AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
149,116 Commits 1 Branch 0 Tags
14b82270837c331adc30bf2a747b3cb720917d67
Commit Graph

3503 Commits

Author SHA1 Message Date
Rhys Perry 14b8227083 nir: add some missing nir_alu_type_get_base_type 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13436>
 2022-01-20 22:54:42 +00:00
Rhys Perry f2fbba7920 nir/algebraic: optimize open-coded fmulz/ffmaz 
This pattern will be found in future versions of D3D9 DXVK.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13436>
 2022-01-20 22:54:42 +00:00
Rhys Perry 312a284980 nir/algebraic: add ignore_exact() wrapper 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13436>
 2022-01-20 22:54:42 +00:00
Rhys Perry 7f05ea3793 nir: add nir_op_fmulz and nir_op_ffmaz 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13436>
 2022-01-20 22:54:42 +00:00
Emma Anholt cac6f633b2 nir/opt_offsets: Use nir_ssa_scalar to chase offset additions. 
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>
 2022-01-19 22:28:34 +00:00
Emma Anholt 645ca56425 nir/opt_offsets: Also apply the max offset to top-level constant folding. 
nir_to_tgsi wants this for disabling folding into shared var accesses at
all.

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>
 2022-01-19 22:28:34 +00:00
Emma Anholt ec4b9909f0 nir/opt_offsets: Disable unsigned wrap checks on non-native-integers HW. 
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>
 2022-01-19 22:28:34 +00:00
Emma Anholt 700d2fbd0a nir: Add a .base field to nir_load_ubo_vec4. 
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>
 2022-01-19 22:28:34 +00:00
Dave Airlie ccbf700d6c nir: remove gl.h include from nir headers. 
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>
 2022-01-19 21:54:58 +00:00
Dave Airlie 1352e0ba0c mesa/*: add a shader primitive type to get away from GL types. 
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>
 2022-01-19 21:54:58 +00:00
Connor Abbott 9c9e8c3349 nir: Reorder ffma and fsub combining 
It's relatively common to do something like "a * b - c", which on most
GPUs can be implemented in a single instruction. Before
opt_algebraic_late this will be something like
"fadd(fmul(a, b), fneg(c))", and we want to turn it info
"ffma(a, b, fneg(c))". But because the fsub pattern was first we
instead turned it into "fsub(fmul(a, b), c)". Fix this by reordering
them.

Selected shader-db results on freedreno:

total instructions in shared programs: 1561330 -> 1551619 (-0.62%)
instructions in affected programs: 780272 -> 770561 (-1.24%)
helped: 1941
HURT: 491
helped stats (abs) min: 1 max: 147 x̄: 7.98 x̃: 4
helped stats (rel) min: 0.07% max: 30.77% x̄: 4.36% x̃: 3.17%
HURT stats (abs)   min: 1 max: 307 x̄: 11.76 x̃: 5
HURT stats (rel)   min: 0.09% max: 18.71% x̄: 2.26% x̃: 1.38%
95% mean confidence interval for instructions value: -4.57 -3.41
95% mean confidence interval for instructions %-change: -3.21% -2.84%
Instructions are helped.

total nops in shared programs: 358926 -> 356263 (-0.74%)
nops in affected programs: 167116 -> 164453 (-1.59%)
helped: 1395
HURT: 859
helped stats (abs) min: 1 max: 108 x̄: 6.80 x̃: 3
helped stats (rel) min: 0.17% max: 100.00% x̄: 19.15% x̃: 10.57%
HURT stats (abs)   min: 1 max: 307 x̄: 7.95 x̃: 3
HURT stats (rel)   min: 0.00% max: 381.82% x̄: 20.04% x̃: 10.00%
95% mean confidence interval for nops value: -1.77 -0.59
95% mean confidence interval for nops %-change: -5.55% -2.87%
Nops are helped.

total non-nops in shared programs: 1202404 -> 1195356 (-0.59%)
non-nops in affected programs: 496682 -> 489634 (-1.42%)
helped: 1951
HURT: 265
helped stats (abs) min: 1 max: 39 x̄: 4.02 x̃: 3
helped stats (rel) min: 0.07% max: 15.38% x̄: 2.97% x̃: 1.96%
HURT stats (abs)   min: 1 max: 22 x̄: 2.97 x̃: 2
HURT stats (rel)   min: 0.05% max: 10.00% x̄: 1.14% x̃: 0.75%
95% mean confidence interval for non-nops value: -3.38 -2.99
95% mean confidence interval for non-nops %-change: -2.60% -2.36%
Non-nops are helped.

total systall in shared programs: 288317 -> 292975 (1.62%)
systall in affected programs: 87876 -> 92534 (5.30%)
helped: 388
HURT: 431
helped stats (abs) min: 1 max: 214 x̄: 14.39 x̃: 8
helped stats (rel) min: 0.25% max: 100.00% x̄: 22.12% x̃: 11.96%
HURT stats (abs)   min: 1 max: 232 x̄: 23.77 x̃: 7
HURT stats (rel)   min: 0.00% max: 1300.00% x̄: 51.71% x̃: 17.30%
95% mean confidence interval for systall value: 3.07 8.30
95% mean confidence interval for systall %-change: 9.49% 23.97%
Systall are HURT.

(The systall hurt is probably just due to having having fewer
instructions to hide latency with.)

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14554>
 2022-01-18 17:44:50 +00:00
Qiang Yu 2cee73f0f7 nir: fix nir_tex_instr hash not count is_sparse field 
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>
 2022-01-18 16:10:35 +08:00
Rhys Perry d95a0b52e4 nir/unsigned_upper_bound: don't follow 64-bit f2u32() 
Fixes Doom Eternal crash.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes: 72ac3f6026 ("nir: add nir_unsigned_upper_bound and nir_addition_might_overflow")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14555>
 2022-01-17 10:59:21 +00:00
Emma Anholt f6ffefba3e nir: Apply nir_opt_offsets to nir_intrinsic_load_uniform as well. 
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>
 2022-01-16 19:11:29 +00:00
Emma Anholt b024102d7c freedreno/ir3: Use nir_opt_offset for removing constant adds for shared vars. 
Saves some work in carchase and manhattan31:

instructions in affected programs: 2842 -> 2818 (-0.84%)
nops in affected programs: 1131 -> 1105 (-2.30%)
non-nops in affected programs: 1236 -> 1238 (0.16%)
mov in affected programs: 57 -> 61 (7.02%)
dwords in affected programs: 2144 -> 2150 (0.28%)
cat0 in affected programs: 1195 -> 1169 (-2.18%)
cat1 in affected programs: 151 -> 155 (2.65%)
cat2 in affected programs: 142 -> 140 (-1.41%)
sstall in affected programs: 190 -> 178 (-6.32%)
(ss) in affected programs: 63 -> 63 (0.00%)
systall in affected programs: 532 -> 511 (-3.95%)

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14023>
 2022-01-16 19:11:29 +00:00
Daniel Schürmann 79a987ad2a nir/opt_if: also merge break statements with ones after the branch 
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>
 2022-01-13 02:30:32 +00:00
Daniel Schürmann dad609d152 nir/opt_if: merge two break statements from both branch legs 
This optimization turns

     loop {
        ...
        if (cond) {
           do_work_1();
           break;
        } else {
           do_work_2();
           break;
        }
     }

 into:

     loop {
        ...
        if (cond) {
           do_work_1();
        } else {
           do_work_2();
        }
        break;
     }

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7587>
 2022-01-13 02:30:32 +00:00
Daniel Schürmann 8a78706643 nir: refactor nir_opt_move 
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>
 2022-01-12 13:41:54 +00:00
Marcin Ślusarz f286ecf906 nir: handle per-view clip/cull distances 
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14263>
 2022-01-11 22:45:23 +00:00
Marcin Ślusarz 0d6f83cbf1 nir: remove invalid assert affecting per-view variables 
per-view variables can have arbitrary (but > 0) number of array levels

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14263>
 2022-01-11 22:45:23 +00:00
Marcin Ślusarz 4fed440724 nir: add load_mesh_view_count and load_mesh_view_indices intrinsics 
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14263>
 2022-01-11 22:45:23 +00:00
Rhys Perry 67fc7a1763 nir/uniform_atomics: fix is_atomic_already_optimized without workgroups 
dims_needed would have been zero, so this would always returned true for
non-compute stages.

Also fix this for variable workgroup sizes.

Improves Shadow of the Tomb Raider RX 6800 performance by 10.6%, 11.5% and
4.5% (day_of_dead, jungle and paititi scenes).

radv_perf before and after:
{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'day_of_dead', 'avg_fps': '62.913333333333334', 'min_fps': '62.81', 'max_fps': '62.98', 'interations': '3'}
{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'jungle', 'avg_fps': '64.02666666666666', 'min_fps': '63.93', 'max_fps': '64.11', 'interations': '3'}
{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'paititi', 'avg_fps': '74.81666666666666', 'min_fps': '74.72', 'max_fps': '74.88', 'interations': '3'}

{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'day_of_dead', 'avg_fps': '69.57', 'min_fps': '69.52', 'max_fps': '69.63', 'interations': '3'}
{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'jungle', 'avg_fps': '71.41000000000001', 'min_fps': '71.31', 'max_fps': '71.5', 'interations': '3'}
{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'paititi', 'avg_fps': '78.16666666666667', 'min_fps': '78.07', 'max_fps': '78.23', 'interations': '3'}

Performance now seems slightly better than AMDVLK 2021.Q4.3:
{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'day_of_dead', 'avg_fps': '68.02666666666666', 'min_fps': '67.95', 'max_fps': '68.16', 'interations': '3'}
{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'jungle', 'avg_fps': '70.24666666666667', 'min_fps': '69.83', 'max_fps': '70.51', 'interations': '3'}
{'app': 'SotTR', 'resolution': '3840x2160', 'preset': 'VeryHigh', 'antialiasing': 'off', 'scene': 'paititi', 'avg_fps': '77.19', 'min_fps': '77.18', 'max_fps': '77.2', 'interations': '3'}

fossil-db (Sienna Cichlid):
Totals from 40 (0.03% of 134621) affected shaders:
CodeSize: 62676 -> 65996 (+5.30%)
Instrs: 11372 -> 12111 (+6.50%)
Latency: 144122 -> 142848 (-0.88%); split: -1.09%, +0.21%
InvThroughput: 19686 -> 19847 (+0.82%); split: -0.06%, +0.87%
VClause: 304 -> 306 (+0.66%)
SClause: 603 -> 604 (+0.17%); split: -0.83%, +1.00%
Copies: 780 -> 858 (+10.00%)
Branches: 235 -> 329 (+40.00%)
PreSGPRs: 1072 -> 1083 (+1.03%); split: -0.37%, +1.40%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14407>
 2022-01-10 19:57:38 +00:00
Rhys Perry b00138090e nir/lower_shader_calls: fix store_scratch write_mask 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14447>
 2022-01-10 19:01:04 +00:00
Danylo Piliaiev b8d486f298 nir/algebraic: Separate has_dot_4x8 into has_sdot_4x8 and has_udot_4x8 
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>
 2022-01-10 13:20:39 +02:00
Gert Wollny 6f348d9c99 nir_lower_io: propagate the "invariant" flag to outputs 
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>
 2022-01-07 16:35:43 +00:00
Emma Anholt 558a600629 nir_to_tgsi: Enable fdot_replicates flag. 
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>
 2022-01-07 09:58:24 +00:00
Jesse Natalie c09c0b351f nir_opt_dead_cf: Remove dead ifs 
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>
 2022-01-07 05:15:48 +00:00
Alyssa Rosenzweig 24ea7cbb06 nir: Extend store_combined_output_pan 
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>
 2022-01-02 01:12:05 +00:00
Alyssa Rosenzweig 5c168f09eb nir: Eliminate store_combined_output_pan BASE 
It's meaningless for this intrinsic and is just adding noise to the
lowering pass.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13714>
 2022-01-02 01:12:05 +00:00
Daniel Schürmann 17ecd0b31a nir/opt_algebraic: lower fneg_hi/lo to fmul 
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>
 2021-12-21 13:23:37 +01:00
Caio Oliveira 729df14e45 nir: Handle volatile semantics for loading HelperInvocation builtin 
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>
 2021-12-17 16:37:14 -08:00
Rhys Perry a65285f54b nir/opt_access: infer CAN_REORDER for global access 
fossil-db (Sienna Cichlid):
Totals from 352 (0.26% of 134621) affected shaders:
VGPRs: 17240 -> 17272 (+0.19%)
CodeSize: 1753640 -> 1755744 (+0.12%); split: -0.04%, +0.16%
Instrs: 323190 -> 323801 (+0.19%); split: -0.03%, +0.22%
Latency: 3241205 -> 3241293 (+0.00%); split: -0.10%, +0.10%
InvThroughput: 568927 -> 568067 (-0.15%); split: -0.16%, +0.00%
SClause: 12109 -> 10444 (-13.75%); split: -13.76%, +0.01%
Copies: 27802 -> 27717 (-0.31%); split: -0.56%, +0.26%
PreSGPRs: 14699 -> 14690 (-0.06%)
PreVGPRs: 15793 -> 15799 (+0.04%)

fossil-db (Polaris10):
Totals from 348 (0.26% of 135668) affected shaders:
SGPRs: 21446 -> 21574 (+0.60%); split: -0.15%, +0.75%
VGPRs: 17004 -> 16996 (-0.05%); split: -0.09%, +0.05%
CodeSize: 1782796 -> 1783060 (+0.01%); split: -0.03%, +0.05%
Instrs: 337828 -> 337921 (+0.03%); split: -0.03%, +0.06%
Latency: 3726328 -> 3726721 (+0.01%); split: -0.09%, +0.10%
InvThroughput: 1307917 -> 1299841 (-0.62%); split: -0.62%, +0.00%
VClause: 4327 -> 4337 (+0.23%); split: -0.09%, +0.32%
SClause: 12178 -> 10529 (-13.54%); split: -13.55%, +0.01%
Copies: 40227 -> 40244 (+0.04%); split: -0.19%, +0.24%
PreSGPRs: 14946 -> 14937 (-0.06%)
PreVGPRs: 15637 -> 15643 (+0.04%)

fossil-db (Pitcairn):
Totals from 351 (0.26% of 135668) affected shaders:
SGPRs: 20382 -> 20619 (+1.16%); split: -0.79%, +1.95%
CodeSize: 1789732 -> 1789836 (+0.01%); split: -0.04%, +0.04%
MaxWaves: 1947 -> 1949 (+0.10%)
Instrs: 352274 -> 352318 (+0.01%); split: -0.04%, +0.06%
Latency: 4057829 -> 4058226 (+0.01%); split: -0.08%, +0.09%
InvThroughput: 1332245 -> 1317578 (-1.10%); split: -1.11%, +0.01%
VClause: 8581 -> 8583 (+0.02%); split: -0.13%, +0.15%
SClause: 12187 -> 10552 (-13.42%); split: -13.43%, +0.02%
Copies: 44906 -> 44915 (+0.02%); split: -0.24%, +0.26%
PreSGPRs: 16571 -> 16562 (-0.05%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14227>
 2021-12-17 18:51:24 +00:00
Rhys Perry 403ae3b48e nir/algebraic: optimize more 64-bit imul with constant source 
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>
 2021-12-17 18:51:24 +00:00
Rhys Perry c56cf157c5 nir/opt_load_store_vectorize: improve ssbo/global alias analysis 
If either the global access or the ssbo access is restrict, they shouldn't
alias.

No fossil-db changes.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14227>
 2021-12-17 18:51:24 +00:00
Jason Ekstrand deec7a590b anv,nir: Use sample_pos_or_center in lower_wpos_center 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14198>
 2021-12-17 16:02:16 +00:00
Jason Ekstrand e8acc5a7ea nir: Add a new sample_pos_or_center system value 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14198>
 2021-12-17 16:02:16 +00:00
Marcin Ślusarz 504e5cb4e8 nir/print: print const value near each use of const ssa variable 
Without/with NIR_DEBUG=print,print_const:

-vec4 32 ssa_60 = fadd ssa_59, ssa_58
+vec4 32 ssa_60 = fadd ssa_59 /*(0xbf800000, 0x3e800000, 0x00000000, 0x3f800000) = (-1.000000, 0.250000, 0.000000, 1.000000)*/, ssa_58

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13880>
 2021-12-17 10:04:50 +00:00
Marcin Ślusarz 23f8f836e0 nir/print: group hex and float vectors together 
Vectors are much easier to follow in this format, because developer cares
either about hex or float values, never both.

Before/after:

-vec4 32 ssa_222 = load_const (0x00000000 /* 0.000000 */, 0x00000000 /* 0.000000 */, 0x3f800000 /* 1.000000 */, 0x3f800000 /* 1.000000 */)
+vec4 32 ssa_222 = load_const (0x00000000, 0x00000000, 0x3f800000, 0x3f800000) = (0.000000, 0.000000, 1.000000, 1.000000)

-vec1 32 ssa_174 = load_const (0xbf800000 /* -1.000000 */)
+vec1 32 ssa_174 = load_const (0xbf800000 = -1.000000)

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13880>
 2021-12-17 10:04:50 +00:00
Marcin Ślusarz d2b4051ea9 nir/print: move print_load_const_instr up 
... to avoid forward declarations in future commit

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13880>
 2021-12-17 10:04:50 +00:00
Marcin Ślusarz f7e63ec5d8 nir/print: compact printing of intrinsic indices 
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14222>
 2021-12-16 09:43:13 +00:00
Marcin Ślusarz d8fa625bb3 nir/print: expand printing of io semantics.gs_streams 
gs_streams can be set for at least 2 other intrinsics.

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14222>
 2021-12-16 09:43:13 +00:00
Marcin Ślusarz be25db9f0f nir/print: simplify printing of IO semantics 
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>
 2021-12-16 09:43:13 +00:00
Caio Oliveira b1156f23a2 Revert "nir: disable a NIR test due to undebuggable & locally unreproducible CI failures" 
This reverts commit 6eb3fe2d4f.  The root cause was
a bug in Meson when using the new gtest protocol and a test failed before producing
the XML file expected by it.  This was fixed in later versions of Meson, so
we've bumped the required meson version to use that feature.  The failure should
now be properly identified, so re-enabling the NIR test.

Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14204>
 2021-12-15 23:28:09 +00:00
Caio Oliveira dcc7b19cae nir: Initialize nir_register::divergent 
Fixes: c7fc44f9eb ("nir/from_ssa: Respect and populate divergence information")
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14205>
 2021-12-15 22:39:06 +00:00
Juan A. Suarez Romero b8f6685bb5 nir: use call_once() to init debug variable 
For data-race safety, let's use this function to ensure NIR debug is
initialized only once.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14057>
 2021-12-14 08:01:17 +00:00
Juan A. Suarez Romero 18c039b2e1 tgsi-to-nir: initialize NIR_DEBUG envvar 
This envvar is initialized when creating a NIR shader, but it needs to
be used before. So initialize it here.

v2 (Juan):
 - Use static variable for first initialization.

Fixes: f77ccdfb4a ("nir: add NIR_DEBUG envvar")
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14057>
 2021-12-14 08:01:17 +00:00
Jordan Justen 211e0606c7 nir/lower_tex: Add filter for tex offset lowering 
Rework:
 * Add callback_data (s-b Jason)

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14142>
 2021-12-13 16:56:23 -08:00
Samuel Pitoiset be53b3d1bf nir/lower_tex: add lower_lod_zero_width 
On AMD, the hardware will return 0 for the raw LOD if the sum of the
absolute values of derivatives is 0 but Vulkan expects the value to
be in the [-inf, -22.0f] range.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14147>
 2021-12-13 10:00:07 +00:00
Marcin Ślusarz 87f03b1662 nir: limit lower_clip_cull_distance_arrays input to traditional stages 
Compute, task, mesh & raytracing stages don't support
ClipDistance/CullDistance as input.

This change is not needed for correctness. Just something I stumbled on.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14149>
 2021-12-13 08:32:23 +00:00
Marek Olšák 2785141c16 nir: add nir_has_divergent_loop function 
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13966>
 2021-12-11 20:07:35 +00:00