Marek Olšák [Fri, 7 May 2021 00:57:52 +0000 (20:57 -0400)]
radeonsi/gfx11: enable NGG-only draw paths
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Thu, 6 May 2021 21:40:25 +0000 (17:40 -0400)]
radeonsi/gfx11: expect packed threadID VGPRs
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Thu, 6 May 2021 21:38:21 +0000 (17:38 -0400)]
radeonsi/gfx11: always allow DCC stores
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Wed, 11 Nov 2020 19:37:00 +0000 (14:37 -0500)]
radeonsi/gfx11: increase the hw screen offset alignment
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Fri, 7 May 2021 00:41:47 +0000 (20:41 -0400)]
ac/surface: add gfx11 support to modifiers tests
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Fri, 7 May 2021 09:54:26 +0000 (05:54 -0400)]
ac/surface: define gfx11 modifiers
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Fri, 7 May 2021 09:59:45 +0000 (05:59 -0400)]
ac/surface: gfx11 changes
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Wed, 8 Dec 2021 16:42:49 +0000 (11:42 -0500)]
ac/gpu_info: set cu_mask correctly for gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Wed, 27 Apr 2022 03:23:33 +0000 (23:23 -0400)]
ac/llvm: don't set GLC for stores on gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Tue, 22 Mar 2022 14:19:41 +0000 (10:19 -0400)]
ac/llvm: update pknorm and waitcnt for gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Fri, 11 Feb 2022 10:45:26 +0000 (05:45 -0500)]
ac/llvm: don't set DLC on gfx11 because it means something else there
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Fri, 17 Dec 2021 19:08:45 +0000 (14:08 -0500)]
ac/llvm: export mrt0 instead of null on gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Sun, 27 Feb 2022 02:32:42 +0000 (21:32 -0500)]
ac: implement register shadowing for gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Thu, 27 Jan 2022 04:36:30 +0000 (23:36 -0500)]
ac: scratch buffer register changes for gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Fri, 17 Dec 2021 16:28:54 +0000 (11:28 -0500)]
ac: implement ac_get_tbuffer_format for gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Pierre-Eric Pelloux-Prayer [Tue, 27 Jul 2021 12:36:30 +0000 (14:36 +0200)]
amd: update gfx10_format_table.py for gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Wed, 4 May 2022 00:32:27 +0000 (20:32 -0400)]
ac: don't align VGPRs to 8 or 16 for gfx11
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Fri, 7 May 2021 11:02:18 +0000 (07:02 -0400)]
amd: add Mesa-only addrlib changes
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Fri, 7 May 2021 10:12:19 +0000 (06:12 -0400)]
amd: import gfx11 addrlib
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Wed, 22 Sep 2021 08:58:53 +0000 (04:58 -0400)]
amd: add gfx11 to packet definitions
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Sun, 28 Mar 2021 06:12:47 +0000 (02:12 -0400)]
amd: enable gfx11 in header generator, fix drivers with renamed gfx6-10 defs
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Sun, 28 Mar 2021 08:59:54 +0000 (04:59 -0400)]
amd/registers: add gfx11-rsrc.json
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Sun, 28 Mar 2021 06:11:42 +0000 (02:11 -0400)]
amd/registers: add gfx11.json
Other files are also updated to due regeneration.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Sun, 28 Mar 2021 05:57:44 +0000 (01:57 -0400)]
amd/registers: add gfx11 to the json generator
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Marek Olšák [Tue, 3 May 2022 18:42:57 +0000 (14:42 -0400)]
amd/registers: hardcode GC base offsets in the json generator
gfx11 doesn't have the ip_offset file, so we have to do it this way.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16328>
Dave Airlie [Mon, 9 May 2022 04:28:32 +0000 (14:28 +1000)]
radv: precalculate tess ring sizes/offsets.
These are all static per device, so just calculate at device init
time instead of preamble
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16392>
Dave Airlie [Mon, 9 May 2022 04:19:35 +0000 (14:19 +1000)]
radv: precalculate hs offchip parameters.
These are per device static, just just precalc them instead of
doing it on every command buffer submission.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16392>
Karol Herbst [Thu, 28 Apr 2022 20:08:00 +0000 (22:08 +0200)]
nir: add a nir_remove_non_entrypoints helper
This code just got duplicated a lot. There is still more, but the
remaining instances do a bit more than just removing other functions.
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16348>
Jason Ekstrand [Tue, 12 Apr 2022 17:41:14 +0000 (12:41 -0500)]
nir: Fix constant folding for non-32-bit ifind_msb and clz
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16348>
Karol Herbst [Tue, 12 Apr 2022 00:33:27 +0000 (02:33 +0200)]
iris/cs: take buffer offsets into account for CL
Sadly we pass in an offset, which the driver can't ignore
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16348>
Karol Herbst [Sun, 3 Apr 2022 10:47:58 +0000 (12:47 +0200)]
llvmpipe/fence: make the fence id counter atomic
Multiple threads can race on the function static variable
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16348>
Karol Herbst [Fri, 25 Feb 2022 05:11:22 +0000 (06:11 +0100)]
llvmpipe: PIPE_COMPUTE_CAP_GRID_DIMENSION is uint64_t
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16348>
Mike Blumenkrantz [Thu, 5 May 2022 17:34:47 +0000 (13:34 -0400)]
zink: split renderpasses for TextureBarrierNV() usage
if no fbfetch is present, then this is a draw -> sampler read barrier,
and zink does not emit a self-dependency for this by default
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16354>
Charmaine Lee [Mon, 9 May 2022 22:07:13 +0000 (15:07 -0700)]
svga/nir: enable PIPE_CAP_TGSI_TEXCOORD
Use texcoord semantic instead of generic with nir.
Fixes assert in nir_gather_info with offsetted varying slots.
Reviewed-by: Neha Bhende <bhenden@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16414>
Michel Zou [Thu, 5 May 2022 17:51:10 +0000 (19:51 +0200)]
vulkan/wsi: fix missing unistd include
fixes:
c72ff19a
Closes #6428
Reviewed-by: Simon Ser <contact@emersion.fr>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16353>
Emma Anholt [Thu, 5 May 2022 19:21:19 +0000 (12:21 -0700)]
ci/iris: Cut the glk-deqp test coverage in half.
It's taking 13-14 minutes of deqp-runner time, not counting booting, or
the LAVA-side job getting being queued behind other jobs. Well past our
10-minute runtime target, and we saw load on these boards causing the
queue to get quite long
(https://gitlab.freedesktop.org/mesa/mesa/-/issues/6409#note_1368750)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16359>
Alyssa Rosenzweig [Mon, 9 May 2022 20:09:59 +0000 (16:09 -0400)]
agx: Restore Valve copyright header
Parallel copy code is from ir3_lower_parallel_copy.c. This was attributed in the
commit message but lost in the copyright header due to a copypaste mistake.
Rectify this.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16413>
Rhys Perry [Fri, 6 May 2022 10:56:30 +0000 (11:56 +0100)]
aco: fix cmpswap global atomic definition on GFX6
Missed this one.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Fixes:
2f0bb39e162 ("aco: ensure that definitions fixed to operands have matching regclasses")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16367>
Emma Anholt [Tue, 26 Apr 2022 04:11:38 +0000 (21:11 -0700)]
Revert "ci: remove nouveau from shader-db runs"
This reverts commit
0464117ad9bd47f079175058771220e8dad4f00b. Now that
the shim back-channel communicates with nouveau that the "GPU" is always
idle, we can get the nouveau compiler back into the CI path.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16150>
Emma Anholt [Wed, 8 Dec 2021 23:21:42 +0000 (15:21 -0800)]
nouveau: disable fences when running under drm-shim.
Otherwise, you get a hang at the end of shader-db.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16150>
Pavel Ondračka [Sun, 8 May 2022 16:04:37 +0000 (18:04 +0200)]
r300: skip draws instead of using a dummy vertex shader
When we fail to compile some vertex shader, we currently use a very
simple dummy one, setting gl_Position to (0,0,0,1), effectively
rendering nothing. Unfortunately, the dummy vertex shader leads to
hangs with RV370 in some rare circumstances. Instead of trying to
fix the shader, just skip the draws altogether when the compilation
fails.
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5870
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16387>
Emma Anholt [Wed, 20 Apr 2022 23:37:46 +0000 (16:37 -0700)]
glsl: Stop lowering ir_quadop_vector.
Now that everybody goes through NIR, glsl_to_nir is happy to handle the
instruction and turn it into nir_op_vec4 instead of going to a temp
variable and back.
No changes on freedreno shader-db.
Reviewed-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16363>
Charmaine Lee [Fri, 6 May 2022 00:11:12 +0000 (17:11 -0700)]
svga: fix shader IR type passed to draw create shader function
In the SVGA create shader functions, the original shader IR could have been
deleted after pipe_shader_state_to_tgsi_tokens() if it is to be converted
from NIR to TGSI. So to avoid accessing the deleted NIR IR,
set the shader state type to TGSI before passing the shader
state to the draw function.
Reviewed-by: Neha Bhende <bhenden@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16365>
Alyssa Rosenzweig [Mon, 9 May 2022 14:40:46 +0000 (10:40 -0400)]
pan/va: Add whitespace after disassembled branches
To make the disassembly easier to read, add whitespace after disassembled
branches. This makes the basic blocks of the original control flow graph more
obvious, to aid comparison with the IR.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16409>
Alyssa Rosenzweig [Fri, 6 May 2022 16:57:14 +0000 (12:57 -0400)]
pan/va: Add some whitespace to Valhall disassembly
Makes it easier to read.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16409>
Sathishkumar S [Wed, 27 Apr 2022 13:12:57 +0000 (18:42 +0530)]
radeon/vcn: engage all available jpeg engines
use multiple contexts and submit in a round robin scheme to make
use of all the available jpeg engines simultaneously. During mjpeg
decode context need not be same across frames as they are discrete
jpeg images.
V2: number of ctx to be equal to number of engines and fix indent (Leo)
V3: decide ctx count in create_decoder, don't add a video param (Boyuan)
Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16355>
Mihai Preda [Mon, 9 May 2022 12:04:52 +0000 (15:04 +0300)]
Revert "gallivm: use LLVM opaque pointers in lp_bld_tgsi_soa.c"
This reverts commit
32a55651cf4ecb830801acafe6410df3f66afca9.
Fixes: #6439 #6453
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16400>
Rhys Perry [Wed, 4 May 2022 12:04:04 +0000 (13:04 +0100)]
ac/nir: skip s_barrier if TCS patches are within subgroup
fossil-db (Sienna Cichlid):
Totals from 538 (0.33% of 162293) affected shaders:
Instrs: 125288 -> 123682 (-1.28%)
CodeSize: 712384 -> 705960 (-0.90%)
Latency: 632139 -> 623596 (-1.35%)
InvThroughput: 218491 -> 215600 (-1.32%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16356>
Rhys Perry [Wed, 4 May 2022 12:03:52 +0000 (13:03 +0100)]
aco: skip s_barrier if TCS patches are within subgroup
fossil-db (Sienna Cichlid):
Totals from 518 (0.32% of 162293) affected shaders:
Instrs: 124943 -> 123908 (-0.83%)
CodeSize: 708764 -> 704624 (-0.58%)
Latency: 618380 -> 618279 (-0.02%)
InvThroughput: 214061 -> 214051 (-0.00%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16356>
Patrick Lerda [Mon, 9 May 2022 14:22:43 +0000 (16:22 +0200)]
panfrost: Fix unwanted valgrind message related to restart_index
As a reminder primitive_restart should always be checked before any access to restart_index.
It seems that restart_index is only initialized when primitive_restart is set to a non-zero
value. This patch is equivalent to the previous code but written in a way that the compiler
will test primitive_restart first before trying to read restart_index.
With commit
ad864a7c150a15221fb9c85d3214d4bcb6db7518:
Conditional jump or move depends on uninitialised value(s)
at 0xD33F1EC: panfrost_is_implicit_prim_restart (pan_cmdstream.c:2907)
by 0xD33F1EC: panfrost_emit_primitive (pan_cmdstream.c:3073)
by 0xD33F1EC: panfrost_draw_emit_tiler (pan_cmdstream.c:3440)
by 0xD33F1EC: panfrost_direct_draw (pan_cmdstream.c:3595)
by 0xD340467: panfrost_draw_vbo (pan_cmdstream.c:3889)
by 0xD219119: u_vbuf_draw_vbo (u_vbuf.c:1498)
by 0xD1C81F9: cso_multi_draw (cso_context.c:1644)
by 0xCFBA19B: _mesa_draw_arrays.part.11 (draw.c:1324)
by 0xCFBADA1: _mesa_draw_arrays (draw.c:1295)
by 0xCFBADA1: _mesa_DrawArrays (draw.c:1533)
by 0xD32EB: gl_vao_draw_data (in /usr/local/bin/mpv)
Uninitialised value was created by a stack allocation
at 0xCFBA14E: _mesa_draw_arrays.part.11 (draw.c:1289)
With mesa-22.1.0-rc4:
Conditional jump or move depends on uninitialised value(s)
at 0xD36369C: panfrost_is_implicit_prim_restart (pan_cmdstream.c:2895)
by 0xD36369C: panfrost_draw_emit_tiler (pan_cmdstream.c:3023)
by 0xD36369C: panfrost_direct_draw (pan_cmdstream.c:3215)
by 0xD3649BF: panfrost_draw_vbo (pan_cmdstream.c:3494)
by 0xD23DE7D: u_vbuf_draw_vbo (u_vbuf.c:1498)
by 0xD1ECBD1: cso_multi_draw (cso_context.c:1644)
by 0xCFD60FF: _mesa_draw_arrays.part.11 (draw.c:1324)
by 0xCFD6D11: _mesa_draw_arrays (draw.c:1295)
by 0xCFD6D11: _mesa_DrawArrays (draw.c:1533)
by 0xD32EB: gl_vao_draw_data (in /usr/local/bin/mpv)
Uninitialised value was created by a stack allocation
at 0xCFD60B2: _mesa_draw_arrays.part.11 (draw.c:1289)
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Reviewed-by: Alyssa Rosenzweig alyssa@collabora.com
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16389>
Alyssa Rosenzweig [Mon, 25 Apr 2022 20:54:11 +0000 (16:54 -0400)]
panvk: Call nir_opt_trivial_continues
Fixes
dEQP-VK.glsl.indexing.tmp_array.vec2_static_loop_write_static_loop_read_vertex
which otherwise fails due to nir_opt_sink being "clever" around unused
loop exit blocks.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16155>
Alyssa Rosenzweig [Mon, 25 Apr 2022 15:32:39 +0000 (11:32 -0400)]
panvk: Conform viewport code to Vulkan spec
The depth equations weren't quite right, with spec citations to prove it. This
didn't fix the test I was debugging, but it surely fixed /something/.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16155>
Alyssa Rosenzweig [Mon, 25 Apr 2022 13:55:59 +0000 (09:55 -0400)]
panvk: Stub pipeline cache using the common code
Moves the needle from Crash to Fail on:
dEQP-VK.synchronization.op.single_queue.fence.write_clear_color_image_read_image_compute.image_64x64x8_r32_sfloat
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16155>
Alyssa Rosenzweig [Sun, 8 May 2022 01:09:11 +0000 (21:09 -0400)]
mesa/st: Fix building tests on macOS
Due to an upstream bug, macOS can't link empty static libraries.
There is open Meson bug about this use case [1], though arguably the issue
is macOS's implementation of ar. Of course, the functionality is mostly
useless.
The removal of GLSL-to-TGSI trivialized a static library, causing
linking to fail. This commit garbage collects the useless library.
This fixes the build on macOS:
FAILED: src/mesa/state_tracker/tests/libmesa_st_test_common.a
rm -f src/mesa/state_tracker/tests/libmesa_st_test_common.a && ar csr src/mesa/state_tracker/tests/libmesa_st_test_common.a
ar: no archive members specified
usage: ar -d [-TLsv] archive file ...
ar -m [-TLsv] archive file ...
ar -m [-abiTLsv] position archive file ...
ar -p [-TLsv] archive [file ...]
ar -q [-cTLsv] archive file ...
ar -r [-cuTLsv] archive file ...
ar -r [-abciuTLsv] position archive file ...
ar -t [-TLsv] archive [file ...]
ar -x [-ouTLsv] archive [file ...]
[1] https://github.com/mesonbuild/meson/issues/3735
Fixes:
214c774ba6c ("mesa/st: Remove st_glsl_to_tgsi.")
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Emma Anholt <emma@anholt.net>
Tested-by: Vinson Lee <vlee@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16385>
Iago Toral Quiroga [Mon, 9 May 2022 07:14:28 +0000 (09:14 +0200)]
v3dv: don't leak variant QPU when pipeline compile fails
Typically we free them when we upload the QPU code from the variant
to the assembly BO in the pipeline, however, if there is an error
during pipeline compilation that may not happen and we would leak
the QPU code from the variants.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16370>
Iago Toral Quiroga [Fri, 6 May 2022 10:01:09 +0000 (12:01 +0200)]
v3dv: expose VK_KHR_pipeline_executable_properties
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16370>
Iago Toral Quiroga [Fri, 6 May 2022 08:59:45 +0000 (10:59 +0200)]
v3dv: implement vkGetPipelineExecutableStatisticsKHR
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16370>
Iago Toral Quiroga [Fri, 6 May 2022 09:24:07 +0000 (11:24 +0200)]
v3d/compiler: add more stats to prog_data
So we can expose them via VK_KHR_pipeline_executable_properties.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16370>
Iago Toral Quiroga [Fri, 6 May 2022 08:56:49 +0000 (10:56 +0200)]
v3dv: implement vkGetPipelineExecutablePropertiesKHR
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16370>
Iago Toral Quiroga [Thu, 5 May 2022 10:51:05 +0000 (12:51 +0200)]
v3dv: implement vkGetPipelineExecutableInternalRepresentationsKHR
We can output the final NIR form (which we store in the pipeline
stage) and the final QPU (which we can retrive from the assembly BO).
We should be careful not to fetch the shaders from the cache when
VK_PIPELINE_CREATE_CAPTURE_INTERNAL_REPRESENTATIONS_BIT_KHR is present,
since we don't store NIR shader in the pipeline shader data that is
cached, so a cache hit would leave us without the NIR shader. The spec
already contemplates this scenario:
"Enabling this flag must not affect the final compiled pipeline but
may disable pipeline caching or otherwise affect pipeline creation
time."
We also prevent disposing of the pipeline stages the variants when this
flag is requested to ensure this information is available later when
calling vkGetPipelineExecutableInternalRepresentationsKHR.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16370>
Iago Toral Quiroga [Mon, 9 May 2022 10:57:13 +0000 (12:57 +0200)]
v3dv: expose separateDepthStencilLayouts
This is actually required by Vulkan 1.2 and to expose the extension,
so let's conform to this requirement, we don't really care since
image layouts are not relevant to our current implementation.
Fixes:
1442d77bc5 ('v3dv: trivially implement VK_KHR_separate_depth_stencil_layouts')
Fixes: dEQP-VK.info.device_mandatory_features
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16398>
Boris Brezillon [Wed, 27 Apr 2022 15:57:25 +0000 (17:57 +0200)]
vulkan/wsi: Make wsi_win32_surface_get_capabilities() return the current extent
We can get that information with GetClientRect(), and some applications
complain when the window size doesn't match the currentExtent values.
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16394>
Lionel Landwerlin [Wed, 20 Apr 2022 07:51:48 +0000 (10:51 +0300)]
nir/lower_shader_calls: put inserted instructions into a dummy block
When moving code into the main block or loop blocks, put the code into
its own :
if(true) { ... }
block so that we avoid break/continue/return issues.
v2: Also take care of the main block with return instructions
v3: Make deletion more obvious with dummy if blocks (Jason)
v4: Fixup assert for loops (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes:
8dfb240b1f06 ("nir: Add raytracing shader call lowering pass.")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16036>
Lionel Landwerlin [Wed, 20 Apr 2022 07:54:36 +0000 (10:54 +0300)]
nir/lower_shader_calls: don't insert code after break/continue
When moving code from below to the insertion cursor point, if the
cursor points to a jump instruction, don't bother inserting the code.
It would break the break/continue assumptions of NIR and would not be
executed anyway.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes:
8dfb240b1f06 ("nir: Add raytracing shader call lowering pass.")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16036>
Lionel Landwerlin [Tue, 19 Apr 2022 12:11:17 +0000 (15:11 +0300)]
nir/lower_shader_calls: don't use nop instructions as cursors
Stop using nop instructions which are causing issues with
break/continue, instead use a nir_cursor (which brings its share of
pain).
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes:
8dfb240b1f06 ("nir: Add raytracing shader call lowering pass.")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16036>
Jason Ekstrand [Tue, 26 Apr 2022 13:53:31 +0000 (08:53 -0500)]
nir/cf: Return a cursor from nir_cf_extract as well
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16036>
Lionel Landwerlin [Tue, 19 Apr 2022 12:05:04 +0000 (15:05 +0300)]
nir/cf: return cursor after insertion of cf_list
This will be useful to cut code from one location and paste it at
another place and later keep pasting after the previous insertions.
v2: update comment (Jason)
deal with stiching 2 empty blocks (Jason)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16036>
Manas Chaudhary [Thu, 21 Apr 2022 04:12:02 +0000 (09:42 +0530)]
panvk: Allow PAN_BO_INVISIBLE to be mapped with dump flag
Signed-off-by: Manas Chaudhary <manas.chaudhary@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16072>
Manas Chaudhary [Thu, 21 Apr 2022 03:51:34 +0000 (09:21 +0530)]
panvk: Add PANVK_DEBUG_DUMP dump mappings
Signed-off-by: Manas Chaudhary <manas.chaudhary@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16072>
Danylo Piliaiev [Tue, 3 May 2022 09:21:07 +0000 (12:21 +0300)]
tu: Implement VK_EXT_image_view_min_lod
Relevant tests:
dEQP-VK.texture.mipmap.*.image_view_min_lod.*
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16292>
Dave Airlie [Tue, 3 May 2022 01:42:24 +0000 (11:42 +1000)]
gallivm: move to new pass manager to handle coroutines change.
LLVM 15 ripped out the legacy coroutine passes. This means moving
to the new pass manager is the best option to move forward and is
long overdue.
I've tried to recreate the same set of passes in the new pass mgr
as the old, but I expect some tweaking may be needed to confirm this.
Acked-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16289>
Juan A. Suarez Romero [Fri, 6 May 2022 11:56:17 +0000 (13:56 +0200)]
v3d: allow TFU blitting for single layer textures
TFU just handles 2D textures.
But for 2D array textures, cubemaps and 3D textures, if only one layer needs
to be blitted, then it is like a simple 2D texture, and thus we can also
use the TFU.
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16372>
David Heidelberg [Wed, 4 May 2022 15:23:43 +0000 (17:23 +0200)]
ci: traces: virgl: reenable previously crashing Godot trace
Visually verified, screenshot does look correctly.
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16335>
Emma Anholt [Thu, 5 May 2022 00:16:47 +0000 (17:16 -0700)]
svga: Set lower_bitops for vpu9 screen.
Since these don't have native integers, nir_lower_io generating ishls
instead of amuls for uniform addressing math runs afoul of
nir_lower_int_to_float.
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16339>
Thomas Debesse [Mon, 25 Apr 2022 07:01:01 +0000 (09:01 +0200)]
gallium/clover: pass -no-opaque-pointers to Clang
Clang opaque pointers have been enabled by default
but the Clover code currently expects them to be disabled.
Signed-off-by: Thomas Debesse <dev@illwieckz.net>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6342
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16129>
Bas Nieuwenhuizen [Wed, 23 Feb 2022 01:27:46 +0000 (02:27 +0100)]
vulkan/wsi/x11: Ensure we have the required number of images for acquire.
For games that needs >1 at the same time the existing check wasn't enough.
Cc: mesa-stable
Reviewed-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15132>
Georg Lehmann [Sat, 7 May 2022 23:30:42 +0000 (01:30 +0200)]
wsi/x11: Don't leak xcb_get_geometry_reply_t.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6443
Fixes:
44a20baeb8b ("wsi/x11: Avoid using xcb_wait_for_special_event in FIFO modes")
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16384>
Georg Lehmann [Wed, 4 May 2022 10:42:00 +0000 (12:42 +0200)]
radeonsi: Use nir_fold_16bit_image_load_store_conversions.
Remove conversions using d16 image_load/image_store.
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16319>
Icecream95 [Thu, 14 Apr 2022 01:30:39 +0000 (13:30 +1200)]
nir/lower_tex: Copy more fields in lower_tex_to_txd and friends
Fixes NIR validation errors for OpenMW on Panfrost.
Fixes:
1f97819fbe5 ("panfrost: Emulate GL_CLAMP on Bifrost")
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15939>
Yiwei Zhang [Fri, 6 May 2022 19:30:01 +0000 (19:30 +0000)]
venus: extend buffer cache to cover layering usage
Layering clients, e.g. angle and zink, use wide sets of buffer usage
flags because they don't know what a resource will be used for in the
majority cases, which is on the other hand making it easier for layering
to optimize resource management.
This change adds a super-set usage to the buffer cache entries, that
will mostly ensure no cache-miss for non-sparsed buffer usages. Since
that involves usage bits from extensions, we'll mask out those disabled
ones upon querying but will use the static cache create info for
checking cache hit for code simplicity.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16379>
Thomas Debesse [Mon, 18 Apr 2022 02:21:02 +0000 (04:21 +0200)]
gallium/clover: LLVM setLangDefaults moved from clangFrontend to clangBasic
Signed-off-by: Thomas Debesse <dev@illwieckz.net>
Reviewed-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16002>
Chia-I Wu [Thu, 17 Mar 2022 18:03:18 +0000 (11:03 -0700)]
anv: advertise rectangularLines only for Gen10+
We use the non-strict algorithm (with parallelograms) prior to Gen10 for
wide lines. We can not advertise rectangularLines.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Fixes:
f6e7de41d7b ("anv: Implement VK_EXT_line_rasterization")
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15432>
Mike Blumenkrantz [Thu, 5 May 2022 13:55:28 +0000 (09:55 -0400)]
llvmpipe: never infer early zs tests when fbfetch is active
this breaks zs fbfetch
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16346>
Mike Blumenkrantz [Thu, 5 May 2022 13:54:18 +0000 (09:54 -0400)]
nir/gather_info: flag fbfetch on subpass image loads
might not be able to determine which output is being read, but these
are definitely fbfetch uses (from lavapipe)
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16346>
Rhys Perry [Fri, 25 Mar 2022 18:00:56 +0000 (18:00 +0000)]
aco: split load_sbt_amd result
fossil-db (Sienna Cichlid):
Totals from 11 (0.01% of 162293) affected shaders:
Instrs: 47857 -> 47738 (-0.25%)
CodeSize: 261556 -> 261080 (-0.18%)
Latency: 1176822 -> 1176245 (-0.05%)
InvThroughput: 784549 -> 784165 (-0.05%)
Copies: 5959 -> 5840 (-2.00%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16203>
Rhys Perry [Fri, 15 Apr 2022 12:30:41 +0000 (13:30 +0100)]
radv: use SMEM for a few load_global
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16203>
Rhys Perry [Fri, 15 Apr 2022 12:15:10 +0000 (13:15 +0100)]
radv: remove subtractions in address calculations
Additions by positive integers can more easily be combined into the
access.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16203>
Rhys Perry [Fri, 15 Apr 2022 11:57:36 +0000 (12:57 +0100)]
radv: allow LBVH on GFX6+
Use integer atomics on GFX8/9.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16203>
Rhys Perry [Thu, 5 May 2022 11:35:09 +0000 (12:35 +0100)]
radv: add radv_has_shader_buffer_float_minmax
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16203>
Daniel Schürmann [Wed, 16 Mar 2022 14:14:29 +0000 (15:14 +0100)]
aco/optimizer: re-combine and copy-propagate p_create_vector(p_split_vector)
Totals from 309 (0.23% of 134913) affected shaders: (GFX10.3)
CodeSize: 1853812 -> 1857732 (+0.21%); split: -0.05%, +0.27%
Instrs: 340810 -> 341789 (+0.29%); split: -0.07%, +0.36%
Latency: 3301814 -> 3301774 (-0.00%); split: -0.02%, +0.02%
InvThroughput: 590473 -> 590914 (+0.07%); split: -0.00%, +0.08%
Copies: 28751 -> 29731 (+3.41%); split: -0.87%, +4.28%
PreSGPRs: 14010 -> 14028 (+0.13%); split: -0.01%, +0.14%
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15414>
Daniel Schürmann [Wed, 16 Mar 2022 16:25:35 +0000 (17:25 +0100)]
aco/ra: improve split_vector register assignment if the operand is not killed
This allows for more coalescing when lowering the copies.
Totals from 44801 (33.21% of 134913) affected shaders: (GFX10.3)
VGPRs: 1513264 -> 1513248 (-0.00%)
CodeSize:
113354240 ->
113172872 (-0.16%); split: -0.16%, +0.00%
Instrs:
21648793 ->
21603397 (-0.21%); split: -0.21%, +0.00%
Latency:
95762290 ->
95757403 (-0.01%); split: -0.01%, +0.00%
InvThroughput:
15427354 ->
15427341 (-0.00%); split: -0.00%, +0.00%
Copies: 2065330 -> 2019933 (-2.20%); split: -2.20%, +0.00%
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15414>
Daniel Schürmann [Wed, 16 Mar 2022 14:49:05 +0000 (15:49 +0100)]
aco: don't re-create vectors for load_barycentric_* intrinsics
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15414>
Mike Blumenkrantz [Thu, 5 May 2022 13:08:33 +0000 (09:08 -0400)]
zink: restore conditional ordering for query begin/end
this is the case which (also) handles query[0] for primgen queries,
so it can't be an else case
Fixes:
a9451f25995 ("zink: use VK_EXT_primitives_generated_query when available")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16345>
Mike Blumenkrantz [Thu, 10 Jun 2021 11:13:05 +0000 (07:13 -0400)]
radv: emit fewer framebuffer registers
only the changed cbuf index registers need to be updated, not all of them
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11299>
Mike Blumenkrantz [Tue, 26 Apr 2022 12:32:54 +0000 (08:32 -0400)]
util/blitter: fix sampler restore with 0 saved samplers
in a sequence where a driver saves 0 sampler/views before calling
u_blitter, the previous state of having 0 sampler/views bound would
not be restored as expected, resulting in stale sampler/views which
could affect behavior before new sampler/views were bound
cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16178>
Lionel Landwerlin [Wed, 4 May 2022 23:37:27 +0000 (02:37 +0300)]
intel: fix stall debug option
Missing the parsing bit.
Fixes:
317512e0387a ("anv/intel: add a new debug flag for stalling after every draw/dispatch")
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16338>
Marek Olšák [Sat, 2 Apr 2022 01:43:42 +0000 (21:43 -0400)]
frontend/dri: allow swapped BGR->RGB channel order for MSAA color buffers
This only applies to MSAA visuals. The MSAA channel order doesn't have to
match the visual, but some drivers don't support MSAA with BGR.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16327>
Marek Olšák [Wed, 27 Apr 2022 11:35:48 +0000 (07:35 -0400)]
mesa: consider the sample count when choosing a texture format
The set of supported MSAA formats can be smaller than the set of supported
non-MSAA formats.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16326>
Emma Anholt [Tue, 12 Apr 2022 17:46:42 +0000 (10:46 -0700)]
glsl: Remove unused lower_variable_index_to_cond_assign.
It's been replaced by nir_lower_indirect_derefs().
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8044>
Emma Anholt [Tue, 12 Apr 2022 05:15:46 +0000 (22:15 -0700)]
glsl: Remove EmitNoLoops and the associated lower_jumps(lower_break=true) code.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8044>