384f4448e1
nak: Get rid of meta instructions
...
They're mostly in the way at this point. Instead, make map_instrs return
a Vec<Instr>. We can reaplace that with a None, One, Some enum type
later if we want.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
438c67c1cb
nak: Use Instr::num_srcs/dsts() less
...
Instead use Iterator::enumerate() and is_empty().
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
fafb72e2be
HACK: Support old meson
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
44243c4741
nak: Lower 8 and 16-bit types
...
We can do some of these on some hardware but we'll lower them all for
now.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
8f121036ea
nak: Implement boolean logic ops
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
871bb59875
nak: Rework instruction encoding
...
Use a struct to store the temporary data and the SM number and over-all
reduce the amount of stuff we have to pass around. Also, rename
everything from tu102 to sm75 which is a more apt description.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
5207d9e3a0
nak: Encode load/store correctly on SM80
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
5daff0876f
nak: Plumb the SM through to nak::Shader
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
611ed96557
nvk: Pass an actual nak_compiler to nak_compile_shader()
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
bf72e78e0a
nak: Add nak_compiler_create/destroy
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
c10aa56efb
nak/meson: Use bindgen dependencies
...
These are new in meson 1.0 and lets us avoid all that manual include
directory mangling which was going to break one of these days anyway.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
e69a30cecb
nak: Rework ALU instruction encode
...
Instead of pulling sources from fixed locations in the instruction, take
a bunch of Option<&Ref> and let the individual encode func take care of
it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
0e755f9964
nak: Implement bcsel
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
b3d6dafc7d
nak: Implement integer comparisons
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
ac2a56f56f
nak: Add support for instruction predicates
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
de073a10e6
nak: Implement 32-bit logic ops
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
8cf12d6a3f
nak: Add a pass for computing instruction dependencies
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
93eb2f884a
nak: Implement iadd and ishl
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
241caaf5ff
nak: Add an instruction fuzzing tool
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
90d3fd10e1
nak: Zero out the .w component of descriptors
...
This saves us from having to constantly add it in as we compute
addresses.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
c212506771
nak: Implement load/store_global
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
c76331fb51
nak: Implement load_ubo
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
62f5e65c80
nak: Implement load_sysval_nv as S2R
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
fb96aaf3fa
nak: Lower system values to a new load_sysval_nak intrinsic
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
481a29a912
nak: Header stuff
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:07 +00:00
80376146ed
nak: Encode program headers
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
50b1deefa7
nak: Add encoding for a few instructions
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
83f8d6e2a7
nak: Add bitset infrastructure
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
812255fc79
nak: Add a lowering pass for ZERO sources and destinations
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
f694f1f8a8
nak: Add a lowering pass for VEC and SPLIT instructions
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
d44b31f3bf
nak: Add a trivial register allocator
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
08eb906111
nak: Add a util library
...
Currently, all we do here is implement div_ceil() and next_multiple_of()
because these helpers are super useful but currently an experimental
Rust feature. This should be a drop-in for the feature so we can delete
it once the feature stabilizes with little to no effect.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
4e95cb908f
nak: Add a dead-code pass
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
9769f19c7e
nak: Add a copy-prop pass
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
5585f57c5c
nak: Add initial translation from NIR
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
a4a5fe2270
nak: Add Rust bindings for NIR
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
771d9c8856
nak: Add the core IR
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
62695614d7
nvk: Run shaders through NAK
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
79ff2d9a33
nak: Add initial stubs for rust code
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
c778d39fa4
nak: Add a bunch of shader lowering code in NIR
...
The nifty thing about the way NVIDIA hardware does inputs and outputs is
that it maps really well to how we do them in NIR. We can make the
driver_locations match exactly to the attribute address space. For
fragment shader outputs, we make them register numbers.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
1b70f39ef6
nak: Copy the optimization loop from Intel
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
40f9327c8a
nouveau: Add initial headers and meson for the new compoiler
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24998 >
2023-11-14 00:48:06 +00:00
9ef621ec2e
d3d12: ARB_query_buffer_object and GL4.4
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26156 >
2023-11-14 00:18:53 +00:00
6384ccd1cd
d3d12: Reference count queries in a batch
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26156 >
2023-11-14 00:18:53 +00:00
9caea4c3e9
d3d12: Enable ARB_enhanced_layouts and ARB_texture_mirror_clamp_to_edge
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26156 >
2023-11-14 00:18:53 +00:00
707cdac7e7
d3d12: Fix GS variant I/O slot counts
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26156 >
2023-11-14 00:18:53 +00:00
9fe88dd9b9
d3d12: Support enhanced layouts for VS inputs
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26156 >
2023-11-14 00:18:53 +00:00
72341747f4
d3d12: Split dvec3 interpolatns into devc2 and double
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26156 >
2023-11-14 00:18:53 +00:00
1aa56d486c
d3d12: Fix location_frac_mask bitfield size
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26156 >
2023-11-14 00:18:53 +00:00
c77f8b4db0
microsoft/compiler: Stop lowering all I/O to temps
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26156 >
2023-11-14 00:18:53 +00:00
Faith Ekstrand