From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: [PATCH V3 0/3] xen-blkfront/xen-blkback trim support Date: Fri, 26 Aug 2011 13:10:48 -0400 Message-ID: <20110826171048.GB2343@dumpdata.com> References: <1314177825-22360-1-git-send-email-lidongyang@novell.com> <20110824141754.GZ32373@reaktio.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Return-path: Content-Disposition: inline In-Reply-To: <20110824141754.GZ32373@reaktio.net> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?= Cc: xen-devel@lists.xensource.com, JBeulich@novell.com, owen.smith@citrix.com, Li Dongyang List-Id: xen-devel@lists.xenproject.org On Wed, Aug 24, 2011 at 05:17:54PM +0300, Pasi K=E4rkk=E4inen wrote: > On Wed, Aug 24, 2011 at 05:23:42PM +0800, Li Dongyang wrote: > > Dear list, > > this is the V3 of the trim support for xen-blkfront/blkback, >=20 > Isn't the generic name for this functionality "discard" in Linux? >=20 > and "trim" being the ATA specific discard-implementation, > and "scsi unmap" the SAS/SCSI specific discard-implementation? Yeah. I think you are right. The 'feature-discard' sounds much much more generic than the 'trim'. Since we are still implementing this and I think we can take the liberty of making it 'feature-discard'. Albeit it would mean that the Citrix folks would have to rewrite their drivers. >=20 > Just wondering.. >=20 > -- Pasi >=20 > > thanks for all your reviews. > > and when I looked back at Owen's patch in Dec 2010, > > http://lists.xensource.com/archives/html/xen-devel/2010-12/msg00299.h= tml > > this patch above also add the trim union to blkif_x86_{32|64}_request= , > > and take care of trim request in blkif_get_x86{32|64}_req(), > > however, in the later versions, the part is just gone. I wonder if it= is > > needed here? Thanks. > >=20 > > Changelog V3: > > rebased on linus's tree > > enum backend types in blkif instead of flags in the interface hea= der > > more reasonable names in xenstore > > move trim requesting handling to a separate function > > do not re-enable interrupts unconditionally when handling respons= e > > set info->feature-trim only when we have all info needed for requ= est queue > > Changelog V2: > > rebased on Jeremy's tree > > fixes according to Jan Beulich's comments > >=20 > >=20 > > _______________________________________________ > > Xen-devel mailing list > > Xen-devel@lists.xensource.com > > http://lists.xensource.com/xen-devel >=20 > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel