From mboxrd@z Thu Jan 1 00:00:00 1970 From: George Dunlap Subject: Re: [PATCH 1/5] xen: Remove redundant __attribute__((packed)) statements Date: Thu, 13 Mar 2014 10:49:04 +0000 Message-ID: <53218D20.3060507@eu.citrix.com> References: <1394651319-8893-1-git-send-email-andrew.cooper3@citrix.com> <1394651319-8893-2-git-send-email-andrew.cooper3@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1394651319-8893-2-git-send-email-andrew.cooper3@citrix.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Andrew Cooper , Xen-devel Cc: Eddie Dong , Keir Fraser , Jun Nakajima , Jan Beulich List-Id: xen-devel@lists.xenproject.org On 03/12/2014 07:08 PM, Andrew Cooper wrote: > All of these structure packing statements are redundant, and was confirmed by > diffing xen-syms with and without the change in place. > > Removing struct xgt_desc in its entirety (as it was unreferenced) results in > shuffling the symbol index, but makes no change to the code. > > Signed-off-by: Andrew Cooper I guess I'm becoming a crotchety old maintainer; at least re the tracing code, I don't see how the cost -- your initial effort in figuring out which structures need packing, the cost of review time, and the risk of regression, no matter how small -- is worth the benefit. It seems to me like having all the tracing structures tagged with "__packed" will remind people to think about packing when they're writing or modifying structures. But if Jan and/or Keir think it's a good idea, and are satisfied that it's correct, I won't oppose it. -George