[Mesa-dev] [Bug 96550] 12.0.0-rc3: mesa_dri_drivers.so linking fails with: relocation R_X86_64_32S against `V4F_COUNT' can not be used when making a shared object

https://bugs.freedesktop.org/show_bug.cgi?id=96550 --- Comment #1 from war...@o2.pl --- FYI: ppl from Phoronix forums located the cause of the problem: gen_matypes.c is passed to the compiler in order to generate assembly code in text form (compiler switch: -S), but with -flto GCC doesn't output

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On Aug 10, 2016 1:02 AM, "Erik Faye-Lund" wrote: > > On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke wrote: > > On Haswell (GL 3.3): > > > > total instructions in shared programs: 6208759 -> 6203860 (-0.08%) > > instructions in affected programs:

Re: [Mesa-dev] [PATCH 09/19] gallium/radeon: simplify radeon_llvm_emit_fetch for direct array addressing

On Tue, Aug 09, 2016 at 12:36:38PM +0200, Nicolai Hähnle wrote: > From: Nicolai Hähnle > > We can use the pointer stored in the temps array directly. Reviewed-by: Tom Stellard > --- > src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 5

Re: [Mesa-dev] [PATCH 10/19] gallium/radeon: simplify radeon_llvm_emit_store for direct array addressing

On Tue, Aug 09, 2016 at 12:36:39PM +0200, Nicolai Hähnle wrote: > From: Nicolai Hähnle > > We can use the pointer stored in the temps array directly. Reviewed-by: Tom Stellard > --- > src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 7

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On Wed, Aug 10, 2016 at 5:48 PM, Jason Ekstrand wrote: > On Aug 10, 2016 1:02 AM, "Erik Faye-Lund" wrote: >> >> On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke >> wrote: >> > On Haswell (GL 3.3): >> > >> > total instructions in

Re: [Mesa-dev] [PATCH 08/19] gallium/radeon: clean up emit_declaration for temporaries

On Tue, Aug 09, 2016 at 12:36:37PM +0200, Nicolai Hähnle wrote: > From: Nicolai Hähnle > > In the alloca'd array case, no longer create redundant and unused allocas > for the individual elements; create getelementptrs instead. Reviewed-by: Tom Stellard

Re: [Mesa-dev] [PATCH 4/8] st/mesa: determine states used or affected by shaders at compile time

On Wed, Aug 10, 2016 at 4:34 PM, Nicolai Hähnle wrote: > On 10.08.2016 13:05, Marek Olšák wrote: >> >> On Tue, Aug 9, 2016 at 12:56 PM, Nicolai Hähnle >> wrote: >>> >>> On 07.08.2016 03:12, Marek Olšák wrote: From: Marek Olšák

Re: [Mesa-dev] [PATCH 1/2] anv/blit2d: Add a format parameter to bind_dst and create_iview

On Tue, Aug 02, 2016 at 10:00:29AM -0700, Jason Ekstrand wrote: > Signed-off-by: Jasosn Ekstrand > Cc: "12.0" > --- > src/intel/vulkan/anv_meta_blit2d.c | 15 ++- > 1 file changed, 10 insertions(+), 5 deletions(-) > > diff

Re: [Mesa-dev] [PATCH 11/19] gallium/radeon: more descriptive names for LLVM temporaries in debug builds

On Tue, Aug 09, 2016 at 12:36:40PM +0200, Nicolai Hähnle wrote: > From: Nicolai Hähnle > This is a great idea. Reviewed-by: Tom Stellard > --- > src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 14 -- > 1 file changed, 12

Re: [Mesa-dev] [PATCH 1/6] nir: Convert ineg(b2i(a)) to a if it's a boolean.

On Wed, Aug 10, 2016 at 5:40 PM, Jason Ekstrand wrote: > On Aug 10, 2016 1:00 AM, "Erik Faye-Lund" wrote: >> >> On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke >> wrote: >> > On Haswell (GL 3.3): >> > >> > total instructions in

Re: [Mesa-dev] [PATCH 0/2] *** Aubinator tool for Intel Gen platforms ***

On Tue, Aug 09, 2016 at 04:52:12PM -0700, Sirisha Gandikota wrote: > From: Sirisha Gandikota > > This is a patch series for adding the aubinator tool to the codebase. > > The aubinator tool is designed to help the driver developers to debug > the driver

Re: [Mesa-dev] [PATCH 1/2] anv/blit2d: Add a format parameter to bind_dst and create_iview

On Wed, Aug 10, 2016 at 10:12:17AM -0700, Nanley Chery wrote: > On Tue, Aug 02, 2016 at 10:00:29AM -0700, Jason Ekstrand wrote: > > Signed-off-by: Jasosn Ekstrand > > Cc: "12.0" I just noticed a bug in this patch. With that fixed the

Re: [Mesa-dev] [PATCH 1/6] nir: Convert ineg(b2i(a)) to a if it's a boolean.

On Aug 10, 2016 1:00 AM, "Erik Faye-Lund" wrote: > > On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke wrote: > > On Haswell (GL 3.3): > > > > total instructions in shared programs: 6211678 -> 6211584 (-0.00%) > > instructions in affected programs:

Re: [Mesa-dev] [PATCH 0/2] *** Aubinator tool for Intel Gen platforms ***

On Wed, Aug 10, 2016 at 09:43:24AM -0700, Nanley Chery wrote: > On Tue, Aug 09, 2016 at 04:52:12PM -0700, Sirisha Gandikota wrote: > > From: Sirisha Gandikota > > > > This is a patch series for adding the aubinator tool to the codebase. > > > > The aubinator tool is

Re: [Mesa-dev] [PATCH 00/19] gallium, radeonsi: temporary array improvements

FYI, I think I'm going to drop most of the gallium part of this series and replace it by a pass over the resulting TGSI to look at which components of the arrays are actually being used. Nicolai On 09.08.2016 12:36, Nicolai Hähnle wrote: Hi, this series was originally motivated by fixing a

Re: [Mesa-dev] [PATCH] st/mesa: in ATI fs don't assume TEMP0=REG0

Pushed, thanks. Marek On Mon, Aug 8, 2016 at 12:48 AM, Miklós Máté wrote: > The temporaries are allocated dynamically. > > Signed-off-by: Miklós Máté > --- > src/mesa/state_tracker/st_atifs_to_tgsi.c | 5 +++-- > 1 file changed, 3 insertions(+), 2

Re: [Mesa-dev] Moving amdgpu/addrlib into a git submodule

On 09.08.2016 22:12, Dave Airlie wrote: tbh, git submodules are more annoying than they need to be, and I'm not really terribly excited to use that for something that is a build dependency. Maybe just move it into libdrm instead? I've only had to use git submodules once with spice project,

Re: [Mesa-dev] Moving amdgpu/addrlib into a git submodule

On 10.08.2016 06:22, Edward O'Callaghan wrote: Well taking a step back here for a second. My initial thoughts are - what, conceptually, is the primary user of addrlib from an extremely high level view? If the case is mesa3d then perhaps mesa should be treated as the upstream from which other

Re: [Mesa-dev] [PATCH 3/3] gallium/radeon: use lp_build_alloca_undef

For the series: Reviewed-by: Marek Olšák Marek On Tue, Aug 9, 2016 at 12:38 PM, Nicolai Hähnle wrote: > From: Nicolai Hähnle > > Avoid building all those store 0 / store undef instrucction pairs that > end up getting removed

[Mesa-dev] [Bug 97231] GL_DEPTH_CLAMP doesn't clamp to the far plane

https://bugs.freedesktop.org/show_bug.cgi?id=97231 Jules Blok changed: What|Removed |Added Attachment #125650|0 |1 is

Re: [Mesa-dev] [PATCH 4/8] st/mesa: determine states used or affected by shaders at compile time

On 10.08.2016 13:05, Marek Olšák wrote: On Tue, Aug 9, 2016 at 12:56 PM, Nicolai Hähnle wrote: On 07.08.2016 03:12, Marek Olšák wrote: From: Marek Olšák At compile time, each shader determines which ST_NEW flags should be set at shader bind time.

Re: [Mesa-dev] [PATCH] st/nine: Fix invalid attempt to use indirect draws.

Pushed. Marek On Wed, Aug 10, 2016 at 1:16 PM, trevor.davenp...@gmail.com wrote: > I don't have commit access so someone else will need to push this out. > > Trevor > > > On Aug 10, 2016 2:56 AM, "Marek Olšák" wrote: >> >> Reviewed-by: Marek Olšák

Re: [Mesa-dev] [PATCH] [rfc] glsl: allow invariant on fragment shader outputs.

Hi, On Mon, 2016-05-23 at 14:18 +1000, Dave Airlie wrote: > From: Dave Airlie > > A CTS test manages to generate this: > GL45-CTS.shading_language_420pack.qualifier_order > > I cannot find definitive evidence in the spec that it isn't > allowed. The specs mentions some

Re: [Mesa-dev] [PATCH 0/4] RadeonSI: GPU hang fix + 3 small changes

On 10.08.2016 01:34, Marek Olšák wrote: Hi, The first patch is pretty important and fixes the root cause of Overlord and Witcher 2 hangs. The GLSL revert can be reverted after that. I wonder what else this might fix. Anyway, nice catch! The series is Reviewed-by: Nicolai Hähnle

Re: [Mesa-dev] [PATCH] glsl/tests: fix segfault in uniform initializer test

On Wed, Aug 10, 2016 at 7:34 PM, Timothy Arceri < timothy.arc...@collabora.com> wrote: > Cause by 549222f5 > s/Cause/Caused/ ? Fixes the make check failure for me (haven't done a full piglit run, just build/check/install test). Thanks for the quick turnaround. --Aaron > > Cc: Aaron Watry

[Mesa-dev] [Bug 97285] Darkness in Dota 2 after Patch "Make Gallium's BlitFramebuffer follow the GL 4.4 sRGB rules"

https://bugs.freedesktop.org/show_bug.cgi?id=97285 --- Comment #2 from Kenneth Graunke --- On i965, both Dota 2 (Source 2) and Left 4 Dead 2 (Source 1) appear to be working. I took screenshots from the same point in a Dota 2 tournament replay both with and without the

[Mesa-dev] [PATCH] glsl/tests: fix segfault in uniform initializer test

Cause by 549222f5 Cc: Aaron Watry --- src/compiler/glsl/tests/set_uniform_initializer_tests.cpp | 5 + 1 file changed, 5 insertions(+) diff --git a/src/compiler/glsl/tests/set_uniform_initializer_tests.cpp b/src/compiler/glsl/tests/set_uniform_initializer_tests.cpp index

Re: [Mesa-dev] [PATCH 3/6] nir: Turn -(b2f(a) + b2f(b) >= 0 into !(a || b).

> > > > For now, this patch is > > > > Reviewed-by: Ian Romanick > I had a hard time parsing the title: "Turn -(b2f(a) + b2f(b) >= 0 into !(a || b)" at first, until I saw the replacement instructions. You're missing a ')' on the commit line. :)

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On 08/10/2016 11:24 AM, Kenneth Graunke wrote: > On Wednesday, August 10, 2016 10:02:12 AM PDT Erik Faye-Lund wrote: >> On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke >> wrote: >>> On Haswell (GL 3.3): >>> >>> total instructions in shared programs: 6208759 -> 6203860

[Mesa-dev] [PATCH 13/16] gallium/radeon: use tgsi_scan_arrays for temp arrays

From: Nicolai Hähnle --- src/gallium/drivers/radeon/radeon_llvm.h| 3 ++- src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 9 ++--- src/gallium/drivers/radeonsi/si_shader.c| 3 ++- 3 files changed, 10 insertions(+), 5 deletions(-) diff

[Mesa-dev] [PATCH 14/16] gallium/radeon: reduce alloca of temporaries based on usagemask

From: Nicolai Hähnle v2: take actual writemasks into account --- src/gallium/drivers/radeon/radeon_llvm.h | 2 + .../drivers/radeon/radeon_setup_tgsi_llvm.c| 62 ++ 2 files changed, 54 insertions(+), 10 deletions(-) diff --git

[Mesa-dev] [PATCH 06/16] gallium/radeon: more descriptive names for LLVM temporaries in debug builds

From: Nicolai Hähnle Reviewed-by: Tom Stellard --- src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 14 -- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c

[Mesa-dev] [PATCH 12/16] gallium/radeon: allocate temps array info in radeon_llvm_context_init

From: Nicolai Hähnle Also, prepare for using tgsi_array_info. This also opens the door for properly handling allocation failures, but I'm leaving that for a separate change. --- src/gallium/drivers/radeon/radeon_llvm.h | 11 ++--

[Mesa-dev] [PATCH 16/16] gallium/radeon: protect against out of bounds temporary array accesses

From: Nicolai Hähnle They can lead to VM faults and worse, which goes against the GL robustness promises. --- src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 15 +++ 1 file changed, 15 insertions(+) diff --git

[Mesa-dev] [PATCH 10/16] gallium/radeon: extract common getelementptr logic into get_pointer_into_array

From: Nicolai Hähnle --- .../drivers/radeon/radeon_setup_tgsi_llvm.c| 105 + 1 file changed, 66 insertions(+), 39 deletions(-) diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c

[Mesa-dev] [PATCH 09/16] gallium/radeon: pass indirect register info into get_alloca_for_array

From: Nicolai Hähnle To have the same signature as get_array_range. --- src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 11 ++- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c

[Mesa-dev] [PATCH 11/16] gallium/radeon: always do the full store in store_value_to_array

From: Nicolai Hähnle Doing the write-back of the temporary vector in radeon_llvm_emit_store makes no sense. This also allows us to get rid of get_alloca_for_array. --- .../drivers/radeon/radeon_setup_tgsi_llvm.c| 77 -- 1 file changed, 28

[Mesa-dev] [PATCH 15/16] gallium/radeon: add radeon_llvm_bound_index for bounds checking

From: Nicolai Hähnle --- src/gallium/drivers/radeon/radeon_llvm.h | 4 +++ .../drivers/radeon/radeon_setup_tgsi_llvm.c| 29 ++ src/gallium/drivers/radeonsi/si_shader.c | 19 +- 3 files changed, 34

[Mesa-dev] [PATCH 01/16] tgsi/scan: add tgsi_scan_arrays

From: Nicolai Hähnle --- src/gallium/auxiliary/tgsi/tgsi_scan.c | 76 ++ src/gallium/auxiliary/tgsi/tgsi_scan.h | 17 2 files changed, 93 insertions(+) diff --git a/src/gallium/auxiliary/tgsi/tgsi_scan.c

[Mesa-dev] [PATCH 03/16] gallium/radeon: clean up emit_declaration for temporaries

From: Nicolai Hähnle In the alloca'd array case, no longer create redundant and unused allocas for the individual elements; create getelementptrs instead. Reviewed-by: Tom Stellard --- .../drivers/radeon/radeon_setup_tgsi_llvm.c| 27

[Mesa-dev] [PATCH v2 00/16] radeonsi: improve handling of temporary arrays

Hi, this is a respin of the series which scans the shader's TGSI to determine which channels of an array are actually written to. Most of the st/mesa changes have become unnecessary. Most of the radeon-specific part stays the same. For one F1 2015 shader, it reduces the scratch size from 132096

[Mesa-dev] [PATCH 04/16] gallium/radeon: simplify radeon_llvm_emit_fetch for direct array addressing

From: Nicolai Hähnle We can use the pointer stored in the temps array directly. Reviewed-by: Tom Stellard --- src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 5 - 1 file changed, 5 deletions(-) diff --git

[Mesa-dev] [PATCH 02/16] st_glsl_to_tgsi: use calloc the way it's meant to be used

From: Nicolai Hähnle --- 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/mesa/state_tracker/st_glsl_to_tgsi.cpp index 362559f..89e5c4d 100644 ---

[Mesa-dev] [PATCH 05/16] gallium/radeon: simplify radeon_llvm_emit_store for direct array addressing

From: Nicolai Hähnle We can use the pointer stored in the temps array directly. Reviewed-by: Tom Stellard --- src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 7 --- 1 file changed, 7 deletions(-) diff --git

[Mesa-dev] [PATCH 07/16] gallium/radeon: clarify the comment on the array alloca heuristic

From: Nicolai Hähnle --- .../drivers/radeon/radeon_setup_tgsi_llvm.c| 29 ++ 1 file changed, 19 insertions(+), 10 deletions(-) diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c

[Mesa-dev] [PATCH 08/16] gallium/radeon: extract common lookup code into get_temp_array function

From: Nicolai Hähnle --- .../drivers/radeon/radeon_setup_tgsi_llvm.c| 73 -- 1 file changed, 40 insertions(+), 33 deletions(-) diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c

Re: [Mesa-dev] [PATCH] st/mesa: in ATI fs don't assume TEMP0=REG0

Thanks. I forgot to add that this should also be applied to the 12.0 stable branch. MM On 08/10/2016 03:07 PM, Marek Olšák wrote: Pushed, thanks. Marek On Mon, Aug 8, 2016 at 12:48 AM, Miklós Máté wrote: The temporaries are allocated dynamically. Signed-off-by: Miklós

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On 08/09/2016 07:30 PM, Kenneth Graunke wrote: > On Haswell (GL 3.3): > > total instructions in shared programs: 6208759 -> 6203860 (-0.08%) > instructions in affected programs: 856541 -> 851642 (-0.57%) > helped: 3157 > HURT: 113 > LOST: 7 > GAINED: 15 > > On Broadwell (GL 4.4): > > total

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On Wednesday, August 10, 2016 10:02:12 AM PDT Erik Faye-Lund wrote: > On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke > wrote: > > On Haswell (GL 3.3): > > > > total instructions in shared programs: 6208759 -> 6203860 (-0.08%) > > instructions in affected programs: 856541

Re: [Mesa-dev] [PATCH] st/mesa: in ATI fs don't assume TEMP0=REG0

On Wed, Aug 10, 2016 at 8:40 PM, Miklós Máté wrote: > Thanks. I forgot to add that this should also be applied to the 12.0 stable > branch. OK. I've sent it to mesa-stable for inclusion into 12.0. Marek ___ mesa-dev mailing list

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

Kenneth Graunke writes: > On Haswell (GL 3.3): > > total instructions in shared programs: 6208759 -> 6203860 (-0.08%) > instructions in affected programs: 856541 -> 851642 (-0.57%) > helped: 3157 > HURT: 113 > LOST: 7 > GAINED: 15 > > On Broadwell (GL 4.4): > > total

Re: [Mesa-dev] [PATCH 3/6] nir: Turn -(b2f(a) + b2f(b) >= 0 into !(a || b).

On Wednesday, August 10, 2016 10:58:49 AM PDT Ian Romanick wrote: > On 08/09/2016 07:30 PM, Kenneth Graunke wrote: > > On Haswell (GL 3.3): > > > > total instructions in shared programs: 6211485 -> 6211427 (-0.00%) > > instructions in affected programs: 16260 -> 16202 (-0.36%) > > helped: 25 > >

Re: [Mesa-dev] [PATCH 4/6] nir: Introduce a nir_opt_move_comparisons() pass.

On 08/09/2016 07:30 PM, Kenneth Graunke wrote: > This tries to move comparisons (a common source of boolean values) > closer to their first use. For GPUs which use condition codes, > this can eliminate a lot of temporary booleans and comparisons > which reload the condition code register based on

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On Wed, Aug 10, 2016 at 1:53 PM, Eric Anholt wrote: > Kenneth Graunke writes: > >> On Haswell (GL 3.3): >> >> total instructions in shared programs: 6208759 -> 6203860 (-0.08%) >> instructions in affected programs: 856541 -> 851642 (-0.57%) >> helped: 3157

Re: [Mesa-dev] [PATCH 1/6] nir: Convert ineg(b2i(a)) to a if it's a boolean.

On Wednesday, August 10, 2016 10:53:17 AM PDT Ian Romanick wrote: > My most "recent" work in this area is in my appropriately named > assume-CMS-in-precompile branch. I should just rename all my branches > idr-got-distracted-#. Thanks, I tried to look for your recent work on this, but in

[Mesa-dev] [PATCH] st/mesa: in ATI fs don't assume TEMP0=REG0

From: Miklós Máté The temporaries are allocated dynamically. Signed-off-by: Miklós Máté Signed-off-by: Marek Olšák Cc: 12.0 --- src/mesa/state_tracker/st_atifs_to_tgsi.c | 5 +++-- 1 file changed, 3

Re: [Mesa-dev] [PATCH 1/6] nir: Convert ineg(b2i(a)) to a if it's a boolean.

My most "recent" work in this area is in my appropriately named assume-CMS-in-precompile branch. I should just rename all my branches idr-got-distracted-#. On 08/09/2016 07:30 PM, Kenneth Graunke wrote: > On Haswell (GL 3.3): > > total instructions in shared programs: 6211678 -> 6211584

Re: [Mesa-dev] [PATCH 6/6] i965: Use the nir_move_comparisons pass.

On 08/09/2016 07:30 PM, Kenneth Graunke wrote: > On Haswell (GL 3.3): > > total instructions in shared programs: 6211427 -> 6210079 (-0.02%) > instructions in affected programs: 219356 -> 218008 (-0.61%) > helped: 567 > HURT: 132 > > No spill/fill changes. > > LOST: 0 > GAINED: 4 > > On

Re: [Mesa-dev] [PATCH 2/7] gallium/auxiliary: Add u_bitcast.h header.

On Thu, Jul 28, 2016 at 9:55 PM, Connor Abbott wrote: > On Thu, Jul 28, 2016 at 7:58 PM, Roland Scheidegger > wrote: >> Am 29.07.2016 um 00:35 schrieb Matt Turner: >>> --- >>> src/gallium/auxiliary/Makefile.sources | 1 + >>>

Re: [Mesa-dev] [PATCH] swrast: fix active attribs with atifragshader

On 08/04/2016 06:51 PM, Brian Paul wrote: On 08/04/2016 10:40 AM, Miklós Máté wrote: On 07/14/2016 10:43 PM, Miklós Máté wrote: On 07/07/2016 07:11 PM, Miklós Máté wrote: On 06/26/2016 09:48 PM, Miklós Máté wrote: Only include the ones that can be used by the shader. This fixes texture

Re: [Mesa-dev] [PATCH 4/6] nir: Introduce a nir_opt_move_comparisons() pass.

On Wed, Aug 10, 2016 at 11:14 AM, Ian Romanick wrote: > On 08/09/2016 07:30 PM, Kenneth Graunke wrote: > > This tries to move comparisons (a common source of boolean values) > > closer to their first use. For GPUs which use condition codes, > > this can eliminate a lot of

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

Connor Abbott writes: > On Wed, Aug 10, 2016 at 1:53 PM, Eric Anholt wrote: >> Kenneth Graunke writes: >> >>> On Haswell (GL 3.3): >>> >>> total instructions in shared programs: 6208759 -> 6203860 (-0.08%) >>> instructions in

Re: [Mesa-dev] [PATCH] glsl: look for frag data bindings with [0] tacked onto the end for arrays

Ilia Mirkin writes: > The GL spec is very unclear on this point. Apparently this is discussed > without resolution in the closed Khronos bugtracker at > https://cvs.khronos.org/bugzilla/show_bug.cgi?id=7829 . The > recommendation is to allow dropping the [0] for looking up

[Mesa-dev] [Bug 97214] X not running with error "Failed to make EGL context current"

https://bugs.freedesktop.org/show_bug.cgi?id=97214 Nicolas Boichat changed: What|Removed |Added Status|NEW |ASSIGNED ---

[Mesa-dev] [PATCH] egl/dri2: dri2_make_current: Release previous context's display

eglMakeCurrent can also be used to change the active display. In that case, we need to decrement ref_count of the previous display (possibly destroying it), and increment it on the next display. Also, old_dsurf/old_rsurf cannot be non-NULL if old_ctx is NULL, so we only need to test if old_ctx is

Re: [Mesa-dev] [PATCH] nir: Add an IO scalarizing pass using the intrinsic's first_component.

On Friday, August 5, 2016 4:27:03 PM PDT Eric Anholt wrote: > vc4 wants to have per-scalar IO load/stores so that dead code elimination > can happen on a more granular basis, which it has been doing in the > backend using a multiplication by 4 of the intrinsic's driver_location. > We can represent

[Mesa-dev] [PATCH] nir: Add support for scalarizing load_interpolated_input.

Signed-off-by: Kenneth Graunke --- src/compiler/nir/nir_lower_io_to_scalar.c | 7 +-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/src/compiler/nir/nir_lower_io_to_scalar.c b/src/compiler/nir/nir_lower_io_to_scalar.c index f2345d5..8be75e4 100644 ---

Re: [Mesa-dev] [PATCH 02/13] glsl: remove dead builtins before assigning varying locations

Timothy Arceri writes: > Builtins already have locations assigned so this shouldn't > changing anything. We want to call it earlier so we can tranform "change" Other than that, 1-4 are: Reviewed-by: Eric Anholt and I'll see if I can get to

Re: [Mesa-dev] [PATCH 1/2] aubinator: Add a new tool called Aubinator to the src/intel/tools folder.

On Tue, Aug 9, 2016 at 4:52 PM, Sirisha Gandikota wrote: > From: Kristian Høgsberg Kristensen > > The Aubinator tool is designed to help the driver developers in debugging > the driver functionality by decoding the data in the .aub files. >

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On Wed, Aug 10, 2016 at 2:24 PM, Kenneth Graunke wrote: > On Wednesday, August 10, 2016 10:02:12 AM PDT Erik Faye-Lund wrote: >> On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke >> wrote: >> > On Haswell (GL 3.3): >> > >> > total instructions in

Re: [Mesa-dev] [PATCH 2/2] aubinator: Fix the tool to correctly decode the DWords

On Tue, Aug 9, 2016 at 4:52 PM, Sirisha Gandikota wrote: > From: Sirisha Gandikota > > Several fixes have been added as part of this as listed below: > > 1) Fix the mask and add disassembler handling for STATE_DS, STATE_HS > as the mask

[Mesa-dev] [Bug 97231] GL_DEPTH_CLAMP doesn't clamp to the far plane

https://bugs.freedesktop.org/show_bug.cgi?id=97231 --- Comment #16 from Roland Scheidegger --- (In reply to Jules Blok from comment #15) > Created attachment 125671 [details] > apitrace file llvmpipe > > Finally got an apitrace running on llvmpipe by using a VM. Sorry for

[Mesa-dev] [Bug 97285] Darkness in Dota 2 after Patch "Make Gallium's BlitFramebuffer follow the GL 4.4 sRGB rules"

https://bugs.freedesktop.org/show_bug.cgi?id=97285 --- Comment #3 from Kenneth Graunke --- It might also be worth testing with and without multisampling... -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the

Re: [Mesa-dev] Moving amdgpu/addrlib into a git submodule

Hi, On Wednesday, 10 August 2016 15:19:06 CEST Nicolai Hähnle wrote: > On 09.08.2016 22:12, Dave Airlie wrote: > >> > >> tbh, git submodules are more annoying than they need to be, and I'm > >> not really terribly excited to use that for something that is a build > >> dependency. Maybe just move

Re: [Mesa-dev] [PATCH 3/6] nir: Turn -(b2f(a) + b2f(b) >= 0 into !(a || b).

On 08/09/2016 07:30 PM, Kenneth Graunke wrote: > On Haswell (GL 3.3): > > total instructions in shared programs: 6211485 -> 6211427 (-0.00%) > instructions in affected programs: 16260 -> 16202 (-0.36%) > helped: 25 > HURT: 37 > > On Broadwell (GL 4.4): > > total instructions in shared programs:

Re: [Mesa-dev] [PATCH] nir: Add support for scalarizing load_interpolated_input.

Kenneth Graunke writes: > Signed-off-by: Kenneth Graunke > --- > src/compiler/nir/nir_lower_io_to_scalar.c | 7 +-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git a/src/compiler/nir/nir_lower_io_to_scalar.c >

Re: [Mesa-dev] [PATCH 2/2] aubinator: Fix the tool to correctly decode the DWords

On Tue, Aug 9, 2016 at 4:52 PM, Sirisha Gandikota < sirisha.gandik...@intel.com> wrote: > From: Sirisha Gandikota > > Several fixes have been added as part of this as listed below: > > 1) Fix the mask and add disassembler handling for STATE_DS, STATE_HS > as the mask

Re: [Mesa-dev] [PATCH v2 00/16] radeonsi: improve handling of temporary arrays

On Wed, Aug 10, 2016 at 9:23 PM, Nicolai Hähnle wrote: > Hi, > > this is a respin of the series which scans the shader's TGSI to determine > which channels of an array are actually written to. Most of the st/mesa > changes have become unnecessary. Most of the radeon-specific

Re: [Mesa-dev] [PATCH] egl/dri2: dri2_make_current: Release previous context's display

On Wed, Aug 10, 2016 at 9:44 AM, Michel Dänzer wrote: > On 10/08/16 03:00 PM, Nicolas Boichat wrote: >> eglMakeCurrent can also be used to change the active display. In that >> case, we need to decrement ref_count of the previous display (possibly >> destroying it), and

[Mesa-dev] [Bug 97285] Darkness in Dota 2 after Patch "Make Gallium's BlitFramebuffer follow the GL 4.4 sRGB rules"

https://bugs.freedesktop.org/show_bug.cgi?id=97285 Matthew Dawson changed: What|Removed |Added Status|NEEDINFO|REOPENED ---

Re: [Mesa-dev] [PATCH 1/6] nir: Convert ineg(b2i(a)) to a if it's a boolean.

On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke wrote: > On Haswell (GL 3.3): > > total instructions in shared programs: 6211678 -> 6211584 (-0.00%) > instructions in affected programs: 18968 -> 18874 (-0.50%) > helped: 62 > HURT: 0 > LOST: 0 > GAINED: 4 > > On Broadwell

Re: [Mesa-dev] [PATCH] egl/dri2: dri2_make_current: Release previous context's display

On 10/08/16 03:00 PM, Nicolas Boichat wrote: > eglMakeCurrent can also be used to change the active display. In that > case, we need to decrement ref_count of the previous display (possibly > destroying it), and increment it on the next display. > > Also, old_dsurf/old_rsurf cannot be non-NULL if

Re: [Mesa-dev] [PATCH 0/8] More state optimizations for st/mesa

On Tue, Aug 9, 2016 at 12:58 PM, Nicolai Hähnle wrote: > I sent comments on patches 4 & 6. Apart from that, the series is > > Reviewed-by: Nicolai Hähnle > > On 07.08.2016 03:12, Marek Olšák wrote: >> >> PS: In order to make reviewing easier, all my

Re: [Mesa-dev] [PATCH 4/8] st/mesa: determine states used or affected by shaders at compile time

On Tue, Aug 9, 2016 at 12:56 PM, Nicolai Hähnle wrote: > On 07.08.2016 03:12, Marek Olšák wrote: >> >> From: Marek Olšák >> >> At compile time, each shader determines which ST_NEW flags should be set >> at shader bind time. >> >> This just sets the new

Re: [Mesa-dev] [PATCH] st/nine: Fix invalid attempt to use indirect draws.

I don't have commit access so someone else will need to push this out. Trevor On Aug 10, 2016 2:56 AM, "Marek Olšák" wrote: > Reviewed-by: Marek Olšák > > Marek > > On Wed, Aug 10, 2016 at 5:28 AM, Trevor Davenport > wrote: >

Re: [Mesa-dev] [PATCH 0/4] RadeonSI: GPU hang fix + 3 small changes

This series is, Reviewed-by: Edward O'Callaghan On 08/10/2016 09:34 AM, Marek Olšák wrote: > Hi, > > The first patch is pretty important and fixes the root cause of Overlord and > Witcher 2 hangs. The GLSL revert can be reverted after that. > > Please review. >

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On Wed, Aug 10, 2016 at 4:30 AM, Kenneth Graunke wrote: > On Haswell (GL 3.3): > > total instructions in shared programs: 6208759 -> 6203860 (-0.08%) > instructions in affected programs: 856541 -> 851642 (-0.57%) > helped: 3157 > HURT: 113 > LOST: 7 > GAINED: 15 > > On

[Mesa-dev] [Bug 97285] Darkness in Dota 2 after Patch "Make Gallium's BlitFramebuffer follow the GL 4.4 sRGB rules"

https://bugs.freedesktop.org/show_bug.cgi?id=97285 Bug ID: 97285 Summary: Darkness in Dota 2 after Patch "Make Gallium's BlitFramebuffer follow the GL 4.4 sRGB rules" Product: Mesa Version: unspecified Hardware: Other

[Mesa-dev] [Bug 97285] Darkness in Dota 2 after Patch "Make Gallium's BlitFramebuffer follow the GL 4.4 sRGB rules"

https://bugs.freedesktop.org/show_bug.cgi?id=97285 Ian Romanick changed: What|Removed |Added Status|NEW |NEEDINFO

Re: [Mesa-dev] [PATCH 02/13] glsl: remove dead builtins before assigning varying locations

On Wed, 2016-08-10 at 00:01 -0700, Eric Anholt wrote: > Timothy Arceri writes: > > > > > Builtins already have locations assigned so this shouldn't > > changing anything. We want to call it earlier so we can tranform > >   "change" > > Other than that, 1-4 are: >

Re: [Mesa-dev] [PATCH 2/6] nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences.

On 08/10/2016 12:50 PM, Eric Anholt wrote: > Connor Abbott writes: > >> On Wed, Aug 10, 2016 at 1:53 PM, Eric Anholt wrote: >>> Kenneth Graunke writes: >>> On Haswell (GL 3.3): total instructions in shared programs:

Re: [Mesa-dev] [PATCH] st/nine: Fix invalid attempt to use indirect draws.

Reviewed-by: Marek Olšák Marek On Wed, Aug 10, 2016 at 5:28 AM, Trevor Davenport wrote: > Since commit 6d7177f01b231e9fe79a558c28d2b562a218d7ea, radeonsi > would take a different path if info->indirect_params was not > initialized properly.

[Mesa-dev] Bisected: Make check is failing as of 549222f5

Hello, I just got home and did a pull/build, and make check is failing as of a very recent commit today... ${MESA_ROOT}/bin/test-driver: line 107: 5863 Segmentation fault (core dumped) "$@" > $log_file 2>&1 FAIL: glsl/tests/uniform-initializer-test PASS: glsl/glcpp/tests/glcpp-test