For GLSL, we want to optimize code like
memoryBarrierBuffer();
controlBarrier();
into a single scoped_barrier intrinsic for the backend to consume. Now that
backends can get scoped_barriers everywhere, what's left is enabling backends to
combine these barriers together. We already have an Intel-specific pass for
combining memory barriers; it just needs a teensy bit of generalization to allow
combining all sorts of barriers together.
This avoids code quality regression on Asahi when switching to purely scoped
barriers. It's probably useful for other backends too.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21661>
This probably doesn't affect Vulkan or GL because they can't have
anything bigger than a vec4 anyway unless it's a u64vec4 and those have
to be at least 8B aligned. This may affect CL apps if they use
__attribute__((packed)) on something with big vectors, depending on how
LLVM decides to translate that.
Fixes: f8aa83f0c8 ("intel/nir: Use nir_lower_mem_access_bit_sizes()")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21524>
Rob added these new helpers a while back, which freedreno and radeonsi
both share. We should use them too. The new helpers use variables and
system value intrinsics, so we can drop the explicit binding table
creation and just use the normal paths.
Because we have to rewrite the system value uploading anyway, we drop
the scrambling of the default tessellation levels on upload, and instead
let the compiler go ahead and remap components like any normal shader.
In theory, this results in more shuffling in the shader. In practice,
we already do MOVs for message setup. In the passthrough shaders I
looked at, this resulted in no extra instructions on Icelake (SIMD8
SINGLE_PATCH) and Tigerlake (8_PATCH). On Haswell, one shader grew by
a single instruction for a pittance of cycles in a stage that isn't a
performance bottleneck anyway. Avoiding remapping wasn't so much of an
optimization as just the way that I originally wrote it. Not worth it.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20809>
This allows us to communicate to the back-end that we don't actually
know if the framebuffer is multisampled or not. No drivers set anything
but ALWAYS/NEVER and we still have a few ALWAYS/NEVER assumptions but
those should be asserted.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21094>
This allows for the possibility that we may not know at compile time if
sample shading is enabled through the API. While we're here, also
document exactly what this bit means so we don't confuse ourselves.
v2: Fixup coarse pixel values (Lionel)
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21094>
This gets our logical atomic messages using the lsc_opcode enum rather
than the legacy BRW_AOP_* defines. We have to translate one way or
another, and using the modern set makes sense going forward.
One advantage is that the lsc_opcode encoding has opcodes for both
integer and floating point atomics in the same enum, whereas the legacy
encoding used overlapping values (BRW_AOP_AND == 1 == BRW_AOP_FMAX),
which made it impossible to handle both sensibly in common code.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20604>
This option didn't exist half a decade ago when I first implemented base
workgroup support in ANV. It's cleaner to just have split system values
like all the other zero_base+base things do.
We currently only do this for COMPUTE and not KERNEL because it lets us
avoid changing intel_clc for now. We can add KERNEL later if needed.
We also don't do this lowering for task/mesh.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20068>
Gfx8 and Gfx9 platforms are helped for cycles because now many
instructions like
mul(8) g12<1>D g10<8,8,1>D 6D
become
mul(8) g12<1>D g10<8,8,1>D 6W
It is the same number of instructions, but the 32x16 multiply is a
little faster.
v2: Fix transposed hi and lo in "(hi >= INT16_MIN && lo <= INT16_MAX)".
Noticed by Caio. Use nir_src_is_const instead of open coding it.
Suggested by Caio.
Broadwell and Skylake had similar results. (Skylake shown)
total cycles in shared programs: 845748380 -> 845145547 (-0.07%)
cycles in affected programs: 446346348 -> 445743515 (-0.14%)
helped: 6017
HURT: 0
helped stats (abs) min: 2 max: 7380 x̄: 100.19 x̃: 8
helped stats (rel) min: <.01% max: 3.72% x̄: 0.41% x̃: 0.39%
95% mean confidence interval for cycles value: -113.37 -87.00
95% mean confidence interval for cycles %-change: -0.42% -0.41%
Cycles are helped.
Skylake
Cycles in all programs: 8844820715 -> 8828897462 (-0.2%)
Cycles helped: 47914
Cycles hurt: 1
No shader-db or fossil-db changes on any other Intel platform.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17718>
nir_opt_large_constants balks at seeing a store_deref of a variable
where the source is a vecN operation of multiple load_consts, and thinks
that isn't a constant, so it should not bother promoting it.
Unfortunately, we were running nir_lower_load_const_to_scalar before
nir_opt_large_constants, so this prevented a ton of constant promotion.
This commit /used to help/ some shaders in shader-db. Presumably since
!16770 landed, those shaders were already helped. Currently ther are
no shader-db changes on any Intel platform.
Fossil-db results:
All Intel platforms had similar results. (Ice Lake shown)
Instructions in all programs: 141998227 -> 141421756 (-0.4%)
Instructions helped: 12515
Instructions hurt: 237
SENDs in all programs: 7437925 -> 7468033 (+0.4%)
SENDs hurt: 12806
Cycles in all programs: 9161655753 -> 9132869800 (-0.3%)
Cycles helped: 10163
Cycles hurt: 2637
Spills in all programs: 19977 -> 18678 (-6.5%)
Spills helped: 384
Spills hurt: 40
Fills in all programs: 32863 -> 31396 (-4.5%)
Fills helped: 385
Fills hurt: 42
Lost: 1
Lots of Shadow of the Tomb Raider fragment shaders and Batman Arkham
Origins vertex shaders were hurt for SENDs in this commit. A couple
Aztec Ruins compute shaders and Spaceship shaders (multiple stages)
were also hurt.
All of the shaders hurt for spills or fills were Spaceship compute
shaders. Nearly all of the shaders helped were Shadow of the Tomb
Raider fragmenet shaders. One Spaceship shader was reall, REALLY helped:
Spills helped fossils/fossil-db/Spaceship.run.9f90a2a226fcc57f.1.foz/0b507d3abe2e3c28/compute: 321 -> 13 (-96.0%)
Fills helped fossils/fossil-db/Spaceship.run.9f90a2a226fcc57f.1.foz/0b507d3abe2e3c28/compute: 279 -> 21 (-92.5%)
Overall this seems like an improvement, but we may want to actually
run these few benchmarks before landing.
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16539>
NIR has two implementations of lower_idiv, keyed on the
imprecise_32bit_lowering flag. This flag is misleading: the results when
setting this flag "imprecise", they're completely wrong for some values.
If a backend has a native implementation of umul_high, the correct path
isn't that much more expensive. If it doesn't, it's substantially slower
for highp integer divison... but in practice, non-constant highp integer
division is pretty rare.
After a painful migration of the tree, this code path has no more users.
Remove it so nobody else gets the bright idea of using it again.
Closes: #6555
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19303>
Setting the NIR options takes care of iris thanks to the common st/mesa
linking code, and updating brw_nir_link_shaders should handle anv.
The main effort here is updating remap_tess_levels, which needs to
handle vector stores, writemasking, and swizzling. Unfortunately,
we also need to continue handling the existing single-component
access because it's used for TES inputs, which we don't vectorize.
We could try to vectorize TES inputs too, but they're all pushed
anyway, so it wouldn't buy us much other than deleting this code.
Also, we do have opt_combine_stores, but not one for loads.
One limitation of using nir_vectorize_tess_levels is that it works
on variables, and so isn't able to combine outer/inner writes that
happen to live in the same vec4 slot (for triangle domains). That
said, it's still better than before.
For writes, we allow the intrinsics to supply up to the full size
of the variable (vec4 for outer, vec2 for inner) even if the domain
only requires a subset of those components (i.e. triangles needs 3).
shader-db results on Icelake:
total instructions in shared programs: 19600314 -> 19597528 (-0.01%)
instructions in affected programs: 65338 -> 62552 (-4.26%)
helped: 271 / HURT: 0
helped stats (abs) min: 6 max: 24 x̄: 10.28 x̃: 12
helped stats (rel) min: 1.30% max: 18.18% x̄: 5.80% x̃: 7.59%
95% mean confidence interval for instructions value: -10.71 -9.85
95% mean confidence interval for instructions %-change: -6.17% -5.43%
Instructions are helped.
total cycles in shared programs: 851842332 -> 851808165 (<.01%)
cycles in affected programs: 618577 -> 584410 (-5.52%)
helped: 271 / HURT: 0
helped stats (abs) min: 64 max: 540 x̄: 126.08 x̃: 111
helped stats (rel) min: 2.57% max: 37.97% x̄: 6.12% x̃: 5.06%
95% mean confidence interval for cycles value: -135.35 -116.80
95% mean confidence interval for cycles %-change: -6.67% -5.57%
Cycles are helped.
total sends in shared programs: 1025238 -> 1024308 (-0.09%)
sends in affected programs: 6454 -> 5524 (-14.41%)
helped: 271 / HURT: 0
helped stats (abs) min: 2 max: 8 x̄: 3.43 x̃: 4
helped stats (rel) min: 5.71% max: 25.00% x̄: 14.98% x̃: 17.39%
95% mean confidence interval for sends value: -3.57 -3.29
95% mean confidence interval for sends %-change: -15.42% -14.54%
Sends are helped.
According to Felix DeGrood, this results in a 10% improvement in
the draw call time for certain draw calls from Strange Brigade.
v2: Fix assertions about number of components and add more of them.
Combine the quads and triangles handling as it's nearly identical.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> [v1]
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19061>
Setting the NIR options takes care of iris thanks to the common st/mesa
linking code, and updating brw_nir_link_shaders should handle anv.
The main effort here is updating remap_tess_levels, which needs to
handle vector stores, writemasking, and swizzling. Unfortunately,
we also need to continue handling the existing single-component
access because it's used for TES inputs, which we don't vectorize.
We could try to vectorize TES inputs too, but they're all pushed
anyway, so it wouldn't buy us much other than deleting this code.
Also, we do have opt_combine_stores, but not one for loads.
One limitation of using nir_vectorize_tess_levels is that it works
on variables, and so isn't able to combine outer/inner writes that
happen to live in the same vec4 slot (for triangle domains). That
said, it's still better than before.
For writes, we allow the intrinsics to supply up to the full size
of the variable (vec4 for outer, vec2 for inner) even if the domain
only requires a subset of those components (i.e. triangles needs 3).
shader-db results on Icelake:
total instructions in shared programs: 19605070 -> 19602284 (-0.01%)
instructions in affected programs: 65338 -> 62552 (-4.26%)
helped: 271 / HURT: 0
helped stats (abs) min: 6 max: 24 x̄: 10.28 x̃: 12
helped stats (rel) min: 1.30% max: 18.18% x̄: 5.80% x̃: 7.59%
95% mean confidence interval for instructions value: -10.71 -9.85
95% mean confidence interval for instructions %-change: -6.17% -5.43%
Instructions are helped.
total cycles in shared programs: 851854659 -> 851820320 (<.01%)
cycles in affected programs: 618749 -> 584410 (-5.55%)
helped: 271 / HURT: 0
helped stats (abs) min: 69 max: 540 x̄: 126.71 x̃: 108
helped stats (rel) min: 2.57% max: 37.97% x̄: 6.17% x̃: 5.06%
95% mean confidence interval for cycles value: -135.89 -117.54
95% mean confidence interval for cycles %-change: -6.72% -5.63%
Cycles are helped.
total sends in shared programs: 1025285 -> 1024355 (-0.09%)
sends in affected programs: 6454 -> 5524 (-14.41%)
helped: 271 / HURT: 0
helped stats (abs) min: 2 max: 8 x̄: 3.43 x̃: 4
helped stats (rel) min: 5.71% max: 25.00% x̄: 14.98% x̃: 17.39%
95% mean confidence interval for sends value: -3.57 -3.29
95% mean confidence interval for sends %-change: -15.42% -14.54%
Sends are helped.
According to Felix DeGrood, this results in a 10% improvement in
the draw call time for certain draw calls from Strange Brigade.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17944>
With this option enabled range of input values for fsin and fcos is
limited to [-2*pi : 2*pi] by calculating the reminder after 2*pi modulo
division. This helps to improve calculation precision for large input
arguments on Intel.
-v2: Add limit_trig_input_range option to prog_key to update shader
cache (Lionel)
Signed-off-by: Vadym Shovkoplias <vadym.shovkoplias@globallogic.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16388>
This controls the whole lowering of "make tex ops with implicit
derivatives on non-implicit-derivative stages be tex ops with an explicit
lod of 0 instead", but it's really hard to describe that in a git commit
summary.
All existing callers get it added except:
- nir_to_tgsi which didn't want it.
- nouveau, which didn't want it (fixes regressions in shadowcube and
shadow2darray with NIR, since the shading languages don't expose txl of
those sampler types and thus it's not supported in HW)
- optional lowering passes in mesa/st (lower_rect, YUV lowering, etc)
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16156>
DG2 can only do sample_d and sample_d_c on 1D and 2D surfaces. Cube
maps and 3D surfaces were already handled, but 1D array and 2D array
surfaces were not.
Fixes the following Vulkan CTS failures on DG2:
dEQP-VK.glsl.texture_functions.texturegradclamp.isampler1darray_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.isampler2darray_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler1darray_fixed_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler1darray_float_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler2darray_fixed_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.sampler2darray_float_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.usampler1darray_fragment
dEQP-VK.glsl.texture_functions.texturegradclamp.usampler2darray_fragment
The Fixes: tag below is a bit misleading. This commit adds another
lowering, similar to the one in the Fixes: commit, that probably should
have been added at the same time. I just want to make sure this commit
gets applied everywhere that commit was also applied.
Fixes: 635ed58e52 ("intel/compiler: Lower txd for 3D samplers on XeHP.")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15681>
In general, an atomic intrinsic may perform separate atomics for every
enabled SIMD channel, as each channel may operate on different memory.
However, an extremely common case is for all channels to access the same
memory location. In this case, we can simply perform a reduction/scan
across the subgroup, and perform one atomic for the whole subgroup,
rather than one per channel. For example, if an intrinsic says to take
the minimum value of the existing memory and the value in each channel,
we can do a thread-local minimum of all enabled channels, then do a
single atomic to take the minimum of that and the existing memory.
Our hardware doesn't optimize the case where multiple channels ask for
atomics on the same memory location; it assumes the compiler will do so.
nir_opt_uniform_atomics() uses divergence analysis to detect this case,
adds the necessary subgroup operations, and moves the atomic inside a
conditional that disables all but a single invocation. It even detects
cases where the shader code already performs this kind of optimization,
and avoids doing it a second time.
This may not be the optimal solution for us. In the backend, we could
detect this case and emit send(1) instructions with NoMask, rather than
generating if...send(16)...endif, and a lot of unnecessary ALU ops. But
it's simple to do, reuses the same path as ACO, and still provides most
of the benefit by cutting up to 16x atomics down to a single atomic,
which is more merciful to the memory bus.
Improves performance of Shadow of the Tomb Raider by 5.5% on XeHP.
Improves performance of a customer-internal benchmark on XeHP at
3840x2160 and low settings by approximately 30%.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15484>
