STINNER Victor added the comment: I pushed a change to optimize the ASCII decoder.
Attached bench.py script: microbenchmark on the ASCII decoder. My results follows. Common platform: Platform: Linux-4.1.5-200.fc22.x86_64-x86_64-with-fedora-22-Twenty_Two Bits: int=32, long=64, long long=64, size_t=64, void*=64 Timer: time.perf_counter CFLAGS: -Wno-unused-result -Wsign-compare -Wunreachable-code -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes CPU model: Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz Python unicode implementation: PEP 393 Timer info: namespace(adjustable=False, implementation='clock_gettime(CLOCK_MONOTONIC)', monotonic=True, resolution=1e-09) Platform of campaign before: Timer precision: 57 ns Date: 2015-09-21 23:48:00 SCM: hg revision=73867bf953e6 branch=default date="2015-09-21 22:40 +0200" Python version: 3.6.0a0 (default:73867bf953e6, Sep 21 2015, 23:47:35) [GCC 5.1.1 20150618 (Red Hat 5.1.1-4)] Platform of campaign after: Timer precision: 55 ns Date: 2015-09-21 23:52:55 Python version: 3.6.0a0 (default:bb0f55f1ec22+, Sep 21 2015, 23:52:43) [GCC 5.1.1 20150618 (Red Hat 5.1.1-4)] SCM: hg revision=bb0f55f1ec22+ tag=tip branch=default date="2015-09-21 23:06 +0200" ------------------+-------------+--------------- ignore | before | after ------------------+-------------+--------------- 256 x 10**1 bytes | 314 us (*) | 5.25 us (-98%) 256 x 10**3 bytes | 31.9 ms (*) | 509 us (-98%) 256 x 10**2 bytes | 3.13 ms (*) | 50.2 us (-98%) 256 x 10**4 bytes | 315 ms (*) | 5.12 ms (-98%) ------------------+-------------+--------------- Total | 351 ms (*) | 5.69 ms (-98%) ------------------+-------------+--------------- ------------------+-------------+--------------- replace | before | after ------------------+-------------+--------------- 256 x 10**1 bytes | 352 us (*) | 6.32 us (-98%) 256 x 10**3 bytes | 35.2 ms (*) | 608 us (-98%) 256 x 10**2 bytes | 3.52 ms (*) | 60.9 us (-98%) 256 x 10**4 bytes | 354 ms (*) | 6.19 ms (-98%) ------------------+-------------+--------------- Total | 393 ms (*) | 6.87 ms (-98%) ------------------+-------------+--------------- ------------------+-------------+--------------- surrogateescape | before | after ------------------+-------------+--------------- 256 x 10**1 bytes | 369 us (*) | 5.92 us (-98%) 256 x 10**3 bytes | 36.8 ms (*) | 570 us (-98%) 256 x 10**2 bytes | 3.69 ms (*) | 56.9 us (-98%) 256 x 10**4 bytes | 371 ms (*) | 5.79 ms (-98%) ------------------+-------------+--------------- Total | 412 ms (*) | 6.43 ms (-98%) ------------------+-------------+--------------- ------------------+-------------+-------- backslashreplace | before | after ------------------+-------------+-------- 256 x 10**1 bytes | 357 us (*) | 361 us 256 x 10**3 bytes | 35.1 ms (*) | 36.1 ms 256 x 10**2 bytes | 3.52 ms (*) | 3.59 ms 256 x 10**4 bytes | 357 ms (*) | 365 ms ------------------+-------------+-------- Total | 396 ms (*) | 405 ms ------------------+-------------+-------- -----------------+--------------+--------------- Summary | before | after -----------------+--------------+--------------- ignore | 351 ms (*) | 5.69 ms (-98%) replace | 393 ms (*) | 6.87 ms (-98%) surrogateescape | 412 ms (*) | 6.43 ms (-98%) backslashreplace | 396 ms (*) | 405 ms -----------------+--------------+--------------- Total | 1.55 sec (*) | 424 ms (-73%) -----------------+--------------+--------------- ---------- Added file: http://bugs.python.org/file40539/bench.py _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue24870> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
