d07cc7d35d
We have conflict to handle the NVMf subsystem shut down. The situation is that: If there is shutdown request (e.g., ctrlr+c), we may have subsystem finalization and subsystem initialization conflict (e.g., have NVMf subsystem fini and intialization together), we will have coredump issue like #682. If we interrupt the initialization of the subsystem, following works should do: 1 Do not initilize the next subsystem. 2 Recycle the resources in each subsystem via the spdk_subsystem_fini related function. And this patch will do the general thing, but will not consider the detailed interrupt policy in each subsystem. Change-Id: I2438b4a2462acb05d8c8e06dfff3da3d388d4b70 Signed-off-by: Ziye Yang <ziye.yang@intel.com> Reviewed-on: https://review.gerrithub.io/c/446189 (master) Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/447459 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> |
||
|---|---|---|
| .. | ||
| rpc | ||
| subsystems | ||
| app.c | ||
| json_config.c | ||
| json_config.h | ||
| Makefile | ||
| reactor.c | ||
| rpc.c | ||
| subsystem.c | ||