Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:
 "This is a re-do of the net-next pull request for the current merge
  window.  The only difference from the one I made the other day is that
  this has Eliezer's interface renames and the timeout handling changes
  made based upon your feedback, as well as a few bug fixes that have
  trickeled in.

  Highlights:

   1) Low latency device polling, eliminating the cost of interrupt
      handling and context switches.  Allows direct polling of a network
      device from socket operations, such as recvmsg() and poll().

      Currently ixgbe, mlx4, and bnx2x support this feature.

      Full high level description, performance numbers, and design in
      commit 0a4db187a999 ("Merge branch 'll_poll'")

      From Eliezer Tamir.

   2) With the routing cache removed, ip_check_mc_rcu() gets exercised
      more than ever before in the case where we have lots of multicast
      addresses.  Use a hash table instead of a simple linked list, from
      Eric Dumazet.

   3) Add driver for Atheros CQA98xx 802.11ac wireless devices, from
      Bartosz Markowski, Janusz Dziedzic, Kalle Valo, Marek Kwaczynski,
      Marek Puzyniak, Michal Kazior, and Sujith Manoharan.

   4) Support reporting the TUN device persist flag to userspace, from
      Pavel Emelyanov.

   5) Allow controlling network device VF link state using netlink, from
      Rony Efraim.

   6) Support GRE tunneling in openvswitch, from Pravin B Shelar.

   7) Adjust SOCK_MIN_RCVBUF and SOCK_MIN_SNDBUF for modern times, from
      Daniel Borkmann and Eric Dumazet.

   8) Allow controlling of TCP quickack behavior on a per-route basis,
      from Cong Wang.

   9) Several bug fixes and improvements to vxlan from Stephen
      Hemminger, Pravin B Shelar, and Mike Rapoport.  In particular,
      support receiving on multiple UDP ports.

  10) Major cleanups, particular in the area of debugging and cookie
      lifetime handline, to the SCTP protocol code.  From Daniel
      Borkmann.

  11) Allow packets to cross network namespaces when traversing tunnel
      devices.  From Nicolas Dichtel.

  12) Allow monitoring netlink traffic via AF_PACKET sockets, in a
      manner akin to how we monitor real network traffic via ptype_all.
      From Daniel Borkmann.

  13) Several bug fixes and improvements for the new alx device driver,
      from Johannes Berg.

  14) Fix scalability issues in the netem packet scheduler's time queue,
      by using an rbtree.  From Eric Dumazet.

  15) Several bug fixes in TCP loss recovery handling, from Yuchung
      Cheng.

  16) Add support for GSO segmentation of MPLS packets, from Simon
      Horman.

  17) Make network notifiers have a real data type for the opaque
      pointer that's passed into them.  Use this to properly handle
      network device flag changes in arp_netdev_event().  From Jiri
      Pirko and Timo Teräs.

  18) Convert several drivers over to module_pci_driver(), from Peter
      Huewe.

  19) tcp_fixup_rcvbuf() can loop 500 times over loopback, just use a
      O(1) calculation instead.  From Eric Dumazet.

  20) Support setting of explicit tunnel peer addresses in ipv6, just
      like ipv4.  From Nicolas Dichtel.

  21) Protect x86 BPF JIT against spraying attacks, from Eric Dumazet.

  22) Prevent a single high rate flow from overruning an individual cpu
      during RX packet processing via selective flow shedding.  From
      Willem de Bruijn.

  23) Don't use spinlocks in TCP md5 signing fast paths, from Eric
      Dumazet.

  24) Don't just drop GSO packets which are above the TBF scheduler's
      burst limit, chop them up so they are in-bounds instead.  Also
      from Eric Dumazet.

  25) VLAN offloads are missed when configured on top of a bridge, fix
      from Vlad Yasevich.

  26) Support IPV6 in ping sockets.  From Lorenzo Colitti.

  27) Receive flow steering targets should be updated at poll() time
      too, from David Majnemer.

  28) Fix several corner case regressions in PMTU/redirect handling due
      to the routing cache removal, from Timo Teräs.

  29) We have to be mindful of ipv4 mapped ipv6 sockets in
      upd_v6_push_pending_frames().  From Hannes Frederic Sowa.

  30) Fix L2TP sequence number handling bugs, from James Chapman."

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1214 commits)
  drivers/net: caif: fix wrong rtnl_is_locked() usage
  drivers/net: enic: release rtnl_lock on error-path
  vhost-net: fix use-after-free in vhost_net_flush
  net: mv643xx_eth: do not use port number as platform device id
  net: sctp: confirm route during forward progress
  virtio_net: fix race in RX VQ processing
  virtio: support unlocked queue poll
  net/cadence/macb: fix bug/typo in extracting gem_irq_read_clear bit
  Documentation: Fix references to defunct linux-net@vger.kernel.org
  net/fs: change busy poll time accounting
  net: rename low latency sockets functions to busy poll
  bridge: fix some kernel warning in multicast timer
  sfc: Fix memory leak when discarding scattered packets
  sit: fix tunnel update via netlink
  dt:net:stmmac: Add dt specific phy reset callback support.
  dt:net:stmmac: Add support to dwmac version 3.610 and 3.710
  dt:net:stmmac: Allocate platform data only if its NULL.
  net:stmmac: fix memleak in the open method
  ipv6: rt6_check_neigh should successfully verify neigh if no NUD information are available
  net: ipv6: fix wrong ping_v6_sendmsg return value
  ...

1 files changed, 218 insertions, 7 deletions

diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
index 4b9bab18ebd9..ab8a6744d402 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
@@ -15,6 +15,7 @@
 #define QLC_83XX_MINIDUMP_FLASH		0x520000
 #define QLC_83XX_OCM_INDEX			3
 #define QLC_83XX_PCI_INDEX			0
+#define QLC_83XX_DMA_ENGINE_INDEX		8
 
 static const u32 qlcnic_ms_read_data[] = {
 	0x410000A8, 0x410000AC, 0x410000B8, 0x410000BC
@@ -32,6 +33,16 @@ static const u32 qlcnic_ms_read_data[] = {
 
 #define QLCNIC_DUMP_MASK_MAX	0xff
 
+struct qlcnic_pex_dma_descriptor {
+	u32	read_data_size;
+	u32	dma_desc_cmd;
+	u32	src_addr_low;
+	u32	src_addr_high;
+	u32	dma_bus_addr_low;
+	u32	dma_bus_addr_high;
+	u32	rsvd[6];
+} __packed;
+
 struct qlcnic_common_entry_hdr {
 	u32     type;
 	u32     offset;
@@ -90,7 +101,10 @@ struct __ocm {
 } __packed;
 
 struct __mem {
-	u8	rsvd[24];
+	u32	desc_card_addr;
+	u32	dma_desc_cmd;
+	u32	start_dma_cmd;
+	u32	rsvd[3];
 	u32	addr;
 	u32	size;
 } __packed;
@@ -466,12 +480,12 @@ skip_poll:
 	return l2->no_ops * l2->read_addr_num * sizeof(u32);
 }
 
-static u32 qlcnic_read_memory(struct qlcnic_adapter *adapter,
-			      struct qlcnic_dump_entry *entry, __le32 *buffer)
+static u32 qlcnic_read_memory_test_agent(struct qlcnic_adapter *adapter,
+					 struct __mem *mem, __le32 *buffer,
+					 int *ret)
 {
-	u32 addr, data, test, ret = 0;
+	u32 addr, data, test;
 	int i, reg_read;
-	struct __mem *mem = &entry->region.mem;
 
 	reg_read = mem->size;
 	addr = mem->addr;
@@ -480,7 +494,8 @@ static u32 qlcnic_read_memory(struct qlcnic_adapter *adapter,
 		dev_info(&adapter->pdev->dev,
 			 "Unaligned memory addr:0x%x size:0x%x\n",
 			 addr, reg_read);
-		return -EINVAL;
+		*ret = -EINVAL;
+		return 0;
 	}
 
 	mutex_lock(&adapter->ahw->mem_lock);
@@ -499,7 +514,7 @@ static u32 qlcnic_read_memory(struct qlcnic_adapter *adapter,
 			if (printk_ratelimit()) {
 				dev_err(&adapter->pdev->dev,
 					"failed to read through agent\n");
-				ret = -EINVAL;
+				*ret = -EIO;
 				goto out;
 			}
 		}
@@ -516,6 +531,181 @@ out:
 	return mem->size;
 }
 
+/* DMA register base address */
+#define QLC_DMA_REG_BASE_ADDR(dma_no)	(0x77320000 + (dma_no * 0x10000))
+
+/* DMA register offsets w.r.t base address */
+#define QLC_DMA_CMD_BUFF_ADDR_LOW	0
+#define QLC_DMA_CMD_BUFF_ADDR_HI	4
+#define QLC_DMA_CMD_STATUS_CTRL		8
+
+#define QLC_PEX_DMA_READ_SIZE		(PAGE_SIZE * 16)
+
+static int qlcnic_start_pex_dma(struct qlcnic_adapter *adapter,
+				struct __mem *mem)
+{
+	struct qlcnic_dump_template_hdr *tmpl_hdr;
+	struct device *dev = &adapter->pdev->dev;
+	u32 dma_no, dma_base_addr, temp_addr;
+	int i, ret, dma_sts;
+
+	tmpl_hdr = adapter->ahw->fw_dump.tmpl_hdr;
+	dma_no = tmpl_hdr->saved_state[QLC_83XX_DMA_ENGINE_INDEX];
+	dma_base_addr = QLC_DMA_REG_BASE_ADDR(dma_no);
+
+	temp_addr = dma_base_addr + QLC_DMA_CMD_BUFF_ADDR_LOW;
+	ret = qlcnic_83xx_wrt_reg_indirect(adapter, temp_addr,
+					   mem->desc_card_addr);
+	if (ret)
+		return ret;
+
+	temp_addr = dma_base_addr + QLC_DMA_CMD_BUFF_ADDR_HI;
+	ret = qlcnic_83xx_wrt_reg_indirect(adapter, temp_addr, 0);
+	if (ret)
+		return ret;
+
+	temp_addr = dma_base_addr + QLC_DMA_CMD_STATUS_CTRL;
+	ret = qlcnic_83xx_wrt_reg_indirect(adapter, temp_addr,
+					   mem->start_dma_cmd);
+	if (ret)
+		return ret;
+
+	/* Wait for DMA to complete */
+	temp_addr = dma_base_addr + QLC_DMA_CMD_STATUS_CTRL;
+	for (i = 0; i < 400; i++) {
+		dma_sts = qlcnic_ind_rd(adapter, temp_addr);
+
+		if (dma_sts & BIT_1)
+			usleep_range(250, 500);
+		else
+			break;
+	}
+
+	if (i >= 400) {
+		dev_info(dev, "PEX DMA operation timed out");
+		ret = -EIO;
+	}
+
+	return ret;
+}
+
+static u32 qlcnic_read_memory_pexdma(struct qlcnic_adapter *adapter,
+				     struct __mem *mem,
+				     __le32 *buffer, int *ret)
+{
+	struct qlcnic_fw_dump *fw_dump = &adapter->ahw->fw_dump;
+	u32 temp, dma_base_addr, size = 0, read_size = 0;
+	struct qlcnic_pex_dma_descriptor *dma_descr;
+	struct qlcnic_dump_template_hdr *tmpl_hdr;
+	struct device *dev = &adapter->pdev->dev;
+	dma_addr_t dma_phys_addr;
+	void *dma_buffer;
+
+	tmpl_hdr = fw_dump->tmpl_hdr;
+
+	/* Check if DMA engine is available */
+	temp = tmpl_hdr->saved_state[QLC_83XX_DMA_ENGINE_INDEX];
+	dma_base_addr = QLC_DMA_REG_BASE_ADDR(temp);
+	temp = qlcnic_ind_rd(adapter,
+			     dma_base_addr + QLC_DMA_CMD_STATUS_CTRL);
+
+	if (!(temp & BIT_31)) {
+		dev_info(dev, "%s: DMA engine is not available\n", __func__);
+		*ret = -EIO;
+		return 0;
+	}
+
+	/* Create DMA descriptor */
+	dma_descr = kzalloc(sizeof(struct qlcnic_pex_dma_descriptor),
+			    GFP_KERNEL);
+	if (!dma_descr) {
+		*ret = -ENOMEM;
+		return 0;
+	}
+
+	/* dma_desc_cmd  0:15  = 0
+	 * dma_desc_cmd 16:19  = mem->dma_desc_cmd 0:3
+	 * dma_desc_cmd 20:23  = pci function number
+	 * dma_desc_cmd 24:31  = mem->dma_desc_cmd 8:15
+	 */
+	dma_phys_addr = fw_dump->phys_addr;
+	dma_buffer = fw_dump->dma_buffer;
+	temp = 0;
+	temp = mem->dma_desc_cmd & 0xff0f;
+	temp |= (adapter->ahw->pci_func & 0xf) << 4;
+	dma_descr->dma_desc_cmd = (temp << 16) & 0xffff0000;
+	dma_descr->dma_bus_addr_low = LSD(dma_phys_addr);
+	dma_descr->dma_bus_addr_high = MSD(dma_phys_addr);
+	dma_descr->src_addr_high = 0;
+
+	/* Collect memory dump using multiple DMA operations if required */
+	while (read_size < mem->size) {
+		if (mem->size - read_size >= QLC_PEX_DMA_READ_SIZE)
+			size = QLC_PEX_DMA_READ_SIZE;
+		else
+			size = mem->size - read_size;
+
+		dma_descr->src_addr_low = mem->addr + read_size;
+		dma_descr->read_data_size = size;
+
+		/* Write DMA descriptor to MS memory*/
+		temp = sizeof(struct qlcnic_pex_dma_descriptor) / 16;
+		*ret = qlcnic_83xx_ms_mem_write128(adapter, mem->desc_card_addr,
+						   (u32 *)dma_descr, temp);
+		if (*ret) {
+			dev_info(dev, "Failed to write DMA descriptor to MS memory at address 0x%x\n",
+				 mem->desc_card_addr);
+			goto free_dma_descr;
+		}
+
+		*ret = qlcnic_start_pex_dma(adapter, mem);
+		if (*ret) {
+			dev_info(dev, "Failed to start PEX DMA operation\n");
+			goto free_dma_descr;
+		}
+
+		memcpy(buffer, dma_buffer, size);
+		buffer += size / 4;
+		read_size += size;
+	}
+
+free_dma_descr:
+	kfree(dma_descr);
+
+	return read_size;
+}
+
+static u32 qlcnic_read_memory(struct qlcnic_adapter *adapter,
+			      struct qlcnic_dump_entry *entry, __le32 *buffer)
+{
+	struct qlcnic_fw_dump *fw_dump = &adapter->ahw->fw_dump;
+	struct device *dev = &adapter->pdev->dev;
+	struct __mem *mem = &entry->region.mem;
+	u32 data_size;
+	int ret = 0;
+
+	if (fw_dump->use_pex_dma) {
+		data_size = qlcnic_read_memory_pexdma(adapter, mem, buffer,
+						      &ret);
+		if (ret)
+			dev_info(dev,
+				 "Failed to read memory dump using PEX DMA: mask[0x%x]\n",
+				 entry->hdr.mask);
+		else
+			return data_size;
+	}
+
+	data_size = qlcnic_read_memory_test_agent(adapter, mem, buffer, &ret);
+	if (ret) {
+		dev_info(dev,
+			 "Failed to read memory dump using test agent method: mask[0x%x]\n",
+			 entry->hdr.mask);
+		return 0;
+	} else {
+		return data_size;
+	}
+}
+
 static u32 qlcnic_dump_nop(struct qlcnic_adapter *adapter,
 			   struct qlcnic_dump_entry *entry, __le32 *buffer)
 {
@@ -893,6 +1083,12 @@ flash_temp:
 
 	tmpl_hdr = ahw->fw_dump.tmpl_hdr;
 	tmpl_hdr->drv_cap_mask = QLCNIC_DUMP_MASK_DEF;
+
+	if ((tmpl_hdr->version & 0xffffff) >= 0x20001)
+		ahw->fw_dump.use_pex_dma = true;
+	else
+		ahw->fw_dump.use_pex_dma = false;
+
 	ahw->fw_dump.enable = 1;
 
 	return 0;
@@ -910,7 +1106,9 @@ int qlcnic_dump_fw(struct qlcnic_adapter *adapter)
 	struct qlcnic_fw_dump *fw_dump = &adapter->ahw->fw_dump;
 	struct qlcnic_dump_template_hdr *tmpl_hdr = fw_dump->tmpl_hdr;
 	static const struct qlcnic_dump_operations *fw_dump_ops;
+	struct device *dev = &adapter->pdev->dev;
 	struct qlcnic_hardware_context *ahw;
+	void *temp_buffer;
 
 	ahw = adapter->ahw;
 
@@ -944,6 +1142,16 @@ int qlcnic_dump_fw(struct qlcnic_adapter *adapter)
 	tmpl_hdr->sys_info[0] = QLCNIC_DRIVER_VERSION;
 	tmpl_hdr->sys_info[1] = adapter->fw_version;
 
+	if (fw_dump->use_pex_dma) {
+		temp_buffer = dma_alloc_coherent(dev, QLC_PEX_DMA_READ_SIZE,
+						 &fw_dump->phys_addr,
+						 GFP_KERNEL);
+		if (!temp_buffer)
+			fw_dump->use_pex_dma = false;
+		else
+			fw_dump->dma_buffer = temp_buffer;
+	}
+
 	if (qlcnic_82xx_check(adapter)) {
 		ops_cnt = ARRAY_SIZE(qlcnic_fw_dump_ops);
 		fw_dump_ops = qlcnic_fw_dump_ops;
@@ -1002,6 +1210,9 @@ int qlcnic_dump_fw(struct qlcnic_adapter *adapter)
 		return 0;
 	}
 error:
+	if (fw_dump->use_pex_dma)
+		dma_free_coherent(dev, QLC_PEX_DMA_READ_SIZE,
+				  fw_dump->dma_buffer, fw_dump->phys_addr);
 	vfree(fw_dump->data);
 	return -EINVAL;
 }