diff options
| author | Mike Marciniszyn <[email protected]> | 2021-07-29 18:26:22 +0000 |
|---|---|---|
| committer | Jason Gunthorpe <[email protected]> | 2021-08-02 15:45:22 +0000 |
| commit | db4657afd10e45855ac1d8437fcc9a86bd3d741d (patch) | |
| tree | 116e50fdcc3af066f6954a3ca15dc960d754271a /lib/debugobjects.c | |
| parent | RDMA/mlx5: Delay emptying a cache entry when a new MR is added to it recently (diff) | |
| download | kernel-db4657afd10e45855ac1d8437fcc9a86bd3d741d.tar.gz kernel-db4657afd10e45855ac1d8437fcc9a86bd3d741d.zip | |
RDMA/cma: Revert INIT-INIT patch
The net/sunrpc/xprtrdma module creates its QP using rdma_create_qp() and
immediately post receives, implicitly assuming the QP is in the INIT state
and thus valid for ib_post_recv().
The patch noted in Fixes: removed the RESET->INIT modifiy from
rdma_create_qp(), breaking NFS rdma for verbs providers that fail the
ib_post_recv() for a bad state.
This situation was proven using kprobes in rvt_post_recv() and
rvt_modify_qp(). The traces showed that the rvt_post_recv() failed before
ANY modify QP and that the current state was RESET.
Fix by reverting the patch below.
Fixes: dc70f7c3ed34 ("RDMA/cma: Remove unnecessary INIT->INIT transition")
Link: https://lore.kernel.org/r/1627583182-81330-1-git-send-email-mike.marciniszyn@cornelisnetworks.com
Cc: Haakon Bugge <[email protected]>
Cc: Chuck Lever III <[email protected]>
Signed-off-by: Mike Marciniszyn <[email protected]>
Signed-off-by: Jason Gunthorpe <[email protected]>
Diffstat (limited to 'lib/debugobjects.c')
0 files changed, 0 insertions, 0 deletions