Closed fmoletta closed 4 months ago
**Hyper Thereading Benchmark results**
hyperfine -r 2 -n "hyper_threading_main threads: 1" 'RAYON_NUM_THREADS=1 ./hyper_threading_main' -n "hyper_threading_pr threads: 1" 'RAYON_NUM_THREADS=1 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 1
Time (mean ± σ): 27.125 s ± 0.011 s [User: 26.331 s, System: 0.790 s]
Range (min … max): 27.117 s … 27.132 s 2 runs
Benchmark 2: hyper_threading_pr threads: 1
Time (mean ± σ): 27.119 s ± 0.034 s [User: 26.279 s, System: 0.837 s]
Range (min … max): 27.095 s … 27.143 s 2 runs
Summary
'hyper_threading_pr threads: 1' ran
1.00 ± 0.00 times faster than 'hyper_threading_main threads: 1'
hyperfine -r 2 -n "hyper_threading_main threads: 2" 'RAYON_NUM_THREADS=2 ./hyper_threading_main' -n "hyper_threading_pr threads: 2" 'RAYON_NUM_THREADS=2 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 2
Time (mean ± σ): 14.610 s ± 0.011 s [User: 26.946 s, System: 0.797 s]
Range (min … max): 14.602 s … 14.618 s 2 runs
Benchmark 2: hyper_threading_pr threads: 2
Time (mean ± σ): 14.355 s ± 0.029 s [User: 26.352 s, System: 0.806 s]
Range (min … max): 14.334 s … 14.375 s 2 runs
Summary
'hyper_threading_pr threads: 2' ran
1.02 ± 0.00 times faster than 'hyper_threading_main threads: 2'
hyperfine -r 2 -n "hyper_threading_main threads: 4" 'RAYON_NUM_THREADS=4 ./hyper_threading_main' -n "hyper_threading_pr threads: 4" 'RAYON_NUM_THREADS=4 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 4
Time (mean ± σ): 10.967 s ± 0.260 s [User: 38.781 s, System: 0.996 s]
Range (min … max): 10.783 s … 11.151 s 2 runs
Benchmark 2: hyper_threading_pr threads: 4
Time (mean ± σ): 10.970 s ± 0.330 s [User: 38.516 s, System: 0.947 s]
Range (min … max): 10.736 s … 11.203 s 2 runs
Summary
'hyper_threading_main threads: 4' ran
1.00 ± 0.04 times faster than 'hyper_threading_pr threads: 4'
hyperfine -r 2 -n "hyper_threading_main threads: 6" 'RAYON_NUM_THREADS=6 ./hyper_threading_main' -n "hyper_threading_pr threads: 6" 'RAYON_NUM_THREADS=6 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 6
Time (mean ± σ): 10.732 s ± 0.239 s [User: 39.096 s, System: 0.946 s]
Range (min … max): 10.563 s … 10.901 s 2 runs
Benchmark 2: hyper_threading_pr threads: 6
Time (mean ± σ): 10.948 s ± 0.010 s [User: 38.642 s, System: 0.965 s]
Range (min … max): 10.941 s … 10.955 s 2 runs
Summary
'hyper_threading_main threads: 6' ran
1.02 ± 0.02 times faster than 'hyper_threading_pr threads: 6'
hyperfine -r 2 -n "hyper_threading_main threads: 8" 'RAYON_NUM_THREADS=8 ./hyper_threading_main' -n "hyper_threading_pr threads: 8" 'RAYON_NUM_THREADS=8 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 8
Time (mean ± σ): 10.610 s ± 0.134 s [User: 39.196 s, System: 1.060 s]
Range (min … max): 10.515 s … 10.704 s 2 runs
Benchmark 2: hyper_threading_pr threads: 8
Time (mean ± σ): 10.555 s ± 0.118 s [User: 38.930 s, System: 0.910 s]
Range (min … max): 10.472 s … 10.639 s 2 runs
Summary
'hyper_threading_pr threads: 8' ran
1.01 ± 0.02 times faster than 'hyper_threading_main threads: 8'
hyperfine -r 2 -n "hyper_threading_main threads: 16" 'RAYON_NUM_THREADS=16 ./hyper_threading_main' -n "hyper_threading_pr threads: 16" 'RAYON_NUM_THREADS=16 ./hyper_threading_pr'
Benchmark 1: hyper_threading_main threads: 16
Time (mean ± σ): 10.656 s ± 0.326 s [User: 39.576 s, System: 0.990 s]
Range (min … max): 10.425 s … 10.886 s 2 runs
Benchmark 2: hyper_threading_pr threads: 16
Time (mean ± σ): 10.601 s ± 0.070 s [User: 39.252 s, System: 1.002 s]
Range (min … max): 10.551 s … 10.650 s 2 runs
Summary
'hyper_threading_pr threads: 16' ran
1.01 ± 0.03 times faster than 'hyper_threading_main threads: 16'
Benchmark Results for unmodified programs :rocket:
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base big_factorial |
2.062 ± 0.011 | 2.050 | 2.088 | 1.00 ± 0.01 |
head big_factorial |
2.056 ± 0.014 | 2.038 | 2.073 | 1.00 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base big_fibonacci |
2.027 ± 0.035 | 2.000 | 2.121 | 1.01 ± 0.02 |
head big_fibonacci |
2.006 ± 0.020 | 1.979 | 2.040 | 1.00 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base blake2s_integration_benchmark |
7.622 ± 0.143 | 7.484 | 7.895 | 1.00 |
head blake2s_integration_benchmark |
7.633 ± 0.162 | 7.450 | 7.998 | 1.00 ± 0.03 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base compare_arrays_200000 |
2.159 ± 0.084 | 2.101 | 2.377 | 1.01 ± 0.04 |
head compare_arrays_200000 |
2.134 ± 0.035 | 2.093 | 2.221 | 1.00 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base dict_integration_benchmark |
1.407 ± 0.005 | 1.400 | 1.413 | 1.00 |
head dict_integration_benchmark |
1.423 ± 0.007 | 1.416 | 1.434 | 1.01 ± 0.01 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base field_arithmetic_get_square_benchmark |
1.280 ± 0.006 | 1.272 | 1.290 | 1.00 |
head field_arithmetic_get_square_benchmark |
1.286 ± 0.012 | 1.270 | 1.304 | 1.00 ± 0.01 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base integration_builtins |
7.643 ± 0.082 | 7.528 | 7.770 | 1.00 ± 0.02 |
head integration_builtins |
7.637 ± 0.091 | 7.506 | 7.727 | 1.00 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base keccak_integration_benchmark |
8.054 ± 0.266 | 7.783 | 8.664 | 1.02 ± 0.04 |
head keccak_integration_benchmark |
7.907 ± 0.085 | 7.754 | 7.982 | 1.00 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base linear_search |
2.095 ± 0.014 | 2.076 | 2.126 | 1.00 |
head linear_search |
2.143 ± 0.045 | 2.103 | 2.258 | 1.02 ± 0.02 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base math_cmp_and_pow_integration_benchmark |
1.712 ± 0.061 | 1.678 | 1.878 | 1.01 ± 0.04 |
head math_cmp_and_pow_integration_benchmark |
1.703 ± 0.010 | 1.693 | 1.729 | 1.00 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base math_integration_benchmark |
1.597 ± 0.005 | 1.588 | 1.605 | 1.00 |
head math_integration_benchmark |
1.609 ± 0.027 | 1.586 | 1.680 | 1.01 ± 0.02 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base memory_integration_benchmark |
1.197 ± 0.005 | 1.190 | 1.206 | 1.00 |
head memory_integration_benchmark |
1.197 ± 0.010 | 1.185 | 1.215 | 1.00 ± 0.01 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base operations_with_data_structures_benchmarks |
1.833 ± 0.054 | 1.797 | 1.983 | 1.01 ± 0.03 |
head operations_with_data_structures_benchmarks |
1.813 ± 0.010 | 1.802 | 1.832 | 1.00 |
Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
---|---|---|---|---|
base pedersen |
514.4 ± 2.3 | 510.8 | 518.1 | 1.00 |
head pedersen |
517.0 ± 7.4 | 512.0 | 537.0 | 1.01 ± 0.02 |
Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
---|---|---|---|---|
base poseidon_integration_benchmark |
954.2 ± 6.8 | 942.9 | 963.9 | 1.01 ± 0.01 |
head poseidon_integration_benchmark |
948.6 ± 5.5 | 940.4 | 958.1 | 1.00 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base secp_integration_benchmark |
1.896 ± 0.049 | 1.855 | 2.006 | 1.03 ± 0.03 |
head secp_integration_benchmark |
1.850 ± 0.024 | 1.830 | 1.912 | 1.00 |
Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
---|---|---|---|---|
base set_integration_benchmark |
638.4 ± 3.2 | 635.0 | 644.3 | 1.00 |
head set_integration_benchmark |
639.6 ± 1.8 | 636.7 | 642.1 | 1.00 ± 0.01 |
Command | Mean [s] | Min [s] | Max [s] | Relative |
---|---|---|---|---|
base uint256_integration_benchmark |
4.245 ± 0.040 | 4.161 | 4.296 | 1.00 ± 0.02 |
head uint256_integration_benchmark |
4.235 ± 0.066 | 4.137 | 4.360 | 1.00 |
Attention: Patch coverage is 99.54751%
with 1 lines
in your changes are missing coverage. Please review.
Project coverage is 94.76%. Comparing base (
258dd84
) to head (0896a0d
). Report is 6 commits behind head on main.:exclamation: Current head 0896a0d differs from pull request most recent head 6eb6768. Consider uploading reports for the commit 6eb6768 to get more accurate results
Files | Patch % | Lines |
---|---|---|
vm/src/vm/runners/cairo_runner.rs | 97.67% | 1 Missing :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
Cairo 0 programs don't allow input arguments (aka arguments to the main function). This is circumvented in cairo 1 by creating the arguments before calling the main function, by creating instructions that declare the input arguments as constants and allocate them as tempvars. This creates a "fake stack", with the builtin bases and the return_fp & return_pc and then a "real_stack" (used by main) with the builtin bases and the arguments. An example of the execution segment after cairo 1 initialization (both VM & added instructions) would be:
While this solves the issue of cairo 0 not being compatible with arguments (as the initial_stack doesn't contain the arguments), it inserts the values of the arguments into the instructions, making the program hash vary depending on the arguments passed to the program. This PR aims to solve this last program by creating both the initial (inputless) stack and the real stack in an initialization method (similar to the VM's initialize) and removing the instructions responsible for creating the real stack in
create_entrypoint
. The result of this changes should not modify cairo 1 executions, as the starting execution segment before calling the main function will remain the sameOther solutions previously explored: Removing create_entry_point altogether and using the normal cairo-vm initialization & segment_arena builtin runner: This lead to problems concerning CairoPie as segment_arena & argument creation disrupted the segment order required by cairo pie validations, and arguments were not expected in the initial_stack