From: Qu Wenruo <wqu@suse.com>
To: sungbyeongchan <tjdqudcks0424@naver.com>,
David Sterba <dsterba@suse.com>, Chris Mason <mason@kernel.org>
Cc: linux-btrfs@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] btrfs: preserve caller ownership of anon_dev on root insert retry
Date: Tue, 6 Oct 2026 08:09:56 +1030 [thread overview]
Message-ID: <4dcfce5e-032b-4f18-b4f1-bde5e29e2292@suse.com> (raw)
In-Reply-To: <20261005212355.428030-1-tjdqudcks0424@naver.com>
在 2026/10/6 07:53, sungbyeongchan 写道:
> create_subvol() preallocates an anonymous block-device ID and passes it to
> btrfs_get_new_fs_root(). On a cache miss, btrfs_get_root_ref() assigns that ID
> to a temporary root before attempting to insert the root in the cache.
>
> If another caller inserts the same root first, the creator gets -EEXIST and
> drops the temporary root. The final put frees root->anon_dev, but the caller's
> *anon_dev still contains the same ID. The retry finds the cached root and
> frees the caller's ID again, producing an ida_free warning.
Please provide the warning message.
>
> Keep ownership with the caller across the retry by clearing the losing
> temporary root's anon_dev when the ID was supplied by the caller. Roots that
> allocated their own ID retain the existing destruction behavior.
>
> The unfixed sequence was reproduced in two clean boots.
And reproducer.
> With this change the
> creator-side -EEXIST branch was exercised twice, each caller-owned ID was
> released exactly once on retry, no ida_free warning occurred, and ordinary
> subvolume creation, writes, qgroup display, and clean unmount passed.
>
> Fixes: 2dfb1e43f57d ("btrfs: preallocate anon block device at first phase of snapshot creation")
> Reported-by: sungbyeongchan <tjdqudcks0424@naver.com>
> Tested-by: sungbyeongchan <tjdqudcks0424@naver.com>
> Assisted-by: LLM
> Signed-off-by: sungbyeongchan <tjdqudcks0424@naver.com>
> Cc: stable@vger.kernel.org
> ---
> fs/btrfs/disk-io.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
> index dc7ad92876c0..4a5e43dfde6f 100644
> --- a/fs/btrfs/disk-io.c
> +++ b/fs/btrfs/disk-io.c
> @@ -1355,6 +1355,8 @@ again:
> ret = btrfs_insert_fs_root(fs_info, root);
> if (ret) {
> if (ret == -EEXIST) {
> + if (anon_dev && *anon_dev)
> + root->anon_dev = 0;
> btrfs_put_root(root);
> goto again;
> }
prev parent reply other threads:[~2026-10-05 21:40 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-05 21:23 sungbyeongchan
2026-10-05 21:39 ` Qu Wenruo [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=4dcfce5e-032b-4f18-b4f1-bde5e29e2292@suse.com \
--to=wqu@suse.com \
--cc=dsterba@suse.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mason@kernel.org \
--cc=tjdqudcks0424@naver.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
all inboxes | Powered by JetHome®