This is a note to let you know that I've just added the patch titled
IB/{qib, hfi1}: Avoid flow control testing for RDMA write operation
to the 4.9-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
ib-qib-hfi1-avoid-flow-control-testing-for-rdma-write-operation.patch
and it can be found in the queue-4.9 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From 5b0ef650bd0f820e922fcc42f1985d4621ae19cf Mon Sep 17 00:00:00 2001
From: Mike Marciniszyn <mike.marciniszyn@xxxxxxxxx>
Date: Mon, 21 Aug 2017 18:26:20 -0700
Subject: IB/{qib, hfi1}: Avoid flow control testing for RDMA write operation
From: Mike Marciniszyn <mike.marciniszyn@xxxxxxxxx>
commit 5b0ef650bd0f820e922fcc42f1985d4621ae19cf upstream.
Section 9.7.7.2.5 of the 1.3 IBTA spec clearly says that receive
credits should never apply to RDMA write.
qib and hfi1 were doing that. The following situation will result
in a QP hang:
- A prior SEND or RDMA_WRITE with immmediate consumed the last
credit for a QP using RC receive buffer credits
- The prior op is acked so there are no more acks
- The peer ULP fails to post receive for some reason
- An RDMA write sees that the credits are exhausted and waits
- The peer ULP posts receive buffers
- The ULP posts a send or RDMA write that will be hung
The fix is to avoid the credit test for the RDMA write operation.
Reviewed-by: Kaike Wan <kaike.wan@xxxxxxxxx>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@xxxxxxxxx>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@xxxxxxxxx>
Signed-off-by: Doug Ledford <dledford@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/infiniband/hw/hfi1/rc.c | 3 ++-
drivers/infiniband/hw/qib/qib_rc.c | 4 ++--
2 files changed, 4 insertions(+), 3 deletions(-)
--- a/drivers/infiniband/hw/hfi1/rc.c
+++ b/drivers/infiniband/hw/hfi1/rc.c
@@ -551,7 +551,7 @@ int hfi1_make_rc_req(struct rvt_qp *qp,
case IB_WR_RDMA_WRITE:
if (newreq && !(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
qp->s_lsn++;
- /* FALLTHROUGH */
+ goto no_flow_control;
case IB_WR_RDMA_WRITE_WITH_IMM:
/* If no credit, return. */
if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) &&
@@ -559,6 +559,7 @@ int hfi1_make_rc_req(struct rvt_qp *qp,
qp->s_flags |= RVT_S_WAIT_SSN_CREDIT;
goto bail;
}
+no_flow_control:
put_ib_reth_vaddr(
wqe->rdma_wr.remote_addr,
&ohdr->u.rc.reth);
--- a/drivers/infiniband/hw/qib/qib_rc.c
+++ b/drivers/infiniband/hw/qib/qib_rc.c
@@ -357,7 +357,7 @@ int qib_make_rc_req(struct rvt_qp *qp, u
case IB_WR_RDMA_WRITE:
if (newreq && !(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
qp->s_lsn++;
- /* FALLTHROUGH */
+ goto no_flow_control;
case IB_WR_RDMA_WRITE_WITH_IMM:
/* If no credit, return. */
if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) &&
@@ -365,7 +365,7 @@ int qib_make_rc_req(struct rvt_qp *qp, u
qp->s_flags |= RVT_S_WAIT_SSN_CREDIT;
goto bail;
}
-
+no_flow_control:
ohdr->u.rc.reth.vaddr =
cpu_to_be64(wqe->rdma_wr.remote_addr);
ohdr->u.rc.reth.rkey =
Patches currently in stable-queue which might be from mike.marciniszyn@xxxxxxxxx are
queue-4.9/ib-qib-hfi1-avoid-flow-control-testing-for-rdma-write-operation.patch