mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Maoyi Xie <maoyixie.tju@gmail.com>
To: Matthew Sakai <msakai@redhat.com>
Cc: dm-devel@lists.linux.dev, linux-kernel@vger.kernel.org
Subject: dm-vdo: iterator after loop end in mark_slab_journal_dirty?
Date: Tue, 19 May 2026 17:41:32 +0800	[thread overview]
Message-ID: <20260519094132.1964544-1-maoyixie.tju@gmail.com> (raw)

Hi all,

While reading drivers/md/dm-vdo/slab-depot.c I noticed
something that looks like a past the end iterator pattern.
I would appreciate it if you could take a look and let me
know whether this is a real issue, and whether it is worth
fixing.

The site is mark_slab_journal_dirty() (linux-7.1-rc1, around
line 156):

    list_for_each_entry_reverse(dirty_journal, dirty_list,
                                dirty_entry) {
            if (dirty_journal->recovery_lock <=
                journal->recovery_lock)
                    break;
    }

    list_move_tail(&journal->dirty_entry,
                   dirty_journal->dirty_entry.next);

When the loop walks all entries without break (the new
journal's recovery_lock is smaller than every existing
entry), dirty_journal is past the end.
dirty_journal->dirty_entry then aliases dirty_list (the
list head) via container_of offset cancellation. So
dirty_journal->dirty_entry.next is head->next, the first
real entry. list_move_tail then inserts journal just
before the first entry, i.e. at the front of the list.
That is the intended behaviour for a sorted insertion
where the new entry has the smallest recovery_lock. The
dereference of the past the end iterator is undefined per
C11.

Jakob Koschel cleaned up many such sites in 2022, for
example commits 99d8ae4ec8a (tracing: Remove usage of list
iterator variable after the loop), 2966a9918df (clockevents:
Use dedicated list iterator variable) and dc1acd5c946 (dlm:
replace usage of found with dedicated list iterator
variable). This site was not covered.

A candidate fix would track an explicit insertion target.
Declare `struct list_head *insert_after = dirty_list` before
the loop. Overwrite it to `&dirty_journal->dirty_entry` only
when the loop breaks early. The final list_move_tail then
reads `insert_after->next`. On break that points to the
entry right after the insertion position. On fall-through
it points to `head->next`, the first real entry. The
behaviour is unchanged in all cases, including an empty
list and a list with one entry.

If this is intentional or already known, please disregard.
Otherwise, I am happy to send a [PATCH] or to leave the fix
to you. Thank you for your time, and sorry for the noise if
this is not actually worth fixing or has already been
spotted.

Thanks,
Maoyi Xie
https://maoyixie.com/

             reply	other threads:[~2026-05-19  9:41 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-19  9:41 Maoyi Xie [this message]
2026-05-19 23:15 ` Matthew Sakai
2026-05-20  6:55   ` Maoyi Xie

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=20260519094132.1964544-1-maoyixie.tju@gmail.com \
    --to=maoyixie.tju@gmail.com \
    --cc=dm-devel@lists.linux.dev \
    --cc=linux-kernel@vger.kernel.org \
    --cc=msakai@redhat.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®