Carl Friedrich Bolz-Tereick pushed to branch branch/py3.8 at PyPy / pypy
Commits: d2007bbe by Matti Picus at 2022-04-14T21:19:19+03:00 MSVC has a copysign, remove macro definition (bpo 42120) - - - - - a39a358a by Carl Friedrich Bolz-Tereick at 2022-04-17T19:10:33+02:00 ouch: the variable 'name' in this function was used for something else (and it's not even needed in the loop) - - - - - f2a542ac by Carl Friedrich Bolz-Tereick at 2022-04-17T19:33:17+02:00 use %T - - - - - 7b0f244e by Carl Friedrich Bolz-Tereick at 2022-04-18T12:44:13+02:00 don't load varname twice in the LOAD_NAME path - - - - - 4d9c202f by Carl Friedrich Bolz-Tereick at 2022-04-18T22:21:51+02:00 experiment with globals cache again this time most of the logic is in in the celldict strategy itself. it is actually not using the version! instead, we introduce a second cell for every name that is ever cached in a code object. --HG-- branch : global-caching-again - - - - - 98c1f47d by Carl Friedrich Bolz-Tereick at 2022-04-19T20:49:17+02:00 Module.w_dict is properly immutable I think? --HG-- branch : global-caching-again - - - - - 7b155534 by Carl Friedrich Bolz-Tereick at 2022-04-19T21:11:08+02:00 optimize STORE_GLOBAL too --HG-- branch : global-caching-again - - - - - ee3b78c7 by Carl Friedrich Bolz-Tereick at 2022-04-19T21:23:37+02:00 support for LOAD_NAME too --HG-- branch : global-caching-again - - - - - fa3c9f72 by Carl Friedrich Bolz-Tereick at 2022-04-21T14:21:31+02:00 support for caching of builtins --HG-- branch : global-caching-again - - - - - 8c031a8e by Carl Friedrich Bolz-Tereick at 2022-04-21T14:36:24+02:00 add a shortcut for __len__ - - - - - 72d4bac8 by Carl Friedrich Bolz-Tereick at 2022-04-22T13:07:06+02:00 this still failed on 32bit linux. weird. trying an extreme number to see whether it passes then. - - - - - 66028b48 by Carl Friedrich Bolz-Tereick at 2022-04-22T13:16:34+02:00 try to fix these two on 32bit - - - - - 3464b428 by Carl Friedrich Bolz-Tereick at 2022-04-22T13:54:42+02:00 restructure that code --HG-- branch : global-caching-again - - - - - 8075497f by Carl Friedrich Bolz-Tereick at 2022-04-22T13:56:45+02:00 just use struct to implement the simulation, it's like 100x faster (some mapdict tests were seriously slow because of these functions) - - - - - 62639a37 by Carl Friedrich Bolz-Tereick at 2022-04-22T21:33:43+02:00 a comment and an assert --HG-- branch : global-caching-again - - - - - 6e989419 by Carl Friedrich Bolz-Tereick at 2022-04-23T13:29:33+02:00 merge global-caching-again: speed up the interpreter by caching global and builtin lookups on the code object - - - - - 12ec098d by Carl Friedrich Bolz-Tereick at 2022-04-23T13:55:06+02:00 maybe this fixes 32bit - - - - - bbd602dd by Carl Friedrich Bolz-Tereick at 2022-04-23T19:32:53+02:00 merge default --HG-- branch : py3.8 - - - - - f143075a by Carl Friedrich Bolz-Tereick at 2022-04-23T19:40:15+02:00 this part of the test is just wrong - - - - - e1499e99 by Carl Friedrich Bolz-Tereick at 2022-04-23T19:44:01+02:00 merge default --HG-- branch : py3.8 - - - - - 1e972794 by Carl Friedrich Bolz-Tereick at 2022-04-24T21:16:59+02:00 fix --HG-- branch : py3.8 - - - - - 752667f5 by Carl Friedrich Bolz-Tereick at 2022-04-24T21:17:25+02:00 merge heads - - - - - 65efbbf4 by Carl Friedrich Bolz-Tereick at 2022-04-24T21:17:49+02:00 merge default --HG-- branch : py3.8 - - - - - 16 changed files: - pypy/interpreter/baseobjspace.py - pypy/interpreter/module.py - pypy/interpreter/pycode.py - pypy/interpreter/pyopcode.py - pypy/interpreter/typedef.py - pypy/module/cpyext/include/floatobject.h - pypy/module/pypyjit/test_pypy_c/test_instance.py - pypy/objspace/descroperation.py - pypy/objspace/std/celldict.py - pypy/objspace/std/dictmultiobject.py - pypy/objspace/std/test/test_celldict.py - pypy/objspace/std/test/test_dictmultiobject.py - pypy/objspace/test/test_descroperation.py - rpython/rlib/longlong2float.py - rpython/rlib/test/test_longlong2float.py - rpython/translator/c/src/ll_math.c View it on Heptapod: https://foss.heptapod.net/pypy/pypy/-/compare/e3bda87f45b1408fcddf64ff730d538dea2b69eb...65efbbf4a2362e2d5e3b7b9fc9a469fd636709e6 -- View it on Heptapod: https://foss.heptapod.net/pypy/pypy/-/compare/e3bda87f45b1408fcddf64ff730d538dea2b69eb...65efbbf4a2362e2d5e3b7b9fc9a469fd636709e6 You're receiving this email because of your account on foss.heptapod.net.
_______________________________________________ pypy-commit mailing list -- pypy-commit@python.org To unsubscribe send an email to pypy-commit-le...@python.org https://mail.python.org/mailman3/lists/pypy-commit.python.org/ Member address: arch...@mail-archive.com
