[lustre-devel] [PATCH 20/22] lnet: lnd: gracefully handle unexpected events
James Simmons
jsimmons at infradead.org
Tue Jun 2 17:59:59 PDT 2020
From: Amir Shehata <ashehata at whamcloud.com>
When a tx completes kiblnd_tx_complete() callback is invoked.
We ensure:
LASSERT (tx->tx_sending > 0);
However this assert is being triggered in some rare scenarios.
The reason tx_sending would be 0 at this point is because:
1. ib_post_send() failed but OFED stack is still sending
a tx complete event.
2. We're getting two different events for the same tx
Instead of asserting, ignore that tx_complete event and print
the tx pointer and its status.
WC-bug-id: https://jira.whamcloud.com/browse/LU-13553
Lustre-commit: 60f9f539e686f ("LU-13553 lnd: gracefully handle unexpected events")
Signed-off-by: Amir Shehata <ashehata at whamcloud.com>
Reviewed-on: https://review.whamcloud.com/38669
Reviewed-by: Andreas Dilger <adilger at whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov at whamcloud.com>
Reviewed-by: Oleg Drokin <green at whamcloud.com>
Signed-off-by: James Simmons <jsimmons at infradead.org>
---
net/lnet/klnds/o2iblnd/o2iblnd_cb.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/net/lnet/klnds/o2iblnd/o2iblnd_cb.c b/net/lnet/klnds/o2iblnd/o2iblnd_cb.c
index 09a46d6..40e196d 100644
--- a/net/lnet/klnds/o2iblnd/o2iblnd_cb.c
+++ b/net/lnet/klnds/o2iblnd/o2iblnd_cb.c
@@ -969,7 +969,11 @@ static int kiblnd_map_tx(struct lnet_ni *ni, struct kib_tx *tx,
struct kib_conn *conn = tx->tx_conn;
int idle;
- LASSERT(tx->tx_sending > 0);
+ if (tx->tx_sending <= 0) {
+ CERROR("Received an event on a freed tx: %p status %d\n",
+ tx, tx->tx_status);
+ return;
+ }
if (failed) {
if (conn->ibc_state == IBLND_CONN_ESTABLISHED)
--
1.8.3.1
More information about the lustre-devel
mailing list