From: Dave Airlie
vertex inputs are counted differently in some cases, with
vertex inputs we need to make sure we don't double count them.
Signed-off-by: Dave Airlie
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 19 ++-
1 file changed, 14 insertions(+), 5 deletions(-)
diff
From: Dave Airlie
This code takes into account double inputs in the array
shrinking code. This fixes some issues with doubles
and geom/tess inputs.
Signed-off-by: Dave Airlie
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --g
From: Dave Airlie
It's important for the double instruction emission code that
the writemasks are correct going in for double so it know
which channels to replicate.
This fixes it for the array and matrix cases.
Signed-off-by: Dave Airlie
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 21 ++
From: Dave Airlie
This fixes fp64 relative addressing, in the upcoming
dmat-vs-gs-tcs-tes.shader_test.
Signed-off-by: Dave Airlie
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
b/src
From: Dave Airlie
Otherwise we end up emitting the wrong index for the second
double.
This fixes dmat-vs-gs-tcs-tes.shader_test and dvec3-vs-gs-tcs-tes.shader_test
Signed-off-by: Dave Airlie
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(
From: Dave Airlie
This is just a precursor patch to a fix for doubles with
tessellation that I've written.
We need to descend into output arrays in that case and
mark dst's as double.
Signed-off-by: Dave Airlie
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 13 ++---
1 file changed,
From: Dave Airlie
This handles the case where a double output is stored
in an array, and tracks it for use in the double
instruction emit code.
Signed-off-by: Dave Airlie
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 35 ++
1 file changed, 31 insertions(+), 4 del
From: Dave Airlie
While playing with fp64, I disable varying packing to debug
something else, and noticed we never emitted half the output
movs for double matrix arrays.
We should be moving the left index two slots for dual
source doubles, and the right index two slots for non-vs
input doubles.
From: Dave Airlie
This handles matrix initialisation properly.
Signed-off-by: Dave Airlie
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 59 --
1 file changed, 48 insertions(+), 11 deletions(-)
diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
b/src/mesa/s
This fixes a bunch of cases with the GLSL->TGSI convertor when
it comes to doubles.
The main areas it addresses are proper handling of doubles in arrays,
double matricies and the ARB_vertex_attrib_64bit vertex input
special double handling.
These fixes contribute to fixing the CTS test
GL41-CTS.g
https://bugs.freedesktop.org/show_bug.cgi?id=91806
Jonathan Gray changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
From: Dave Airlie
If oViewport is written, vertex reuse need to be turned off.
If oViewport is constant, vertex reuse is fine, and VPORT_PROVOKE_DISABLE
need to be set.
Fixes: arb_viewport_array-render-viewport-2 and some CTS tests.
Signed-off-by: Dave Airlie
---
src/gallium/drivers/radeonsi/
From: Dave Airlie
If oViewport is written, vertex reuse need to be turned off.
If oViewport is constant, vertex reuse is fine, and VPORT_PROVOKE_DISABLE
need to be set.
Fixes: arb_viewport_array-render-viewport-2 and some CTS tests.
Signed-off-by: Dave Airlie
---
src/gallium/drivers/r600/ever
From: Dave Airlie
This function deals with vertex inputs and fragment
outputs, so we should count the attribute locations
correctly for the vertex inputs.
Signed-off-by: Dave Airlie
---
src/glsl/linker.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/glsl/linker.cpp
Reviewed-by: Mark Janes
Tested-by: Mark Janes
Emil Velikov writes:
> With an earlier commit we've spit the flags parsing to a separate
> function, but forgot to update all the dri modules to use it.
>
> Noticed when we've enabled KHR_debug for every dri module - fdo#93048
>
> Fixes: 38366c0c6e
On Thu, Dec 17, 2015 at 11:01 PM, Nicolai Hähnle wrote:
> From: Nicolai Hähnle
>
> This option allows replacing a single shader by a pre-compiled ELF object
> as generated by LLVM's llc, for example. This can be useful for debugging a
> deterministically occuring error in shaders (and has in fact
Reviewed-by: Marek Olšák
On Thu, Dec 17, 2015 at 11:01 PM, Nicolai Hähnle wrote:
> From: Nicolai Hähnle
>
> This changes the count slightly (because of si_generate_gs_copy_shader), but
> this is only relevant for the driver-specific num-compilations query. It sets
> the stage for the next commi
On 12/17/2015 01:06 PM, Marek Olšák wrote:
On Wed, Dec 16, 2015 at 11:30 PM, Miklós Máté wrote:
On 12/16/2015 05:27 PM, Marek Olšák wrote:
What is this good for?
Marek
KotOR uses a series of scratch framebuffers for drawing the framebuffer
effects. These have no depth and no stencil, so chec
On Fri, Dec 18, 2015 at 11:45 PM, Miklós Máté wrote:
> On 12/17/2015 01:06 PM, Marek Olšák wrote:
>>
>> On Wed, Dec 16, 2015 at 11:30 PM, Miklós Máté wrote:
>>>
>>> On 12/16/2015 05:27 PM, Marek Olšák wrote:
What is this good for?
Marek
>>>
>>> KotOR uses a series of scratch f
On 19 December 2015 at 02:57, Emil Velikov wrote:
> On 17 December 2015 at 15:45, Rob Herring wrote:
>> Add the virtio-gpu PCI ID so the driver probing works.
>>
> Dave, I'd imagine that currently this works ala platform device style
> ? I.e. we fail to get the pciid and fall back to the kernel m
On Fri, Dec 18, 2015 at 1:26 AM, Pohjolainen, Topi
wrote:
> On Fri, Dec 11, 2015 at 07:14:21PM -0800, Anuj Phogat wrote:
>> Overlapping blits are anyway undefined in OpenGL. So no need
>> of overlap check here.
>
> Does this mean that earlier checks prevent us from getting here or
> are we going t
On Thu, Dec 17, 2015 at 10:42 AM, Rob Herring wrote:
> Android builds with -Werror=pointer-to-int-cast causing an error on 32-bit
> builds.
>
> Signed-off-by: Rob Herring
thanks, pushed.. and added CC-stable tag for good measure
BR,
-R
> ---
> src/gallium/drivers/freedreno/ir3/ir3_print.c | 2
On 12/18/2015 11:32 AM, Brad King wrote:
> Prior to this change the test renders what we expect. After the
> change the test renders *nothing* on top of the background.
Ugh, nevermind, sorry for the noise.
It turns out this was a branch-on-uninitialized-value bug in VTK
that happened to trigger
Reviewed-by: Matt Turner
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Fri, Dec 18, 2015 at 11:01 AM, Emil Velikov wrote:
> Thank you Rob !
>
> On 17 December 2015 at 15:45, Rob Herring wrote:
>> This is just a copy-n-paste and rename of vc4 Android makefiles.
>>
> Looks great.
>
> Rob, I take it that you've at least compile tested this patch ?
Better than that,
On 17.12.2015 22:13, Michel Dänzer wrote:
On 18.12.2015 07:00, Nicolai Hähnle wrote:
From: Nicolai Hähnle
Unify the cleanup paths of the function rather than duplicating code.
This should probably be backported to the stable branches? If so, add
True. It's not like it's a critical fix, giv
Thank you Rob !
On 17 December 2015 at 15:45, Rob Herring wrote:
> This is just a copy-n-paste and rename of vc4 Android makefiles.
>
Looks great.
Rob, I take it that you've at least compile tested this patch ?
Cc: "11.1"
Reviewed-by: Emil Velikov
I'll apply this some time next week.
-Emil
_
On 17 December 2015 at 15:45, Rob Herring wrote:
> Add the virtio-gpu PCI ID so the driver probing works.
>
Dave, I'd imagine that currently this works ala platform device style
? I.e. we fail to get the pciid and fall back to the kernel module
name being our dri module name.
If the PCI id is sta
As far as I remember, parts of it were.
The r300 driver, which is now the r600 driver, was a side project in 2005
and was hosted there. I don't remember anymore, but based on this, it
wasn't even svn.
sourceforge.net/p/r300/code/
On Fri, Dec 18, 2015 at 9:34 AM, Giuseppe Bilotta <
giuseppe.bilo..
On Fri, Dec 18, 2015 at 11:32 AM, Brad King wrote:
> Hi Folks,
>
> I run nightly testing of VTK with nightly 'master' of mesa.
> For the last few days some tests have been failing. According to
> `git bisect`, mesa commit 183c606066b1b260acb189e46a40cb71e63b44aa
> (glsl: simplify interface matchi
Hi Folks,
I run nightly testing of VTK with nightly 'master' of mesa.
For the last few days some tests have been failing. According to
`git bisect`, mesa commit 183c606066b1b260acb189e46a40cb71e63b44aa
(glsl: simplify interface matching, 2015-12-02) introduced the
regression.
For example, see th
With an earlier commit we've spit the flags parsing to a separate
function, but forgot to update all the dri modules to use it.
Noticed when we've enabled KHR_debug for every dri module - fdo#93048
Fixes: 38366c0c6e7 "dri_util: Don't assume __DRIcontext->driverPrivate
is a gl_context"
Cc: Mark Ja
On Fri, Dec 18, 2015 at 7:48 AM, Emil Velikov wrote:
> Hi Alex,
>
> On 20 November 2015 at 23:31, Alex Deucher wrote:
>> From: Boyuan Zhang
>>
>> v2: set the behaviour default for future ASICs.
>>
>> Signed-off-by: Boyuan Zhang
>> Reviewed-by: Leo Liu
>> Cc: mesa-sta...@lists.freedesktop.org
>
Hi Christian,
I've just sent a few comments/suggestions. With or without the
nitpicks (but the remaining addressed) the series is
Reviewed-by: Emil Velikov
Thanks
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.or
On 16 December 2015 at 20:14, Christian König wrote:
> @@ -174,6 +175,51 @@ static VAStatus vlVaPostProcBlit(vlVaDriver *drv,
> vlVaContext *context,
> return VA_STATUS_SUCCESS;
> }
>
> +static struct pipe_video_buffer *
> +vlVaApplyDeint(vlVaDriver *drv, vlVaContext *context,
> +
On 16 December 2015 at 20:14, Christian König wrote:
> @@ -367,7 +385,9 @@ vl_deint_filter_cleanup(struct vl_deint_filter *filter)
> assert(filter);
>
> filter->pipe->delete_sampler_state(filter->pipe, filter->sampler[0]);
> - filter->pipe->delete_blend_state(filter->pipe, filter->blend
Suggested by Ilia Mirkin.
Signed-off-by: Samuel Pitoiset
---
Untested patch.
src/gallium/drivers/nouveau/nv50/nv50_query_hw_sm.c | 8 ++--
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/src/gallium/drivers/nouveau/nv50/nv50_query_hw_sm.c
b/src/gallium/drivers/nouveau/nv50/n
Suggested by Ilia Mirkin.
Signed-off-by: Samuel Pitoiset
---
Untested patch.
src/gallium/drivers/nouveau/nv50/nv50_query_hw_sm.c | 9 +++--
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/src/gallium/drivers/nouveau/nv50/nv50_query_hw_sm.c
b/src/gallium/drivers/nouveau/nv50/
This might be useful to avoid breaking the current compute state when
monitoring MP perf counters because we use a compute kernel to read out
those counters. This has been suggested by Ilia Mirkin.
Signed-off-by: Samuel Pitoiset
---
src/gallium/drivers/nouveau/nv50/nv50_query_hw_sm.c | 2 ++
src
Hi Christian,
An alternative solution would be to refcount on handle_table_get/put
(first we need the latter even :-P) although I think that route will
be longer (i.e. larger patch). Just thinking out loud, than a
suggestion really.
On 16 December 2015 at 20:14, Christian König wrote:
> @@ -165
Hello list,
The candidate for the Mesa 11.0.8 is now available. Currently we have:
- 54 queued
- 17 nominated (outstanding)
- and 1 rejected/obsolete patches
Current queue covers a significant amount of fixes - from radeonsi
(Fiji, Hyper-Z), r600 (geom. shaders), nouveau (ir), freedreno (pigli
Hi Alex,
On 20 November 2015 at 23:31, Alex Deucher wrote:
> From: Boyuan Zhang
>
> v2: set the behaviour default for future ASICs.
>
> Signed-off-by: Boyuan Zhang
> Reviewed-by: Leo Liu
> Cc: mesa-sta...@lists.freedesktop.org
Just double-checking and noticed that isn't tagged for 11.0. I'd
im
Patch is:
Reviewed-by: Eduardo Lima Mitev
On 12/10/2015 05:23 AM, Jason Ekstrand wrote:
> It's not really doing enough anymore to justify a helper function.
> ---
> src/mesa/drivers/dri/i965/brw_vec4.h | 2 --
> src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 37
> ++---
On 18 December 2015 at 12:20, Christian König wrote:
> On 15.12.2015 11:16, Emil Velikov wrote:
>>
>> On 11 December 2015 at 12:33, Christian König
>> wrote:
>>
>>> +static void vlVaBoxAdjust(struct pipe_video_buffer *buf, unsigned idx,
>>> + struct pipe_box *box)
>>> +{
Iago Toral writes:
> On Thu, 2015-12-17 at 16:29 +0200, Francisco Jerez wrote:
>> Iago Toral Quiroga writes:
>>
>> > Some drivers can disable the FS unit if there is nothing in the shader code
>> > that writes to an output (i.e. color, depth, etc). Right now, mesa has
>> > a function to check f
On 15.12.2015 11:16, Emil Velikov wrote:
On 11 December 2015 at 12:33, Christian König wrote:
+static void vlVaBoxAdjust(struct pipe_video_buffer *buf, unsigned idx,
+ struct pipe_box *box)
+{
+ if (buf->interlaced) {
+ box->y /= 2;
+ box->height /= 2;
+
I was cleverly using one iteration to obtain a pointer to the last item
in ralloc's singly list child list, while also setting parents.
Unfortunately, I forgot to set the parent on that last item.
Signed-off-by: Kenneth Graunke
Cc: "11.1 11.0 10.6"
---
src/util/ralloc.c | 1 +
1 file changed,
On Fri, Dec 11, 2015 at 07:14:21PM -0800, Anuj Phogat wrote:
> Overlapping blits are anyway undefined in OpenGL. So no need
> of overlap check here.
Does this mean that earlier checks prevent us from getting here or
are we going to blit here (with undefined results)? If the latter,
is it always sa
The i965 driver uses this function to decide if it can disable the
FS unit in the absence of color/depth writes. We don't want to disable
the unit in the presence of SSBOs, since the fragment shader could
be writing to it.
We could go a step further and check not just for the presence of SSBOs
but
On Haswell we need to set the UAV_ONLY WM state bit when there are no colour
or depth buffer writes and on all hardware we should set the early
depth/stencil control field to PSEXEC unless early fragment tests are enabled
to make sure that the fragment shader is executed regardless of whether
per-f
Some drivers can disable the FS unit if there is nothing in the shader code
that writes to an output (i.e. color, depth, etc). Right now, mesa has
a function to check for atomic buffers and the i965 driver also checks for
images. Refactor this logic into a generic function that we can use for
any s
On Fri, Dec 18, 2015 at 9:12 AM, Boris Peterbarg wrote:
> As far as I remember, parts of it were.
> The r300 driver, which is now the r600 driver, was a side project in 2005
> and was hosted there. I don't remember anymore, but based on this, it wasn't
> even svn.
>
> sourceforge.net/p/r300/code/
52 matches
Mail list logo
