All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mikulas Patocka <mpatocka@redhat.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Jens Axboe <axboe@kernel.dk>,
	Andrea Arcangeli <aarcange@redhat.com>, Jan Kara <jack@suse.cz>,
	dm-devel@redhat.com, linux-kernel@vger.kernel.org,
	Jeff Moyer <jmoyer@redhat.com>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	kosaki.motohiro@jp.fujitsu.com, linux-fsdevel@vger.kernel.org,
	lwoodman@redhat.com, "Alasdair G. Kergon" <agk@redhat.com>
Subject: Re: [dm-devel] [PATCH 2/3] Introduce percpu rw semaphores
Date: Sun, 29 Jul 2012 01:13:34 -0400 (EDT)	[thread overview]
Message-ID: <Pine.LNX.4.64.1207290107460.29910@file.rdu.redhat.com> (raw)
In-Reply-To: <1343508252.2626.13184.camel@edumazet-glaptop>



On Sat, 28 Jul 2012, Eric Dumazet wrote:

> On Sat, 2012-07-28 at 12:41 -0400, Mikulas Patocka wrote:
> > Introduce percpu rw semaphores
> > 
> > When many CPUs are locking a rw semaphore for read concurrently, cache
> > line bouncing occurs. When a CPU acquires rw semaphore for read, the
> > CPU writes to the cache line holding the semaphore. Consequently, the
> > cache line is being moved between CPUs and this slows down semaphore
> > acquisition.
> > 
> > This patch introduces new percpu rw semaphores. They are functionally
> > identical to existing rw semaphores, but locking the percpu rw semaphore
> > for read is faster and locking for write is slower.
> > 
> > The percpu rw semaphore is implemented as a percpu array of rw
> > semaphores, each semaphore for one CPU. When some thread needs to lock
> > the semaphore for read, only semaphore on the current CPU is locked for
> > read. When some thread needs to lock the semaphore for write, semaphores
> > for all CPUs are locked for write. This avoids cache line bouncing.
> > 
> > Note that the thread that is locking percpu rw semaphore may be
> > rescheduled, it doesn't cause bug, but cache line bouncing occurs in
> > this case.
> > 
> > Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
> 
> I am curious to see how this performs with 4096 cpus ?

Each cpu should have its own rw semaphore in its cache, so I don't see a 
problem there.

When you change block size, all 4096 rw semaphores are locked for write, 
but changing block size is not a performance sensitive operation.

> Really you shouldnt use rwlock in a path if this might hurt performance.
> 
> RCU is probably a better answer.

RCU is meaningless here. RCU allows lockless dereference of a pointer. 
Here the problem is not pointer dereference, the problem is that integer 
bd_block_size may change.

> (bdev->bd_block_size should be read exactly once )

Rewrite all direct and non-direct io code so that it reads block size just 
once ...

Mikulas

  reply	other threads:[~2012-07-29  5:13 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-06-28  3:04 Crash when IO is being submitted and block size is changed Mikulas Patocka
2012-06-28 11:15 ` Jan Kara
2012-06-28 15:44   ` Mikulas Patocka
2012-06-28 16:53     ` Jan Kara
2012-07-16  0:55   ` Mikulas Patocka
2012-07-17 19:19     ` Jeff Moyer
2012-07-19  2:27       ` Mikulas Patocka
2012-07-19 13:33         ` Jeff Moyer
2012-07-28 16:40           ` [PATCH 1/3] Fix " Mikulas Patocka
2012-07-28 16:41             ` [PATCH 2/3] Introduce percpu rw semaphores Mikulas Patocka
2012-07-28 16:42               ` [PATCH 3/3] blockdev: turn a rw semaphore into a percpu rw semaphore Mikulas Patocka
2012-07-28 20:44               ` [PATCH 2/3] Introduce percpu rw semaphores Eric Dumazet
2012-07-29  5:13                 ` Mikulas Patocka [this message]
2012-07-29 10:10                   ` [dm-devel] " Eric Dumazet
2012-07-29 18:36                     ` Eric Dumazet
2012-08-01 20:07                       ` Mikulas Patocka
2012-08-01 20:09                       ` [PATCH 4/3] " Mikulas Patocka
2012-08-31 18:40                         ` [PATCH 0/4] Fix a crash when block device is read and block size is changed at the same time Mikulas Patocka
2012-08-31 18:41                           ` [PATCH 1/4] Add a lock that will be needed by the next patch Mikulas Patocka
2012-08-31 18:42                             ` [PATCH 2/4] blockdev: fix a crash when block size is changed and I/O is issued simultaneously Mikulas Patocka
2012-08-31 18:43                               ` [PATCH 3/4] blockdev: turn a rw semaphore into a percpu rw semaphore Mikulas Patocka
2012-08-31 18:43                                 ` [PATCH 4/4] New percpu lock implementation Mikulas Patocka
2012-08-31 19:27                           ` [PATCH 0/4] Fix a crash when block device is read and block size is changed at the same time Mikulas Patocka
2012-08-31 20:11                             ` Jeff Moyer
2012-08-31 20:34                               ` Mikulas Patocka
2012-09-17 21:19                               ` Jeff Moyer
2012-09-17 21:19                                 ` Jeff Moyer
2012-09-18 17:04                                 ` Mikulas Patocka
2012-09-18 17:22                                   ` Jeff Moyer
2012-09-18 18:55                                     ` Mikulas Patocka
2012-09-18 18:58                                       ` Jeff Moyer
2012-09-18 20:11                                   ` Jeff Moyer
2012-09-25 17:49                                     ` Jeff Moyer
2012-09-25 17:59                                       ` Jens Axboe
2012-09-25 17:59                                         ` Jens Axboe
2012-09-25 18:11                                         ` Jens Axboe
2012-09-25 18:11                                           ` Jens Axboe
2012-09-25 22:49                                           ` [PATCH 1/2] " Mikulas Patocka
2012-09-26  5:48                                             ` Jens Axboe
2012-09-26  5:48                                               ` Jens Axboe
2012-11-16 22:02                                             ` Jeff Moyer
2012-09-25 22:50                                           ` [PATCH 2/2] " Mikulas Patocka
2012-09-25 22:58                                       ` [PATCH 0/4] " Mikulas Patocka
2012-09-26 13:47                                         ` Jeff Moyer
2012-09-26 14:35                                           ` Mikulas Patocka
2012-07-30 17:00                   ` [dm-devel] [PATCH 2/3] Introduce percpu rw semaphores Paul E. McKenney
2012-07-31  0:00                     ` Mikulas Patocka
2012-08-01 17:15                       ` Paul E. McKenney
2012-06-29  6:25 ` Crash when IO is being submitted and block size is changed Vyacheslav Dubeyko

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=Pine.LNX.4.64.1207290107460.29910@file.rdu.redhat.com \
    --to=mpatocka@redhat.com \
    --cc=aarcange@redhat.com \
    --cc=agk@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=dm-devel@redhat.com \
    --cc=eric.dumazet@gmail.com \
    --cc=jack@suse.cz \
    --cc=jmoyer@redhat.com \
    --cc=kosaki.motohiro@jp.fujitsu.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lwoodman@redhat.com \
    --cc=viro@zeniv.linux.org.uk \
    /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.