I noticed the SGE case while looking at the output of
shaders/closed/steam/trine-2/fp-3.shader_test on i915g. These are
especially bad on i915 that needs two instructions to implement SNE.
An alternative would be to duplicate the sne(sXX(a, b), 0.0) rules in an
algebraic pass that occurs after bool_to_float. Doing the work earlier
seems preferable.
i915
total instructions in shared programs: 788274 -> 788223 (<.01%)
instructions in affected programs: 666 -> 615 (-7.66%)
helped: 5
HURT: 0
helped stats (abs) min: 9 max: 12 x̄: 10.20 x̃: 9
helped stats (rel) min: 5.00% max: 11.11% x̄: 8.12% x̃: 8.16%
95% mean confidence interval for instructions value: -12.24 -8.16
95% mean confidence interval for instructions %-change: -10.81% -5.43%
Instructions are helped.
LOST: 0
GAINED: 2
The two gained shaders are assembly fragment programs in Euro Truck
Simulator 2.
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15210>
This should reduce follow-on optimization work to copy-propagate and
dead-code away the movs generated in construction of vectors.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14865>
Trivial, but will help users avoid some struct constructions that can be
awkward in C.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14865>
Many users of nir_vec() do so by nir_channel()-ing a new ssa defs as movs
from other vectors to put the new vector together, which then just have to
get copy-propagated into the ALU srcs and DCEed away the temporary movs.
If they instead take nir_ssa_scalar, we can avoid that extra work.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14865>
moved from radeonsi without the vectorization, which won't be needed for
now. We will lower IO in st/mesa instead of radeonsi to get the transform
feedback info into store instructions.
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14388>
This is for drivers that have separate store instructions for varyings,
system value outputs (such as clip distances), and transform feedback.
The flags tell the driver not to store the output to those locations.
This will be used by radeonsi initially, and then maybe by a new linker.
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14388>
This will allow compaction of transform feedback varyings because they
are no longer tied to varying slots with this information.
It will also make transform feedback info available to all NIR passes
after IO is lowered. It's meant to replace pipe_stream_output_info.
Other intrinsics are not used with transform feedback.
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14388>
for this pass to work with xfb, the original value in the shader must be
preserved when xfb is active, and the driver must export only the newly
created output
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15184>
creating this at the start of the shader means it will get optimized out
when the pass is used to overwrite existing psiz values, and creating it
at the end means it will get optimized out in geometry shaders, so instead
just walk the instructions and create another store right after the existing one
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15184>
This required relaxing a core NIR assertion which I don't think is doing
any important validation.
The shader-db effects here are small, but they're important for avoiding a
regression when we start doing per-component DCE in opt_shrink_vectors
(https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12468)
softpipe shader-db:
total instructions in shared programs: 2859777 -> 2859454 (-0.01%)
instructions in affected programs: 18881 -> 18558 (-1.71%)
total temps in shared programs: 293994 -> 293914 (-0.03%)
temps in affected programs: 418 -> 338 (-19.14%)
i915g:
total instructions in shared programs: 407562 -> 407544 (<.01%)
instructions in affected programs: 570 -> 552 (-3.16%)
r300:
total instructions in shared programs: 1414450 -> 1414459 (<.01%)
instructions in affected programs: 44494 -> 44503 (0.02%)
total vinst in shared programs: 473782 -> 473727 (-0.01%)
vinst in affected programs: 1102 -> 1047 (-4.99%)
total sinst in shared programs: 231224 -> 231216 (<.01%)
sinst in affected programs: 432 -> 424 (-1.85%)
total temps in shared programs: 197605 -> 197607 (<.01%)
temps in affected programs: 103 -> 105 (1.94%)
crocus hsw:
total instructions in shared programs: 8158185 -> 8158134 (<.01%)
instructions in affected programs: 10927 -> 10876 (-0.47%)
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15178>
Task shader outputs work differently than other shaders, so they
need special consideration. Essentially, they have two kinds of
outputs:
1. Number of mesh shader workgroups to launch.
Will be still represented by a shader output.
2. Optional payload of up to (at least) 16K bytes.
These payload variables behave similarly to shared memory, but
the spec doesn't actually define them as shared memory (also, they
may be implemented differently by each backend), so we need to add
a new NIR variable mode for them.
These payload variables can't be represented by shader outputs
because the 16K bytes don't fit the 32x vec4 model that NIR uses
for its output variables.
This patch adds a new NIR variable mode: nir_var_mem_task_payload
and corresponding explicit I/O intrinsics, as well as support for
this new mode in nir_lower_io.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14930>
It won't work if the blob is fixed-size and we overrun the size, which
will be the case with the Vulkan pipeline cache.
This gets a bit tricky for the repeated-header optimization, because we
can't read the header from the blob. Instead we have to store the header
itself.
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15028>
nir_alu_instr_is_comparison needs to consider all comparison opcodes regardless
of size. Otherwise, they will be missed by nir_opt_move/sink.
Without this change, lowering booleans to integers regresses register
pressure (and spills/fills) significantly in certain shaders on Panfrost,
like android/com.miHoYo.GenshinImpact/1420.shader_test.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15073>
Matches the expected use by callers. We do need to fix up a few callers which
use this call for external shaders.
v2: Fix up a radv call site (Rhys).
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net> [v1]
Acked-by: Rhys Perry <pendingchaos02@gmail.com>
Acked-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14936>
Commit 38800b38 changed nir_opcodes.py, but that doesn't seem to have
triggered nir_opt_algebraic.py. The change in 75ef5991 depends on
opt_algebraic lowering 16-bit versions of slt, but if opt_algebraic is
not rebuilt, this may not happen. This resulted in some people seeing
assertion failures in, for example,
dEQP-VK.spirv_assembly.instruction.compute.float16.arithmetic_3.step,
due to the backend seeing nir_op_slt that it didn't know how to handle.
v2: Add nir_opcodes.py to nir_algebraic_py so that all the per-driver
algebraic passes pick up the dependency too. Rename it to
nir_algebraic_depends. Suggested by Emma.
Closes: #6047
Fixes: d1992255bb ("meson: Add build Intel "anv" vulkan driver")
Reviewed-by: Emma Anholt <emma@anholt.net>
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15050>
memcpy is divided into chunks that are vec4 sized max. The problem
here happens with a structure of 24 bytes :
struct {
float3 a;
float3 b;
}
If you memcpy that struct, the lowering will emit 2 load/store, one of
sized 8, next one sized 16. But both end up located at offset 0, so we
effectively drop 2 floats.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: a3177cca99 ("nir: Add a lowering pass to lower memcpy")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15049>
I was doing some RE on freedreno and we had some questions about when the
hardware might need non-uniform or non-constant array access for various
descriptor types, so let's leave some notes for the next person.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13621>
This intrinsic specific to RADV will be used to load VRS rates from
an user SGPR when RADV_FORCE_VRS is enabled by the application.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14713>
This patch moves the shrinking of store sources into
a separate pass.
The reasoning behind this is that this pass usually only
needs to be called once while nir_shrink_vectors might
better be called several times. This allows to move
the pass(es) out of the optimization loops.
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14480>
Properly handling NaN adversely affects several hundred shaders in
shader-db (lots of Skia and a few others from various synthetic
benchmarks) and fossil-db (mostly Talos and some Doom 2016). Only apply
the NaN handling work-around when the shader demands it.
v2: Add comment explaining the 1.0*y_over_x. Suggested by Caio.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Fixes: 2098ae16c8 ("nir/builder: Move nir_atan and nir_atan2 from SPIR-V translator")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13999>
frexp_sig of ±0, ±Inf, or NaN should just return the input unmodified.
frexp_exp of ±Inf or NaN is undefined, and frexp_exp of ±0 should return
the input unmodified. This seems to already work.
No shader-db or fossil-db changes on any Intel platform.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Fixes: 23d30f4099 ("spirv,nir: lower frexp_exp/frexp_sig inside a new NIR pass")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13999>
Extend 4195a9450b so that the next poor fool doesn't come along and
say, "sge does the right thing for 16-bit sources, but slt gives a NIR
validation failure. What the deuce?"
NOTE: This commit is necessary to prevent regressions in GLSLstd450Step
tests of 16-bit sources at "spriv: Produce correct result for
GLSLstd450Step with NaN".
Fixes: 4195a9450b ("nir: sge operation is defined for floating-point types")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13999>
When slots is 64 only the first bit was being set, instead of setting
all 64 bits of the variable, so for that case the function
get_variable_io_mask() always returned 0.
This behaviour caused variables that are being used both on producer and
consumer to be considered unused and thus being removed on
nir_remove_unused_io_vars().
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14955>
There is a comment in nir_fold_16bit_sampler_conversions saying that these
are the same, but the code only checks for i2i16.
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14893>
It was introduced for nir-to-tgsi, and I found that it was the wrong
approach. There's a reason nobody else does RA this way.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14404>
v2: Add helper for acceleration->root_node computation (Caio)
v3: Update comment on "done" bit (Caio)
Remove progress bool value for impl function (Caio)
Don't use nir_shader_instructions_pass to search the shader (Caio)
v4: Rename variable for if/else block (Caio)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13719>
We'll use this to apply ray tracing operations in our trivial return
shader based on the stage we're in.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13719>
In the future we'll want to reuse this intrinsic to deal with ray
queries. Ray queries will use a different global pointer and
programmatically change the control/level arguments of the trace send
instruction.
v2: Comment on barrier after sync trace instruction (Caio)
Generalize lsc helper (Caio)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13719>
This will allow to reuse the same intrinsic for various topology based
ID.
v2: fix intrinsic comment (Caio)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13719>
This pass (lowering discard_if to control flow and unconditional
discard) was originally written for Zink, but is useful for hardware
that lacks conditional discard instructions like AGX. In theory AGX
could implement a conditional discard with CSEL, but the vendor
compiler uses a lowering like this one. Since I like not writing code,
I'd like to use the pass that's already in tree.
v2: Don't preserve dominance (Jason). Assert we don't see demotes or
terminates (Jason). Add Mike's ack.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14217>
Qualcomm doesn't natively support shuffle, but it does natively support
relative shuffles where the delta is a constant. Therefore we'll expose
emulated support for both. Add support for this emulation of
subgroupShuffle() to NIR.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14412>
