This is necessary to allow optimizing VS inputs after nir_lower_io, which
is currently impossible because the loss of dual-slot information in NIR
would break VS inputs. With this, driver locations can be recomputed by
calling nir_recompute_io_bases. It's a prerequisite for optimizing varyings
with lowered IO.
When this is used, we will be able to eliminate unused dual-slot VS inputs
as well as unused low and high halves of dual-slot VS inputs for the first
time, which can happen due to optimizations of varyings. Without this,
st/mesa binds vertex buffers for dual-slot inputs that are fully or
partially unused in the shader.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25394>
There are two new variable modes:
- nir_var_mem_node_payload
- nir_var_mem_node_payload_in
Also add a few more intrinsics and some shader info.
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24512>
We already renamed the type, we just need to rename the enum and the
casting helper functions.
Generated with sed:
sed -i -e 's/nir_instr_type_ssa_undef/nir_instr_type_undef/g' src/**/*.h src/**/*.c src/**/*.cpp
sed -i -e 's/nir_instr_as_ssa_undef/nir_instr_as_undef/g' src/**/*.h src/**/*.c src/**/*.cpp
and two tiny whitespace fixups in lima.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24703>
Instead, we replace every use of it with nir_def. Most of this commit
was generated by sed:
sed -i -e 's/dest.ssa/def/g' src/**/*.h src/**/*.c src/**/*.cpp
A few manual fixups were required in lima and the nir_legacy code.
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24674>
Instead, we replace it directly with nir_def. We could replace it with
nir_dest but the next commit gets rid of that so this avoids unnecessary
churn. Most of this commit was generated by sed:
sed -i -e 's/dest.dest.ssa/def/g' src/**/*.h src/**/*.c src/**/*.cpp
There were a few manual fixups required in the nir_legacy.c and
nir_from_ssa.c as nir_legacy_reg and nir_parallel_copy_entry both have a
similar pattern.
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24674>
Nothing produces them any more, so remove them from NIR. This massively reduces
the size of nir_src, which should improve performance all over.
nir_src size reduced from 56 bytes -> 40 bytes (pahole results on arm64, x86_64
should be similar.)
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24253>
Makes it easier to copy snippets of shaders into code or
test comments without worrying about conflict with `/* */`.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23564>
Follow the same syntax as the intrinsic indices, since they
are conceptually similar.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23564>
- For SSA destination, padding is applied before `%`.
- For Reg destination, pad to the SSA size (to align div/con),
then remaining padding is applied before `r`.
- For instructions without destination, padding is applied so
they start right after the ` = ` of the cases above.
If the block doesn't have any destinations, there's no padding
is applied to the instructions without destinations in that
block.
For now registers with array access will be unaligned.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23564>
This frees up the shorter names for the intrinsic-based versions that will
replace them.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23956>
In addition to the hexadecimal and float (when applicable), print the
signed and unsigned representations. Representations may be omitted based
on information about the value:
- If gather types has unambiguous information, we use it;
- Float is omitted for 8 bit values;
- Signed decimal is omitted for positive values;
- Unsigned decimal is omitted for small values (representation is same as hex);
Note for now the "terse form" that appear in SSA uses is unchaged.
Based on a patch by Mike Blumenkrantz.
Examples:
```
// Just used as float. Omitted decimals.
vec4 32 ssa_81 = load_const (0x3f800000, 0x3f800000, 0x3e4ccccd, 0x3f800000) = (1.000000, 1.000000, 0.200000, 1.000000)
vec1 32 ssa_28 = load_const (0x3e4ccccd = 0.200000)
// Just a small integer. Omitted float and decimal.
vec1 32 ssa_45 = load_const (0x00000001)
// Larger positive integers. Omitted float.
vec1 32 ssa_39 = load_const (0x00002000 = 8192)
vec1 32 ssa_30 = load_const (0x000000ff = 255)
vec1 32 ssa_28 = load_const (0x00000010 = 16)
// Integers with negative values.
load_const (0xff = -1 = 255)
load_const (0xff80 = -128 = 65408)
load_const (0xffff = -1 = 65535)
// Same value, in the first case we know is used as an integer.
load_const (0xffffffe0 = -32 = 4294967264)
load_const (0xffffffe0 = -nan = -32 = 4294967264)
```
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Acked-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23562>
If the src_type is not available, untie by looking at the results from
nir_gather_ssa_types(). If that is ambiguous, just pick uint.
Now in print_const_from_load() when the type is invalid, print the full
constant form (with both padded hex and float); when the passed type
is valid, print the terse form based on it.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Acked-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23562>
In many cases the raw value is not really helpful,
since we only work with enums and the raw value is
already printed for indices without special printing.
If an index benefits from having special printing AND the
raw value, we can include the printing of the raw value
as part of its handler.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23375>
This is a prepare step to remove depends on p_defines.h in src/util/*
This is done by:
replace pipe_prim_type with mesa_prim
replace shader_prim with mesa_prim
replace PIPE_PRIM_MAX with MESA_PRIM_COUNT
replace SHADER_PRIM_ with MESA_PRIM_
replace PIPE_PRIM_ with MESA_PRIM_
This patch only replace code only
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23369>
Intel HW has multiple ways to access resources like UBO/SSBO/images :
- binding tables : a small ~240 heap of surfaces
- bindless surfaces : a 64Mb heap of surfaces up to Gfx12+, 4Gb on Gfx12.5+
- surfaces : a 4Gb heap on Gfx12.5+ (mostly unused at the moment,
only available through the LSC)
For samplers, we have 2 options since Gfx11+ :
- samplers indexed from the Dynamic State Heap (4Gb)
- samplers indexed from the Bindless Sampler Heap (4Gb)
Additionally our whole push constant promotion mechanism is based
around binding table indices. This is problematic if you want to also
promote to push constants things that would be accessed through the
bindless heap.
To solve this issue, we introduce a new intrinsic that will cary a
block index that is not based off the binding table index nor the
bindless table offset.
We will also use this intrinsic to identify whether the buffer/surface
index in load_ubo/load_ssbo/store_ssbo/etc... is relative to the
binding table or the bindless heap.
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/21645>
Currently, we have an atomic intrinsic for each combination of memory type
(global, shared, image, etc) and atomic operation (add, sub, etc). So for m
types of memory supported by the driver and n atomic opcodes, the driver has to
handle O(mn) intrinsics. This makes a total mess in every single backend I've
looked at, without fail.
It would be a lot nicer to unify the intrinsics. There are two obvious ways:
1. Make the memory type a constant index, keep different intrinsics for
different operations. The problem with this is that different memory types
imply different intrinsic signatures (number of sources, etc). As an
example, it doesn't make sense to unify global_atomic_amd with
global_atomic_2x32, as an example. The first takes 3 scalar sources, the
second takes 1 vector and 1 scalar. Also, in any single backend, there are a
lot more operations than there are memory types.
2. Make the opcode a constant index, keep different intrinsics for different
operations. This works well, with one exception: compswap and fcompswap
take an extra argument that other atomics don't, so there's an extra axis of
variation for the intrinsic signatures.
So, the solution is to have 2 intrinsics for each memory type -- for atomics
taking 1 argument and atomics taking 2 respectively. Both of these intrinsics
take an nir_atomic_op enum to describe its operation. We don't use a nir_op for
this purpose, as there are some atomics (cmpxchg, inc_wrap, etc) that don't
cleanly map to any ALU op and it would be weird to force it.
The plan is to transition to these new opcodes gradually. This series adds a
lowering pass producing these opcodes from the existing opcodes, so that
backends can opt-in to the new forms one-by-one. Then we can convert backends
separately without any cross-tree flag day. Once everything is converted, we can
convert the producers and core NIR as a flag day, but we have far fewer
producers than backends so this should be fine. Finally we can drop the old
stuff.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22914>
