linux-kernel - Re: [PATCH] cgroup/rstat: change cgroup_base

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: <ftih3vsbegjha32tjoktayjyak6zs6n7cbfuwzjasutexqiluo@pjauxm2b6xmn>
Date: Tue, 17 Jun 2025 15:16:30 +0200
From: Michal Koutný <mkoutny@...e.com>
To: Bertrand Wlodarczyk <bertrand.wlodarczyk@...el.com>
Cc: tj@...nel.org, hannes@...xchg.org, cgroups@...r.kernel.org, 
	linux-kernel@...r.kernel.org, JP Kobryn <inwardvessel@...il.com>
Subject: Re: [PATCH] cgroup/rstat: change cgroup_base_stat to atomic

Hello.

This is interesting.

On Tue, Jun 17, 2025 at 12:26:45PM +0200, Bertrand Wlodarczyk <bertrand.wlodarczyk@...el.com> wrote:
> The kernel currently faces scalability issues when multiple userspace
> programs attempt to read cgroup statistics concurrently.

Does "currently" mean that you didn't observe this before per-subsys
split?

> The primary bottleneck is the css_cgroup_lock in cgroup_rstat_flush,
> which prevents access and updates to the statistics
> of the css from multiple CPUs in parallel.

I think this description needs some refresh on top of the current
mainline (at least after the commit 748922dcfabdd ("cgroup: use
subsystem-specific rstat locks to avoid contention") to be clear which
lock (and locking functions) is apparently contentious.

> Given that rstat operates on a per-CPU basis and only aggregates
> statistics in the parent cgroup, there is no compelling reason
> why these statistics cannot be atomic.
> By eliminating the lock, each CPU can traverse its rstat hierarchy
> independently, without blocking. Synchronization is achieved during
> parent propagation through atomic operations.
> 
> This change significantly enhances performance in scenarios
> where multiple CPUs access CPU rstat within a single cgroup hierarchy,
> yielding a performance improvement of around 50 times compared
> to the mainline version.

Given the recent changes and to be on the same page, it'd be better if
you referred to particular commits/tags when benchmarking so that it's
unambiguous what is compared with what.

> Notably, performance for memory and I/O rstats remains unchanged,
> as these are managed in separate submodules.
 

> Additionally, this patch addresses a race condition detectable
> in the current mainline by KCSAN in __cgroup_account_cputime,
> which occurs when attempting to read a single hierarchy
> from multiple CPUs.

Could you please extract this fix and send it separately?

Thanks,
Michal

Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)