From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: Full hostlock pushdown available Date: Tue, 2 Nov 2010 10:21:11 +0100 Message-ID: <20101102092111.GK25817@basil.fritz.box> References: <20101028150508.GA2385@basil.fritz.box> <4CCD5F7F.8020808@panasas.com> <4CCE271D.7040400@garzik.org> <20101101135338.GA25817@basil.fritz.box> <4CCEE33F.5030300@garzik.org> <20101101175742.GG25817@basil.fritz.box> <4CCF0DFE.4050106@garzik.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <4CCF0DFE.4050106@garzik.org> Sender: linux-ide-owner@vger.kernel.org To: Jeff Garzik Cc: Andi Kleen , Stefan Richter , Boaz Harrosh , James.Bottomley@suse.de, nab@linux-iscsi.org, linux-scsi@vger.kernel.org, Linux IDE mailing list List-Id: linux-scsi@vger.kernel.org > It's a bit disappointing that libata's lock profile in this patch is > quite different than that of current upstream: with your patch, > libata holds the scsi host lock for a considerably longer period of > time, while also holding the ATA port/host spinlock. The goal here is not really what comes out of this patch, but dropping the host lock completely. This is just the first step. > > IOW, it's doing the exact opposite of what the previous code did > (release the scsi host lock, before acquiring the ATA port/host > spinlock), not at all an equivalent transformation. > > The following sequence would seem to better preserve the existing > lock profile, correct? Possibly, but it's not a mechanic change. -Andi