On 08/21/2013 03:31 PM, Ian Romanick wrote:
On 08/17/2013 10:42 AM, Henri Verbeet wrote:
This was originally introduced by commit
ba47aabc9868b410cdfe3bc8b6d25a44a598cba2, but unfortunately the commit
message
doesn't go into much detail about why +INF would be a problem here. I
don't
see anythin
Now that we use a fixed set of register classes, we can set up the
register set and conflict graphs once, at context creation, rather than
on every VS compile. This is obviously less expensive, and also what
we already do in the FS backend.
Signed-off-by: Kenneth Graunke
Cc: Paul Berry
---
src
For now, nothing else can get allocated over them. That may change at
some point in the future.
This also means that base_reg_count can be computed without knowing the
number of registers used for the payload, which is required if we want
to allocate the register set once at context creation time
We're soon going to be calling brw_alloc_reg_set() from outside of the
visitor, where we don't have the precomputed "max_grf" variable handy.
Signed-off-by: Kenneth Graunke
Cc: Paul Berry
---
src/mesa/drivers/dri/i965/brw_vec4_reg_allocate.cpp | 7 ---
1 file changed, 4 insertions(+), 3 del
Arrays, structures, and matrices use large VGRFs of arbitrary sizes.
However, split_virtual_grfs() breaks those down into VGRFs of size 1.
For reference, commit 5d90b988791e51cfb6413109271ad102fd7a304c is the
analogous change to the FS backend.
Signed-off-by: Kenneth Graunke
Cc: Paul Berry
---
On Wed, Aug 21, 2013 at 9:14 PM, Ian Romanick wrote:
> ping
Committed yesterday to master. Should I cherry-pick it to 9.2 now?
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 08/21/2013 11:09 AM, Chad Versace wrote:
Fixes inconsistent failure of gles2conform/GL2Tests/glUniform/glUniform.test
under gnome-shell. What follows is a description of the bug and its fix.
When intel_update_renderbuffers() allocates a miptree for a winsys
renderbuffer, it propagates the ren
On 08/21/2013 08:47 PM, Maxence Le Doré wrote:
As ARB_seamless_cubemap_per_texture is word-to-word same as
AMD_seamless_cubemap_per_texture and this last already implemented we
can enable the ARB extension. This patch is a candidate for it :
They're actually a tiny bit different. From the ARB
On 08/21/2013 08:46 PM, Maxence Le Doré wrote:
As ARB_seamless_cubemap_per_texture is word-to-word same as
AMD_seamless_cubemap_per_texture and this last already implemented we
can enable the ARB extension. This patch is a candidate for it :
Since the extensions are identical, we should expose
ping
On 08/06/2013 11:09 AM, Ross Burton wrote:
The rules were writing files to e.g. util/u_indices_gen.py, but in an
out-of-tree build this directory doesn't exist in the build directory. So,
create the directories just in case.
NOTE: This is a candidate for the stable branches.
Cc: mesa-sta
With my change to patch 4 (lol), the series is:
Reviewed-by: Ian Romanick
On 08/15/2013 11:27 AM, Matt Turner wrote:
The variable means that UBO qualifiers are allowed in a particular
context (e.g., not allowed in a struct field declaration), rather than a
particular set of UBO qualifiers are
On 08/21/2013 09:02 PM, Ian Romanick wrote:
On 08/15/2013 11:27 AM, Matt Turner wrote:
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=59648
---
src/glsl/ast_to_hir.cpp | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_t
On 08/21/2013 04:22 PM, Chad Versace wrote:
On 08/21/2013 11:57 AM, Ian Romanick wrote:
On 08/21/2013 11:09 AM, Chad Versace wrote:
Fixes inconsistent failure of
gles2conform/GL2Tests/glUniform/glUniform.test
under gnome-shell. What follows is a description of the bug and its fix.
When intel_u
On 08/15/2013 11:27 AM, Matt Turner wrote:
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=59648
---
src/glsl/ast_to_hir.cpp | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_to_hir.cpp
index c2fdbd5..f35c11f 100644
--- a
Your mail client line wrapped the patch, so it can't be applied.
Please resend with git send-email.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
As ARB_seamless_cubemap_per_texture is word-to-word same as
AMD_seamless_cubemap_per_texture and this last already implemented we
can enable the ARB extension. This patch is a candidate for it :
>From eb2cf312a7c7ba70f22f8eb8d66aab8a1d78b6d9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Maxence=20Le=20
As ARB_seamless_cubemap_per_texture is word-to-word same as
AMD_seamless_cubemap_per_texture and this last already implemented we
can enable the ARB extension. This patch is a candidate for it :
>From eb2cf312a7c7ba70f22f8eb8d66aab8a1d78b6d9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Maxence=20Le=20
From: Ian Romanick
SEQ and SNE are not native i915 instructions, so they each generate at
least 3 instructions. If both operands are uniforms or constants, we
get 5 instructions like:
U[1] = MOV CONST[1]
U[0].xyz = SGE CONST[0]., U[1]
U[1] = M
From: Roland Scheidegger
Just a copy & paste error.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=68409.
Note that the test passing before probably simply means it doesn't verify
clamping of the border color itself as required by the OpenGL spec.
---
src/gallium/auxiliary/gallivm/lp_bld_sam
https://bugs.freedesktop.org/show_bug.cgi?id=68409
Priority: medium
Bug ID: 68409
Keywords: regression
CC: srol...@vmware.com
Assignee: mesa-dev@lists.freedesktop.org
Summary: [llvmpipe] piglit texwrap GL_ARB_texture_rg-int
From: Roland Scheidegger
block size depth is always 1 even for compressed formats (unless someone
invents true 3d compressed formats at least which we can't represent).
Nearest (and soa) path had it right.
---
src/gallium/auxiliary/gallivm/lp_bld_sample_aos.c |4 ++--
1 file changed, 2 inser
From: Roland Scheidegger
Except for explicit derivs with cube maps which are very bogus anyway.
Just like explicit lod this is only used if no_quad_lod is set in
GALLIVM_DEBUG env var.
Minification is terrible on cpus which don't support true vector shifts
(but should work correctly). Cannot do t
Fixes a bug where if an uniform array is passed to a function the accesses
to the array are not propagated so later all but the first vector of the
uniform array are removed in parcel_out_uniform_storage resulting in
broken shaders and out of bounds access to arrays in
brw::vec4_visitor::pack_unifo
On Wed, Aug 21, 2013 at 5:19 PM, Marek Olšák wrote:
> I think the correct solution is to make the two conflicting libs
> mutually exclusive. If both are enabled in the build at the same time,
> then let's print a configure error. I don't think installing a lib to
> a non-standard directory like /u
I think the correct solution is to make the two conflicting libs
mutually exclusive. If both are enabled in the build at the same time,
then let's print a configure error. I don't think installing a lib to
a non-standard directory like /usr/lib/gallium would help anybody.
Marek
On Thu, Aug 22, 20
On 08/21/2013 05:03 PM, Matt Turner wrote:
On Wed, Aug 21, 2013 at 7:06 AM, Brian Paul wrote:
On 08/20/2013 03:18 PM, Matt Turner wrote:
The Gallium implementation is apparently not ready for regular
consumption, so as much as I hate adding more build-time options, here's
another.
The inte
Le mercredi 21 août 2013 14:00:12 Tom Stellard a écrit :
> From: Tom Stellard
>
> This fixes a crash in Unigine Tropics.
> ---
This fixes the following failure and now Tropics woks fine for me:
Tropics: SIInstrInfo.cpp:94: virtual void
llvm::SIInstrInfo::copyPhysReg(llvm::MachineBasicBlock&,
ll
On 08/21/2013 11:57 AM, Ian Romanick wrote:
On 08/21/2013 11:09 AM, Chad Versace wrote:
Fixes inconsistent failure of gles2conform/GL2Tests/glUniform/glUniform.test
under gnome-shell. What follows is a description of the bug and its fix.
When intel_update_renderbuffers() allocates a miptree for
On Wed, Aug 21, 2013 at 7:06 AM, Brian Paul wrote:
> On 08/20/2013 03:18 PM, Matt Turner wrote:
>>
>> The Gallium implementation is apparently not ready for regular
>> consumption, so as much as I hate adding more build-time options, here's
>> another.
>>
>
> The intention was for the old/swrast v
On 08/17/2013 10:42 AM, Henri Verbeet wrote:
This was originally introduced by commit
ba47aabc9868b410cdfe3bc8b6d25a44a598cba2, but unfortunately the commit message
doesn't go into much detail about why +INF would be a problem here. I don't
see anything in the spec that would allow 1.0f here.
S
On 08/19/2013 05:05 PM, Ian Romanick wrote:
Mesa 9.2 release candidate 1 is now available for testing.
The tag in the GIT repository for Mesa 9.2-rc1 is 'mesa-9.2-rc1'.
Mesa 9.2 release candidate 1 is available for download at
ftp://freedesktop.org/pub/mesa/9.2/
md5sums:
866e9a1b3ce72b822671e
I'm not sure about the lit test, but this definitely made it much more
obvious to me what was failing in my VP8 decoder on R600 in OpenCL.
Old:
Stack dump:
0.Running pass 'Function Pass Manager' on module 'radeon'.
1.Running pass 'R600 Texture Intrinsics Replacer' on function
'@vp8_loop_fi
On 08/21/2013 11:09 AM, Chad Versace wrote:
Fixes inconsistent failure of gles2conform/GL2Tests/glUniform/glUniform.test
under gnome-shell. What follows is a description of the bug and its fix.
When intel_update_renderbuffers() allocates a miptree for a winsys
renderbuffer, it propagates the ren
From: Tom Stellard
This pass was segfaulting when it ran into a non-intrinsic function
call. Function calls are not supported, so now instead of segfaulting,
we will get an assertion failure with a nice error message.
I'm not sure how to test this using lit.
---
lib/Target/R600/R600TextureIntr
Hi,
The attached patches fix some LDS bugs on SI and add support for atomic
add for R600 and SI.
Please Review.
-Tom
>From 0447b0918efca9000e66414e9ff1bd291936f702 Mon Sep 17 00:00:00 2001
From: Tom Stellard
Date: Wed, 21 Aug 2013 11:35:27 -0400
Subject: [PATCH 1/4] R600/SI: Don't emit S_WQM_B6
Fixes inconsistent failure of gles2conform/GL2Tests/glUniform/glUniform.test
under gnome-shell. What follows is a description of the bug and its fix.
When intel_update_renderbuffers() allocates a miptree for a winsys
renderbuffer, it propagates the renderbuffer's format to become also the
miptree'
On 08/20/2013 11:30 AM, Paul Berry wrote:
[snip]
diff --git a/src/mesa/drivers/dri/i965/brw_eu.h
b/src/mesa/drivers/dri/i965/brw_eu.h
index ae4cab5..9053ea2 100644
--- a/src/mesa/drivers/dri/i965/brw_eu.h
+++ b/src/mesa/drivers/dri/i965/brw_eu.h
@@ -252,6 +252,12 @@ enum brw_urb_write_flags {
On 08/20/2013 11:30 AM, Paul Berry wrote:
---
src/mesa/drivers/dri/i965/brw_defines.h | 9 +
src/mesa/drivers/dri/i965/brw_eu.h | 6 ++
src/mesa/drivers/dri/i965/brw_eu_emit.c | 4 ++--
src/mesa/drivers/dri/i965/brw_shader.cpp| 5 -
src/mesa/drivers/
From: Tom Stellard
This fixes a crash in Unigine Tropics.
---
lib/Target/R600/SIFixSGPRCopies.cpp | 2 +-
test/CodeGen/R600/sgpr-copy.ll | 145 +++-
2 files changed, 143 insertions(+), 4 deletions(-)
diff --git a/lib/Target/R600/SIFixSGPRCopies.cpp
b/lib/
On 08/21/2013 07:54 AM, Paul Berry wrote:
On 20 August 2013 11:58, Kenneth Graunke mailto:kenn...@whitecape.org>> wrote:
On 08/20/2013 11:30 AM, Paul Berry wrote:
---
src/mesa/drivers/dri/i965/brw___program.h | 8
src/mesa/drivers/dri/i965/brw___shade
Am 21.08.2013 18:30, schrieb Jose Fonseca:
> Series looks good.
>
> You might want to promote XXX to FIXME. Fixing wouldn't be hard --
> just need to pass some additional parameters to this function, so
> that we can check that MIN/MAG filter are same.
Ok I'll make it FIXME. I thought it would cau
From: Michel Dänzer
They are defined as constant 0.0/0.0/1.0.
Three more little piglits.
Cc: mesa-sta...@lists.freedesktop.org
Signed-off-by: Michel Dänzer
---
src/gallium/drivers/radeonsi/radeonsi_shader.c | 18 ++
1 file changed, 18 insertions(+)
diff --git a/src/gallium/dr
Series looks good.
You might want to promote XXX to FIXME. Fixing wouldn't be hard -- just need to
pass some additional parameters to this function, so that we can check that
MIN/MAG filter are same.
Jose
- Original Message -
> From: Roland Scheidegger
>
> Detected this hunting some
On 21.8.2013 18:07, Brian Paul wrote:
On 08/21/2013 08:44 AM, Armin K wrote:
---
src/mesa/drivers/osmesa/Makefile.am | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/mesa/drivers/osmesa/Makefile.am
b/src/mesa/drivers/osmesa/Makefile.am
index a1c0bb5..d09c18e 100644
--- a/src/mesa/drive
On 08/21/2013 08:44 AM, Armin K wrote:
---
src/mesa/drivers/osmesa/Makefile.am | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/mesa/drivers/osmesa/Makefile.am
b/src/mesa/drivers/osmesa/Makefile.am
index a1c0bb5..d09c18e 100644
--- a/src/mesa/drivers/osmesa/Makefile.am
+++ b/src/mesa/d
Am 21.08.2013 17:50, schrieb jfons...@vmware.com:
> From: José Fonseca
>
> Same as PIPE_FORMAT_B10G10R10A2_UINT but without the swizzling.
> ---
> src/gallium/auxiliary/gallivm/lp_bld_sample_soa.c | 1 +
> src/gallium/auxiliary/util/u_format.csv | 1 +
> src/gallium/include/pipe/p_form
On 20 August 2013 14:04, Kenneth Graunke wrote:
> On 08/20/2013 11:30 AM, Paul Berry wrote:
>
>> Previously, we would always use the same push constant allocation
>> regardless of what shader programs were being run: the available push
>> constant space was split into 2 equal size partitions, one
From: José Fonseca
Same as PIPE_FORMAT_B10G10R10A2_UINT but without the swizzling.
---
src/gallium/auxiliary/gallivm/lp_bld_sample_soa.c | 1 +
src/gallium/auxiliary/util/u_format.csv | 1 +
src/gallium/include/pipe/p_format.h | 1 +
3 files changed, 3 insertions(+)
diff
From: Roland Scheidegger
Was using wrong (undefined) vector element (the elements are at 0/2 position,
not 0/1).
---
src/gallium/auxiliary/gallivm/lp_bld_sample.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/auxiliary/gallivm/lp_bld_sample.c
b/src/gallium/a
From: Roland Scheidegger
Detected this hunting some other bug, not sure if it really needs fixing but
it is definitely wrong.
---
src/gallium/auxiliary/gallivm/lp_bld_sample.c |8
src/gallium/auxiliary/gallivm/lp_bld_sample_aos.c |2 +-
src/gallium/auxiliary/gallivm/lp_bld_s
On 20 August 2013 11:58, Kenneth Graunke wrote:
> On 08/20/2013 11:30 AM, Paul Berry wrote:
>
>> ---
>> src/mesa/drivers/dri/i965/brw_**program.h | 8
>> src/mesa/drivers/dri/i965/brw_**shader.cpp | 2 +-
>> src/mesa/drivers/dri/i965/brw_**shader.h | 7 ++-
>> src/mesa/dri
On 08/21/2013 10:06 AM, Brian Paul wrote:
> On 08/20/2013 03:18 PM, Matt Turner wrote:
>> The Gallium implementation is apparently not ready for regular
>> consumption, so as much as I hate adding more build-time options, here's
>> another.
>>
>
> The intention was for the old/swrast version of OS
---
src/mesa/drivers/osmesa/Makefile.am | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/mesa/drivers/osmesa/Makefile.am
b/src/mesa/drivers/osmesa/Makefile.am
index a1c0bb5..d09c18e 100644
--- a/src/mesa/drivers/osmesa/Makefile.am
+++ b/src/mesa/drivers/osmesa/Makefile.am
@@ -56,6 +56,7 @@
On 08/21/2013 04:06 PM, Brian Paul wrote:
> On 08/20/2013 03:18 PM, Matt Turner wrote:
>> The Gallium implementation is apparently not ready for regular
>> consumption, so as much as I hate adding more build-time options, here's
>> another.
>>
>
> The intention was for the old/swrast version of OS
On 20 August 2013 13:27, Chris Forbes wrote:
> Ken,
>
> I think the big win from using pass-by-pointer is that you can see
> what's happening at the call site. References' value syntax and
> pointer semantics is a weird mix.
>
> But that might just be the C programmer talking, as you suggested.
>
Am 21.08.2013 02:05, schrieb Roland Mainz:
> On Wed, Aug 21, 2013 at 1:45 AM, Roland Scheidegger
> wrote:
>> Am 21.08.2013 01:28, schrieb Roland Mainz:
>>> On Wed, Aug 21, 2013 at 1:12 AM, wrote:
From: Roland Scheidegger
> [snip]
>>> Can you check whether the new code supports -nan correc
On 08/20/2013 03:18 PM, Matt Turner wrote:
The Gallium implementation is apparently not ready for regular
consumption, so as much as I hate adding more build-time options, here's
another.
The intention was for the old/swrast version of OSMesa would wind up in
lib/ while the gallium version wo
https://bugs.freedesktop.org/show_bug.cgi?id=68380
Priority: medium
Bug ID: 68380
Assignee: mesa-dev@lists.freedesktop.org
Summary: flink race on wayland
Severity: normal
Classification: Unclassified
OS: All
Reporte
Reviewed and committed both patches.
Thanks for the help,
Christian.
Am 21.08.2013 10:06, schrieb Rico Schüller:
---
src/gallium/state_trackers/vdpau/query.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
___
mesa-dev mailing list
me
---
src/gallium/state_trackers/vdpau/decode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/vdpau/decode.c b/src/gallium/state_trackers/vdpau/decode.c
index 684ded8..47ca229 100644
--- a/src/gallium/state_trackers/vdpau/decode.c
+++ b/src/gallium/
---
src/gallium/state_trackers/vdpau/query.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/vdpau/query.c b/src/gallium/state_trackers/vdpau/query.c
index 72b1fe9..1d35252 100644
--- a/src/gallium/state_trackers/vdpau/query.c
+++ b/src/galliu
61 matches
Mail list logo
