4.0-stable review patch. If anyone has any objections, please let me know.
------------------
From: Bart Van Assche <bart.vanassche@xxxxxxxxxxx>
commit 8de9fe3a1d4ac8c3e4953fa4b7d81f863f5196ad upstream.
Avoid that receiving a DREQ while RDMA channels are being
established causes target->qp_in_error to be reset.
Signed-off-by: Bart Van Assche <bart.vanassche@xxxxxxxxxxx>
Cc: Sagi Grimberg <sagig@xxxxxxxxxxxx>
Cc: Sebastian Parschauer <sebastian.riemer@xxxxxxxxxxxxxxxx>
Signed-off-by: Doug Ledford <dledford@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/infiniband/ulp/srp/ib_srp.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -992,8 +992,6 @@ static int srp_connect_ch(struct srp_rdm
WARN_ON_ONCE(!multich && target->connected);
- target->qp_in_error = false;
-
ret = srp_lookup_path(ch);
if (ret)
return ret;
@@ -1242,6 +1240,9 @@ static int srp_rport_reconnect(struct sr
for (j = 0; j < target->queue_size; ++j)
list_add(&ch->tx_ring[j]->list, &ch->free_tx);
}
+
+ target->qp_in_error = false;
+
for (i = 0; i < target->ch_count; i++) {
ch = &target->ch[i];
if (ret || !ch->target) {
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html