Commit graph

161795 commits

Author SHA1 Message Date
Mike Blumenkrantz
b6f6224810 zink: clamp feedback loop layout to GENERAL if the ext isn't supported
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19140>
2022-10-19 06:03:40 +00:00
Mike Blumenkrantz
bf3edad94c zink: enable tc cpu storage on some buffers
matching freedreno usage

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19142>
2022-10-19 05:43:40 +00:00
Marek Olšák
211c2aebd2 cso: make memcmp and hash computation use a literal key size for blend state
This results in faster lookups because memcmp and the hash computation
can be unrolled.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19129>
2022-10-19 04:56:55 +00:00
Marek Olšák
f7b513c9b1 cso: inline more functions because some parameters like key_size are literals
One function is moved to the header file, the other two functions are inlined
manually.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19129>
2022-10-19 04:56:55 +00:00
Marek Olšák
2f0cd4029c cso: don't destroy CSOs that are saved
I think this can't happen in practice, but better safe than sorry.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19129>
2022-10-19 04:56:55 +00:00
Marek Olšák
94240f561c cso: fix broken optimization for sampler state lookups
Since the key size wasn't a constant expression, all the function inlining
didn't do much. This makes it a constant expression.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Fixes: c4e18cd4dd - mesa/st: add PIPE_QUIRK_TEXTURE_BORDER_COLOR_SWIZZLE_FREEDRENO
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19129>
2022-10-19 04:56:55 +00:00
Marek Olšák
224735abaf cso: constify some parameters to remove typecasts
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19129>
2022-10-19 04:56:55 +00:00
Marek Olšák
97ec1de8f3 cso: start without u_vbuf by default if it's not always used
no functional change, but it's convenient for future work

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19129>
2022-10-19 04:56:55 +00:00
Marek Olšák
8e024e2870 glthread: don't sync for glIsEnabled(GL_BLEND, GL_LIGHTING, GL_POLYGON_STIPPLE)
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:06 +00:00
Marek Olšák
513ac52e47 glthread remove the unused *last pointer from unmarshal functions
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
6e6245f189 glthread: merge and collapse glBindBuffer calls that unbind and then bind
This is a small optimization for viewperf. See the comment in the code.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
9b6e2783eb gl_marshal.py: C style fixups
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
5db7dc599a gallium/u_threaded: don't call simplify_draw_info redundantly
It's always called for the first and second draw in a row, and if they
can't be merged, the second draw becomes the first draw in the next call,
and simplify_draw_info is called again on that. Thus, it's called twice
on any draw that isn't merged. To prevent that, call it when we add
the draws.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
3cb96b18f6 glthread: rewrite CallList merging and do it in the app thread
This looks simpler and hopefully faster, but it may just move the overhead
to the app thread.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
6f8cbbfce2 glthread: demystify Draw function names
Since we had the freedom to write our own marshal and unmarshal Draw
functions, we turned it into a mess by doing whatever we want in those
functions. For example, DrawElementsInstancedBaseVertex actually
implemented DrawArraysInstancedBaseInstanceBaseInstance.

Add 4 internal GL functions that pass user buffers through marshal
/unmarshal functions, and clean up the other names to match their
behavior.

The new functions don't need any parameters in their definitions because
we don't call them directly.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
5679ef99b8 glapi: remove EXT and ARB suffixes from Draw functions
This swaps the function names with aliased names that don't have those
suffixes.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
58f738bf5e glthread: add more DrawArrays/Elements variants with fewer fields
The idea is to save 8 bytes per call by removing 2 fields that are not
specified by the user such as instance_count=1, basevertex=0, and
baseinstance=0.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
83ce647c15 glthread: use a constant expression instead of cmd_size in custom functions
cmd_size is constant in these cases

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
b09a88fb72 gl_marshal.py: simplify print_sync_call and rename to print_call
no spaces around = is a python convention for implicit arguments only

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
05dddaeb80 gl_marshal.py: inline functions for readability
All functions have only one use. Jumping constantly between functions
in the file is confusing.

No spaces around = is a python convention for implicit arguments only.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
146836f4ba gl_marshal.py: remove the -O1 hack and manual SET_* inlining
The compile times aren't so bad anymore now that the initialization is
split among all 8 files.

Also add one assertion.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
67c7260571 gl_marshal.py: rework how the marshal dispatch table is initialized
Instead of setting all function pointers in marshal_generated0.c,
set the function pointers in the file that contains the functions,
and remove all the forward declarations of marshal functions
in marshal_generated.h.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
5e66e26998 gl_marshal.py: move the unmarshal table into a separately generated file
It's unrelated to the rest of the script and it's in the way of bigger
changes.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Marek Olšák
5c65a3a98f gl_marshal.py: remove/simplify parameters
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18199>
2022-10-19 04:23:05 +00:00
Timothy Arceri
b4b2fd0bb4 glsl: move lower instructions logic inside that pass
There is now only a single called of this pass so tidy things up
and move all this logic inside the pass.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19112>
2022-10-19 03:52:21 +00:00
Timothy Arceri
e5102a406f glsl: always do {CARRY,BORROW}_TO_ARITH lowering
The only caller always sets these so here we just remove the
option to disable it.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19112>
2022-10-19 03:52:21 +00:00
Timothy Arceri
9f14c5dae2 glsl: drop sub to add neg lowering in GLSL IR
NIR opt algebraic does this for us so no need to have it
implemented here also.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19112>
2022-10-19 03:52:21 +00:00
Timothy Arceri
a31c547206 glsl: move rule inside lower_packing_builtins()
We only have a single user of this pass so lets tidy things up and
move all the rules in the pass itself.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19112>
2022-10-19 03:52:20 +00:00
Mark Janes
141703f906 anv: work around improper buffer usage in hitman3
Avoid memory errors by returning an error immediately instead of
attempting to allocate the buffer view with an unsupported format.

Fixes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6998

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17972>
2022-10-19 03:24:36 +00:00
Lionel Landwerlin
a31e8a7568 anv: add a layer for hitman3
This title is trying to create a buffer view with usage not supported by Anv.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17972>
2022-10-19 03:24:36 +00:00
Mark Janes
74d0be27ae anv: compile anv_acceleration_structure.c
This file was unintentionally excluded from meson build support.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17972>
2022-10-19 03:24:36 +00:00
Rhys Perry
36703a60bf aco: add ACO_DEBUG=force-waitdeps
GFX11 has a lot of complicated data dependency hazards.

For debugging GFX10+ data dependency hazards. This creates an excessive
amount of s_waitcnt_depctr.

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/18273>
2022-10-19 02:46:03 +00:00
Rhys Perry
6f368431fd aco/gfx11: workaround VALUMaskWriteHazard
fossil-db (gfx1100):
Totals from 62812 (46.52% of 135032) affected shaders:
Instrs: 43971580 -> 44069887 (+0.22%)
CodeSize: 233473420 -> 233866648 (+0.17%)
Latency: 463487489 -> 463520688 (+0.01%); split: -0.00%, +0.01%
InvThroughput: 86505748 -> 86509679 (+0.00%); split: -0.00%, +0.00%

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/18273>
2022-10-19 02:46:03 +00:00
Rhys Perry
5806f0246f aco/gfx11: workaround VALUPartialForwardingHazard
fossil-db (gfx1100):
Totals from 18121 (13.42% of 135032) affected shaders:
Instrs: 28272367 -> 28323468 (+0.18%)
CodeSize: 152581916 -> 152786320 (+0.13%)
Latency: 366477785 -> 366482318 (+0.00%); split: -0.00%, +0.00%
InvThroughput: 74475188 -> 74475758 (+0.00%); split: -0.00%, +0.00%

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/18273>
2022-10-19 02:46:03 +00:00
Rhys Perry
b0b48b7687 aco/gfx11: workaround VALUTransUseHazard
fossil-db (gfx1100):
Totals from 116990 (86.64% of 135032) affected shaders:
Instrs: 67942325 -> 69493991 (+2.28%)
CodeSize: 366448984 -> 372655648 (+1.69%)
Latency: 673236871 -> 673269808 (+0.00%); split: -0.00%, +0.00%
InvThroughput: 128266905 -> 128270175 (+0.00%); split: -0.00%, +0.00%

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/18273>
2022-10-19 02:46:03 +00:00
Rhys Perry
98ee3e1468 aco/gfx11: workaround LdsDirectVMEMHazard
fossil-db (gfx1100):
Totals from 27217 (20.16% of 135032) affected shaders:
Instrs: 18010853 -> 18047277 (+0.20%)
CodeSize: 99369568 -> 99515264 (+0.15%)
Latency: 207454040 -> 207464932 (+0.01%); split: -0.00%, +0.01%
InvThroughput: 39810158 -> 39810628 (+0.00%); split: -0.00%, +0.00%

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/18273>
2022-10-19 02:46:03 +00:00
Rhys Perry
296b4d95a3 aco/gfx11: workaround LdsDirectVALUHazard
fossil-db (gfx1100):
Totals from 57858 (42.85% of 135032) affected shaders:

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/18273>
2022-10-19 02:46:03 +00:00
Rhys Perry
1a9840b341 aco/gfx11: workaround VcmpxPermlaneHazard
Same as GFX10, but in a separate pass because it's the only hazard that's
shared.

No fossil-db changes.

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/18273>
2022-10-19 02:46:03 +00:00
Rhys Perry
484df5b30e aco: add search_backwards helper
This will be useful for VALUPartialForwardingHazard.

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/18273>
2022-10-19 02:46:03 +00:00
Pedro J. Estébanez
6558a9e1fb nir2dxil: Lower texture projections
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19115>
2022-10-19 02:30:16 +00:00
Erik Faye-Lund
95fe418def mesa/st: remove translate_mode
This no longer does anything useful, so let's just drop it.

Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19034>
2022-10-19 01:33:23 +00:00
Erik Faye-Lund
a125f2d8c7 mesa/st: move static asserts out of translate_prim
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19034>
2022-10-19 01:33:23 +00:00
Sil Vilerino
2d504bc5a0 CI: Add gallium-va and video-codecs in windows-vs2019 and debian-mingw32-x86_64
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19063>
2022-10-19 00:45:48 +00:00
Sil Vilerino
e08df85923 d3d12: resource_from_handle to validate importing resource ptr from same d3d12 device
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19063>
2022-10-19 00:45:48 +00:00
Sil Vilerino
b557ceb75b frontends/va: Add windows VA frontend support via vl_winsys_win32 and libva-win32
Acked-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19063>
2022-10-19 00:45:48 +00:00
Sil Vilerino
05f0d27601 gallium/vl: Add vl_winsys_win32 support
Acked-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19063>
2022-10-19 00:45:48 +00:00
Iván Briano
cde0778d3d hasvk: use DX rules for point rasterization
It seems that's what the reference renderer in the CTS expects for
Vulkan. This mostly matters if the edges of a point primitive fall
exactly on a pixel sampling point.

Fixes some upcoming tests under
dEQP-VK.pipeline.monolithic.depth.format.*.point_list*

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19026>
2022-10-19 00:20:48 +00:00
Iván Briano
63b2f735b1 anv: use DX rules for point rasterization
It seems that's what the reference renderer in the CTS expects for
Vulkan. This mostly matters if the edges of a point primitive fall
exactly on a pixel sampling point.

Fixes some upcoming tests under
dEQP-VK.pipeline.monolithic.depth.format.*.point_list*

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19026>
2022-10-19 00:20:48 +00:00
Iván Briano
d042669747 anv: compile_upload_rt_shader expects a valid pointer
Fixes crashes on almost every CTS test that uses raytracing pipelines.

Fixes: ff91c5ca42 ("anv: add analysis for push descriptor uses and store it in shader cache")

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19130>
2022-10-18 23:55:48 +00:00
Yonggang Luo
095dfc6caa util: Move the implementation of futex_wake and futex_wait from futex.h to futex.c
Doing this is to avoid include bloat on Windows that pulls windows.h,
for other unix platforms, it's also benefit include the used symbols only when use
futex_wake and futex_wait.
No functional change, there is no performance incur because function call is consume
much less time than syscall

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19120>
2022-10-18 23:20:25 +00:00