rtc_probe(struct platform_device *pdev) return -ENOENT; } - s3c_rtc_mem = request_mem_region(res->start, resource_size(res), - pdev->name); - - if (s3c_rtc_mem == NULL) { - dev_err(&pdev->dev, "failed to reserve memory region\n"); - ret = -ENOENT; - goto err_nores; - } - - s3c_rtc_base = ioremap(res->start, resource_size(res)); + s3c_rtc_base = devm_request_and_ioremap(&pdev->dev, res); if (s3c_rtc_base == NULL) { - dev_err(&pdev->dev, "failed ioremap()\n"); - ret = -EINVAL; - goto err_nomap; + dev_err(&pdev->dev, "failed to ioremap memory region\n"); + return -EINVAL; } - rtc_clk = clk_get(&pdev->dev, "rtc"); + rtc_clk = devm_clk_get(&pdev->dev, "rtc"); if (IS_ERR(rtc_clk)) { dev_err(&pdev->dev, "failed to find rtc clock source\n"); ret = PTR_ERR(rtc_clk); rtc_clk = NULL; - goto err_clk; + return ret; } clk_enable(rtc_clk); @@ -576,28 +556,24 @@ static int __devinit s3c_rtc_probe(struct platform_device *pdev) s3c_rtc_setfreq(&pdev->dev, 1); - ret = request_irq(s3c_rtc_alarmno, s3c_rtc_alarmirq, + ret = devm_request_irq(&pdev->dev, s3c_rtc_alarmno, s3c_rtc_alarmirq, 0, "s3c2410-rtc alarm", rtc); if (ret) { dev_err(&pdev->dev, "IRQ%d error %d\n", s3c_rtc_alarmno, ret); goto err_alarm_irq; } - ret = request_irq(s3c_rtc_tickno, s3c_rtc_tickirq, + ret = devm_request_irq(&pdev->dev, s3c_rtc_tickno, s3c_rtc_tickirq, 0, "s3c2410-rtc tick", rtc); if (ret) { dev_err(&pdev->dev, "IRQ%d error %d\n", s3c_rtc_tickno, ret); - free_irq(s3c_rtc_alarmno, rtc); - goto err_tick_irq; + goto err_alarm_irq; } clk_disable(rtc_clk); return 0; - err_tick_irq: - free_irq(s3c_rtc_alarmno, rtc); - err_alarm_irq: platform_set_drvdata(pdev, NULL); rtc_device_unregister(rtc); @@ -605,15 +581,7 @@ static int __devinit s3c_rtc_probe(struct platform_device *pdev) err_nortc: s3c_rtc_enable(pdev, 0); clk_disable(rtc_clk); - clk_put(rtc_clk); - err_clk: - iounmap(s3c_rtc_base); - - err_nomap: - release_resource(s3c_rtc_mem); - - err_nores: return ret; } @@ -695,8 +663,6 @@ static const struct of_device_id s3c_rtc_dt_match[] = { {}, }; MODULE_DEVICE_TABLE(of, s3c_rtc_dt_match); -#else -#define s3c_rtc_dt_match NULL #endif static struct platform_device_id s3c_rtc_driver_ids[] = { @@ -727,7 +693,7 @@ static struct platform_driver s3c_rtc_driver = { .driver = { .name = "s3c-rtc", .owner = THIS_MODULE, - .of_match_table = s3c_rtc_dt_match, + .of_match_table = of_match_ptr(s3c_rtc_dt_match), }, }; diff --git a/drivers/rtc/rtc-spear.c b/drivers/rtc/rtc-spear.c index bb507d23f6ce..141fc945295f 100644 --- a/drivers/rtc/rtc-spear.c +++ b/drivers/rtc/rtc-spear.c @@ -363,35 +363,42 @@ static int __devinit spear_rtc_probe(struct platform_device *pdev) dev_err(&pdev->dev, "no resource defined\n"); return -EBUSY; } - if (!request_mem_region(res->start, resource_size(res), pdev->name)) { - dev_err(&pdev->dev, "rtc region already claimed\n"); - return -EBUSY; - } - config = kzalloc(sizeof(*config), GFP_KERNEL); + config = devm_kzalloc(&pdev->dev, sizeof(*config), GFP_KERNEL); if (!config) { dev_err(&pdev->dev, "out of memory\n"); - status = -ENOMEM; - goto err_release_region; + return -ENOMEM; } - config->clk = clk_get(&pdev->dev, NULL); - if (IS_ERR(config->clk)) { - status = PTR_ERR(config->clk); - goto err_kfree; + /* alarm irqs */ + irq = platform_get_irq(pdev, 0); + if (irq < 0) { + dev_err(&pdev->dev, "no update irq?\n"); + return irq; } - status = clk_enable(config->clk); - if (status < 0) - goto err_clk_put; + status = devm_request_irq(&pdev->dev, irq, spear_rtc_irq, 0, pdev->name, + config); + if (status) { + dev_err(&pdev->dev, "Alarm interrupt IRQ%d already claimed\n", + irq); + return status; + } - config->ioaddr = ioremap(res->start, resource_size(res)); + config->ioaddr = devm_request_and_ioremap(&pdev->dev, res); if (!config->ioaddr) { - dev_err(&pdev->dev, "ioremap fail\n"); - status = -ENOMEM; - goto err_disable_clock; + dev_err(&pdev->dev, "request-ioremap fail\n"); + return -ENOMEM; } + config->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(config->clk)) + return PTR_ERR(config->clk); + + status = clk_prepare_enable(config->clk); + if (status < 0) + return status; + spin_lock_init(&config->lock); platform_set_drvdata(pdev, config); @@ -401,42 +408,19 @@ static int __devinit spear_rtc_probe(struct platform_device *pdev) dev_err(&pdev->dev, "can't register RTC device, err %ld\n", PTR_ERR(config->rtc)); status = PTR_ERR(config->rtc); - goto err_iounmap; - } - - /* alarm irqs */ - irq = platform_get_irq(pdev, 0); - if (irq < 0) { - dev_err(&pdev->dev, "no update irq?\n"); - status = irq; - goto err_clear_platdata; + goto err_disable_clock; } - status = request_irq(irq, spear_rtc_irq, 0, pdev->name, config); - if (status) { - dev_err(&pdev->dev, "Alarm interrupt IRQ%d already \ - claimed\n", irq); - goto err_clear_platdata; - } + config->rtc->uie_unsupported = 1; if (!device_can_wakeup(&pdev->dev)) device_init_wakeup(&pdev->dev, 1); return 0; -err_clear_platdata: - platform_set_drvdata(pdev, NULL); - rtc_device_unregister(config->rtc); -err_iounmap: - iounmap(config->ioaddr); err_disable_clock: - clk_disable(config->clk); -err_clk_put: - clk_put(config->clk); -err_kfree: - kfree(config); -err_release_region: - release_mem_region(res->start, resource_size(res)); + platform_set_drvdata(pdev, NULL); + clk_disable_unprepare(config->clk); return status; } @@ -444,24 +428,11 @@ err_release_region: static int __devexit spear_rtc_remove(struct platform_device *pdev) { struct spear_rtc_config *config = platform_get_drvdata(pdev); - int irq; - struct resource *res; - /* leave rtc running, but disable irqs */ + rtc_device_unregister(config->rtc); spear_rtc_disable_interrupt(config); + clk_disable_unprepare(config->clk); device_init_wakeup(&pdev->dev, 0); - irq = platform_get_irq(pdev, 0); - if (irq) - free_irq(irq, pdev); - clk_disable(config->clk); - clk_put(config->clk); - iounmap(config->ioaddr); - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (res) - release_mem_region(res->start, resource_size(res)); - platform_set_drvdata(pdev, NULL); - rtc_device_unregister(config->rtc); - kfree(config); return 0; } diff --git a/drivers/rtc/rtc-test.c b/drivers/rtc/rtc-test.c index 7e96254bd365..974b9ae252ab 100644 --- a/drivers/rtc/rtc-test.c +++ b/drivers/rtc/rtc-test.c @@ -152,24 +152,24 @@ static int __init test_init(void) if ((test1 = platform_device_alloc("rtc-test", 1)) == NULL) { err = -ENOMEM; - goto exit_free_test0; + goto exit_put_test0; } if ((err = platform_device_add(test0))) - goto exit_free_test1; + goto exit_put_test1; if ((err = platform_device_add(test1))) - goto exit_device_unregister; + goto exit_del_test0; return 0; -exit_device_unregister: - platform_device_unregister(test0); +exit_del_test0: + platform_device_del(test0); -exit_free_test1: +exit_put_test1: platform_device_put(test1); -exit_free_test0: +exit_put_test0: platform_device_put(test0); exit_driver_unregister: diff --git a/drivers/rtc/rtc-tps65910.c b/drivers/rtc/rtc-tps65910.c index 073108dcf9e7..22eb4ebfa1a6 100644 --- a/drivers/rtc/rtc-tps65910.c +++ b/drivers/rtc/rtc-tps65910.c @@ -247,6 +247,13 @@ static int __devinit tps65910_rtc_probe(struct platform_device *pdev) return ret; dev_dbg(&pdev->dev, "Enabling rtc-tps65910.\n"); + + /* Enable RTC digital power domain */ + ret = regmap_update_bits(tps65910->regmap, TPS65910_DEVCTRL, + DEVCTRL_RTC_PWDN_MASK, 0 << DEVCTRL_RTC_PWDN_SHIFT); + if (ret < 0) + return ret; + rtc_reg = TPS65910_RTC_CTRL_STOP_RTC; ret = regmap_write(tps65910->regmap, TPS65910_RTC_CTRL, rtc_reg); if (ret < 0) @@ -261,7 +268,7 @@ static int __devinit tps65910_rtc_probe(struct platform_device *pdev) ret = devm_request_threaded_irq(&pdev->dev, irq, NULL, tps65910_rtc_interrupt, IRQF_TRIGGER_LOW, - "rtc-tps65910", &pdev->dev); + dev_name(&pdev->dev), &pdev->dev); if (ret < 0) { dev_err(&pdev->dev, "IRQ is not free.\n"); return ret; diff --git a/drivers/rtc/rtc-vt8500.c b/drivers/rtc/rtc-vt8500.c index 07bf19364a74..14e2d8cfcc83 100644 --- a/drivers/rtc/rtc-vt8500.c +++ b/drivers/rtc/rtc-vt8500.c @@ -210,7 +210,8 @@ static int __devinit vt8500_rtc_probe(struct platform_device *pdev) struct vt8500_rtc *vt8500_rtc; int ret; - vt8500_rtc = kzalloc(sizeof(struct vt8500_rtc), GFP_KERNEL); + vt8500_rtc = devm_kzalloc(&pdev->dev, + sizeof(struct vt8500_rtc), GFP_KERNEL); if (!vt8500_rtc) return -ENOMEM; @@ -220,15 +221,13 @@ static int __devinit vt8500_rtc_probe(struct platform_device *pdev) vt8500_rtc->res = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!vt8500_rtc->res) { dev_err(&pdev->dev, "No I/O memory resource defined\n"); - ret = -ENXIO; - goto err_free; + return -ENXIO; } vt8500_rtc->irq_alarm = platform_get_irq(pdev, 0); if (vt8500_rtc->irq_alarm < 0) { dev_err(&pdev->dev, "No alarm IRQ resource defined\n"); - ret = -ENXIO; - goto err_free; + return -ENXIO; } vt8500_rtc->res = request_mem_region(vt8500_rtc->res->start, @@ -236,8 +235,7 @@ static int __devinit vt8500_rtc_probe(struct platform_device *pdev) "vt8500-rtc"); if (vt8500_rtc->res == NULL) { dev_err(&pdev->dev, "failed to request I/O memory\n"); - ret = -EBUSY; - goto err_free; + return -EBUSY; } vt8500_rtc->regbase = ioremap(vt8500_rtc->res->start, @@ -278,8 +276,6 @@ err_unmap: err_release: release_mem_region(vt8500_rtc->res->start, resource_size(vt8500_rtc->res)); -err_free: - kfree(vt8500_rtc); return ret; } @@ -297,7 +293,6 @@ static int __devexit vt8500_rtc_remove(struct platform_device *pdev) release_mem_region(vt8500_rtc->res->start, resource_size(vt8500_rtc->res)); - kfree(vt8500_rtc); platform_set_drvdata(pdev, NULL); return 0; diff --git a/drivers/scsi/fcoe/fcoe_ctlr.c b/drivers/scsi/fcoe/fcoe_ctlr.c index 2ebe03a4b51d..4a909d7cfde1 100644 --- a/drivers/scsi/fcoe/fcoe_ctlr.c +++ b/drivers/scsi/fcoe/fcoe_ctlr.c @@ -2144,7 +2144,7 @@ static void fcoe_ctlr_vn_restart(struct fcoe_ctlr *fip) */ port_id = fip->port_id; if (fip->probe_tries) - port_id = prandom32(&fip->rnd_state) & 0xffff; + port_id = prandom_u32_state(&fip->rnd_state) & 0xffff; else if (!port_id) port_id = fip->lp->wwpn & 0xffff; if (!port_id || port_id == 0xffff) @@ -2169,7 +2169,7 @@ static void fcoe_ctlr_vn_restart(struct fcoe_ctlr *fip) static void fcoe_ctlr_vn_start(struct fcoe_ctlr *fip) { fip->probe_tries = 0; - prandom32_seed(&fip->rnd_state, fip->lp->wwpn); + prandom_seed_state(&fip->rnd_state, fip->lp->wwpn); fcoe_ctlr_vn_restart(fip); } diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c index 75c0c4f5fdf2..ab34497bcfee 100644 --- a/drivers/spi/spi-atmel.c +++ b/drivers/spi/spi-atmel.c @@ -20,6 +20,7 @@ #include <linux/spi/spi.h> #include <linux/slab.h> #include <linux/platform_data/atmel.h> +#include <linux/of.h> #include <asm/io.h> #include <asm/gpio.h> @@ -768,6 +769,10 @@ static int atmel_spi_setup(struct spi_device *spi) /* chipselect must have been muxed as GPIO (e.g. in board setup) */ npcs_pin = (unsigned int)spi->controller_data; + + if (gpio_is_valid(spi->cs_gpio)) + npcs_pin = spi->cs_gpio; + asd = spi->controller_state; if (!asd) { asd = kzalloc(sizeof(struct atmel_spi_device), GFP_KERNEL); @@ -937,8 +942,9 @@ static int atmel_spi_probe(struct platform_device *pdev) /* the spi->mode bits understood by this driver: */ master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH; + master->dev.of_node = pdev->dev.of_node; master->bus_num = pdev->id; - master->num_chipselect = 4; + master->num_chipselect = master->dev.of_node ? 0 : 4; master->setup = atmel_spi_setup; master->transfer = atmel_spi_transfer; master->cleanup = atmel_spi_cleanup; @@ -1064,11 +1070,20 @@ static int atmel_spi_resume(struct platform_device *pdev) #define atmel_spi_resume NULL #endif +#if defined(CONFIG_OF) +static const struct of_device_id atmel_spi_dt_ids[] = { + { .compatible = "atmel,at91rm9200-spi" }, + { /* sentinel */ } +}; + +MODULE_DEVICE_TABLE(of, atmel_spi_dt_ids); +#endif static struct platform_driver atmel_spi_driver = { .driver = { .name = "atmel_spi", .owner = THIS_MODULE, + .of_match_table = of_match_ptr(atmel_spi_dt_ids), }, .suspend = atmel_spi_suspend, .resume = atmel_spi_resume, diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c index 4dd7b7ce5c5a..ad93231a8038 100644 --- a/drivers/spi/spi-s3c64xx.c +++ b/drivers/spi/spi-s3c64xx.c @@ -215,6 +215,10 @@ static void flush_fifo(struct s3c64xx_spi_driver_data *sdd) writel(0, regs + S3C64XX_SPI_PACKET_CNT); val = readl(regs + S3C64XX_SPI_CH_CFG); + val &= ~(S3C64XX_SPI_CH_RXCH_ON | S3C64XX_SPI_CH_TXCH_ON); + writel(val, regs + S3C64XX_SPI_CH_CFG); + + val = readl(regs + S3C64XX_SPI_CH_CFG); val |= S3C64XX_SPI_CH_SW_RST; val &= ~S3C64XX_SPI_CH_HS_EN; writel(val, regs + S3C64XX_SPI_CH_CFG); @@ -248,10 +252,6 @@ static void flush_fifo(struct s3c64xx_spi_driver_data *sdd) val = readl(regs + S3C64XX_SPI_MODE_CFG); val &= ~(S3C64XX_SPI_MODE_TXDMA_ON | S3C64XX_SPI_MODE_RXDMA_ON); writel(val, regs + S3C64XX_SPI_MODE_CFG); - - val = readl(regs + S3C64XX_SPI_CH_CFG); - val &= ~(S3C64XX_SPI_CH_RXCH_ON | S3C64XX_SPI_CH_TXCH_ON); - writel(val, regs + S3C64XX_SPI_CH_CFG); } static void s3c64xx_spi_dmacb(void *data) @@ -771,8 +771,6 @@ static int s3c64xx_spi_transfer_one_message(struct spi_master *master, if (list_is_last(&xfer->transfer_list, &msg->transfers)) cs_toggle = 1; - else - disable_cs(sdd, spi); } msg->actual_length += xfer->len; diff --git a/drivers/spi/spi-sh-hspi.c b/drivers/spi/spi-sh-hspi.c index 32f7b55fce09..60cfae51c713 100644 --- a/drivers/spi/spi-sh-hspi.c +++ b/drivers/spi/spi-sh-hspi.c @@ -290,7 +290,7 @@ static int hspi_probe(struct platform_device *pdev) } clk = clk_get(NULL, "shyway_clk"); - if (!clk) { + if (IS_ERR(clk)) { dev_err(&pdev->dev, "shyway_clk is required\n"); ret = -EINVAL; goto error0; diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index ab095acdb2a8..19ee901577da 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -824,6 +824,7 @@ static void of_register_spi_devices(struct spi_master *master) struct spi_device *spi; struct device_node *nc; const __be32 *prop; + char modalias[SPI_NAME_SIZE + 4]; int rc; int len; @@ -887,7 +888,9 @@ static void of_register_spi_devices(struct spi_master *master) spi->dev.of_node = nc; /* Register the new device */ - request_module(spi->modalias); + snprintf(modalias, sizeof(modalias), "%s%s", SPI_MODULE_PREFIX, + spi->modalias); + request_module(modalias); rc = spi_add_device(spi); if (rc) { dev_err(&master->dev, "spi_device register error %s\n", diff --git a/drivers/staging/android/binder.c b/drivers/staging/android/binder.c index 4a36e9ab8cf7..2d12e8a1f82e 100644 --- a/drivers/staging/android/binder.c +++ b/drivers/staging/android/binder.c @@ -35,6 +35,7 @@ #include <linux/uaccess.h> #include <linux/vmalloc.h> #include <linux/slab.h> +#include <linux/pid_namespace.h> #include "binder.h" #include "binder_trace.h" @@ -2320,7 +2321,7 @@ retry: if (t->from) { struct task_struct *sender = t->from->proc->tsk; tr.sender_pid = task_tgid_nr_ns(sender, - current->nsproxy->pid_ns); + task_active_pid_ns(current)); } else { tr.sender_pid = 0; } diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c index 57cc9c6eaa9f..f1c6c5470b92 100644 --- a/drivers/usb/musb/musb_core.c +++ b/drivers/usb/musb/musb_core.c @@ -251,7 +251,7 @@ void musb_write_fifo(struct musb_hw_ep *hw_ep, u16 len, const u8 *src) /* best case is 32bit-aligned source address */ if ((0x02 & (unsigned long) src) == 0) { if (len >= 4) { - writesl(fifo, src + index, len >> 2); + iowrite32_rep(fifo, src + index, len >> 2); index += len & ~0x03; } if (len & 0x02) { @@ -260,7 +260,7 @@ void musb_write_fifo(struct musb_hw_ep *hw_ep, u16 len, const u8 *src) } } else { if (len >= 2) { - writesw(fifo, src + index, len >> 1); + iowrite16_rep(fifo, src + index, len >> 1); index += len & ~0x01; } } @@ -268,7 +268,7 @@ void musb_write_fifo(struct musb_hw_ep *hw_ep, u16 len, const u8 *src) musb_writeb(fifo, 0, src[index]); } else { /* byte aligned */ - writesb(fifo, src, len); + iowrite8_rep(fifo, src, len); } } @@ -294,7 +294,7 @@ void musb_read_fifo(struct musb_hw_ep *hw_ep, u16 len, u8 *dst) /* best case is 32bit-aligned destination address */ if ((0x02 & (unsigned long) dst) == 0) { if (len >= 4) { - readsl(fifo, dst, len >> 2); + ioread32_rep(fifo, dst, len >> 2); index = len & ~0x03; } if (len & 0x02) { @@ -303,7 +303,7 @@ void musb_read_fifo(struct musb_hw_ep *hw_ep, u16 len, u8 *dst) } } else { if (len >= 2) { - readsw(fifo, dst, len >> 1); + ioread16_rep(fifo, dst, len >> 1); index = len & ~0x01; } } @@ -311,7 +311,7 @@ void musb_read_fifo(struct musb_hw_ep *hw_ep, u16 len, u8 *dst) dst[index] = musb_readb(fifo, 0); } else { /* byte aligned */ - readsb(fifo, dst, len); + ioread8_rep(fifo, dst, len); } } #endif diff --git a/drivers/usb/musb/musb_io.h b/drivers/usb/musb/musb_io.h index 565ad1617832..eebeed78edd6 100644 --- a/drivers/usb/musb/musb_io.h +++ b/drivers/usb/musb/musb_io.h @@ -37,27 +37,6 @@ #include <linux/io.h> -#if !defined(CONFIG_ARM) && !defined(CONFIG_SUPERH) \ - && !defined(CONFIG_AVR32) && !defined(CONFIG_PPC32) \ - && !defined(CONFIG_PPC64) && !defined(CONFIG_BLACKFIN) \ - && !defined(CONFIG_MIPS) && !defined(CONFIG_M68K) \ - && !defined(CONFIG_XTENSA) -static inline void readsl(const void __iomem *addr, void *buf, int len) - { insl((unsigned long)addr, buf, len); } -static inline void readsw(const void __iomem *addr, void *buf, int len) - { insw((unsigned long)addr, buf, len); } -static inline void readsb(const void __iomem *addr, void *buf, int len) - { insb((unsigned long)addr, buf, len); } - -static inline void writesl(const void __iomem *addr, const void *buf, int len) - { outsl((unsigned long)addr, buf, len); } -static inline void writesw(const void __iomem *addr, const void *buf, int len) - { outsw((unsigned long)addr, buf, len); } -static inline void writesb(const void __iomem *addr, const void *buf, int len) - { outsb((unsigned long)addr, buf, len); } - -#endif - #ifndef CONFIG_BLACKFIN /* NOTE: these offsets are all in bytes */ diff --git a/drivers/usb/musb/tusb6010.c b/drivers/usb/musb/tusb6010.c index 8bde6fc5eb75..3969813c217d 100644 --- a/drivers/usb/musb/tusb6010.c +++ b/drivers/usb/musb/tusb6010.c @@ -22,6 +22,7 @@ #include <linux/prefetch.h> #include <linux/usb.h> #include <linux/irq.h> +#include <linux/io.h> #include <linux/platform_device.h> #include <linux/dma-mapping.h> #include <linux/usb/nop-usb-xceiv.h> @@ -198,7 +199,7 @@ void musb_write_fifo(struct musb_hw_ep *hw_ep, u16 len, const u8 *buf) /* Best case is 32bit-aligned destination address */ if ((0x02 & (unsigned long) buf) == 0) { if (len >= 4) { - writesl(fifo, buf, len >> 2); + iowrite32_rep(fifo, buf, len >> 2); buf += (len & ~0x03); len &= 0x03; } @@ -245,7 +246,7 @@ void musb_read_fifo(struct musb_hw_ep *hw_ep, u16 len, u8 *buf) /* Best case is 32bit-aligned destination address */ if ((0x02 & (unsigned long) buf) == 0) { if (len >= 4) { - readsl(fifo, buf, len >> 2); + ioread32_rep(fifo, buf, len >> 2); buf += (len & ~0x03); len &= 0x03; } diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index 9c31277b3a81..e7068c508800 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -2140,14 +2140,16 @@ config FB_UDL To compile as a module, choose M here: the module name is udlfb. config FB_IBM_GXT4500 - tristate "Framebuffer support for IBM GXT4500P adaptor" + tristate "Framebuffer support for IBM GXT4000P/4500P/6000P/6500P adaptors" depends on FB && PPC select FB_CFB_FILLRECT select FB_CFB_COPYAREA select FB_CFB_IMAGEBLIT ---help--- - Say Y here to enable support for the IBM GXT4500P display - adaptor, found on some IBM System P (pSeries) machines. + Say Y here to enable support for the IBM GXT4000P/6000P and + GXT4500P/6500P display adaptor based on Raster Engine RC1000, + found on some IBM System P (pSeries) machines. This driver + doesn't use Geometry Engine GT1000. config FB_PS3 tristate "PS3 GPU framebuffer driver" diff --git a/drivers/video/backlight/88pm860x_bl.c b/drivers/video/backlight/88pm860x_bl.c index b7ec34c57f46..c072ed9aea36 100644 --- a/drivers/video/backlight/88pm860x_bl.c +++ b/drivers/video/backlight/88pm860x_bl.c @@ -117,8 +117,8 @@ static int pm860x_backlight_set(struct backlight_device *bl, int brightness) data->current_brightness = value; return 0; out: - dev_dbg(chip->dev, "set brightness %d failure with return " - "value:%d\n", value, ret); + dev_dbg(chip->dev, "set brightness %d failure with return value: %d\n", + value, ret); return ret; } @@ -208,22 +208,19 @@ static int pm860x_backlight_probe(struct platform_device *pdev) res = platform_get_resource_byname(pdev, IORESOURCE_REG, "duty cycle"); if (!res) { dev_err(&pdev->dev, "No REG resource for duty cycle\n"); - ret = -ENXIO; - goto out; + return -ENXIO; } data->reg_duty_cycle = res->start; res = platform_get_resource_byname(pdev, IORESOURCE_REG, "always on"); if (!res) { dev_err(&pdev->dev, "No REG resorce for always on\n"); - ret = -ENXIO; - goto out; + return -ENXIO; } data->reg_always_on = res->start; res = platform_get_resource_byname(pdev, IORESOURCE_REG, "current"); if (!res) { dev_err(&pdev->dev, "No REG resource for current\n"); - ret = -ENXIO; - goto out; + return -ENXIO; } data->reg_current = res->start; @@ -231,8 +228,7 @@ static int pm860x_backlight_probe(struct platform_device *pdev) sprintf(name, "backlight-%d", pdev->id); data->port = pdev->id; data->chip = chip; - data->i2c = (chip->id == CHIP_PM8606) ? chip->client \ - : chip->companion; + data->i2c = (chip->id == CHIP_PM8606) ? chip->client : chip->companion; data->current_brightness = MAX_BRIGHTNESS; if (pm860x_backlight_dt_init(pdev, data, name)) { if (pdata) { @@ -263,8 +259,6 @@ static int pm860x_backlight_probe(struct platform_device *pdev) return 0; out_brt: backlight_device_unregister(bl); -out: - devm_kfree(&pdev->dev, data); return ret; } diff --git a/drivers/video/backlight/atmel-pwm-bl.c b/drivers/video/backlight/atmel-pwm-bl.c index df1cbb7ef6ca..de5e5e74e2a7 100644 --- a/drivers/video/backlight/atmel-pwm-bl.c +++ b/drivers/video/backlight/atmel-pwm-bl.c @@ -106,10 +106,9 @@ static int atmel_pwm_bl_init_pwm(struct atmel_pwm_bl *pwmbl) pwm_channel_writel(&pwmbl->pwmc, PWM_CPRD, pwmbl->pdata->pwm_compare_max); - dev_info(&pwmbl->pdev->dev, "Atmel PWM backlight driver " - "(%lu Hz)\n", pwmbl->pwmc.mck / - pwmbl->pdata->pwm_compare_max / - (1 << prescale)); + dev_info(&pwmbl->pdev->dev, "Atmel PWM backlight driver (%lu Hz)\n", + pwmbl->pwmc.mck / pwmbl->pdata->pwm_compare_max / + (1 << prescale)); return pwm_channel_enable(&pwmbl->pwmc); } diff --git a/drivers/video/backlight/backlight.c b/drivers/video/backlight/backlight.c index 297db2fa91f5..345f6660d4b3 100644 --- a/drivers/video/backlight/backlight.c +++ b/drivers/video/backlight/backlight.c @@ -370,6 +370,35 @@ void backlight_device_unregister(struct backlight_device *bd) } EXPORT_SYMBOL(backlight_device_unregister); +#ifdef CONFIG_OF +static int of_parent_match(struct device *dev, void *data) +{ + return dev->parent && dev->parent->of_node == data; +} + +/** + * of_find_backlight_by_node() - find backlight device by device-tree node + * @node: device-tree node of the backlight device + * + * Returns a pointer to the backlight device corresponding to the given DT + * node or NULL if no such backlight device exists or if the device hasn't + * been probed yet. + * + * This function obtains a reference on the backlight device and it is the + * caller's responsibility to drop the reference by calling put_device() on + * the backlight device's .dev field. + */ +struct backlight_device *of_find_backlight_by_node(struct device_node *node) +{ + struct device *dev; + + dev = class_find_device(backlight_class, NULL, node, of_parent_match); + + return dev ? to_backlight_device(dev) : NULL; +} +EXPORT_SYMBOL(of_find_backlight_by_node); +#endif + static void __exit backlight_class_exit(void) { class_destroy(backlight_class); diff --git a/drivers/video/backlight/corgi_lcd.c b/drivers/video/backlight/corgi_lcd.c index eaaebf21993e..e323fcbe884e 100644 --- a/drivers/video/backlight/corgi_lcd.c +++ b/drivers/video/backlight/corgi_lcd.c @@ -6,8 +6,8 @@ * Based on Sharp's 2.4 Backlight Driver * * Copyright (c) 2008 Marvell International Ltd. - * Converted to SPI device based LCD/Backlight device driver - * by Eric Miao <eric.miao@marvell.com> + * Converted to SPI device based LCD/Backlight device driver + * by Eric Miao <eric.miao@marvell.com> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -192,7 +192,7 @@ static void lcdtg_set_phadadj(struct corgi_lcd *lcd, int mode) { int adj; - switch(mode) { + switch (mode) { case CORGI_LCD_MODE_VGA: /* Setting for VGA */ adj = sharpsl_param.phadadj; @@ -409,10 +409,10 @@ static int corgi_bl_set_intensity(struct corgi_lcd *lcd, int intensity) cont = !!(intensity & 0x20) ^ lcd->gpio_backlight_cont_inverted; if (gpio_is_valid(lcd->gpio_backlight_cont)) - gpio_set_value(lcd->gpio_backlight_cont, cont); + gpio_set_value_cansleep(lcd->gpio_backlight_cont, cont); if (gpio_is_valid(lcd->gpio_backlight_on)) - gpio_set_value(lcd->gpio_backlight_on, intensity); + gpio_set_value_cansleep(lcd->gpio_backlight_on, intensity); if (lcd->kick_battery) lcd->kick_battery(); @@ -495,8 +495,9 @@ static int setup_gpio_backlight(struct corgi_lcd *lcd, err = devm_gpio_request(&spi->dev, pdata->gpio_backlight_on, "BL_ON"); if (err) { - dev_err(&spi->dev, "failed to request GPIO%d for " - "backlight_on\n", pdata->gpio_backlight_on); + dev_err(&spi->dev, + "failed to request GPIO%d for backlight_on\n", + pdata->gpio_backlight_on); return err; } @@ -508,8 +509,9 @@ static int setup_gpio_backlight(struct corgi_lcd *lcd, err = devm_gpio_request(&spi->dev, pdata->gpio_backlight_cont, "BL_CONT"); if (err) { - dev_err(&spi->dev, "failed to request GPIO%d for " - "backlight_cont\n", pdata->gpio_backlight_cont); + dev_err(&spi->dev, + "failed to request GPIO%d for backlight_cont\n", + pdata->gpio_backlight_cont); return err; } diff --git a/drivers/video/backlight/da903x_bl.c b/drivers/video/backlight/da903x_bl.c index 573c7ece0fde..8179cef0730f 100644 --- a/drivers/video/backlight/da903x_bl.c +++ b/drivers/video/backlight/da903x_bl.c @@ -2,10 +2,10 @@ * Backlight driver for Dialog Semiconductor DA9030/DA9034 * * Copyright (C) 2008 Compulab, Ltd. - * Mike Rapoport <mike@compulab.co.il> + * Mike Rapoport <mike@compulab.co.il> * * Copyright (C) 2006-2008 Marvell International Ltd. - * Eric Miao <eric.miao@marvell.com> + * Eric Miao <eric.miao@marvell.com> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -164,15 +164,14 @@ static int da903x_backlight_remove(struct platform_device *pdev) #ifdef CONFIG_PM static int da903x_backlight_suspend(struct device *dev) { - struct platform_device *pdev = to_platform_device(dev); - struct backlight_device *bl = platform_get_drvdata(pdev); + struct backlight_device *bl = dev_get_drvdata(dev); + return da903x_backlight_set(bl, 0); } static int da903x_backlight_resume(struct device *dev) { - struct platform_device *pdev = to_platform_device(dev); - struct backlight_device *bl = platform_get_drvdata(pdev); + struct backlight_device *bl = dev_get_drvdata(dev); backlight_update_status(bl); return 0; @@ -199,7 +198,7 @@ static struct platform_driver da903x_backlight_driver = { module_platform_driver(da903x_backlight_driver); MODULE_DESCRIPTION("Backlight Driver for Dialog Semiconductor DA9030/DA9034"); -MODULE_AUTHOR("Eric Miao <eric.miao@marvell.com>" - "Mike Rapoport <mike@compulab.co.il>"); +MODULE_AUTHOR("Eric Miao <eric.miao@marvell.com>"); +MODULE_AUTHOR("Mike Rapoport <mike@compulab.co.il>"); MODULE_LICENSE("GPL"); MODULE_ALIAS("platform:da903x-backlight"); diff --git a/drivers/video/backlight/da9052_bl.c b/drivers/video/backlight/da9052_bl.c index ac196181fe45..842da5a3ac4f 100644 --- a/drivers/video/backlight/da9052_bl.c +++ b/drivers/video/backlight/da9052_bl.c @@ -34,7 +34,7 @@ enum { DA9052_TYPE_WLED3, }; -static unsigned char wled_bank[] = { +static const unsigned char wled_bank[] = { DA9052_LED1_CONF_REG, DA9052_LED2_CONF_REG, DA9052_LED3_CONF_REG, diff --git a/drivers/video/backlight/generic_bl.c b/drivers/video/backlight/generic_bl.c index 8c660fcd250d..0ae155be9c89 100644 --- a/drivers/video/backlight/generic_bl.c +++ b/drivers/video/backlight/generic_bl.c @@ -97,8 +97,8 @@ static int genericbl_probe(struct platform_device *pdev) props.max_brightness = machinfo->max_intensity; bd = backlight_device_register(name, &pdev->dev, NULL, &genericbl_ops, &props); - if (IS_ERR (bd)) - return PTR_ERR (bd); + if (IS_ERR(bd)) + return PTR_ERR(bd); platform_set_drvdata(pdev, bd); diff --git a/drivers/video/backlight/hp680_bl.c b/drivers/video/backlight/hp680_bl.c index c99966342448..5cefd73526f8 100644 --- a/drivers/video/backlight/hp680_bl.c +++ b/drivers/video/backlight/hp680_bl.c @@ -26,7 +26,7 @@ #define HP680_DEFAULT_INTENSITY 10 static int hp680bl_suspended; -static int current_intensity = 0; +static int current_intensity; static DEFINE_SPINLOCK(bl_lock); static void hp680bl_send_intensity(struct backlight_device *bd) @@ -168,7 +168,7 @@ static int __init hp680bl_init(void) static void __exit hp680bl_exit(void) { platform_device_unregister(hp680bl_device); - platform_driver_unregister(&hp680bl_driver); + platform_driver_unregister(&hp680bl_driver); } module_init(hp680bl_init); diff --git a/drivers/video/backlight/ili9320.c b/drivers/video/backlight/ili9320.c index 66cc313185ad..1235bf9defc4 100644 --- a/drivers/video/backlight/ili9320.c +++ b/drivers/video/backlight/ili9320.c @@ -45,7 +45,7 @@ static inline int ili9320_write_spi(struct ili9320 *ili, /* second message is the data to transfer */ data[0] = spi->id | ILI9320_SPI_DATA | ILI9320_SPI_WRITE; - data[1] = value >> 8; + data[1] = value >> 8; data[2] = value; return spi_sync(spi->dev, &spi->message); @@ -56,11 +56,10 @@ int ili9320_write(struct ili9320 *ili, unsigned int reg, unsigned int value) dev_dbg(ili->dev, "write: reg=%02x, val=%04x\n", reg, value); return ili->write(ili, reg, value); } - EXPORT_SYMBOL_GPL(ili9320_write); int ili9320_write_regs(struct ili9320 *ili, - struct ili9320_reg *values, + const struct ili9320_reg *values, int nr_values) { int index; @@ -74,7 +73,6 @@ int ili9320_write_regs(struct ili9320 *ili, return 0; } - EXPORT_SYMBOL_GPL(ili9320_write_regs); static void ili9320_reset(struct ili9320 *lcd) @@ -260,7 +258,6 @@ int ili9320_probe_spi(struct spi_device *spi, return ret; } - EXPORT_SYMBOL_GPL(ili9320_probe_spi); int ili9320_remove(struct ili9320 *ili) @@ -271,7 +268,6 @@ int ili9320_remove(struct ili9320 *ili) return 0; } - EXPORT_SYMBOL_GPL(ili9320_remove); #ifdef CONFIG_PM @@ -296,20 +292,17 @@ int ili9320_suspend(struct ili9320 *lcd, pm_message_t state) return 0; } - EXPORT_SYMBOL_GPL(ili9320_suspend); int ili9320_resume(struct ili9320 *lcd) { dev_info(lcd->dev, "resuming from power state %d\n", lcd->power); - if (lcd->platdata->suspend == ILI9320_SUSPEND_DEEP) { + if (lcd->platdata->suspend == ILI9320_SUSPEND_DEEP) ili9320_write(lcd, ILI9320_POWER1, 0x00); - } return ili9320_power(lcd, FB_BLANK_UNBLANK); } - EXPORT_SYMBOL_GPL(ili9320_resume); #endif @@ -318,7 +311,6 @@ void ili9320_shutdown(struct ili9320 *lcd) { ili9320_power(lcd, FB_BLANK_POWERDOWN); } - EXPORT_SYMBOL_GPL(ili9320_shutdown); MODULE_AUTHOR("Ben Dooks <ben-linux@fluff.org>"); diff --git a/drivers/video/backlight/ili9320.h b/drivers/video/backlight/ili9320.h index e388eca7cac5..e0db738f7bb9 100644 --- a/drivers/video/backlight/ili9320.h +++ b/drivers/video/backlight/ili9320.h @@ -63,7 +63,7 @@ extern int ili9320_write(struct ili9320 *ili, unsigned int reg, unsigned int value); extern int ili9320_write_regs(struct ili9320 *ili, - struct ili9320_reg *values, + const struct ili9320_reg *values, int nr_values); /* Device probe */ diff --git a/drivers/video/backlight/jornada720_bl.c b/drivers/video/backlight/jornada720_bl.c index 16f593b64427..fef6ce4fad71 100644 --- a/drivers/video/backlight/jornada720_bl.c +++ b/drivers/video/backlight/jornada720_bl.c @@ -48,7 +48,7 @@ static int jornada_bl_get_brightness(struct backlight_device *bd) jornada_ssp_end(); - return (BL_MAX_BRIGHT - ret); + return BL_MAX_BRIGHT - ret; } static int jornada_bl_update_status(struct backlight_device *bd) @@ -77,18 +77,23 @@ static int jornada_bl_update_status(struct backlight_device *bd) goto out; } - /* at this point we expect that the mcu has accepted - our command and is waiting for our new value - please note that maximum brightness is 255, - but due to physical layout it is equal to 0, so we simply - invert the value (MAX VALUE - NEW VALUE). */ - if (jornada_ssp_byte(BL_MAX_BRIGHT - bd->props.brightness) != TXDUMMY) { + /* + * at this point we expect that the mcu has accepted + * our command and is waiting for our new value + * please note that maximum brightness is 255, + * but due to physical layout it is equal to 0, so we simply + * invert the value (MAX VALUE - NEW VALUE). + */ + if (jornada_ssp_byte(BL_MAX_BRIGHT - bd->props.brightness) + != TXDUMMY) { pr_err("set brightness failed\n"); ret = -ETIMEDOUT; } - /* If infact we get an TXDUMMY as output we are happy and dont - make any further comments about it */ + /* + * If infact we get an TXDUMMY as output we are happy and dont + * make any further comments about it + */ out: jornada_ssp_end(); @@ -121,9 +126,11 @@ static int jornada_bl_probe(struct platform_device *pdev) bd->props.power = FB_BLANK_UNBLANK; bd->props.brightness = BL_DEF_BRIGHT; - /* note. make sure max brightness is set otherwise - you will get seemingly non-related errors when - trying to change brightness */ + /* + * note. make sure max brightness is set otherwise + * you will get seemingly non-related errors when + * trying to change brightness + */ jornada_bl_update_status(bd); platform_set_drvdata(pdev, bd); diff --git a/drivers/video/backlight/l4f00242t03.c b/drivers/video/backlight/l4f00242t03.c index f5aa0a5961d6..9a35196d12d7 100644 --- a/drivers/video/backlight/l4f00242t03.c +++ b/drivers/video/backlight/l4f00242t03.c @@ -4,7 +4,7 @@ * Copyright 2007-2009 Freescale Semiconductor, Inc. All Rights Reserved. * * Copyright (c) 2009 Alberto Panizzo <maramaopercheseimorto@gmail.com> - * Inspired by Marek Vasut work in l4f00242t03.c + * Inspired by Marek Vasut work in l4f00242t03.c * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -33,7 +33,6 @@ struct l4f00242t03_priv { struct regulator *core_reg; }; - static void l4f00242t03_reset(unsigned int gpio) { pr_debug("l4f00242t03_reset.\n"); diff --git a/drivers/video/backlight/lcd.c b/drivers/video/backlight/lcd.c index a5d0d024bb92..34fb6bd798c8 100644 --- a/drivers/video/backlight/lcd.c +++ b/drivers/video/backlight/lcd.c @@ -108,7 +108,7 @@ static ssize_t lcd_show_power(struct device *dev, struct device_attribute *attr, static ssize_t lcd_store_power(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - int rc = -ENXIO; + int rc; struct lcd_device *ld = to_lcd_device(dev); unsigned long power; @@ -116,6 +116,8 @@ static ssize_t lcd_store_power(struct device *dev, if (rc) return rc; + rc = -ENXIO; + mutex_lock(&ld->ops_lock); if (ld->ops && ld->ops->set_power) { pr_debug("set power to %lu\n", power); @@ -144,7 +146,7 @@ static ssize_t lcd_show_contrast(struct device *dev, static ssize_t lcd_store_contrast(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - int rc = -ENXIO; + int rc; struct lcd_device *ld = to_lcd_device(dev); unsigned long contrast; @@ -152,6 +154,8 @@ static ssize_t lcd_store_contrast(struct device *dev, if (rc) return rc; + rc = -ENXIO; + mutex_lock(&ld->ops_lock); if (ld->ops && ld->ops->set_contrast) { pr_debug("set contrast to %lu\n", contrast); diff --git a/drivers/video/backlight/lm3630_bl.c b/drivers/video/backlight/lm3630_bl.c index 0207bc0a4407..a6d637b5c68f 100644 --- a/drivers/video/backlight/lm3630_bl.c +++ b/drivers/video/backlight/lm3630_bl.c @@ -37,7 +37,7 @@ enum lm3630_leds { BLED_2 }; -static const char *bled_name[] = { +static const char * const bled_name[] = { [BLED_ALL] = "lm3630_bled", /*Bank1 controls all string */ [BLED_1] = "lm3630_bled1", /*Bank1 controls bled1 */ [BLED_2] = "lm3630_bled2", /*Bank1 or 2 controls bled2 */ diff --git a/drivers/video/backlight/lm3639_bl.c b/drivers/video/backlight/lm3639_bl.c index b0e1e8ba4d9f..7ab2d2a04e41 100644 --- a/drivers/video/backlight/lm3639_bl.c +++ b/drivers/video/backlight/lm3639_bl.c @@ -214,7 +214,7 @@ out_input: } -static DEVICE_ATTR(bled_mode, 0666, NULL, lm3639_bled_mode_store); +static DEVICE_ATTR(bled_mode, S_IWUSR, NULL, lm3639_bled_mode_store); /* torch */ static void lm3639_torch_brightness_set(struct led_classdev *cdev, diff --git a/drivers/video/backlight/lms283gf05.c b/drivers/video/backlight/lms283gf05.c index b29c7071c9db..55819b384701 100644 --- a/drivers/video/backlight/lms283gf05.c +++ b/drivers/video/backlight/lms283gf05.c @@ -31,7 +31,7 @@ struct lms283gf05_seq { }; /* Magic sequences supplied by manufacturer, for details refer to datasheet */ -static struct lms283gf05_seq disp_initseq[] = { +static const struct lms283gf05_seq disp_initseq[] = { /* REG, VALUE, DELAY */ { 0x07, 0x0000, 0 }, { 0x13, 0x0000, 10 }, @@ -78,7 +78,7 @@ static struct lms283gf05_seq disp_initseq[] = { { 0x22, 0x0000, 0 } }; -static struct lms283gf05_seq disp_pdwnseq[] = { +static const struct lms283gf05_seq disp_pdwnseq[] = { { 0x07, 0x0016, 30 }, { 0x07, 0x0004, 0 }, @@ -104,7 +104,7 @@ static void lms283gf05_reset(unsigned long gpio, bool inverted) } static void lms283gf05_toggle(struct spi_device *spi, - struct lms283gf05_seq *seq, int sz) + const struct lms283gf05_seq *seq, int sz) { char buf[3]; int i; @@ -158,13 +158,10 @@ static int lms283gf05_probe(struct spi_device *spi) int ret = 0; if (pdata != NULL) { - ret = devm_gpio_request(&spi->dev, pdata->reset_gpio, - "LMS285GF05 RESET"); - if (ret) - return ret; - - ret = gpio_direction_output(pdata->reset_gpio, - !pdata->reset_inverted); + ret = devm_gpio_request_one(&spi->dev, pdata->reset_gpio, + GPIOF_DIR_OUT | (!pdata->reset_inverted ? + GPIOF_INIT_HIGH : GPIOF_INIT_LOW), + "LMS285GF05 RESET"); if (ret) return ret; } diff --git a/drivers/video/backlight/locomolcd.c b/drivers/video/backlight/locomolcd.c index 3a6d5419e3e3..146fea8aa431 100644 --- a/drivers/video/backlight/locomolcd.c +++ b/drivers/video/backlight/locomolcd.c @@ -107,7 +107,6 @@ void locomolcd_power(int on) } EXPORT_SYMBOL(locomolcd_power); - static int current_intensity; static int locomolcd_set_intensity(struct backlight_device *bd) @@ -122,13 +121,25 @@ static int locomolcd_set_intensity(struct backlight_device *bd) intensity = 0; switch (intensity) { - /* AC and non-AC are handled differently, but produce same results in sharp code? */ - case 0: locomo_frontlight_set(locomolcd_dev, 0, 0, 161); break; - case 1: locomo_frontlight_set(locomolcd_dev, 117, 0, 161); break; - case 2: locomo_frontlight_set(locomolcd_dev, 163, 0, 148); break; - case 3: locomo_frontlight_set(locomolcd_dev, 194, 0, 161); break; - case 4: locomo_frontlight_set(locomolcd_dev, 194, 1, 161); break; - + /* + * AC and non-AC are handled differently, + * but produce same results in sharp code? + */ + case 0: + locomo_frontlight_set(locomolcd_dev, 0, 0, 161); + break; + case 1: + locomo_frontlight_set(locomolcd_dev, 117, 0, 161); + break; + case 2: + locomo_frontlight_set(locomolcd_dev, 163, 0, 148); + break; + case 3: + locomo_frontlight_set(locomolcd_dev, 194, 0, 161); + break; + case 4: + locomo_frontlight_set(locomolcd_dev, 194, 1, 161); + break; default: return -ENODEV; } @@ -175,9 +186,11 @@ static int locomolcd_probe(struct locomo_dev *ldev) locomo_gpio_set_dir(ldev->dev.parent, LOCOMO_GPIO_FL_VR, 0); - /* the poodle_lcd_power function is called for the first time + /* + * the poodle_lcd_power function is called for the first time * from fs_initcall, which is before locomo is activated. - * We need to recall poodle_lcd_power here*/ + * We need to recall poodle_lcd_power here + */ if (machine_is_poodle()) locomolcd_power(1); @@ -190,8 +203,8 @@ static int locomolcd_probe(struct locomo_dev *ldev) &ldev->dev, NULL, &locomobl_data, &props); - if (IS_ERR (locomolcd_bl_device)) - return PTR_ERR (locomolcd_bl_device); + if (IS_ERR(locomolcd_bl_device)) + return PTR_ERR(locomolcd_bl_device); /* Set up frontlight so that screen is readable */ locomolcd_bl_device->props.brightness = 2; @@ -226,7 +239,6 @@ static struct locomo_driver poodle_lcd_driver = { .resume = locomolcd_resume, }; - static int __init locomolcd_init(void) { return locomo_driver_register(&poodle_lcd_driver); diff --git a/drivers/video/backlight/lp855x_bl.c b/drivers/video/backlight/lp855x_bl.c index fd985e0681e9..6e4db0c874c8 100644 --- a/drivers/video/backlight/lp855x_bl.c +++ b/drivers/video/backlight/lp855x_bl.c @@ -15,6 +15,7 @@ #include <linux/backlight.h> #include <linux/err.h> #include <linux/platform_data/lp855x.h> +#include <linux/pwm.h> /* Registers */ #define BRIGHTNESS_CTRL 0x00 @@ -34,22 +35,19 @@ struct lp855x { struct i2c_client *client; struct backlight_device *bl; struct device *dev; - struct mutex xfer_lock; struct lp855x_platform_data *pdata; + struct pwm_device *pwm; }; static int lp855x_read_byte(struct lp855x *lp, u8 reg, u8 *data) { int ret; - mutex_lock(&lp->xfer_lock); ret = i2c_smbus_read_byte_data(lp->client, reg); if (ret < 0) { - mutex_unlock(&lp->xfer_lock); dev_err(lp->dev, "failed to read 0x%.2x\n", reg); return ret; } - mutex_unlock(&lp->xfer_lock); *data = (u8)ret; return 0; @@ -57,13 +55,7 @@ static int lp855x_read_byte(struct lp855x *lp, u8 reg, u8 *data) static int lp855x_write_byte(struct lp855x *lp, u8 reg, u8 data) { - int ret; - - mutex_lock(&lp->xfer_lock); - ret = i2c_smbus_write_byte_data(lp->client, reg, data); - mutex_unlock(&lp->xfer_lock); - - return ret; + return i2c_smbus_write_byte_data(lp->client, reg, data); } static bool lp855x_is_valid_rom_area(struct lp855x *lp, u8 addr) @@ -121,6 +113,28 @@ static int lp855x_init_registers(struct lp855x *lp) return ret; } +static void lp855x_pwm_ctrl(struct lp855x *lp, int br, int max_br) +{ + unsigned int period = lp->pdata->period_ns; + unsigned int duty = br * period / max_br; + struct pwm_device *pwm; + + /* request pwm device with the consumer name */ + if (!lp->pwm) { + pwm = devm_pwm_get(lp->dev, lp->chipname); + if (IS_ERR(pwm)) + return; + + lp->pwm = pwm; + } + + pwm_config(lp->pwm, duty, period); + if (duty) + pwm_enable(lp->pwm); + else + pwm_disable(lp->pwm); +} + static int lp855x_bl_update_status(struct backlight_device *bl) { struct lp855x *lp = bl_get_data(bl); @@ -130,12 +144,10 @@ static int lp855x_bl_update_status(struct backlight_device *bl) bl->props.brightness = 0; if (mode == PWM_BASED) { - struct lp855x_pwm_data *pd = &lp->pdata->pwm_data; int br = bl->props.brightness; int max_br = bl->props.max_brightness; - if (pd->pwm_set_intensity) - pd->pwm_set_intensity(br, max_br); + lp855x_pwm_ctrl(lp, br, max_br); } else if (mode == REGISTER_BASED) { u8 val = bl->props.brightness; @@ -150,14 +162,7 @@ static int lp855x_bl_get_brightness(struct backlight_device *bl) struct lp855x *lp = bl_get_data(bl); enum lp855x_brightness_ctrl_mode mode = lp->pdata->mode; - if (mode == PWM_BASED) { - struct lp855x_pwm_data *pd = &lp->pdata->pwm_data; - int max_br = bl->props.max_brightness; - - if (pd->pwm_get_intensity) - bl->props.brightness = pd->pwm_get_intensity(max_br); - - } else if (mode == REGISTER_BASED) { + if (mode == REGISTER_BASED) { u8 val = 0; lp855x_read_byte(lp, BRIGHTNESS_CTRL, &val); @@ -266,8 +271,6 @@ static int lp855x_probe(struct i2c_client *cl, const struct i2c_device_id *id) lp->chip_id = id->driver_data; i2c_set_clientdata(cl, lp); - mutex_init(&lp->xfer_lock); - ret = lp855x_init_registers(lp); if (ret) { dev_err(lp->dev, "i2c communication err: %d", ret); diff --git a/drivers/video/backlight/max8925_bl.c b/drivers/video/backlight/max8925_bl.c index c6bec7aab87b..2c9bce050aa9 100644 --- a/drivers/video/backlight/max8925_bl.c +++ b/drivers/video/backlight/max8925_bl.c @@ -120,15 +120,13 @@ static int max8925_backlight_probe(struct platform_device *pdev) res = platform_get_resource(pdev, IORESOURCE_REG, 0); if (!res) { dev_err(&pdev->dev, "No REG resource for mode control!\n"); - ret = -ENXIO; - goto out; + return -ENXIO; } data->reg_mode_cntl = res->start; res = platform_get_resource(pdev, IORESOURCE_REG, 1); if (!res) { dev_err(&pdev->dev, "No REG resource for control!\n"); - ret = -ENXIO; - goto out; + return -ENXIO; } data->reg_cntl = res->start; @@ -142,8 +140,7 @@ static int max8925_backlight_probe(struct platform_device *pdev) &max8925_backlight_ops, &props); if (IS_ERR(bl)) { dev_err(&pdev->dev, "failed to register backlight\n"); - ret = PTR_ERR(bl); - goto out; + return PTR_ERR(bl); } bl->props.brightness = MAX_BRIGHTNESS; @@ -166,8 +163,6 @@ static int max8925_backlight_probe(struct platform_device *pdev) return 0; out_brt: backlight_device_unregister(bl); -out: - devm_kfree(&pdev->dev, data); return ret; } diff --git a/drivers/video/backlight/omap1_bl.c b/drivers/video/backlight/omap1_bl.c index 9a046a4c98f5..af31c269baa6 100644 --- a/drivers/video/backlight/omap1_bl.c +++ b/drivers/video/backlight/omap1_bl.c @@ -42,12 +42,12 @@ struct omap_backlight { struct omap_backlight_config *pdata; }; -static void inline omapbl_send_intensity(int intensity) +static inline void omapbl_send_intensity(int intensity) { omap_writeb(intensity, OMAP_PWL_ENABLE); } -static void inline omapbl_send_enable(int enable) +static inline void omapbl_send_enable(int enable) { omap_writeb(enable, OMAP_PWL_CLK_ENABLE); } diff --git a/drivers/video/backlight/pandora_bl.c b/drivers/video/backlight/pandora_bl.c index 4ec30748b447..633b0a22fd64 100644 --- a/drivers/video/backlight/pandora_bl.c +++ b/drivers/video/backlight/pandora_bl.c @@ -71,8 +71,7 @@ static int pandora_backlight_update_status(struct backlight_device *bl) * set PWM duty cycle to max. TPS61161 seems to use this * to calibrate it's PWM sensitivity when it starts. */ - twl_i2c_write_u8(TWL4030_MODULE_PWM0, MAX_VALUE, - TWL_PWM0_OFF); + twl_i2c_write_u8(TWL_MODULE_PWM, MAX_VALUE, TWL_PWM0_OFF); /* first enable clock, then PWM0 out */ twl_i2c_read_u8(TWL4030_MODULE_INTBR, &r, TWL_INTBR_GPBR1); @@ -90,8 +89,7 @@ static int pandora_backlight_update_status(struct backlight_device *bl) usleep_range(2000, 10000); } - twl_i2c_write_u8(TWL4030_MODULE_PWM0, MIN_VALUE + brightness, - TWL_PWM0_OFF); + twl_i2c_write_u8(TWL_MODULE_PWM, MIN_VALUE + brightness, TWL_PWM0_OFF); done: if (brightness != 0) @@ -132,7 +130,7 @@ static int pandora_backlight_probe(struct platform_device *pdev) platform_set_drvdata(pdev, bl); /* 64 cycle period, ON position 0 */ - twl_i2c_write_u8(TWL4030_MODULE_PWM0, 0x80, TWL_PWM0_ON); + twl_i2c_write_u8(TWL_MODULE_PWM, 0x80, TWL_PWM0_ON); bl->props.state |= PANDORABL_WAS_OFF; bl->props.brightness = MAX_USER_VALUE; diff --git a/drivers/video/backlight/pcf50633-backlight.c b/drivers/video/backlight/pcf50633-backlight.c index 0087396007e4..e87c7a3394f3 100644 --- a/drivers/video/backlight/pcf50633-backlight.c +++ b/drivers/video/backlight/pcf50633-backlight.c @@ -52,7 +52,7 @@ int pcf50633_bl_set_brightness_limit(struct pcf50633 *pcf, unsigned int limit) pcf_bl->brightness_limit = limit & 0x3f; backlight_update_status(pcf_bl->bl); - return 0; + return 0; } static int pcf50633_bl_update_status(struct backlight_device *bl) @@ -136,8 +136,10 @@ static int pcf50633_bl_probe(struct platform_device *pdev) pcf50633_reg_write(pcf_bl->pcf, PCF50633_REG_LEDDIM, pdata->ramp_time); - /* Should be different from bl_props.brightness, so we do not exit - * update_status early the first time it's called */ + /* + * Should be different from bl_props.brightness, so we do not exit + * update_status early the first time it's called + */ pcf_bl->brightness = pcf_bl->bl->props.brightness + 1; backlight_update_status(pcf_bl->bl); diff --git a/drivers/video/backlight/platform_lcd.c b/drivers/video/backlight/platform_lcd.c index 894bfc5ce422..17a6b83f97af 100644 --- a/drivers/video/backlight/platform_lcd.c +++ b/drivers/video/backlight/platform_lcd.c @@ -27,7 +27,7 @@ struct platform_lcd { struct plat_lcd_data *pdata; unsigned int power; - unsigned int suspended : 1; + unsigned int suspended:1; }; static inline struct platform_lcd *to_our_lcd(struct lcd_device *lcd) diff --git a/drivers/video/backlight/s6e63m0.c b/drivers/video/backlight/s6e63m0.c index 484e10dd1a8e..3e1c1135f6df 100644 --- a/drivers/video/backlight/s6e63m0.c +++ b/drivers/video/backlight/s6e63m0.c @@ -757,7 +757,7 @@ static int s6e63m0_probe(struct spi_device *spi) lcd->spi = spi; lcd->dev = &spi->dev; - lcd->lcd_pd = (struct lcd_platform_data *)spi->dev.platform_data; + lcd->lcd_pd = spi->dev.platform_data; if (!lcd->lcd_pd) { dev_err(&spi->dev, "platform data is NULL.\n"); return -EFAULT; diff --git a/drivers/video/backlight/tdo24m.c b/drivers/video/backlight/tdo24m.c index 146ffb9404d1..ad2325f3d652 100644 --- a/drivers/video/backlight/tdo24m.c +++ b/drivers/video/backlight/tdo24m.c @@ -2,7 +2,7 @@ * tdo24m - SPI-based drivers for Toppoly TDO24M series LCD panels * * Copyright (C) 2008 Marvell International Ltd. - * Eric Miao <eric.miao@marvell.com> + * Eric Miao <eric.miao@marvell.com> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -47,7 +47,7 @@ struct tdo24m { ((x1) << 9) | 0x100 | (x2)) #define CMD_NULL (-1) -static uint32_t lcd_panel_reset[] = { +static const uint32_t lcd_panel_reset[] = { CMD0(0x1), /* reset */ CMD0(0x0), /* nop */ CMD0(0x0), /* nop */ @@ -55,7 +55,7 @@ static uint32_t lcd_panel_reset[] = { CMD_NULL, }; -static uint32_t lcd_panel_on[] = { +static const uint32_t lcd_panel_on[] = { CMD0(0x29), /* Display ON */ CMD2(0xB8, 0xFF, 0xF9), /* Output Control */ CMD0(0x11), /* Sleep out */ @@ -63,7 +63,7 @@ static uint32_t lcd_panel_on[] = { CMD_NULL, }; -static uint32_t lcd_panel_off[] = { +static const uint32_t lcd_panel_off[] = { CMD0(0x28), /* Display OFF */ CMD2(0xB8, 0x80, 0x02), /* Output Control */ CMD0(0x10), /* Sleep in */ @@ -71,7 +71,7 @@ static uint32_t lcd_panel_off[] = { CMD_NULL, }; -static uint32_t lcd_vga_pass_through_tdo24m[] = { +static const uint32_t lcd_vga_pass_through_tdo24m[] = { CMD1(0xB0, 0x16), CMD1(0xBC, 0x80), CMD1(0xE1, 0x00), @@ -80,7 +80,7 @@ static uint32_t lcd_vga_pass_through_tdo24m[] = { CMD_NULL, }; -static uint32_t lcd_qvga_pass_through_tdo24m[] = { +static const uint32_t lcd_qvga_pass_through_tdo24m[] = { CMD1(0xB0, 0x16), CMD1(0xBC, 0x81), CMD1(0xE1, 0x00), @@ -89,8 +89,8 @@ static uint32_t lcd_qvga_pass_through_tdo24m[] = { CMD_NULL, }; -static uint32_t lcd_vga_transfer_tdo24m[] = { - CMD1(0xcf, 0x02), /* Blanking period control (1) */ +static const uint32_t lcd_vga_transfer_tdo24m[] = { + CMD1(0xcf, 0x02), /* Blanking period control (1) */ CMD2(0xd0, 0x08, 0x04), /* Blanking period control (2) */ CMD1(0xd1, 0x01), /* CKV timing control on/off */ CMD2(0xd2, 0x14, 0x00), /* CKV 1,2 timing control */ @@ -102,7 +102,7 @@ static uint32_t lcd_vga_transfer_tdo24m[] = { CMD_NULL, }; -static uint32_t lcd_qvga_transfer[] = { +static const uint32_t lcd_qvga_transfer[] = { CMD1(0xd6, 0x02), /* Blanking period control (1) */ CMD2(0xd7, 0x08, 0x04), /* Blanking period control (2) */ CMD1(0xd8, 0x01), /* CKV timing control on/off */ @@ -115,7 +115,7 @@ static uint32_t lcd_qvga_transfer[] = { CMD_NULL, }; -static uint32_t lcd_vga_pass_through_tdo35s[] = { +static const uint32_t lcd_vga_pass_through_tdo35s[] = { CMD1(0xB0, 0x16), CMD1(0xBC, 0x80), CMD1(0xE1, 0x00), @@ -123,7 +123,7 @@ static uint32_t lcd_vga_pass_through_tdo35s[] = { CMD_NULL, }; -static uint32_t lcd_qvga_pass_through_tdo35s[] = { +static const uint32_t lcd_qvga_pass_through_tdo35s[] = { CMD1(0xB0, 0x16), CMD1(0xBC, 0x81), CMD1(0xE1, 0x00), @@ -131,8 +131,8 @@ static uint32_t lcd_qvga_pass_through_tdo35s[] = { CMD_NULL, }; -static uint32_t lcd_vga_transfer_tdo35s[] = { - CMD1(0xcf, 0x02), /* Blanking period control (1) */ +static const uint32_t lcd_vga_transfer_tdo35s[] = { + CMD1(0xcf, 0x02), /* Blanking period control (1) */ CMD2(0xd0, 0x08, 0x04), /* Blanking period control (2) */ CMD1(0xd1, 0x01), /* CKV timing control on/off */ CMD2(0xd2, 0x00, 0x1e), /* CKV 1,2 timing control */ @@ -144,7 +144,7 @@ static uint32_t lcd_vga_transfer_tdo35s[] = { CMD_NULL, }; -static uint32_t lcd_panel_config[] = { +static const uint32_t lcd_panel_config[] = { CMD2(0xb8, 0xff, 0xf9), /* Output control */ CMD0(0x11), /* sleep out */ CMD1(0xba, 0x01), /* Display mode (1) */ @@ -175,10 +175,11 @@ static uint32_t lcd_panel_config[] = { CMD_NULL, }; -static int tdo24m_writes(struct tdo24m *lcd, uint32_t *array) +static int tdo24m_writes(struct tdo24m *lcd, const uint32_t *array) { struct spi_transfer *x = &lcd->xfer; - uint32_t data, *p = array; + const uint32_t *p = array; + uint32_t data; int nparams, err = 0; for (; *p != CMD_NULL; p++) { diff --git a/drivers/video/backlight/tosa_bl.c b/drivers/video/backlight/tosa_bl.c index a0521abdcd8a..588682cc1614 100644 --- a/drivers/video/backlight/tosa_bl.c +++ b/drivers/video/backlight/tosa_bl.c @@ -92,14 +92,12 @@ static int tosa_bl_probe(struct i2c_client *client, data->comadj = sharpsl_param.comadj == -1 ? COMADJ_DEFAULT : sharpsl_param.comadj; - ret = devm_gpio_request(&client->dev, TOSA_GPIO_BL_C20MA, "backlight"); + ret = devm_gpio_request_one(&client->dev, TOSA_GPIO_BL_C20MA, + GPIOF_OUT_INIT_LOW, "backlight"); if (ret) { dev_dbg(&data->bl->dev, "Unable to request gpio!\n"); return ret; } - ret = gpio_direction_output(TOSA_GPIO_BL_C20MA, 0); - if (ret) - return ret; i2c_set_clientdata(client, data); data->i2c = client; @@ -163,7 +161,6 @@ static const struct i2c_device_id tosa_bl_id[] = { { }, }; - static struct i2c_driver tosa_bl_driver = { .driver = { .name = "tosa-bl", diff --git a/drivers/video/backlight/tosa_lcd.c b/drivers/video/backlight/tosa_lcd.c index 86fff88c2e4a..96bae941585a 100644 --- a/drivers/video/backlight/tosa_lcd.c +++ b/drivers/video/backlight/tosa_lcd.c @@ -63,7 +63,7 @@ static int tosa_tg_send(struct spi_device *spi, int adrs, uint8_t data) int tosa_bl_enable(struct spi_device *spi, int enable) { /* bl_enable GP04=1 otherwise GP04=0*/ - return tosa_tg_send(spi, TG_GPODR2, enable? 0x01 : 0x00); + return tosa_tg_send(spi, TG_GPODR2, enable ? 0x01 : 0x00); } EXPORT_SYMBOL(tosa_bl_enable); @@ -91,15 +91,17 @@ static void tosa_lcd_tg_on(struct tosa_lcd_data *data) tosa_tg_send(spi, TG_PNLCTL, value); /* TG LCD pannel power up */ - tosa_tg_send(spi, TG_PINICTL,0x4); + tosa_tg_send(spi, TG_PINICTL, 0x4); mdelay(50); /* TG LCD GVSS */ - tosa_tg_send(spi, TG_PINICTL,0x0); + tosa_tg_send(spi, TG_PINICTL, 0x0); if (!data->i2c) { - /* after the pannel is powered up the first time, we can access the i2c bus */ - /* so probe for the DAC */ + /* + * after the pannel is powered up the first time, + * we can access the i2c bus so probe for the DAC + */ struct i2c_adapter *adap = i2c_get_adapter(0); struct i2c_board_info info = { .type = "tosa-bl", @@ -115,11 +117,11 @@ static void tosa_lcd_tg_off(struct tosa_lcd_data *data) struct spi_device *spi = data->spi; /* TG LCD VHSA off */ - tosa_tg_send(spi, TG_PINICTL,0x4); + tosa_tg_send(spi, TG_PINICTL, 0x4); mdelay(50); /* TG LCD signal off */ - tosa_tg_send(spi, TG_PINICTL,0x6); + tosa_tg_send(spi, TG_PINICTL, 0x6); mdelay(50); /* TG Off */ @@ -193,17 +195,13 @@ static int tosa_lcd_probe(struct spi_device *spi) data->spi = spi; dev_set_drvdata(&spi->dev, data); - ret = devm_gpio_request(&spi->dev, TOSA_GPIO_TG_ON, "tg #pwr"); + ret = devm_gpio_request_one(&spi->dev, TOSA_GPIO_TG_ON, + GPIOF_OUT_INIT_LOW, "tg #pwr"); if (ret < 0) goto err_gpio_tg; mdelay(60); - ret = gpio_direction_output(TOSA_GPIO_TG_ON, 0); - if (ret < 0) - goto err_gpio_tg; - - mdelay(60); tosa_lcd_tg_init(data); tosa_lcd_tg_on(data); diff --git a/drivers/video/backlight/vgg2432a4.c b/drivers/video/backlight/vgg2432a4.c index 712b0acfd339..45e81b4cf8b4 100644 --- a/drivers/video/backlight/vgg2432a4.c +++ b/drivers/video/backlight/vgg2432a4.c @@ -26,7 +26,7 @@ /* Device initialisation sequences */ -static struct ili9320_reg vgg_init1[] = { +static const struct ili9320_reg vgg_init1[] = { { .address = ILI9320_POWER1, .value = ILI9320_POWER1_AP(0) | ILI9320_POWER1_BT(0), @@ -43,7 +43,7 @@ static struct ili9320_reg vgg_init1[] = { }, }; -static struct ili9320_reg vgg_init2[] = { +static const struct ili9320_reg vgg_init2[] = { { .address = ILI9320_POWER1, .value = (ILI9320_POWER1_AP(3) | ILI9320_POWER1_APE | @@ -54,7 +54,7 @@ static struct ili9320_reg vgg_init2[] = { } }; -static struct ili9320_reg vgg_gamma[] = { +static const struct ili9320_reg vgg_gamma[] = { { .address = ILI9320_GAMMA1, .value = 0x0000, @@ -89,7 +89,7 @@ static struct ili9320_reg vgg_gamma[] = { }; -static struct ili9320_reg vgg_init0[] = { +static const struct ili9320_reg vgg_init0[] = { [0] = { /* set direction and scan mode gate */ .address = ILI9320_DRIVER, @@ -217,7 +217,7 @@ static int vgg2432a4_resume(struct spi_device *spi) } #else #define vgg2432a4_suspend NULL -#define vgg2432a4_resume NULL +#define vgg2432a4_resume NULL #endif static struct ili9320_client vgg2432a4_client = { diff --git a/drivers/video/gxt4500.c b/drivers/video/gxt4500.c index 0e9afa41d163..4bdea6e9bd55 100644 --- a/drivers/video/gxt4500.c +++ b/drivers/video/gxt4500.c @@ -1,5 +1,6 @@ /* - * Frame buffer device for IBM GXT4500P and GXT6000P display adaptors + * Frame buffer device for IBM GXT4500P/6500P and GXT4000P/6000P + * display adaptors * * Copyright (C) 2006 Paul Mackerras, IBM Corp. <paulus@samba.org> */ @@ -14,6 +15,8 @@ #include <linux/string.h> #define PCI_DEVICE_ID_IBM_GXT4500P 0x21c +#define PCI_DEVICE_ID_IBM_GXT6500P 0x21b +#define PCI_DEVICE_ID_IBM_GXT4000P 0x16e #define PCI_DEVICE_ID_IBM_GXT6000P 0x170 /* GXT4500P registers */ @@ -173,6 +176,8 @@ static const struct fb_videomode defaultmode __devinitconst = { /* List of supported cards */ enum gxt_cards { GXT4500P, + GXT6500P, + GXT4000P, GXT6000P }; @@ -182,6 +187,8 @@ static const struct cardinfo { const char *cardname; } cardinfo[] = { [GXT4500P] = { .refclk_ps = 9259, .cardname = "IBM GXT4500P" }, + [GXT6500P] = { .refclk_ps = 9259, .cardname = "IBM GXT6500P" }, + [GXT4000P] = { .refclk_ps = 40000, .cardname = "IBM GXT4000P" }, [GXT6000P] = { .refclk_ps = 40000, .cardname = "IBM GXT6000P" }, }; @@ -736,6 +743,10 @@ static void __devexit gxt4500_remove(struct pci_dev *pdev) static const struct pci_device_id gxt4500_pci_tbl[] = { { PCI_DEVICE(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_GXT4500P), .driver_data = GXT4500P }, + { PCI_DEVICE(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_GXT6500P), + .driver_data = GXT6500P }, + { PCI_DEVICE(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_GXT4000P), + .driver_data = GXT4000P }, { PCI_DEVICE(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_GXT6000P), .driver_data = GXT6000P }, { 0 } @@ -768,7 +779,7 @@ static void __exit gxt4500_exit(void) module_exit(gxt4500_exit); MODULE_AUTHOR("Paul Mackerras <paulus@samba.org>"); -MODULE_DESCRIPTION("FBDev driver for IBM GXT4500P/6000P"); +MODULE_DESCRIPTION("FBDev driver for IBM GXT4500P/6500P and GXT4000P/6000P"); MODULE_LICENSE("GPL"); module_param(mode_option, charp, 0); MODULE_PARM_DESC(mode_option, "Specify resolution as \"<xres>x<yres>[-<bpp>][@<refresh>]\""); diff --git a/drivers/virt/fsl_hypervisor.c b/drivers/virt/fsl_hypervisor.c index 4939e0ccc4e5..d294f67d6f84 100644 --- a/drivers/virt/fsl_hypervisor.c +++ b/drivers/virt/fsl_hypervisor.c @@ -796,9 +796,6 @@ static int has_fsl_hypervisor(void) struct device_node *node; int ret; - if (!(mfmsr() & MSR_GS)) - return 0; - node = of_find_node_by_path("/hypervisor"); if (!node) return 0; |