From: Kaike Wan <kaike.wan@xxxxxxxxx>
The following message occurs when running an AI application
with TID RDMA enabled:
hfi1 0000:7f:00.0: hfi1_0: [QP74] hfi1_tid_timeout 4084
hfi1 0000:7f:00.0: hfi1_0: [QP70] hfi1_tid_timeout 4084
The issue happens when TID RDMA WRITE request is followed by an
IB_WR_RDMA_WRITE_WITH_IMM request, the latter could be completed
first on the responder side. As a result, no ACK packet for the
latter could be sent because the TID RDMA WRITE request is still
being processed on the responder side.
When the TID RDMA WRITE request is eventually completed, the requester
will wait for the IB_WR_RDMA_WRITE_WITH_IMM request to be acknowledged.
If the next request is another TID RDMA WRITE request, no
TID RDMA WRITE DATA packet could be sent because the preceding
IB_WR_RDMA_WRITE_WITH_IMM request is not completed yet.
Consequently the IB_WR_RDMA_WRITE_WITH_IMM will be retried but
it will be ignored on the responder side because the responder
thinks it has already been completed. Eventually the retry will
be exhausted and the qp will be put into error state on the requester
side. On the responder side, the TID resource timer will eventually
expire because no TID RDMA WRITE DATA packets will be received for
the second TID RDMA WRITE request. There is also risk of a
write-after-write memory corruption due to the issue.
Fix by adding a requester side interlock to prevent any potential
data corruption and TID RDMA protocol error.
Fixes: a0b34f75ec20 ("IB/hfi1: Add interlock between a TID RDMA request and other requests")
Cc: <stable@xxxxxxxxxxxxxxx> # 5.4.x+
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@xxxxxxxxx>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@xxxxxxxxx>
Signed-off-by: Kaike Wan <kaike.wan@xxxxxxxxx>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@xxxxxxxxx>
---
drivers/infiniband/hw/hfi1/tid_rdma.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c
index 9af82ff..73d197e 100644
--- a/drivers/infiniband/hw/hfi1/tid_rdma.c
+++ b/drivers/infiniband/hw/hfi1/tid_rdma.c
@@ -3215,6 +3215,7 @@ bool hfi1_tid_rdma_wqe_interlock(struct rvt_qp *qp, struct rvt_swqe *wqe)
case IB_WR_ATOMIC_CMP_AND_SWP:
case IB_WR_ATOMIC_FETCH_AND_ADD:
case IB_WR_RDMA_WRITE:
+ case IB_WR_RDMA_WRITE_WITH_IMM:
switch (prev->wr.opcode) {
case IB_WR_TID_RDMA_WRITE:
req = wqe_to_tid_req(prev);
