fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-05-27 18:48:14 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 3
180759 commits 66 branches 936 tags 956 MiB
Commit graph

1526 commits

Author SHA1 Message Date
Faith Ekstrand
23e1f3c373 nvk: Use nak_shader_info natively 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:13 +00:00
Faith Ekstrand
c074ea6215 nak: Handle the num_gpr offsetting inside nak 
This makes the thing in the nak_shader_info exactly the thing that gets
plugged into the hardware.  Makes the driver a bit simpler.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:13 +00:00
Faith Ekstrand
d8551cd328 nak: Add a writes_layer bit to nak_shader_info::vtg 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
a232050204 nak: Move clip, cull, and XFB into a nak_shader_info.vtg 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
440adf7970 nak: Properly prefix nak_xfb_info 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
4e6e814f5e nak: Rename TLS to SLM 
Shader Local Memory is what NVIDIA calls it in the shader header docs as
well as the command stream headers.  Better to be consistent even if it
gets my Intel brain confused.  (Intel uses SLM for shared memory.)

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
a946071546 nvk: Use nak_fs_key instead of rolling our own 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
0f086401e3 nvk: Move even more lowering into nvk_codegen.c 
At this point, we're fully trusting NAK to do its own lowering and we
only lower stuff in nvk_shader.c if it's relevant for Vulkan.  This also
assumes that NAK is already doing the right thing everywhere.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
67bb8e8165 nvk: Move the guts of nvk_compile_nir() to nvk_codegen.c 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
0405f494e8 nvk: Move the optimization loop to the nvk_codegen.c 
We also call it from nak_preprocess_nir and lower var copies there.  NAK
should already be doing this for us.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
7f8fbacb8a nvk: Move a bunch of codegen-specific lowering to helpers 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
c3a44f6264 nvk: Add a codegen helper for nir_shader_compiler_options 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
845e7d2911 nvk: Only lower outputs to temporaries 
Also, move it up to right after we parse the SPIR-V and remove some now
unnecessary clean-up passes.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
26bb5f4972 nak/nir: Lower indirect FS inputs 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
e507d70333 nvk: Handle load_first_vertex in nvk_nir_lower_descriptors() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
82061b1b9d nvk: Only advertise VK_KHR_shader_terminate_invocation if using NAK 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26197>
 2023-11-15 02:24:12 +00:00
Faith Ekstrand
618bdb8571 nak: Rework FS input interpolation 
This gives FS I/O the same treatment as we did for vertex attributes in
that we now have a NIR intrinsic which pretty closely matches the
hardware and we lower to that before going into NAK.  This gives us a
bit more control in the NIR.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26181>
 2023-11-14 16:38:03 +00:00
Faith Ekstrand
d3c5688cf5 nak: Plumb the nak_compiler through to lower_fs_input_intrin 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26181>
 2023-11-14 16:38:02 +00:00
Faith Ekstrand
f5ba0751e2 nak: Make encode_sm75 a method of Shader 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26181>
 2023-11-14 16:38:02 +00:00
Faith Ekstrand
a6376705e4 nak: Make ALD/AST.PHYS a boolean 
The generic flags field was originally copied from codegen but a boolean
makes way more sense.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26181>
 2023-11-14 16:38:02 +00:00
Faith Ekstrand
8e00ee6fe8 nak: Drop OpAtomCas in favor of OpAtom with atom_op == CmpExch 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26181>
 2023-11-14 16:38:02 +00:00
Faith Ekstrand
ea453b373d nak: Fix copy-prop for OpPLop3 sources 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26181>
 2023-11-14 16:38:02 +00:00
Faith Ekstrand
a65518b625 nvk: Free NAK shaders 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26181>
 2023-11-14 16:38:02 +00:00
Faith Ekstrand
ade73d2e7f nvk: Stop asserting 11-bit storage image handles 
Now that NAK is the default for Turing+, we can just chalk any storage
image descriptor handle overruns up to codegen bugs.  We could plumb
shader stages all the way through to here and only assert when codegen
is in use but that's a lot of work just for an assert.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
92cb4cc7f6 nvk: Default to NAK on Turing+ 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
M Henning
f2495f47a7 nvk: Use load_global_constant for ubo loads 
and support load_global_constant in nak

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
M Henning
1d7e051a7f nak: Add MemOrder::Constant 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
M Henning
2a02a0d0b3 nak: Bind nir_intrinsic_access 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
M Henning
dfb6260483 nak: Specify MemScope on MemOrder::Strong 
The scope isn't relevant on MemOrder::Weak. In fact, ptxas refuses
any input that specifies a scope on weak loads. So, don't make those
combinations representable in our IR.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
e43e8f3f78 nak: Require meson 1.3.0 and clean up a couple bits 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
7feba721c5 nak: Rely on Rust 1.73 for next_multiple_of() and div_ceil() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
c7c73d6d17 nvk: Enable subgroups features 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
4bd9c99ca0 nak: Document a bit in encode_lds() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
0177373e2d nak: Insert an OpNop after OpBar 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
e111bfdb5c nak: Emit MemBar before Bar 
This is what NVIDIA does. I'm not sure why this order but it seems
necessary to pass some of the 1.1 memory model tests.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
bc58620e1e nak: Use the simplified BAR.SYNC encoding 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
4f2f9c6ab3 nak: Use strong ordering for Image load/store 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
ec2c866a77 nak: Emit CCtl in barriers with acq/rel semantics 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
8781acba18 nak: Allow 1-component image load/store 
We get these when we get an atomic image load/store

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
ee11b473b5 nak: Implement read_invocation and shuffle_* 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
edf015eb7f nak: Fix the encoding of OpShfl 
We weren't handling Zero.  Also, we need to mask immediates or else the
encoder blows up.  The hardware automatically masks them when they come
in as sources but when we get immediates, they're not guaranteed to fit
in the bitfield.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
5dd7a76c8b nak: Implement vote and ballot 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
3df9065296 nak/nir: Zero-pad subgroup masks 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
37de14e2e2 nak/nir: Allow boolean vote_ieq 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
143d88dcc3 nak: Lower subgroup_id and num_subgroups 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
42a305416a nak: Use nir_shader_intrinsics_pass for system values 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
ce40d81d46 nak: Call nir_lower_subgroups() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
bf8642e984 nak: Remove unnecessary control barriers 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
3468f70ca1 nak: Use barriers for re-convergence 
Now that we're getting total re-convergence through barriers, we can
drop our OpWarpSyncs in front of barriers and derivatives

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00
Faith Ekstrand
22468439ff nak: Handle control-flow barriers 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998>
 2023-11-14 00:48:15 +00:00