JinYu b8769cdb08 nvmf: Add the Keep Alive feature ... The controller shall treat a Keep Alive Timeout in the same manner as connection loss. If the Keep Alive feature is in use and the timer expires, then the controller shall: 1, stop processing commands and set the Controller Fatal Status (CSTS,CFS) bit to '1'; 2, terminate the NVMe Transport connection; 3, break the host to controller association; A timer poller is added to each subsystem to monitor timeout event. Change-Id: I001afab8a6764f30c39df37fa96384180d117486 Signed-off-by: JinYu <jin.yu@intel.com> Reviewed-on: https://review.gerrithub.io/c/439330 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Changpeng Liu <changpeng.liu@intel.com> Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>		2019-01-24 04:24:11 +00:00
..
ctrlr_bdev.c	trace: inline tpoint_mask checking	2018-09-04 17:09:25 +00:00
ctrlr_discovery.c	test/unit: remove unnecessary mock definitions	2019-01-13 00:48:20 +00:00
ctrlr.c	nvmf: Add the Keep Alive feature	2019-01-24 04:24:11 +00:00
rdma.c	rdma: use the new common poll group data buffer cache.	2019-01-18 16:57:37 +00:00
request.c	trace: inline tpoint_mask checking	2018-09-04 17:09:25 +00:00
subsystem.c	test: Use ut_multithread framework in nvmf subsystem unit test	2018-12-18 23:31:08 +00:00
tcp.c	nvmf/tcp: Use the common buffer cache for each polling group	2019-01-18 19:21:58 +00:00
Makefile	test/nvmf: Add unit tests for NVMe-oF TCP transport	2018-12-13 02:42:12 +00:00