Rob Clark [Mon, 22 Jun 2020 23:18:13 +0000 (16:18 -0700)]
freedreno/ir3/ra: fix pre-color edge case
Fixes a case where you have something like:
aVecOutput.z = aScalarInput;
In particular, skipping over things that are not the first component is
wrong.. in the above case the input we need to precolor is the 3rd
component. But we need to adjust the target register according to the
offset.
Fixes android.hardware.nativehardware.cts.AHardwareBufferNativeTests
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5601>
Jonathan Marek [Mon, 22 Jun 2020 02:08:45 +0000 (22:08 -0400)]
turnip: disable early_z for VK_FORMAT_S8_UINT
This format doesn't have depth, and apparently having earlyz enabled can
cause issues. Fixes at least these tests:
dEQP-VK.renderpass.suballocation.multisample.s8_uint.samples_*
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5586>
Jonathan Marek [Sun, 21 Jun 2020 03:34:53 +0000 (23:34 -0400)]
turnip: fix update_stencil_mask
The previous value was not being cleared, resulting in some dynamic stencil
state failures. Fixes these two tests:
dEQP-VK.dynamic_state.ds_state.stencil_params_advanced
dEQP-VK.dynamic_state.ds_state.stencil_params_basic_1
Fixes: 233610f8cf8d8810 ("turnip: refactor draw states and dynamic states")
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5586>
Jonathan Marek [Sun, 21 Jun 2020 03:20:32 +0000 (23:20 -0400)]
turnip: fix empty scissor case
Fixes these two tests:
dEQP-VK.draw.scissor.empty_dynamic_scissor_first_draw
dEQP-VK.draw.scissor.empty_static_scissor
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5586>
Rob Clark [Wed, 24 Jun 2020 18:28:32 +0000 (11:28 -0700)]
freedreno: handle batch flush in resource tracking
In rare cases, we can get into situations where the tracking of read/
written resources triggers a flush of the current batch.
To handle that, (1) take a reference to the current batch, so it doesn't
disappear under us, and (2) check after resource tracking whether the
current batch was flushed. If it is, we have to re-do the resource
tracking, but since we have a fresh batch, it should not get flushed the
second time around.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3160
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5634>
Rob Clark [Wed, 24 Jun 2020 16:31:03 +0000 (09:31 -0700)]
freedreno: split out batch clear tracking helper
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5634>
Rob Clark [Wed, 24 Jun 2020 16:23:50 +0000 (09:23 -0700)]
freedreno: split out batch draw tracking helper
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5634>
Rob Clark [Wed, 24 Jun 2020 16:06:50 +0000 (09:06 -0700)]
freedreno: make foreach_bit() declare it's cursor
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5634>
Jonathan Marek [Wed, 24 Jun 2020 23:56:01 +0000 (19:56 -0400)]
turnip: implement VK_EXT_vertex_attribute_divisor
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5640>
Eric Engestrom [Wed, 24 Jun 2020 23:48:24 +0000 (01:48 +0200)]
docs: fix 20.1.2 relnotes
I manually converted them from html and didn't double-check the
result...
Fixes: e94f81e9df8f038813a4 ("docs: Add release notes for 20.1.2")
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5639>
Eric Engestrom [Wed, 24 Jun 2020 22:58:56 +0000 (00:58 +0200)]
docs: update calendar and link releases notes for 20.1.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5638>
Eric Engestrom [Wed, 24 Jun 2020 21:44:30 +0000 (23:44 +0200)]
docs: Add release notes for 20.1.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5638>
Rob Clark [Sun, 21 Jun 2020 20:26:57 +0000 (13:26 -0700)]
freedreno/ir3: switch PIPE_CAP_TGSI_TEXCOORD
We don't really need the varying remapping, and it seems to somehow
happen twice when shader-cache comes into the picture. But we can
just choose not to have this problem.
Now that everything is using the ir3_point_sprite() helper, we can
flip this pipe cap without it being a massive flag-day.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5595>
Rob Clark [Mon, 22 Jun 2020 18:54:43 +0000 (11:54 -0700)]
freedreno: convert builtin blit VS prog to ureg builder
The correct varying semantic to use depends on PIPE_CAP_TGSI_TEXCOORD.
To handle this transition switch it over to ureg builder, and query the
pipe-cap to choose the appropriate semantic.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5595>
Rob Clark [Mon, 22 Jun 2020 15:47:33 +0000 (08:47 -0700)]
freedreno/a3xx: use point-coord helper
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5595>
Rob Clark [Mon, 22 Jun 2020 15:46:35 +0000 (08:46 -0700)]
freedreno/a4xx: use point-coord helper
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5595>
Rob Clark [Mon, 22 Jun 2020 15:45:13 +0000 (08:45 -0700)]
freedreno/a5xx: use point-coord helper
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5595>
Rob Clark [Mon, 22 Jun 2020 15:43:17 +0000 (08:43 -0700)]
freedreno/a6xx: use point-coord helper
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5595>
Rob Clark [Mon, 22 Jun 2020 15:39:12 +0000 (08:39 -0700)]
freedreno/a6xx: de-duplicate vinterp/vpsrepl state building
When we flip the texcoord patch, we'll setup PNTC input slot in the
pre-built interp stateobj, rather than this being a draw-time (slow-
path) built stateobj. But rather than duplicate more of the slow-
path logic, refactor it out into a helper that is reused in both
cases.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5595>
Rob Clark [Mon, 22 Jun 2020 15:24:55 +0000 (08:24 -0700)]
freedreno/ir3: add helper to determine point-coord inputs
This will simplify a bit the logic for setting up vinterp/vprepl in the
driver backend, and also avoid it being a flag-day when we switch the
texcoord pipe cap.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5595>
Jonathan Marek [Sat, 13 Jun 2020 02:12:51 +0000 (22:12 -0400)]
turnip: move some logic out of create_render_pass_common
CreateRenderPass2 is the common path now, it doesn't make sense to have a
create_render_pass_common. Rename it to tu_render_pass_gmem_config and
move logic not related to gmem config out of it.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5451>
Jonathan Marek [Sat, 13 Jun 2020 02:05:26 +0000 (22:05 -0400)]
turnip: use RenderPassCreateInfo for render_pass_add_implicit_deps
This gets rid of the some unnecessary values that were stored in
tu_render_pass for this. It also makes the render_pass_add_implicit_deps
more generic, with very few references to the tu_render_pass.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5451>
Jonathan Marek [Sat, 13 Jun 2020 01:43:24 +0000 (21:43 -0400)]
turnip: replace a memset(0) with zalloc in CreateRenderPass
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5451>
Jonathan Marek [Sat, 13 Jun 2020 01:19:20 +0000 (21:19 -0400)]
turnip: translate CreateRenderPass to CreateRenderPass2
It doesn't cut down the code size by much, and might not be the ideal for
performance (unless the compiler is unexpectedly smart), but makes it
easier to maintain (no modifying the same code in two places) and will
allow some simplifications since we wont have to worry about trying to
share code between the two versions.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5451>
Jonathan Marek [Tue, 23 Jun 2020 22:45:32 +0000 (18:45 -0400)]
turnip: implement depthBounds
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5628>
Jonathan Marek [Tue, 23 Jun 2020 22:44:42 +0000 (18:44 -0400)]
freedreno/registers: a6xx depth bounds test registers
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5628>
Rhys Perry [Tue, 16 Jun 2020 13:58:15 +0000 (14:58 +0100)]
aco: remove outdated assert in handle_operands()
"target" is no longer expected to be completely inside "swap".
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5626>
Rhys Perry [Wed, 24 Jun 2020 12:23:12 +0000 (13:23 +0100)]
aco: ignore blocked registers when checking edges in get_reg_impl()
If the only two registers available are consecutive and used by killed
operands, both of them will be blocked and fail the edge check.
Totals from 903 (0.66% of 135946) affected shaders:
VGPRs: 30892 -> 30884 (-0.03%)
CodeSize:
1584468 ->
1584044 (-0.03%); split: -0.05%, +0.02%
MaxWaves: 14374 -> 14378 (+0.03%)
Instrs: 306482 -> 306399 (-0.03%); split: -0.06%, +0.03%
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5626>
Samuel Pitoiset [Wed, 24 Jun 2020 13:38:19 +0000 (15:38 +0200)]
radv: fix checking the return value of cs_finalize()
cs_finalize() now returns a Vulkan error code and VK_SUCCESS is 0.
Fixes: 64a92ef7a26 ("radv/winsys: Distinguish device/host memory errors.")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5629>
Samuel Pitoiset [Wed, 24 Jun 2020 12:30:46 +0000 (14:30 +0200)]
gitlab-ci: update the list of expected failures for Pitcairn
These tests have been fixed as part of
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5625>
Bas Nieuwenhuizen [Tue, 23 Jun 2020 23:28:54 +0000 (01:28 +0200)]
radv: Make radv_alloc_shader_memory static.
Just a cleanup.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5578>
Bas Nieuwenhuizen [Sun, 21 Jun 2020 17:36:16 +0000 (19:36 +0200)]
radv/winsys: Distinguish device/host memory errors.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5578>
Bas Nieuwenhuizen [Sat, 20 Jun 2020 19:12:01 +0000 (21:12 +0200)]
radv: Handle mmap failures.
Which can happen if we have to many mmaps active in the process.
CC: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5578>
Bas Nieuwenhuizen [Sat, 20 Jun 2020 18:34:59 +0000 (20:34 +0200)]
radv/winsys: Deal with realloc failures in BO lists.
Otherwise if realloc fails we silently try to use it. Make recording fail instead.
CC: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5578>
Rhys Perry [Mon, 25 May 2020 14:36:12 +0000 (15:36 +0100)]
aco: improve vectorization of 8/16-bit loads/stores
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Tue, 23 Jun 2020 18:19:00 +0000 (19:19 +0100)]
aco: fix when sub-dword create_vector operand cannot be placed perfectly
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Daniel Schürmann [Tue, 23 Jun 2020 09:19:27 +0000 (10:19 +0100)]
aco: don't allow partial copies on GFX6/7
These are not supported due to missing SDWA instructions
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Daniel Schürmann [Tue, 23 Jun 2020 10:55:34 +0000 (11:55 +0100)]
aco: align swap operations to 4 bytes on GFX6/7
GFX6/7 can only swap full registers
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Thu, 11 Jun 2020 14:38:07 +0000 (15:38 +0100)]
aco: don't create byte-aligned short loads
The ISA docs don't seem to say if this is allowed, so just assume short
loads require short alignment.
In practice, the only situation this should affect are byte-aligned u8vec2
loads.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Tue, 26 May 2020 16:45:56 +0000 (17:45 +0100)]
aco: add missing bld.scc() in byte_align_scalar()
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Tue, 26 May 2020 15:43:05 +0000 (16:43 +0100)]
aco: don't store byte-aligned short stores
The ISA docs don't seem to say if this is allowed, so just assume short
stores require short alignment.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Tue, 26 May 2020 15:33:12 +0000 (16:33 +0100)]
aco: fix copy+paste error in split_buffer_store
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Mon, 25 May 2020 13:49:53 +0000 (14:49 +0100)]
radv/aco,aco: allow SMEM SSBO loads on GFX6/7
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Mon, 25 May 2020 13:47:25 +0000 (14:47 +0100)]
aco: allow SMEM for some sub-dword accesses
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Fri, 22 May 2020 16:55:15 +0000 (17:55 +0100)]
aco: only use SMEM if we can prove it's safe
Totals from 26 (0.02% of 127638) affected shaders:
SGPRs: 1680 -> 1664 (-0.95%)
VGPRs: 1492 -> 1504 (+0.80%)
CodeSize: 233140 -> 233016 (-0.05%); split: -0.09%, +0.04%
Instrs: 47121 -> 47114 (-0.01%); split: -0.08%, +0.06%
VMEM: 4930 -> 4655 (-5.58%); split: +0.12%, -5.70%
SMEM: 2030 -> 2001 (-1.43%); split: +3.79%, -5.22%
VClause: 891 -> 947 (+6.29%)
SClause: 876 -> 816 (-6.85%)
Copies: 4734 -> 4716 (-0.38%); split: -0.40%, +0.02%
Branches: 2048 -> 2047 (-0.05%)
PreSGPRs: 1400 -> 1396 (-0.29%)
PreVGPRs: 1440 -> 1443 (+0.21%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Mon, 15 Jun 2020 16:37:52 +0000 (17:37 +0100)]
radv: fix image variable types in meta shaders
We write to these variables using image intrinsics.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Rhys Perry [Thu, 21 May 2020 19:08:37 +0000 (20:08 +0100)]
spirv: set variables to restrict by default
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5207>
Mauro Rossi [Sat, 20 Jun 2020 18:47:05 +0000 (20:47 +0200)]
android: freedreno/ir3: simplify generated sources rules
Simplification and alignment with meson's sources generation rules
Changelog:
- move rules from src/gallium/drivers/freedreno/Android.gen.mk to Android.ir3.mk
- simplify LOCAL_GENERATED_SOURCES based on $(ir3_GENERATED_FILES)
- remove includes of src/gallium/drivers/freedreno/Android.gen.mk
- remove src/gallium/drivers/freedreno/Android.gen.mk
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Acked-by: Rob Clark <robdclark@gmail.com>
Acked-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5580>
Mauro Rossi [Sat, 20 Jun 2020 17:07:44 +0000 (19:07 +0200)]
android: freedreno/ir3: add missing generated sources and rules
Changelog:
- Makefile.sources: add ir3_lexer.c and ir3_parser.{c,h} generated sources
- Android.ir3.mk: add the necessary generated sources rules
- Android.ir3.mk: add the necessary include paths
- src/gallium/drivers/freedreno/Android.gen.mk: generate only ir3_nir_{imul,trig}.c for the moment
Fixes the following building error:
target C: libfreedreno_ir3 <= external/mesa/src/freedreno/ir3/ir3_assembler.c
FAILED: out/target/product/x86_64/obj/STATIC_LIBRARIES/libfreedreno_ir3_intermediates/ir3/ir3_assembler.o
...
external/mesa/src/freedreno/ir3/ir3_assembler.c:28:10: fatal error: 'ir3_parser.h' file not found
^~~~~~~~~~~~~~
1 error generated.
Fixes: 1e8808a4a0f ("freedreno/ir3: refactor out helper to compile shader from asm")
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Acked-by: Rob Clark <robdclark@gmail.com>
Acked-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5580>
Mauro Rossi [Sat, 20 Jun 2020 16:22:31 +0000 (18:22 +0200)]
android: freedreno: add fd5_layout.c to Makefile.sources
Fixes the following building error:
FAILED: out/target/product/x86_64/obj/SHARED_LIBRARIES/gallium_dri_intermediates/LINKED/gallium_dri.so
...
ld.lld: error: undefined symbol: fdl5_layout
clang-9: error: linker command failed with exit code 1 (use -v to see invocation)
Fixes: a1a739995ba ("freedreno/a5xx: Move resource layout to fdl.")
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Acked-by: Rob Clark <robdclark@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5580>
Bas Nieuwenhuizen [Sat, 28 Mar 2020 21:43:32 +0000 (22:43 +0100)]
vulkan/wsi/x11: Ensure we create at least minImageCount images.
Doom Eternal happily creates a swapchain with 2 images for IMMEDIATE...
This fixes a 10% performance issues with Doom Eternal for me.
Since the game only sets a minImageCount increasing till our own minimum
is totally okay.
CC: <stable@lists.freedesktop.org>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/2684
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3156
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4368>
Mike Blumenkrantz [Fri, 29 May 2020 18:38:02 +0000 (14:38 -0400)]
zink: clamp VkImageCreateInfo.arrayLayers to 1 for image resource creation
this is required by spec, so we can generally assume that any time it's 0 here
this is the result of us being lazy elsewhere in the zink driver when we're
manually creating this sort of buffer
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5614>
Samuel Pitoiset [Fri, 19 Jun 2020 13:18:27 +0000 (15:18 +0200)]
gitlab-ci: update the list of expected CTS failures for RADV/ACO
Based on Vulkan CTS 1.2.3.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5593>
Kenneth Graunke [Tue, 5 May 2020 18:03:12 +0000 (11:03 -0700)]
nir: Fix divergence analysis for tessellation input/outputs
The load_per_vertex_{input,output} intrinsics simply mean that they're
reading an arrayed input/output, which have one element per invocation.
Most accesses to those use gl_InvocationID as the subscript. However,
it's totally possible to read any element of the array. For example,
an evaluation shader might read gl_in[2].gl_Position, or a control
shader might read output[0].
For threads processing a single patch, an input/output load is
convergent if and only if both sources (the per-vertex-array subscript
and the offset) are convergent. For threads processing multiple
patches, we continued to mark them divergent.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5613>
Kenneth Graunke [Tue, 23 Jun 2020 21:04:42 +0000 (14:04 -0700)]
intel: Disable loading drivers on DG1 devices for now
Kernel support for DG1 has not yet been merged upstream; per our
long-standing DRM subsystem policy, we should not enable the platform
in userspace until the kernel patches are merged and functional.
We will re-enable this in the future. In the meantime, we retain all
of the infrastructure and code for the platform so that we can continue
developing DG1 support in upstream.
See a discussion here:
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4956#note_547775
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5617>
Kenneth Graunke [Wed, 24 Jun 2020 02:23:32 +0000 (19:23 -0700)]
CI: Disable Panfrost Mali-T820, Lima Mali-400 and Lima Mali-450 jobs
The runners appear to be unhealthy. Disable for now so people can merge
patches for other drivers in the meantime.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5619>
Jordan Justen [Wed, 13 Feb 2019 19:10:39 +0000 (11:10 -0800)]
iris/compute: Split out iris_load_indirect_location
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5571>
Jordan Justen [Wed, 13 Feb 2019 02:11:16 +0000 (18:11 -0800)]
iris: Split walker and state update into iris_upload_gpgpu_walker
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5571>
Jordan Justen [Wed, 14 Nov 2018 19:41:23 +0000 (11:41 -0800)]
anv/cmd_buffer: Split GPGPU_WALKER out to emit_gpgpu_walker
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5571>
Jordan Justen [Wed, 14 Nov 2018 18:49:05 +0000 (10:49 -0800)]
anv/pipeline: Split VFE/INTERFACE_DESCRIPTOR out to emit_media_cs_state
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5571>
Jonathan Marek [Tue, 23 Jun 2020 14:37:56 +0000 (10:37 -0400)]
turnip: use pipeline cs for shader programs instead of separate bo
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5606>
Jonathan Marek [Tue, 23 Jun 2020 14:37:18 +0000 (10:37 -0400)]
turnip: fix ts_cs_memory typo
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5606>
Kristian H. Kristensen [Tue, 23 Jun 2020 17:39:14 +0000 (10:39 -0700)]
freedreno: Handle DRM_FORMAT_MOD_INVALID in shared code
layout_resource_for_modifier() needs to handle DRM_FORMAT_MOD_INVALID
as well, since src/gallium/frontends/dri/dri2.c uses this to indicate
"no modifier" when it's called through the older non-modifier entry
points.
This is similar to
334788d4 ("freedreno: allow INVALID modifier") but
for the generic implementation.
Fixes: 98910626 ("freedreno/a6xx: Implement layout for DRM_FORMAT_MOD_QCOM_COMPRESSED")
Closes: #3154
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5611>
Jason Ekstrand [Thu, 22 Aug 2019 16:15:50 +0000 (11:15 -0500)]
intel/eu: Add the RNDU opcode
We don't want to use it on gen5 and earlier because only RNDD can be
done with a single instruction and we can implement RNDU(x) as -RNDD(-x)
so it's better to just do that when we have the instruction. On gen6
and above, we may as well just use the right instruction.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5596>
Jason Ekstrand [Thu, 17 Oct 2019 14:54:02 +0000 (09:54 -0500)]
intel/eu: Set the right subnr for ALIGN16 destinations
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5596>
Jason Ekstrand [Thu, 12 Sep 2019 18:23:46 +0000 (13:23 -0500)]
intel/eu: Add a brw_urb_dest_msg_type helper
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5596>
Kenneth Graunke [Fri, 9 Aug 2019 02:16:32 +0000 (19:16 -0700)]
intel/eu: Add a brw_urb_desc helper
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5596>
Jason Ekstrand [Tue, 13 Aug 2019 16:08:40 +0000 (11:08 -0500)]
intel/compiler: Expose brw_texture_offset to C
Some day we probably want to move it out of brw_shader if we're going to
share it with IBC but that can be another day.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5596>
Jason Ekstrand [Thu, 24 Oct 2019 22:31:18 +0000 (17:31 -0500)]
intel/fs: Move more prog_data setup into populate_wm_prog_data
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5596>
Jason Ekstrand [Thu, 18 Jul 2019 14:23:47 +0000 (09:23 -0500)]
intel/fs: Break wm_prog_data setup into a helper
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5596>
Jason Ekstrand [Thu, 18 Jul 2019 14:23:23 +0000 (09:23 -0500)]
intel/fs: Expose a couple of NIR lowering helpers
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5596>
Kenneth Graunke [Mon, 22 Jun 2020 22:49:31 +0000 (15:49 -0700)]
iris: Delete useless #define
When I was bringing up the driver, I had BLORP use #ifdefs for
softpin-mode vs. relocation-mode. That all got reworked during
review, but apparently this #define is still kicking around,
even though nothing uses it.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5610>
Jose Maria Casanova Crespo [Wed, 17 Jun 2020 00:30:23 +0000 (02:30 +0200)]
nir: only uniforms with dynamically_uniform offset are dynamically_uniform
Previously all nir_intrinsic_load_uniform that were used as sources were
considered to be dynamically_uniform but when offsets of load_uniform
are indirect it can not be determined.
This fixes artefacts in Google Maps 3D view in V3D.
Fixes: 886d46b08971 ("nir: Add a function to determine if a source is dynamically uniform")
Reviewed-by: Neil Roberts <nroberts@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5587>
Eric Engestrom [Fri, 19 Jun 2020 10:44:41 +0000 (12:44 +0200)]
bin/symbols-check: explain C++ symbols workaround
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5556>
Jonathan Marek [Sat, 20 Jun 2020 19:09:05 +0000 (15:09 -0400)]
turnip: implement CmdDrawIndirectByteCountEXT
Fixes these deqp tests:
dEQP-VK.transform_feedback.simple.backward_dependency*
dEQP-VK.transform_feedback.simple.draw_indirect*
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5579>
Jonathan Marek [Sat, 20 Jun 2020 19:02:10 +0000 (15:02 -0400)]
turnip: enable VK_EXT_index_type_uint8
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5579>
Jonathan Marek [Mon, 22 Jun 2020 16:47:47 +0000 (12:47 -0400)]
turnip: refactor CmdDraw* functions (and a few fixes)
This cleans up the CmdDraw* functions to be more straightforward. And a few
fixes applied while going through it:
* Fix indirect draw commands not adding the buffer->bo_offset, and ignoring
drawCount/stride parameters (deqp tests not testing indirect draws very
much apparently).
* Fixed a potential issue with RESTART_INDEX + secondary command bufs.
* Add missing logic for 8-bit indices
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
# Conflicts:
# src/freedreno/vulkan/tu_cmd_buffer.c
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5579>
Jonathan Marek [Sat, 20 Jun 2020 17:46:01 +0000 (13:46 -0400)]
turnip: rework streamout state and add missing counter buffer read/writes
Rework the streamout state and at the same time fix some issues, the
biggest one being to actually use the counter buffers instead of ignoring
them completely.
(note it appears the dEQP tests are bad and able to pass with the previous
broken behavior of not ever reading/writing from the counter buffers)
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5579>
Eric Engestrom [Wed, 27 May 2020 23:40:11 +0000 (01:40 +0200)]
docs: add planning for 20.2
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5239>
Eric Engestrom [Wed, 17 Jun 2020 22:10:51 +0000 (00:10 +0200)]
docs: add some padding to the release calendar
This extra padding allows for `-rcX` suffixes.
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5239>
Rob Clark [Sat, 20 Jun 2020 16:21:02 +0000 (09:21 -0700)]
ci: remove some freedreno a6xx skips
These don't seem to be flakey anymore. I did still see a flake with
dEQP-GLES31.functional.layout_binding.ssbo.fragment_binding_array so
I put that one back in.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5577>
Marek Olšák [Sat, 13 Jun 2020 21:19:51 +0000 (17:19 -0400)]
driconf: add workarounds for SPECviewperf13
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5459>
Marek Olšák [Sat, 13 Jun 2020 21:18:34 +0000 (17:18 -0400)]
glsl,driconf: add allow_glsl_120_subset_in_110 for SPECviewperf13
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5459>
Marek Olšák [Tue, 16 Jun 2020 18:39:04 +0000 (14:39 -0400)]
radeonsi: replace ctx->screen with sscreen in si_flush_gfx_cs
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5506>
Marek Olšák [Mon, 4 May 2020 21:41:41 +0000 (17:41 -0400)]
radeonsi: don't wait for idle at the end of gfx IBs
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5506>
Samuel Pitoiset [Thu, 18 Jun 2020 10:28:42 +0000 (12:28 +0200)]
gitlab-ci: stop testing RADV with LLVM
ACO is going to be our default compiler soon and it seems useless
to waste CI resources for LLVM.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5543>
Samuel Pitoiset [Mon, 22 Jun 2020 11:33:21 +0000 (13:33 +0200)]
aco: fix printing ASM on GFX6-7 again
Checking errno is actually wrong because it's only updated if
popen() fails (ie. NULL). One solution is to check if the first
line is empty.
Fixes: c95d258d1bc ("aco: fix printing ASM on GFX6-7 if clrxdisasm is not found")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5591>
Tomeu Vizoso [Mon, 15 Jun 2020 13:39:40 +0000 (15:39 +0200)]
gitlab-ci: Update CTS runner
We need a newer version to be able to successfully run the OpenGL suites
in dEQP.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5494>
Tomeu Vizoso [Wed, 13 May 2020 07:46:06 +0000 (09:46 +0200)]
gitlab-ci: Test virgl with Khronos' OpenGL CTS
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5494>
Tomeu Vizoso [Tue, 12 May 2020 08:18:48 +0000 (10:18 +0200)]
gitlab-ci: Add manual tests for Virgl using GLES on the host
The ones that run automatically will use big GL on the host.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5494>
Tomeu Vizoso [Tue, 12 May 2020 08:03:25 +0000 (10:03 +0200)]
gitlab-ci: Run more dEQP tests for virgl
Llvmpipe seems to have become faster, and we can run more tests while
still being under 5 minutes per job.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5494>
Marek Olšák [Sat, 30 May 2020 21:42:46 +0000 (17:42 -0400)]
glthread: handle ARB_vertex_attrib_binding
This handles ARB_vertex_attrib_binding for vertex uploads correctly.
Before this, the extension might have led to crashes if non-VBO vertex
attribs were present.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5303>
Marek Olšák [Sun, 31 May 2020 02:39:54 +0000 (22:39 -0400)]
glthread: rename non_vbo_attrib_mask -> user_buffer_mask, attribs -> buffers
just a cleanup, no change in behavior
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5303>
Marek Olšák [Sat, 30 May 2020 21:21:32 +0000 (17:21 -0400)]
glapi: fix incorrect param names in ARB_vertex_attrib_binding functions
no change in behavior
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5303>
Marek Olšák [Wed, 17 Jun 2020 19:25:28 +0000 (15:25 -0400)]
ac/nir: fix 64-bit division for GL CTS
This fixes: KHR-GL45.gpu_shader_fp64.builtin.mod_*
Fixes: ba2ec1f3 "ac/nir: use llvm.amdgcn.rcp in ac_build_fdiv()"
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5531>
Marek Olšák [Wed, 17 Jun 2020 22:04:53 +0000 (18:04 -0400)]
radeonsi: compact MRTs to save PS export memory space
If there are holes between color outputs (e.g. a shader exports MRT1, but
not MRT0), we can remove the holes by moving higher MRTs lower.
The hardware will remap the MRTs to their correct locations if we remove
holes in SPI_SHADER_COL_FORMAT but not CB_SHADER_MASK.
This is a performance optimization, but MRTs with holes are pretty rare,
so there is most likely no effect on any app.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5535>
Jason Ekstrand [Mon, 22 Jun 2020 21:15:02 +0000 (16:15 -0500)]
anv: Bump the advertised patch version to 145
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5597>
Jason Ekstrand [Mon, 22 Jun 2020 21:12:39 +0000 (16:12 -0500)]
vulkan: Update Vulkan XML and headers to 1.2.145
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5597>
Eric Engestrom [Fri, 19 Jun 2020 22:35:58 +0000 (00:35 +0200)]
docs: cat maintainer keys to a single file
The original issue asked for all the keys in a single file, but I didn't
do that because it's much easier to manage and verify the keys as
separate files, but sphinx doesn't provide a way to expose a folder so
we'd need to create an index.html and have it list all the keys
manually, which is very error prone.
At this point, we might as well just concatenate the keys and expose
a single file, so let's do that.
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5568>
Eric Engestrom [Fri, 19 Jun 2020 22:41:53 +0000 (00:41 +0200)]
docs: drop deleted file from extra sphinx files
Fixes: 3e37b7e6bbad007c168a ("docs: remove plain-text copy of versions.rst")
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5567>