Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <facbfec3-837a-51ed-85fa-31021c17d6ef@gmail.com> <5c7455db-4ed8-b54f-e2d5-d2811908123d@leemhuis.info>
In-Reply-To: <5c7455db-4ed8-b54f-e2d5-d2811908123d@leemhuis.info>
From:   Suren Baghdasaryan <surenb@google.com>
Date:   Mon, 3 Jul 2023 11:08:33 -0700
Message-ID: <CAJuCfpH7BOBYGEG=op09bZrh1x3WA8HMcGBXXRhe6M5RJaen5A@mail.gmail.com>
Subject: Re: Fwd: Memory corruption in multithreaded user space program while
 calling fork
To:     Linux regressions mailing list <regressions@lists.linux.dev>
Cc:     Bagas Sanjaya <bagasdotme@gmail.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Jacob Young <jacobly.alt@gmail.com>,
        Laurent Dufour <ldufour@linux.ibm.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux Memory Management <linux-mm@kvack.org>,
        Linux PowerPC <linuxppc-dev@lists.ozlabs.org>,
        Linux ARM <linux-arm-kernel@lists.infradead.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Mon, Jul 3, 2023 at 2:53=E2=80=AFAM Linux regression tracking (Thorsten
Leemhuis) <regressions@leemhuis.info> wrote:
>
> On 02.07.23 14:27, Bagas Sanjaya wrote:
> > I notice a regression report on Bugzilla [1]. Quoting from it:
> >
> >> After upgrading to kernel version 6.4.0 from 6.3.9, I noticed frequent=
 but random crashes in a user space program.  After a lot of reduction, I h=
ave come up with the following reproducer program:
> > [...]
> >> After tuning the various parameters for my computer, exit code 2, whic=
h indicates that memory corruption was detected, occurs approximately 99% o=
f the time.  Exit code 1, which occurs approximately 1% of the time, means =
it ran out of statically-allocated memory before reproducing the issue, and=
 increasing the memory usage any more only leads to diminishing returns.  T=
here is also something like a 0.1% chance that it segfaults due to memory c=
orruption elsewhere than in the statically-allocated buffer.
> >>
> >> With this reproducer in hand, I was able to perform the following bise=
ction:
> > [...]
> >
> > See Bugzilla for the full thread.
>
> Additional details from
> https://bugzilla.kernel.org/show_bug.cgi?id=3D217624#c5 :
>
> ```
> I can confirm that v6.4 with 0bff0aaea03e2a3ed6bfa302155cca8a432a1829
> reverted no longer causes any memory corruption with either my
> reproducer or the original program.
> ```
>
> FWIW: 0bff0aaea03 ("x86/mm: try VMA lock-based page fault handling
> first") [merged for v6.4-rc1, authored by Suren Baghdasaryan [already CCe=
d]]
>
> That's the same commit that causes build problems with go:
>
> https://lore.kernel.org/all/dbdef34c-3a07-5951-e1ae-e9c6e3cdf51b@kernel.o=
rg/

Thanks! I'll investigate this later today. After discussing with
Andrew, we would like to disable CONFIG_PER_VMA_LOCK by default until
the issue is fixed. I'll post a patch shortly.

>
> Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
> --
> Everything you wanna know about Linux kernel regression tracking:
> https://linux-regtracking.leemhuis.info/about/#tldr
> If I did something stupid, please tell me, as explained on that page.
>
> #regzbot introduced: 0bff0aaea03e2a3