summaryrefslogtreecommitdiff
path: root/src/compiler/nir/nir_range_analysis.c
Commit message (Collapse)AuthorAgeFilesLines
...
* nir/range_analysis: Make sure the table validation only occurs onceIan Romanick2019-11-221-38/+58
| | | | | | | | | | | | | All of the tables are static const, so they only need to be validated once. As noted in the previous commit, the compiler should be able to eliminate all of this code when the assertions would pass. Even with the help of the previous commit, this does not always occur. -Og: -95.688 +/- 3.91935 (-24.9562% +/- 1.0222%) N=5 -O1: No difference proven at 95.0% confidence. N=5 -O2: -1.962 +/- 0.85001 (-0.860013% +/- 0.372589%) N=5 Reviewed-by: Eric Anholt <eric@anholt.net>
* nir/range-analysis: Add pragmas to help loop unrollingIan Romanick2019-11-221-0/+10
| | | | | | | | | | | | | | | | | | | | I was pretty liberal with these assertions when I wrote this code because I had assumed that GCC would unroll the loops, inline the look ups of static const arrays with now constant indices, and then elmininate all the actuall assertions. It seems none of this happens even at -O3. Adding the pragmas helps encourage loop unrolling at some optimization levels. I tested by running shader-db with NIR_VALIDATE=false on a Core i7 Haswell desktop system. -Og: No difference proven at 95.0% confidence. N=5 -O1: -48.304 +/- 1.221 (-16.3343% +/- 0.412888%) N=5 -O2: -49.94 +/- 1.23521 (-17.9634% +/- 0.444303%) N=5 v2: Add a _Pragma to an inner loop that was accidentally dropped during a rebase. Reviewed-by: Eric Anholt <eric@anholt.net>
* nir: Keep the range analysis HT around intra-pass until we make a change.Eric Anholt2019-10-041-9/+4
| | | | | | | | | This lets us memoize range analysis work across instructions. Reduces runtime of shader-db on Intel by -30.0288% +/- 2.1693% (n=3). Fixes: 405de7ccb6cb ("nir/range-analysis: Rudimentary value range analysis pass") Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
* nir/range-analysis: Use types to provide better ranges from bcsel and movIan Romanick2019-09-251-25/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> All Gen7+ platforms had similar results. (Ice Lake shown) total instructions in shared programs: 16328255 -> 16315391 (-0.08%) instructions in affected programs: 218318 -> 205454 (-5.89%) helped: 988 HURT: 0 helped stats (abs) min: 1 max: 72 x̄: 13.02 x̃: 10 helped stats (rel) min: 0.33% max: 16.04% x̄: 6.27% x̃: 4.88% 95% mean confidence interval for instructions value: -13.69 -12.35 95% mean confidence interval for instructions %-change: -6.55% -5.99% Instructions are helped. total cycles in shared programs: 363683977 -> 363615417 (-0.02%) cycles in affected programs: 1475193 -> 1406633 (-4.65%) helped: 923 HURT: 36 helped stats (abs) min: 1 max: 624 x̄: 75.78 x̃: 48 helped stats (rel) min: 0.08% max: 13.89% x̄: 5.20% x̃: 5.08% HURT stats (abs) min: 1 max: 179 x̄: 38.58 x̃: 4 HURT stats (rel) min: 0.06% max: 16.56% x̄: 3.33% x̃: 0.29% 95% mean confidence interval for cycles value: -75.88 -67.10 95% mean confidence interval for cycles %-change: -5.10% -4.66% Cycles are helped. Sandy Bridge total instructions in shared programs: 10785779 -> 10785654 (<.01%) instructions in affected programs: 13855 -> 13730 (-0.90%) helped: 67 HURT: 0 helped stats (abs) min: 1 max: 15 x̄: 1.87 x̃: 1 helped stats (rel) min: 0.20% max: 3.45% x̄: 0.97% x̃: 0.78% 95% mean confidence interval for instructions value: -2.47 -1.26 95% mean confidence interval for instructions %-change: -1.13% -0.81% Instructions are helped. total cycles in shared programs: 153704799 -> 153704481 (<.01%) cycles in affected programs: 101509 -> 101191 (-0.31%) helped: 38 HURT: 13 helped stats (abs) min: 1 max: 38 x̄: 12.53 x̃: 16 helped stats (rel) min: 0.07% max: 2.69% x̄: 0.87% x̃: 0.53% HURT stats (abs) min: 1 max: 36 x̄: 12.15 x̃: 7 HURT stats (rel) min: 0.06% max: 2.53% x̄: 0.73% x̃: 0.44% 95% mean confidence interval for cycles value: -10.24 -2.24 95% mean confidence interval for cycles %-change: -0.75% -0.17% Cycles are helped. LOST: 2 GAINED: 0 No shader-db change on Iron Lake or GM45.
* nir/range-analysis: Use types in the hash keyIan Romanick2019-09-251-38/+98
| | | | | | | This allows the reslut of mov and bcsel to be separately interpreted as float or int depending on the use. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
* nir/range-analysis: Bail if the types don't matchIan Romanick2019-09-251-0/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Some shaders are hurt by this change because now a load_const(0x00000000) is not recognized as eq_zero when loaded as a float. This behavior is restored in a later patch (nir/range-analysis: Use types to provide better ranges from bcsel and mov). v2: Add a comment about reinterpretation of int/uint/bool. Suggested by Caio. Rewrite condition the check for types being float versus checking for types not being all the things that aren't float. Fixes: 405de7ccb6c ("nir/range-analysis: Rudimentary value range analysis pass") Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> All Gen7+ platforms had similar results. (Ice Lake shown) total instructions in shared programs: 16327543 -> 16328255 (<.01%) instructions in affected programs: 55928 -> 56640 (1.27%) helped: 0 HURT: 208 HURT stats (abs) min: 1 max: 16 x̄: 3.42 x̃: 3 HURT stats (rel) min: 0.33% max: 6.74% x̄: 1.31% x̃: 1.12% 95% mean confidence interval for instructions value: 3.06 3.79 95% mean confidence interval for instructions %-change: 1.17% 1.46% Instructions are HURT. total cycles in shared programs: 363682759 -> 363683977 (<.01%) cycles in affected programs: 325758 -> 326976 (0.37%) helped: 44 HURT: 133 helped stats (abs) min: 1 max: 179 x̄: 33.61 x̃: 5 helped stats (rel) min: 0.06% max: 14.21% x̄: 2.47% x̃: 0.29% HURT stats (abs) min: 1 max: 157 x̄: 20.28 x̃: 14 HURT stats (rel) min: 0.07% max: 14.44% x̄: 1.42% x̃: 0.73% 95% mean confidence interval for cycles value: 0.38 13.39 95% mean confidence interval for cycles %-change: -0.06% 0.96% Inconclusive result (%-change mean confidence interval includes 0). Sandy Bridge total instructions in shared programs: 10787433 -> 10787443 (<.01%) instructions in affected programs: 1842 -> 1852 (0.54%) helped: 0 HURT: 10 HURT stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1 HURT stats (rel) min: 0.33% max: 1.85% x̄: 0.73% x̃: 0.49% 95% mean confidence interval for instructions value: 1.00 1.00 95% mean confidence interval for instructions %-change: 0.36% 1.10% Instructions are HURT. total cycles in shared programs: 153724543 -> 153724563 (<.01%) cycles in affected programs: 8407 -> 8427 (0.24%) helped: 1 HURT: 3 helped stats (abs) min: 18 max: 18 x̄: 18.00 x̃: 18 helped stats (rel) min: 0.98% max: 0.98% x̄: 0.98% x̃: 0.98% HURT stats (abs) min: 4 max: 18 x̄: 12.67 x̃: 16 HURT stats (rel) min: 0.21% max: 0.75% x̄: 0.56% x̃: 0.72% 95% mean confidence interval for cycles value: -21.31 31.31 95% mean confidence interval for cycles %-change: -1.11% 1.46% Inconclusive result (value mean confidence interval includes 0). No shader-db changes on Iron Lake or GM45.
* nir: do not assume that the result of fexp2(a) is always an integralSamuel Pitoiset2019-09-021-0/+1
| | | | | | | | | It's only correct when 'a' is an integral greater or equal to 0. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111493 Fixes: 5544b2cbbd2 ("nir/algebraic: Use value range analysis to eliminate useless unary ops") Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
* nir/range-analysis: Add a lot more assertions about the contents of tablesIan Romanick2019-08-291-6/+128
| | | | | | | | | v2: Update several of the comments. Drop some redundant uses of ASSERT_UNION_OF_OTHERS_MATCHES_UNKNOWN_*_SOURCE source. Suggested by Caio. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> Suggested-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
* nir/range-analysis: Range tracking for fpowIan Romanick2019-08-291-0/+66
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | One shader from Metro Last Light and the rest from Rochard. In the Rochard cases, something like: min(1.0, max(pow(saturate(x), y), z)) was transformed to saturate(max(pow(saturate(x), y), z)) because the result of the pow must be >= 0. The Metro Last Light case was similar. An instance of min(pow(abs(x), y), 1.0) became saturate(pow(abs(x), y)) v2: Fix some comments. Suggested by Caio. v3: Fix setting is_intgral when the exponent might be negative. See also Mesa MR !1778. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> All Intel platforms had similar results. (Ice Lake shown) total instructions in shared programs: 16280670 -> 16280659 (<.01%) instructions in affected programs: 1130 -> 1119 (-0.97%) helped: 11 HURT: 0 helped stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1 helped stats (rel) min: 0.72% max: 1.43% x̄: 1.03% x̃: 0.97% 95% mean confidence interval for instructions value: -1.00 -1.00 95% mean confidence interval for instructions %-change: -1.19% -0.86% Instructions are helped. total cycles in shared programs: 367168430 -> 367168270 (<.01%) cycles in affected programs: 10281 -> 10121 (-1.56%) helped: 10 HURT: 1 helped stats (abs) min: 16 max: 18 x̄: 17.00 x̃: 17 helped stats (rel) min: 1.31% max: 2.43% x̄: 1.79% x̃: 1.70% HURT stats (abs) min: 10 max: 10 x̄: 10.00 x̃: 10 HURT stats (rel) min: 3.10% max: 3.10% x̄: 3.10% x̃: 3.10% 95% mean confidence interval for cycles value: -20.06 -9.04 95% mean confidence interval for cycles %-change: -2.36% -0.32% Cycles are helped.
* nir/range-analysis: Handle constants in nir_op_mov just like nir_op_bcselIan Romanick2019-08-291-2/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I discovered this while looking at a shader that was hurt by some other work I'm doing. When I examined the changes, I was confused that one instance of a comparison that was used in a discard_if was (incorrectly) eliminated, while another instance used by a bcsel was (correctly) not eliminated. I had to use NIR_PRINT=true to see exactly where things when wrong. A bunch of shaders in Goat Simulator, Dungeon Defenders, Sanctum 2, and Strike Suit Zero were impacted. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> Fixes: 405de7ccb6c ("nir/range-analysis: Rudimentary value range analysis pass") All Intel platforms had similar results. (Ice Lake shown) total instructions in shared programs: 16280659 -> 16281075 (<.01%) instructions in affected programs: 21042 -> 21458 (1.98%) helped: 0 HURT: 136 HURT stats (abs) min: 1 max: 9 x̄: 3.06 x̃: 3 HURT stats (rel) min: 1.16% max: 6.12% x̄: 2.23% x̃: 2.03% 95% mean confidence interval for instructions value: 2.93 3.19 95% mean confidence interval for instructions %-change: 2.08% 2.37% Instructions are HURT. total cycles in shared programs: 367168270 -> 367170313 (<.01%) cycles in affected programs: 172020 -> 174063 (1.19%) helped: 14 HURT: 111 helped stats (abs) min: 2 max: 80 x̄: 21.21 x̃: 9 helped stats (rel) min: 0.10% max: 4.47% x̄: 1.35% x̃: 0.79% HURT stats (abs) min: 2 max: 584 x̄: 21.08 x̃: 5 HURT stats (rel) min: 0.12% max: 17.28% x̄: 1.55% x̃: 0.40% 95% mean confidence interval for cycles value: 5.41 27.28 95% mean confidence interval for cycles %-change: 0.64% 1.81% Cycles are HURT.
* nir/range-analysis: Fix incorrect fadd range result for (ne_zero, ne_zero)Ian Romanick2019-08-291-3/+8
| | | | | | | | | | | | | | | Found by inspection. I tried really, really hard to make a test case that would trigger this problem, but I was unsuccesful. It's very hard to get an instruction to produce a ne_zero result without ne_zero sources. The most plausible way is using bcsel. That proves problematic because bcsel interprets its sources as integers, so it cannot currently be used to "clean" values for floating point instructions. No shader-db changes on any Intel platform. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> Fixes: 405de7ccb6c ("nir/range-analysis: Rudimentary value range analysis pass")
* nir/range-analysis: Adjust result range of multiplication to account for ↵Ian Romanick2019-08-291-31/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | flush-to-zero Fixes piglit tests (new in piglit!110): - fs-underflow-fma-compare-zero.shader_test - fs-underflow-mul-compare-zero.shader_test v2: Add back part of comment accidentally deleted. Noticed by Caio. Remove is_not_zero function as it is no longer used. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111308 Fixes: fa116ce357b ("nir/range-analysis: Range tracking for ffma and flrp") Fixes: 405de7ccb6c ("nir/range-analysis: Rudimentary value range analysis pass") Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> All Gen7+ platforms** had similar results. (Ice Lake shown) total instructions in shared programs: 16278465 -> 16279492 (<.01%) instructions in affected programs: 16765 -> 17792 (6.13%) helped: 0 HURT: 23 HURT stats (abs) min: 7 max: 275 x̄: 44.65 x̃: 8 HURT stats (rel) min: 1.15% max: 17.51% x̄: 4.23% x̃: 1.62% 95% mean confidence interval for instructions value: 9.57 79.74 95% mean confidence interval for instructions %-change: 1.85% 6.61% Instructions are HURT. total cycles in shared programs: 367135159 -> 367154270 (<.01%) cycles in affected programs: 279306 -> 298417 (6.84%) helped: 0 HURT: 23 HURT stats (abs) min: 13 max: 6029 x̄: 830.91 x̃: 54 HURT stats (rel) min: 0.17% max: 45.67% x̄: 7.33% x̃: 0.49% 95% mean confidence interval for cycles value: 100.89 1560.94 95% mean confidence interval for cycles %-change: 0.94% 13.71% Cycles are HURT. total spills in shared programs: 8870 -> 8869 (-0.01%) spills in affected programs: 19 -> 18 (-5.26%) helped: 1 HURT: 0 total fills in shared programs: 21904 -> 21901 (-0.01%) fills in affected programs: 81 -> 78 (-3.70%) helped: 1 HURT: 0 LOST: 0 GAINED: 1 ** On Broadwell, a shader was hurt for spills / fills instead of helped. No changes on any earlier platforms.
* nir/range-analysis: Adjust result range of exp2 to account for flush-to-zeroIan Romanick2019-08-291-2/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes piglit tests (new in piglit!110): - fs-underflow-exp2-compare-zero.shader_test Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111308 Fixes: 405de7ccb6c ("nir/range-analysis: Rudimentary value range analysis pass") Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> Most of the shaders affected are, unsurprisingly, in Unigine Heaven. All Gen6+ platforms had similar results. (Ice Lake shown) total instructions in shared programs: 16278207 -> 16278465 (<.01%) instructions in affected programs: 11374 -> 11632 (2.27%) helped: 0 HURT: 58 HURT stats (abs) min: 2 max: 13 x̄: 4.45 x̃: 4 HURT stats (rel) min: 0.54% max: 4.11% x̄: 2.42% x̃: 2.82% 95% mean confidence interval for instructions value: 3.77 5.13 95% mean confidence interval for instructions %-change: 2.19% 2.64% Instructions are HURT. total cycles in shared programs: 367134284 -> 367135159 (<.01%) cycles in affected programs: 81207 -> 82082 (1.08%) helped: 17 HURT: 36 helped stats (abs) min: 6 max: 356 x̄: 90.35 x̃: 6 helped stats (rel) min: 0.69% max: 21.45% x̄: 5.71% x̃: 0.78% HURT stats (abs) min: 4 max: 235 x̄: 66.97 x̃: 16 HURT stats (rel) min: 0.35% max: 27.58% x̄: 5.34% x̃: 1.09% 95% mean confidence interval for cycles value: -20.36 53.38 95% mean confidence interval for cycles %-change: -1.08% 4.67% Inconclusive result (value mean confidence interval includes 0). No changes on any earlier platforms.
* nir/range-analysis: Fail gracefully on non-SSA sourcesIan Romanick2019-08-141-0/+3
| | | | | Tested-by: Rob Clark <robdclark@chromium.org> Reviewed-by: Eric Anholt <eric@anholt.net>
* nir/range-analysis: Range tracking for ffma and flrpIan Romanick2019-08-051-114/+138
| | | | | | | | | A similar technique could be used for fmin3, fmax3, and fmid3. This could be squashed with the previous commit. I kept it separate to ease review. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
* nir/range-analysis: Range tracking for bcselIan Romanick2019-08-051-0/+77
| | | | | | | | | | | This could be squashed with the previous commit. I kept it separate to ease review. v2: Add some missing cases. Use nir_src_is_const helper. Both suggested by Caio. Use a table for mapping source ranges to a result range. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
* nir/range-analysis: Tighten the range of fsat based on the range of its sourceIan Romanick2019-08-051-1/+22
| | | | | | | | | | This could be squashed with the previous commit. I kept it separate to ease review. v2: Use a switch statement and add more comments. Both suggested by Caio. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
* nir/range-analysis: Rudimentary value range analysis passIan Romanick2019-08-051-0/+653
Most integer operations are omitted because dealing with integer overflow is hard. There are a few things that could be smarter if there was a small amount more tracking of ranges of integer types (i.e., operands are Boolean, operand values fit in 16 bits, etc.). The changes to nir_search_helpers.h are included in this patch to simplify reordering the changes to nir_opt_algebraic.py. v2: Memoize range analysis results. Without this, some shaders appear to get stuck in infinite loops. v3: Rebase on many months of Mesa changes, including 1-bit Boolean changes. v4: Rebase on "nir: Drop imov/fmov in favor of one mov instruction". v5: Use nir_alu_srcs_equal for detecting (a*a). Previously just the SSA value was compared, and this incorrectly matched (a.x*a.y). v6: Many code improvements including (but not limited to) better names, more comments, and better use of helper functions. All suggested by Caio. Rework the handling of several opcodes to use a table for mapping source ranges to a result range. This change fixed a bug that caused fmax(gt_zero, ge_zero) to be incorrectly recognized as ge_zero. Slightly tighten the range of fmul by recognizing that x*x is gt_zero if x is gt_zero. Add similar handling for -x*x. v7: Use _______ in the tables as an alias for unknown. Suggested by Caio. Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
View Lorry Controller Status