linux-kernel - Re: [PATCH PoC] kernel/cgroup/pids: refactor "pids.forks" into "pids.stat"

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] [day] [month] [year] [list]

Message-ID: <m2ak6nayh25g7zzsnm6ouik5uovctaa6kfstxbfboupvbsik3u@7f3wzsfayhhg>
Date: Mon, 12 May 2025 12:20:10 +0200
From: Michal Koutný <mkoutny@...e.com>
To: Max Kellermann <max.kellermann@...os.com>
Cc: tj@...nel.org, hannes@...xchg.org, cgroups@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH PoC] kernel/cgroup/pids: refactor "pids.forks" into
 "pids.stat"

Hi.

On Tue, May 06, 2025 at 05:30:56PM +0200, Max Kellermann <max.kellermann@...os.com> wrote:
> Proof-of-concept patch as reply to
> https://lore.kernel.org/cgroups/aBkqeM0DoXUHHdgq@slm.duckdns.org/ to
> be applied on top of
> https://lore.kernel.org/lkml/20250502121930.4008251-1-max.kellermann@ionos.com/
> 

I'm still curious about the reasoning for the introduction of permanent
metric like this -- why isn't cgroup's cpu.stat and memory.stat (that
are more tied to actual resources) sufficient to troubleshoot. I'd like
to see that in the commit message that introduces pids.stat.

> This is quick'n'dirty, with a lot of code copied from mm/memcontrol.c
> and adjusted.

I think the transfer of PIDS_CHARGE_BATCH into the implementation is
overkill at this stage (and it skews the stat field).

Thanks,
Michal

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