[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a0480aa7-4be5-49c5-a20a-3bdf936d29d4@kernel.dk>
Date: Tue, 10 Sep 2024 08:55:36 -0600
From: Jens Axboe <axboe@...nel.dk>
To: Felix Moessbauer <felix.moessbauer@...mens.com>
Cc: asml.silence@...il.com, linux-kernel@...r.kernel.org,
io-uring@...r.kernel.org, cgroups@...r.kernel.org, dqminh@...udflare.com,
longman@...hat.com, adriaan.schmidt@...mens.com, florian.bezdeka@...mens.com
Subject: Re: [PATCH 1/2] io_uring/io-wq: do not allow pinning outside of
cpuset
On 9/10/24 8:33 AM, Felix Moessbauer wrote:
> The io work queue polling threads are userland threads that just never
> exit to the userland. By that, they are also assigned to a cgroup (the
> group of the creating task).
They are not polling threads, they are just worker threads.
> When changing the affinity of the io_wq thread via syscall, we must only
> allow cpumasks within the ambient limits. These are defined by the cpuset
> controller of the cgroup (if enabled).
ambient limits? Not quite sure that's the correct term to use here.
Outside of commit message oddities, change looks good.
--
Jens Axboe
Powered by blists - more mailing lists