From: Jens Axboe <jaxboe@fusionio.com>
To: Kent Overstreet <koverstreet@google.com>
Cc: linux-kernel@vger.kernel.org, linux-aio@kvack.org,
linux-fsdevel@vger.kernel.org, zab@redhat.com, bcrl@kvack.org,
jmoyer@redhat.com, viro@zeniv.linux.org.uk
Subject: Re: [PATCH 00/26] AIO performance improvements/cleanups, v2
Date: Thu, 13 Dec 2012 22:18:08 +0100 [thread overview]
Message-ID: <20121213211808.GJ25017@kernel.dk> (raw)
In-Reply-To: <1354568322-29029-1-git-send-email-koverstreet@google.com>
On Mon, Dec 03 2012, Kent Overstreet wrote:
> Last posting: http://thread.gmane.org/gmane.linux.kernel.aio.general/3169
>
> Changes since the last posting should all be noted in the individual
> patch descriptions.
>
> * Zach pointed out the aio_read_evt() patch was calling functions that
> could sleep in TASK_INTERRUPTIBLE state, that patch is rewritten.
> * Ben pointed out some synchronize_rcu() usage was problematic,
> converted it to call_rcu()
> * The flush_dcache_page() patch is new
> * Changed the "use cancellation list lazily" patch so as to remove
> ki_flags from struct kiocb.
Kent, I ran a few tests, and the below patches still don't seem as fast
as the approach I took. To keep it fair, I used your aio branch and
applied by dio speedups too. As a sanity check, I ran with your branch
alone as well. The quick results below - kaio is kent-aio, just your
branch. kaio-dio is with the direct IO speedups too. jaio is my branch,
which already has the dio changes too.
Devices Branch IOPS
1 kaio ~915K
1 kaio-dio ~930K
1 jaio ~1220K
6 kaio ~3050K
6 kaio-dio ~3080K
6 jaio 3500K
The box runs out of CPU driving power, which is why it doesn't scale
linearly, otherwise I know that jaio at least does. It's basically
completion limited for the 6 device test at the moment.
I'll run some profiling tomorrow morning and get you some better
results. Just thought I'd share these at least.
--
Jens Axboe
--
To unsubscribe, send a message with 'unsubscribe linux-aio' in
the body to majordomo@kvack.org. For more info on Linux AIO,
see: http://www.kvack.org/aio/
Don't email: <a href=mailto:"aart@kvack.org">aart@kvack.org</a>
next prev parent reply other threads:[~2012-12-13 21:18 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-12-03 20:58 [PATCH 00/26] AIO performance improvements/cleanups, v2 Kent Overstreet
2012-12-03 20:58 ` [PATCH 01/26] mm: remove old aio use_mm() comment Kent Overstreet
2012-12-03 20:58 ` [PATCH 02/26] aio: remove dead code from aio.h Kent Overstreet
2012-12-03 20:58 ` [PATCH 03/26] gadget: remove only user of aio retry Kent Overstreet
2012-12-03 20:58 ` [PATCH 04/26] aio: remove retry-based AIO Kent Overstreet
2012-12-27 10:11 ` Fubo Chen
2012-12-03 20:58 ` [PATCH 05/26] char: add aio_{read,write} to /dev/{null,zero} Kent Overstreet
2012-12-03 20:58 ` [PATCH 06/26] aio: Kill return value of aio_complete() Kent Overstreet
2012-12-03 20:58 ` [PATCH 07/26] aio: kiocb_cancel() Kent Overstreet
2012-12-03 20:58 ` [PATCH 08/26] aio: Move private stuff out of aio.h Kent Overstreet
2012-12-03 20:58 ` [PATCH 09/26] aio: dprintk() -> pr_debug() Kent Overstreet
2012-12-03 20:58 ` [PATCH 10/26] aio: do fget() after aio_get_req() Kent Overstreet
2012-12-03 20:58 ` [PATCH 11/26] aio: Make aio_put_req() lockless Kent Overstreet
2012-12-03 20:58 ` [PATCH 12/26] aio: Refcounting cleanup Kent Overstreet
2012-12-03 20:58 ` [PATCH 13/26] aio: Convert read_events() to hrtimers Kent Overstreet
2012-12-03 20:58 ` [PATCH 14/26] aio: Make aio_read_evt() more efficient Kent Overstreet
2012-12-03 20:58 ` [PATCH 15/26] aio: Use flush_dcache_page() Kent Overstreet
2012-12-03 20:58 ` [PATCH 16/26] aio: Use cancellation list lazily Kent Overstreet
2012-12-03 20:58 ` [PATCH 17/26] aio: Change reqs_active to include unreaped completions Kent Overstreet
2012-12-03 20:58 ` [PATCH 18/26] aio: Kill batch allocation Kent Overstreet
2012-12-03 20:58 ` [PATCH 19/26] aio: Kill struct aio_ring_info Kent Overstreet
2012-12-03 20:58 ` [PATCH 20/26] aio: Give shared kioctx fields their own cachelines Kent Overstreet
2012-12-03 20:58 ` [PATCH 21/26] aio: reqs_active -> reqs_available Kent Overstreet
2012-12-03 20:58 ` [PATCH 22/26] aio: percpu reqs_available Kent Overstreet
2012-12-03 20:58 ` [PATCH 23/26] Generic dynamic per cpu refcounting Kent Overstreet
2012-12-27 13:47 ` Fubo Chen
2012-12-03 20:58 ` [PATCH 24/26] aio: Percpu ioctx refcount Kent Overstreet
2012-12-03 20:58 ` [PATCH 25/26] aio: use xchg() instead of completion_lock Kent Overstreet
2012-12-03 20:58 ` [PATCH 26/26] aio: Don't include aio.h in sched.h Kent Overstreet
2012-12-13 21:18 ` Jens Axboe [this message]
2012-12-14 2:26 ` [PATCH 00/26] AIO performance improvements/cleanups, v2 Jack Wang
2012-12-14 7:35 ` Jens Axboe
2012-12-15 9:25 ` Kent Overstreet
2012-12-15 9:46 ` Jens Axboe
2012-12-15 10:36 ` Kent Overstreet
2012-12-15 12:59 ` Jens Axboe
2012-12-15 13:16 ` Jens Axboe
2012-12-18 19:16 ` Kent Overstreet
2012-12-19 6:45 ` Kent Overstreet
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=20121213211808.GJ25017@kernel.dk \
--to=jaxboe@fusionio.com \
--cc=bcrl@kvack.org \
--cc=jmoyer@redhat.com \
--cc=koverstreet@google.com \
--cc=linux-aio@kvack.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=viro@zeniv.linux.org.uk \
--cc=zab@redhat.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).