From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: [PATCH v2] block: rename trace_block_remap to trace_block_bio_remap Date: Tue, 16 Nov 2010 12:51:48 +0100 Message-ID: <4CE27054.3000309@fusionio.com> References: <4AB3A393.5000709@ce.jp.nec.com> <4AB4F908.60307@ce.jp.nec.com> <20101103215047.GC26473@redhat.com> <20101115205815.GA25567@redhat.com> <20101115213444.GA25705@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20101115213444.GA25705@redhat.com> Sender: linux-kernel-owner@vger.kernel.org To: Mike Snitzer Cc: device-mapper development , Jun'ichi Nomura , Alasdair G Kergon , Li Zefan , "linux-kernel@vger.kernel.org" List-Id: dm-devel.ids On 2010-11-15 22:34, Mike Snitzer wrote: > On Mon, Nov 15 2010 at 3:58pm -0500, > Mike Snitzer wrote: > >> Hi Jens, >> >> This may have slipped through while you were traveling? >> >> Patch is also available here: >> https://patchwork.kernel.org/patch/300552/ >> >> Not sure if this change is concerning on a kABI level? > > To be clear: that concern/question was prompted from having taken a > quick glance at lwn's kernel summit "KS2010: ABI status for tracepoints" > coverage: http://lwn.net/Articles/412685/ > > But in that article it clearly states: > "there are to be no stable tracepoints in drivers or filesystems". > > Not sure where that leaves tracepoints for the block layer. Anyway, > seems tracepoint kABI is likely a concern for the future (not the > immediate future). > > So this change would be nice to get in before there is a new tracepoint > kABI-aware sheriff in town. As far as I'm concerned, trace point changes in the block area cannot break existing blktrace (as in cause malfunction). So this patch is fine, I'll pick it up for .38. -- Jens Axboe