This means we respect the pattern order better because
simple replacements like bcsel(False, a, b) -> b no longer
insert movs that can block more specialized patterns.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40399>
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40190>
This was used because the exact bit meant something different for
comparisons than it did for the replacement expression, but that isn't the
case anymore.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39809>
match_expression() only accesses the first instr->def.num_components
elements, so we don't need to ensure the rest are zero.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39808>
Removes the runtime code for this, and means we propergate the
signed zero/inf/nan checks to subexpessions too, not just exact.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39616>
Unifies nir per instruction float control.
In the future this can be split into contract/reassoc/transform
like SPIR-V.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com> (except SPIR-V)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39103>
We don't need one bit per bitsize per instruction if only one actually
matters in the end.
First step towards moving NIR in the direction of full float_controls2
only.
Also rename this from fp_fast_math, because that name implied that 0 is
the no fast math mode, while the opposite was the case.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39026>
We add a bunch of new helpers to avoid the need to touch >parent_instr,
including the full set of:
* nir_def_is_*
* nir_def_as_*_or_null
* nir_def_as_* [assumes the right instr type]
* nir_src_is_*
* nir_src_as_*
* nir_scalar_is_*
* nir_scalar_as_*
Plus nir_def_instr() where there's no more suitable helper.
Also an existing helper is renamed to unify all the names, while we're
churning the tree:
* nir_src_as_alu_instr -> nir_src_as_alu
..and then we port the tree to use the helpers as much as possible, using
nir_def_instr() where that does not work.
Acked-by: Marek Olšák <maraeo@gmail.com>
---
To eliminate nir_def::parent_instr we need to churn the tree anyway, so I'm
taking this opportunity to clean up a lot of NIR patterns.
Co-authored-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38313>
Slight differences due to different optimization order.
Totals from 135 (0.17% of 79839) affected shaders: (Navi48)
Instrs: 287852 -> 287527 (-0.11%); split: -0.15%, +0.03%
CodeSize: 1522972 -> 1521764 (-0.08%); split: -0.12%, +0.04%
Latency: 1806803 -> 1825754 (+1.05%); split: -0.08%, +1.12%
InvThroughput: 242693 -> 244703 (+0.83%); split: -0.02%, +0.84%
VClause: 4092 -> 4084 (-0.20%)
SClause: 7462 -> 7478 (+0.21%)
Copies: 20509 -> 20401 (-0.53%); split: -0.74%, +0.21%
Branches: 6395 -> 6386 (-0.14%)
PreSGPRs: 7334 -> 7337 (+0.04%); split: -0.03%, +0.07%
PreVGPRs: 6375 -> 6382 (+0.11%)
VALU: 151787 -> 151595 (-0.13%); split: -0.15%, +0.02%
SALU: 52967 -> 52910 (-0.11%); split: -0.23%, +0.12%
VMEM: 6704 -> 6696 (-0.12%)
SMEM: 12099 -> 12129 (+0.25%)
Tested on a small collection of 2518 shaders from Dredge with callgrind using RADV:
baseline:
nir_opt_algebraic was called 12917 times from radv_optimize_nir()
nir_opt_cse was called 15204 times from radv_optimize_nir()
relative time spent in radv_optimize_nir(): 31.48%
total instruction fetch cost: 28,642,638,021
with nir/algebraic: ad-hoc constant-fold ALU instructions
nir_opt_algebraic was called 12797 times from radv_optimize_nir()
nir_opt_cse was called 12963 times from radv_optimize_nir()
relative time spent in radv_optimize_nir(): 30.63%
total instruction fetch cost: 28,284,386,123
=> ~1.27% improvement in total compile times
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37195>
Most of the time, we can infer the type to append in
util_dynarray_append using __typeof__, which is standardized in C23 and
support in Jesse's MSMSVCV. This patch drops the type argument most of
the time, making util_dynarray a little more ergonomic to use.
This is done in four steps.
First, rename util_dynarray_append -> util_dynarray_append_typed
bash -c "find . -type f -exec sed -i -e 's/util_dynarray_append(/util_dynarray_append_typed(/g' \{} \;"
Then, add a new append that infers the type. This is much more ergonomic
for what you want most of the time.
Next, use type-inferred append as much as possible, via Coccinelle
patch (plus manual fixup):
@@
expression dynarray, element;
type type;
@@
-util_dynarray_append_typed(dynarray, type, element);
+util_dynarray_append(dynarray, element);
Finally, hand fixup cases that Coccinelle missed or incorrectly
translated, of which there were several because we can't used the
untyped append with a literal (since the sizeof won't do what you want).
All four steps are squashed to produce a single patch changing every
util_dynarray_append call site in tree to either drop a type parameter
(if possible) or insert a _typed suffix (if we can't infer). As such,
the final patch is best reviewed by hand even though it was
tool-assisted.
No Long Linguine Meals were involved in the making of this patch.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38038>
_mesa_hash_table_clear() memsets the entries, even if it's already empty.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36760>
In the C23 standard unreachable() is now a predefined function-like
macro in <stddef.h>
See https://android.googlesource.com/platform/bionic/+/HEAD/docs/c23.md#is-now-a-predefined-function_like-macro-in
And this causes build errors when building for C23:
-----------------------------------------------------------------------
In file included from ../src/util/log.h:30,
from ../src/util/log.c:30:
../src/util/macros.h:123:9: warning: "unreachable" redefined
123 | #define unreachable(str) \
| ^~~~~~~~~~~
In file included from ../src/util/macros.h:31:
/usr/lib/gcc/x86_64-linux-gnu/14/include/stddef.h:456:9: note: this is the location of the previous definition
456 | #define unreachable() (__builtin_unreachable ())
| ^~~~~~~~~~~
-----------------------------------------------------------------------
So don't redefine it with the same name, but use the name UNREACHABLE()
to also signify it's a macro.
Using a different name also makes sense because the behavior of the
macro was extending the one of __builtin_unreachable() anyway, and it
also had a different signature, accepting one argument, compared to the
standard unreachable() with no arguments.
This change improves the chances of building mesa with the C23 standard,
which for instance is the default in recent AOSP versions.
All the instances of the macro, including the definition, were updated
with the following command line:
git grep -l '[^_]unreachable(' -- "src/**" | sort | uniq | \
while read file; \
do \
sed -e 's/\([^_]\)unreachable(/\1UNREACHABLE(/g' -i "$file"; \
done && \
sed -e 's/#undef unreachable/#undef UNREACHABLE/g' -i src/intel/isl/isl_aux_info.c
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36437>
The fmul+fadd -> fma rules in nir_opt_algebraic are marked imprecise,
because they are a contraction. However, they respect signed zero/Inf/NaN rules.
As such, it is legal to do this fusion with shader float controls as long as the
exact bit is not set (mapping to SPIR-V NoContract).
Unfortunately, NIR's imprecise rules do not distinguish between contraction
issues versus float special case issues, forcing nir_search to skip all
imprecise rules when any shader float control modes are used. This notably
affects DXVK, which sets shader float controls to get D3D11 float behaviour and
hence loses FMA fusing.
Therefore, we plumb in the exact bit to express NoContract independent of the
float controls, and weaken the requirement for fma fusion to allowable
contraction. For fma splitting, it's a similar issue, as inexact GLSL fma in
SPIR-V is just a multiply add that we're allowed to contract rather than the
real deal.
Drivers that use their own FMA fusing passes (notably, Intel and AMD) are
unaffected, but DXVK-capable drivers using fuse_ffma should like this. Results
on hk shown:
Totals from 2194 (4.06% of 54019) affected shaders:
MaxWaves: 2174272 -> 2175936 (+0.08%); split: +0.08%, -0.01%
Instrs: 1173283 -> 1131494 (-3.56%); split: -3.57%, +0.01%
CodeSize: 8568168 -> 8381724 (-2.18%); split: -2.18%, +0.01%
Spills: 1094 -> 747 (-31.72%)
Fills: 988 -> 681 (-31.07%)
Scratch: 4444 -> 3820 (-14.04%)
ALU: 953032 -> 913149 (-4.18%); split: -4.19%, +0.01%
FSCIB: 953032 -> 913149 (-4.18%); split: -4.19%, +0.01%
IC: 215398 -> 215274 (-0.06%)
GPRs: 139865 -> 139032 (-0.60%); split: -1.56%, +0.96%
Uniforms: 414886 -> 414466 (-0.10%); split: -0.14%, +0.04%
Preamble instrs: 646398 -> 644017 (-0.37%); split: -0.43%, +0.07%
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35989>
This is required to optimize FP64 and Int64 shaders generated by
virglrenderer. It generates pack/unpack around every 64-bit op,
which NIR currently can't eliminate. This fixes that.
There is a new constraint ".y", which means that the use of an instruction
should have swizzle.y. This allows us to add patterns that have Y swizzle
on results of instructions.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32172>
With float_controls2, shaders can decide on the behavior of
NaN/Inf/SignedZero preservation by decorating specific instructions, on
top of having a default for the whole program.
Add where to track these to nir_alu_instr and propagate them to new
instructions everywhere that exact is being done already.
v2: use less bits for fp_fast_math in nir_alu_instr (Alyssa)
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27281>
First, we need to give the parent_instr field a unique name to be able to
replace with a helper. We have parent_instr fields for both nir_src and
nir_def, so let's rename nir_src::parent_instr in preparation for rework.
This was done with a combination of sed and manual fix-ups.
Then we use semantic patches plus manual fixups:
@@
expression s;
@@
-s->renamed_parent_instr
+nir_src_parent_instr(s)
@@
expression s;
@@
-s.renamed_parent_instr
+nir_src_parent_instr(&s)
@@
expression s;
@@
-s->parent_if
+nir_src_parent_if(s)
@@
expression s;
@@
-s.renamed_parent_if
+nir_src_parent_if(&s)
@@
expression s;
@@
-s->is_if
+nir_src_is_if(s)
@@
expression s;
@@
-s.is_if
+nir_src_is_if(&s)
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24671>
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>
