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

Pull networking updates from David Miller:
 "Highlights:

   1) Steady transitioning of the BPF instructure to a generic spot so
      all kernel subsystems can make use of it, from Alexei Starovoitov.

   2) SFC driver supports busy polling, from Alexandre Rames.

   3) Take advantage of hash table in UDP multicast delivery, from David
      Held.

   4) Lighten locking, in particular by getting rid of the LRU lists, in
      inet frag handling.  From Florian Westphal.

   5) Add support for various RFC6458 control messages in SCTP, from
      Geir Ola Vaagland.

   6) Allow to filter bridge forwarding database dumps by device, from
      Jamal Hadi Salim.

   7) virtio-net also now supports busy polling, from Jason Wang.

   8) Some low level optimization tweaks in pktgen from Jesper Dangaard
      Brouer.

   9) Add support for ipv6 address generation modes, so that userland
      can have some input into the process.  From Jiri Pirko.

  10) Consolidate common TCP connection request code in ipv4 and ipv6,
      from Octavian Purdila.

  11) New ARP packet logger in netfilter, from Pablo Neira Ayuso.

  12) Generic resizable RCU hash table, with intial users in netlink and
      nftables.  From Thomas Graf.

  13) Maintain a name assignment type so that userspace can see where a
      network device name came from (enumerated by kernel, assigned
      explicitly by userspace, etc.) From Tom Gundersen.

  14) Automatic flow label generation on transmit in ipv6, from Tom
      Herbert.

  15) New packet timestamping facilities from Willem de Bruijn, meant to
      assist in measuring latencies going into/out-of the packet
      scheduler, latency from TCP data transmission to ACK, etc"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1536 commits)
  cxgb4 : Disable recursive mailbox commands when enabling vi
  net: reduce USB network driver config options.
  tg3: Modify tg3_tso_bug() to handle multiple TX rings
  amd-xgbe: Perform phy connect/disconnect at dev open/stop
  amd-xgbe: Use dma_set_mask_and_coherent to set DMA mask
  net: sun4i-emac: fix memory leak on bad packet
  sctp: fix possible seqlock seadlock in sctp_packet_transmit()
  Revert "net: phy: Set the driver when registering an MDIO bus device"
  cxgb4vf: Turn off SGE RX/TX Callback Timers and interrupts in PCI shutdown routine
  team: Simplify return path of team_newlink
  bridge: Update outdated comment on promiscuous mode
  net-timestamp: ACK timestamp for bytestreams
  net-timestamp: TCP timestamping
  net-timestamp: SCHED timestamp on entering packet scheduler
  net-timestamp: add key to disambiguate concurrent datagrams
  net-timestamp: move timestamp flags out of sk_flags
  net-timestamp: extend SCM_TIMESTAMPING ancillary data struct
  cxgb4i : Move stray CPL definitions to cxgb4 driver
  tcp: reduce spurious retransmits due to transient SACK reneging
  qlcnic: Initialize dcbnl_ops before register_netdev
  ...

1 files changed, 188 insertions, 4 deletions

diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index 7db82b24302b..f25df5276c22 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -33,6 +33,7 @@
 #include <linux/moduleparam.h>
 #include <linux/debugfs.h>
 #include <linux/vmalloc.h>
+#include <linux/math64.h>
 
 #include <rdma/ib_verbs.h>
 
@@ -55,6 +56,15 @@ module_param(allow_db_coalescing_on_t5, int, 0644);
 MODULE_PARM_DESC(allow_db_coalescing_on_t5,
 		 "Allow DB Coalescing on T5 (default = 0)");
 
+int c4iw_wr_log = 0;
+module_param(c4iw_wr_log, int, 0444);
+MODULE_PARM_DESC(c4iw_wr_log, "Enables logging of work request timing data.");
+
+int c4iw_wr_log_size_order = 12;
+module_param(c4iw_wr_log_size_order, int, 0444);
+MODULE_PARM_DESC(c4iw_wr_log_size_order,
+		 "Number of entries (log2) in the work request timing log.");
+
 struct uld_ctx {
 	struct list_head entry;
 	struct cxgb4_lld_info lldi;
@@ -103,6 +113,117 @@ static ssize_t debugfs_read(struct file *file, char __user *buf, size_t count,
 	return simple_read_from_buffer(buf, count, ppos, d->buf, d->pos);
 }
 
+void c4iw_log_wr_stats(struct t4_wq *wq, struct t4_cqe *cqe)
+{
+	struct wr_log_entry le;
+	int idx;
+
+	if (!wq->rdev->wr_log)
+		return;
+
+	idx = (atomic_inc_return(&wq->rdev->wr_log_idx) - 1) &
+		(wq->rdev->wr_log_size - 1);
+	le.poll_sge_ts = cxgb4_read_sge_timestamp(wq->rdev->lldi.ports[0]);
+	getnstimeofday(&le.poll_host_ts);
+	le.valid = 1;
+	le.cqe_sge_ts = CQE_TS(cqe);
+	if (SQ_TYPE(cqe)) {
+		le.qid = wq->sq.qid;
+		le.opcode = CQE_OPCODE(cqe);
+		le.post_host_ts = wq->sq.sw_sq[wq->sq.cidx].host_ts;
+		le.post_sge_ts = wq->sq.sw_sq[wq->sq.cidx].sge_ts;
+		le.wr_id = CQE_WRID_SQ_IDX(cqe);
+	} else {
+		le.qid = wq->rq.qid;
+		le.opcode = FW_RI_RECEIVE;
+		le.post_host_ts = wq->rq.sw_rq[wq->rq.cidx].host_ts;
+		le.post_sge_ts = wq->rq.sw_rq[wq->rq.cidx].sge_ts;
+		le.wr_id = CQE_WRID_MSN(cqe);
+	}
+	wq->rdev->wr_log[idx] = le;
+}
+
+static int wr_log_show(struct seq_file *seq, void *v)
+{
+	struct c4iw_dev *dev = seq->private;
+	struct timespec prev_ts = {0, 0};
+	struct wr_log_entry *lep;
+	int prev_ts_set = 0;
+	int idx, end;
+
+#define ts2ns(ts) div64_ul((ts) * dev->rdev.lldi.cclk_ps, 1000)
+
+	idx = atomic_read(&dev->rdev.wr_log_idx) &
+		(dev->rdev.wr_log_size - 1);
+	end = idx - 1;
+	if (end < 0)
+		end = dev->rdev.wr_log_size - 1;
+	lep = &dev->rdev.wr_log[idx];
+	while (idx != end) {
+		if (lep->valid) {
+			if (!prev_ts_set) {
+				prev_ts_set = 1;
+				prev_ts = lep->poll_host_ts;
+			}
+			seq_printf(seq, "%04u: sec %lu nsec %lu qid %u opcode "
+				   "%u %s 0x%x host_wr_delta sec %lu nsec %lu "
+				   "post_sge_ts 0x%llx cqe_sge_ts 0x%llx "
+				   "poll_sge_ts 0x%llx post_poll_delta_ns %llu "
+				   "cqe_poll_delta_ns %llu\n",
+				   idx,
+				   timespec_sub(lep->poll_host_ts,
+						prev_ts).tv_sec,
+				   timespec_sub(lep->poll_host_ts,
+						prev_ts).tv_nsec,
+				   lep->qid, lep->opcode,
+				   lep->opcode == FW_RI_RECEIVE ?
+							"msn" : "wrid",
+				   lep->wr_id,
+				   timespec_sub(lep->poll_host_ts,
+						lep->post_host_ts).tv_sec,
+				   timespec_sub(lep->poll_host_ts,
+						lep->post_host_ts).tv_nsec,
+				   lep->post_sge_ts, lep->cqe_sge_ts,
+				   lep->poll_sge_ts,
+				   ts2ns(lep->poll_sge_ts - lep->post_sge_ts),
+				   ts2ns(lep->poll_sge_ts - lep->cqe_sge_ts));
+			prev_ts = lep->poll_host_ts;
+		}
+		idx++;
+		if (idx > (dev->rdev.wr_log_size - 1))
+			idx = 0;
+		lep = &dev->rdev.wr_log[idx];
+	}
+#undef ts2ns
+	return 0;
+}
+
+static int wr_log_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, wr_log_show, inode->i_private);
+}
+
+static ssize_t wr_log_clear(struct file *file, const char __user *buf,
+			    size_t count, loff_t *pos)
+{
+	struct c4iw_dev *dev = ((struct seq_file *)file->private_data)->private;
+	int i;
+
+	if (dev->rdev.wr_log)
+		for (i = 0; i < dev->rdev.wr_log_size; i++)
+			dev->rdev.wr_log[i].valid = 0;
+	return count;
+}
+
+static const struct file_operations wr_log_debugfs_fops = {
+	.owner   = THIS_MODULE,
+	.open    = wr_log_open,
+	.release = single_release,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.write   = wr_log_clear,
+};
+
 static int dump_qp(int id, void *p, void *data)
 {
 	struct c4iw_qp *qp = p;
@@ -241,12 +362,32 @@ static int dump_stag(int id, void *p, void *data)
 	struct c4iw_debugfs_data *stagd = data;
 	int space;
 	int cc;
+	struct fw_ri_tpte tpte;
+	int ret;
 
 	space = stagd->bufsize - stagd->pos - 1;
 	if (space == 0)
 		return 1;
 
-	cc = snprintf(stagd->buf + stagd->pos, space, "0x%x\n", id<<8);
+	ret = cxgb4_read_tpte(stagd->devp->rdev.lldi.ports[0], (u32)id<<8,
+			      (__be32 *)&tpte);
+	if (ret) {
+		dev_err(&stagd->devp->rdev.lldi.pdev->dev,
+			"%s cxgb4_read_tpte err %d\n", __func__, ret);
+		return ret;
+	}
+	cc = snprintf(stagd->buf + stagd->pos, space,
+		      "stag: idx 0x%x valid %d key 0x%x state %d pdid %d "
+		      "perm 0x%x ps %d len 0x%llx va 0x%llx\n",
+		      (u32)id<<8,
+		      G_FW_RI_TPTE_VALID(ntohl(tpte.valid_to_pdid)),
+		      G_FW_RI_TPTE_STAGKEY(ntohl(tpte.valid_to_pdid)),
+		      G_FW_RI_TPTE_STAGSTATE(ntohl(tpte.valid_to_pdid)),
+		      G_FW_RI_TPTE_PDID(ntohl(tpte.valid_to_pdid)),
+		      G_FW_RI_TPTE_PERM(ntohl(tpte.locread_to_qpid)),
+		      G_FW_RI_TPTE_PS(ntohl(tpte.locread_to_qpid)),
+		      ((u64)ntohl(tpte.len_hi) << 32) | ntohl(tpte.len_lo),
+		      ((u64)ntohl(tpte.va_hi) << 32) | ntohl(tpte.va_lo_fbo));
 	if (cc < space)
 		stagd->pos += cc;
 	return 0;
@@ -259,7 +400,7 @@ static int stag_release(struct inode *inode, struct file *file)
 		printk(KERN_INFO "%s null stagd?\n", __func__);
 		return 0;
 	}
-	kfree(stagd->buf);
+	vfree(stagd->buf);
 	kfree(stagd);
 	return 0;
 }
@@ -282,8 +423,8 @@ static int stag_open(struct inode *inode, struct file *file)
 	idr_for_each(&stagd->devp->mmidr, count_idrs, &count);
 	spin_unlock_irq(&stagd->devp->lock);
 
-	stagd->bufsize = count * sizeof("0x12345678\n");
-	stagd->buf = kmalloc(stagd->bufsize, GFP_KERNEL);
+	stagd->bufsize = count * 256;
+	stagd->buf = vmalloc(stagd->bufsize);
 	if (!stagd->buf) {
 		ret = -ENOMEM;
 		goto err1;
@@ -348,6 +489,7 @@ static int stats_show(struct seq_file *seq, void *v)
 		   dev->rdev.stats.act_ofld_conn_fails);
 	seq_printf(seq, "PAS_OFLD_CONN_FAILS: %10llu\n",
 		   dev->rdev.stats.pas_ofld_conn_fails);
+	seq_printf(seq, "AVAILABLE IRD: %10u\n", dev->avail_ird);
 	return 0;
 }
 
@@ -583,6 +725,12 @@ static int setup_debugfs(struct c4iw_dev *devp)
 	if (de && de->d_inode)
 		de->d_inode->i_size = 4096;
 
+	if (c4iw_wr_log) {
+		de = debugfs_create_file("wr_log", S_IWUSR, devp->debugfs_root,
+					 (void *)devp, &wr_log_debugfs_fops);
+		if (de && de->d_inode)
+			de->d_inode->i_size = 4096;
+	}
 	return 0;
 }
 
@@ -696,7 +844,20 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev)
 		pr_err(MOD "error allocating status page\n");
 		goto err4;
 	}
+
+	if (c4iw_wr_log) {
+		rdev->wr_log = kzalloc((1 << c4iw_wr_log_size_order) *
+				       sizeof(*rdev->wr_log), GFP_KERNEL);
+		if (rdev->wr_log) {
+			rdev->wr_log_size = 1 << c4iw_wr_log_size_order;
+			atomic_set(&rdev->wr_log_idx, 0);
+		} else {
+			pr_err(MOD "error allocating wr_log. Logging disabled\n");
+		}
+	}
+
 	rdev->status_page->db_off = 0;
+
 	return 0;
 err4:
 	c4iw_rqtpool_destroy(rdev);
@@ -710,6 +871,7 @@ err1:
 
 static void c4iw_rdev_close(struct c4iw_rdev *rdev)
 {
+	kfree(rdev->wr_log);
 	free_page((unsigned long)rdev->status_page);
 	c4iw_pblpool_destroy(rdev);
 	c4iw_rqtpool_destroy(rdev);
@@ -768,6 +930,27 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
 	}
 	devp->rdev.lldi = *infop;
 
+	/* init various hw-queue params based on lld info */
+	PDBG("%s: Ing. padding boundary is %d, egrsstatuspagesize = %d\n",
+	     __func__, devp->rdev.lldi.sge_ingpadboundary,
+	     devp->rdev.lldi.sge_egrstatuspagesize);
+
+	devp->rdev.hw_queue.t4_eq_status_entries =
+		devp->rdev.lldi.sge_ingpadboundary > 64 ? 2 : 1;
+	devp->rdev.hw_queue.t4_max_eq_size = 65520;
+	devp->rdev.hw_queue.t4_max_iq_size = 65520;
+	devp->rdev.hw_queue.t4_max_rq_size = 8192 -
+		devp->rdev.hw_queue.t4_eq_status_entries - 1;
+	devp->rdev.hw_queue.t4_max_sq_size =
+		devp->rdev.hw_queue.t4_max_eq_size -
+		devp->rdev.hw_queue.t4_eq_status_entries - 1;
+	devp->rdev.hw_queue.t4_max_qp_depth =
+		devp->rdev.hw_queue.t4_max_rq_size;
+	devp->rdev.hw_queue.t4_max_cq_depth =
+		devp->rdev.hw_queue.t4_max_iq_size - 2;
+	devp->rdev.hw_queue.t4_stat_len =
+		devp->rdev.lldi.sge_egrstatuspagesize;
+
 	/*
 	 * For T5 devices, we map all of BAR2 with WC.
 	 * For T4 devices with onchip qp mem, we map only that part
@@ -818,6 +1001,7 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
 	mutex_init(&devp->rdev.stats.lock);
 	mutex_init(&devp->db_mutex);
 	INIT_LIST_HEAD(&devp->db_fc_list);
+	devp->avail_ird = devp->rdev.lldi.max_ird_adapter;
 
 	if (c4iw_debugfs_root) {
 		devp->debugfs_root = debugfs_create_dir(