Sending any data over a queue pair associated with a closed
connection is wrong. The HCA will send the data anyway and such
data may be sent to another system to a queue pair that is in use.
The data will get processed and a response will be sent back. That
can result in ib_srp complaining about "Null scmnd for RSP ..."
followed by a kernel oops.
Signed-off-by: Bart Van Assche <redacted>
Cc: David Dillow <redacted>
Cc: Roland Dreier <redacted>
---
drivers/infiniband/ulp/srp/ib_srp.c | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index ff3e15a..b7a5e92 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -1338,6 +1338,7 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
if (target->state == SRP_TARGET_DEAD ||
target->state == SRP_TARGET_REMOVED ||
+ !target->connected ||
target->qp_in_error) {
scmnd->result = DID_BAD_TARGET << 16;
scmnd->scsi_done(scmnd);@@ -1690,6 +1691,7 @@ static int srp_send_tsk_mgmt(struct srp_target_port *target,
if (target->state == SRP_TARGET_DEAD ||
target->state == SRP_TARGET_REMOVED ||
+ !target->connected ||
target->qp_in_error)
return -1;
--
1.7.7
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html