linux-kernel - Re: [GIT PULL] hotfixes for 6.2

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAHk-=wiBrY+O-4=2mrbVyxR+hOqfdJ=Do6xoucfJ9_5az01L4Q@mail.gmail.com>
Date:   Mon, 13 Feb 2023 14:19:19 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Andrew Morton <akpm@...ux-foundation.org>
Cc:     linux-mm@...ck.org, mm-commits@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [GIT PULL] hotfixes for 6.2

On Mon, Feb 13, 2023 at 2:08 PM Andrew Morton <akpm@...ux-foundation.org> wrote:
>
> Kuan-Ying Lee (1):
>       mm/gup: add folio to list when folio_isolate_lru() succeed

Ugh. I really hate fixes like this.

The problem came from mis-understanding the return value of
folio_isolate_lru(), and thinking that it was a boolean
success/failure thing.

It wasn't, it was an integer "success/errno" thing, and the sense of
the test was wrong. So the patch is

-       if (!folio_isolate_lru(folio))
+       if (folio_isolate_lru(folio))
                continue;

but at no point was the code *clarified*.

Wouldn't it have been much better to write the new code to be

        if (folio_isolate_lru(folio) < 0)
                continue;

to actually make it clear that this is a "negative error return check".

I've pulled this, but I really think that when somebody notices that
we had a silly bug because of a misunderstanding like this, it's not
just that the bug should be fixed, the code should also be *clarified*
at the same time.

                 Linus