From owner-cvs-src@FreeBSD.ORG Fri Mar 21 11:03:43 2008 Return-Path: Delivered-To: cvs-src@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id A23581065670; Fri, 21 Mar 2008 11:03:43 +0000 (UTC) (envelope-from jroberson@chesapeake.net) Received: from webaccess-cl.virtdom.com (webaccess-cl.virtdom.com [216.240.101.25]) by mx1.freebsd.org (Postfix) with ESMTP id 6F7C78FC2C; Fri, 21 Mar 2008 11:03:43 +0000 (UTC) (envelope-from jroberson@chesapeake.net) Received: from [192.168.1.107] (cpe-24-94-75-93.hawaii.res.rr.com [24.94.75.93]) (authenticated bits=0) by webaccess-cl.virtdom.com (8.13.6/8.13.6) with ESMTP id m2LB3W4J054418; Fri, 21 Mar 2008 07:03:36 -0400 (EDT) (envelope-from jroberson@chesapeake.net) Date: Fri, 21 Mar 2008 01:04:35 -1000 (HST) From: Jeff Roberson X-X-Sender: jroberson@desktop To: Poul-Henning Kamp In-Reply-To: <4851.1206094122@critter.freebsd.dk> Message-ID: <20080321010256.J910@desktop> References: <4851.1206094122@critter.freebsd.dk> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: cvs-src@FreeBSD.org, Jeff Roberson , src-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/kern vfs_bio.c X-BeenThere: cvs-src@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: CVS commit messages for the src tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 21 Mar 2008 11:03:43 -0000 On Fri, 21 Mar 2008, Poul-Henning Kamp wrote: > In message <200803211000.m2LA05KN096467@repoman.freebsd.org>, Jeff Roberson wri > tes: >> jeff 2008-03-21 10:00:05 UTC >> >> FreeBSD src repository >> >> Modified files: >> sys/kern vfs_bio.c >> Log: >> - Reduce contention on the global bdonelock and bpinlock by using >> a pool mutex to protect these sleep/wakeup/counter races. This >> still is preferable to bloating each bio with a mtx. > > Shouldn't this be the BOLOCK ? As mentioned on irc; I believe these originally were not done this way because we may have a buf without a bufobj. I think that happens when we use a pager buf for physio. The other reason not to do that would be to attempt to reduce contention on the bo lock which is increasingly becoming a problem. Patches are underway to address that. Thanks, Jeff > > -- > Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 > phk@FreeBSD.ORG | TCP/IP since RFC 956 > FreeBSD committer | BSD since 4.3-tahoe > Never attribute to malice what can adequately be explained by incompetence. >