From: Al Viro <viro@ftp.linux.org.uk>
To: Akinobu Mita <mita@miraclelinux.com>
Cc: Greg KH <greg@kroah.com>, Jens Axboe <axboe@suse.de>,
linux-kernel@vger.kernel.org, akpm@osdl.org,
Nick Piggin <nickpiggin@yahoo.com.au>
Subject: Re: [patch 3/3] use kref for bio
Date: Wed, 26 Apr 2006 08:39:44 +0100 [thread overview]
Message-ID: <20060426073944.GL27946@ftp.linux.org.uk> (raw)
In-Reply-To: <20060426072030.GA7693@miraclelinux.com>
On Wed, Apr 26, 2006 at 03:20:30PM +0800, Akinobu Mita wrote:
> If this is good one and the places where Al Viro pointed out really affect
> performance, should we propagate this faster one by introducing helper
> function like:
>
> static inline int refcount_test(atomic_t *refcount)
> {
> return (atomic_read(refcount) == 1) || (atomic_dec_and_test(refcount));
> }
>
> and replace atomic_dec_and_test with it?
No. It's obviously slower than atomic_dec_and_test() if refcount is
greater than 1. And I'm less than sure that you can show that benefits
in case when it is 1 outweight that. Moreover, for dentries, inodes,
superblocks and vfsmounts you'd have to pull spin_lock() in front of
it, which would _definitely_ hurt (these are atomic_dec_and_lock()).
next prev parent reply other threads:[~2006-04-26 7:39 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-04-26 2:10 [patch 0/3] use kref Akinobu Mita
2006-04-26 2:11 ` [patch 1/3] use kref for blk_queue_tag Akinobu Mita
2006-04-26 2:42 ` Nick Piggin
2006-04-26 5:17 ` Greg KH
2006-04-26 2:11 ` [patch 2/3] use kref for io_context Akinobu Mita
2006-04-26 2:11 ` [patch 3/3] use kref for bio Akinobu Mita
2006-04-26 2:26 ` Al Viro
2006-04-26 5:13 ` Jens Axboe
2006-04-26 5:18 ` Greg KH
2006-04-26 7:20 ` Akinobu Mita
2006-04-26 7:39 ` Al Viro [this message]
2006-04-26 9:29 ` Nick Piggin
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=20060426073944.GL27946@ftp.linux.org.uk \
--to=viro@ftp.linux.org.uk \
--cc=akpm@osdl.org \
--cc=axboe@suse.de \
--cc=greg@kroah.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mita@miraclelinux.com \
--cc=nickpiggin@yahoo.com.au \
/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