From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: [RFC PATCH] Xen blkback for 2.6.40. Date: Mon, 18 Apr 2011 16:12:16 -0400 Message-ID: <20110418201216.GB12562@dumpdata.com> References: <1303149464-875-1-git-send-email-konrad.wilk@oracle.com> <1303151824.9965.72.camel@agari.van.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <1303151824.9965.72.camel@agari.van.xensource.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Daniel Stodden Cc: "xen-devel@lists.xensource.com" List-Id: xen-devel@lists.xenproject.org On Mon, Apr 18, 2011 at 11:37:04AM -0700, Daniel Stodden wrote: > On Mon, 2011-04-18 at 13:57 -0400, Konrad Rzeszutek Wilk wrote: > > I am proposing this patch for 2.6.40. This is an RFC at this point. > > > > The git tree for this specific patch is > > git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen.git devel/xen-blkback-v3.squashed > > > > (the full tree is devel/xen-blkback-v3) > > > > Please take a look at the patch. I know that I need to rewrite the description > > to have more details for the upstream community - and that will be shortly supplied. > > > > Any thoughts on what I should address/remove/augment/etc would be appreciated. And whether > > the code has any obvious pitfalls? > > > > The code was originally lifted from 2.6.18, then ported to 2.6.32, and now I have backported > > it to 2.6.38/.39. scripts/checkpatch.pl had been applied and also some re-ordering of the code > > has been done to make it easier to read. > > If propose to consider eliminating vbd.c and interface.c, inlining the > critical bits where they belong, respectively. Ok. Let me take an axe to that. > > Leaving only blkback.c for the data- and xenbus.c for the control path. > > I don't think there's anything notable in there which has more than a > single position to call it from. It always looked like out of some > abstraction layer to me, maybe with blktap1 in mind. Beyond that, I > don't see that getting anywhere. > > Daniel >