On Sat, Aug 13, 2016 at 5:07 PM, Daniel Scharrer wrote:
> This was first fixed in commit b3f9c5c and then broken again in commit
> fe2d2c7, which removed the abs modifier from input registers.
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91342
> Signed-off-by:
It's fairly rare that the BB layout puts BBs after the exit block, which
is likely the reason these issues lingered for so long.
This fixes a fraction of issues with the giant pixmark piano shader.
Signed-off-by: Ilia Mirkin
Cc:
---
This was first fixed in commit b3f9c5c and then broken again in commit
fe2d2c7, which removed the abs modifier from input registers.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91342
Signed-off-by: Daniel Scharrer
Cc: "12.0"
https://bugs.freedesktop.org/show_bug.cgi?id=77449
Bug 77449 depends on bug 91342, which changed state.
Bug 91342 Summary: Very dark textures on some objects in indoors environments
in Postal 2
https://bugs.freedesktop.org/show_bug.cgi?id=91342
What|Removed
On Saturday, August 13, 2016 12:59:46 PM PDT Chris Wilson wrote:
> If the buffer has been freed by the kernel under memory pressure, it is
> invalid to try and access the backing storage for that buffer in the
> future - the backing storage is not recreated automatically. As such we
> need to mark
https://bugs.freedesktop.org/show_bug.cgi?id=97285
Kai changed:
What|Removed |Added
Resolution|FIXED |---
https://bugs.freedesktop.org/show_bug.cgi?id=97285
--- Comment #14 from Vladimir Usikov ---
Why fixed? Patch not in git master. Current git master stil broken.
--
You are receiving this mail because:
You are the assignee for the bug.
You are the QA Contact for the
On Sat, Aug 13, 2016 at 3:33 PM, Ilia Mirkin wrote:
> On Sat, Aug 13, 2016 at 3:26 PM, Connor Abbott wrote:
>> So, I don't know much about how nv50 ir works, but to me this just
>> seems like a pretty slow implementation of a very limited instruction
>>
https://bugs.freedesktop.org/show_bug.cgi?id=97285
Dmytro Zverev changed:
What|Removed |Added
Version|unspecified |git
https://bugs.freedesktop.org/show_bug.cgi?id=97285
Dmytro Zverev changed:
What|Removed |Added
Status|REOPENED|RESOLVED
2016-08-13 21:33 GMT+02:00 Ilia Mirkin :
> On Sat, Aug 13, 2016 at 3:26 PM, Connor Abbott wrote:
>> So, I don't know much about how nv50 ir works, but to me this just
>> seems like a pretty slow implementation of a very limited instruction
>> scheduler.
https://bugs.freedesktop.org/show_bug.cgi?id=77449
Bug 77449 depends on bug 91342, which changed state.
Bug 91342 Summary: Very dark textures on some objects in indoors environments
in Postal 2
https://bugs.freedesktop.org/show_bug.cgi?id=91342
What|Removed
On Sat, Aug 13, 2016 at 3:26 PM, Connor Abbott wrote:
> So, I don't know much about how nv50 ir works, but to me this just
> seems like a pretty slow implementation of a very limited instruction
> scheduler. In addition to the runtime complexity problems you
> mentioned,
So, I don't know much about how nv50 ir works, but to me this just
seems like a pretty slow implementation of a very limited instruction
scheduler. In addition to the runtime complexity problems you
mentioned, you're going to get a lot more benefit even from a very
simple list scheduler compared
On Sat, Aug 13, 2016 at 2:23 PM, Rob Clark wrote:
> Signed-off-by: Rob Clark
> ---
> src/gallium/auxiliary/util/u_blitter.c | 38
> ++
> src/gallium/auxiliary/util/u_blitter.h | 5 +
> 2 files changed, 30
Signed-off-by: Rob Clark
---
src/gallium/drivers/freedreno/a4xx/fd4_context.c | 3 +-
src/gallium/drivers/freedreno/a4xx/fd4_draw.c| 216 +--
2 files changed, 4 insertions(+), 215 deletions(-)
diff --git
Signed-off-by: Rob Clark
---
src/gallium/drivers/freedreno/a3xx/fd3_context.c | 3 +-
src/gallium/drivers/freedreno/a3xx/fd3_draw.c| 201 +--
2 files changed, 4 insertions(+), 200 deletions(-)
diff --git
Signed-off-by: Rob Clark
---
src/gallium/auxiliary/util/u_blitter.c | 38 ++
src/gallium/auxiliary/util/u_blitter.h | 5 +
2 files changed, 30 insertions(+), 13 deletions(-)
diff --git a/src/gallium/auxiliary/util/u_blitter.c
Since clears are more or less just normal draws, there isn't that much
benefit in having hand-rolled clear path. Add support to use u_blitter
instead if gen specific backend doesn't implement ctx->clear().
Signed-off-by: Rob Clark
---
So, for adreno, clears are basically just normal draws. So I've been
meaning to get rid of a bunch of generation specific hand-rolled clear
draw code. I got more motivated when I realized a bunch more stuff
would need to be fixed up to make MSAA work if I kept the hand-rolled
clear path.
The
Not (currently) state that is overwridden by u_blitter itself, but
drivers with custom blit/clear which are reusing part of the u_blitter
infrastructure will use it.
Signed-off-by: Rob Clark
---
src/gallium/auxiliary/util/u_blitter.c | 10 ++
Signed-off-by: Rob Clark
---
src/gallium/auxiliary/util/u_blitter.c | 145 +
src/gallium/auxiliary/util/u_blitter.h | 9 ++
2 files changed, 84 insertions(+), 70 deletions(-)
diff --git a/src/gallium/auxiliary/util/u_blitter.c
On Sat, Aug 13, 2016 at 10:43 AM, Tobias Klausmann
wrote:
>
>
>
> On 13.08.2016 12:02, Karol Herbst wrote:
>>
>> Signed-off-by: Karol Herbst
>> ---
>> src/gallium/drivers/nouveau/codegen/nv50_ir_target.h| 1 +
>>
2016-08-13 19:27 GMT+02:00 Ilia Mirkin :
> On Sat, Aug 13, 2016 at 1:24 PM, karol herbst wrote:
>> 2016-08-13 18:17 GMT+02:00 Ilia Mirkin :
>>> On Sat, Aug 13, 2016 at 6:02 AM, Karol Herbst wrote:
no
On Sat, Aug 13, 2016 at 1:24 PM, karol herbst wrote:
> 2016-08-13 18:17 GMT+02:00 Ilia Mirkin :
>> On Sat, Aug 13, 2016 at 6:02 AM, Karol Herbst wrote:
>>> no changes without a dual_issue pass
>>>
>>> changes with for ./GpuTest
2016-08-13 18:17 GMT+02:00 Ilia Mirkin :
> On Sat, Aug 13, 2016 at 6:02 AM, Karol Herbst wrote:
>> no changes without a dual_issue pass
>>
>> changes with for ./GpuTest /test=pixmark_piano /benchmark /no_scorebox
>> /msaa=0
>>
2016-08-13 17:43 GMT+02:00 Tobias Klausmann
:
> Hi Karol,
>
> one question inline.
>
>
> On 13.08.2016 12:02, Karol Herbst wrote:
>>
>> min/max pairs can be dual issued on Kepler1
>>
>> changes for ./GpuTest /test=pixmark_piano /benchmark /no_scorebox /msaa=0
On Sat, Aug 13, 2016 at 6:02 AM, Karol Herbst wrote:
> no changes without a dual_issue pass
>
> changes with for ./GpuTest /test=pixmark_piano /benchmark /no_scorebox /msaa=0
> /benchmark_duration_ms=6 /width=1024 /height=640:
>
> inst_executed: 1.03G
> inst_issued1:
Hi Karol,
one question inline.
On 13.08.2016 12:02, Karol Herbst wrote:
min/max pairs can be dual issued on Kepler1
changes for ./GpuTest /test=pixmark_piano /benchmark /no_scorebox /msaa=0
/benchmark_duration_ms=6 /width=1024 /height=640:
without dual_issue pass:
inst_executed: 1.03G
On 13.08.2016 12:02, Karol Herbst wrote:
no changes without a dual_issue pass
changes with for ./GpuTest /test=pixmark_piano /benchmark /no_scorebox /msaa=0
/benchmark_duration_ms=6 /width=1024 /height=640:
inst_executed: 1.03G
inst_issued1: 538M -> 535M
inst_issued2: 251M -> 254M
On 13.08.2016 12:02, Karol Herbst wrote:
Signed-off-by: Karol Herbst
---
src/gallium/drivers/nouveau/codegen/nv50_ir_target.h| 1 +
src/gallium/drivers/nouveau/codegen/nv50_ir_target_nvc0.cpp | 7 ++-
https://bugs.freedesktop.org/show_bug.cgi?id=97285
--- Comment #12 from Dmytro Zverev ---
Please add this patch to git master branch. This patch fix bug 91342 too.
--
You are receiving this mail because:
You are the assignee for the bug.
You are the QA Contact for the
On Sat, Aug 13, 2016 at 12:59:46PM +0100, Chris Wilson wrote:
> If the buffer has been freed by the kernel under memory pressure, it is
> invalid to try and access the backing storage for that buffer in the
> future - the backing storage is not recreated automatically. As such we
> need to mark
If the buffer has been freed by the kernel under memory pressure, it is
invalid to try and access the backing storage for that buffer in the
future - the backing storage is not recreated automatically. As such we
need to mark the GL object as being freed for unretained buffers and so
recreate the
Hi,
I made an attempt to run Enemy Territories Quake Wars using the OpenSWR
driver and I found a number of bugs in the implementation
The final result of the 'hacks' below is the first one or two frames of the
actual game rendering but with most of the rendered textures painted in red
(perhaps
https://bugs.freedesktop.org/show_bug.cgi?id=77449
Bug 77449 depends on bug 91342, which changed state.
Bug 91342 Summary: Very dark textures on some objects in indoors environments
in Postal 2
https://bugs.freedesktop.org/show_bug.cgi?id=91342
What|Removed
> On Aug 12, 2016, at 05:11, Tapani Pälli wrote:
>
>
>
> On 08/12/2016 01:31 PM, Jeremy Huddleston Sequoia wrote:
>>
>>> On Aug 12, 2016, at 00:41, Tapani Pälli wrote:
>>>
>>>
>>> On 07/18/2016 07:07 PM, Jeremy Huddleston Sequoia wrote:
no changes without a dual_issue pass
changes with for ./GpuTest /test=pixmark_piano /benchmark /no_scorebox /msaa=0
/benchmark_duration_ms=6 /width=1024 /height=640:
inst_executed: 1.03G
inst_issued1: 538M -> 535M
inst_issued2: 251M -> 254M
score: 1038 -> 1052
Signed-off-by: Karol Herbst
Signed-off-by: Karol Herbst
---
src/gallium/drivers/nouveau/codegen/nv50_ir_target.h| 1 +
src/gallium/drivers/nouveau/codegen/nv50_ir_target_nvc0.cpp | 7 ++-
src/gallium/drivers/nouveau/codegen/nv50_ir_target_nvc0.h | 1 +
3 files changed, 8 insertions(+),
this paths reorders instructions in a way, so that the hardware is able to
dual issue more instructions leading to higher performance.
with all previos commits put together
changes for ./GpuTest /test=pixmark_piano /benchmark /no_scorebox /msaa=0
/benchmark_duration_ms=6 /width=1024
this series improves the dual issueing code for nvc0 and newer gpus.
the pass has some outstanding runtime complexity issues I still need to solve,
but I included the patch for the sake of completeness.
patches 2-3 can be merged without the 1st and 4th one, but the performance
gains without the
min/max pairs can be dual issued on Kepler1
changes for ./GpuTest /test=pixmark_piano /benchmark /no_scorebox /msaa=0
/benchmark_duration_ms=6 /width=1024 /height=640:
without dual_issue pass:
inst_executed: 1.03G
inst_issued1: 614M -> 580M
inst_issued2: 213M -> 230M
score: 1021 -> 1030
Hi,
On Friday, 12 August 2016 12:15:52 CEST Nicolai Hähnle wrote:
> I take it this means `git bisect` becomes unusable? At least it looks
> like it after a bit of playing around with it. That seems like a
> show-stopper :(
git bisect skip does this job.
But it's pretty surprising if you step
On Friday, August 12, 2016 9:46:59 PM PDT Jason Ekstrand wrote:
> On Aug 12, 2016 8:13 PM, "Kenneth Graunke" wrote:
> >
> > We'll do blending in the shader in this case, so just disable the
> > hardware blending.
> >
> > Signed-off-by: Kenneth Graunke
44 matches
Mail list logo