fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-05-20 04:48:07 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 4
200743 commits 66 branches 936 tags 981 MiB
Commit graph

16750 commits

Author SHA1 Message Date
Georg Lehmann
997ea2e273 aco: update is_dual_issue_capable for gfx11.5+ 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Daniel Schürmann <None>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33111>
 2025-01-27 11:59:45 +00:00
Rhys Perry
e18e293e6c aco: don't use divergence information for most ALU defs 
If one of the sources are divergent but all are SGPRs, then this can
cause issues (for example, imul with two SGPR sources and a VGPR definition).

No fossil-db changes (navi21)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12454
Backport-to: 24.3
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32997>
 2025-01-27 10:48:07 +00:00
Samuel Pitoiset
ce3a137892 radv: fix the number of drm modifier planes for DCC on GFX12 
It's always 1 plane because DCC isn't allocated from the userspace
driver.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33213>
 2025-01-27 08:44:48 +00:00
Samuel Pitoiset
cd31a61fda radv: fix an assertion about DCC and modifier on GFX12 
radv_image_has_dcc() should always returns FALSE.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33213>
 2025-01-27 08:44:48 +00:00
Samuel Pitoiset
48d199f3dc ac/gpu_info: add gfx12_supports_display_dcc 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33212>
 2025-01-27 08:08:15 +00:00
Marek Olšák
f7e3689fe1 ac/nir: lower sample_pos to load_sample_positions_amd when frag_coord is center 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
0a228dfef2 ac/nir: compute ddx/ddy for barycentric_at_offset at the beginning of shaders 
to make it work after terminate (discard).

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
f98613d47c aco: implement replacement of sample_mask_in with helper_invocation in PS prolog 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
a842f198d7 aco: simplify how broadcast_last_cbuf is implemented in PS epilog 
So PS epilogs only need a single bool flag that determines whether all
enabled color buffers should be written.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
5c4f737b84 aco: implement replacing frag_coord with pixel_coord in PS prolog 
This adds an option to replace frag_coord.xy with pixel_coord when sample
shading is disabled, which is most of the time. This reduces the number of
input VGPRs.

It's already implement in ac_nir_lower_ps_early for monolithic shaders.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
eddd063420 ac/nir: cosmetic stuff for ac_nir_lower_ps 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
247c0593eb ac/nir: eliminate sample_mask_in without MSAA in ac_nir_lower_ps_early 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
e57b52ff6c ac/nir: optimize frag_coord <-> pixel_coord in ac_nir_lower_ps_early 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
33134f9503 ac/nir: optimize barycentric_at_sample(sample_id) in ac_lower_ps_early 
Replace it with barycentric_sample.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:26 -05:00
Marek Olšák
43f6b2655e ac/nir: simplify force_*_center_interp options in ac_nir_lower_ps_early 
This only indicates whether MSAA is disabled.

Having a separate option for each sysval is better for the PS prolog,
but not for monolithic compilation.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
3bccfa72cc ac/nir: simplify force_*_sample_interp options in ac_nir_lower_ps_early 
The only thing we need here is whether sample shading is enabled
and how many samples.

Having a separate option for each sysval is better for the PS prolog,
but not for monolithic compilation.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
d7d4d56f5b ac,aco,radeonsi: replace SampleMaskIn with 1 << SampleID if full sample shading 
Since the sample mask is always 1 << sample_id with full sample shading,
just use that instead of loading sample_mask_in. Set it to 0 if it's
a helper invocation. This removes the sample mask input VGPR.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
1026402b7c ac/nir: clamp vertex color outputs in the right place 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
4f63b21df0 ac/nir: drop 16x EQAA support from ac_get_ps_iter_mask 
We don't support 16x EQAA anymore.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
51bbe2606d ac/nir: switch passes to use nir_shader_intrinsics_pass 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
d31b3513da ac/nir: handle FRAG_RESULT_COLOR with dual src blending in ac_nir_lower_ps_early 
If FRAG_RESULT_COLOR with dual_src_blend_kndex=1 is present,
FRAG_RESULT_COLOR really means FRAG_RESULT_DATA0.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
38d3fc7a6a ac/nir: return progress from ac_nir_lower_ps_late 
This changes the creation of barycentric coordinate variables to on-demand.
There is also some reordering of export code to return progress.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
231009be65 ac/nir: return progress from ac_nir_lower_ps_early 
This changes the creation of barycentric coordinate variables to on-demand.
Everything else is ready to return progress.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
234b416ffb ac/nir: lower fbfetch_output in ac_nir_lower_ps_early 
so that we can gather shader_info after this and all system values that
this adds will be gathered.

shader_info won't be gathered after si_nir_lower_abi, which is why we
have to lower fbfetch_output here.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
1570387aaa ac/nir: lower barycentric_at_offset/sample in ac_nir_lower_ps_early 
i.e. before future linking optimizations and shader_info gathering

They are lowered together because one depends on the other.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
b9b1f3a047 ac/nir: lower sample_pos in ac_nir_lower_ps_early 
i.e. before future linking optimizations and shader_info gathering

The ac_nir_lower_ps_early call is also added for non-monolithic shaders
because it's required now.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
580304350b ac/nir: optimize front_face in ac_nir_lower_ps_early 
i.e. before future linking optimizations and shader_info gathering

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Marek Olšák
0939fc6fa8 radv: don't call ac_nir_lower_ps_early 
The current version doesn't do anything, and later commits require significant
reordering of NIR passes to make it usable.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33024>
 2025-01-25 12:20:25 -05:00
Detlev Casanova
8f4efda257 ci/fluster: Add radeonsi-raven-vaapi-fluster jobs 
The CI job runs each supported decoder on 1 LAVA job:
- vp9
- h.264
- h.265

They each run the gstreamer VAAPI fluster decoder tests on a radeonsi
device on LAVA.

Signed-off-by: Detlev Casanova <detlev.casanova@collabora.com>
Co-authored-by: Sergi Blanch Torne <sergi@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26113>
 2025-01-25 08:48:04 +00:00
Samuel Pitoiset
f8e7c037e3 radv: add a new helper to set image BO metadata 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33193>
 2025-01-25 02:10:03 +00:00
Samuel Pitoiset
ee4a1021d1 radv: add support for BO metadata on GFX12 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33193>
 2025-01-25 02:10:02 +00:00
Saroj Kumar
433004dcff ac/surface: fix missing NULL check in gfx12_select_swizle_mode() 
Add null check for surf pointer.
Fixes segfault issue during start of gdm on gfx12.

Signed-off-by: Saroj Kumar <saroj.kumar@amd.com>

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33199>
 2025-01-24 21:51:50 +00:00
Rhys Perry
5cc977bee4 radv: set has_image_bvh_intersect_ray for null winsys 
This is needed for fossilize-replay.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Fixes: 14e3231b56 ("radv: add a flag to indicate ray tracing support")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33201>
 2025-01-24 18:08:12 +00:00
David Rosca
604fb68876 radeonsi: Report surface alignment for AV1 encode 
This is also useful for AV1 to allow applications to use container crop.

Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33083>
 2025-01-24 13:04:47 +00:00
Samuel Pitoiset
d41a3bb744 radv: program COMPUTE_DISPATCH_INTERLEAVE on GFX12 
To the same value as GFX11.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33185>
 2025-01-24 11:08:59 +00:00
Samuel Pitoiset
c039caeb53 ac/cmdbuf: program SPI_SHADER_GS_MESHLET_CTRL to 0 in the GFX12 preamble 
Otherwise, it's derivative group quads by default.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33185>
 2025-01-24 11:08:59 +00:00
Rhys Perry
0eb5f66660 nir/validate: validate ssa dominance by default 
This no longer modifies dominance metadata, so enable it by default.

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/32005>
 2025-01-23 23:35:44 +00:00
Konstantin Seurer
eecaf80554 radv/rmv: Use radv_rmv_log_resource_destroy more 
Gets rid of some duplicate code with the added benefit of removing a
memory leak warning when destroying the device because
radv_rmv_log_border_color_palette_destroy did not destroy the resource
ID.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33167>
 2025-01-23 23:07:49 +00:00
Samuel Pitoiset
4b741338ac radv: exclude layer when recomputing FS input bases 
This is always exported as a sysval.

Closes: mesa/mesa#12501
Fixes: dd00b3f5 ("radv: Implement FS layer ID input as a system value.")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33168>
 2025-01-23 13:21:03 +00:00
Valentine Burley
2d99b77f2e amd/ci: Run full radeonsi-raven-va job pre-merge 
The full job run takes approximately 8 minutes now, as the issues
previously mentioned have been resolved.

Signed-off-by: Valentine Burley <valentine.burley@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33155>
 2025-01-23 11:33:25 +00:00
Valentine Burley
0624cd9c51 amd/ci: Add lava-hp-x360-14a-cb0001xx-zork and use it for VA-API testing 
Move the existing radeonsi-raven-va job to this new device.

Signed-off-by: Valentine Burley <valentine.burley@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33155>
 2025-01-23 11:33:25 +00:00
Connor Abbott
47a85815b0 radv: Delete acceleration structure stubs 
These are now provided by common code.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33153>
 2025-01-23 05:16:58 +00:00
Daniel Schürmann
08560b8ff8 aco/lower_branches: stitch linear blocks if there is exactly one successor with one predecessor 
Totals from 12906 (16.26% of 79395) affected shaders: (Navi31)

Instrs: 22051521 -> 22049488 (-0.01%); split: -0.01%, +0.00%
CodeSize: 116591240 -> 116583920 (-0.01%)
Latency: 196625178 -> 196538410 (-0.04%); split: -0.04%, +0.00%
InvThroughput: 33943045 -> 33930615 (-0.04%); split: -0.04%, +0.00%
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32477>
 2025-01-23 00:11:06 +00:00
Daniel Schürmann
c90ae5f773 aco: delete aco_jump_threading.cpp 
This is now handled by lower_branches().

Totals from 47236 (59.49% of 79395) affected shaders: (Navi31)
Instrs: 29490400 -> 29490507 (+0.00%)
CodeSize: 152316812 -> 152317248 (+0.00%); split: -0.00%, +0.00%
Latency: 229665459 -> 229665106 (-0.00%); split: -0.00%, +0.00%
InvThroughput: 36870605 -> 36870504 (-0.00%); split: -0.00%, +0.00%
Copies: 1966751 -> 2233467 (+13.56%)
SALU: 3122941 -> 3123048 (+0.00%)

Note, that only about 20 shaders are actually affected.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32477>
 2025-01-23 00:11:06 +00:00
Daniel Schürmann
c677809f25 aco/lower_branches: allow for non-fallthrough loop exits in try_merge_break_with_continue() 
Totals from 211 (0.27% of 79395) affected shaders: (Navi31)

Instrs: 276961 -> 276545 (-0.15%)
CodeSize: 1404356 -> 1402248 (-0.15%)
Latency: 1344722 -> 1344887 (+0.01%); split: -0.00%, +0.01%
InvThroughput: 165624 -> 165622 (-0.00%); split: -0.00%, +0.00%
Branches: 6149 -> 5987 (-2.63%)
SALU: 25722 -> 25468 (-0.99%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32477>
 2025-01-23 00:11:06 +00:00
Daniel Schürmann
12656ea5f5 aco: move try_merge_break_with_continue() to lower_branches() 
Totals from 3 (0.00% of 79395) affected shaders: (Navi31)

Instrs: 12888 -> 12882 (-0.05%)
Latency: 83253 -> 83246 (-0.01%)
InvThroughput: 9251 -> 9249 (-0.02%)
Branches: 483 -> 480 (-0.62%)
SALU: 1329 -> 1326 (-0.23%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32477>
 2025-01-23 00:11:06 +00:00
Daniel Schürmann
13ad3db43f aco/lower_branches: implement try_remove_simple_block() in lower_branches() 
This is mostly the same as in jump_threading, but can handle
multiple predecessors.

Totals from 3523 (4.44% of 79395) affected shaders: (Navi31)

Instrs: 10244892 -> 10244753 (-0.00%); split: -0.00%, +0.00%
CodeSize: 54171500 -> 54168540 (-0.01%); split: -0.01%, +0.00%
Latency: 75070425 -> 75059570 (-0.01%); split: -0.02%, +0.00%
InvThroughput: 11606911 -> 11605767 (-0.01%); split: -0.01%, +0.00%
Branches: 331778 -> 331675 (-0.03%); split: -0.05%, +0.02%
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32477>
 2025-01-23 00:11:06 +00:00
Daniel Schürmann
2b5a893e29 aco/lower_branches: do eliminate_useless_exec_writes_in_block() during branch lowering. 
Totals from 728 (0.92% of 79395) affected shaders: (Navi31)

Instrs: 452926 -> 452161 (-0.17%)
CodeSize: 2255536 -> 2252504 (-0.13%)
Latency: 1683404 -> 1683470 (+0.00%); split: -0.01%, +0.01%
InvThroughput: 210887 -> 210888 (+0.00%); split: -0.00%, +0.00%
SALU: 77865 -> 77106 (-0.97%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32477>
 2025-01-23 00:11:06 +00:00
Daniel Schürmann
eecdb45d61 aco: consider s_cbranch_exec* instructions in needs_exec_mask() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32477>
 2025-01-23 00:11:06 +00:00
Daniel Schürmann
de1e38e214 aco/assembler: Find loop exits using the successor's loop nest depth 
Previously, we just used the next block after a loop that
has a back-edge. This assumes that loop-exit blocks can
only be removed when falling through to the next block,
when in fact it can also be a jump to somewhere else,
in future even to some block before the actual loop.

12 (0.02% of 79395) affected shaders.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32477>
 2025-01-23 00:11:06 +00:00