Mike Blumenkrantz [Mon, 17 Oct 2022 12:22:18 +0000 (08:22 -0400)]
zink: move spec constant emission to the types/consts block
spec constants are used as constants, so emitting them in the instructions
block can create mismatched ordering of the constant usage with other types
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19116>
Tapani Pälli [Wed, 19 Oct 2022 08:48:38 +0000 (11:48 +0300)]
intel/compiler: run nir_opt_idiv_const before nir_lower_idiv
Integer div lowering can potentially create a lot of code that is
not removed later on. Running const lowering pass first can be used
to eliminate that code.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19157>
Samuel Pitoiset [Tue, 18 Oct 2022 01:11:26 +0000 (01:11 +0000)]
radv: drop the ES vertex count requirement on GFX11
Ported from RadeonSI.
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/19155>
Samuel Pitoiset [Mon, 17 Oct 2022 09:03:50 +0000 (09:03 +0000)]
radv: do not use memory for waiting for cache flushes on GFX11
There is a different mechanism with an internal counter.
Ported from RadeonSI.
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/19155>
Samuel Pitoiset [Mon, 17 Oct 2022 08:20:30 +0000 (08:20 +0000)]
radv: use correct VGT_TESS_DISTRIBUTION settings on GFX11
Ported from RadeonSI.
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/19155>
Samuel Pitoiset [Mon, 17 Oct 2022 08:14:34 +0000 (08:14 +0000)]
radv: set SAMPLE_MASK_TRACKER_WATERMARK to 15 on GFX11
Ported from RadeonSI.
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/19155>
Tomeu Vizoso [Wed, 12 Oct 2022 06:46:17 +0000 (08:46 +0200)]
radeonsi,radv/ci: Increase coverage
12 more machines with Stoney Ridge have been added to Collabora's lab.
Use the new ones for testing more of RadeonSI, and repurpose the freed
ones for more RADV coverage.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19161>
Tomeu Vizoso [Wed, 12 Oct 2022 07:11:36 +0000 (09:11 +0200)]
ci: Run RadeonSI jobs on changes to src/amd/ci
Now that RadeonSI jobs are defined in src/amd/ci, make sure that those
jobs are run on changes to that directory.
Fixes:
a1d34d8c9577 ("ci/amd: move radeonsi in amd directory")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19161>
Qiang Yu [Sun, 12 Jun 2022 10:27:21 +0000 (18:27 +0800)]
ac/nir/ngg: use nir_load_provoking_vtx_in_prim_amd in ngg lower
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19166>
Qiang Yu [Sun, 12 Jun 2022 09:57:00 +0000 (17:57 +0800)]
nir,ac/llvm,radv: add nir_intrinsic_load_provoking_vtx_in_prim_amd
For radeonsi which load this from arg.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19166>
Martin Roukala (né Peres) [Tue, 18 Oct 2022 13:30:50 +0000 (16:30 +0300)]
radv/ci: run two jobs in parallel to reduce the execution time
This should bring the overall execution time at around 15 minutes \o/,
but it exposes more issues in the driver:
- [...].linear_interpolation.no_offset_8_samples (#7220)
- Timeouts on some dedicated_alloc.image.full.1048577.* tests
- Timeout on memory.pipeline_barrier.host_read_host_write
More performance improvements should come later, as the load average
remains in the [12, 16] range when running on my machine on a navi22
gets me a 31 load average and a lower execution time by a couple of
minutes.
References: #7220
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19111>
Martin Roukala (né Peres) [Mon, 17 Oct 2022 10:44:16 +0000 (13:44 +0300)]
radv/ci: bump the job count on navi21
We got shiny new hosts for our navi21 which sport twice as many cores
as the previous hosts. Let's try to keep them fed!
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19111>
Vinson Lee [Sun, 16 Oct 2022 18:36:38 +0000 (11:36 -0700)]
r600/sfn: Initialize TestInstrFromString member m_instr_factory.
Fix defect reported by Coverity Scan.
Uninitialized pointer field (UNINIT_CTOR)
uninit_member: Non-static class member m_instr_factory is not
initialized in this constructor nor in any functions that it calls.
Fixes:
e840645bb79 ("r600/sfn:explicitly initialize the memory pool")
Suggested-by: Gert Wollny <gert.wollny@collabora.com>
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19098>
Yonggang Luo [Fri, 14 Oct 2022 16:48:34 +0000 (00:48 +0800)]
util: Move the function implementation from u_thread.h to u_thread.c
Move the platform dependent functions from u_thread.h to u_thread.c,
so that avoid pull unneeded headers when using u_thread.h, including the big windows.h
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Sat, 3 Sep 2022 10:04:49 +0000 (18:04 +0800)]
util: Remove include of windows.h in rwlock.h
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Sat, 3 Sep 2022 09:21:58 +0000 (17:21 +0800)]
util: Remove include of windows.h in memstream.h
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Sat, 3 Sep 2022 10:22:53 +0000 (18:22 +0800)]
llvmpipe: #include <windows.h> in lp_rast.c
It's accessed GetExitCodeThread
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Sat, 3 Sep 2022 10:13:56 +0000 (18:13 +0800)]
util: #include <windows.h> in disk_cache_os.c
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Thu, 15 Sep 2022 14:53:03 +0000 (22:53 +0800)]
c11: Getting the macro guard of HAVE_PTHREAD in c11/threads.h to be consistence with util/u_uthread.h
The macro guard of #if defined(_WIN32) && !defined(__CYGWIN__) is comes from yohhoy's implementation
and that's not consistence with util/u_uthread.h, this caused it's hard to understand.
Now we change the behavior that's always rely on how meson detecting HAVE_PTHREAD.
So we always disable detecting of threads on Win32 as it's always included in the kernel32 library to
avoid detecting pthreads in mingw.
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Tue, 6 Sep 2022 09:39:28 +0000 (17:39 +0800)]
util: Remove the guard WIN32_WINNT >= 0x0600 in u_thread.h
We already require Windows Vista or later before this commit;
No need check WIN32_WINNT >= 0x0600
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Tue, 6 Sep 2022 09:32:29 +0000 (17:32 +0800)]
meson/mesa/wide: define WIN32_LEAN_AND_MEAN with pre_args
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Sat, 3 Sep 2022 10:40:26 +0000 (18:40 +0800)]
c11: Remove the code for Windows XP
We already require Windows Vista or later before this commit;
So the code for Windows XP can be removed.
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Sat, 3 Sep 2022 10:13:24 +0000 (18:13 +0800)]
util: #include "util/macros.h" in u_cpu_detect.h for ATTRIBUTE_CONST
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Yonggang Luo [Tue, 23 Aug 2022 19:32:57 +0000 (03:32 +0800)]
util: #include "c11/threads.h" when necessary in simple_mtx.h
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18425>
Luis Felipe Strano Moraes [Wed, 19 Oct 2022 02:16:14 +0000 (19:16 -0700)]
anv: adding parsetab.py to the .gitignore for grl
This file is autogenerated by python's ply package and is used during
compilation to cache tables built by it for quicker parsing.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19153>
Luis Felipe Strano Moraes [Wed, 19 Oct 2022 00:48:55 +0000 (17:48 -0700)]
anv: added proper handling for input argument in intel_clc
That was previously listed on the getopt_long struct but not actually
being used. This makes intel_clc argument processing easier as now
all of its arguments are handled with getopt and anything after the
special argument '--' is passed along to clang to form the final build
command.
Thanks to Dylan Baker for help with changes to the meson file.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19153>
Luis Felipe Strano Moraes [Wed, 19 Oct 2022 00:09:29 +0000 (17:09 -0700)]
anv: fixing typo on description of output flag for intel_clc
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19153>
Luis Felipe Strano Moraes [Wed, 19 Oct 2022 00:06:52 +0000 (17:06 -0700)]
anv: add missing separator to help for intel_clc
intel_clc relies on the special argument '--' for getopt to be given so
it knows when to start expecting purely input files or clang arguments.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19153>
Luis Felipe Strano Moraes [Wed, 19 Oct 2022 00:01:51 +0000 (17:01 -0700)]
anv: reword info flag in intel_clc's getopt to avoid clash
The info keyword was using the same short description that
was listed for input files on the struct for long_options.
Rewording it to 'v' and 'verbose' to be more in line with
expectations.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19153>
Yonggang Luo [Tue, 23 Aug 2022 20:00:04 +0000 (04:00 +0800)]
mesa: Use util_call_once_data instead of mtx_t init_once_lock
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 23 Aug 2022 20:07:20 +0000 (04:07 +0800)]
egl/dri2: Use util_call_once_data to replace usage of mtx_t glFlushMutex
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 23 Aug 2022 20:07:00 +0000 (04:07 +0800)]
egl/wgl: Use util_call_once_data to replace usage of mtx_t glFlushMutex
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 6 Sep 2022 15:02:41 +0000 (23:02 +0800)]
gallium/auxiliary: Remove the need of _MTX_INITIALIZER_NP by using simple_mtx_t/SIMPLE_MTX_INITIALIZER
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 6 Sep 2022 15:00:34 +0000 (23:00 +0800)]
mapi: Remove the need of _MTX_INITIALIZER_NP by using simple_mtx_t/SIMPLE_MTX_INITIALIZER in stub.c
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Mon, 17 Oct 2022 18:36:42 +0000 (02:36 +0800)]
mapi: Use util_call_once to init exec_mem and mutex instead _MTX_INITIALIZER_NP
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 6 Sep 2022 14:15:16 +0000 (22:15 +0800)]
util: Remove the need of _MTX_INITIALIZER_NP by using simple_mtx_t/SIMPLE_M in u_debug_refcnt.c
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 6 Sep 2022 14:12:45 +0000 (22:12 +0800)]
util: Remove the need of _MTX_INITIALIZER_NP by using simple_mtx_t/SIMPLE_MTX_INITIALIZER in u_debug_memory.c
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 6 Sep 2022 14:09:29 +0000 (22:09 +0800)]
util: Remove the need of _MTX_INITIALIZER_NP by using simple_mtx_t/SIMPLE_MTX_INITIALIZER in u_debug_stack_android.cpp
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 6 Sep 2022 14:09:05 +0000 (22:09 +0800)]
util: Remove the need of _MTX_INITIALIZER_NP by using simple_mtx_t/SIMPLE_MTX_INITIALIZER in u_debug_symbol.c
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 6 Sep 2022 13:55:23 +0000 (21:55 +0800)]
util: Call function mtx_init on exit_mutex to replace the usage of _MTX_INITIALIZER_NP in u_queue.c
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Tue, 23 Aug 2022 19:08:48 +0000 (03:08 +0800)]
util: Remove the need of _MTX_INITIALIZER_NP by using simple_mtx_t/SIMPLE_MTX_INITIALIZER in u_debug_stack.c
This removed the thread-unsafe function initialize_backtrace_mutex.
And remove the _MTX_INITIALIZER_NP that's doesn't belong to c11
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Yonggang Luo [Thu, 22 Sep 2022 15:59:46 +0000 (23:59 +0800)]
meson: add dep_valgrind dependency for 'dxil_compiler'
The error when include simple_mtx.h in glsl_types.h:
../src/microsoft/compiler/dxil_function.c
In file included from ../src/compiler/glsl_types.h:37,
from ../src/compiler/nir_types.h:36,
from ../src/compiler/nir/nir.h:44,
from ../src/microsoft/compiler/dxil_signature.h:28,
from ../src/microsoft/compiler/dxil_module.h:37,
from ../src/microsoft/compiler/dxil_function.h:48,
from ../src/microsoft/compiler/dxil_function.c:24:
../src/util/simple_mtx.h:37:12: fatal error: valgrind.h: No such file or directory
37 | # include <valgrind.h>
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19154>
Jesse Natalie [Mon, 29 Aug 2022 19:17:37 +0000 (12:17 -0700)]
d3d12: Release sharing contract in flush_frontbuffer
Just a leak I happened to notice.
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18305>
Jesse Natalie [Sun, 28 Aug 2022 13:40:23 +0000 (06:40 -0700)]
d3d12: Use ID3D12DeviceConfiguration for root signature serialization
If we were able to create a device factory from a DLL-local D3D12Core,
we want to use that one to do root signature serialization, instead of
going to the globally-configured D3D12Core.
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18305>
Jesse Natalie [Sun, 28 Aug 2022 13:34:31 +0000 (06:34 -0700)]
d3d12: Attempt to use the Agility SDK
The ideal would be to use a D3D12Core.dll next to the GL DLL, via
ID3D12SDKConfiguration1, but this requires a new D3D12.dll to be
present on the machine, which only very new OSes will currently have.
For older OSes, we can at least try to use ID3D12SDKConfiguration to
set a global Agility SDK override, but that only works in developer
mode and requires the DLL to be at a relative path from the app EXE,
so only try that if we have environment variables telling us to do so.
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18305>
Jesse Natalie [Sun, 28 Aug 2022 13:32:36 +0000 (06:32 -0700)]
d3d12: Support device factories in addition to global device creation
As explained in a comment, a device factory allows modification of state
like debug layer enablement without touching global state / triggering
device removal on a singleton device, and can even create devices that are
not singletons if the driver supports that.
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18305>
Jesse Natalie [Mon, 29 Aug 2022 19:33:25 +0000 (12:33 -0700)]
d3d12: Store the d3d12 module reference on the screen
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18305>
Jesse Natalie [Mon, 29 Aug 2022 19:32:48 +0000 (12:32 -0700)]
d3d12: Move some things from screen late-init to early-init
The GLSL singleton ref was too late, and we could end up with an un-paired
dec-ref if screen init failed. Also the transfer slab init was in the wrong
init function, so we'd re-init that struct if the screen was re-initialized.
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18305>
Jesse Natalie [Wed, 31 Aug 2022 19:32:33 +0000 (12:32 -0700)]
deps: Update DirectX-Headers req to 1.606.4
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18305>
Iván Briano [Wed, 21 Sep 2022 19:03:28 +0000 (12:03 -0700)]
anv: enable rayTraversalPrimitiveCulling feature
It is mandatory if VK_KHR_ray_query is supported.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19152>
Iván Briano [Mon, 26 Sep 2022 20:26:36 +0000 (13:26 -0700)]
anv: support VK_PIPELINE_CREATE_RAY_TRACING_SKIP_*
VK_PIPELINE_CREATE_RAY_TRACING_SKIP_AABBS_BIT_KHR and
VK_PIPELINE_CREATE_RAY_TRACING_SKIP_TRIANGLES_BIT_KHR, when specified,
make TraceRay behave as if the corresponding shader flags were set, but
without affecting the value of IncomingRayFlags in shaders.
v2 (Lionel): Improve comments
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19152>
Samuel Pitoiset [Tue, 18 Oct 2022 07:04:19 +0000 (09:04 +0200)]
ac/llvm: add missing nir_intrinsic_bindless_image_atomic_{fmin,fmax}
Fix various atomic operations.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19134>
LingMan [Wed, 5 Oct 2022 21:59:14 +0000 (23:59 +0200)]
rusticl/api: Interpret `lengths` as a slice of Option<NonZeroUsize>
So far `lengths` as been interpreted as a slice of usize. However, 0 is specified as a sentinel
value signalling that the corresponding string is nul terminated. Since checking for sentinel
values is frequently forgotten, Option types should be preferred if possible.
Option<NonZeroUsize> is layout compatible with usize. The None variant is then represented as 0,
which is exaclty what we need here.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18978>
LingMan [Wed, 5 Oct 2022 21:36:53 +0000 (23:36 +0200)]
rusticl/api: Don't check the program source for nul bytes needlessly
At this point we already know that the Vec we've constructed cannot contain a nul byte. Avoid the
needless scan.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18978>
LingMan [Wed, 5 Oct 2022 21:07:08 +0000 (23:07 +0200)]
rusticl/api: Shrink unsafe block
There's only two unsafe operations left here, so shrink the unsafe block to make them obvious.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18978>
LingMan [Fri, 7 Oct 2022 17:56:34 +0000 (19:56 +0200)]
rusticl/api: Factor pushing to the Vec out of the if
Both branches do the same here.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18978>
LingMan [Wed, 5 Oct 2022 20:00:56 +0000 (22:00 +0200)]
rusticl/api: Fix creating a program if a nul byte is within the specified string length
The code assumed that if the length of a string was specified and greater than zero, the string
would not contain a nul byte.
Legal or not, there are apparently applications which violate that assumption. Since the spec
doesn't say anything about this case, take the likely most compatible route and treat a nul byte as
terminating the string early.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7408
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18978>
LingMan [Wed, 5 Oct 2022 20:49:44 +0000 (22:49 +0200)]
rusticl/api: Drop UTF-8 conversion of input source
The spec doesn't define any charset for the source code. While the vast majority of inputs are
likely to be ASCII (which a subset of UTF-8), it is better not to make assumptions.
As a nice side effect this should be a minor speedup.
`CString`s can't currently be pushed to, so use a `Vec` as intermediate.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18978>
LingMan [Wed, 5 Oct 2022 17:43:22 +0000 (19:43 +0200)]
rusticl/api: Use iterators in create_program_with_source
This replaces several instances of unsafe pointer arithmetic and dereferencing with a single unsafe
creation of a slice, which we then use normal iterators on.
The spec mandates that a null pointer is to be interpreted as if a slice filled with zeros had been
given. That case is represented by an infinite iterator returning only zero.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18978>
Pavel Ondračka [Wed, 19 Oct 2022 08:23:11 +0000 (10:23 +0200)]
r300: fix generation of invalid swizzles in regalloc
rc_adjust_channels is inteded for moving the swizzles to a new channels
when rewriting the writemask of an instruction. However for readers one
needs to keep the swizzles in the old channels but rather convert to the
new values, so use the proper helper rc_rewrite_swizzle.
With the new swizzle fixed, we should properly detect that it would be
invalid and thus we can select the proper register class to prevent the
writemask rewrite in the regalloc.
Documentation was added to rc_adjust_channels to make it more clear what
it actually does.
Fixes a bunch of dEQP tests.
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7521
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19158>
Pedro J. Estébanez [Wed, 19 Oct 2022 16:33:20 +0000 (18:33 +0200)]
nir2dxil: Use native helper lane intrinsic on SM >= 6.6
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19170>
Bas Nieuwenhuizen [Tue, 18 Oct 2022 21:56:31 +0000 (23:56 +0200)]
aco: Pre-split result of bvh64_intersect_ray_amd.
Avoids later moves with extractions from the vector.
Reduces VALU operation in the raytrace loop by ~6%, increasing
the RT performance in Q2RTX on a 6800 XT by about ~1.3%.
Suggested by Georg.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19148>
Dylan Baker [Wed, 12 Oct 2022 22:25:02 +0000 (15:25 -0700)]
util/indicies: convert u_unfilled_gen.py to write a file
Which avoids meson needing to wrap the generator to capture the output,
and makes it faster
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19058>
Dylan Baker [Wed, 12 Oct 2022 22:13:19 +0000 (15:13 -0700)]
util/indicies: use itertools.product in u_unfilled_gen.py
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19058>
Dylan Baker [Wed, 12 Oct 2022 21:54:25 +0000 (14:54 -0700)]
util/indicies: simplify some of u_indicies_gen.py
v2:
- Give variable more descriptive name to avoid shadowing
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
and avoid shadowing
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19058>
Dylan Baker [Wed, 12 Oct 2022 21:32:01 +0000 (14:32 -0700)]
util/indicies: write a file with u_indicies_gen.py
This avoids meson creating a wrapper to redirect stdout, and makes the
generator faster
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19058>
Dylan Baker [Wed, 12 Oct 2022 20:46:07 +0000 (13:46 -0700)]
util/indicies: use itertools.product instead of deeply nested loops
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19058>
Dylan Baker [Tue, 11 Oct 2022 21:08:17 +0000 (14:08 -0700)]
util/indicies: move common static functions to private header
less duplication is good
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19058>
Dylan Baker [Wed, 19 Oct 2022 17:11:43 +0000 (10:11 -0700)]
docs: Extend calendar entries for 22.2 by 2 releases.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19172>
Dylan Baker [Wed, 19 Oct 2022 17:10:41 +0000 (10:10 -0700)]
docs: update calendar and link releases notes for 22.2.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19172>
Dylan Baker [Wed, 19 Oct 2022 17:06:27 +0000 (10:06 -0700)]
docs: Add sha256 sum for 22.2.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19172>
Dylan Baker [Wed, 19 Oct 2022 16:50:22 +0000 (09:50 -0700)]
docs: add release notes for 22.2.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19172>
Ruijing Dong [Tue, 18 Oct 2022 20:33:42 +0000 (16:33 -0400)]
frontends/va: fixed an av1 decoding image corruption issue.
problem:
When playing back some clips with loop restoration parameters
enabled, the display image could be corrupted.
solution:
correct loop restoration unit size logic in vaapi interface.
CC: 22.2 <mesa-stable>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Ruijing Dong <ruijing.dong@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19146>
Mike Blumenkrantz [Wed, 19 Oct 2022 11:15:05 +0000 (07:15 -0400)]
zink: add ci flake
fixes #7442
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19160>
Mark Collins [Mon, 17 Oct 2022 08:57:56 +0000 (08:57 +0000)]
tu: Clean up variable usage in `tu6_draw_common`
Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19107>
Mark Collins [Mon, 17 Oct 2022 08:47:48 +0000 (08:47 +0000)]
tu: Only write `A6XX_PC_PRIMITIVE_CNTL_0` if changed
Increases the score in the `draw` test in `vkoverhead` to 71809
from 67170 on a HDK 888.
Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19107>
Alyssa Rosenzweig [Fri, 14 Oct 2022 18:01:48 +0000 (14:01 -0400)]
gallium: Default to PIPE_CAP_MIXED_COLORBUFFER_FORMATS
There are three tiers of drivers:
* Drivers that support MRT and support mixed colorbuffer formats. All modern
hardware fits this as it becomes a spec requirement.
* Drivers that do not support MRT. Then this CAP is a no-op, so we might as well
set it by default even here (this commit trivially enables the CAP for lima,
vc4, etanviv).
* Drivers that support MRT but do not support mixed colorbuffer formats! Very
little hardware fits this category as it doesn't suffice for MRT in most APIs.
Unfortunately we have a few drivers that are in this category, preventing us
from bulldozing the CAP altogether.
Given that the CAP only exists for a few legacy drivers, default it to being
enabled to avoid new drivers falling into the trap of forgetting to enable it.
Failing to set this CAP causes failures in
dEQP-GLES3.functional.fbo.completeness.*
Drivers which still do not set this CAP: nv30, r300 (older than r500), virgl in
some cases. r300/r400 is due to a hardware requirement as Emma points out.
v2: Advertise the cap on lima like the commit message claims (delete the special
case).
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19079>
Vinson Lee [Sat, 15 Oct 2022 06:04:28 +0000 (23:04 -0700)]
zink: Remove duplicate variable zm.
Fix defect reported by Coverity Scan.
Evaluation order violation (EVALUATION_ORDER)
write_write_typo: In
zm = zm = create_shader_module_for_stage_optimal(ctx, screen, prog->shaders[i], prog, i, state),
zm is written twice with the same value.
Fixes:
325c703624a ("zink: add 'optimal_keys' handling for shader keys")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19086>
Karmjit Mahil [Tue, 13 Sep 2022 16:05:02 +0000 (17:05 +0100)]
pvr: Remove outdated comments.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19110>
Karmjit Mahil [Tue, 31 May 2022 14:53:35 +0000 (15:53 +0100)]
pvr: Handle pipeline barrier vk_sync.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19110>
Karmjit Mahil [Fri, 14 Oct 2022 12:25:46 +0000 (13:25 +0100)]
pvr: Fix overflow before widen warning for rgn headers size calculation.
This fixes CID 1515968.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19136>
Karmjit Mahil [Fri, 14 Oct 2022 12:09:41 +0000 (13:09 +0100)]
pvr: Fix possible seg fault on csb copy.
When copying the secondary command buffer's deferred control stream
to the main stream we have to first allocate space in the main
stream. In case the allocation failed we were attempting to
memcpy() to a NULL destination causing a NULL dereference.
This fixes CID 1515977.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19136>
Rob Clark [Sun, 16 Oct 2022 20:01:38 +0000 (13:01 -0700)]
freedreno/a6xx: Drop max_scissor tracking
We don't actually really use this on a6xx, since SQE can early-exit IB2
when there are no more remaining primitives in the bin.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Rob Clark [Sun, 16 Oct 2022 15:35:08 +0000 (08:35 -0700)]
freedreno/a6xx: Update caps
We should be doing all the 64b lowering.. I think that should be enough
to get us at least glsl400.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Rob Clark [Tue, 18 Oct 2022 18:51:49 +0000 (11:51 -0700)]
freedreno/a6xx: Fix primitives-generated query
RBBM_PRIMCTR_7 is pre-clipped, whereas RBBM_PRIMCTR_8 is after clipping.
I believe we want pre-clipping, and this is what tu does.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Rob Clark [Tue, 18 Oct 2022 17:02:10 +0000 (10:02 -0700)]
freedreno/a6xx: Fix MAX_GEOMETRY_OUTPUT_VERTICES cap
Limited by the size of PC_PRIMITIVE_CNTL_5.GS_VERTICES_OUT
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Rob Clark [Sun, 16 Oct 2022 15:34:42 +0000 (08:34 -0700)]
freedreno/ir3: Lower all the 64b
Just need to enable some existing lowering.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Rob Clark [Mon, 17 Oct 2022 17:25:10 +0000 (10:25 -0700)]
freedreno/ir3: Fix clipvertex with GS+tess
If we have both GS and tess, GS is the stage we should run lower_clip_vs
on.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Rob Clark [Sun, 16 Oct 2022 19:45:09 +0000 (12:45 -0700)]
freedreno/a6xx: Remove unused param
Minor cleanup, make it more clear that we aren't actually emitting in
IB2.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Rob Clark [Sun, 16 Oct 2022 16:40:59 +0000 (09:40 -0700)]
mesa/st: ARB_vertex_attrib_64bit depend on glsl>=410
It is possible to expose ARB_gpu_shader_fp64 without supporting
ARB_vertex_attrib_64bit. The supported GLSL version should tell
us if 64b vertex attribs are also supported or not.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Rob Clark [Sun, 16 Oct 2022 16:27:09 +0000 (09:27 -0700)]
docs: Update freedreno features
Another manual re-sync with reality.
There are a few extension, notably GL_ARB_texture_compression_bptc,
which are exposed but mesamatrix doesn't show as supported. I can't
really figure out why.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19100>
Marek Olšák [Tue, 18 Oct 2022 20:49:53 +0000 (16:49 -0400)]
driconf: add a workaround for Cossacks 3
Wine issue: https://bugs.winehq.org/show_bug.cgi?id=44560
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19147>
Marek Olšák [Tue, 18 Oct 2022 20:49:06 +0000 (16:49 -0400)]
driconf: add a workaround for a cursor issue with kwin_wayland
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7351
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19147>
Erik Faye-Lund [Wed, 19 Oct 2022 08:20:46 +0000 (10:20 +0200)]
docs: correct spelling of "freedreno"
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19139>
Erik Faye-Lund [Tue, 18 Oct 2022 15:25:02 +0000 (17:25 +0200)]
docs: correct spelling of "traveling"
While "travelling" isn't incorrect, we generally prefer US english in the
docs, where "traveling" is the common form.
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19139>
Erik Faye-Lund [Tue, 18 Oct 2022 15:24:04 +0000 (17:24 +0200)]
docs: correct spelling of "correctly"
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19139>
Erik Faye-Lund [Tue, 18 Oct 2022 14:52:02 +0000 (16:52 +0200)]
docs: ie. -> i.e.
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19139>
Erik Faye-Lund [Tue, 18 Oct 2022 14:41:47 +0000 (16:41 +0200)]
docs: fractionary -> fractional
While "fractionary" isn't incorrect language, it's much less common than
the more modern "fractional" term.
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19139>
Erik Faye-Lund [Tue, 18 Oct 2022 14:58:38 +0000 (16:58 +0200)]
docs: correct spelling of "rasterizer"
We use rasterizer all over the docs, so let's be consistent.
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19139>
Erik Faye-Lund [Tue, 18 Oct 2022 14:56:56 +0000 (16:56 +0200)]
docs: correct spelling of "explicitly"
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19139>