On 20 February 2015 at 12:49, Ilia Mirkin wrote:
> On Thu, Feb 19, 2015 at 9:43 PM, Dave Airlie wrote:
>> From: Dave Airlie
>>
>> I'm not sure we really care about this, but we need to
>> write better support if we do. For now just disable it.
>>
>> piglit test:
>> tests/spec/arb_gpu_shader_fp6
From: Dave Airlie
I'm not sure we really care about this, but we need to
write better support if we do. For now just disable it.
piglit test:
tests/spec/arb_gpu_shader_fp64/execution/vs-out-fs-in-double-2.shader_test
v2: move code down below to avoid getting array type.
Signed-off-by: Dave Ai
Missed a few drivers in the earlier changes, this should fix up all the
ones that print unknown caps or don't have a default statement.
Signed-off-by: Ilia Mirkin
---
src/gallium/drivers/freedreno/freedreno_screen.c | 2 ++
src/gallium/drivers/i915/i915_screen.c | 4
src/gallium/d
On 20 February 2015 at 14:46, Ilia Mirkin wrote:
> Missed a few drivers in the earlier changes, this should fix up all the
> ones that print unknown caps or don't have a default statement.
>
> +++ b/src/gallium/drivers/vc4/vc4_screen.c
> @@ -318,6 +318,10 @@ vc4_screen_get_shader_param(struct pi
On Thu, Feb 19, 2015 at 11:54 PM, Dave Airlie wrote:
> On 20 February 2015 at 14:46, Ilia Mirkin wrote:
>> Missed a few drivers in the earlier changes, this should fix up all the
>> ones that print unknown caps or don't have a default statement.
>>
>
>
>> +++ b/src/gallium/drivers/vc4/vc4_screen.
On Thu, Feb 19, 2015 at 03:42:05PM -0800, Jordan Justen wrote:
> For fragment programs, we pull this mask from the payload header. The same
> mask doesn't exist for compute shaders, so we set all bits to enabled.
>
> Note: this mask is ANDed with the execution mask, so some channels may not end
>
We were already do this for ALU operations but we haven't for non-ALU
operations. This changes that.
total NIR instructions in shared programs: 2039883 -> 2022338 (-0.86%)
NIR instructions in affected programs: 1768850 -> 1751305 (-0.99%)
helped:14244
HURT:
On Thu, Feb 19, 2015 at 3:42 PM, Jordan Justen
wrote:
> For fragment programs, we pull this mask from the payload header. The same
> mask doesn't exist for compute shaders, so we set all bits to enabled.
>
> Note: this mask is ANDed with the execution mask, so some channels may not end
> up issuin
On Thursday, February 19, 2015 10:03:15 PM Jason Ekstrand wrote:
> We were already do this for ALU operations but we haven't for non-ALU
> operations. This changes that.
>
> total NIR instructions in shared programs: 2039883 -> 2022338 (-0.86%)
> NIR instructions in affected programs: 1768850
When under dispatch_width=16 the previous code would allocate 2 registers for
the payload when only one is needed. This manifested itself through bugs on SKL
which needs to mess with this instruction.
Ken though this might impact shader-db, but apparently it doesn't
Cc: Kenneth Graunke
Bugzilla:
I agree with Ken that the regressions are small enough, and it seems
they're mostly stuff we can prevent by being smarter when doing the
sel peephole, so it seems like the cleanup that will probably help
other passes is worth it.
Reviewed-by: Connor Abbott
On Fri, Feb 20, 2015 at 1:03 AM, Jason
On Thu, Feb 19, 2015 at 11:01 PM, Connor Abbott wrote:
> I agree with Ken that the regressions are small enough, and it seems
> they're mostly stuff we can prevent by being smarter when doing the
> sel peephole, so it seems like the cleanup that will probably help
> other passes is worth it.
So,
On Wednesday 18 February 2015 11:51:45 Matt Turner wrote:
> On Tue, Dec 9, 2014 at 2:52 AM, Eduardo Lima Mitev wrote:
> > From: Samuel Iglesias Gonsalvez
> >
> > GLSL 1.50 and GLSL 4.40 specs, they both say the same in
> > "Interface Blocks" section:
> >
> > "If no optional qualifier is used in
On Thursday 19 February 2015 11:39:13 Ian Romanick wrote:
> On 12/09/2014 02:52 AM, Eduardo Lima Mitev wrote:
> > From: Samuel Iglesias Gonsalvez
> >
> > GLSL 1.50 and GLSL 4.40 specs, they both say the same in
> > "Interface Blocks" section:
> >
> > "If no optional qualifier is used in a member
In gen6 we need to compute the primitive count in the generated GS program.
The current implementation only counts full primitives, that is, if the
output primitive type is a triangle strip, it won't count individual
triangles in the strip, only complete strips.
If we want to count basic primitive
On 2015-02-19 21:40:37, Ben Widawsky wrote:
> On Thu, Feb 19, 2015 at 03:42:05PM -0800, Jordan Justen wrote:
> > For fragment programs, we pull this mask from the payload header. The same
> > mask doesn't exist for compute shaders, so we set all bits to enabled.
> >
> > Note: this mask is ANDed wi
On Fri, Feb 20, 2015 at 08:21:25AM +0100, Iago Toral Quiroga wrote:
> In gen6 we need to compute the primitive count in the generated GS program.
> The current implementation only counts full primitives, that is, if the
> output primitive type is a triangle strip, it won't count individual
> triang
On Thu, Feb 19, 2015 at 11:25:56PM -0800, Jordan Justen wrote:
> On 2015-02-19 21:40:37, Ben Widawsky wrote:
> > On Thu, Feb 19, 2015 at 03:42:05PM -0800, Jordan Justen wrote:
> > > For fragment programs, we pull this mask from the payload header. The same
> > > mask doesn't exist for compute shade
On 18/02/2015 20:10, Ian Romanick wrote:
On 02/17/2015 07:59 AM, Martin Peres wrote:
Hey Laura,
Thanks for this code. I'll be sending reviews throughout the week :)
On 12/02/15 04:05, Laura Ekstrand wrote:
diff --git a/src/mesa/main/teximage.c b/src/mesa/main/teximage.c
index 336feff..ce6f446
On Tue, Feb 17, 2015 at 11:46 AM, Matt Turner wrote:
> total instructions in shared programs: 8013221 -> 8010869 (-0.03%)
> instructions in affected programs: 475925 -> 473573 (-0.49%)
> helped:2350
> ---
Patches 1 and 3 have been reviewed, but I'm this one has
On 20/02/2015 02:15, Laura Ekstrand wrote:
This is NamedBufferStorage, not NamedBufferData. The storage function
uses a bitfield instead of an enum.
Oops, sorry. I was trying to match these patches with their relative
piglit tests and as the order is not always the same, I must have got
confu
On 20/02/2015 02:12, Laura Ekstrand wrote:
On Wed, Feb 18, 2015 at 7:40 AM, Martin Peres
mailto:martin.pe...@linux.intel.com>>
wrote:
On 12/02/15 04:05, Laura Ekstrand wrote:
v2: review from Ian Romanick
- Fix space in ARB_direct_state_access.xml.
- Rem
101 - 122 of 122 matches
Mail list logo
