Samuel Pitoiset [Tue, 20 Oct 2020 14:56:29 +0000 (16:56 +0200)]
radv,aco: fix use of texop_samples_identical in the resolve meta path
The return value of this texture intrinsic should be a NIR 1-bit bool.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7236>
Samuel Pitoiset [Tue, 20 Oct 2020 14:54:45 +0000 (16:54 +0200)]
ac/nir: do not sign-extend the result of texop_samples_identical
The return value should be a NIR 1-bit bool.
This fixes a regression with piglit
ext_shader_samples_identical-simple-fs 2 on RadeonSI.
Fixes:
e690a1b78bf ("ac/llvm: don't lower bool to int32, switch to native i1 bool")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7236>
Alejandro Piñeiro [Wed, 14 Oct 2020 07:48:44 +0000 (09:48 +0200)]
docs/features: add v3dv driver
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7125>
Thong Thai [Thu, 15 Oct 2020 19:32:47 +0000 (15:32 -0400)]
frontends/va/postproc: Un-break field flag
Fixes an issue where deinterlaced videos would play at half the
framerate, since only one field was repeated, instead of using both
fields. Reverts a change I made previously which broke this.
Fixes:
78786a219ea ("frontends/va: Fix deinterlace bottom field first flag")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3621
Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7194>
Tony Wasserka [Tue, 6 Oct 2020 20:52:02 +0000 (22:52 +0200)]
aco/isel: Miscellaneous cleanups using the new Stage API
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7094>
Tony Wasserka [Wed, 7 Oct 2020 16:21:48 +0000 (18:21 +0200)]
aco: Clean up symbol names and comments related to NGG
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7094>
Tony Wasserka [Mon, 5 Oct 2020 15:50:37 +0000 (17:50 +0200)]
aco: Use strong typing to model SW<->HW stage mappings
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7094>
Eric Anholt [Tue, 20 Oct 2020 20:43:03 +0000 (13:43 -0700)]
nir: Only validate in passes that might have changed things.
If a pass returning boolean progress reports no change, we shouldn't need
to re-validate. If a pass breaks the NIR but also fails to report
progress correctly, it would be up to the next pass to catch that.
This should hopefully help with test timeouts on
KHR-GL33.texture_swizzle.functional since switching softpipe to
nir-to-tgsi and enabling NIR validation in CI (27s to 20s on my system).
Suggested-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7239>
Timothy Arceri [Fri, 16 Oct 2020 08:26:51 +0000 (19:26 +1100)]
glsl: relax rule on varying matching for shaders older than 4.00
Please see new code commment for full justification.
Fixes:
18004c338f6b ("glsl: fail when a shader's input var has not an equivalent out var in previous")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3648
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7184>
Icecream95 [Wed, 14 Oct 2020 11:08:36 +0000 (00:08 +1300)]
panfrost: Only enable occlusion queries when active
Fixes piglit tests:
- occlusion_query_meta_fragments
- occlusion_query_meta_save
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7074>
Icecream95 [Fri, 9 Oct 2020 07:01:16 +0000 (20:01 +1300)]
panfrost: Precise occlusion query support
In the "counter" occlusion query mode, each shader core has a separate
counter.
When MSAA is disabled, counter values need to be divided by four.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7074>
Eric Anholt [Wed, 26 Aug 2020 19:51:15 +0000 (12:51 -0700)]
st/mesa: Drop the TGSI paths for drawpixels and use nir-to-tgsi if needed.
Now that we have a NIR translator in gallium, we can drop this duplicated
code in the API implementation.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6477>
Eric Anholt [Wed, 26 Aug 2020 19:42:36 +0000 (12:42 -0700)]
st/mesa: Drop the TGSI paths for PBOs and use nir-to-tgsi if needed.
Now that we have a NIR translator in gallium, we can drop this duplicated
code in the API implementation.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6477>
Eric Anholt [Wed, 26 Aug 2020 19:14:42 +0000 (12:14 -0700)]
gallium/ntt: Add default compiler options for non-native-NIR drivers.
For using NIR internally in mesa/st, we need some compiler options. This
is the options wrote for softpipe, which should be enough for whatever NTT
we do.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6477>
Vinson Lee [Thu, 15 Oct 2020 00:20:14 +0000 (17:20 -0700)]
v3dv: Fix assert using assign instead of compare.
Fix defect reported by Coverity Scan.
Assign instead of compare (PW.ASSIGN_WHERE_COMPARE_MEANT)
assign_where_compare_meant: use of "=" where "==" may have been intended
Fixes:
ca86c7c65a82 ("v3dv: assert command buffers are executable when submitting to a queue")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7156>
Bas Nieuwenhuizen [Fri, 18 Sep 2020 01:11:18 +0000 (03:11 +0200)]
radv: Advertise VK_KHR_shader_terminate_invocation.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7226>
Bas Nieuwenhuizen [Fri, 18 Sep 2020 01:09:19 +0000 (03:09 +0200)]
amd/llvm: Add VK_KHR_shader_terminate_invocation support.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7226>
Bas Nieuwenhuizen [Fri, 18 Sep 2020 01:08:43 +0000 (03:08 +0200)]
aco: Add VK_KHR_shader_terminate_invocation support.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7226>
Eric Anholt [Mon, 19 Oct 2020 18:54:24 +0000 (11:54 -0700)]
ci: Enable Werror on meson-arm64-build-test.
I found that it was warnings-clean already, let's make sure it stays that
way.
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7224>
Eric Anholt [Mon, 19 Oct 2020 19:17:38 +0000 (12:17 -0700)]
turnip: Add error path handling for descriptor pool init.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7224>
Eric Anholt [Mon, 19 Oct 2020 19:14:26 +0000 (12:14 -0700)]
turnip: Handle the error path for tu/drm's vkResetFences().
OUT_OF_MEMORY is the only valid error code from this function, but this
error is more of a "things went horribly wrong, you can't talk to the GPU"
case. Set the device to be in error.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7224>
Eric Anholt [Mon, 19 Oct 2020 19:03:33 +0000 (12:03 -0700)]
turnip: Handle some error paths in allocating CS space from a command buffer.
Fixes some release-build warnings.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7224>
Eric Anholt [Mon, 19 Oct 2020 18:56:17 +0000 (11:56 -0700)]
freedreno/fdperf: Silence a compiler warning about current counter.
It seems like selecting the first here is a fine choice if we can't
find the counter.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7224>
Eric Anholt [Mon, 19 Oct 2020 18:46:43 +0000 (11:46 -0700)]
freedreno/tools: Fix compiler warnings about using sz in the error paths.
If we don't check for a NULL str, then sz might be undefined (as was
happening in the match_compatible path, and returning 0 makes us not match
as we should).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7224>
Eric Anholt [Mon, 19 Oct 2020 18:41:57 +0000 (11:41 -0700)]
freedreno/cffdec: Fix format overflow warning.
../src/freedreno/decode/cffdec.c: In function ‘reg_disasm_gpuaddr’:
../src/freedreno/decode/cffdec.c:404:29: error: ‘sprintf’ writing a
terminating nul past the end of the destination [-Werror=format-overflow=]
404 | sprintf(filename, "%04d.%s", n++, ext);
../src/freedreno/decode/cffdec.c:404:3: note: ‘sprintf’ output between
9 and 16 bytes into a destination of size 8
404 | sprintf(filename, "%04d.%s", n++, ext);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7224>
Karol Herbst [Sun, 11 Oct 2020 21:17:20 +0000 (23:17 +0200)]
llvmpipe: enable CL images
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Dave Airlie [Sun, 11 Oct 2020 23:45:40 +0000 (09:45 +1000)]
llvmpipe: fix sampler/image binding for clover.
Clover uses these APIs a bit different, avoid crashes
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Karol Herbst [Thu, 8 Oct 2020 11:18:07 +0000 (13:18 +0200)]
nvc0/CL: enable images
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Karol Herbst [Sat, 15 Aug 2020 10:57:03 +0000 (12:57 +0200)]
nouveau: hide SVM support behing a variable for now as kernel space is broken
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Karol Herbst [Thu, 8 Oct 2020 11:20:16 +0000 (13:20 +0200)]
clover/nir: set kernel_image cap
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Jason Ekstrand [Mon, 31 Aug 2020 23:09:44 +0000 (18:09 -0500)]
clover/nir: Add an image lowering pass
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Jason Ekstrand [Wed, 2 Sep 2020 23:19:35 +0000 (18:19 -0500)]
clover/nir: Calculate sizes of images and samplers properly
Clover uses very specific sizes and alignments for images and samplers
to pass various bits of data. We need to add a new size/align helper
for inputs which matches the standard CL size/align for most types but
also has the right size/align for images and samplers.
v2 (Karol): use sizeof(cl_mem) instead of 8 to fix 32 bit runtimes.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Karol Herbst [Thu, 15 Oct 2020 14:31:10 +0000 (16:31 +0200)]
clover/device: use PIPE_MAX_SHADER_SAMPLER_VIEWS for max_images_read
Read images are really just normal textures and OpenCL requires 128 anyway.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Serge Martin [Sun, 11 Oct 2020 18:12:12 +0000 (20:12 +0200)]
clover: clCreateImage: calculate image row_pitch and slice_pitch when not provided
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Karol Herbst [Wed, 14 Oct 2020 23:19:57 +0000 (01:19 +0200)]
clover: support custom driver strides
This is required by llvmpipe as it sets explicit strides on buffers and
textures.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Serge Martin [Sun, 11 Oct 2020 20:24:26 +0000 (22:24 +0200)]
clover: validate image_row_pitch and image_slice_pitch in clEnqueueMapImage
v2 (Karol Herbst): remove filling values as it was incorrect.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Aaron Watry [Wed, 6 Nov 2019 04:26:58 +0000 (22:26 -0600)]
clover: Fix incorrect error check in clGetSupportedImageFormats
From CL1.2 Section 5.3.2:
returns CL_INVALID_VALUE ... if num_entries is 0 and image_formats is not NULL.
We were checking the num_image_formats param, not num_entries.
Signed-off-by: Aaron Watry <awatry@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Karol Herbst [Wed, 7 Oct 2020 21:08:43 +0000 (23:08 +0200)]
clover: use pipe_image_view for images instead of set_compute_resources
Long term we want to git rid of set_compute_resources, this is just one
piece. The other bit would be to use the proper const buffer interfaces,
but because that path is only hit with r600/radeonsi I won't touch it.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Jason Ekstrand [Tue, 6 Oct 2020 17:40:45 +0000 (12:40 -0500)]
compiler/types: Allow images and samplers in get_explicit_type_for_size_align
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Jason Ekstrand [Thu, 8 Oct 2020 07:14:47 +0000 (02:14 -0500)]
compiler/types: Assert non-zero alignments in get_explicit_type_for_size_align
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Jason Ekstrand [Thu, 8 Oct 2020 07:27:24 +0000 (02:27 -0500)]
nir/lower_io: Assert non-zero power-of-two alignments
The way the ALIGN_POT macro works, an alignment of 0 may cause
ALIGN_POT(x, 0) to return 0 for any x. Throw in an assert to guard
against this case.
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Jason Ekstrand [Tue, 1 Sep 2020 00:33:51 +0000 (19:33 -0500)]
spirv: Add 0.5 to integer coordinates for OpImageSampleExplicitLod
Just casting to a float is insufficient because that gives us the upper-left corner
of the texel rather than the center.
Fixes:
701cb9d60c03 "nir/vtn: Handle integer sampling coordinates"
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7069>
Jason Ekstrand [Tue, 20 Oct 2020 03:36:07 +0000 (22:36 -0500)]
anv,iris: Use the data cache for UBO pulls on Gen12+
Now that we have the HDC, using the data cache for UBO pulls seems to
help things quite a bit:
GTA V DXVK 104.0%
Talos Principle GL 102.8%
Rise of Tomb Raider VK 102.8%
Dark Souls 3 DXVK 101.4%
Witcher3 DXVK 101.3%
Bioshock Infinite GL 100.5%
Doom 2016 VK 97.7%
Doom is a bit of a loss but it helps enough other stuff, it's probably
worth the hit.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7230>
Jason Ekstrand [Tue, 20 Oct 2020 03:33:05 +0000 (22:33 -0500)]
iris: Flush caches based on brw_compiler::indirect_ubos_use_sampler
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7230>
Marijn Suijten [Tue, 20 Oct 2020 18:00:18 +0000 (20:00 +0200)]
android: gallium/auxiliary: Deduplicate nir_to_tgsi.c inclusion
Both commits add nir_to_tgsi.c to a different variable, causing a
build-time error when compiling in an AOSP tree:
build/make/core/binary.mk:970: error: overriding commands for target `..../obj/STATIC_LIBRARIES/libmesa_gallium_intermediates/nir/nir_to_tgsi.o', previously defined at build/make/core/binary.mk:970
Move all sources into NIR_SOURCES to resolve this issue.
Fixes:
d0f8fe59091 ("softpipe: Switch to using NIR as the shader format from mesa/st.")
Fixes:
34cc6a804ec ("gallium: Add a nir-to-TGSI pass.")
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7237>
Eric Anholt [Fri, 16 Oct 2020 22:50:40 +0000 (15:50 -0700)]
docs: Document how to build and install Android drivers.
This is what I've been using so far.
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7227>
Eric Anholt [Mon, 19 Oct 2020 20:47:58 +0000 (13:47 -0700)]
meson: Don't enable libunwind by in 'auto' mode on Android.
On Android we're expected to use their backtrace library.
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7227>
Eric Anholt [Mon, 19 Oct 2020 20:46:11 +0000 (13:46 -0700)]
meson: Don't try to build GLX by default on Android.
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7227>
Eric Anholt [Mon, 19 Oct 2020 21:19:27 +0000 (14:19 -0700)]
freedreno: Use Android's libsync instead of libdrm's.
This should be equivalent, given that we only support the DRM backend so
far, but we'll want to have the compatibility for future KGSL freedreno.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7227>
Eric Anholt [Mon, 19 Oct 2020 19:28:17 +0000 (12:28 -0700)]
meson: Only require libexpat when a part of the build needs it.
Now that xmlconfig can be built without libexpat on Android, we can make
android builds not require the presence of libexpat for many drivers.
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7223>
Michael Olbrich [Sun, 8 Sep 2019 11:14:40 +0000 (13:14 +0200)]
meson.build: xxf86vm is not needed for -Dglx-direct=false
It is only used in src/glx/glxcmds.c and when GLX_DIRECT_RENDERING is
defined.
So only depend on it if GLX direct rendering is actually enabled.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1905>
Eric Anholt [Wed, 1 Jan 2020 00:20:52 +0000 (16:20 -0800)]
softpipe: Switch to using NIR as the shader format from mesa/st.
This causes our TGSI to use far more temps, since NTT is currently not
releasing temps from registers. On the other hand, this interpreter is
already spectacularly slow, and if we wanted to go fast we should probably
write a scalar NIR intrepeter.
For now, using NTT means that we test that codepath in preparation for
switching TGSI-consuming HW drivers over, so that we can eventually
garbage collect st_glsl_to_tgsi.
As this is a major restructuring, there are some impacts on piglit:
- Several tests start assert failing about 64-bit NIR registers for temp
arrays not getting split to vec2s:
- fs-frexp-dvec4-variable-index.shader_test
- arb_gpu_shader_fp64/uniform_buffers/{vs,fs,gs}-array-copy.shader_test
- arb_gpu_shader_int64/execution/indirect-array-two-accesses.shader_test
- dEQP-GLES31.functional.primitive_bounding_box.wide_points.global_state.vertex_geometry_fragment.fbo_bbox_larger
starts crashing depending on various bits of state (previous tests run
before it, presence of valgrind, presence of glib's memcheck). Doesn't
seem really NTT-specific, added to flakes list with other GS flakes.
- Almost 200 fp64/int64-related tests start passing, mostly around i/o loayout.
shader-db:
total instructions in shared programs: 3492656 -> 3081674 (-11.77%)
total loops in shared programs: 1418 -> 1387 (-2.19%)
total temps in shared programs: 340041 -> 615527 (81.02%)
total const in shared programs: 3158970 -> 1528630 (-51.61%)
total imm in shared programs: 117586 -> 101349 (-13.81%)
Total CPU time (seconds): 430.36 -> 900.94 (109.35%)
FPS results:
glmark2 texture +7.32484% +/- 3.76528% (n=10)
glmark2 desktop:effect=shadow +20% +/- 0% (n=10)
glmark2 shadow +6.49351% +/- 3.65335% (n=7)
glmark2 conditionals +18.75% +/- 2.74658% (n=9)
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3395>
Eric Anholt [Wed, 26 Aug 2020 03:58:27 +0000 (20:58 -0700)]
softpipe: Fix buffer overflows in SSBO atomics.
SSBO atomics are always to the .x channel, but we were doing reads on
.xyzw and writes to whatever the writemask was.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3395>
Eric Anholt [Wed, 5 Nov 2014 20:41:08 +0000 (12:41 -0800)]
gallium: Add a nir-to-TGSI pass.
The goal is to replace glsl_to_tgsi.cpp and its supporting code (~10k
LOC). This code ends up being smaller because NIR has many lowering
passes that help it map better to TGSI than GLSL IR does.
As a benefit, this brings NIR optimizations to TGSI-only drivers.
Many of the softpipe shaders I've looked at end up being significantly
shorter. Some potentially relevant changes to TGSI consumers:
- All immediates are now UINT typed. This means they're less legible
in printouts, but means that they get deduplicated better (no more
multiple copies of 0x0!)
- Sampler views are not currently declared.
- nir_registers don't have their live ranges tracked, so TGSI temp usage
may go up with a lot of control flow.
- nir_lower_vec_to_mov naively inserts movs instead of trying to coalesce
the movs with the generators of the ssa values, sometimes increasing
instruction count.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3395>
Eric Anholt [Fri, 24 Jul 2020 22:13:40 +0000 (15:13 -0700)]
nir: Add an option to not lower source mods for f64/u64/i64.
TGSI can't handle them, but we want to use this pass for nir-to-tgsi.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3395>
Eric Anholt [Wed, 22 Jul 2020 16:06:48 +0000 (09:06 -0700)]
nir: Add a call to get a struct describing SSA liveness per instruction.
nir-to-tgsi will use this to release release temporaries for SSA storage
back to ureg's linear register allocation once they're dead.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3395>
Eric Anholt [Wed, 14 Oct 2020 19:11:20 +0000 (12:11 -0700)]
nir: Add a block start/end ip to live instr index metadata.
I wanted it for the per-instruction live intervals metadata, and it's not
much to store in general. Make the ip explicitly 32-bit, on suggestion by
jekstrand.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3395>
Eric Anholt [Tue, 29 Sep 2020 22:44:57 +0000 (15:44 -0700)]
nir: Replace nir_ssa_def->live_index with nir_instr->index.
live_index had two things going on: 0 meant the instr was an undef and
always dead, and otherwise ssa defs had increasing numbers by instruction
order. We already have a field in the instruction for storing instruction
order, and ssa defs don't need that number to be contiguous (if you want a
compact per-ssa-def number, use ssa->index after reindexing).
We don't use ssa->index for this, because reindexing those would change
nir_print, and that would be rude to people trying to track what's
happening in optimization passes.
This openend up a hole in nir_ssa_def, so we move nir_ssa_def->index
toward the end to shrink the struct from 64 bytes to 56.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3395>
Eric Anholt [Tue, 29 Sep 2020 22:33:39 +0000 (15:33 -0700)]
nir: Introduce nir_metadata_instr_index for nir_index_instr() being current.
This will be useful to remove the live_index field from nir_ssa_def.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3395>
Eric Anholt [Sat, 17 Oct 2020 00:20:11 +0000 (17:20 -0700)]
ci: Enable NIR_VALIDATE everywhere.
I wasted a bunch of time today tracking down a spurious test results
change due to a driver invoking UB by running tests where NIR validation
had failed (instruction reading from components beyond vector size). If
we need to shrink our coverage to get runtimes down, it will still be
better to be catching validation errors in CI.
To keep the test jobs runtime under 10 minutes, I've split a530's gles2 to
two different jobs.
Acked-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7203>
Samuel Pitoiset [Thu, 8 Oct 2020 06:11:30 +0000 (08:11 +0200)]
radv: move all NIR pass outside of ACO
This has several advantages:
- it generates roughly the same NIR for both compiler backends
(this might help for debugging purposes)
- it might allow to move around some NIR pass to improve compile time
- it might help for RadeonSI support
- it improves fossils-db stats for RADV/LLVM (this shouldn't matter
much but it's a win for free)
fossil-db (Navi/LLVM):
Totals from 80732 (59.18% of 136420) affected shaders:
SGPRs: 5390036 -> 5382843 (-0.13%); split: -3.38%, +3.24%
VGPRs: 3910932 -> 3890320 (-0.53%); split: -2.38%, +1.85%
SpillSGPRs: 319212 -> 283149 (-11.30%); split: -17.69%, +6.39%
SpillVGPRs: 14668 -> 14324 (-2.35%); split: -7.53%, +5.18%
CodeSize:
265360860 ->
267572132 (+0.83%); split: -0.47%, +1.30%
Scratch: 5338112 -> 6134784 (+14.92%); split: -2.65%, +17.57%
MaxWaves: 1077230 -> 1086902 (+0.90%); split: +2.79%, -1.90%
No fossils-db changes on RADV/ACO.
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/7077>
Samuel Pitoiset [Fri, 9 Oct 2020 06:43:46 +0000 (08:43 +0200)]
ac/nir: handle non-const offset with txf/txf_ms
It might be a vec2. If it's a constant, LLVM will fold it.
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/7077>
Marek Olšák [Tue, 8 Sep 2020 08:40:02 +0000 (04:40 -0400)]
ac/llvm: don't lower bool to int32, switch to native i1 bool
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7077>
James Park [Mon, 19 Oct 2020 10:51:46 +0000 (03:51 -0700)]
util: Fix rwlock Windows include for MinGW
MinGW uses windows.h, and not Windows.h.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7215>
Lionel Landwerlin [Fri, 16 Oct 2020 09:46:26 +0000 (12:46 +0300)]
genxml: drop gen10
Finishing off the job started in !6899
v2: Remove remaining gen10_pack.h include (Sagar)
v3: Forgot isl gen10 removal (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7185>
Samuel Pitoiset [Fri, 16 Oct 2020 08:41:49 +0000 (10:41 +0200)]
Revert "radv/aco: disable NGG GS support because it randomly hangs the GPU"
This reverts commit
b84d1a0c42c5a1973ebc53a49fd941590e540a82.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7213>
Timur Kristóf [Mon, 19 Oct 2020 09:49:05 +0000 (11:49 +0200)]
aco/ngg: Add assertion to make sure we always know the vertex count.
Just a sanity check to avoid hangs caused by missing this
in the future.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7213>
Timur Kristóf [Mon, 19 Oct 2020 09:45:35 +0000 (11:45 +0200)]
nir: Emit set_vertex_and_primitive_count for inactive streams.
This fixes issues in backends such as ACO which rely on always
getting this intrinsic to know the correct vertex and primitive
count.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7213>
Samuel Pitoiset [Mon, 19 Oct 2020 16:42:22 +0000 (18:42 +0200)]
radv: add missing 'discardtodemote' option in the debug list
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/7220>
Marcin Ślusarz [Fri, 16 Oct 2020 18:07:28 +0000 (20:07 +0200)]
vulkan/wsi: fix possible random stalls in wsi_display_wait_for_event
pthread_cond_broadcast man page says this:
"The pthread_cond_broadcast() or pthread_cond_signal() functions may
be called by a thread whether or not it currently owns the mutex that
threads calling pthread_cond_wait() or pthread_cond_timedwait() have
associated with the condition variable during their waits; however,
if predictable scheduling behavior is required, then that mutex shall
be locked by the thread calling pthread_cond_broadcast() or
pthread_cond_signal()."
Found by reading the code.
Compile tested only.
Fixes:
da997ebec92 ("vulkan: Add KHR_display extension using DRM [v10]")
Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7197>
Vinson Lee [Thu, 15 Oct 2020 23:14:16 +0000 (16:14 -0700)]
panfrost: Fix stride for AFBC_FORMAT_MOD_BLOCK_SIZE_32x8.
Fix defect reported by Coverity Scan.
Identical code for different branches (IDENTICAL_BRANCHES)
identical_branches: Ternary expression on condition width has
identical then and else expressions: 32. Should one of the
expressions be modified, or the entire ternary expression
replaced?
Fixes:
8bb1d61f276c ("panfrost: Add panfrost_block_dim helper")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7177>
Caio Marcelo de Oliveira Filho [Thu, 7 May 2020 04:56:29 +0000 (21:56 -0700)]
anv: Advertise VK_KHR_shader_terminate_invocation
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7221>
Caio Marcelo de Oliveira Filho [Mon, 19 Oct 2020 17:04:32 +0000 (10:04 -0700)]
vulkan: Update XML and headers to 1.2.158
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7221>
Dave Airlie [Mon, 19 Oct 2020 00:38:16 +0000 (10:38 +1000)]
clover/nir: add a constant folding pass before lowering mem const
If we lower mem constants first, then direct array accesses to
constants never get lowered, so do a constant fold pass first to
remove direct const array accesses.
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7209>
Kenneth Graunke [Wed, 7 Oct 2020 16:45:35 +0000 (09:45 -0700)]
isl: Enable Tigerlake HDC:L1 caches via MOCS in various cases.
Thanks to Felix Degrood for discovering that we missed enabling this
additional caching on Tigerlake! Felix also benchmarked the changes.
We now use MOCS 48 (HDC:L1 + L3 + LLC) for render targets, textures,
and pull constant buffers. We leave storage buffers & images, as well
as stateless messages, using the previous MOCS 2 value. We can't use
HDC:L1 with atomics, and we don't know a priori whether storage buffers
will be used with atomics or not. Similarly, the Vulkan buffer device
address feature allows atomics to be performed on buffers via stateless
messages, and we only can control MOCS at the base address level, so
we can't do much there.
This is closer to what the Windows Vulkan and OpenGL drivers do,
though it isn't quite the same - they also disable LLC in some cases,
but we observed this to have noticable performance regressions when
we tried (though a couple titles benefited). We may try experiment
with that in the future.
Improves performance in a number of titles:
- Unreal Engine 4 Shooter Demo [VK]: 11.8%
- Witcher 3 [DXVK]: 3.9%
- Rise of the Tomb Raider [VK]: 1.5%
- Shadow of the Tomb Raider [VK]: 1.0%
- Grand Theft Auto V [DXVK]: 0.8%
We did not observe any performance regressions.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7104>
Kenneth Graunke [Wed, 7 Oct 2020 14:44:56 +0000 (07:44 -0700)]
isl, anv, iris: Add a centralized helper to select MOCS based on usage
On Gen12+, we can enable additional caches in certain usage situations.
This routes that decision making to a central place in ISL, based on
surface usage flags, and updates both drivers to use it. (i965 doesn't
need to change because it doesn't support Gen12.)
We continue handling the "external" decision via an anv_mocs() wrapper
for now, since we store that flag in anv_bo, which isl doesn't know
about. (We could introduce an ISL_SURF_USAGE_EXTERNAL, but I'm not
actually sure that would be cleaner.)
This patch should not have any functional nor performance effects, as
we continue selecting the exact same MOCS values for now.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7104>
Kenneth Graunke [Wed, 30 Sep 2020 22:05:54 +0000 (15:05 -0700)]
anv: Set only one ISL usage bit (RT/texture) for CopyBuffer sources
Most uses of this function deal with destination buffers, but for
copy_buffer_to_image, the buffer is the source, and isn't rendered
to. We should avoid setting ISL_SURF_USAGE_RENDER_TARGET_BIT.
Also, we should avoid setting ISL_SURF_USAGE_TEXTURE_BIT for the
destination, which isn't sampled from.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7104>
Eric Anholt [Fri, 18 Sep 2020 19:43:18 +0000 (12:43 -0700)]
turnip: Only link libdrm in the DRM case, not KGSL.
libvulkan's not a fan of opening my libdrm.so.2 from /vendor/lib64 or
/vendor/lib64/hw, but then we shouldn't need it, anyway.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6821>
Eric Anholt [Fri, 18 Sep 2020 21:36:36 +0000 (14:36 -0700)]
turnip: Don't link the WSI code if we don't have a WSI extension.
I don't like the TU_HAS_SURFACE duplication, but this is a step to having
a non-libdrm-dependent turnip on Android with KGSL (which doesn't have
drm for rendering).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6821>
Eric Anholt [Fri, 18 Sep 2020 20:16:23 +0000 (13:16 -0700)]
turnip: Use Mesa's libsync.h instead of libdrm's libsync.h.
Given that we already link to Android's libsync, use it instead of using a
build-time dependency on libdrm for the KGSL path. This also would help
for older kernel compat with KGSL.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6821>
Eric Anholt [Wed, 16 Sep 2020 02:18:12 +0000 (19:18 -0700)]
turnip: Drop a dead error checking path in device init.
The only result != SUCCESS setters above all jump across to the fail
label.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6821>
Eric Anholt [Tue, 22 Sep 2020 20:02:37 +0000 (13:02 -0700)]
android: Add pre-4.7 Android kernel compatibility to our libsync header.
The downstream Android kernels had a different API than was merged
upstream, and libsync on Android abstracts over that for us. Use their
sync_merge() and sync_wait(), at the cost of linking against libsync
(which Android.mk and meson both do).
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6821>
Eric Anholt [Tue, 22 Sep 2020 19:59:24 +0000 (12:59 -0700)]
util: Import a copy of drm's libsync.h
libsync.h is one of the common dependencies on libdrm from drivers that
otherwise don't want libdrm. Given that it's header-only code, just
import it to Mesa instead of forcing library dependencies on our users.
Copied from libdrm commit
a84caff71be9 (" intel: Add PCI ID support to RKL
platform")
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6821>
Eric Anholt [Thu, 24 Sep 2020 23:44:50 +0000 (16:44 -0700)]
ci/android: Switch build to using platform SDK version 26.
In SDK 25 (Android 7, 2016), the kernel only went up to around 4.4, so
the upstreamed sync API didn't exist, and libsync didn't expose
sync_merge().
In SDK 26 (Android 8, 2017), the kernel is sometimes bumped to 4.9 or
4.14, but libsync has a sync_merge() that operates on either the
downstream or the upstream API.
Since our android-targeting drivers in general use sync objects (requiring
a kernel newer than 7 got) and sync_merge() (suggesting interaction with
external fencing that started in Android 8), make CI build against an SDK
version with the sync API. I think really doing SDK 25 right at this
point would involve backporting mesa with some #ifdefs to not expose sync
APIs that wouldn't work.
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6821>
Eric Anholt [Tue, 22 Sep 2020 17:55:53 +0000 (10:55 -0700)]
android_stub: Update platform headers to include gralloc1.h.
This header is used in anv and radv, and soon turnip. Since the script
just checks out master, this also bumps the headers to upstream
02dfcc7c1562 ("Merge "Merge Android R"")
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6821>
Samuel Pitoiset [Thu, 15 Oct 2020 09:08:26 +0000 (11:08 +0200)]
radv: report latest extension spec versions
Found with the vulkan extension checker tool.
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/7158>
Duncan Hopkins [Thu, 15 Oct 2020 11:14:57 +0000 (12:14 +0100)]
meson: Add xcb-fixes to loader when using x11 and dri3. Fixes undefined symbol for xcb_xfixes_create_region in loader_dri3_helper.c
loader_dr3_helper.c uses xcb_xfixes_create_region() that requires dep_xcb_xfixes to link. This is dependent on with_platform_x11 and with_dri3.
But the source meson file does not set this up dependent on with_dri3.
The build was initialsed using platforms=x11 and gallium-drivers=zink,swrast.
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7164>
Ian Romanick [Mon, 12 Oct 2020 23:53:37 +0000 (16:53 -0700)]
mesa: Pass the correct caller string to _mesa_lookup_or_create_texture
An actual bug found by the 'unused parameter' warning. :D
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7167>
Ian Romanick [Mon, 12 Oct 2020 23:18:13 +0000 (16:18 -0700)]
mesa/st: Silence unused parameter warnings in st_context.c
src/mesa/state_tracker/st_context.c: In function ‘st_Enable’:
src/mesa/state_tracker/st_context.c:101:57: warning: unused parameter ‘state’ [-Wunused-parameter]
101 | st_Enable(struct gl_context *ctx, GLenum cap, GLboolean state)
| ~~~~~~~~~~^~~~~
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7167>
Ian Romanick [Mon, 12 Oct 2020 23:16:34 +0000 (16:16 -0700)]
i965: Silence unused parameter warnings
src/mesa/drivers/dri/i965/intel_screen.c: In function ‘brw_driconf_get_xml’:
src/mesa/drivers/dri/i965/intel_screen.c:103:33: warning: unused parameter ‘driver_name’ [-Wunused-parameter]
103 | brw_driconf_get_xml(const char *driver_name)
| ~~~~~~~~~~~~^~~~~~~~~~~
src/mesa/drivers/dri/i965/intel_screen.c: In function ‘intel_unmap_image’:
src/mesa/drivers/dri/i965/intel_screen.c:882:33: warning: unused parameter ‘context’ [-Wunused-parameter]
882 | intel_unmap_image(__DRIcontext *context, __DRIimage *image, void *map_info)
| ~~~~~~~~~~~~~~^~~~~~~
src/mesa/drivers/dri/i965/intel_screen.c:882:54: warning: unused parameter ‘image’ [-Wunused-parameter]
882 | intel_unmap_image(__DRIcontext *context, __DRIimage *image, void *map_info)
| ~~~~~~~~~~~~^~~~~
src/mesa/drivers/dri/i965/intel_screen.c: In function ‘intelReleaseBuffer’:
src/mesa/drivers/dri/i965/intel_screen.c:2904:33: warning: unused parameter ‘dri_screen’ [-Wunused-parameter]
2904 | intelReleaseBuffer(__DRIscreen *dri_screen, __DRIbuffer *buffer)
| ~~~~~~~~~~~~~^~~~~~~~~~
src/mesa/drivers/dri/i965/brw_context.c: In function ‘brw_set_background_context’:
src/mesa/drivers/dri/i965/brw_context.c:144:58: warning: unused parameter ‘queue_info’ [-Wunused-parameter]
144 | struct util_queue_monitoring *queue_info)
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7167>
Nanley Chery [Fri, 16 Oct 2020 17:49:48 +0000 (10:49 -0700)]
isl: Fix the aux-map encoding for D24_UNORM_X8
Bspec: 53911 now defines the encoding for this format.
Cc: mesa-stable
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7198>
Jason Ekstrand [Fri, 16 Oct 2020 03:56:09 +0000 (22:56 -0500)]
anv: Implement VariableDescriptorCount
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7180>
Jason Ekstrand [Fri, 16 Oct 2020 17:52:46 +0000 (12:52 -0500)]
anv: Add a descriptor_count to descriptor sets
This is useful for asserting in-bounds descriptor set access.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7180>
Jason Ekstrand [Fri, 16 Oct 2020 01:33:12 +0000 (20:33 -0500)]
anv: Bump the number of update-after-bind descriptors to 1M
It's a bit hard to exactly map our implementation to the limits
described by Vulkan. The bindless surface handle in the extended
message descriptors is 20 bits and it's an index into the table of
RENDER_SURFACE_STATE structs that starts at bindless surface base
address. This means that we can have at must 1M surface states
allocated at any given time. Since most image views take two
descriptors, this means we have a limit of about 500K image views.
However, since we allocate surface states at vkCreateImageView time,
this means our limit is actually something on the order of 500K image
views allocated at any time. The actual limit describe by Vulkan, on
the other hand, is a limit of how many you can have in a descriptor set.
Assuming anyone using 1M descriptors will be using the same image view
twice a bunch of times (or a bunch of null descriptors), we can safely
advertise a larger limit. 1M is what's required by D3D12, so let's
advertise that.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3335
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7180>
Erik Faye-Lund [Mon, 19 Oct 2020 09:17:30 +0000 (11:17 +0200)]
zink: reject resource-imports with modifiers
Zink doesn't support forwarding DRM modifiers yet, so whenever those are
used, we end up ignoring them. That's not going to do the right thing in
most cases, so let's reject them instead.
Since
d686835171c, the dri2 code tries to create a 0x0 surface without
any format when trying to import. This makes this go from
rendering-issues to asserting in debug builds, making things even worse.
Fixes:
d686835171c ("gallium/dri2: Support I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS import")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3654
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7214>
Mike Blumenkrantz [Sat, 17 Oct 2020 13:40:18 +0000 (09:40 -0400)]
doc/features: remove zink entries for GL 3.3 items
zink fully supports 3.3, so we don't need to be listed for these
Reviewed-by: Hoe Hao Cheng <haochengho12907@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7204>
James Park [Tue, 4 Aug 2020 17:58:11 +0000 (10:58 -0700)]
util,ac,aco,radv: Cross-platform memstream API
POSIX memstream is not available on Windows.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7143>
James Park [Wed, 14 Oct 2020 04:06:21 +0000 (21:06 -0700)]
util,radv,radv/winsys: Cross-platform rwlock API
POSIX rwlock is not available on Windows.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7144>
Ricardo Garcia [Thu, 15 Oct 2020 16:14:52 +0000 (18:14 +0200)]
anv: Ignore continue flag in primary cmd buffers
According to the Vulkan specification, the
VK_COMMAND_BUFFER_USAGE_RENDER_PASS_CONTINUE_BIT flag will be ignored if
included in a VkCommandBufferBeginInfo for a primary command buffer.
This also implies pBeginInfo->pInheritanceInfo should not be read even
if the flag is present, and makes it legal to include the flag knowing
it will be ignored.
Signed-off-by: Ricardo Garcia <rgarcia@igalia.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7128>
Samuel Pitoiset [Fri, 16 Oct 2020 11:46:08 +0000 (13:46 +0200)]
ac/nir: implement missing nir_op_pack_half_2x16_split
RADV now lowers lower_pack_half_2x16.
Fixes:
2f125908b35 ("radv,aco: lower_pack_half_2x16")
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/7187>