linux/fs/dlm
Alexander Aring 00f30c051b fs: dlm: add unbound flag to dlm_io workqueue
This patch will add the WQ_UNBOUND flag to the lowcomms dlm_io workqueue
which handles socket io handling to send and receive dlm messages.
The amount of sockets will be 2 for a 3 node cluster. Each socket has
two different workers for doing send and receive work by calling socket
API functionality. Each worker will do their task in order to send dlm
messages in a ordered stream based socket communication. On receive
side the receive buffer will be queued up for an ordered dlm_process
workqueue to parse received dlm messages. The parsing need to be done
currently in an ordered synchronized way because the dlm message processing
is not being made to parse parallel.

After explaining all those workqueue behaviours in lowcomms, the dlm_io
workqueue is only being used for socket handling. Each socket handling
has 2 workers (send and receive). In a 3 cluster node we will end up
with 4 workers. Without the WQ_UNBOUND flag the workers are tight to a
CPU and can never switch, this could be an advantage because local CPU
execution. However with dlm_locktorture testcase I expierenced not all
workers are always in use and my assumption is that some workers are
bound to the same CPU. We should always send or receive when we are
ready to do so, one reason why we disable nigel algorithm on sockets.
We should be safe to do the socket io handling on any CPU which can be
switched during runtime. There is no assumption that the worker stays on
the same CPU. There is no need to respect any workqueue concurrency
model that each worker can only run on one CPU. Lowcomms queue_work()
mechanism has an higher level flag to be sure that it can't schedule
work if the previous worker did not signal it to keep ordered socket
handling. Therefore this patch sets the WQ_UNBOUND flag to allow workers
being executed by any available CPU.

Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
2023-03-06 15:49:07 -06:00
..
ast.c fs: dlm: fix DLM_IFL_CB_PENDING gets overwritten 2023-03-06 15:49:07 -06:00
ast.h fs: dlm: use a non-static queue for callbacks 2022-11-08 12:59:41 -06:00
config.c fs: dlm: use listen sock as dlm running indicator 2022-11-21 09:45:49 -06:00
config.h fs: dlm: don't use deprecated timeout features by default 2022-08-01 09:31:38 -05:00
debug_fs.c fs: dlm: use a non-static queue for callbacks 2022-11-08 12:59:41 -06:00
dir.c dlm: use __le types for dlm header 2022-04-06 14:02:28 -05:00
dir.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 193 2019-05-30 11:29:21 -07:00
dlm_internal.h fs: dlm: fix DLM_IFL_CB_PENDING gets overwritten 2023-03-06 15:49:07 -06:00
Kconfig fs/dlm: Remove "select SRCU" 2023-01-05 15:40:02 -06:00
lock.c fs: dlm: remove ls_remove_wait waitqueue 2022-11-08 12:59:41 -06:00
lock.h fs: dlm: const void resource name parameter 2022-08-23 15:02:47 -05:00
lockspace.c Driver core changes for 6.3-rc1 2023-02-24 12:58:55 -08:00
lockspace.h fs: dlm: remove DLM_LSFL_FS from uapi 2022-08-23 14:54:54 -05:00
lowcomms.c fs: dlm: add unbound flag to dlm_io workqueue 2023-03-06 15:49:07 -06:00
lowcomms.h fs: dlm: remove socket shutdown handling 2022-11-21 09:45:49 -06:00
lvb_table.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 193 2019-05-30 11:29:21 -07:00
main.c fs: dlm: add midcomms init/start functions 2022-11-21 09:45:49 -06:00
Makefile fs: dlm: don't use deprecated timeout features by default 2022-08-01 09:31:38 -05:00
member.c fs: dlm: catch dlm_add_member() error 2022-11-08 12:59:41 -06:00
member.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 193 2019-05-30 11:29:21 -07:00
memory.c fs: dlm: fix return value check in dlm_memory_init() 2023-01-05 15:37:51 -06:00
memory.h fs: dlm: allow different allocation context per _create_message 2022-11-08 12:59:41 -06:00
midcomms.c fs: dlm: remove unnecessary waker_up() calls 2023-01-23 16:27:28 -06:00
midcomms.h fs: dlm: wait until all midcomms nodes detect version 2023-01-23 14:58:19 -06:00
netlink.c genetlink: start to validate reserved header bytes 2022-08-29 12:47:15 +01:00
plock.c filelock: move file locking definitions to separate header file 2023-01-11 06:52:32 -05:00
rcom.c fd: dlm: trace send/recv of dlm message and rcom 2022-11-08 12:59:41 -06:00
rcom.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 193 2019-05-30 11:29:21 -07:00
recover.c dlm: replace usage of found with dedicated list iterator variable 2022-04-06 14:03:14 -05:00
recover.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 193 2019-05-30 11:29:21 -07:00
recoverd.c fs: dlm: handle recovery result outside of ls_recover 2022-06-24 11:57:48 -05:00
recoverd.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 193 2019-05-30 11:29:21 -07:00
requestqueue.c fs: dlm: avoid false-positive checker warning 2022-11-21 09:45:49 -06:00
requestqueue.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 193 2019-05-30 11:29:21 -07:00
user.c fs: dlm: fix DLM_IFL_CB_PENDING gets overwritten 2023-03-06 15:49:07 -06:00
user.h fs: dlm: use a non-static queue for callbacks 2022-11-08 12:59:41 -06:00
util.c dlm: use __le types for dlm messages 2022-04-06 14:02:37 -05:00
util.h dlm: use __le types for dlm messages 2022-04-06 14:02:37 -05:00