RDMA/cxgb4: Don't hang threads forever waiting on WR replies

In c4iw_wait_for_reply(), if a FW6_MSG WR reply is not received after
C4IW_WR_TO seconds, fail the WR operation and mark the device as fatally
dead.  Further, if the device is marked fatally dead, then fail the WR
wait immediately.

Also change the timeout to 60 seconds.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>

1 files changed, 14 insertions, 15 deletions

diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index b5678ac97393..d87e1650f643 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -196,7 +196,7 @@ static inline int c4iw_num_stags(struct c4iw_rdev *rdev)
         return (int)(rdev->lldi.vr->stag.size >> 5);
 }
 
-#define C4IW_WR_TO (30*HZ)
+#define C4IW_WR_TO (60*HZ)
 
 struct c4iw_wr_wait {
         struct completion completion;
@@ -220,22 +220,21 @@ static inline int c4iw_wait_for_reply(struct c4iw_rdev *rdev,
                                  u32 hwtid, u32 qpid,
                                  const char *func)
 {
-        unsigned to = C4IW_WR_TO;
         int ret;
 
-        do {
-                ret = wait_for_completion_timeout(&wr_waitp->completion, to);
-                if (!ret) {
-                        printk(KERN_ERR MOD "%s - Device %s not responding - "
-                               "tid %u qpid %u\n", func,
-                               pci_name(rdev->lldi.pdev), hwtid, qpid);
-                        if (c4iw_fatal_error(rdev)) {
-                                wr_waitp->ret = -EIO;
-                                break;
-                        }
-                        to = to << 2;
-                }
-        } while (!ret);
+        if (c4iw_fatal_error(rdev)) {
+                wr_waitp->ret = -EIO;
+                goto out;
+        }
+
+        ret = wait_for_completion_timeout(&wr_waitp->completion, C4IW_WR_TO);
+        if (!ret) {
+                PDBG("%s - Device %s not responding (disabling device) - tid %u qpid %u\n",
+                     func, pci_name(rdev->lldi.pdev), hwtid, qpid);
+                rdev->flags |= T4_FATAL_ERROR;
+                wr_waitp->ret = -EIO;
+        }
+out:
         if (wr_waitp->ret)
                 PDBG("%s: FW reply %d tid %u qpid %u\n",
                      pci_name(rdev->lldi.pdev), wr_waitp->ret, hwtid, qpid);