[PATCH 5.15 096/121] RDMA/irdma: Return correct WC error for bind operation failure
From: Greg Kroah-Hartman
Date: Tue Sep 13 2022 - 10:45:41 EST
From: Sindhu-Devale <sindhu.devale@xxxxxxxxx>
[ Upstream commit dcb23bbb1de7e009875fdfac2b8a9808a9319cc6 ]
When a QP and a MR on a local host are in different PDs, the HW generates
an asynchronous event (AE). The same AE is generated when a QP and a MW
are in different PDs during a bind operation. Return the more appropriate
IBV_WC_MW_BIND_ERR for the latter case by checking the OP type from the
CQE in error.
Fixes: 551c46edc769 ("RDMA/irdma: Add user/kernel shared libraries")
Signed-off-by: Sindhu-Devale <sindhu.devale@xxxxxxxxx>
Signed-off-by: Shiraz Saleem <shiraz.saleem@xxxxxxxxx>
Link: https://lore.kernel.org/r/20220906223244.1119-4-shiraz.saleem@xxxxxxxxx
Signed-off-by: Leon Romanovsky <leon@xxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/infiniband/hw/irdma/uk.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/irdma/uk.c b/drivers/infiniband/hw/irdma/uk.c
index 9b544a3b12886..7e6c3ba8df6ab 100644
--- a/drivers/infiniband/hw/irdma/uk.c
+++ b/drivers/infiniband/hw/irdma/uk.c
@@ -1068,6 +1068,7 @@ irdma_uk_cq_poll_cmpl(struct irdma_cq_uk *cq, struct irdma_cq_poll_info *info)
enum irdma_status_code ret_code;
bool move_cq_head = true;
u8 polarity;
+ u8 op_type;
bool ext_valid;
__le64 *ext_cqe;
@@ -1250,7 +1251,6 @@ irdma_uk_cq_poll_cmpl(struct irdma_cq_uk *cq, struct irdma_cq_poll_info *info)
do {
__le64 *sw_wqe;
u64 wqe_qword;
- u8 op_type;
u32 tail;
tail = qp->sq_ring.tail;
@@ -1267,6 +1267,8 @@ irdma_uk_cq_poll_cmpl(struct irdma_cq_uk *cq, struct irdma_cq_poll_info *info)
break;
}
} while (1);
+ if (op_type == IRDMA_OP_TYPE_BIND_MW && info->minor_err == FLUSH_PROT_ERR)
+ info->minor_err = FLUSH_MW_BIND_ERR;
qp->sq_flush_seen = true;
if (!IRDMA_RING_MORE_WORK(qp->sq_ring))
qp->sq_flush_complete = true;
--
2.35.1