From: Sagi Grimberg <sagig@xxxxxxxxxxxx>
In case of the HCA going into catasrophic error flow, the
beacon post_send is likely to fail, so surely there will
be no completion for it.
In this case, use a best effort approach and don't wait for
beacon completion if we failed to post send it.
Reported-by: Alex Tabachnik <alext@xxxxxxxxxxxx>
Signed-off-by: Sagi Grimberg <sagig@xxxxxxxxxxxx>
Signed-off-by: Or Gerlitz <ogerlitz@xxxxxxxxxxxx>
---
drivers/infiniband/ulp/iser/iser_verbs.c | 4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c
index 2ccbc64..7f78af5 100644
--- a/drivers/infiniband/ulp/iser/iser_verbs.c
+++ b/drivers/infiniband/ulp/iser/iser_verbs.c
@@ -677,8 +677,10 @@ int iser_conn_terminate(struct iser_conn *iser_conn)
/* post an indication that all flush errors were consumed */
err = ib_post_send(ib_conn->qp, &ib_conn->beacon, &bad_wr);
- if (err)
+ if (err) {
iser_err("conn %p failed to post beacon", ib_conn);
+ return 1;
+ }
wait_for_completion(&ib_conn->flush_comp);
}
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
