AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
204,297 Commits 1 Branch 0 Tags
b44dc98cde52c409fb7bc82d2972f8f5dfbdc099
Commit Graph

204297 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Samuel Pitoiset b44dc98cde radv/sdma: remove redundant check for compression when getting metadata 
It's already checked by the caller.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34517>
 2025-04-16 06:57:00 +00:00
Samuel Pitoiset d3d5d2fe86 radv/sdma: use SDMA5_DCC_xxx bitfields 
It's cleaner.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34517>
 2025-04-16 06:57:00 +00:00
Samuel Pitoiset f44342199a radv/sdma: simplify configuring the number of uncompressed DCC blocks 
SDMA doesn't support MSAA, so the value can be
V_028C78_MAX_BLOCK_SIZE_256B.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34517>
 2025-04-16 06:57:00 +00:00
Samuel Pitoiset 13db408e59 ac/perfcounter: add support for GFX12 
Sourced from PAL to add SPM support.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34524>
 2025-04-16 06:35:33 +00:00
Samuel Pitoiset c42d43e8eb radv: print more error messages during SPM initialization 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34524>
 2025-04-16 06:35:33 +00:00
Marek Olšák 177427877b radeonsi: use nir_opt_shrink_vectors 
It reduces VGPR usage, but the impact is almost none.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34432>
 2025-04-16 06:08:48 +00:00
Marek Olšák b7eff9cd87 radeonsi: always scalarize shared memory instructions 
to get ds_load_2addr/ds_store_2addr more often and to prevent code size
regressions from nir_opt_shrink_vectors.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34432>
 2025-04-16 06:08:48 +00:00
Marek Olšák 78cacfd9ce ac/surface: select 3D tile mode without overallocating too much for gfx6-8 
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12466
Fixes: c87ce78d - ac/surface: enable thick tiling for 3D textures for better perf on gfx6-8

Reviewed-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/34432>
 2025-04-16 06:08:48 +00:00
Marek Olšák 195e7b4f75 ac/surface: make gfx12_estimate_size reusable by gfx6 
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12466
Fixes: c87ce78d - ac/surface: enable thick tiling for 3D textures for better perf on gfx6-8

Reviewed-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/34432>
 2025-04-16 06:08:48 +00:00
Marek Olšák 2c122d478b ac/nir: set X=0 for task->mesh shader dispatch when Y or Z is 0 
The code set X=0 when Y and Z is 0, not "or".

Reviewed-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/34432>
 2025-04-16 06:08:48 +00:00
Marek Olšák 963147d7fd ac/gpu_info: add 256 to payload_entry_size to increase future task shader perf 
It has no effect because num_entries is 1K, but the table shows a lot of
potential.

Reviewed-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/34432>
 2025-04-16 06:08:48 +00:00
Marek Olšák d7c903f258 ac/gpu_info: add payload_entry_size into ac_task_info 
to stop causing full RADV recompiles when it's changed.

Reviewed-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/34432>
 2025-04-16 06:08:48 +00:00
Marek Olšák 0dafd04695 ac/gpu_info: remove has_tmz_support function 
It's not needed since:
    8b3056343f - ac/gpu_info: bump required DRM minor version to 3.42.0 (kernel 5.15+)

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-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/34432>
 2025-04-16 06:08:48 +00:00
Marek Olšák 0be5a3559a ac/gpu_info: increase the attribute ring size for gfx12 
Reviewed-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/34432>
 2025-04-16 06:08:48 +00:00
Ian Romanick e783930b10 elk/algebraic: Don't optimize float SEL.CMOD to MOV 
Floating point SEL.CMOD may flush denorms to zero. We don't have enough
information at this point in compilation to know whether or not it is
safe to remove that.

Integer SEL or SEL without a conditional modifier is just a fancy
MOV. Those are always safe to eliminate.

See also 3f782cdd25.

Fixes: fab92fa1cb ("i965/fs: Optimize SEL with the same sources into a MOV.")
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34192>
 2025-04-15 23:59:31 +00:00
Ian Romanick f4ede9c10a elk/algebraic: Clear condition modifier on optimized SEL instruction 
The condition modifier on SEL means something completely different than
it means on MOV.  On MOV it means to modify the flags based on the value
written to the destination. On SEL it means to compare the sources using
that mode and pick the result (i.e., as min() or max()) without
modifying the flags.

The resulting MOV should not have a condition modifier for the same
reason it (already) doesn't have a predicate. This bug was found by
inspection, so I added a unit test.

Fixes: fab92fa1cb ("i965/fs: Optimize SEL with the same sources into a MOV.")
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34192>
 2025-04-15 23:59:31 +00:00
Ian Romanick 6a19d8915f brw/algebraic: Don't optimize float SEL.CMOD to MOV 
Floating point SEL.CMOD may flush denorms to zero. We don't have enough
information at this point in compilation to know whether or not it is
safe to remove that.

Integer SEL or SEL without a conditional modifier is just a fancy
MOV. Those are always safe to eliminate.

See also 3f782cdd25.

Fixes: fab92fa1cb ("i965/fs: Optimize SEL with the same sources into a MOV.")

No shader-db changes on any Intel platform.

fossil-db:

All Intel platforms had similar results. (Lunar Lake shown)
Totals:
Instrs: 209903490 -> 209903492 (+0.00%)
Cycle count: 30546025224 -> 30546021980 (-0.00%); split: -0.00%, +0.00%
Max live registers: 65516231 -> 65516235 (+0.00%)

Totals from 2 (0.00% of 706657) affected shaders:
Instrs: 3197 -> 3199 (+0.06%)
Cycle count: 361650 -> 358406 (-0.90%); split: -10.05%, +9.15%
Max live registers: 300 -> 304 (+1.33%)

Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34192>
 2025-04-15 23:59:31 +00:00
Ian Romanick 07dc1d4043 brw/algebraic: Clear condition modifier on optimized SEL instruction 
The condition modifier on SEL means something completely different than
it means on MOV.  On MOV it means to modify the flags based on the value
written to the destination. On SEL it means to compare the sources using
that mode and pick the result (i.e., as min() or max()) without
modifying the flags.

The resulting MOV should not have a condition modifier for the same
reason it (already) doesn't have a predicate. This bug was found by
inspection, so I added a unit test.

No shader-db or shader-db changes on any Intel platform.

Fixes: fab92fa1cb ("i965/fs: Optimize SEL with the same sources into a MOV.")
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34192>
 2025-04-15 23:59:31 +00:00
Ian Romanick 1d2ebeca17 nir/algebraic: Allow fmin(a,a) optimization when flush denorm to zero is not set 
I was surprised this had any affect on Intel GPUs because we have been
unconditionally performing this optimization in the backend since June
2014.

Once that error is fixed (later in this MR), this change prevents a
couple dozen regressions in shader-db and around 90 regressions in
fossil-db. Many of the regressions in fossil-db were loss of SIMD32, and
that can be a big deal.

v2: Add 64-bit too. Suggested by Alyssa.

shader-db:

All Intel platforms had similar results. (Lunar Lake shown)
total instructions in shared programs: 16970141 -> 16970139 (<.01%)
instructions in affected programs: 40 -> 38 (-5.00%)
helped: 2 / HURT: 0

total cycles in shared programs: 914617580 -> 914617548 (<.01%)
cycles in affected programs: 3428 -> 3396 (-0.93%)
helped: 2 / HURT: 0

fossil-db:

All Intel platforms had similar results. (Lunar Lake shown)
Totals:
Cycle count: 30546028462 -> 30546025224 (-0.00%); split: -0.00%, +0.00%
Non SSA regs after NIR: 237017827 -> 237017731 (-0.00%)

Totals from 83 (0.01% of 706657) affected shaders:
Cycle count: 3042978 -> 3039740 (-0.11%); split: -0.13%, +0.02%
Non SSA regs after NIR: 78997 -> 78901 (-0.12%)

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34192>
 2025-04-15 23:59:31 +00:00
Faith Ekstrand 44b01b55d5 nvk: Handle shifted QMD cbuf addrs in indirect command processing 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34443>
 2025-04-15 23:38:54 +00:00
Faith Ekstrand 8b2f0be254 nak/qmd: Add QMD version 4.0 for Hopper 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34443>
 2025-04-15 23:38:54 +00:00
Faith Ekstrand a62c59d7e0 nak/qmd: Add support for shifted cbuf addresses 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34443>
 2025-04-15 23:38:54 +00:00
Faith Ekstrand 7050a285a3 nak/qmd: Rework cbuf size suffix handling 
Addresses will also be shifted starting with QMD version 4.0 and we'll
need something a little more general for that.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34443>
 2025-04-15 23:38:54 +00:00
Faith Ekstrand c8a5086213 nvk: Disable VK_EXT_descriptor_buffer pre-Maxwell 
Our strategy of covering the entire address space in buffer views
requires that we be able to create very large buffer views.  The
pre-Maxwell texture unit doesn't allow for this.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34542>
 2025-04-15 18:03:53 -05:00
Faith Ekstrand c135bd6542 nak: Get rid of RegRef::zero 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34538>
 2025-04-15 22:17:38 +00:00
Faith Ekstrand 7125b25800 nak/sm70: Add zero_reg() and true_reg() helpers 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34538>
 2025-04-15 22:17:38 +00:00
Faith Ekstrand c4452c5fa4 nak/sm50: Add zero_reg() and true_reg() helpers 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34538>
 2025-04-15 22:17:38 +00:00
Faith Ekstrand 32570924cf nak: Move some legalization helpers from sm50 to common code 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34538>
 2025-04-15 22:17:38 +00:00
Dave Airlie 6a4c1ac464 nak: add F2FP to sm75 instr latencies 
hw_tests found this was missing, copy it from Ampere.

Fixes: 6b8a4e6bb7 ("nak: Add Turing latency information")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34537>
 2025-04-15 21:56:55 +00:00
Ryan Mckeever 1efa53ac17 panvk: shaderStorageImageWriteWithoutFormat support 
Signed-off-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33787>
 2025-04-15 21:30:41 +00:00
Ryan Mckeever e79a568442 panvk: shaderStorageImageReadWithoutFormat support 
Signed-off-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33787>
 2025-04-15 21:30:41 +00:00
Ryan Mckeever 70b8056df1 panvk: Enable KHR_format_feature_flags2 and use them 
We need to enable this to use
VK_FORMAT_FEATURE_2_STORAGE_READ_WITHOUT_FORMAT_BIT and
VK_FORMAT_FEATURE_2_STORAGE_WRITE_WITHOUT_FORMAT_BIT.

Signed-off-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33787>
 2025-04-15 21:30:41 +00:00
Ryan Mckeever d95423686f pan/format: Add PAN_BIND_STORAGE_IMAGE flag 
The HW does not support STORAGE_IMAGE for all formats that can be sampled
or used as render targets. The HW does not support STORAGE_TEXEL_BUFFER
for all formats that can be used as render targets.

Signed-off-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33787>
 2025-04-15 21:30:41 +00:00
Ryan Mckeever b9a9798c46 pan/format: Update format flags to follow HW spec 
Fixes: 861e7dca ("panfrost: Switch formats to table")

Signed-off-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33787>
 2025-04-15 21:30:41 +00:00
Faith Ekstrand 58321cf2e5 nak: Add stubs for Fermi and Kepler A 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34536>
 2025-04-15 16:04:10 -05:00
Faith Ekstrand c0cd01241a nak: Add stubs for Kepler B 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34536>
 2025-04-15 16:04:10 -05:00
Faith Ekstrand af6093a712 nak/legalize: Add a helper for lowering ineg 
This is similar to the helper we have for lowering float modifiers.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34536>
 2025-04-15 16:04:09 -05:00
Faith Ekstrand d16e75e55f nak: Lower texture inputs for Kepler B 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34535>
 2025-04-15 20:02:52 +00:00
Faith Ekstrand 9c5a0eca95 nak: Add False and True to IntCmpOp 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34535>
 2025-04-15 20:02:52 +00:00
Faith Ekstrand eb4fb70bae nak: Add a plop2 test 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34535>
 2025-04-15 20:02:52 +00:00
Faith Ekstrand e84f210362 nak/hw_tests: Feed predicate/carry sources with 0/1 data 
Instead of doing the &1 on the GPU, do it on the CPU.  This saves an
instruction and also makes things easier to debug.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34535>
 2025-04-15 20:02:51 +00:00
Erik Faye-Lund 26a4a2ee18 docs/panvk: add missing new features 
We missed these from the release notes, let's add them.

Fixes: fbb2805575 ("panvk: Advertise KHR_dynamic_rendering_local_read support")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34525>
 2025-04-15 20:54:09 +02:00
Erik Faye-Lund c6025f6cb6 docs/panvk: document EXT extension aswell 
We're supporting both, let's mention both in the release notes.

Fixes: 4fabd37a3c ("panvk: enable KHR_line_rasterization support")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34525>
 2025-04-15 20:53:55 +02:00
Erik Faye-Lund 469985cfa4 docs/panvk: document ycbr in terms of extensions 
We usually document these features in terms of extensions when they
exist, so let's do that here too.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34525>
 2025-04-15 20:53:55 +02:00
Erik Faye-Lund 37e75da005 docs/panvk: add VK_-prefix for extension name 
We're using the VK_-prefix elsewhere, let's be consistent.

Fixes: d4797b8ab7 ("panvk: enable KHR_spirv_1_4 on v10+")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34525>
 2025-04-15 20:53:55 +02:00
Erik Faye-Lund d5fce25fb0 docs/panvk: fixup docs around float controls 
We both misnamed the extension, and missed the v10+ detail for the
second version.

Fixes: 22fa3e88dd ("panvk: advertise VK_KHR_float_controls2")
Fixes: 7612dc4713 ("panvk: advertise VK_KHR_shader_float_controls")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34525>
 2025-04-15 20:53:55 +02:00
Erik Faye-Lund 68669ba0c1 docs/panvk: remove disabled extension 
We ended up reverting this one, let's remove it from the release notes
and features.txt.

Fixes: 305925c439 ("panvk: disable VK_KHR_shader_quad_control")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34525>
 2025-04-15 20:53:55 +02:00
Erik Faye-Lund 4efb1dac6b docs/panvk: fixup extension support 
This extension is now only supported on PanVK/v10+

Fixes: 23c0d64e24 ("panvk: disable VK_EXT_image_drm_format_modifier for arch < 10")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34525>
 2025-04-15 20:53:55 +02:00
Erik Faye-Lund 790093f625 panvk: enable dualSrcBlend 
This is already implemented, we just didn't report it.

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34520>
 2025-04-15 18:34:36 +00:00
Aleksi Sapon 9301b7098a llvmpipe: improve aniso filtering 
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34438>
 2025-04-15 18:03:22 +00:00