Re: [Mesa-dev] [PATCH 0/5] pb_buffer & radeonsi textures >= 4 GB

bly just have to deal with fallout when the time comes. In the meantime, other than two comments on patch 2, the series is Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> Marek ___ mesa-dev mailing list mesa-dev@lists.freedesktop.

Re: [Mesa-dev] [PATCH 1/2] gallium: add pipe_context::set_active_query_state for pausing queries

value itself. +``set_active_query_state`` Set whether all current queries except TIME_ELAPSED +are active or paused. Perhaps "all current queries" -> "all current non-driver queries". Probably nobody would want to pause driver queries. Either way, both patches are

Re: [Mesa-dev] [PATCH 2/5] winsys/amdgpu: add support for 64-bit buffer sizes

On 11.04.2016 05:38, Marek Olšák wrote: diff --git a/src/gallium/winsys/radeon/drm/radeon_drm_bo.c b/src/gallium/winsys/radeon/drm/radeon_drm_bo.c index c92a66e..2810dca 100644 --- a/src/gallium/winsys/radeon/drm/radeon_drm_bo.c +++ b/src/gallium/winsys/radeon/drm/radeon_drm_bo.c @@ -718,7

Re: [Mesa-dev] [PATCH 2/5] radeonsi: use guard band clipping

On 12.04.2016 07:45, Marek Olšák wrote: On Tue, Apr 12, 2016 at 1:53 AM, Marek Olšák <mar...@gmail.com> wrote: On Mon, Apr 11, 2016 at 7:37 PM, Nicolai Hähnle <nhaeh...@gmail.com> wrote: On 10.04.2016 17:34, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> G

Re: [Mesa-dev] [PATCH 03/20] radeonsi: add shared memory

On 02.04.2016 08:10, Bas Nieuwenhuizen wrote: Declares the shared memory as a global variable so that LLVM is aware of it and it does not conflict with passes like AMDGPUPromoteAlloca. Signed-off-by: Bas Nieuwenhuizen --- src/gallium/drivers/radeon/radeon_llvm.h

Re: [Mesa-dev] [PATCH 08/20] radeonsi: implement TGSI compute shader creation

On 02.04.2016 08:10, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen --- src/gallium/drivers/radeonsi/si_compute.c | 72 +++ 1 file changed, 54 insertions(+), 18 deletions(-) diff --git

Re: [Mesa-dev] [PATCH 05/20] radeonsi: implement shared atomics

On 02.04.2016 08:10, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen --- src/gallium/drivers/radeonsi/si_shader.c | 89 +++- 1 file changed, 88 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c

Re: [Mesa-dev] [PATCH 13/20] radeonsi: only emit compute shader state when switching shaders

There's a dangling-pointer error waiting to happen with emitted_bo when the last launched shader is deleted. It seems safer to use emitted_shader rather than emitted_bo, and then NULL that in si_delete_compute_state if required. Cheers, Nicolai On 02.04.2016 08:10, Bas Nieuwenhuizen wrote:

Re: [Mesa-dev] [PATCH 16/20] radeonsi: split setting graphics and compute descriptors

On 02.04.2016 08:10, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen --- src/gallium/drivers/radeonsi/si_compute.c | 3 ++ src/gallium/drivers/radeonsi/si_descriptors.c | 60 ++- src/gallium/drivers/radeonsi/si_state.h |

Re: [Mesa-dev] [PATCH 00/20] GL compute shaders for radeonsi

-by: Nicolai Hähnle <nicolai.haeh...@amd.com> The barrier and flushing (patches 17-18) look correct to me, but they definitely require some Gfx <-> CS tests to be sure. Cheers, Nicolai Bas Nieuwenhuizen (20): radeonsi: set shader calling conventions radeonsi: lower compute shad

Re: [Mesa-dev] [PATCH] trace: add missing set_shader_images()

Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 05.04.2016 12:37, Samuel Pitoiset wrote: Signed-off-by: Samuel Pitoiset <samuel.pitoi...@gmail.com> --- src/gallium/drivers/trace/tr_context.c| 39 ++ src/gallium/drivers/trace/tr_dump_

Re: [Mesa-dev] [PATCH] radeonsi: Synchronize a streamout write after read hazard.

Sounds right to me. Do you have a test case that fails without it? Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 11.04.2016 09:04, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen <b...@basnieuwenhuizen.nl> --- src/gallium/drivers/radeonsi/si_descriptors.c |

Re: [Mesa-dev] Newbie task to get started

On 11.04.2016 12:13, Jakob Sinclair wrote: On 2016-04-11 17:25, Ilia Mirkin wrote: On Mon, Apr 11, 2016 at 11:18 AM, Jakob Sinclair wrote: Thanks for the tips! I have been looking at the ARB_shader_image_load_store and ARB_shader_image size extensions for GL

Re: [Mesa-dev] [PATCH 1/2] gallium/radeon: removing dead code for sharing stencil buffers

Series is: Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 11.04.2016 05:43, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> This is a remnant of the times when the DDX was allocating depth-stencil buffers for windows. Now, st/dri allocates them and doesn'

Re: [Mesa-dev] [PATCH 2/5] radeonsi: use guard band clipping

On 10.04.2016 17:34, Marek Olšák wrote: From: Marek Olšák Guard band clipping speeds up rasterization for primitives that are partially off-screen. This change in particular results in small framerate improvements in a wide range of games. Started by Grigori Goronzy

Re: [Mesa-dev] [PATCH 0/5] R600, GCN: Guard Band support

Assuming that this has been tested, and with comments on patch #2 addressed, Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 10.04.2016 17:34, Marek Olšák wrote: Hi, This patch series adds Guard Band support into r600g and radeonsi. It first implements the Guard Band in ra

Re: [Mesa-dev] [PATCH] gallium/radeon: unify checking streamout enable state

Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 08.04.2016 04:00, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> --- src/gallium/drivers/r600/r600_state_common.c | 5 ++--- src/gallium/drivers/radeon/r600_pipe_common.h | 6 ++ src/gallium/dr

Re: [Mesa-dev] [PATCH] radeonsi: fix mask checking when emitting scissors and viewports

Thanks for doing this. Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 08.04.2016 04:00, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> --- src/gallium/drivers/radeonsi/si_state.c | 12 1 file changed, 8 insertions(+), 4 deletions(-) diff

[Mesa-dev] [PATCH 15/20] radeonsi: implement coherent memory access (v2)

From: Nicolai Hähnle <nicolai.haeh...@amd.com> v2: set glc=1 for volatile also on buffers --- volatile should behave like coherent in this regard, which it already did for images in v1. --- src/gallium/drivers/radeonsi/si_shader.c | 17 + 1 file changed, 13 insertions

Re: [Mesa-dev] [PATCH 06.5/20] radeonsi: update shader image descriptor for invalidated buffer

On 15.03.2016 22:04, Nicolai Hähnle wrote: From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_descriptors.c | 22 +- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_descriptors.c b/src/g

[Mesa-dev] [PATCH] radeonsi: cache flush/invalidation for missing PIPE_BARRIER_*_BUFFER bits (v2)

From: Nicolai Hähnle <nicolai.haeh...@amd.com> This fixes arb_shader_image_load_store-host-mem-barrier. v2: flush TC L2 for index buffers on <= CIK (Marek) --- src/gallium/drivers/radeonsi/si_state.c | 14 -- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/sr

Re: [Mesa-dev] [PATCH 3/4] st/mesa: emit sampler view decls in PBO upload shader

On 17.03.2016 12:07, Brian Paul wrote: On 03/17/2016 10:47 AM, Nicolai Hähnle wrote: On 16.03.2016 19:43, Brian Paul wrote: The return type is float, which is what was implicitly used before, AFAICT. Isn't this technically incorrect when the uploaded texture is integer? That was my

Re: [Mesa-dev] [PATCH 22/20] radeonsi: disable early Z if the fragment shader writes to memory

I'll push this series. Cheers, Nicolai Marek On Wed, Mar 16, 2016 at 4:09 AM, Nicolai Hähnle <nhaeh...@gmail.com> wrote: From: Nicolai Hähnle <nicolai.haeh...@amd.com> Empirically, both the EXEC_ON_* flags and LATE_Z are necessary. --- src/gallium/drivers/radeonsi/s

[Mesa-dev] [PATCH] radeonsi: fix out-of-bounds indexing of shader images

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Results are undefined but may not crash. Without this change, out-of-bounds indexing can lead to VM faults and GPU hangs. Constant buffers, samplers, and possibly others will eventually need similar treatment to s

Re: [Mesa-dev] [PATCH 3/3] radeonsi: cache flush/invalidation for missing PIPE_BARRIER_*_BUFFER bits

On 21.03.2016 06:07, Marek Olšák wrote: On Sat, Mar 19, 2016 at 4:37 AM, Nicolai Hähnle <nhaeh...@gmail.com> wrote: From: Nicolai Hähnle <nicolai.haeh...@amd.com> This fixes arb_shader_image_load_store-host-mem-barrier. --- src/gallium/drivers/radeonsi/si_state.c | 7 +-- 1

[Mesa-dev] [PATCH 07/11] radeonsi: add shader buffer support to TGSI_OPCODE_LOAD

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_shader.c | 89 +--- 1 file changed, 70 insertions(+), 19 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index 7

[Mesa-dev] [PATCH 06/11] radeonsi: add shader buffer support to TGSI_OPCODE_ATOM*

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_shader.c | 61 1 file changed, 46 insertions(+), 15 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index b

[Mesa-dev] [PATCH 10/11] radeonsi: enable shader buffer pipe caps

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_pipe.c | 4 ++-- src/gallium/drivers/radeonsi/si_state.h | 2 ++ 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/gallium/drivers/radeonsi/si_

[Mesa-dev] [PATCH 03/11] radeonsi: preload shader buffers in shaders

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_shader.c | 17 + 1 file changed, 17 insertions(+) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index efc00f3..0be886e 100644 --- a/src/g

[Mesa-dev] [PATCH 08/11] radeonsi: add shader buffer support to TGSI_OPCODE_STORE

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_shader.c | 129 ++- 1 file changed, 111 insertions(+), 18 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index d

[Mesa-dev] [PATCH 11/11] docs: mark atomic counters and SSBOs as done for radeonsi

From: Nicolai Hähnle <nicolai.haeh...@amd.com> While I'm at it, add the image extensions in the GL ES section of GL3.txt. --- docs/GL3.txt | 12 ++-- docs/relnotes/11.3.0.html | 4 +++- 2 files changed, 9 insertions(+), 7 deletions(-) diff --git a/docs/GL3.txt b/do

[Mesa-dev] [PATCH 09/11] radeonsi: add shader buffer support to TGSI_OPCODE_RESQ

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_shader.c | 17 +++-- 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index 1a0f75b..87dc229

[Mesa-dev] [PATCH 05/11] radeonsi: add offset parameter to buffer_append_args

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_shader.c | 15 ++- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index 1313134..b5e962f

[Mesa-dev] [PATCH 02/11] radeonsi: implement set_shader_buffers

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_descriptors.c | 61 +- src/gallium/drivers/radeonsi/si_pipe.h| 1 + src/gallium/drivers/radeonsi/si_shader.c | 5 +- src/gallium/drivers/radeonsi/si_shader.h

[Mesa-dev] [PATCH 00/11] radeonsi: shader buffer support (atomic counters, ssbo)

Hi, since shader images have laid most of the foundation, here are shader buffers now. This is the last extension missing for OpenGL 4.2 (we still need to turn on GLSL 4.2, but I think that only involves flipping a bit). As with shader images, this extension needs bleeding edge LLVM - this time,

[Mesa-dev] [PATCH 04/11] radeonsi: adjust buffer_append_args to take a 128 bit resource

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Move the buffer resource extraction code out into its own function. --- src/gallium/drivers/radeonsi/si_shader.c | 30 +++--- 1 file changed, 23 insertions(+), 7 deletions(-) diff --git a/src/gallium/drivers/ra

[Mesa-dev] [PATCH 01/11] radeonsi: move resetting of constant buffers into a separate function

From: Nicolai Hähnle <nicolai.haeh...@amd.com> This will be re-used for shader buffers. --- src/gallium/drivers/radeonsi/si_descriptors.c | 41 +-- 1 file changed, 26 insertions(+), 15 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_descriptors.c b/src/g

Re: [Mesa-dev] [PATCH] radeonsi: fix out-of-bounds indexing of shader images

On 21.03.2016 15:41, Nicolai Hähnle wrote: From: Nicolai Hähnle <nicolai.haeh...@amd.com> Results are undefined but may not crash. Without this change, out-of-bounds indexing can lead to VM faults and GPU hangs. Constant buffers, samplers, and possibly others will eventually need s

Re: [Mesa-dev] [PATCH] radeonsi: fix 2D array MSAA failures since image support landed

->array_size; } else if (type == V_008F1C_SQ_RSRC_IMG_CUBE) Looks good to me. Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH] radeonsi: silence a coverity warning

From: Nicolai Hähnle <nicolai.haeh...@amd.com> The following Coverity warning 5378tmpl.fetch_args = atomic_fetch_args; 5379tmpl.emit = atomic_emit; >>> CID 1357115: Uninitialized variables (UNINIT) >>> Using uninitialized value "

Re: [Mesa-dev] [PATCH 3/3] mesa/st: Remove GLSLVersion clamping

If you're going to abolish the clamping entirely (which is admittedly a matter of taste), I'd rather remove the glsl_feature_level variable and assign to consts->GLSLVersion directly. Cheers, Nicolai On 22.03.2016 22:21, Edward O'Callaghan wrote: Signed-off-by: Edward O'Callaghan

[Mesa-dev] [PATCH] st/mesa: use RGBA instead of BGRA for SRGB_ALPHA

From: Nicolai Hähnle <nicolai.haeh...@amd.com> This fixes a regression introduced by commit a8eea696 "st/mesa: honour sized internal formats in st_choose_format (v2)". Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94657 --- src/mesa/state_tracker/st_format.c | 2 +- 1

[Mesa-dev] [PATCH] st/mesa: use the texture view's format for render-to-texture

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Aside from the bug below, it fixes a simplistic test I've written locally, and I see no regression in Piglit for radeonsi. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94595 Cc: mesa-sta...@lists.freedesktop.org --- src/mesa/state_t

Re: [Mesa-dev] [PATCH 3/4] st/mesa: emit sampler view decls in PBO upload shader

On 16.03.2016 19:43, Brian Paul wrote: The return type is float, which is what was implicitly used before, AFAICT. Isn't this technically incorrect when the uploaded texture is integer? To be honest, it's not really clear to me what the return type is supposed to mean. I don't think any of

Re: [Mesa-dev] [PATCH 06/20] radeonsi: implement set_shader_images

On 17.03.2016 18:13, Marek Olšák wrote: On Tue, Mar 15, 2016 at 7:28 PM, Nicolai Hähnle <nhaeh...@gmail.com> wrote: From: Nicolai Hähnle <nicolai.haeh...@amd.com> Whether DCC is disabled depends on the access flags with which the image is bound: image_load supports DCC, but stor

Re: [Mesa-dev] [PATCH mesa 1/6] tgsi_build: Fix return of uninitialized memory in tgsi_*_instruction_memory

: 3243b6fc97 ("tgsi: add Texture and Format to tgsi_instruction_memory") Cc: Nicolai Hähnle <nicolai.haeh...@amd.com> Signed-off-by: Hans de Goede <hdego...@redhat.com> Thanks. Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/auxiliary/tgsi/tg

Re: [Mesa-dev] [PATCH 1/4] gallium/radeon: remove old CS tracing

Series is: Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 17.03.2016 10:15, Marek Olšák wrote: Ping On Fri, Mar 11, 2016 at 4:01 PM, Marek Olšák <mar...@gmail.com> wrote: From: Marek Olšák <marek.ol...@amd.com> Cons: - it was only integrated in r600g - it doesn

Re: [Mesa-dev] [PATCH 1/3] radeon/r600_query.c: Minor style fix

On 22.03.2016 22:21, Edward O'Callaghan wrote: Signed-off-by: Edward O'Callaghan <eocallag...@alterapraxis.com> --- src/gallium/drivers/radeon/r600_query.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> diff --git a

Re: [Mesa-dev] [PATCH 3/4] expose ARB_robust_buffer_access_behavior

Capitalize the commit title. It might be nice to split this patch up into mesa/main and Gallium parts. Also... On 04.04.2016 06:41, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen --- src/gallium/docs/source/screen.rst | 4 +++-

Re: [Mesa-dev] [PATCH 2/4] radeonsi: use bounded indexing for samplers

Patches 1 & 2 are Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 04.04.2016 06:41, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen <b...@basnieuwenhuizen.nl> --- src/gallium/drivers/radeonsi/si_shader.c | 5 - 1 file changed, 4 insertions(+), 1 d

[Mesa-dev] [PATCH] radeonsi: gate PIPE_CAP_SHADER_BUFFER_OFFSET_ALIGNMENT by LLVM version

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Otherwise we incorrectly claim ARB_ssbo support even with older LLVM versions. --- src/gallium/drivers/radeonsi/si_pipe.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/g

Re: [Mesa-dev] [PATCH] radeonsi: don't expose ETC2 on Polaris

Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 13.04.2016 09:34, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> not supported --- src/gallium/drivers/radeonsi/si_state.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/driv

Re: [Mesa-dev] [PATCH] r600g: fix typo in r600 register definitions

Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 08.04.2016 18:12, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> --- src/gallium/drivers/r600/r600d.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/drivers/r600/r600d.h b

Re: [Mesa-dev] [PATCH] build: Fix LLVM shared library detection on systems with libLLVM.so.

On 13.04.2016 08:25, Emil Velikov wrote: On 13 April 2016 at 00:17, Nicolai Hähnle <nhaeh...@gmail.com> wrote: On 12.04.2016 12:58, Kenneth Graunke wrote: Arch Linux ships libLLVM.so.3.7.1 rather than libLLVM-3.7.1.so; Mesa's configure script failed to find LLVM shared libraries in thi

[Mesa-dev] [PATCH] gallium/radeon: handle failure when mapping staging buffer

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Cc: "11.1 11.2" <mesa-sta...@lists.freedesktop.org> --- src/gallium/drivers/radeon/r600_buffer_common.c | 4 1 file changed, 4 insertions(+) diff --git a/src/gallium/drivers/radeon/r600_buffer_common.c b/src/

[Mesa-dev] [PATCH] radeonsi: mark ssbo and images descriptor pointers dirty at beginning of CS

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Without this, we were getting non-deterministic VM faults under high pressure. --- src/gallium/drivers/radeonsi/si_descriptors.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/gallium/drivers/radeonsi/si_descriptors.c b/src/g

[Mesa-dev] [PATCH 1/2] radeonsi: expclear must be disabled on first Z/S clear

From: Nicolai Hähnle <nicolai.haeh...@amd.com> The documentation and the HW team say so. --- src/gallium/drivers/radeonsi/si_blit.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_blit.c b/src/gallium/drivers/radeonsi/si_blit.c

[Mesa-dev] [PATCH 2/2] radeonsi: work around an MSAA fast stencil clear problem

From: Nicolai Hähnle <nicolai.haeh...@amd.com> A piglit test (arb_texture_multisample-stencil-clear) has been sent. This problem was discovered analyzing Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93767 --- src/gallium/drivers/radeonsi/si_state.c | 18 +++---

[Mesa-dev] [PATCH 0/9] radeonsi: decompression cleanups and ZS optimization

Hi, this series contains a bunch of minor cleanups to the various decompression routines. The biggest functional change is that we decompress Z & S in-place simultaneously now when asked for decompressing both, while at the same time being slightly more precise about which of the two we

[Mesa-dev] [PATCH 1/9] gallium/util: add u_bit_consecutive for generating a consecutive range of bits

From: Nicolai Hähnle <nicolai.haeh...@amd.com> There are some undefined behavior subtleties, so having a function to match the u_bit_scan_consecutive_range makes sense. --- src/gallium/auxiliary/util/u_math.h | 12 1 file changed, 12 insertions(+) diff --git a/src/gallium/aux

[Mesa-dev] [PATCH 3/9] radeonsi: use MIN2 instead of expanded ?: operator

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_blit.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_blit.c b/src/gallium/drivers/radeonsi/si_blit.c index 2b00da9..6db73f7 100644 --- a/src/g

Re: [Mesa-dev] Mesa (master): 29 new commits

On 22.04.2016 12:29, Nicolai Hähnle wrote: On 20.04.2016 23:02, Michel Dänzer wrote: On 21.04.2016 02:42, Marek Olšák wrote: On Thu, Apr 14, 2016 at 9:29 AM, Michel Dänzer <mic...@daenzer.net> wrote: On 14.04.2016 11:37, Michel Dänzer wrote: On 12.04.2016 21:33, Marek =?UNKNOWN?B?T2zF

[Mesa-dev] [PATCH 7/9] radeonsi: si_blit_decompress_depth is only used for staging

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_blit.c | 26 +- 1 file changed, 5 insertions(+), 21 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_blit.c b/src/gallium/drivers/radeonsi/si_blit.c index 39ef98e..09f3207

[Mesa-dev] [PATCH 6/9] radeonsi: only decompress the required ZS planes from si_blit

From: Nicolai Hähnle <nicolai.haeh...@amd.com> This happens to "fix" a rendering bug in KotOR2, because it avoids a still not quite understood bug with MSAA fast stencil clear decompress. For the stencil clear bug, I have sent a piglit test (arb_texture_multisample-stencil-c

[Mesa-dev] [PATCH 2/9] radeonsi: fix brace style

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_blit.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_blit.c b/src/gallium/drivers/radeonsi/si_blit.c index 7b028c1..2b00da9 100644 --- a/src/g

[Mesa-dev] [PATCH 5/9] radeonsi: decompress Z & S planes in one pass

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/gallium/drivers/radeonsi/si_blit.c | 101 - 1 file changed, 73 insertions(+), 28 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_blit.c b/src/gallium/drivers/radeonsi/si_blit.c index 5

[Mesa-dev] [PATCH 8/9] radeonsi: use level mask for early out in si_blit_decompress_color

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Mostly for consistency with the other decompress functions, but note that in the non-DCC decompress case, the function can now early-out in slightly more (albeit probably rare) cases. --- src/gallium/drivers/radeonsi/si_blit.c | 14 ---

[Mesa-dev] [PATCH 9/9] radeonsi: move blend choice out of loop in si_blit_decompress_color

From: Nicolai Hähnle <nicolai.haeh...@amd.com> It does not depend on the level or layer. --- src/gallium/drivers/radeonsi/si_blit.c | 18 +- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_blit.c b/src/gallium/drivers/ra

[Mesa-dev] [PATCH 4/9] radeonsi: early out of si_blit_decompress_depth_in_place based on dirty mask

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Avoid dirtying the db_render_state atom when possible. --- src/gallium/drivers/radeonsi/si_blit.c | 20 ++-- 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_blit.c b/src/gallium/d

[Mesa-dev] [PATCH 1/3] radeon/video: always use the reusable buffer pool

From: Nicolai Hähnle <nicolai.haeh...@amd.com> A semantic error was introduced in a past refactoring that caused the bind parameter to be passed into the use_reusable_pool parameter of buffer_create. Since this clearly makes no sense, and there is no clear reason why the cache _shouldn't_ b

[Mesa-dev] [PATCH 2/3] gallium/radeon: remove use_reusable_pool parameter from r600_init_resource

From: Nicolai Hähnle <nicolai.haeh...@amd.com> All callers set it to true. --- src/gallium/drivers/r600/r600_state_common.c| 2 +- src/gallium/drivers/radeon/r600_buffer_common.c | 7 +++ src/gallium/drivers/radeon/r600_pipe_common.h | 3 +-- src/gallium/drivers/radeon/r600_tex

[Mesa-dev] [PATCH 3/3] winsys/radeon: remove use_reusable_pool parameter from buffer_create

From: Nicolai Hähnle <nicolai.haeh...@amd.com> All callers set this parameter to true. --- src/gallium/drivers/r300/r300_query.c | 2 +- src/gallium/drivers/r300/r300_render.c | 2 +- src/gallium/drivers/r300/r300_screen_buffer.c | 4 ++-- src/gallium/driver

Re: [Mesa-dev] [PATCH 1/2] radeonsi: Lock a mutex when checking scratch relocations.

On 21.04.2016 13:51, Bas Nieuwenhuizen wrote: We can use shaders from multiple contexts, and they were not otherwise locked yet. Ouch... I guess this is why compute scratch buffers used to be per-program? I'm still trying to wrap my head around the possible code paths here... are you sure

Re: [Mesa-dev] [PATCH 2/2] radeonsi: Fix memory leak in error path.

Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 21.04.2016 13:51, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen <b...@basnieuwenhuizen.nl> --- src/gallium/drivers/radeonsi/si_compute.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/gallium/driv

Re: [Mesa-dev] [PATCH 2/4] st/mesa: check return value of begin/end_query

On 20.04.2016 11:13, Samuel Pitoiset wrote: On 04/20/2016 05:43 PM, Nicolai Hähnle wrote: From: Nicolai Hähnle <nicolai.haeh...@amd.com> They can only indicate out of memory conditions, since the other error conditions are caught earlier. --- src/mesa/state_tracker/st_cb_queryobj.

Re: [Mesa-dev] [PATCH 4/4] radeonsi: Print a message when scratch allocation fails.

On 21.04.2016 09:14, Bas Nieuwenhuizen wrote: On Wed, Apr 20, 2016 at 8:33 AM, <eocallag...@alterapraxis.com> wrote: On 2016-04-20 11:46, Nicolai Hähnle wrote: On 19.04.2016 17:50, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen <b...@basnieuwenhuizen.nl> ---

Re: [Mesa-dev] [PATCH 03/10] radeonsi: generalize si_set_constant_buffer

On 20.04.2016 10:47, Marek Olšák wrote: From: Marek Olšák this will be used in the next commit --- src/gallium/drivers/radeonsi/si_descriptors.c | 27 +-- 1 file changed, 17 insertions(+), 10 deletions(-) diff --git

Re: [Mesa-dev] [PATCH 2/2] radeonsi: simplify invalidation of RW buffer bindings

On 21.04.2016 10:58, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> this is a follow-up to the RW buffer cleanup. Any chance you could squash this with the patch in the earlier series for better bisect-ability of the history? Either way, Reviewed-by: Nicolai

Re: [Mesa-dev] [PATCH 02/10] radeonsi: make RW buffer descriptor array global, not per shader stage

On 21.04.2016 07:49, Bas Nieuwenhuizen wrote: On Wed, Apr 20, 2016 at 5:47 PM, Marek Olšák wrote: From: Marek Olšák --- src/gallium/drivers/radeonsi/si_descriptors.c | 50 +-- src/gallium/drivers/radeonsi/si_pipe.h|

Re: [Mesa-dev] [PATCH 2/3] gallium/ddebug: Support compute states.

On 21.04.2016 10:42, Bas Nieuwenhuizen wrote: Note that compute states have a different struct than the other shader states, so we cannot reuse the macro. Couldn't you still share bind and delete? This requires dealing with the possibility of not having TGSI, but that'll be useful for

Re: [Mesa-dev] [PATCH 10/10] radeonsi: decrease GS copy shader user SGPRs to 2

Nice cleanup. Patches 1, 3-10 are: Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 20.04.2016 10:47, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> const buffers are no longer used since the clip plane const buffer was moved to RW buffers --- src/gal

Re: [Mesa-dev] Mesa (master): 29 new commits

On 20.04.2016 23:02, Michel Dänzer wrote: On 21.04.2016 02:42, Marek Olšák wrote: On Thu, Apr 14, 2016 at 9:29 AM, Michel Dänzer wrote: On 14.04.2016 11:37, Michel Dänzer wrote: On 12.04.2016 21:33, Marek =?UNKNOWN?B?T2zFocOhaw==?= wrote: URL:

Re: [Mesa-dev] [PATCH 3/3] gallium/ddebug: Implement launch_grid.

On 21.04.2016 10:42, Bas Nieuwenhuizen wrote: Does not implement dumping info. Why not? Signed-off-by: Bas Nieuwenhuizen --- src/gallium/drivers/ddebug/dd_draw.c | 29 + 1 file changed, 29 insertions(+) diff --git

Re: [Mesa-dev] [PATCH 1/3] gallium/ddebug: Add passthrough for get_compute_param.

Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 21.04.2016 10:42, Bas Nieuwenhuizen wrote: Signed-off-by: Bas Nieuwenhuizen <b...@basnieuwenhuizen.nl> --- src/gallium/drivers/ddebug/dd_screen.c | 12 1 file changed, 12 insertions(+) diff --git a/src/gal

Re: [Mesa-dev] [PATCH] gallium/radeon: implement randomized SDMA testing

Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 20.04.2016 14:00, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> --- src/gallium/drivers/r600/r600_pipe.c | 3 + src/gallium/drivers/radeon/Makefile.sources | 1 + src/gallium/dr

Re: [Mesa-dev] [PATCH] gallium/radeon: nuke the final pre LLVN 3.6 codepath

"LLVN" typo in the subject line. Apart from that, Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 21.04.2016 08:37, Emil Velikov wrote: Missed with commit 100796c15c3 "gallium/radeon: drop support for LLVM 3.5" Cc: Marek Olšák <marek.ol...@amd.com>

Re: [Mesa-dev] [PATCH v2 2/2] radeonsi: Enable loading into CE RAM.

For the series: Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 21.04.2016 05:12, Bas Nieuwenhuizen wrote: We need to enable a bit in the CONTEXT_CONTROL packet for the loads to work. v2: Style issues. Signed-off-by: Bas Nieuwenhuizen <b...@basnieuwenhuizen.nl> ---

Re: [Mesa-dev] [PATCH 1/4] gallium: add bool return to pipe_context::end_query

is an out of memory condition that will soon lead to failures elsewhere anyway. Nicolai > I can't quite see why informing the state tracker of it really helps. Roland Am 20.04.2016 um 17:43 schrieb Nicolai Hähnle: From: Nicolai Hähnle <nicolai.haeh...@amd.com> Even when begin_query

Re: [Mesa-dev] [PATCH 1/4] gallium: add bool return to pipe_context::end_query

On 22.04.2016 11:52, Roland Scheidegger wrote: Am 22.04.2016 um 18:22 schrieb Nicolai Hähnle: On 22.04.2016 08:56, Roland Scheidegger wrote: I don't quite understand why this is necessary. Couldn't you just handle such failures in the driver easily? How? When we need a new query buffer due

Re: [Mesa-dev] Mesa (master): 29 new commits

On 20.04.2016 23:02, Michel Dänzer wrote: On 21.04.2016 02:42, Marek Olšák wrote: On Thu, Apr 14, 2016 at 9:29 AM, Michel Dänzer wrote: On 14.04.2016 11:37, Michel Dänzer wrote: On 12.04.2016 21:33, Marek =?UNKNOWN?B?T2zFocOhaw==?= wrote: URL:

Re: [Mesa-dev] [PATCH 2/2] radeonsi: work around an MSAA fast stencil clear problem

ar was added in 11.2, so these two patches should be added there as well. Thanks, Nicolai On 22.04.2016 17:31, Nicolai Hähnle wrote: From: Nicolai Hähnle <nicolai.haeh...@amd.com> A piglit test (arb_texture_multisample-stencil-clear) has been sent. This problem was discovered

Re: [Mesa-dev] [PATCH] gallium/util: s/Elements/ARRAY_SIZE/

Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 27.04.2016 11:44, Brian Paul wrote: --- src/gallium/auxiliary/util/u_blit.c | 6 +++--- src/gallium/auxiliary/util/u_blitter.c| 8 src/gallium/auxiliary/util/u_dump_defines.c | 12 ++--

Re: [Mesa-dev] [PATCH 1/1] radeonsi: check if value is negative

should happen, though. I'd be happier with just assert()s. (I'd also be okay with an assert(!"error message") if you want to keep the if statement). With such asserts (obviously, the second hunk needs to change a bit more), you can add my Reviewed-by: Nicolai Hähnle <nicolai.haeh...@a

Re: [Mesa-dev] [PATCH 3/3] radeonsi: don't flush DB for fast color clears

r600_screen_clear_buffer is always called with last parameter true, also for the initial HTILE clear, so that doesn't seem quite right to me. Nicolai On 22.04.2016 03:52, Marek Olšák wrote: From: Marek Olšák --- src/gallium/drivers/r600/r600_blit.c | 2 +-

Re: [Mesa-dev] [PATCH 2/3] radeonsi: remove needless cache flushes at the end of CP DMA operations

Patches 1 & 2 are Reviewed-by: Nicolai Hähnle <nicolai.haeh...@amd.com> On 22.04.2016 03:52, Marek Olšák wrote: From: Marek Olšák <marek.ol...@amd.com> not needed AFAIK --- src/gallium/drivers/radeonsi/si_cp_dma.c | 8 1 file changed, 8 deletions(-) diff --g

[Mesa-dev] [PATCH 2/9] mesa/main: fix another undefined left shift

From: Nicolai Hähnle <nicolai.haeh...@amd.com> --- src/mesa/main/uniforms.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/mesa/main/uniforms.c b/src/mesa/main/uniforms.c index a9308d0..7758640 100644 --- a/src/mesa/main/uniforms.c +++ b/src/mesa/main/uniforms.c @@

[Mesa-dev] [PATCH 0/9] Enable gcc's undefined behaviour sanitizer

Hi, this patch series enables the use of gcc's -fsanitize=undefined in Mesa and fixes the bulk of reported issues in basic examples like glxgears. However, many issues remain, and some of them in the GLSL compiler look like they could be real - see below for a sampling. By "enables the use" I

[Mesa-dev] [PATCH 4/9] compiler/list: add and use for_range_list_safe

From: Nicolai Hähnle <nicolai.haeh...@amd.com> This macro avoids undefined behaviour that crashes gcc's ubsan. --- src/compiler/glsl/list.h | 13 + src/compiler/glsl/opt_dead_code_local.cpp | 7 +-- src/compiler/glsl/opt_tree_grafting.cpp | 5 + 3

[Mesa-dev] [PATCH 5/9] compiler/glsl: do not downcast list sentinel

From: Nicolai Hähnle <nicolai.haeh...@amd.com> This crashes gcc's undefined behaviour sanitizer. --- src/compiler/glsl/ir.cpp | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/src/compiler/glsl/ir.cpp b/src/compiler/glsl/ir.cpp index 750f617..d69ab13 100644 ---

[Mesa-dev] [PATCH 1/9] mesa/main: define _NEW_xxx flags as unsigned shifts

From: Nicolai Hähnle <nicolai.haeh...@amd.com> Since 1 << 31 complains about undefined behaviour; the others are changed only for consistency. --- src/mesa/main/mtypes.h | 60 +- 1 file changed, 30 insertions(+), 30 deletions(-) diff

<    2   3   4   5   6   7   8   9   10   11   >