[Mesa-dev] [PATCH 3/6] radeonsi: align vertex buffer descriptor list size for optimal prefetch

From: Marek Olšák --- src/gallium/drivers/radeonsi/si_cp_dma.c | 2 +- src/gallium/drivers/radeonsi/si_descriptors.c | 3 ++- src/gallium/drivers/radeonsi/si_state.c | 2 ++ src/gallium/drivers/radeonsi/si_state.h | 2 ++ 4 files changed, 7 insertions(+), 2 deletions(-) diff

[Mesa-dev] [PATCH 5/6] gallium/radeon: ignore the level parameter in buffer_transfer_map

From: Marek Olšák --- src/gallium/drivers/radeon/r600_buffer_common.c | 9 - 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/src/gallium/drivers/radeon/r600_buffer_common.c b/src/gallium/drivers/radeon/r600_buffer_common.c index 11d2319..d7660ef 100644 --- a/src/gallium

[Mesa-dev] [PATCH 1/6] radeonsi: move CP_DMA_ALIGNMENT definition

From: Marek Olšák --- src/gallium/drivers/radeonsi/si_cp_dma.c | 18 -- src/gallium/drivers/radeonsi/si_pipe.h | 2 ++ 2 files changed, 10 insertions(+), 10 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_cp_dma.c b/src/gallium/drivers/radeonsi/si_cp_dma.c index

[Mesa-dev] [PATCH 6/6] gallium/radeon: use staging for texture read mappings from GTT WC

From: Marek Olšák --- src/gallium/drivers/radeon/r600_texture.c | 9 + 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/src/gallium/drivers/radeon/r600_texture.c b/src/gallium/drivers/radeon/r600_texture.c index 55c8e6f..5d84eb4 100644 --- a/src/gallium/drivers/radeon

[Mesa-dev] [PATCH 2/6] radeonsi: align shader binaries to CP DMA alignment for optimal prefetch

From: Marek Olšák --- src/gallium/drivers/radeonsi/si_shader.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index f8eccdf..9ec6c06 100644 --- a/src/gallium/drivers/radeonsi/si_shader.c

Re: [Mesa-dev] [PATCH 4/6] gallium/radeon: fix performance of buffer readbacks

Please ignore this patch. There are a bunch of piglit regressions. Marek On Thu, Feb 9, 2017 at 12:21 PM, Marek Olšák wrote: > From: Marek Olšák > > We want cached GTT for all non-persistent read mappings. > Set level = 0 on purpose. > > Use dma_copy, because resource_co

[Mesa-dev] [PATCH 4/6] gallium/radeon: fix performance of buffer readbacks

From: Marek Olšák We want cached GTT for all non-persistent read mappings. Set level = 0 on purpose. Use dma_copy, because resource_copy_region causes a failure in the PBO read of piglit/getteximage-luminance. If Rocket League used the READ flag, it should get cached GTT. v2: mask out

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

On Thu, Feb 9, 2017 at 1:52 PM, Ian Romanick wrote: > On 02/08/2017 10:26 AM, Nicolai Hähnle wrote: >> On 07.02.2017 23:54, Matt Turner wrote: >>> On Tue, Feb 7, 2017 at 10:56 AM, Marek Olšák wrote: >>>> On Tue, Feb 7, 2017 at 2:57 AM, Kenneth Graunke >>&g

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

On Thu, Feb 9, 2017 at 6:23 PM, Ian Romanick wrote: > On 02/09/2017 03:03 PM, Marek Olšák wrote: >> On Thu, Feb 9, 2017 at 1:52 PM, Ian Romanick wrote: >>> On 02/08/2017 10:26 AM, Nicolai Hähnle wrote: >>>> On 07.02.2017 23:54, Matt Turner wrote: >>>&g

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

On Thu, Feb 9, 2017 at 8:18 PM, Marek Olšák wrote: > On Thu, Feb 9, 2017 at 6:23 PM, Ian Romanick wrote: >> On 02/09/2017 03:03 PM, Marek Olšák wrote: >>> On Thu, Feb 9, 2017 at 1:52 PM, Ian Romanick wrote: >>>> On 02/08/2017 10:26 AM, Nicolai Hähnle wrote: &

Re: [Mesa-dev] [PATCH] tgsi: remove ureg_label_insn

Reviewed-by: Marek Olšák Marek On Thu, Feb 9, 2017 at 3:36 PM, wrote: > From: Marc-André Lureau > > Unused since commit 2897cb3dba9287011f9c43cd2f214100952370c0. > > Signed-off-by: Marc-André Lureau > --- > src/gallium/auxiliary/tgsi/tgsi_ureg.c | 31 -

Re: [Mesa-dev] [PATCH 06/40] glsl: add helper to convert pointers to uint64_t

Why would you wanna store pointers in the cache? Marek On Tue, Feb 7, 2017 at 4:42 AM, Timothy Arceri wrote: > From: Timothy Arceri > > This will be used to store all pointers in the cache as 64bit ints > allowing us to avoid issues when a 32bit program reads a cached > shader that was created

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

The glmark2 issue is not important, because glthread will never support glmark2. (because glmark2 is a SwapBuffers benchmark like glxgears) Marek On Fri, Feb 10, 2017 at 11:32 AM, Eero Tamminen wrote: > Hi, > > On 09.02.2017 20:27, Ernst Sjöstrand wrote: >> >> 2017-02-09 18:06 GMT+01:00 Eero Tam

Re: [Mesa-dev] [PATCH 39/40] mesa: add helper to create cache name string

If this doesn't use gl_context, it can be in src/util. Marek On Tue, Feb 7, 2017 at 4:42 AM, Timothy Arceri wrote: > --- > src/mesa/main/version.c | 9 + > src/mesa/main/version.h | 3 +++ > 2 files changed, 12 insertions(+) > > diff --git a/src/mesa/main/version.c b/src/mesa/main/versi

Re: [Mesa-dev] [PATCH 40/40] r600/radeonsi: enable glsl/tgsi on-disk cache

I don't see anything radeonsi-specific here. Is there any reason this has to be in radeonsi? Also, all pipe_screen functions must be thread-safe, so a mutex around checking and setting disk_shader_cache should be used. if _mesa_create_cache_version_string was in src/util, you wouldn't have to pas

Re: [Mesa-dev] [PATCH 4/6] gallium/radeon: fix performance of buffer readbacks

On Fri, Feb 10, 2017 at 12:22 PM, Emil Velikov wrote: > On 9 February 2017 at 11:56, Marek Olšák wrote: >> From: Marek Olšák >> >> We want cached GTT for all non-persistent read mappings. >> Set level = 0 on purpose. >> >> Use dma_copy, because resource_

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

On Fri, Feb 10, 2017 at 12:26 PM, Edward O'Callaghan wrote: > > > On 02/08/2017 09:13 AM, Timothy Arceri wrote: >> On Tue, 2017-02-07 at 10:56 +0100, Marek Olšák wrote: >>> On Tue, Feb 7, 2017 at 2:57 AM, Kenneth Graunke >> g> wrote: >>>> On M

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

On Fri, Feb 10, 2017 at 12:48 PM, Edward O'Callaghan wrote: > > > On 02/10/2017 10:36 PM, Marek Olšák wrote: >> On Fri, Feb 10, 2017 at 12:26 PM, Edward O'Callaghan >> wrote: >>> >>> >>> On 02/08/2017 09:13 AM, Timothy Arceri wrote:

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

likely be disabling multithreading on the fly than trying to fix the crash. Marek On Feb 10, 2017 12:58 PM, "Edward O'Callaghan" wrote: > > > On 02/10/2017 10:50 PM, Marek Olšák wrote: > > On Fri, Feb 10, 2017 at 12:48 PM, Edward O'Callaghan > > wrote: >

Re: [Mesa-dev] [PATCH 40/40] r600/radeonsi: enable glsl/tgsi on-disk cache

You might also just create the cache unconditionally in screen_create. Marek On Feb 10, 2017 12:58 PM, "Timothy Arceri" wrote: > > > On 10/02/17 22:25, Marek Olšák wrote: > >> I don't see anything radeonsi-specific here. Is there any rea

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

ying, as to reiterate my point was precise and finely scoped; We should get shader caching i's and t's dotted and crossed and help Timothy get though the last 5% there. Kind Regards, Edward. On 02/11/2017 12:01 AM, Marek Olšák wrote: > FYI. Civilization VI is another game that works with

Re: [Mesa-dev] [PATCH 09/26] glapi: Generate GL API marshalling code from the XML.

On Fri, Feb 10, 2017 at 3:23 PM, Nicolai Hähnle wrote: > On 10.02.2017 15:01, Nicolai Hähnle wrote: >> >> On 08.02.2017 19:03, Marek Olšák wrote: >>> >>> From: Paul Berry >>> >>> This is not yet used in the build, just generated. >>>

Re: [Mesa-dev] [RFC PATCH 1/1] st/dri: add a new driconf option override_glsl_version for ARK games

On Mon, Feb 6, 2017 at 9:26 PM, Samuel Pitoiset wrote: > > > On 02/06/2017 04:45 PM, Eero Tamminen wrote: >> >> Hi, >> >> On 05.02.2017 15:19, Samuel Pitoiset wrote: >>> >>> On 02/03/2017 07:48 PM, Bas Nieuwenhuizen wrote: As far as I can see[1], when the game detects GL 4.3+, the engine

Re: [Mesa-dev] [PATCH 38/40] st/mesa: implement a tgsi on-disk shader cache

On Tue, Feb 7, 2017 at 4:42 AM, Timothy Arceri wrote: > Implements a tgsi cache for the OpenGL state tracker. > --- > src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 217 > + > src/mesa/state_tracker/st_program.c| 133 +- > 2 files changed, 345 in

Re: [Mesa-dev] [PATCH 06/26] egl: Implement __DRI_BACKGROUND_CALLABLE

On Feb 10, 2017 5:55 PM, "Emil Velikov" wrote: On 8 February 2017 at 18:03, Marek Olšák wrote: > +const __DRIbackgroundCallableExtension background_callable_extension = { > + { __DRI_BACKGROUND_CALLABLE, __DRI_BACKGROUND_CALLABLE_VERSION }, > + dri_set_background_cont

Re: [Mesa-dev] [PATCH 10/21] winsys/amdgpu: extract amdgpu_do_add_real_buffer

Assuming you address other people's comments, the first 10 are: Reviewed-by: Marek Olšák Marek On Wed, Feb 8, 2017 at 1:42 PM, Nicolai Hähnle wrote: > From: Nicolai Hähnle > > We will use it for delayed adding of sparse buffers' backing buffers. > --- > src/

Re: [Mesa-dev] [PATCH 14/26] mesa: Connect the generated GL command marshalling code to the build.

On Fri, Feb 10, 2017 at 7:04 PM, Steven Newbury wrote: > On Fri, 2017-02-10 at 17:15 +, Emil Velikov wrote: >> On 8 February 2017 at 18:03, Marek Olšák wrote: >> > From: Eric Anholt >> > >> > v2: Rebase on the Begin/End changes, and just disable this

Re: [Mesa-dev] [PATCH 06/26] egl: Implement __DRI_BACKGROUND_CALLABLE

On Feb 11, 2017 11:42 AM, "Nicolai Hähnle" wrote: On 10.02.2017 18:35, Emil Velikov wrote: > On 10 February 2017 at 17:13, Marek Olšák wrote: > >> Yes, the egl stuff is untested. We can drop it if needed. We mainly need >> GLX >> support for now. >> >

[Mesa-dev] [PATCH 4/4] radeonsi: allow unaligned vertex buffer offsets and strides on VI only

From: Marek Olšák so that we can disable u_vbuf for GL core profiles. SI-CIK fail piglit/draw-vertices, while VI doesn't. --- src/gallium/drivers/radeonsi/si_descriptors.c | 1 - src/gallium/drivers/radeonsi/si_pipe.c| 8 +--- 2 files changed, 5 insertions(+), 4 deletions(-)

[Mesa-dev] [PATCH 3/4] radeonsi: implement uploading zero-stride vertex attribs

From: Marek Olšák This is the only kind of user buffer we can get with the GL core profile. --- src/gallium/drivers/radeonsi/si_state.c | 31 +++ 1 file changed, 23 insertions(+), 8 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_state.c b/src/gallium

[Mesa-dev] [PATCH 0/4] RadeonSI: prerequisites for disabling u_vbuf with GL core profile

Please review. Marek ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH 2/4] radeonsi: implement legacy GL_DOUBLE vertex formats

From: Marek Olšák so that we can disable u_vbuf for GL core profiles. --- src/gallium/drivers/radeonsi/si_shader.c | 78 ++-- src/gallium/drivers/radeonsi/si_shader.h | 4 ++ src/gallium/drivers/radeonsi/si_state.c | 56 --- 3 files changed, 117

[Mesa-dev] [PATCH 1/4] radeonsi: clean up si_get_param

From: Marek Olšák has_streamout is always true --- src/gallium/drivers/radeonsi/si_pipe.c | 30 +++--- 1 file changed, 11 insertions(+), 19 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/gallium/drivers/radeonsi/si_pipe.c index 853d850..8806027

[Mesa-dev] [PATCH 1/2] gallium: let state trackers tell u_vbuf whether user VBOs are possible

From: Marek Olšák This can affect whether u_vbuf will be enabled or not. --- src/gallium/auxiliary/cso_cache/cso_context.c | 9 + src/gallium/auxiliary/cso_cache/cso_context.h | 3 ++- src/gallium/auxiliary/util/u_tests.c | 8 src/gallium/auxiliary/util/u_vbuf.c

[Mesa-dev] [PATCH 2/2] st/mesa: tell u_vbuf that GL core doesn't have user VBOs

From: Marek Olšák I think this only affects radeonsi - VI, because all other drivers using u_vbuf probably don't support GL_DOUBLE, so they won't be affected by this. --- src/mesa/state_tracker/st_context.c | 9 - 1 file changed, 8 insertions(+), 1 deletion(-) diff --git

[Mesa-dev] [PATCH 1/4] gallium/os: add per-thread time clock queries

From: Marek Olšák --- src/gallium/auxiliary/os/os_thread.h | 31 +++ 1 file changed, 31 insertions(+) diff --git a/src/gallium/auxiliary/os/os_thread.h b/src/gallium/auxiliary/os/os_thread.h index 21faf4b..0caf955 100644 --- a/src/gallium/auxiliary/os/os_thread.h

[Mesa-dev] [PATCH 2/4] gallium/u_queue: add util_queue_get_thread_time_nano

From: Marek Olšák --- src/gallium/auxiliary/util/u_queue.c | 10 ++ src/gallium/auxiliary/util/u_queue.h | 2 ++ 2 files changed, 12 insertions(+) diff --git a/src/gallium/auxiliary/util/u_queue.c b/src/gallium/auxiliary/util/u_queue.c index 838464f..4da5d8e 100644 --- a/src/gallium

[Mesa-dev] [PATCH 4/4] gallium/hud: add monitoring of API thread busy status

From: Marek Olšák --- src/gallium/auxiliary/hud/hud_context.c | 3 ++ src/gallium/auxiliary/hud/hud_cpu.c | 60 + src/gallium/auxiliary/hud/hud_private.h | 1 + 3 files changed, 64 insertions(+) diff --git a/src/gallium/auxiliary/hud/hud_context.c b/src

[Mesa-dev] [PATCH 0/4] Gallium HUD: Monitoring thread busy status

Hi, This is for monitoring thread activity with the HUD. Please review. Marek ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH 3/4] gallium/radeon: add a HUD query for monitoring the CS thread activity

From: Marek Olšák --- src/gallium/drivers/radeon/r600_query.c | 25 ++- src/gallium/drivers/radeon/r600_query.h | 1 + src/gallium/drivers/radeon/radeon_winsys.h| 1 + src/gallium/winsys/amdgpu/drm/amdgpu_winsys.c | 2 ++ src/gallium/winsys

[Mesa-dev] [PATCH] gallium/radeon: include SDMA in the GPU load query

From: Marek Olšák --- src/gallium/drivers/radeon/r600_gpu_load.c| 10 +- src/gallium/drivers/radeon/r600_pipe_common.h | 3 +++ 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeon/r600_gpu_load.c b/src/gallium/drivers/radeon/r600_gpu_load.c

[Mesa-dev] [PATCH 2/2] gallium/hud: don't use user vertex buffers

From: Marek Olšák --- src/gallium/auxiliary/hud/hud_context.c | 26 +++--- 1 file changed, 19 insertions(+), 7 deletions(-) diff --git a/src/gallium/auxiliary/hud/hud_context.c b/src/gallium/auxiliary/hud/hud_context.c index 15ca471..9de260c 100644 --- a/src/gallium

[Mesa-dev] [PATCH 1/2] gallium/hud: call u_upload_alloc only once

From: Marek Olšák --- src/gallium/auxiliary/hud/hud_context.c | 37 ++--- 1 file changed, 29 insertions(+), 8 deletions(-) diff --git a/src/gallium/auxiliary/hud/hud_context.c b/src/gallium/auxiliary/hud/hud_context.c index aa63171..15ca471 100644 --- a/src/gallium

Re: [Mesa-dev] [PATCH 3/4] gallium/radeon: add a HUD query for monitoring the CS thread activity

On Sat, Feb 11, 2017 at 10:31 PM, Grazvydas Ignotas wrote: > On Sat, Feb 11, 2017 at 9:58 PM, Marek Olšák wrote: >> From: Marek Olšák >> >> --- >> src/gallium/drivers/radeon/r600_query.c | 25 >> ++- >> src/gallium/dri

[Mesa-dev] [PATCH 1/2] gallium/radeon: add an assertion to texture_transfer_map for app bugs

From: Marek Olšák --- src/gallium/drivers/radeon/r600_texture.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/gallium/drivers/radeon/r600_texture.c b/src/gallium/drivers/radeon/r600_texture.c index 5d84eb4..47aa8b1 100644 --- a/src/gallium/drivers/radeon/r600_texture.c +++ b/src

[Mesa-dev] [PATCH 2/2] vdpau: skip vlVdpOutputSurfacePutBitsNative with a zero-area rectangle

From: Marek Olšák This prevents errors: "EE r600_texture.c:1571 r600_texture_transfer_map - failed to create temporary texture to hold untiled copy" Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99542 --- src/gallium/state_trackers/vdpau/output.c | 7 +++ 1 file

Re: [Mesa-dev] [Mesa-stable] [PATCH] st/mesa: don't pass compare mode for stencil-sampled textures

Reviewed-by: Marek Olšák Marek On Sun, Feb 12, 2017 at 8:20 PM, Ilia Mirkin wrote: > Fixes dEQP-GLES31.functional.stencil_texturing.misc.compare_mode_effect > > Signed-off-by: Ilia Mirkin > Cc: mesa-sta...@lists.freedesktop.org > --- > src/mesa/state_tracker/st_atom_sam

Re: [Mesa-dev] [PATCH] r300g: only allow byteswapped formats on big endian

I'd like some evidence that the bug is caused by r300g and not some common code. Marek On Feb 13, 2017 12:10 PM, "Grazvydas Ignotas" wrote: > On Mon, Feb 13, 2017 at 10:22 AM, Michel Dänzer > wrote: > > On 13/02/17 05:17 PM, Michel Dänzer wrote: > >> On 11/02/17 08:01 AM, Grazvydas Ignotas wro

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

On Mon, Feb 13, 2017 at 11:10 AM, Gregory Hainaut wrote: >> On 11/02/17 06:36 AM, gregory hainaut wrote: >> > >> > On GLX side, I can explain some crashes (didn't debug all neither fail) >> > and I suspect there are related to the previous report. >> > >> > At the context creation a glthread dispa

Re: [Mesa-dev] [PATCH 3/3] Changed shader disassembler number formatting to use integers when the "disasm" debug flag is used. Register types and regions are also now formatted more like in the archit

This is a bad commit message format. See the git history for examples of what an i965 commit looks like. Marek On Mon, Feb 13, 2017 at 12:25 PM, Lonnberg, Toni wrote: > --- > src/mesa/drivers/dri/i965/brw_disasm.c | 101 > +++-- > src/mesa/drivers/dri/i965/brw_eu_em

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

On Mon, Feb 13, 2017 at 3:56 PM, Gregory Hainaut wrote: > Hum, I need to check it. I think I disabled EGL because I didn't have > latest waffle on my Debian (and Nvidia didn't support fully EGL back > then). > > However, I forgot to say that a fair number of piglit tests are based > on GL1 direct

Re: [Mesa-dev] [RFC PATCH 1/4] driconf: add new force_compat_profile option

On Mon, Feb 13, 2017 at 11:57 AM, Samuel Pitoiset wrote: > > > On 02/10/2017 10:05 PM, Emil Velikov wrote: >> >> On 10 February 2017 at 21:04, Emil Velikov >> wrote: >>> >>> Hi Samuel, >>> >>> On 10 February 2017 at 13:41, Samuel Pitoiset >>> wrote: Mesa currently doesn't allow to crea

Re: [Mesa-dev] [PATCH 3/4] gallium/radeon: add a HUD query for monitoring the CS thread activity

On Mon, Feb 13, 2017 at 4:37 PM, Nicolai Hähnle wrote: > On 11.02.2017 20:58, Marek Olšák wrote: >> >> From: Marek Olšák >> >> --- >> src/gallium/drivers/radeon/r600_query.c | 25 >> ++- >> src/gallium/drivers/radeon/

Re: [Mesa-dev] [RFC PATCH 1/4] driconf: add new force_compat_profile option

On Mon, Feb 13, 2017 at 5:04 PM, Samuel Pitoiset wrote: > > > On 02/13/2017 04:51 PM, Marek Olšák wrote: >> >> On Mon, Feb 13, 2017 at 11:57 AM, Samuel Pitoiset >> wrote: >>> >>> >>> >>> On 02/10/2017 10:05 PM, Emil Velikov w

Re: [Mesa-dev] [RFC PATCH 4/4] drirc: add force_compat_profile for Tropico 5

Patches 2-4: Reviewed-by: Marek Olšák Marek On Fri, Feb 10, 2017 at 2:41 PM, Samuel Pitoiset wrote: > Signed-off-by: Samuel Pitoiset > --- > src/mesa/drivers/dri/common/drirc | 4 > 1 file changed, 4 insertions(+) > > diff --git a/src/mesa/drivers/dri/common/drirc &g

Re: [Mesa-dev] [PATCH 4/4] radeonsi: allow unaligned vertex buffer offsets and strides on VI only

On Mon, Feb 13, 2017 at 4:42 PM, Nicolai Hähnle wrote: > On 11.02.2017 17:30, Marek Olšák wrote: >> >> From: Marek Olšák >> >> so that we can disable u_vbuf for GL core profiles. >> >> SI-CIK fail piglit/draw-vertices, while VI doesn't. > &g

Re: [Mesa-dev] Time to merge threaded GL dispatch? (aka glthread)

On Mon, Feb 13, 2017 at 4:14 PM, Gregory Hainaut wrote: > If I remember correctly I got something like 4K-8K fails (total is 40K > tests) when I broke the GL1 restore dispatch function. So it is > around 10%-20%. Maybe it was only the test below 3.1. I was afraid > that various modern extension c

[Mesa-dev] [PATCH] radeonsi: allow unaligned vertex buffer offsets and strides on CIK-VI

From: Marek Olšák So that we can disable u_vbuf for GL core profiles. This is a v2 of the previous VI-only patch. It requires SH_MEM_CONFIG.ALIGNMENT_MODE = UNALIGNED on CIK-VI. --- src/gallium/drivers/radeonsi/si_descriptors.c | 1 - src/gallium/drivers/radeonsi/si_pipe.c| 12

Re: [Mesa-dev] [PATCH 4/4] radeonsi: allow unaligned vertex buffer offsets and strides on VI only

On Mon, Feb 13, 2017 at 5:16 PM, Marek Olšák wrote: > On Mon, Feb 13, 2017 at 4:42 PM, Nicolai Hähnle wrote: >> On 11.02.2017 17:30, Marek Olšák wrote: >>> >>> From: Marek Olšák >>> >>> so that we can disable u_vbuf for GL core profiles. >&

Re: [Mesa-dev] [RFC PATCH 1/4] driconf: add new force_compat_profile option

On Mon, Feb 13, 2017 at 5:06 PM, Marek Olšák wrote: > On Mon, Feb 13, 2017 at 5:04 PM, Samuel Pitoiset > wrote: >> >> >> On 02/13/2017 04:51 PM, Marek Olšák wrote: >>> >>> On Mon, Feb 13, 2017 at 11:57 AM, Samuel Pitoiset >>> wrote: >>>

Re: [Mesa-dev] [PATCH 3/3] radeonsi: use common sendmsg emission function.

For 1 & 3: Reviewed-by: Marek Olšák Marek On Mon, Feb 13, 2017 at 11:15 PM, Dave Airlie wrote: > From: Dave Airlie > > This just ports radeonsi to use the sendmsg common code. > > Signed-off-by: Dave Airlie > --- > src/gallium/drivers/

Re: [Mesa-dev] [RFC PATCH 1/4] driconf: add new force_compat_profile option

On Feb 14, 2017 4:11 AM, "Michel Dänzer" wrote: On 14/02/17 09:28 AM, Samuel Pitoiset wrote: > On 02/13/2017 11:43 PM, Marek Olšák wrote: >> On Mon, Feb 13, 2017 at 5:06 PM, Marek Olšák wrote: >>> On Mon, Feb 13, 2017 at 5:04 PM, Samuel Pitoiset >>> wrot

Re: [Mesa-dev] [PATCH] radeonsi: allow unaligned vertex buffer offsets and strides on CIK-VI

On Feb 14, 2017 9:06 AM, "Nicolai Hähnle" wrote: On 13.02.2017 18:01, Marek Olšák wrote: > From: Marek Olšák > > So that we can disable u_vbuf for GL core profiles. > > This is a v2 of the previous VI-only patch. > It requires SH_MEM_CONFIG.ALIGNMENT_MODE = UNALI

Re: [Mesa-dev] [PATCH] r300g: only allow byteswapped formats on big endian

I've changed my mind. The patch can be merged if nobody disagrees. Marek On Feb 13, 2017 2:10 PM, "Marek Olšák" wrote: > I'd like some evidence that the bug is caused by r300g and not some common > code. > > Marek > > On Feb 13, 2017 12:10 PM, "Gra

Re: [Mesa-dev] [RFC PATCH 1/4] driconf: add new force_compat_profile option

On Feb 14, 2017 2:02 PM, "Samuel Pitoiset" wrote: On 02/14/2017 01:54 PM, Edmondo Tommasina wrote: > > > On Feb 14, 2017 11:50 AM, "Marek Olšák" <mailto:mar...@gmail.com>> wrote: > > > > On Feb 14, 2017 4:11 AM, "Michel Dänzer"

Re: [Mesa-dev] [PATCH 4/6] gallium/radeon: fix performance of buffer readbacks

On Tue, Feb 14, 2017 at 3:05 PM, Emil Velikov wrote: > On 13 February 2017 at 03:24, Michel Dänzer wrote: >> On 10/02/17 08:27 PM, Marek Olšák wrote: >>> On Fri, Feb 10, 2017 at 12:22 PM, Emil Velikov >>> wrote: >>>> On 9 February 2017 at 11:56, Mare

Re: [Mesa-dev] [PATCH 32/32] r600/radeonsi: enable glsl/tgsi on-disk cache

On Tue, Feb 14, 2017 at 1:52 AM, Timothy Arceri wrote: > --- > src/gallium/drivers/r600/r600_pipe.c | 10 ++ > src/gallium/drivers/radeon/r600_pipe_common.c | 2 +- > src/gallium/drivers/radeon/r600_pipe_common.h | 1 + > src/gallium/drivers/radeonsi/si_pipe.c| 11 +

Re: [Mesa-dev] [PATCH v2 1/4] driconf: add allow_higher_compat_version option

Reviewed-by: Marek Olšák Marek On Tue, Feb 14, 2017 at 4:18 PM, Samuel Pitoiset wrote: > Mesa currently doesn't allow to create 3.1+ compatibility profiles > mainly because various features are unimplemented and bugs can > happen. > > However, some buggy apps request a co

Re: [Mesa-dev] [PATCH v2 4/4] drirc: add allow_higher_compat_version for Tropico 5

Rb for the last 3 as well. Marek On Tue, Feb 14, 2017 at 4:18 PM, Samuel Pitoiset wrote: > v2: s/force_compat_profile/allow_higher_compat_version > > Signed-off-by: Samuel Pitoiset > Reviewed-by: Marek Olšák (v1) > --- > src/mesa/drivers/dri/common/drirc | 4 &g

Re: [Mesa-dev] [PATCH] gallium/radeon: add HUD queries for GPU temperature and clocks

On Tue, Feb 14, 2017 at 9:23 AM, Nicolai Hähnle wrote: > On 13.02.2017 23:08, Samuel Pitoiset wrote: >> >> Only the Radeon kernel driver exposed the GPU temperature and >> the shader/memory clocks, this implements the same functionality >> for the AMDGPU kernel driver. >> >> These queries will ret

Re: [Mesa-dev] [PATCH 3/4] gallium/radeon: add a HUD query for monitoring the CS thread activity

On Mon, Feb 13, 2017 at 4:57 PM, Marek Olšák wrote: > On Mon, Feb 13, 2017 at 4:37 PM, Nicolai Hähnle wrote: >> On 11.02.2017 20:58, Marek Olšák wrote: >>> >>> From: Marek Olšák >>> >>> --- >>> src/gallium/drivers/radeon/r600_query.c

Re: [Mesa-dev] [PATCH] radeonsi: allow unaligned vertex buffer offsets and strides on CIK-VI

On Tue, Feb 14, 2017 at 7:24 PM, Matt Arsenault wrote: > > On Feb 13, 2017, at 09:01, Marek Olšák wrote: > > So that we can disable u_vbuf for GL core profiles. > > This is a v2 of the previous VI-only patch. > It requires SH_MEM_CONFIG.ALIGNMENT_MODE = UNALIGNED on CIK-VI.

Re: [Mesa-dev] [PATCH 32/32] r600/radeonsi: enable glsl/tgsi on-disk cache

On Tue, Feb 14, 2017 at 9:53 PM, Timothy Arceri wrote: > On 15/02/17 02:14, Marek Olšák wrote: > >> On Tue, Feb 14, 2017 at 1:52 AM, Timothy Arceri >> wrote: >>> >>> --- >>> src/gallium/drivers/r600/r600_pipe.c | 10 ++ >>>

Re: [Mesa-dev] [PATCH 32/32] r600/radeonsi: enable glsl/tgsi on-disk cache

On Tue, Feb 14, 2017 at 11:47 PM, Timothy Arceri wrote: > > > On 15/02/17 08:35, Marek Olšák wrote: >> >> On Tue, Feb 14, 2017 at 9:53 PM, Timothy Arceri >> wrote: >>> >>> On 15/02/17 02:14, Marek Olšák wrote: >>> >>&g

Re: [Mesa-dev] [PATCH 32/32] r600/radeonsi: enable glsl/tgsi on-disk cache

On Wed, Feb 15, 2017 at 1:49 PM, Nicolai Hähnle wrote: > On 14.02.2017 23:51, Marek Olšák wrote: >> >> On Tue, Feb 14, 2017 at 11:47 PM, Timothy Arceri >> wrote: >>> >>> >>> >>> On 15/02/17 08:35, Marek Olšák wrote: >>>>

Re: [Mesa-dev] [PATCH 1/3] radeon/ac: add emit umsb shared code.

Reviewed-by: Marek Olšák Marek On Thu, Feb 16, 2017 at 4:55 AM, Dave Airlie wrote: > From: Dave Airlie > > Since we shared imsb, makes sense to share umsb. > > Signed-off-by: Dave Airlie > --- > src/amd/common/ac_llvm_build.c | 25 + > src/am

Re: [Mesa-dev] [PATCH 3/3] radeonsi: use shared emit_umsb helper.

Reviewed-by: Marek Olšák Marek On Thu, Feb 16, 2017 at 4:55 AM, Dave Airlie wrote: > From: Dave Airlie > > Signed-off-by: Dave Airlie > --- > src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c | 24 > ++- > 1 file changed, 2 insertions(+), 22 deleti

Re: [Mesa-dev] [PATCH 4/4] radeon/ac: use llvm.amdgcn.sffbh intrinsic instead of AMDGPU.flbit.i32

For 1, 3, 4: Reviewed-by: Marek Olšák Marek On Thu, Feb 16, 2017 at 4:48 AM, Dave Airlie wrote: > From: Dave Airlie > > Use the newer intrinsic. > > Signed-off-by: Dave Airlie > --- > src/amd/common/ac_llvm_build.c | 3 ++- > 1 file changed, 2 insertions(+), 1 dele

[Mesa-dev] [PATCH 4/7] gallium/u_suballoc: use clear_buffer if available

From: Marek Olšák --- src/gallium/auxiliary/util/u_suballoc.c | 21 ++--- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/src/gallium/auxiliary/util/u_suballoc.c b/src/gallium/auxiliary/util/u_suballoc.c index 5aaddbc..8c463c9 100644 --- a/src/gallium/auxiliary

[Mesa-dev] [PATCH 3/7] gallium/util: correctly unref a buffer in u_prim_restart

From: Marek Olšák --- src/gallium/auxiliary/util/u_prim_restart.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/auxiliary/util/u_prim_restart.c b/src/gallium/auxiliary/util/u_prim_restart.c index a4d7c14..e45aa56 100644 --- a/src/gallium/auxiliary/util

[Mesa-dev] [PATCH 2/7] gallium/util: remove unused u_index_modify helpers

From: Marek Olšák --- src/gallium/auxiliary/util/u_index_modify.c | 78 - src/gallium/auxiliary/util/u_index_modify.h | 23 - 2 files changed, 101 deletions(-) diff --git a/src/gallium/auxiliary/util/u_index_modify.c b/src/gallium/auxiliary/util

[Mesa-dev] [PATCH 7/7] vl: fix a buffer leak in the bicubic filter by using an uploader

From: Marek Olšák there's no error checking, because the previous code didn't do it either. --- src/gallium/auxiliary/vl/vl_bicubic_filter.c | 27 +++ 1 file changed, 11 insertions(+), 16 deletions(-) diff --git a/src/gallium/auxiliary/vl/vl_bicubic_filte

[Mesa-dev] [PATCH 6/7] gallium/hud: create files after graphs are created to get final names

From: Marek Olšák --- src/gallium/auxiliary/hud/hud_context.c | 25 +++-- src/gallium/auxiliary/hud/hud_cpu.c | 4 src/gallium/auxiliary/hud/hud_driver_query.c | 2 -- src/gallium/auxiliary/hud/hud_fps.c | 2 -- src/gallium/auxiliary/hud

[Mesa-dev] [PATCH 5/7] gallium/u_suballoc: allow setting pipe_resource::flags

From: Marek Olšák --- src/gallium/auxiliary/util/u_suballoc.c | 22 ++ src/gallium/auxiliary/util/u_suballoc.h | 2 +- src/gallium/drivers/r600/r600_pipe.c | 5 +++-- src/gallium/drivers/radeon/r600_pipe_common.c | 2 +- src/gallium/drivers/radeonsi

[Mesa-dev] [PATCH 02/18] radeonsi: add a workaround for clamping unaligned RGB 8 & 16-bit vertex loads

From: Marek Olšák --- src/gallium/drivers/radeonsi/si_shader.c | 38 +++--- src/gallium/drivers/radeonsi/si_shader.h | 4 +++ src/gallium/drivers/radeonsi/si_shader_internal.h | 3 ++ .../drivers/radeonsi/si_shader_tgsi_setup.c| 3 ++ src/gallium

[Mesa-dev] [PATCH 1/7] gallium/util: remove unused helper util_draw_texquad

From: Marek Olšák --- src/gallium/auxiliary/util/u_draw_quad.c | 66 src/gallium/auxiliary/util/u_draw_quad.h | 6 --- 2 files changed, 72 deletions(-) diff --git a/src/gallium/auxiliary/util/u_draw_quad.c b/src/gallium/auxiliary/util/u_draw_quad.c index

[Mesa-dev] [PATCH 06/18] radeonsi: have separate LS and ES main shader parts in the shader selector

From: Marek Olšák This might reduce the on-demand compilation if the initial VS/LS/ES determination is wrong. --- src/gallium/drivers/radeonsi/si_shader.c| 2 +- src/gallium/drivers/radeonsi/si_shader.h| 16 +++ src/gallium/drivers/radeonsi/si_state_shaders.c | 36

[Mesa-dev] [PATCH 08/18] radeonsi: use SI_MAX_ATTRIBS where it should be used

From: Marek Olšák for consistency; no change in behavior --- src/gallium/drivers/radeonsi/si_descriptors.c | 2 +- src/gallium/drivers/radeonsi/si_pipe.c| 2 +- src/gallium/drivers/radeonsi/si_shader.c | 4 ++-- src/gallium/drivers/radeonsi/si_shader.h | 2 +- 4 files changed

[Mesa-dev] [PATCH 05/18] radeonsi: don't compile pure monolithic shaders asynchronously

From: Marek Olšák there is no point, we have to wait anyway. --- src/gallium/drivers/radeonsi/si_state_shaders.c | 8 ++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_state_shaders.c b/src/gallium/drivers/radeonsi/si_state_shaders.c index

[Mesa-dev] [PATCH 09/18] radeonsi: fix UNSIGNED_BYTE index buffer fallback with non-zero start

From: Marek Olšák start can only be non-zero with MultiDrawElements, which is unlikely to occur with UNSIGNED_BYTE indices. --- src/gallium/drivers/radeonsi/si_state_draw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c b/src

[Mesa-dev] [PATCH 07/18] radeonsi: sort members of si_shader_key::part

From: Marek Olšák and improve some comments --- src/gallium/drivers/radeonsi/si_shader.h | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.h b/src/gallium/drivers/radeonsi/si_shader.h index d4b57c9..b7cf7ea 100644 --- a/src

[Mesa-dev] [PATCH 03/18] radeonsi: remove the fix_size3 workaround

From: Marek Olšák not needed with the shader fallback --- src/gallium/drivers/radeonsi/si_descriptors.c | 22 -- src/gallium/drivers/radeonsi/si_state.c | 9 - src/gallium/drivers/radeonsi/si_state.h | 5 - 3 files changed, 36 deletions(-) diff

[Mesa-dev] [PATCH 11/18] radeonsi: use a clever alignment for constant buffer uploads

From: Marek Olšák This results in a very tiny decrease in lgkm wait cycles. --- src/gallium/drivers/radeon/radeon_winsys.h| 1 + src/gallium/drivers/radeonsi/si_descriptors.c | 4 +++- src/gallium/drivers/radeonsi/si_pipe.h| 15 +++ src/gallium/winsys

[Mesa-dev] [PATCH 01/18] radeonsi: make fix_fetch an array of uint8_t

From: Marek Olšák so that we can add 3-component fallbacks. --- src/gallium/drivers/radeonsi/si_shader.c| 8 +-- src/gallium/drivers/radeonsi/si_shader.h| 5 ++--- src/gallium/drivers/radeonsi/si_state.c | 28 - src/gallium/drivers/radeonsi

[Mesa-dev] [PATCH 10/18] radeonsi: move index buffer flushing into a non-upload indexed case

From: Marek Olšák The other codepaths don't need this. --- src/gallium/drivers/radeonsi/si_state_draw.c | 13 ++--- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c b/src/gallium/drivers/radeonsi/si_state_draw.c index 8f

[Mesa-dev] [PATCH 13/18] radeonsi: use a clever alignment for index buffer uploads

From: Marek Olšák --- src/gallium/drivers/radeonsi/si_state_draw.c | 11 +++ 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c b/src/gallium/drivers/radeonsi/si_state_draw.c index ca28f50..8fed61a 100644 --- a/src/gallium/drivers

[Mesa-dev] [PATCH 04/18] radeonsi: allow unaligned vertex buffer offsets and strides on CIK-VI

From: Marek Olšák So that we can disable u_vbuf for GL core profiles. This is a v2 of the previous VI-only patch. It requires SH_MEM_CONFIG.ALIGNMENT_MODE = UNALIGNED on CIK-VI. --- src/gallium/drivers/radeonsi/si_pipe.c | 12 +--- 1 file changed, 9 insertions(+), 3 deletions(-) diff

[Mesa-dev] [PATCH 15/18] radeonsi: upload constants into VRAM instead of GTT

From: Marek Olšák This lowers lgkm wait cycles by 30% on VI and normal conditions. The might be a measurable improvement when CE is disabled (radeon) or under L2 thrashing. --- src/gallium/drivers/radeon/r600_pipe_common.c | 11 --- src/gallium/drivers/radeonsi/si_compute.c | 4

[Mesa-dev] [PATCH 14/18] gallium/radeon: use TCC line size as alignment in other places

From: Marek Olšák --- src/gallium/drivers/radeon/r600_buffer_common.c | 3 ++- src/gallium/drivers/radeon/r600_pipe_common.c | 3 ++- src/gallium/drivers/radeonsi/si_compute.c | 3 ++- src/gallium/drivers/radeonsi/si_descriptors.c | 5 +++-- 4 files changed, 9 insertions(+), 5

[Mesa-dev] [PATCH 16/18] gallium/radeon: change r600_aligned_buffer_create to take flags, not bind

From: Marek Olšák All call sites set bind = 0. The next commit will use this. --- src/gallium/drivers/radeon/r600_buffer_common.c | 6 +++--- src/gallium/drivers/radeon/r600_pipe_common.h | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/src/gallium/drivers/radeon

<    3   4   5   6   7   8   9   10   11   12   >