netdev - Re: [PATCH v2 net-next 1/2] tcp: account for memory pressure signaled by cgroup

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: <CAAVpQUDSnDLxjL2O0xbOJTuV9CSNTE4XMZQ1Z5wxNteeyiCMwg@mail.gmail.com>
Date: Thu, 17 Jul 2025 10:26:33 -0700
From: Kuniyuki Iwashima <kuniyu@...gle.com>
To: Daniel Sedlak <daniel.sedlak@...77.com>
Cc: Shakeel Butt <shakeel.butt@...ux.dev>, "David S. Miller" <davem@...emloft.net>, 
	Eric Dumazet <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, 
	Simon Horman <horms@...nel.org>, Jonathan Corbet <corbet@....net>, Neal Cardwell <ncardwell@...gle.com>, 
	David Ahern <dsahern@...nel.org>, Andrew Morton <akpm@...ux-foundation.org>, 
	Yosry Ahmed <yosry.ahmed@...ux.dev>, linux-mm@...ck.org, netdev@...r.kernel.org, 
	Matyas Hurtik <matyas.hurtik@...77.com>
Subject: Re: [PATCH v2 net-next 1/2] tcp: account for memory pressure signaled
 by cgroup

On Thu, Jul 17, 2025 at 8:31 AM Daniel Sedlak <daniel.sedlak@...77.com> wrote:
>
> On 7/16/25 8:07 PM, Kuniyuki Iwashima wrote:
> >> Incrementing it here will give a very different semantic to this stat
> >> compared to LINUX_MIB_TCPMEMORYPRESSURES. Here the increments mean the
> >> number of times the kernel check if a given socket is under memcg
> >> pressure for a net namespace. Is that what we want?
> >
> > I'm trying to decouple sk_memcg from the global tcp_memory_allocated
> > as you and Wei planned before, and the two accounting already have the
> > different semantics from day1 and will keep that, so a new stat having a
> > different semantics would be fine.
> >
> > But I think per-memcg stat like memory.stat.XXX would be a good fit
> > rather than pre-netns because one netns could be shared by multiple
> > cgroups and multiple sockets in the same cgroup could be spread across
> > multiple netns.
>
> I can move the counter to memory.stat.XXX in favor of this patch

Please do so.  Per-netns stats could be confusing in some setup above.

Thanks!