On Tue, May 7, 2013 at 4:24 AM, Eric Anholt wrote:
> Paul Berry writes:
>
>> Currently the i965 driver uses a single buffer object to hold both batch
>> buffer commands and dynamic state data structures (which are pointed to by
>> batch buffer commands). We use a "stack and heap model", where th
On 16 July 2013 17:52, Grigori Goronzy wrote:
> On 17.07.2013 02:05, Marek Olšák wrote:
>
>> No, it's not faster, but it's not slower either.
>>
>> Now that I think about it, I can't come up with a good shader-based
>> algorithm for the resolve operation.
>>
>> I don't think Christoph's approach
Hi,
The attached three patches along with this one should fix VSELECT on SI
as well.
-Tom
On Tue, Jul 16, 2013 at 05:12:40PM -0500, Aaron Watry wrote:
> Looks good to me.
>
> I've tested on Cedar (HD5400) with no OpenCL regressions, but cannot
> test on SI because SETCC still causes issues (see
Evergreen and later chipsets can sample from compressed colorbuffers.
Cayman and later chipsets cannot even decompress them. On those
chipsets, the decompression code only converts the CMASK+FMASK combo
to a texturable FMASK.
Marek
On Wed, Jul 17, 2013 at 2:52 AM, Grigori Goronzy wrote:
> On 17.
https://bugs.freedesktop.org/show_bug.cgi?id=66175
--- Comment #3 from Aaron Watry ---
Looks like it's infinitely looping in and out of getSetCCResultType in
SIISelLowering.cpp.
I'll have to read up a bit and try to figure out the next move.
--
You are receiving this mail because:
You are the
On 17.07.2013 02:05, Marek Olšák wrote:
No, it's not faster, but it's not slower either.
Now that I think about it, I can't come up with a good shader-based
algorithm for the resolve operation.
I don't think Christoph's approach that an MSAA texture can be viewed
as a larger single-sample textu
No, it's not faster, but it's not slower either.
Now that I think about it, I can't come up with a good shader-based
algorithm for the resolve operation.
I don't think Christoph's approach that an MSAA texture can be viewed
as a larger single-sample texture is correct, because the physical
locati
I forgot about the YNYN to YYNN change in this patch - sorry for that, will
re-submit in ~9 hours.
2013/7/16 Tomasz Lis
> From: Tomasz Lis
>
> glx: Float configs extension strings.
>
> The patch defines fbconfig_float in tables of extensions and enables
> reporting the extension availability.
On 17.07.2013 00:12, Grigori Goronzy wrote:
> On 16.07.2013 19:26, Marek Olšák wrote:
>> Surprisingly all drivers supporting MSAA can already do this (r300g
>> and r600g
>> for sure) and I think Christoph wanted to have this feature for his
>> Nouveau
>> drivers anyway.
>
> OK, they can do it, but
Patch 11 was a bit of a bear. You might have lost me a bit on that one. :)
Reviewed-by: Jordan Justen
On Mon, Jul 15, 2013 at 7:32 PM, Kenneth Graunke wrote:
> Hello!
>
> The GL_ARB_shading_language_420pack allows qualifiers to be specified in
> a flexible order, which is nice now that there ar
Looks good to me.
I've tested on Cedar (HD5400) with no OpenCL regressions, but cannot
test on SI because SETCC still causes issues (see
https://bugs.freedesktop.org/show_bug.cgi?id=66175). Once SETCC is
fixed for SI, we should probably add SI-CHECK lines to vselect.ll
--Aaron
On Tue, Jul 16, 2
On 16.07.2013 19:26, Marek Olšák wrote:
Surprisingly all drivers supporting MSAA can already do this (r300g and r600g
for sure) and I think Christoph wanted to have this feature for his Nouveau
drivers anyway.
OK, they can do it, but is it actually any faster than doing a resolve
and regular b
https://bugs.freedesktop.org/show_bug.cgi?id=66850
--- Comment #6 from Mikko Juola ---
As far as I can tell, it now works perfectly.
--
You are receiving this mail because:
You are the assignee for the bug.
___
mesa-dev mailing list
mesa-dev@lists.fre
https://bugs.freedesktop.org/show_bug.cgi?id=66850
--- Comment #5 from Brian Paul ---
Created attachment 82506
--> https://bugs.freedesktop.org/attachment.cgi?id=82506&action=edit
fix a few more bugs
OK, here's a patch to apply on top of the previous 3. Retest?
--
You are receiving this mai
https://bugs.freedesktop.org/show_bug.cgi?id=66850
--- Comment #4 from Mikko Juola ---
Created attachment 82505
--> https://bugs.freedesktop.org/attachment.cgi?id=82505&action=edit
Test program showing problem with glGetTexImage
After patching in these patches, it no longer crashes. However, I
Hi Emil,
I really have no problem with the patch. I was just asking.
I'll push the patch shortly.
Marek
On Tue, Jul 16, 2013 at 10:36 PM, Emil Velikov wrote:
> On 08/07/13 21:32, Marek Olšák wrote:
>> Hi Emil,
>>
>> What issue does this patch fix? Is there a nonarchaic CPU architecture
>> or a
Em 16-07-2013 16:43, Christoph Bumiller escreveu:
So far I've tried Skyrim, Civilization 5, Anno 1404 and StarCraft 2 on
the nvc0 and r600g drivers, which work pretty well, at up to x2 the fps
I get with wined3d (NOTE: no thorough benchmarking done yet).
Oblivion sucks fps wise nowdays, I wonde
On 08/07/13 21:32, Marek Olšák wrote:
> Hi Emil,
>
> What issue does this patch fix? Is there a nonarchaic CPU architecture
> or a compiler where "int" doesn't have 32 bits?
>
Pardon my arrogance, but should one consider such a (imho) trivial
patch, after things have exploded ? I do agree that it
From: Roland Scheidegger
Instead of just ignoring the srgb/linear conversions, simply call the
corresponding conversion functions, for all of pack/unpack/fetch,
both for float and unorm8 versions (though some don't make a whole
lot of sense, i.e. unorm8/unorm8 srgb/linear combinations).
Refactore
So, about two months ago I had the insane idea to pick up Joakim
Sindholt's Direct3D 9 state tracker that he'd started about 3 years ago
with the goal to make it run StarCraft 2 so I could finally play at a
reasonable frame rate ...
With help from Joakim and advice from the wine developers, as wel
From: Tom Stellard
These are really the same address space in hardware. The only
difference is that CONSTANT_ADDRESS uses a special cache for faster
access. When we are unable to use the constant kcache for some reason
(e.g. smaller types or lack of indirect addresing) then the instruction
sele
https://bugs.freedesktop.org/show_bug.cgi?id=66931
Jonathan Charest changed:
What|Removed |Added
Attachment #82485|0 |1
is obsolete|
From: Tom Stellard
---
lib/Target/R600/AMDGPUISelLowering.cpp | 3 +++
lib/Target/R600/R600ISelLowering.cpp | 3 ---
test/CodeGen/R600/vselect.ll | 30 ++
3 files changed, 33 insertions(+), 3 deletions(-)
diff --git a/lib/Target/R600/AMDGPUISelLowering.
From: Tomasz Lis
This fix splits entries in dispatch table mechanism according to
framebuffer_object EXT and ARB specs. Different bind functions
allow further diferentiation in other calls, based on binded
type of object.
This patch is based on one created by Bartosz Zawistowski.
---
src/mapi/g
glx: Require proper drawableType in init_fbconfig_for_chooser.
This patch makes sure that init_fbconfig_for_chooser sets correct
value of drawableType for visual configs and fbconfigs.
---
src/glx/glxcmds.c |7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/src/glx/glxcm
From: Tomasz Lis
glx: Float configs extension strings.
The patch defines fbconfig_float in tables of extensions and enables
reporting the extension availability.
Signed-off-by: Tomasz Lis
---
src/glx/glxextensions.c |2 ++
src/glx/glxextensions.h |6 +-
2 files changed, 7 insertio
From: Tomasz Lis
dri: Introduce new flags in __DRI_ATTRIB_RENDER_TYPE.
The patch marks __DRI_ATTRIB_FLOAT_MODE as deprecated and introduces
nrw flags to __DRI_ATTRIB_RENDER_TYPE which are to define float modes.
Both signed float (fbconfig_float) and unsigned (packed_float) is
introduced as a fla
glx: Verification of the RENDER_TYPE value.
The change is to correctly handle the value of renderType in GLX context.
In case of the value being incorrect, context creation fails.
---
src/glx/dri2_glx.c |9 +
src/glx/dri_glx.c |5 +
src/glx/drisw_glx.c
glx: Storing the RENDER_TYPE in indirect rendering.
The change allows to store RENDER_TYPE value in GLX context
when indirect rendering is in use.
---
src/glx/create_context.c |4 +++-
src/glx/glxclient.h |7 +++
src/glx/glxcmds.c|1 -
src/glx/indirect_glx.c | 29
glx: Handling RENDER_TYPE in glXCreateContext and init_fbconfig_for_chooser.
This patch sets the correct values of renderType
in glXCreateContext and init_fbconfig_for_chooser routines.
---
src/glx/glxcmds.c | 25 +
1 file changed, 21 insertions(+), 4 deletions(-)
diff
glx: Changes to visual configs initialization.
The change is to correctly handle the value of renderType and
drawableType in fbconfig. This part modifies glXInitializeVisualConfigFromTags
to read the parameter value, or detect it if it's not there.
---
src/glx/glxext.c | 37 ++
glx: Retrieve the value of RENDER_TYPE from GLX attribs array.
This change makes sure that context creation routines are provided
with the value of RENDER_TYPE retrieved from GLX attribs.
---
src/glx/dri2_glx.c |3 ++-
src/glx/dri_common.c |8
src/glx/dri_common.h |2 +-
sr
glx: Store the value of renderType while creating context.
This patch makes sure that renderType property value is stored in GLX context
while it's being created. Further patches will be provided to make the value
correspond to fbconfig's renderType.
---
src/glx/dri2_glx.c |7 +++
src/gl
On 15 July 2013 15:50, Ian Romanick wrote:
> On 07/12/2013 06:25 PM, Paul Berry wrote:
>
>> When the core profile is active, there is no fixed function fragment
>> shader functionality. However, we still need to generate a dummy
>> fragment shader program in case the back-end expects it (e.g. to
Surprisingly all drivers supporting MSAA can already do this (r300g and r600g
for sure) and I think Christoph wanted to have this feature for his Nouveau
drivers anyway.
---
I'll remove PIPE_CAP_SCALED_RESOLVE in a follow-on change if this is acked.
src/mesa/state_tracker/st_extensions.c | 1 +
exp2(0) needs to be exactly 1, if exp2(src) overflows then it has
to be equal to infinity and exp2(nan) has to be equal to a nan.
The patches fixes all three cases.
Signed-off-by: Zack Rusin
---
src/gallium/auxiliary/gallivm/lp_bld_arit.c | 16 +++-
1 file changed, 15 insertions(+)
Both D3D10 and OpenCL say that if one the inputs is nan then
the other should be returned. To preserve that behavior
the patche fixes both the sse and the non-sse paths in both
functions.
Signed-off-by: Zack Rusin
---
src/gallium/auxiliary/gallivm/lp_bld_arit.c | 60 ---
Thank you.
I will change all the VER to 0,0.
I will turn this patch into last one in the series.
Regarding indirect rendering - I didn't tested it. Please let me know if I
should change any of the Y/N switches.
2013/7/15 Ian Romanick
> On 07/15/2013 07:28 AM, Tomasz Lis wrote:
>
>> From: To
Thank you for your feedback.
I will remove GL_TRUE/FALSE and treat these as C logic values.
I will change setting of the mutually exclusive flags. While I think using
bitwise OR emphases the fact that these are bit flags and not integer
values, both ways will work.
Please make sure someone will v
https://bugs.freedesktop.org/show_bug.cgi?id=44618
Chad Versace changed:
What|Removed |Added
CC||chad.vers...@linux.intel.co
On 07/14/2013 12:57 AM, Vinson Lee wrote:
opt_return_value was not initialized if mode != ast_return.
Fixes "Uninitialized pointer field" defect reported by Coverity.
Signed-off-by: Vinson Lee
---
src/glsl/glsl_parser_extras.cpp | 1 +
1 file changed, 1 insertion(+)
Committed to master,
On 07/12/2013 05:01 PM, Vinson Lee wrote:
execinfo.h is not available on OpenBSD.
Signed-off-by: Vinson Lee
---
Committed to master, and
Reviewed-by: Chad Versace
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.or
https://bugs.freedesktop.org/show_bug.cgi?id=66886
--- Comment #1 from Chad Versace ---
>From my understanding of the EGL spec, Mesa behaves correctly.
eglChooseConfig should return false only if an error occured. And, if an error
occurred, then there is always an accompanying error emitted by e
https://bugs.freedesktop.org/show_bug.cgi?id=66931
Jonathan Charest changed:
What|Removed |Added
Attachment #82462|0 |1
is obsolete|
Looks alright be me.
Jose
- Original Message -
> From: Kyle McMartin
>
> MCJIT is the only supported LLVM JIT on AArch64 and ARM (the regular
> JIT has bit-rotted badly on ARM and doesn't exist on AArch64.)
>
> Signed-off-by: Kyle McMartin
>
> --- a/src/gallium/auxiliary/gallivm/lp_b
There is something wrong with indentation in that file, but fixing it
should be a separate commit.
I will change the patch to keep the original whitespace convention.
2013/7/15 Ian Romanick
> On 07/15/2013 07:28 AM, Tomasz Lis wrote:
>
>> From: Tomasz Lis
>>
>> The patch marks __DRI_ATTRIB_F
https://bugs.freedesktop.org/show_bug.cgi?id=66955
--- Comment #9 from Phil Armstrong ---
Sadly I was only referring to the 'dump the old libstdc++' solution to the
immediate problem of the Xserver crashing. I don't have a patch at this point.
If the reality is 'this problem is real, but will be
https://bugs.freedesktop.org/show_bug.cgi?id=66955
--- Comment #8 from Marek Olšák ---
Sorry for my hastiness. If you have a fix, that's great! Feel free to send it
to the appropriate mailing list. Thanks!
--
You are receiving this mail because:
You are the assignee for the bug.
___
https://bugs.freedesktop.org/show_bug.cgi?id=66955
--- Comment #7 from Phil Armstrong ---
(In reply to comment #5)
> The solution is simple: don't use an older libstdc++. Lots of closed source
> apps do that, which breaks them if the Mesa driver was linked against a
> newer version. There is noth
Thank you.
I will rename the function to validate_renderType_against_config and make
it return bool.
I will fix the brace.
2013/7/15 Ian Romanick
> On 07/15/2013 07:28 AM, Tomasz Lis wrote:
>
>> The change is to correctly handle the value of renderType in GLX context.
>> In case of the value b
https://bugs.freedesktop.org/show_bug.cgi?id=66955
--- Comment #6 from Marek Olšák ---
Oh and by the way, the X crash seems to be caused by indirect rendering, which
had been broken according to what Keith Packard said at XDC2012, IIRC. The
issue will be trivially resolved by nuking indirect rend
Thank you for your review.
I will add the white space.
I will remove the unused parameter code.
2013/7/15 Ian Romanick
> On 07/15/2013 07:28 AM, Tomasz Lis wrote:
>
>> The change allows to store RENDER_TYPE value in GLX context
>> when indirect rendering is in use.
>> ---
>> src/glx/create_c
On 07/15/2013 05:38 PM, Andy Li wrote:
> Hi Brian,
>
> Thank you so much for your time and help along, up to this point, the
> code make much more sense to me.
> However, I have a few more questions here, hope that you will be able to
> help me out.
>
> I am wondering
>
> Where the aa_genera
https://bugs.freedesktop.org/show_bug.cgi?id=66955
Marek Olšák changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://bugs.freedesktop.org/show_bug.cgi?id=66850
--- Comment #3 from Brian Paul ---
Created attachment 82482
--> https://bugs.freedesktop.org/attachment.cgi?id=82482&action=edit
patch 3/3
Here's a patch series to try. Let me know if this helps.
--
You are receiving this mail because:
You
https://bugs.freedesktop.org/show_bug.cgi?id=66850
--- Comment #2 from Brian Paul ---
Created attachment 82481
--> https://bugs.freedesktop.org/attachment.cgi?id=82481&action=edit
patch 2/3
--
You are receiving this mail because:
You are the assignee for the bug.
_
https://bugs.freedesktop.org/show_bug.cgi?id=66850
--- Comment #1 from Brian Paul ---
Created attachment 82480
--> https://bugs.freedesktop.org/attachment.cgi?id=82480&action=edit
patch 1/3
--
You are receiving this mail because:
You are the assignee for the bug.
_
https://bugs.freedesktop.org/show_bug.cgi?id=66955
Alex Deucher changed:
What|Removed |Added
Assignee|dri-devel@lists.freedesktop |mesa-dev@lists.freedesktop.
On 07/15/2013 12:33 PM, Ian Romanick wrote:
On 07/15/2013 07:28 AM, Tomasz Lis wrote:
[...]
diff --git a/src/mesa/drivers/x11/fakeglx.c
b/src/mesa/drivers/x11/fakeglx.c
index 969ee7d..7a2cfbe 100644
--- a/src/mesa/drivers/x11/fakeglx.c
+++ b/src/mesa/drivers/x11/fakeglx.c
I'm not sure these l
Thank you for your comments.
GCC Warnings:
There are no warnings in "glxcmds.c" during build. I think GCC only warns
when multiple logic/bitwise operators are used in a row.
Setting context renderType from config renderType:
Invalid value of renderType in configs may be caused by DRI drivers wh
r600g needs explicit flushing before DRI2 buffers are presented on the screen.
A complete implementation in all drivers will follow once this is acked.
---
src/gallium/docs/source/context.rst | 13 +
src/gallium/include/pipe/p_context.h | 13 +++
Thank you for your feedback.
I will modify all patches to have short description below email title. I
will add "glx:" to the GLX patches.
2013/7/16 Kenneth Graunke
> On 07/15/2013 01:41 PM, Ian Romanick wrote:
>
>> On 07/15/2013 07:28 AM, Tomasz Lis wrote:
>>
>>> The change is to correctly ha
https://bugs.freedesktop.org/show_bug.cgi?id=47824
Maarten Lankhorst changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
This patch implements the texture sharing.
Signed-off-by: Zhigang Gong
---
src/mesa/drivers/dri/i965/Makefile.am |2 +
src/mesa/drivers/dri/i965/Makefile.sources |1 +
.../drivers/dri/i965/intel_dri_resource_sharing.c | 192
.../drivers/dri/i96
Signed-off-by: Zhigang Gong
---
src/mesa/drivers/dri/common/dri_util.c | 48
src/mesa/drivers/dri/common/dri_util.h | 10 +++
2 files changed, 58 insertions(+)
diff --git a/src/mesa/drivers/dri/common/dri_util.c
b/src/mesa/drivers/dri/common/dri_util.c
i
OpenCL has some extension can create CL memory object from GL
textures/buffer objects/render buffers. The existing extension
can't satisfied this requirement. As we need to pass the resource
handler directly to the dri driver layer and get some low level
information back which may include the low l
This extension provides entry points for exposing internal GL
resouces to external libraries for sharing access purpose. One good
candiate of the external library is the Open CL GL extension. The
extension lets applications can acquire a low level attributes of the
given texture/buffer object/rende
67 matches
Mail list logo
