linux-kernel - Re: [PATCH v7 1/7] selftests/proc: add /proc/pid/maps tearing from vma split test

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: <CAJuCfpHQf6oVx44VaiyVOb4ABAkhGe8NAN+juU-q99OhJCFhUQ@mail.gmail.com>
Date: Wed, 16 Jul 2025 09:44:27 -0700
From: Suren Baghdasaryan <surenb@...gle.com>
To: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
Cc: David Hildenbrand <david@...hat.com>, akpm@...ux-foundation.org, Liam.Howlett@...cle.com, 
	vbabka@...e.cz, peterx@...hat.com, jannh@...gle.com, hannes@...xchg.org, 
	mhocko@...nel.org, paulmck@...nel.org, shuah@...nel.org, adobriyan@...il.com, 
	brauner@...nel.org, josef@...icpanda.com, yebin10@...wei.com, 
	linux@...ssschuh.net, willy@...radead.org, osalvador@...e.de, 
	andrii@...nel.org, ryan.roberts@....com, christophe.leroy@...roup.eu, 
	tjmercier@...gle.com, kaleshsingh@...gle.com, aha310510@...il.com, 
	linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org, 
	linux-mm@...ck.org, linux-kselftest@...r.kernel.org
Subject: Re: [PATCH v7 1/7] selftests/proc: add /proc/pid/maps tearing from
 vma split test

On Wed, Jul 16, 2025 at 7:20 AM Suren Baghdasaryan <surenb@...gle.com> wrote:
>
> On Wed, Jul 16, 2025 at 3:50 AM Lorenzo Stoakes
> <lorenzo.stoakes@...cle.com> wrote:
> >
> > On Wed, Jul 16, 2025 at 12:44:23PM +0200, David Hildenbrand wrote:
> > > On 16.07.25 05:05, Suren Baghdasaryan wrote:
> > > > The /proc/pid/maps file is generated page by page, with the mmap_lock
> > > > released between pages.  This can lead to inconsistent reads if the
> > > > underlying vmas are concurrently modified. For instance, if a vma split
> > > > or merge occurs at a page boundary while /proc/pid/maps is being read,
> > > > the same vma might be seen twice: once before and once after the change.
> > > > This duplication is considered acceptable for userspace handling.
> > > > However, observing a "hole" where a vma should be (e.g., due to a vma
> > > > being replaced and the space temporarily being empty) is unacceptable.
> > > >
> > > > Implement a test that:
> > > > 1. Forks a child process which continuously modifies its address space,
> > > > specifically targeting a vma at the boundary between two pages.
> > > > 2. The parent process repeatedly reads the child's /proc/pid/maps.
> > > > 3. The parent process checks the last vma of the first page and
> > > > the first vma of the second page for consistency, looking for the
> > > > effects of vma splits or merges.
> > > >
> > > > The test duration is configurable via the -d command-line parameter
> > > > in seconds to increase the likelihood of catching the race condition.
> > > > The default test duration is 5 seconds.
> > > >
> > > > Example Command: proc-maps-race -d 10
> > > >
> > > > Signed-off-by: Suren Baghdasaryan <surenb@...gle.com>
> > >
> > > Why is this selftest not making use of any kselftest framework?
> > >
> > > I'm sure there is a very good reason :)
>
> It used to be a part of proc-pid-vm.c and after the split I kept its
> overall structure. I'll look into using the kselftest framework.
> Thanks!
>
> > >
> > > Reading assert() feels very weird compared to other selftests.
> >
> > Sorry to meta-review via your review again David :P
> >
> > But just to say tools/testing/selftests/kselftest_harness.h is really good, and
> > makes life simple. See tools/testing/selftests/mm/guard-regions.c for an example
> > of how they can be used - pretty straightforward and avoids a lot of kselftest
> > boilerplate.
>
> Thanks for the pointers. I need to figure out a way to pass
> command-line parameters to my test. Maybe I can use fixtures for
> that... Let me read more about it.

Ok, I think I'll use environment variables to set verbosity and test
duration. That seems like the cleanest approach.

>
> >
> > >
> > > --
> > > Cheers,
> > >
> > > David / dhildenb
> > >
> > >