linux-kernel - Re: [PATCH AUTOSEL 5.5 542/542] pipe: use exclusive waits when reading or writing

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-=wjd6BKXEpU0MfEaHuOEK-StRToEcYuu6NpVfR0tR5d6xw@mail.gmail.com>
Date:   Tue, 18 Feb 2020 09:36:08 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Andrei Vagin <avagin@...il.com>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Sasha Levin <sashal@...nel.org>,
        stable <stable@...r.kernel.org>,
        Josh Triplett <josh@...htriplett.org>,
        linux-fsdevel <linux-fsdevel@...r.kernel.org>
Subject: Re: [PATCH AUTOSEL 5.5 542/542] pipe: use exclusive waits when
 reading or writing

On Tue, Feb 18, 2020 at 1:51 AM Andrei Vagin <avagin@...il.com> wrote:
>
> This patch breaks one of CRIU tests. Here is a small reproducer:

Good catch.

> The quick fix looks like this:

That one works, but is not really right.

The things that change the number of readers or writers should simply
use "wake_up_all()".

I thought we did that already, but no - there _was_ one place where we
did it, but that was for the pipe buffer size case, and in that case
it's actually pointless. That case acts just like a "new space or data
was added"

So I think the right fix is the attached patch. Since you had such a
lovely test-case, let me go test it too ;)

                Linus

View attachment "patch.diff" of type "text/x-patch" (1227 bytes)