d/numam-spdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

nvme_rdma: Detect stale connection failures.

Browse Source 
This is the first step in properly reconnecting after a hard power off
event.

Change-Id: I9739bffacd66ec6d9f8f1d376bf42291c84f90f2
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473061
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This commit is contained in:
Seth Howell 2019-11-01 15:31:32 -07:00 committed by Tomasz Zawadzki
parent 208fbb675c
commit 6b87dd8023
1 changed files with 15 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
lib/nvme/nvme_rdma.c
View File

@ -362,6 +362,21 @@ nvme_rdma_validate_cm_event(enum rdma_cm_event_type expected_evt_type,
return 0;
}
switch (expected_evt_type) {
case RDMA_CM_EVENT_ESTABLISHED:
/*
* There is an enum ib_cm_rej_reason in the kernel headers that sets 10 as
* IB_CM_REJ_STALE_CONN. I can't find the corresponding userspace but we get
* the same values here.
*/
if (reaped_evt->event == RDMA_CM_EVENT_REJECTED && reaped_evt->status == 10) {
rc = -ESTALE;
}
break;
default:
break;
}
SPDK_ERRLOG("Expected %s but received %s (%d) from CM event channel (status = %d)\n",
nvme_rdma_cm_event_str_get(expected_evt_type),
nvme_rdma_cm_event_str_get(reaped_evt->event), reaped_evt->event,