On 2015-06-24 07:36:24, Iago Toral wrote:
> On Wed, 2015-06-24 at 15:43 +0300, Francisco Jerez wrote:
> > AFAICT the reason why this (and many of the other changes in GLSL
> > optimization passes) is needed is because SSBO loads have been
> > implemented as ir_expression nodes instead of being lowe
Before considering radeonsi, I felt llvmpipe support would
be a good stepping stone.
Things are messy with SoA, and I've no idea how test AoS,
so I left it alone.
but this set does work and does pass the tests and doesn't regress.
Dave.
___
mesa-dev m
Signed-off-by: Dave Airlie
---
src/gallium/auxiliary/tgsi/tgsi_info.c | 37 ++
1 file changed, 37 insertions(+)
diff --git a/src/gallium/auxiliary/tgsi/tgsi_info.c
b/src/gallium/auxiliary/tgsi/tgsi_info.c
index 9295311..4b16ef3 100644
--- a/src/gallium/auxiliary/
This adds support for ARB_gpu_shader_fp64 and ARB_vertex_attrib_64bit to
llvmpipe.
Two things that don't mix well are SoA and doubles, see
emit_fetch_double, and emit_store_double_chan in this.
I've also had to split emit_data.chan, to add src_chan,
which can be different for doubles.
Open issue
From: Dave Airlie
Signed-off-by: Dave Airlie
---
docs/GL3.txt | 4 ++--
docs/relnotes/10.7.0.html | 2 ++
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/docs/GL3.txt b/docs/GL3.txt
index 54e4574..ce3b134 100644
--- a/docs/GL3.txt
+++ b/docs/GL3.txt
@@ -109,7 +109,7
Signed-off-by: Tapani Pälli
---
src/mesa/drivers/dri/i915/i915_context.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/mesa/drivers/dri/i915/i915_context.c
b/src/mesa/drivers/dri/i915/i915_context.c
index 42ea54e..57b033c 100644
--- a/src/mesa/drivers/dri/i915/i915_context.c
+++ b/s
Signed-off-by: Tapani Pälli
---
src/mesa/drivers/dri/i965/brw_shader.cpp | 4
1 file changed, 4 insertions(+)
diff --git a/src/mesa/drivers/dri/i965/brw_shader.cpp
b/src/mesa/drivers/dri/i965/brw_shader.cpp
index 32c4013..3e3d78b 100644
--- a/src/mesa/drivers/dri/i965/brw_shader.cpp
+++ b/
Hi;
Here's additions to patches I sent earlier (reviewed by curro), these just
set EmitNoIndirectSampler on when there's no ARB_gpu_shader5 support liked
discussed.
Original mail thread:
http://lists.freedesktop.org/archives/mesa-dev/2015-June/086049.html
All patches applied on master:
http://cg
Signed-off-by: Tapani Pälli
---
src/mesa/state_tracker/st_context.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/src/mesa/state_tracker/st_context.c
b/src/mesa/state_tracker/st_context.c
index ed9ed0f..62a0fbe 100644
--- a/src/mesa/state_tracker/st_context.c
+++ b/src/mesa/state_trac
On Mon, 2015-06-29 at 00:11 -0700, Jordan Justen wrote:
> On 2015-06-24 07:36:24, Iago Toral wrote:
> > On Wed, 2015-06-24 at 15:43 +0300, Francisco Jerez wrote:
> > > AFAICT the reason why this (and many of the other changes in GLSL
> > > optimization passes) is needed is because SSBO loads have b
From: Christian König
We need to check what the 3D pipe is able to handle for the mixer, not what
the decoder is able to decode. This fixes output of resolutions like 720x1280.
Signed-off-by: Christian König
CC: mesa-sta...@lists.freedesktop.org
---
src/gallium/state_trackers/vdpau/mixer.c | 2
From: Christian König
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=90728
Signed-off-by: Christian König
CC: mesa-sta...@lists.freedesktop.org
---
src/gallium/auxiliary/vl/vl_mpeg12_decoder.c | 24
src/gallium/auxiliary/vl/vl_mpeg12_decoder.h | 4
2 files c
Mesa 10.6.1 is now available.
This release includes core mesa and glsl patches (amonst which
a Dota2 Reborn bugfix) affecting all dri drivers, nouveau specific
fixes and a selection of shared-glapi commits - from build fixes, to
ones ensuring that the dri modules can be loaded.
Anuj Phogat (4):
These checks were in Mesa prior to commit fbba25bba, but they were
not necessary for the purpose that Mesa intended (check if we could
resolve ReadPixels via memcpy), so that commit took them away.
Unfortunately, it seems that some Gallium drivers rely on these
checks to make the decision of wheth
Tapani Pälli writes:
> Hi;
>
> Here's additions to patches I sent earlier (reviewed by curro), these just
> set EmitNoIndirectSampler on when there's no ARB_gpu_shader5 support liked
> discussed.
>
> Original mail thread:
> http://lists.freedesktop.org/archives/mesa-dev/2015-June/086049.html
>
>
Reviewed-by: Marek Olšák
Marek
On Mon, Jun 29, 2015 at 9:21 AM, Dave Airlie wrote:
> Signed-off-by: Dave Airlie
> ---
> src/gallium/auxiliary/tgsi/tgsi_info.c | 37
> ++
> 1 file changed, 37 insertions(+)
>
> diff --git a/src/gallium/auxiliary/tgsi/tgsi_info.c
Davin McCall writes:
> On 26/06/15 14:53, Francisco Jerez wrote:
>
>> [...]
>>
>> Your first approach seemed quite reasonable IMHO. Were you able to
>> measure any performance regression from it?
>>
>> Thanks.
>>
>
> When I run an apitrace replay of a Dota 2 trace [1] with
> LIBGL_ALWAYS_SOFTWA
Reviewed-by: Marek Olšák
Marek
On Mon, Jun 29, 2015 at 9:35 AM, Tapani Pälli wrote:
> Signed-off-by: Tapani Pälli
> ---
> src/mesa/state_tracker/st_context.c | 5 +
> 1 file changed, 5 insertions(+)
>
> diff --git a/src/mesa/state_tracker/st_context.c
> b/src/mesa/state_tracker/st_contex
Reviewed-by: Marek Olšák
The fix for amdgpu should be in libdrm_amdgpu / amdgpu_device.c.
Marek
On Sun, Jun 28, 2015 at 7:49 AM, Mario Kleiner
wrote:
> Thanks for the review Ilia. Just saw that the new amdgpu winsys driver will
> need the same fix, once this one is confirmed to be ok.
>
> than
On 29/06/15 10:40, Francisco Jerez wrote:
Davin McCall writes:
On 26/06/15 14:53, Francisco Jerez wrote:
[...]
Your first approach seemed quite reasonable IMHO. Were you able to
measure any performance regression from it?
Thanks.
When I run an apitrace replay of a Dota 2 trace [1] with
Davin McCall writes:
> On 29/06/15 10:40, Francisco Jerez wrote:
>> Davin McCall writes:
>>
>>> On 26/06/15 14:53, Francisco Jerez wrote:
>>>
[...]
Your first approach seemed quite reasonable IMHO. Were you able to
measure any performance regression from it?
Thanks
Instead of using symbol table, build mask by inspecting IR. This
change is required by further patches to move resource list creation
to happen later when symbol table does not exist anymore.
Signed-off-by: Tapani Pälli
---
src/glsl/linker.cpp | 11 ---
1 file changed, 8 insertions(+), 3
These changes move program resource list creation to happen later
after backend LinkShader hook has been called. So, there are no changes
in overall logic but functionality is called only after LinkShader that
is still able to remove some variables considered dead.
No Piglit regressions observed.
Resource list can be created properly only after LinkShader hook
has been called to make sure all dead variables have been removed.
Signed-off-by: Tapani Pälli
https://bugs.freedesktop.org/show_bug.cgi?id=90925
---
src/glsl/linker.cpp | 4
src/mesa/program/ir_to_mesa.cpp | 2 ++
This is required so that we can move resource list creation
to happen later.
Signed-off-by: Tapani Pälli
---
src/glsl/linker.cpp | 2 +-
src/glsl/program.h | 4
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/glsl/linker.cpp b/src/glsl/linker.cpp
index 9be48f0..abf7391 10
On 29/06/15 13:26, Francisco Jerez wrote:
Davin McCall writes:
On 29/06/15 10:40, Francisco Jerez wrote:
Davin McCall writes:
On 26/06/15 14:53, Francisco Jerez wrote:
[...]
Your first approach seemed quite reasonable IMHO. Were you able to
measure any performance regression from it?
On 29/06/15 15:35, Tapani Pälli wrote:
Instead of using symbol table, build mask by inspecting IR. This
change is required by further patches to move resource list creation
to happen later when symbol table does not exist anymore.
Signed-off-by: Tapani Pälli
---
src/glsl/linker.cpp | 11 +
Ping
On Fri, Jun 26, 2015 at 11:01 AM, Marek Olšák wrote:
> From: Marek Olšák
>
> No states are initialized at that point yet.
>
> Cc: 10.5 10.6
> ---
> src/mesa/state_tracker/st_atom_constbuf.c | 14 --
> src/mesa/state_tracker/st_atom_sampler.c | 30 +
This just seems super-hacky. Given that we set _MaintainTexEnvProgram
it seems reasonable to assume there will always be one there. IMHO the
core should be fixed to not allow this situation to occur in the first
place...
On Mon, Jun 29, 2015 at 10:08 AM, Marek Olšák wrote:
> Ping
>
> On Fri, Jun
I can imagine an app calling glBlitFramebuffer only. Why would the
driver generate a fixed-func fragment program then?
Marek
On Mon, Jun 29, 2015 at 4:10 PM, Ilia Mirkin wrote:
> This just seems super-hacky. Given that we set _MaintainTexEnvProgram
> it seems reasonable to assume there will alwa
From: Marek Olšák
The reason might be that no commands have been submitted before the flush
and the GPU is idle.
---
src/mesa/state_tracker/st_cb_syncobj.c | 14 +-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/src/mesa/state_tracker/st_cb_syncobj.c
b/src/mesa/state_
From: Marek Olšák
Absolute timeouts are used with the amdgpu kernel driver.
It also makes waiting for several variables and fences at the same time
easier (the timeout doesn't have to be recalculated after every wait call).
---
src/gallium/auxiliary/os/os_time.c | 41
Don't worry about the AoS stuff. Only meant to do simple things.
Looks good overall, I guess it makes sense to not split execution too
(so you'd have native hw vector size there), llvm should handle that
pretty well these days (the sse intrinsics won't get used that way
probably (though there's a
ping?
2015-06-21 23:06 GMT+08:00 Boyan Ding :
> This series of patches add support for EGL_KHR_create_context when
> using swrast. The first patch moves context_attrib filling into a
> separate function for reuse in swrast. The second one uses
> createContextAttribs in swrast extension and the th
In commit c61bc6ed8 ("util: port _mesa_strto[df] to C") we renamed
strtod.cpp -> strtod.c.
Applying this patch on top of a tree that has already built strtod.cpp
will yield the error in the subject. To fix, simply run
sed -i -e 's/strtod\.cpp/strtod.c/' src/util/.deps/libmesautil_la-strtod.Plo
i
Thanks!
Pushed, and sent a note describing how to fix the problems caused by
renaming strtod.cpp -> strtod.c.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Mon, Jun 29, 2015 at 6:40 PM, Matt Turner wrote:
> Thanks!
>
> Pushed, and sent a note describing how to fix the problems caused by
> renaming strtod.cpp -> strtod.c.
Thanks a lot :)
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lis
---
src/mesa/main/context.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/src/mesa/main/context.c b/src/mesa/main/context.c
index fdef412..faa1de7 100644
--- a/src/mesa/main/context.c
+++ b/src/mesa/main/context.c
@@ -377,13 +377,12 @@ one_time_init( struct gl_c
It couldn't have worked anyway. There were calls to undefined functions.
---
src/glx/packrender.h | 46 --
src/glx/packsingle.h | 56
2 files changed, 102 deletions(-)
diff --git a/src/glx/packrender.h b/
On Thu, Jun 25, 2015 at 5:45 AM, Topi Pohjolainen
wrote:
> and simplify the interface to take directly the size and to return
> the offset. The routine does nothing more than allocate, it doesn't
> upload anything.
>
> CC: Kenneth Graunke
> Signed-off-by: Topi Pohjolainen
> ---
> src/mesa/drive
On Thu, Jun 25, 2015 at 5:45 AM, Topi Pohjolainen
wrote:
> Current logic re-writes the same data when existing data is found.
> Not that this actually matters at the moment in practice, the
> contraint for finding matching data is too severe to ever allow
> data to be shared between two items in t
On Fri, Jun 26, 2015 at 01:23:41PM -0700, Anuj Phogat wrote:
> On Mon, Jun 22, 2015 at 5:23 PM, Anuj Phogat wrote:
> > On Mon, Jun 22, 2015 at 2:53 PM, Ben Widawsky wrote:
> >> On Wed, Jun 10, 2015 at 03:30:47PM -0700, Anuj Phogat wrote:
> >>> Buffers with Yf/Ys tiling end up using meta upload /
On 29/06/15 13:26, Francisco Jerez wrote:
Davin McCall writes:
On 29/06/15 10:40, Francisco Jerez wrote:
Davin McCall writes:
On 26/06/15 14:53, Francisco Jerez wrote:
[...]
Your first approach seemed quite reasonable IMHO. Were you able to
measure any performance regression from it?
On Mon 29 Jun 2015, Matt Turner wrote:
> ---
> src/mesa/main/context.c | 13 ++---
> 1 file changed, 6 insertions(+), 7 deletions(-)
Reviewed-by: Chad Versace
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.or
Jason Ekstrand writes:
> Reviewed-by: Iago Toral Quiroga
> Reviewed-by: Topi Pohjolainen
> ---
> src/mesa/drivers/dri/i965/brw_fs.cpp | 6 ++
> 1 file changed, 6 insertions(+)
>
> diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp
> b/src/mesa/drivers/dri/i965/brw_fs.cpp
> index 589b74c..6
On Thu, Jun 25, 2015 at 5:45 AM, Topi Pohjolainen
wrote:
> Items in the program cache consist of three things: key, the data
> representing the instructions and auxiliary data representing
> uniform storage. The data consisting of instructions is stored into
> a drm buffer object while the key and
Kenneth Graunke writes:
> On Wednesday, June 03, 2015 09:21:11 PM Topi Pohjolainen wrote:
>> Shader programs are cached per stage (FS, VS, GS) using the
>> corresponding shader source identifier and compile time choices
>> as key. However, one not only stores the program binary but
>> a pair cons
On 30 June 2015 at 00:58, Roland Scheidegger wrote:
> Don't worry about the AoS stuff. Only meant to do simple things.
>
> Looks good overall, I guess it makes sense to not split execution too
> (so you'd have native hw vector size there), llvm should handle that
> pretty well these days (the sse
On Friday, June 26, 2015 10:06:18 AM Eduardo Lima Mitev wrote:
> The NIR->vec4 pass will be activated if ALL the following conditions are met:
>
> * INTEL_USE_NIR environment variable is defined and is positive (1 or true)
> * The stage is vertex shader
> * The HW generation is either SandyBridge
https://bugs.freedesktop.org/show_bug.cgi?id=91149
Bug ID: 91149
Summary: make check optimization-test regression
Product: Mesa
Version: git
Hardware: x86-64 (AMD64)
OS: All
Status: NEW
Keywords: bisected,
https://bugs.freedesktop.org/show_bug.cgi?id=91149
--- Comment #1 from Erik Faye-Lund ---
Without seeing what the actual problem is (it's lacking from the report), my
guess would be that Matt's work-around from
is the
remedy you're looking for:
sed -i -e 's/strtod\.cpp/strtod.c/' src/util/.deps
https://bugs.freedesktop.org/show_bug.cgi?id=91149
--- Comment #2 from Vinson Lee ---
Reproduced with a clean checkout and build.
== Testing optimization passes ==
Testing ./lower_jumps/lower_returns_main_false.opt_test...FAIL
Traceback (most recent call last):
File "src/glsl/tests/com
On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
> The NIR->vec4 pass will be activated if ALL the following conditions are met:
>
> * INTEL_USE_NIR environment variable is defined and is positive (1 or true)
> * The stage is vertex shader
> * The HW generation is either SandyBridge (gen
https://bugs.freedesktop.org/show_bug.cgi?id=91149
--- Comment #3 from Erik Faye-Lund ---
Created attachment 116806
--> https://bugs.freedesktop.org/attachment.cgi?id=116806&action=edit
Fix for test-regression
Yeah, seems like there were another user of the compiler-internals. And it
doesn't p
On 06/29/2015 11:22 PM, Jason Ekstrand wrote:
> On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
>> The NIR->vec4 pass will be activated if ALL the following conditions are met:
>>
>> * INTEL_USE_NIR environment variable is defined and is positive (1 or true)
>> * The stage is vertex sha
On 06/29/2015 10:20 PM, Kenneth Graunke wrote:
> On Friday, June 26, 2015 10:06:18 AM Eduardo Lima Mitev wrote:
>> The NIR->vec4 pass will be activated if ALL the following
>> conditions are met:
>>
>> * INTEL_USE_NIR environment variable is defined and is positive
>> (1 or true) * The stage is ve
This is the third iteration of a patch to resolve a strict aliasing
problem in the exec_list structure.
At the suggestion of Francisco Jerez this is a return to the original
(v1) style of the patch, which replaces the three 'exec_node *' members
in the list with two 'struct exec_node' members.
On Mon, Jun 29, 2015 at 2:49 PM, Eduardo Lima Mitev wrote:
> On 06/29/2015 11:22 PM, Jason Ekstrand wrote:
>> On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
>>> The NIR->vec4 pass will be activated if ALL the following conditions are
>>> met:
>>>
>>> * INTEL_USE_NIR environment varia
On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
> This implementation sets up a map of input variable offsets to source
> registers
> that are already initialized with the corresponding register offset.
>
> This map will then be queried when processing load_input intrinsic operations,
On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
> From: Iago Toral Quiroga
>
> This is based on similar code existing in vec4_visitor. It builds the
> uniform register file iterating through each uniform variable. It
> also stores the index of each register at the corresponding offset
On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
> From: Alejandro Piñeiro
>
> The new virtual method is more flexible, it has a signature:
>
> dst_reg *make_reg_for_system_value(int location, const glsl_type *type);
>
> so the current method will be chained through this one.
I just gr
https://bugs.freedesktop.org/show_bug.cgi?id=77449
Bug 77449 depends on bug 84570, which changed state.
Bug 84570 Summary: Borderlands 2/Pre-Sequel: Constant frame rate drops while
playing; really bad with additionl lighting
https://bugs.freedesktop.org/show_bug.cgi?id=84570
What|
https://bugs.freedesktop.org/show_bug.cgi?id=84570
Andreas Hartmetz changed:
What|Removed |Added
Status|RESOLVED|REOPENED
Resolution|FIXED
Good work guys! I've started reviewing but review will probably take
a few days so please be patient.
On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
> Hello,
>
> This series adds a new vec4 backend for i965 based on NIR. It is the result
> of working on
> https://bugs.freedesktop.o
On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
> From: Alejandro Piñeiro
>
> Similar to other variable setups, system values will initialize the
> corresponding register inside a 'nir_system_values' map, which will then
> be queried later when processing the different system value int
Am 29.06.2015 um 22:18 schrieb Dave Airlie:
> On 30 June 2015 at 00:58, Roland Scheidegger wrote:
>> Don't worry about the AoS stuff. Only meant to do simple things.
>>
>> Looks good overall, I guess it makes sense to not split execution too
>> (so you'd have native hw vector size there), llvm sho
https://bugs.freedesktop.org/show_bug.cgi?id=84570
--- Comment #36 from Andreas Hartmetz ---
Forgot to mention: Ubuntu kernel 3.19.
--
You are receiving this mail because:
You are the assignee for the bug.
___
mesa-dev mailing list
mesa-dev@lists.free
On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
> It basically allocates registers local to a function in a nir_locals map,
> then emits all its control-flow blocks.
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89580
> ---
> src/mesa/drivers/dri/i965/brw_vec4.h | 1
As a general comment, please remove the Bugzilla tags from the
commits. This is a new feature, not a bugfix. If you want to leave a
reference to the bug, pick a commit and put it there but it doesn't
need to be in all of them.
--Jason
On Mon, Jun 29, 2015 at 4:29 PM, Jason Ekstrand wrote:
> Goo
On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
> New method brw_writemask_for_size() will return a writemask with the first
> 'size'
> components activated.
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89580
> ---
> src/mesa/drivers/dri/i965/brw_reg.h | 9 +
> 1
On 06/30/2015 01:49 AM, Jason Ekstrand wrote:
> As a general comment, please remove the Bugzilla tags from the
> commits. This is a new feature, not a bugfix. If you want to leave a
> reference to the bug, pick a commit and put it there but it doesn't
> need to be in all of them.
> --Jason
>
Ok
On Jun 29, 2015 11:21 AM, "Francisco Jerez" wrote:
>
> Jason Ekstrand writes:
>
> > Reviewed-by: Iago Toral Quiroga
> > Reviewed-by: Topi Pohjolainen
> > ---
> > src/mesa/drivers/dri/i965/brw_fs.cpp | 6 ++
> > 1 file changed, 6 insertions(+)
> >
> > diff --git a/src/mesa/drivers/dri/i965/
This adds support for ARB_gpu_shader_fp64 and ARB_vertex_attrib_64bit to
llvmpipe.
Two things that don't mix well are SoA and doubles, see
emit_fetch_double, and emit_store_double_chan in this.
I've also had to split emit_data.chan, to add src_chan,
which can be different for doubles.
It handles
On 30 June 2015 at 09:36, Roland Scheidegger wrote:
> Am 29.06.2015 um 22:18 schrieb Dave Airlie:
>> On 30 June 2015 at 00:58, Roland Scheidegger wrote:
>>> Don't worry about the AoS stuff. Only meant to do simple things.
>>>
>>> Looks good overall, I guess it makes sense to not split execution t
Hello,
Since git 2.3, there have been a number of new fsck options added which
produce issues when I clone the repository
git://anongit.freedesktop.org/git/mesa/mesa
These new tests are enabled by default when using git fsck. I have been
testing with git version 2.4.4.409.g5b1d901 and thought yo
On 30/06/15 01:27, Jason Ekstrand wrote:
> On Fri, Jun 26, 2015 at 1:06 AM, Eduardo Lima Mitev wrote:
>> From: Alejandro Piñeiro
>>
>> The new virtual method is more flexible, it has a signature:
>>
>> dst_reg *make_reg_for_system_value(int location, const glsl_type *type);
>>
>> so the current
76 matches
Mail list logo
