Linux-NVME Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: John Garry <john.garry@linux.dev>
To: Nilay Shroff <nilay@linux.ibm.com>, linux-nvme@lists.infradead.org
Cc: kbusch@kernel.org, sagi@grimberg.me, hch@lst.de, axboe@fb.com,
	john.g.garry@oracle.com, wenxiong@linux.ibm.com,
	gjoyce@linux.ibm.com
Subject: Re: [PATCH 0/2] nvme: add reference counting for transport modules
Date: Tue, 1 Sep 2026 09:14:28 +0100	[thread overview]
Message-ID: <eb76f8b4-3730-410c-bdcd-3a36c46686ab@linux.dev> (raw)
In-Reply-To: <20260831152006.819471-1-nilay@linux.ibm.com>

On 8/31/26 16:19, Nilay Shroff wrote:
> Hi,
> 
> This patchset adds reference counting for NVMe transport modules while
> the corresponding multipath head node is open. This prevents the
> underlying transport module from being unloaded while it is still in
> use by the multipath head node. Unloading a transport module while it
> is still in use can result in undefined behavior.
> 

What is the undefined behaviour specifically?

When the ctrl ops module ref counting was originally introduced, the 
commit message mentioned a crash which it solves.

So far for this problem we have seen a report that if we remove the 
module, the mounted FS will have IOs fail. The same can be experienced 
if the admin unbinds the device from the driver. However, I have not 
seen a mention of a crash, kernel data corruption, a hang, etc.


> The second patch adds a Clang context annotation for the shared field
> used to track the number of active head node openers, allowing the
> Clang context analyzer to validate accesses to the field.
> 
> As usual, feedback and suggestions are welcome.
> 
> Thanks!
> 
> Nilay Shroff (2):
>    nvme: keep transport module referenced while head node is open
>    nvme: add context annotation for nvme_ns_head::nr_openers
> 
>   drivers/nvme/host/core.c      | 11 +++++++++--
>   drivers/nvme/host/multipath.c | 37 +++++++++++++++++++++++++++++++++--
>   drivers/nvme/host/nvme.h      | 33 +++++++++++++++++++++++++++++++
>   3 files changed, 77 insertions(+), 4 deletions(-)
> 



  parent reply	other threads:[~2026-09-01  8:14 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-31 15:19 [PATCH 0/2] nvme: add reference counting for transport modules Nilay Shroff
2026-08-31 15:19 ` [PATCH 1/2] nvme: keep transport module referenced while head node is open Nilay Shroff
2026-09-01 16:39   ` Keith Busch
2026-09-02  4:55     ` Nilay Shroff
2026-09-02  0:20   ` Wen Xiong
2026-09-02 13:31   ` Christoph Hellwig
2026-09-02 14:09     ` Nilay Shroff
2026-09-02 14:13       ` Christoph Hellwig
2026-09-02 14:25         ` Keith Busch
2026-09-02 14:27         ` Nilay Shroff
2026-09-13 14:03           ` Nilay Shroff
2026-08-31 15:19 ` [PATCH 2/2] nvme: add context annotation for nvme_ns_head::nr_openers Nilay Shroff
2026-09-01  8:14 ` John Garry [this message]
2026-09-01  9:47   ` [PATCH 0/2] nvme: add reference counting for transport modules Nilay Shroff
2026-09-01 10:14     ` John Garry

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=eb76f8b4-3730-410c-bdcd-3a36c46686ab@linux.dev \
    --to=john.garry@linux.dev \
    --cc=axboe@fb.com \
    --cc=gjoyce@linux.ibm.com \
    --cc=hch@lst.de \
    --cc=john.g.garry@oracle.com \
    --cc=kbusch@kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=nilay@linux.ibm.com \
    --cc=sagi@grimberg.me \
    --cc=wenxiong@linux.ibm.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox