Replace load_mesh_global_arg_addr_intel with a more general intrinsic
load_mesh_inline_data_intel, since inline data now hold both
a pointer descriptor information and the first few push constants.
Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14788>
Use texture_index if there is no deref src.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14308>
This pass was originally written for d3d12, but is useful for hardware
that lacks sample compare support like some etnaviv GPU models.
Also rename the lowering pass and some surrounding code to
nir_lower_tex_shadow as suggested by Emma.
I'd like to use the pass that's already in tree.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14308>
I noticed the inefficiency in NIR-to-TGSI output while trying to debug a
failure handling some arrays in r600. While this makes reading CTS
shaders easier, the effect in the real world is pretty limited. From
softpipe shader-db:
total instructions in shared programs: 2929840 -> 2929836 (<.01%)
instructions in affected programs: 118 -> 114 (-3.39%)
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14321>
Zero isn't really a valid write mask. If it's provided, use a full write
mask.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14455>
For nir_to_tgsi, I want to be able to fold into the base from a vector
load_const, which the ad-hoc scalar chasing couldn't handle.
r300:
total instructions in shared programs: 1278731 -> 1256502 (-1.74%)
instructions in affected programs: 457909 -> 435680 (-4.85%)
total flowcontrol in shared programs: 8316 -> 8313 (-0.04%)
flowcontrol in affected programs: 5 -> 2 (-60.00%)
total temps in shared programs: 213687 -> 213774 (0.04%)
temps in affected programs: 13140 -> 13227 (0.66%)
total consts in shared programs: 952850 -> 949929 (-0.31%)
consts in affected programs: 386352 -> 383431 (-0.76%)
Fixes: #5781
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14309>
Since we don't have 32-bit ints, these checks for 32-bit unsigned wrapping
don't help and just reduce optimization opportunities (particularly for
DX9 addressing math).
Doesn't affect any current consumers.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14309>
This lets nir-to-tgsi fold the constant offset of addressing calculations
into the CONST[] reference, which is important for D3D9-era compatibility:
HW of that age has limited uniform space, and if we do the addressing math
as math in the shader for dynamic indexing, the nir_load_consts end up
taking up uniforms we don't have available.
r300:
total instructions in shared programs: 1279699 -> 1279167 (-0.04%)
instructions in affected programs: 134796 -> 134264 (-0.39%)
total instructions in shared programs: 1279699 -> 1279167 (-0.04%)
instructions in affected programs: 134796 -> 134264 (-0.39%)
total temps in shared programs: 213912 -> 213736 (-0.08%)
temps in affected programs: 2166 -> 1990 (-8.13%)
total consts in shared programs: 953237 -> 952973 (-0.03%)
consts in affected programs: 45980 -> 45716 (-0.57%)
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14309>
This saves a lot of pointless gl.h includes across the board,
it moves the one place that needs GLenum into a separate file
only used in those passes that require it.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14605>
This creates an internal shader_prim enum, I've fixed up most
users to use it instead of GL types.
don't store the enum in shader_info as it changes size, and confuses
other things.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14605>
This fixes nir_opt_cse miss replace a non-sparse tex instruction
with a sparse tex instruction and fail the nir_validate_shader().
Fixes: 3a7972f72a ("nir,spirv: add sparse texture fetches")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14362>
Doing this for ir3 required adding a struct for limits of how much base to
fold in (which NTT wants as well for its case of shared vars), otherwise
the later work to lower to the 1<<9 word limit would emit more
instructions.
The shader-db results are that sometimes the reduction in NIR instruction
count results in the fewer sampler prefetches due to the shader being
estimated to be shorter (dota2, nexuiz):
total instructions in shared programs: 8996651 -> 8996776 (<.01%)
total cat5 in shared programs: 86561 -> 86577 (0.02%)
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14023>
This optimizations turns
loop {
...
if (cond1) {
if (cond2) {
do_work_1();
break;
} else {
do_work_2();
}
do_work_3();
break;
} else {
...
}
}
into:
loop {
...
if (cond1) {
if (cond2) {
do_work_1();
} else {
do_work_2();
do_work_3();
}
break;
} else {
...
}
}
As this optimizations moves code into the NIF statement,
it re-iterates on the branch legs in case of success.
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7587>
This patch is a rewrite of nir_opt_move.
Differently from the previous version, each instruction is checked
if it can be moved downwards and then inserted before the first user
of the definition. The advantage is that less insert operations are
performed, the original order is kept if two movable instructions have
the same first user, and instructions without user in the same block
are moved towards the end.
v2: Only return true if an instruction really changed the position.
Don't care for discards, this will be handled by another MR.
v3: fix self-referring phis and update according to nir_can_move_instr().
v4: use nir_can_move_instr() and nir_instr_ssa_def()
v5: deduplicate some code
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3657>
Adreno GPUs has native instruction for unsigned and mixed dot_4x8 but
not signed dot product.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13986>
Ultimately this is consumed by nir-to-tgsi and needed by virglrenderer
to correctly declare output variables.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14423>
That's how the TGSI math opcodes work.
This lets lower_vec_to_regs coalesce the DP output into the .yzw channels,
giving an impressive shader-db win on softpipe:
total instructions in shared programs: 2929840 -> 2794036 (-4.64%)
instructions in affected programs: 1651438 -> 1515634 (-8.22%)
total temps in shared programs: 372730 -> 332744 (-10.73%)
temps in affected programs: 118151 -> 78165 (-33.84%)
and a minor one on r300:
total instructions in shared programs: 51238 -> 51149 (-0.17%)
instructions in affected programs: 2621 -> 2532 (-3.40%)
total vinst in shared programs: 15655 -> 15618 (-0.24%)
vinst in affected programs: 468 -> 431 (-7.91%)
total temps in shared programs: 9838 -> 9828 (-0.10%)
temps in affected programs: 59 -> 49 (-16.95%)
and a bigger one on i915g:
total instructions in shared programs: 398064 -> 395901 (-0.54%)
instructions in affected programs: 29271 -> 27108 (-7.39%)
total tex_indirect in shared programs: 12261 -> 12233 (-0.23%)
tex_indirect in affected programs: 98 -> 70 (-28.57%)
LOST: 0
GAINED: 5
The r300 change is less impressive because it does some backend copy-prop,
but also because intermediate storage of DPs now takes a vec4 instead of a
scalar.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14200>
An if that looks like:
if (x) { } else { }
That has no phis following it is dead. Currently these are only
removed by peephole select, but that means that 'x' is considered
used until that pass is run, which can make it difficult to apply
sane lowering in the case where loading 'x' requires complex or
expensive transformations, but 'x' is *really* unused.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14400>
Extend store_combined_output_pan to take a dual source blend input in
addition to colour, depth, and stencil inputs. Use the last source for
this, and represent the type with the DEST_TYPE index. This is a hack
but there is no SRC2_TYPE and NIR doesn't seem to mind as long as we
know what we mean. This allows the backend to emit a combined "blend
render target #0" instruction taking two sources.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13714>
This pattern, found in the FSR upscaling shader,
helps the vectorization efforts by keeping the
chain of vectorized instructions intact.
Radeon can optimize it to per-component fneg modifiers.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13688>
SPV_EXT_demote_to_helper_invocation added OpDemoteToHelperInvocation
operation to turn an invocation into a helper invocation, but the
value of HelperInvocation (a builtin from Input storage class)
couldn't be modified dynamically without breaking compatibility.
For the extension the operation OpIsHelperInvocation was added to get
the dynamic value.
For SPIR-V 1.6, the demote operation was promoted, but now to get the
dynamic value the shader must issue a load to HelperInvocation with
Volatile memory access semantics.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14209>
Two 64-bit shifts and an addition are usually faster than the several
multiplications nir_lower_int64 creates.
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14227>
Some of the tested flags are set for other intrinsics and they are
printed only when set, so there's no point in checking exact intrinsic
name or shader stage.
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14222>
