[git patch review 2/3] [IB] srp: don't post receive if no send buf available

[Roland Dreier <rolandd@cisco.com>]

Have __srp_get_tx_iu() fail if the target port's request limit will
not allow the initiator to post a send.  This avoids continuing on and
posting a receive, and then failing to post a corresponding send.  If
that happens, then the initiator will end up with an extra receive
posted, and if this happens to much, the receive queue will overflow.

Signed-off-by: Roland Dreier <rolandd@cisco.com>

---

 drivers/infiniband/ulp/srp/ib_srp.c |   15 +++++++++------
 1 files changed, 9 insertions(+), 6 deletions(-)

applies-to: a5f8266c59f39f0a1f3dc3d71a00da7276ac1a80
47f2bce9021b4974ed33b072ebb8348c8145c946
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index a364530..ee9fe22 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -802,13 +802,21 @@ static int srp_post_recv(struct srp_targ
 
 /*
  * Must be called with target->scsi_host->host_lock held to protect
- * req_lim and tx_head.
+ * req_lim and tx_head.  Lock cannot be dropped between call here and
+ * call to __srp_post_send().
  */
 static struct srp_iu *__srp_get_tx_iu(struct srp_target_port *target)
 {
 	if (target->tx_head - target->tx_tail >= SRP_SQ_SIZE)
 		return NULL;
 
+	if (unlikely(target->req_lim < 1)) {
+		if (printk_ratelimit())
+			printk(KERN_DEBUG PFX "Target has req_lim %d\n",
+			       target->req_lim);
+		return NULL;
+	}
+
 	return target->tx_ring[target->tx_head & SRP_SQ_SIZE];
 }
 
@@ -823,11 +831,6 @@ static int __srp_post_send(struct srp_ta
 	struct ib_send_wr wr, *bad_wr;
 	int ret = 0;
 
-	if (target->req_lim < 1) {
-		printk(KERN_ERR PFX "Target has req_lim %d\n", target->req_lim);
-		return -EAGAIN;
-	}
-
 	list.addr   = iu->dma;
 	list.length = len;
 	list.lkey   = target->srp_host->mr->lkey;
---
0.99.9g
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/