Re: [PATCH] build: Conditional build/clean/distclean targets on subsystems

[George Dunlap <george.dunlap@eu.citrix.com>]

On 08/05/13 13:51, Ian Jackson wrote:
> Andrew Cooper writes ("Re: [Xen-devel] [PATCH] build: Conditional build/clean/distclean targets on subsystems"):
>> On 08/05/13 12:18, Ian Campbell wrote:
>>> On Wed, 2013-05-08 at 11:22 +0100, George Dunlap wrote:
>>>> On Tue, May 7, 2013 at 6:27 PM, Ian Jackson <Ian.Jackson@eu.citrix.com> wrote:
>>>>> Andrew Cooper writes ("Re: [Xen-devel] [PATCH] build: Conditional build/clean/distclean targets on subsystems"):
>>>>>> I did attempt to fix this before with
>>>>>> http://lists.xen.org/archives/html/xen-devel/2012-08/msg00134.html
> ...
>>>>> At the time I thought we also needed this patch, before yours:
>>>>>    http://lists.xen.org/archives/html/xen-devel/2012-08/msg00870.html
>>>>>
>>>>> It looks from that thread like I was hoping for an ack from someone
>>>>> else on my patch.
>>>>>
>>>>> These look like bugfixes to me.  George, are they OK for 4.3 ?
>>>> Hrm, I'm not terribly happy about this kind of change this late in the
>>>> day, but I think overall it's best to fix this.
> Perhaps I don't understand the freeze policy.  I thought we were still
> accepting bugfixes ?  I guess at some point we'll start accepting
> fixes only for release-critical bugs, but surely not yet ?

I haven't been using a strict rule; I've been attempting to approach 
each request from a risk/benefits perspective.  In this case, the change 
doesn't look *terribly* risky; but it does seem like there may be a bit 
of risk, and overall the benefit didn't seem to be terribly large 
either.  On balance, in my judgement it comes pretty close to the line.  
It's quite possible I'm judging inaccurately in this case -- either 
overestimating the risk, or underestimating the benefit.  But it's 
probably better to have a discussion about that in a case where we 
actually disagree about the conclusion. :-)

  -George