Skip to content

Instantly share code, notes, and snippets.

@mdboom

mdboom/README.md

Created March 21, 2024 22:19
Show Gist options
  • Save mdboom/b2dd22a6a351db9ce9192bacf7dbd44a to your computer and use it in GitHub Desktop.
Save mdboom/b2dd22a6a351db9ce9192bacf7dbd44a to your computer and use it in GitHub Desktop.
Download ZIP
Raw
README.md

linux x86_64 (linux)

commit change
d0672a4 (d0672a4) 1.02x ↑ 📄 📈
Raw
linux-d0672a4-vs-039d20.md

Benchmarks with tag 'apps':

Benchmark bm-20240319-linux-x86_64-python-039d20ae5428dfe3d704-3.13.0a5+-039d20a bm-20240321-linux-x86_64-faster%2dcpython-fix_gc_counting-3.13.0a5+-d0672a4
2to3 272 ms 268 ms: 1.02x faster
chameleon 7.01 ms 6.95 ms: 1.01x faster
docutils 2.65 sec 2.74 sec: 1.03x slower
html5lib 67.6 ms 65.4 ms: 1.03x faster
tornado_http 97.4 ms 94.8 ms: 1.03x faster
Geometric mean (ref) 1.01x faster

Benchmarks with tag 'asyncio':

Benchmark bm-20240319-linux-x86_64-python-039d20ae5428dfe3d704-3.13.0a5+-039d20a bm-20240321-linux-x86_64-faster%2dcpython-fix_gc_counting-3.13.0a5+-d0672a4
async_tree_memoization_tg 593 ms 442 ms: 1.34x faster
async_tree_io_tg 1.22 sec 917 ms: 1.33x faster
async_tree_io 1.21 sec 917 ms: 1.32x faster
async_tree_none_tg 459 ms 350 ms: 1.31x faster
async_tree_memoization 578 ms 461 ms: 1.25x faster
async_tree_cpu_io_mixed 727 ms 590 ms: 1.23x faster
async_tree_cpu_io_mixed_tg 744 ms 606 ms: 1.23x faster
async_tree_none 451 ms 374 ms: 1.20x faster
Geometric mean (ref) 1.28x faster

Benchmarks with tag 'math':

Benchmark bm-20240319-linux-x86_64-python-039d20ae5428dfe3d704-3.13.0a5+-039d20a bm-20240321-linux-x86_64-faster%2dcpython-fix_gc_counting-3.13.0a5+-d0672a4
nbody 98.6 ms 89.9 ms: 1.10x faster
float 81.8 ms 78.5 ms: 1.04x faster
pidigits 190 ms 190 ms: 1.00x faster
Geometric mean (ref) 1.05x faster

Benchmarks with tag 'regex':

Benchmark bm-20240319-linux-x86_64-python-039d20ae5428dfe3d704-3.13.0a5+-039d20a bm-20240321-linux-x86_64-faster%2dcpython-fix_gc_counting-3.13.0a5+-d0672a4
regex_v8 25.5 ms 24.8 ms: 1.03x faster
regex_dna 212 ms 218 ms: 1.03x slower
Geometric mean (ref) 1.00x slower

Benchmark hidden because not significant (2): regex_compile, regex_effbot

Benchmarks with tag 'serialize':

Benchmark bm-20240319-linux-x86_64-python-039d20ae5428dfe3d704-3.13.0a5+-039d20a bm-20240321-linux-x86_64-faster%2dcpython-fix_gc_counting-3.13.0a5+-d0672a4
unpickle 15.7 us 14.6 us: 1.07x faster
unpickle_pure_python 224 us 220 us: 1.02x faster
xml_etree_generate 87.3 ms 85.5 ms: 1.02x faster
json_dumps 10.5 ms 10.4 ms: 1.01x faster
tomli_loads 2.21 sec 2.19 sec: 1.01x faster
pickle 11.9 us 11.8 us: 1.01x faster
xml_etree_process 60.1 ms 59.8 ms: 1.01x faster
json_loads 28.3 us 28.6 us: 1.01x slower
unpickle_list 5.10 us 5.24 us: 1.03x slower
pickle_dict 33.5 us 35.3 us: 1.06x slower
pickle_list 4.94 us 5.36 us: 1.09x slower
Geometric mean (ref) 1.00x slower

Benchmark hidden because not significant (3): pickle_pure_python, xml_etree_iterparse, xml_etree_parse

Benchmarks with tag 'startup':

Benchmark bm-20240319-linux-x86_64-python-039d20ae5428dfe3d704-3.13.0a5+-039d20a bm-20240321-linux-x86_64-faster%2dcpython-fix_gc_counting-3.13.0a5+-d0672a4
python_startup_no_site 8.92 ms 8.83 ms: 1.01x faster
python_startup 10.3 ms 10.4 ms: 1.01x slower
Geometric mean (ref) 1.00x faster

Benchmarks with tag 'template':

Benchmark bm-20240319-linux-x86_64-python-039d20ae5428dfe3d704-3.13.0a5+-039d20a bm-20240321-linux-x86_64-faster%2dcpython-fix_gc_counting-3.13.0a5+-d0672a4
django_template 34.6 ms 34.0 ms: 1.02x faster
mako 11.2 ms 11.5 ms: 1.03x slower
Geometric mean (ref) 1.00x slower

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

All benchmarks:

Benchmark bm-20240319-linux-x86_64-python-039d20ae5428dfe3d704-3.13.0a5+-039d20a bm-20240321-linux-x86_64-faster%2dcpython-fix_gc_counting-3.13.0a5+-d0672a4
async_tree_memoization_tg 593 ms 442 ms: 1.34x faster
async_tree_io_tg 1.22 sec 917 ms: 1.33x faster
async_tree_io 1.21 sec 917 ms: 1.32x faster
async_tree_none_tg 459 ms 350 ms: 1.31x faster
async_tree_memoization 578 ms 461 ms: 1.25x faster
async_tree_cpu_io_mixed 727 ms 590 ms: 1.23x faster
async_tree_cpu_io_mixed_tg 744 ms 606 ms: 1.23x faster
async_tree_none 451 ms 374 ms: 1.20x faster
mypy2 864 ms 738 ms: 1.17x faster
nbody 98.6 ms 89.9 ms: 1.10x faster
unpickle 15.7 us 14.6 us: 1.07x faster
unpack_sequence 47.0 ns 44.2 ns: 1.06x faster
create_gc_cycles 1.52 ms 1.44 ms: 1.06x faster
gc_traversal 3.83 ms 3.65 ms: 1.05x faster
float 81.8 ms 78.5 ms: 1.04x faster
html5lib 67.6 ms 65.4 ms: 1.03x faster
async_generators 454 ms 440 ms: 1.03x faster
telco 8.67 ms 8.41 ms: 1.03x faster
scimark_lu 116 ms 113 ms: 1.03x faster
scimark_sparse_mat_mult 4.95 ms 4.81 ms: 1.03x faster
regex_v8 25.5 ms 24.8 ms: 1.03x faster
tornado_http 97.4 ms 94.8 ms: 1.03x faster
pycparser 1.20 sec 1.17 sec: 1.02x faster
unpickle_pure_python 224 us 220 us: 1.02x faster
xml_etree_generate 87.3 ms 85.5 ms: 1.02x faster
django_template 34.6 ms 34.0 ms: 1.02x faster
sqlglot_transpile 1.61 ms 1.58 ms: 1.02x faster
2to3 272 ms 268 ms: 1.02x faster
scimark_fft 364 ms 360 ms: 1.01x faster
logging_simple 5.90 us 5.84 us: 1.01x faster
coverage 97.1 ms 96.1 ms: 1.01x faster
python_startup_no_site 8.92 ms 8.83 ms: 1.01x faster
comprehensions 16.5 us 16.4 us: 1.01x faster
logging_format 6.55 us 6.49 us: 1.01x faster
pyflate 487 ms 483 ms: 1.01x faster
chameleon 7.01 ms 6.95 ms: 1.01x faster
json_dumps 10.5 ms 10.4 ms: 1.01x faster
tomli_loads 2.21 sec 2.19 sec: 1.01x faster
pickle 11.9 us 11.8 us: 1.01x faster
xml_etree_process 60.1 ms 59.8 ms: 1.01x faster
pidigits 190 ms 190 ms: 1.00x faster
raytrace 269 ms 268 ms: 1.00x faster
asyncio_tcp_ssl 1.84 sec 1.84 sec: 1.00x slower
sympy_expand 461 ms 463 ms: 1.00x slower
sympy_integrate 20.0 ms 20.0 ms: 1.00x slower
bench_thread_pool 826 us 831 us: 1.01x slower
generators 29.9 ms 30.1 ms: 1.01x slower
aiohttp 1.16 ms 1.17 ms: 1.01x slower
asyncio_tcp 500 ms 503 ms: 1.01x slower
pprint_safe_repr 736 ms 741 ms: 1.01x slower
python_startup 10.3 ms 10.4 ms: 1.01x slower
scimark_monte_carlo 69.1 ms 69.6 ms: 1.01x slower
sympy_sum 153 ms 154 ms: 1.01x slower
deepcopy 350 us 353 us: 1.01x slower
sympy_str 274 ms 277 ms: 1.01x slower
json_loads 28.3 us 28.6 us: 1.01x slower
spectral_norm 113 ms 114 ms: 1.01x slower
richards 45.6 ms 46.1 ms: 1.01x slower
sqlglot_optimize 54.1 ms 54.8 ms: 1.01x slower
nqueens 80.7 ms 81.8 ms: 1.01x slower
scimark_sor 129 ms 131 ms: 1.02x slower
typing_runtime_protocols 111 us 113 us: 1.02x slower
dulwich_log 66.8 ms 67.9 ms: 1.02x slower
fannkuch 399 ms 406 ms: 1.02x slower
logging_silent 98.1 ns 99.8 ns: 1.02x slower
json 5.20 ms 5.31 ms: 1.02x slower
regex_dna 212 ms 218 ms: 1.03x slower
unpickle_list 5.10 us 5.24 us: 1.03x slower
mako 11.2 ms 11.5 ms: 1.03x slower
richards_super 51.5 ms 53.1 ms: 1.03x slower
docutils 2.65 sec 2.74 sec: 1.03x slower
pickle_dict 33.5 us 35.3 us: 1.06x slower
mdp 2.56 sec 2.71 sec: 1.06x slower
pickle_list 4.94 us 5.36 us: 1.09x slower
Geometric mean (ref) 1.02x faster

Benchmark hidden because not significant (28): djangocms, thrift, deepcopy_reduce, bench_mp_pool, dask, crypto_pyaes, sqlite_synth, pprint_pformat, pickle_pure_python, genshi_text, meteor_contest, asyncio_websockets, xml_etree_iterparse, genshi_xml, regex_compile, coroutines, sqlglot_normalize, regex_effbot, hexiom, deepcopy_memo, chaos, go, sqlglot_parse, deltablue, pathlib, gunicorn, xml_etree_parse, pylint

HPT report

  • Reliability score: 95.23% likely to be faster
  • 90% likely to have a speedup of 1.00x
  • 95% likely to have a speedup of 1.00x
  • 99% likely to have a speedup of 1.00x

Memory

  • memory change: 1.03x
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment