Pablo Galindo Salgado <pablog...@gmail.com> added the comment:
I may not know the what is the average airspeed velocity of a laden swallow, but I know the average speed of adding a LOAD_METHOD opcode cache as in PR 23503 (measured with PGO + LTO + CPU isolation): +-------------------------+--------------------------------------+-------------------------------------------+ | Benchmark | 2020-11-24_18-08-master-0ec34cab9dd4 | 2020-11-24_19-41-load_method-0c43ca99644b | +=========================+======================================+===========================================+ | pidigits | 266 ms | 233 ms: 1.14x faster (-12%) | +-------------------------+--------------------------------------+-------------------------------------------+ | regex_v8 | 32.7 ms | 30.4 ms: 1.07x faster (-7%) | +-------------------------+--------------------------------------+-------------------------------------------+ | unpickle | 17.7 us | 17.1 us: 1.04x faster (-4%) | +-------------------------+--------------------------------------+-------------------------------------------+ | scimark_sparse_mat_mult | 6.21 ms | 5.99 ms: 1.04x faster (-3%) | +-------------------------+--------------------------------------+-------------------------------------------+ | go | 283 ms | 274 ms: 1.03x faster (-3%) | +-------------------------+--------------------------------------+-------------------------------------------+ | mako | 20.4 ms | 19.9 ms: 1.03x faster (-3%) | +-------------------------+--------------------------------------+-------------------------------------------+ | pickle_pure_python | 546 us | 531 us: 1.03x faster (-3%) | +-------------------------+--------------------------------------+-------------------------------------------+ | logging_simple | 9.58 us | 9.34 us: 1.03x faster (-3%) | +-------------------------+--------------------------------------+-------------------------------------------+ | telco | 8.21 ms | 8.03 ms: 1.02x faster (-2%) | +-------------------------+--------------------------------------+-------------------------------------------+ | regex_effbot | 3.97 ms | 3.88 ms: 1.02x faster (-2%) | +-------------------------+--------------------------------------+-------------------------------------------+ | regex_dna | 267 ms | 261 ms: 1.02x faster (-2%) | +-------------------------+--------------------------------------+-------------------------------------------+ | pathlib | 21.2 ms | 21.7 ms: 1.02x slower (+2%) | +-------------------------+--------------------------------------+-------------------------------------------+ | xml_etree_iterparse | 129 ms | 132 ms: 1.02x slower (+2%) | +-------------------------+--------------------------------------+-------------------------------------------+ | xml_etree_parse | 186 ms | 193 ms: 1.03x slower (+3%) | +-------------------------+--------------------------------------+-------------------------------------------+ Not significant (46): scimark_fft; nbody; pickle; pickle_list; pickle_dict; logging_format; scimark_lu; deltablue; chaos; meteor_contest; sqlalchemy_imperative; scimark_sor; raytrace; scimark_monte_carlo; sympy_integrate; unpickle_pure_python; dulwich_log; logging_silent; nqueens; json_loads; crypto_pyaes; hexiom; django_template; sympy_str; regex_compile; sympy_expand; xml_etree_generate; unpack_sequence; spectral_norm; xml_etree_process; sqlite_synth; genshi_xml; pyflate; json_dumps; python_startup_no_site; 2to3; python_startup; genshi_text; unpickle_list; fannkuch; sympy_sum; sqlalchemy_declarative; chameleon; float; richards; tornado_http; ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <https://bugs.python.org/issue42115> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
