On 30/09/14 23:32, Mathias Fröhlich wrote:
Hi all,
Thanks and pushed.
Good Night!
Mathias
Thanks Mathias!
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Hi all,
Thanks and pushed.
Good Night!
Mathias
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Am 01.10.2014 00:17, schrieb Mathias Fröhlich:
>
> Brian,
>
> at least here, I get a build that runs glxgears which
> previously did not run with 3.3, 3.5.
> Currently the compile test runs with 3.6.
> If this succeeds, ok to push the attached fix
> (The same than before but with a more descripti
On 09/30/2014 04:17 PM, Mathias Fröhlich wrote:
Brian,
at least here, I get a build that runs glxgears which
previously did not run with 3.3, 3.5.
Currently the compile test runs with 3.6.
If this succeeds, ok to push the attached fix
(The same than before but with a more descriptive commit mes
Brian,
at least here, I get a build that runs glxgears which
previously did not run with 3.3, 3.5.
Currently the compile test runs with 3.6.
If this succeeds, ok to push the attached fix
(The same than before but with a more descriptive commit message)?
Greetings
Mathias>From 39a8625423f85327eef
Hi Brian,
On Tuesday, September 30, 2014 15:42:21 Brian Paul wrote:
> Yeah, reverting that patch clears up the regression here. Please go
> ahead and do the revert if you don't think you can solve the problem
> otherwise. Thanks!
I could even reproduce the segfault with the previous patch an
On 09/30/2014 03:34 PM, Mathias Fröhlich wrote:
Hi,
On Tuesday, September 30, 2014 21:25:54 Jose Fonseca wrote:
Your patch looks good AFAICT.
Not sure why the crash, and I'm afraid I won't have time to look into it.
I am currently looking into that.
I think it might help to '#define USE_MC
Hi,
On Tuesday, September 30, 2014 21:25:54 Jose Fonseca wrote:
> Your patch looks good AFAICT.
>
> Not sure why the crash, and I'm afraid I won't have time to look into it.
I am currently looking into that.
> I think it might help to '#define USE_MCJIT 1' for now, ie, enable MCJIT for
> all L
builds/tests on master.
Jose
From: Brian Paul
Sent: 30 September 2014 21:47
To: Mathias Fröhlich; mesa-dev@lists.freedesktop.org
Cc: Jose Fonseca
Subject: Re: [Mesa-dev] [PATCH] gallivm: fix build for LLVM 3.2
On 09/30/2014 02:40 PM, Mathias Fröhlich wrote:
>
> Hi
On 09/30/2014 02:40 PM, Mathias Fröhlich wrote:
Hi Brian,
On Tuesday, September 30, 2014 13:30:21 Brian Paul wrote:
Move the USE_MCJIT / HAVE_AVX determination logic to lp_bld.h. If we
don't have MCJIT define a dummy LLVMMCJITMemoryManagerRef type to avoid
excessive #ifdef testing elsewhere.
Hi Brian,
On Tuesday, September 30, 2014 13:30:21 Brian Paul wrote:
> Move the USE_MCJIT / HAVE_AVX determination logic to lp_bld.h. If we
> don't have MCJIT define a dummy LLVMMCJITMemoryManagerRef type to avoid
> excessive #ifdef testing elsewhere.
[...]
> @@ -219,7 +188,7 @@ gallivm_free_code
Move the USE_MCJIT / HAVE_AVX determination logic to lp_bld.h. If we
don't have MCJIT define a dummy LLVMMCJITMemoryManagerRef type to avoid
excessive #ifdef testing elsewhere.
---
src/gallium/auxiliary/gallivm/lp_bld.h | 40 +++
src/gallium/auxiliary/gallivm/lp_bld
12 matches
Mail list logo