From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: [PATCH for-4.5 v20 07/12] xl/remus: change bool to defbool Date: Fri, 26 Sep 2014 10:02:13 -0400 Message-ID: <20140926140213.GH30097@laptop.dumpdata.com> References: <1411625784-4060-1-git-send-email-yanghy@cn.fujitsu.com> <1411625784-4060-8-git-send-email-yanghy@cn.fujitsu.com> <20140925192130.GG29663@laptop.dumpdata.com> <21540.42889.978272.747892@mariner.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <21540.42889.978272.747892@mariner.uk.xensource.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: Ian Jackson Cc: ian.campbell@citrix.com, wency@cn.fujitsu.com, Jiang Yunhong , eddie.dong@intel.com, xen-devel@lists.xen.org, rshriram@cs.ubc.ca, Yang Hongyang , laijs@cn.fujitsu.com List-Id: xen-devel@lists.xenproject.org On Fri, Sep 26, 2014 at 12:38:49AM +0100, Ian Jackson wrote: > Shriram Rajagopalan writes ("Re: [Xen-devel] [PATCH for-4.5 v20 07/12] xl/remus: change bool to defbool"): > > On Sep 25, 2014 3:21 PM, "Konrad Rzeszutek Wilk" > > wrote: > > > On Thu, Sep 25, 2014 at 02:16:19PM +0800, Yang Hongyang wrote: > > > > Use defbool instead of bool for boolean flags in remus_info struct. > > > > > > While that change by itself looks OK, the change in 'libxl_types.idl' > > > break the ABI. > > (We don't provide ABI stability in libxl (across Xen releases), only > API stability. However, this change does break the API too:) > > > > Could you say a bit of why that is OK? As in, would there had > > > been in the past any users of this ABI that now would have issues with this? > > > > There were no users of the libxl Remus api in the past. Certainly not an API > > level user like libvirt. > > This is IMO a good answer. As I said earlier, the libxl Remus API in > previous version does not really work properly so it seems unlikely > that this change will break any uses. > > > > Also, how important is this patch? Does it have to go in or > > > can it be dropped from the patchset? > > > > Well, this defbool thing came up as part of the feedback. > > Ian C and I think that this variable should be a defbool. That it > wasn't, beforehand, was a mistake. We have an opportinity to fix that > now, but we won't after 4.5 if we release working libxl Remus support > in 4.5 (as we intend.) OK, defbool it is. Thank you for clearing that up.