linux-nvme.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v2 0/5] Handle update hardware queues and queue freeze more carefully
@ 2021-07-08  9:27 Daniel Wagner
  2021-07-08  9:27 ` [PATCH v2 1/5] nvme-fc: Update hardware queues before using them Daniel Wagner
                   ` (4 more replies)
  0 siblings, 5 replies; 11+ messages in thread
From: Daniel Wagner @ 2021-07-08  9:27 UTC (permalink / raw)
  To: linux-nvme
  Cc: linux-kernel, James Smart, Keith Busch, Jens Axboe, Ming Lei,
	Sagi Grimberg, Daniel Wagner

Hi,

I've tested this on top of Ming's patches 'blk-mq: fix
blk_mq_alloc_request_hctx'[1] which fixes all problems (including the
hanger in nvme_wait_freeze()).

Thanks,
Danie

[1] https://lore.kernel.org/linux-nvme/20210629074951.1981284-1-ming.lei@redhat.com/

v1:
 - https://lore.kernel.org/linux-nvme/20210625101649.49296-1-dwagner@suse.de/
v2:
 - reviewed tags collected
 - added 'update hardware queues' for all transport
 - added fix for fc hanger in nvme_wait_freeze_timeout


Initial cover letter:

this is a followup on the crash I reported in

  https://lore.kernel.org/linux-block/20210608183339.70609-1-dwagner@suse.de/

By moving the hardware check up the crash was gone. Unfortuntatly, I
don't understand why this fixes the crash. The per-cpu access is
crashing but I can't see why the blk_mq_update_nr_hw_queues() is
fixing this problem.

Even though I can't explain why it fixes it, I think it makes sense to
update the hardware queue mapping bevore we recreate the IO
queues. Thus I avoided in the commit message to say it fixes
something.

Also during testing I observed the we hang indivinetly in
blk_mq_freeze_queue_wait(). Again I can't explain why we get stuck
there but given a common pattern for the nvme_wait_freeze() is to use
it with a timeout I think the timeout should be used too :)

Anyway, someone with more undertanding of the stack can explain the
problems.


Daniel Wagner (4):
  nvme-fc: Update hardware queues before using them
  nvme-rdma: Update number of hardware queues before using them
  nvme-fc: Wait with a timeout for queue to freeze
  nvme-fc: Freeze queues before destroying them

Hannes Reinecke (1):
  nvme-tcp: Update number of hardware queues before using them

 drivers/nvme/host/fc.c   | 26 +++++++++++++++++---------
 drivers/nvme/host/rdma.c | 13 ++++++-------
 drivers/nvme/host/tcp.c  | 14 ++++++--------
 3 files changed, 29 insertions(+), 24 deletions(-)

-- 
2.29.2


_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme

^ permalink raw reply	[flat|nested] 11+ messages in thread

end of thread, other threads:[~2021-07-09 16:43 UTC | newest]

Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-07-08  9:27 [PATCH v2 0/5] Handle update hardware queues and queue freeze more carefully Daniel Wagner
2021-07-08  9:27 ` [PATCH v2 1/5] nvme-fc: Update hardware queues before using them Daniel Wagner
2021-07-08 10:08   ` Hannes Reinecke
2021-07-08  9:27 ` [PATCH v2 2/5] nvme-tcp: Update number of " Daniel Wagner
2021-07-08  9:27 ` [PATCH v2 3/5] nvme-rdma: " Daniel Wagner
2021-07-08  9:27 ` [PATCH v2 4/5] nvme-fc: Wait with a timeout for queue to freeze Daniel Wagner
2021-07-08 10:12   ` Hannes Reinecke
2021-07-08  9:27 ` [PATCH v2 5/5] nvme-fc: Freeze queues before destroying them Daniel Wagner
2021-07-08 10:14   ` Hannes Reinecke
2021-07-09 16:14   ` James Smart
2021-07-09 16:42     ` Daniel Wagner

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).