w/o RXE_START_MASK, the last_psn of IB_OPCODE_RC_SEND_ONLY_INV
will not be updated in update_wqe_psn, and the corresponding
wqe will not be acked in rxe_completer due to its last_psn is
zero. Finally, the other wqe will also not be able to be acked,
because the wqe of IB_OPCODE_RC_SEND_ONLY_INV with last_psn 0
is still there. This causes large amount of io timeout when
nvmeof is over rxe.
Add RXE_START_MASK for IB_OPCODE_RC_SEND_ONLY_INV to fix this.
Signed-off-by: Jianchao Wang <[email protected]>
---
drivers/infiniband/sw/rxe/rxe_opcode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/infiniband/sw/rxe/rxe_opcode.c b/drivers/infiniband/sw/rxe/rxe_opcode.c
index 61927c1..4cf1106 100644
--- a/drivers/infiniband/sw/rxe/rxe_opcode.c
+++ b/drivers/infiniband/sw/rxe/rxe_opcode.c
@@ -390,7 +390,7 @@ struct rxe_opcode_info rxe_opcode[RXE_NUM_OPCODE] = {
.name = "IB_OPCODE_RC_SEND_ONLY_INV",
.mask = RXE_IETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK
| RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK
- | RXE_END_MASK,
+ | RXE_END_MASK | RXE_START_MASK,
.length = RXE_BTH_BYTES + RXE_IETH_BYTES,
.offset = {
[RXE_BTH] = 0,
--
2.7.4
On 2018/4/26 11:52, Jianchao Wang wrote:
> w/o RXE_START_MASK, the last_psn of IB_OPCODE_RC_SEND_ONLY_INV
> will not be updated in update_wqe_psn, and the corresponding
> wqe will not be acked in rxe_completer due to its last_psn is
> zero. Finally, the other wqe will also not be able to be acked,
> because the wqe of IB_OPCODE_RC_SEND_ONLY_INV with last_psn 0
> is still there. This causes large amount of io timeout when
> nvmeof is over rxe.
>
> Add RXE_START_MASK for IB_OPCODE_RC_SEND_ONLY_INV to fix this.
>
> Signed-off-by: Jianchao Wang <[email protected]>
Reviewed-by: Zhu Yanjun <[email protected]>
> ---
> drivers/infiniband/sw/rxe/rxe_opcode.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/infiniband/sw/rxe/rxe_opcode.c b/drivers/infiniband/sw/rxe/rxe_opcode.c
> index 61927c1..4cf1106 100644
> --- a/drivers/infiniband/sw/rxe/rxe_opcode.c
> +++ b/drivers/infiniband/sw/rxe/rxe_opcode.c
> @@ -390,7 +390,7 @@ struct rxe_opcode_info rxe_opcode[RXE_NUM_OPCODE] = {
> .name = "IB_OPCODE_RC_SEND_ONLY_INV",
> .mask = RXE_IETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK
> | RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK
> - | RXE_END_MASK,
> + | RXE_END_MASK | RXE_START_MASK,
> .length = RXE_BTH_BYTES + RXE_IETH_BYTES,
> .offset = {
> [RXE_BTH] = 0,
On Thu, 2018-04-26 at 12:02 +0800, Yanjun Zhu wrote:
> On 2018/4/26 11:52, Jianchao Wang wrote:
> > w/o RXE_START_MASK, the last_psn of IB_OPCODE_RC_SEND_ONLY_INV
> > will not be updated in update_wqe_psn, and the corresponding
> > wqe will not be acked in rxe_completer due to its last_psn is
> > zero. Finally, the other wqe will also not be able to be acked,
> > because the wqe of IB_OPCODE_RC_SEND_ONLY_INV with last_psn 0
> > is still there. This causes large amount of io timeout when
> > nvmeof is over rxe.
> >
> > Add RXE_START_MASK for IB_OPCODE_RC_SEND_ONLY_INV to fix this.
> >
> > Signed-off-by: Jianchao Wang <[email protected]>
>
> Reviewed-by: Zhu Yanjun <[email protected]>
Thanks, applied to for-rc.
--
Doug Ledford <[email protected]>
GPG KeyID: B826A3330E572FDD
Key fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDD