From: Pekka Enberg <penberg@iki.fi>
To: Jens Axboe <axboe@kernel.dk>, Jann Horn <jannh@google.com>
Cc: io-uring <io-uring@vger.kernel.org>,
Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>,
joseph qi <joseph.qi@linux.alibaba.com>,
Jiufei Xue <jiufei.xue@linux.alibaba.com>,
Pavel Begunkov <asml.silence@gmail.com>,
Christoph Lameter <cl@linux.com>,
Pekka Enberg <penberg@kernel.org>,
David Rientjes <rientjes@google.com>,
Joonsoo Kim <iamjoonsoo.kim@lge.com>,
Andrew Morton <akpm@linux-foundation.org>,
Linux-MM <linux-mm@kvack.org>
Subject: Re: [PATCH RFC} io_uring: io_kiocb alloc cache
Date: Wed, 13 May 2020 23:31:58 +0300 [thread overview]
Message-ID: <d3ff604d-2955-f8f6-dcbd-25ae90569dc3@iki.fi> (raw)
In-Reply-To: <fb43ddb4-693b-5c07-775f-3142502495de@kernel.dk>
Hi Jens,
On 5/13/20 1:20 PM, Pekka Enberg wrote:
>> So I assume if someone does "perf record", they will see significant
>> reduction in page allocator activity with Jens' patch. One possible way
>> around that is forcing the page allocation order to be much higher. IOW,
>> something like the following completely untested patch:
On 5/13/20 11:09 PM, Jens Axboe wrote:
> Now tested, I gave it a shot. This seems to bring performance to
> basically what the io_uring patch does, so that's great! Again, just in
> the microbenchmark test case, so freshly booted and just running the
> case.
Great, thanks for testing!
On 5/13/20 11:09 PM, Jens Axboe wrote:
> Will this patch introduce latencies or non-deterministic behavior for a
> fragmented system?
You have to talk to someone who is more up-to-date with how the page
allocator operates today. But yeah, I assume people still want to avoid
higher-order allocations as much as possible, because they make
allocation harder when memory is fragmented.
That said, perhaps it's not going to the page allocator as much as I
thought, but the problem is that the per-CPU cache size is just to small
for these allocations, forcing do_slab_free() to take the slow path
often. Would be interesting to know if CONFIG_SLAB does better here
because the per-CPU cache size is much larger IIRC.
- Pekka
next prev parent reply other threads:[~2020-05-13 20:32 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-13 16:30 [PATCH RFC} io_uring: io_kiocb alloc cache Jens Axboe
2020-05-13 17:42 ` Jann Horn
2020-05-13 17:42 ` Jann Horn
2020-05-13 18:34 ` Jens Axboe
2020-05-13 19:20 ` Pekka Enberg
2020-05-13 20:09 ` Jens Axboe
2020-05-13 20:31 ` Pekka Enberg [this message]
2020-05-13 20:44 ` Jens Axboe
2020-05-14 8:25 ` Xiaoguang Wang
2020-05-14 14:22 ` Jens Axboe
2020-05-14 14:33 ` Jens Axboe
2020-05-14 14:53 ` Pavel Begunkov
2020-05-14 15:15 ` Jens Axboe
2020-05-14 15:37 ` Pavel Begunkov
2020-05-14 15:53 ` Jens Axboe
2020-05-14 16:18 ` Pavel Begunkov
2020-05-14 16:21 ` Jens Axboe
2020-05-14 16:25 ` Pavel Begunkov
2020-05-14 17:01 ` Jens Axboe
2020-05-14 17:41 ` Jens Axboe
2020-05-16 9:20 ` Xiaoguang Wang
2020-05-16 16:15 ` Xiaoguang Wang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=d3ff604d-2955-f8f6-dcbd-25ae90569dc3@iki.fi \
--to=penberg@iki.fi \
--cc=akpm@linux-foundation.org \
--cc=asml.silence@gmail.com \
--cc=axboe@kernel.dk \
--cc=cl@linux.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=io-uring@vger.kernel.org \
--cc=jannh@google.com \
--cc=jiufei.xue@linux.alibaba.com \
--cc=joseph.qi@linux.alibaba.com \
--cc=linux-mm@kvack.org \
--cc=penberg@kernel.org \
--cc=rientjes@google.com \
--cc=xiaoguang.wang@linux.alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.