From mboxrd@z Thu Jan 1 00:00:00 1970 From: Boaz Harrosh Subject: Re: [BUG] 2.6.24-git usb reset problems Date: Tue, 29 Jan 2008 22:26:48 +0200 Message-ID: <479F8C08.2060207@panasas.com> References: <20080129141108.GV15220@kernel.dk> <200801291531.39825.oliver@neukum.org> <20080129143109.GW15220@kernel.dk> <20080129183910.GI15220@kernel.dk> <20080129191024.GO14375@one-eyed-alien.net> <1201635195.3069.36.camel@localhost.localdomain> <20080129193520.GQ15220@kernel.dk> <20080129194543.GR15220@kernel.dk> <479F856C.4090900@panasas.com> <479F880F.20709@panasas.com> <20080129201328.GZ15220@kernel.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20080129201328.GZ15220-tSWWG44O7X1aa/9Udqfwiw@public.gmane.org> Sender: linux-usb-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Jens Axboe Cc: James Bottomley , Matthew Dharm , Oliver Neukum , Greg KH , linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-usb-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-scsi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-scsi@vger.kernel.org On Tue, Jan 29 2008 at 22:13 +0200, Jens Axboe wrote: > On Tue, Jan 29 2008, Boaz Harrosh wrote: >>> Ok this is not in Linus tree is it? Hence I did not have that failure. >>> >>> Boaz >>> >>> >> actually James bidi tree has a fix for this in the scsi_data_buffer patch. >> >> what you sent is not enough there are other places. look at this patch I >> sent to the list. >> >> http://www.spinics.net/lists/linux-scsi/msg21938.html > > Hard to compare, since its on different bases. Yes in this patchset I have taken your sg branch at the time, and rebased it ontop of scsi_data_buffer patch. Because I felt that it is more natural for this patch to come after the scsi total cleanup that is scsi_data_buffer. Then the extraction to sg_table is simple and trivial. What I meant to point out with this patch is that all the exact same places that are touched there should be fixed when moving to sg_table. Look at it. It is a revised version of your patch. > >> Could we take the 2 SG patches and submit them through the scsi >> bidi tree? It is much more natural to have them in one tree as one >> patchset then try coordinate with git-merge. Actually if you look at it, >> the biggest change is to SCSI. So I think it is more natural this way > > What 2 sg patches do you mean? > I mean the patches that where in sg branch of the linux-block tree, But I see that it is now to late, and that they are in Linus already James the most simple is to submit the scsi_data_buff patch that fixes all these places. If not do you want that I send in fixes? Boaz