[prev in list] [next in list] [prev in thread] [next in thread] 

List:       mesa3d-announce
Subject:    Re: [Mesa-announce] [Mesa-dev]  Mesa 10.3 release candidate 1
From:       Matt Turner <mattst88 () gmail ! com>
Date:       2014-11-07 9:19:21
Message-ID: CAEdQ38HS_wr+UiWjJvtA13_gP55U9U9iyve5sYXtNj0F6aA41g () mail ! gmail ! com
[Download RAW message or body]

On Fri, Nov 7, 2014 at 1:07 AM, Thierry Vignaud
<thierry.vignaud@gmail.com> wrote:
> On 5 November 2014 04:44, Matt Turner <mattst88@gmail.com> wrote:
>>>>>I tried to reproduce this today and couldn't.
>
> (...)
>
>>>> Thanks. Maybe you could give a little more information, like an error
>>>> message or something?
>>>
>>> Same error as Thierry reported in this thread in August:
>>
>> Unfortunately Thierry's was from a re-run of make, so it wasn't useful.
>
> No It wasn't a re-run!
> It was a clean build in our build system with make -jXX with XX auto set to
> the number of cores and is always reproducable given enough cores

Oh, weird.

>> I've gone over this all and can't spot the problem. The dependencies
>> look fine. I tried automake-1.13 and 1.14, and make-3.82 and 4.0.
>> Maybe I'll have more luck on a 40 core system.
>
> As already explained, in order to be able to reproduce, you must either have
> a large system or force make -j to a high value (eg: -j24)

Did you see the rest of the thread where I said I couldn't reproduce
on a 40 core system?

Perhaps someone who can reproduce could try to take a look?
_______________________________________________
mesa-announce mailing list
mesa-announce@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-announce

[prev in list] [next in list] [prev in thread] [next in thread]