Re: [xen-unstable test] 53884: regressions - FAIL

[Julien Grall <julien.grall@citrix.com>]

Hi,

On 11/05/15 10:05, Ian Campbell wrote:
> On Mon, 2015-05-11 at 09:32 +0100, Jan Beulich wrote:
> 
>>> While it'd be really nice to do so, the flight here didn't test the tip of
>>> staging (onto which I'll push the revert in a few minutes), so a force
>>> push may be a little difficult to arrange. I'm also afraid that doing so
>>> would yield the XSM build failure an allowable one going forward.
>>> Otoh, considering that we got two almost-pushes recently, I'm quite
>>> hopeful that we'll get a real push pretty soon.
>>
>> And sadly 6ea9f0481e ("xen: guestcopy: Provide an helper to
>> safely copy string from guest") breaks the x86 XSM build too (due
>> to type issues with the compat mode layer of flask's hypercall
>> handling). Reverting that one would be more cumbersome (as I
>> suppose subsequent changes depend on it), but I very much
>> hope that you, Julien, will provide a timely fix for this.
> 
> Yes, please.
> 
> There's no point my killing 55228 until then.
> 
> Perhaps I should enable XSM for my precommit build test, I don't want to
> do both (life's too short) but building XSM is more likely to be
> forgotten by contributors I think.

Sorry for the breakage, I did test XSM but only with ARM build but I
forgot to do it for x86.

I will send a fix today.

Regards,

-- 
Julien Grall