Skip to content

Latest commit

 

History

History
185 lines (158 loc) · 22.3 KB

bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483-vs-base.md

File metadata and controls

185 lines (158 loc) · 22.3 KB

Results vs. base

  • fork: iritkatriel
  • ref: binary_subscr_to_op
  • machine: linux-x86_64
  • commit hash: c05e483
  • commit date: 2025-02-05
  • overall geometric mean: 1.002x slower
  • HPT reliability: 97.05%
  • HPT 99th percentile: 1.00x slower
  • Memory change: 1.00x

Benchmarks with tag 'apps':

Benchmark bm-20250205-linux-x86_64-python-cdcacec79f7a216c3c98-3.14.0a4+-cdcacec bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483
docutils 2.58 sec 2.59 sec: 1.00x slower
html5lib 59.8 ms 61.6 ms: 1.03x slower
sphinx 994 ms 1.01 sec: 1.01x slower
Geometric mean (ref) 1.01x slower

Benchmark hidden because not significant (1): 2to3

Benchmarks with tag 'asyncio':

Benchmark bm-20250205-linux-x86_64-python-cdcacec79f7a216c3c98-3.14.0a4+-cdcacec bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483
async_tree_cpu_io_mixed 499 ms 488 ms: 1.02x faster
async_tree_cpu_io_mixed_tg 488 ms 481 ms: 1.02x faster
async_tree_memoization_tg 321 ms 316 ms: 1.02x faster
async_generators 383 ms 381 ms: 1.01x faster
coroutines 23.3 ms 23.5 ms: 1.01x slower
Geometric mean (ref) 1.01x faster

Benchmark hidden because not significant (6): async_tree_none, async_tree_io, async_tree_memoization, async_tree_none_tg, asyncio_websockets, async_tree_io_tg

Benchmarks with tag 'math':

Benchmark bm-20250205-linux-x86_64-python-cdcacec79f7a216c3c98-3.14.0a4+-cdcacec bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483
float 70.5 ms 68.8 ms: 1.03x faster
pidigits 186 ms 186 ms: 1.00x faster
nbody 92.7 ms 94.7 ms: 1.02x slower
Geometric mean (ref) 1.00x faster

Benchmarks with tag 'regex':

Benchmark bm-20250205-linux-x86_64-python-cdcacec79f7a216c3c98-3.14.0a4+-cdcacec bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483
regex_compile 126 ms 125 ms: 1.01x faster
regex_dna 210 ms 213 ms: 1.01x slower
regex_effbot 3.30 ms 3.36 ms: 1.02x slower
Geometric mean (ref) 1.01x slower

Benchmark hidden because not significant (1): regex_v8

Benchmarks with tag 'serialize':

Benchmark bm-20250205-linux-x86_64-python-cdcacec79f7a216c3c98-3.14.0a4+-cdcacec bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483
pickle_pure_python 317 us 312 us: 1.02x faster
json_dumps 11.9 ms 11.8 ms: 1.01x faster
xml_etree_process 58.4 ms 58.1 ms: 1.01x faster
xml_etree_generate 82.9 ms 83.5 ms: 1.01x slower
unpickle_pure_python 216 us 218 us: 1.01x slower
tomli_loads 1.94 sec 1.96 sec: 1.01x slower
json_loads 28.5 us 28.9 us: 1.01x slower
Geometric mean (ref) 1.00x slower

Benchmark hidden because not significant (2): xml_etree_parse, xml_etree_iterparse

Benchmarks with tag 'startup':

Benchmark bm-20250205-linux-x86_64-python-cdcacec79f7a216c3c98-3.14.0a4+-cdcacec bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483
python_startup_no_site 7.05 ms 7.10 ms: 1.01x slower
python_startup 12.8 ms 13.0 ms: 1.01x slower
Geometric mean (ref) 1.01x slower

Benchmarks with tag 'template':

Benchmark bm-20250205-linux-x86_64-python-cdcacec79f7a216c3c98-3.14.0a4+-cdcacec bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483
mako 11.0 ms 10.8 ms: 1.02x faster
genshi_xml 47.8 ms 48.1 ms: 1.00x slower
Geometric mean (ref) 1.00x faster

Benchmark hidden because not significant (2): genshi_text, django_template

All benchmarks:

Benchmark bm-20250205-linux-x86_64-python-cdcacec79f7a216c3c98-3.14.0a4+-cdcacec bm-20250205-linux-x86_64-iritkatriel-binary_subscr_to_op-3.14.0a4+-c05e483
logging_silent 107 ns 103 ns: 1.04x faster
go 119 ms 115 ms: 1.04x faster
sqlglot_parse 1.26 ms 1.23 ms: 1.03x faster
float 70.5 ms 68.8 ms: 1.03x faster
logging_format 6.12 us 5.98 us: 1.02x faster
async_tree_cpu_io_mixed 499 ms 488 ms: 1.02x faster
deltablue 3.22 ms 3.16 ms: 1.02x faster
mako 11.0 ms 10.8 ms: 1.02x faster
pathlib 16.1 ms 15.8 ms: 1.02x faster
richards_super 50.2 ms 49.3 ms: 1.02x faster
pickle_pure_python 317 us 312 us: 1.02x faster
async_tree_cpu_io_mixed_tg 488 ms 481 ms: 1.02x faster
scimark_sor 121 ms 119 ms: 1.02x faster
async_tree_memoization_tg 321 ms 316 ms: 1.02x faster
sqlglot_transpile 1.56 ms 1.54 ms: 1.01x faster
json 5.14 ms 5.08 ms: 1.01x faster
richards 43.8 ms 43.4 ms: 1.01x faster
logging_simple 5.48 us 5.42 us: 1.01x faster
regex_compile 126 ms 125 ms: 1.01x faster
thrift 766 us 759 us: 1.01x faster
json_dumps 11.9 ms 11.8 ms: 1.01x faster
sympy_expand 454 ms 450 ms: 1.01x faster
async_generators 383 ms 381 ms: 1.01x faster
pprint_safe_repr 707 ms 703 ms: 1.01x faster
comprehensions 16.3 us 16.2 us: 1.01x faster
xml_etree_process 58.4 ms 58.1 ms: 1.01x faster
scimark_monte_carlo 67.4 ms 67.1 ms: 1.01x faster
sqlglot_optimize 51.8 ms 51.7 ms: 1.00x faster
create_gc_cycles 2.48 ms 2.47 ms: 1.00x faster
pidigits 186 ms 186 ms: 1.00x faster
gc_traversal 5.03 ms 5.03 ms: 1.00x faster
docutils 2.58 sec 2.59 sec: 1.00x slower
nqueens 79.6 ms 79.9 ms: 1.00x slower
genshi_xml 47.8 ms 48.1 ms: 1.00x slower
deepcopy 255 us 257 us: 1.01x slower
xml_etree_generate 82.9 ms 83.5 ms: 1.01x slower
raytrace 269 ms 271 ms: 1.01x slower
python_startup_no_site 7.05 ms 7.10 ms: 1.01x slower
coroutines 23.3 ms 23.5 ms: 1.01x slower
sympy_str 265 ms 267 ms: 1.01x slower
bench_thread_pool 864 us 872 us: 1.01x slower
hexiom 6.27 ms 6.33 ms: 1.01x slower
unpickle_pure_python 216 us 218 us: 1.01x slower
tomli_loads 1.94 sec 1.96 sec: 1.01x slower
shortest_path 471 ms 476 ms: 1.01x slower
python_startup 12.8 ms 13.0 ms: 1.01x slower
sphinx 994 ms 1.01 sec: 1.01x slower
sqlalchemy_declarative 128 ms 130 ms: 1.01x slower
deepcopy_reduce 2.68 us 2.71 us: 1.01x slower
scimark_lu 115 ms 116 ms: 1.01x slower
connected_components 434 ms 439 ms: 1.01x slower
crypto_pyaes 70.2 ms 71.1 ms: 1.01x slower
json_loads 28.5 us 28.9 us: 1.01x slower
regex_dna 210 ms 213 ms: 1.01x slower
many_optionals 932 us 946 us: 1.02x slower
bench_mp_pool 81.0 ms 82.3 ms: 1.02x slower
regex_effbot 3.30 ms 3.36 ms: 1.02x slower
generators 27.6 ms 28.1 ms: 1.02x slower
chaos 57.6 ms 58.7 ms: 1.02x slower
sympy_integrate 19.5 ms 19.9 ms: 1.02x slower
pyflate 452 ms 460 ms: 1.02x slower
nbody 92.7 ms 94.7 ms: 1.02x slower
typing_runtime_protocols 157 us 161 us: 1.02x slower
fannkuch 396 ms 407 ms: 1.03x slower
html5lib 59.8 ms 61.6 ms: 1.03x slower
sympy_sum 147 ms 152 ms: 1.04x slower
dulwich_log 64.0 ms 66.9 ms: 1.05x slower
sqlalchemy_imperative 16.4 ms 17.3 ms: 1.05x slower
scimark_sparse_mat_mult 4.53 ms 4.85 ms: 1.07x slower
Geometric mean (ref) 1.00x slower

Benchmark hidden because not significant (27): async_tree_none, async_tree_io, telco, sqlite_synth, meteor_contest, regex_v8, async_tree_memoization, sqlglot_normalize, async_tree_none_tg, genshi_text, subparsers, coverage, asyncio_websockets, pprint_pformat, 2to3, scimark_fft, mdp, xml_etree_parse, django_template, bpe_tokeniser, spectral_norm, xml_etree_iterparse, deepcopy_memo, pylint, pycparser, async_tree_io_tg, k_core

  • Geometric mean (including insignificant results): 1.002x slower

HPT report

  • Reliability score: 97.05% likely to be slow
  • 90% likely to have a slowdown of 1.00x
  • 95% likely to have a slowdown of 1.00x
  • 99% likely to have a slowdown of 1.00x

Memory

  • memory change: 1.00x