From: sagi@grimberg.me (Sagi Grimberg)
Subject: [PATCH rfc 2/2] nvme: fix possible use-after-free condition when controller reset is racing namespace scanning
Date: Thu, 1 Aug 2019 11:46:34 -0700 [thread overview]
Message-ID: <97583db4-8ff5-9d13-bbc8-d133bbc3f28d@grimberg.me> (raw)
In-Reply-To: <b6b929e7-3da5-f274-f7c8-4e98f8585917@suse.de>
>> nvme_stop/start_queue are quiescing the queue, not freezing it which
>> does not block requests in blk_queue_enter.
>> So I don't think that blk_cleanup_queue will hang for unquiesced queues.
>
> My worry here is for quiesced queues; they are not unquiesced when
> blk_clean_queue() is running. So I'm not sure if we're able to flush all
> commands then.
blk_cleanup_queue does not block on quiesced queues, it blocks on frozen
queues.
> But be it as it may, I've sent a patchset for checking the NS_REMOVING
> flag when traversing the namespace list. That should take care of this
> issue, and should render this patch obsolete.
I've shared feedback on that one, I need to understand what is the
crash that you are talking about.. and why not promoting the list
removal in nvme_ns_remove() doesn't address it.
prev parent reply other threads:[~2019-08-01 18:46 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-29 23:31 [PATCH rfc 0/2] nvme controller reset and namespace scan work race conditions Sagi Grimberg
2019-07-29 23:32 ` [PATCH rfc 1/2] nvme: don't remove namespace if revalidate failed because of controller reset Sagi Grimberg
2019-07-30 0:59 ` Keith Busch
2019-07-30 1:04 ` Sagi Grimberg
2019-07-30 1:04 ` Ming Lei
2019-07-30 1:06 ` Sagi Grimberg
2019-07-30 1:10 ` Ming Lei
2019-07-30 1:19 ` Sagi Grimberg
2019-07-30 1:30 ` Ming Lei
2019-07-30 1:40 ` Sagi Grimberg
2019-07-30 2:09 ` Ming Lei
2019-07-30 17:12 ` Sagi Grimberg
2019-07-30 17:30 ` Keith Busch
2019-07-30 18:15 ` Sagi Grimberg
2019-07-31 7:13 ` Hannes Reinecke
2019-07-31 18:08 ` Sagi Grimberg
2019-07-31 7:01 ` Hannes Reinecke
2019-07-31 14:16 ` Keith Busch
2019-07-31 18:03 ` Sagi Grimberg
2019-07-31 19:32 ` Keith Busch
2019-07-31 20:08 ` Sagi Grimberg
2019-07-31 20:16 ` Keith Busch
2019-07-31 20:45 ` Sagi Grimberg
2019-07-31 20:58 ` Keith Busch
2019-07-31 21:14 ` Sagi Grimberg
2019-07-31 21:54 ` Keith Busch
2019-08-01 1:13 ` Sagi Grimberg
2019-08-01 14:33 ` Keith Busch
2019-08-01 18:52 ` Sagi Grimberg
2019-07-31 6:58 ` Hannes Reinecke
2019-07-31 18:11 ` Sagi Grimberg
2019-07-31 20:02 ` Hannes Reinecke
2019-07-31 20:16 ` Sagi Grimberg
2019-07-31 12:18 ` Hannes Reinecke
2019-07-31 18:16 ` Sagi Grimberg
2019-07-31 20:04 ` Hannes Reinecke
2019-07-31 20:37 ` Sagi Grimberg
2019-07-29 23:32 ` [PATCH rfc 2/2] nvme: fix possible use-after-free condition when controller reset is racing namespace scanning Sagi Grimberg
2019-07-31 12:23 ` Hannes Reinecke
2019-07-31 18:21 ` Sagi Grimberg
2019-08-01 7:24 ` Hannes Reinecke
2019-08-01 18:46 ` Sagi Grimberg [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=97583db4-8ff5-9d13-bbc8-d133bbc3f28d@grimberg.me \
--to=sagi@grimberg.me \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.