[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <774dee70-69bd-9f95-d197-4cff83e4e633@kernel.dk>
Date: Tue, 7 Sep 2021 10:43:46 -0600
From: Jens Axboe <axboe@...nel.dk>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: kernel test robot <oliver.sang@...el.com>,
Vasily Averin <vvs@...tuozzo.com>,
Shakeel Butt <shakeelb@...gle.com>,
Alexander Viro <viro@...iv.linux.org.uk>,
Alexey Dobriyan <adobriyan@...il.com>,
Andrei Vagin <avagin@...il.com>,
Borislav Petkov <bp@...en8.de>, Borislav Petkov <bp@...e.de>,
Christian Brauner <christian.brauner@...ntu.com>,
Dmitry Safonov <0x7f454c46@...il.com>,
"Eric W. Biederman" <ebiederm@...ssion.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
"H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...hat.com>,
"J. Bruce Fields" <bfields@...ldses.org>,
Jeff Layton <jlayton@...nel.org>,
Jiri Slaby <jirislaby@...nel.org>,
Johannes Weiner <hannes@...xchg.org>,
Kirill Tkhai <ktkhai@...tuozzo.com>,
Michal Hocko <mhocko@...nel.org>,
Oleg Nesterov <oleg@...hat.com>, Roman Gushchin <guro@...com>,
Serge Hallyn <serge@...lyn.com>, Tejun Heo <tj@...nel.org>,
Thomas Gleixner <tglx@...utronix.de>,
Vladimir Davydov <vdavydov.dev@...il.com>,
Yutian Yang <nglaive@...il.com>,
Zefan Li <lizefan.x@...edance.com>,
Andrew Morton <akpm@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org,
kernel test robot <lkp@...el.com>,
"Huang, Ying" <ying.huang@...el.com>,
Feng Tang <feng.tang@...el.com>,
Zhengjun Xing <zhengjun.xing@...ux.intel.com>
Subject: Re: [memcg] 0f12156dff: will-it-scale.per_process_ops -33.6%
regression
On 9/7/21 10:39 AM, Linus Torvalds wrote:
> On Tue, Sep 7, 2021 at 8:46 AM Jens Axboe <axboe@...nel.dk> wrote:
>>
>> Are we at all worried about these? There's been a number of them
>> reported, basically for all the accounting enablements that have been
>> done in this merge window.
>
> We are worried about them. I'm considering reverting several of them
> because I think the problems are
>
> (a) big
>
> (b) nontrivial
>
> and the patches clearly weren't ready and people weren't aware of this issue.
I think that is prudent. When I first enabled it for io_uring it was a
bit of a shit show in terms of performance degradations, and some work
had to be done before it could get enabled in a sane fashion.
The accounting needs to be more efficient if we're seeing 30-50%
slowdowns simply by enabling it on a kmem cache.
--
Jens Axboe
Powered by blists - more mailing lists