arigato changed the topic of #pypy to: PyPy, the flexible snake (IRC logs: https://quodlibet.duckdns.org/irc/pypy/latest.log.html#irc-end ) | use cffi for calling C | mac OS and Fedora are not Windows
tsaka__ has quit [Ping timeout: 258 seconds]
chelz has joined #pypy
dustinm- has quit [Quit: Leaving]
dustinm has joined #pypy
chelz has quit [Ping timeout: 250 seconds]
chelz has joined #pypy
dddddd has quit [Remote host closed the connection]
_whitelogger has joined #pypy
speeder39_ has joined #pypy
ronan has quit [Ping timeout: 258 seconds]
ronan has joined #pypy
chelz has quit [Quit: chelz]
chelz has joined #pypy
riddle has quit [Ping timeout: 268 seconds]
riddle has joined #pypy
xcm has quit [Read error: Connection reset by peer]
xcm has joined #pypy
tsaka__ has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
speeder39_ has quit [Quit: Connection closed for inactivity]
jacob22 has quit [Ping timeout: 245 seconds]
xcm has quit [Remote host closed the connection]
speeder39_ has joined #pypy
xcm has joined #pypy
jacob22 has joined #pypy
antocuni has joined #pypy
infernix has quit [Ping timeout: 268 seconds]
dddddd has joined #pypy
irclogs_io_bot has quit [Remote host closed the connection]
infernix has joined #pypy
irclogs_io_bot has joined #pypy
speeder39_ has quit [Quit: Connection closed for inactivity]
<kenaan> arigo cffi/cffi 8097b9f9ab09 /cffi/cparser.py: (brendan on mailing list) Ignore preprocessor stuff when looking for '"'
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
antocuni has quit [Ping timeout: 245 seconds]
xcm has quit [Read error: Connection reset by peer]
xcm has joined #pypy
wleslie has joined #pypy
lritter has joined #pypy
marky1991 has joined #pypy
antocuni has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
jcea has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
Rhy0lite has joined #pypy
marky1991 has quit [Ping timeout: 245 seconds]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
marky1991 has joined #pypy
marky1991 has quit [Remote host closed the connection]
marky1991 has joined #pypy
wleslie has quit [Quit: ~~~ Crash in JIT!]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
xcm is now known as Guest92251
Guest92251 has quit [Killed (card.freenode.net (Nickname regained by services))]
xcm has joined #pypy
<ronan> ^^^ I think this shows a fundamental limitation of cffi: that it has a lot of overhead compared to plain Python code
<ronan> I'm beginning to think we should just drop _decimal
vstinner has joined #pypy
<vstinner> hello. i wrote an article about my work on the C API in Python 3.8: https://vstinner.github.io/split-include-directory-python38.html
<vstinner> in the meanwhile, they are discussions about adding more C functions per performance :-) https://mail.python.org/archives/list/capi-sig@python.org/thread/X6HJOEX6RRFLNKAQSQR6HLD7K4QZ4OTZ/ & https://github.com/python/cpython/pull/14228
piem has left #pypy [#pypy]
<tos9> ronan: what was the original reason to write a cffi implementation?
<ronan> tos9: I don't know. Better compatibility, maybe?
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
<mattip> vstinner: thanks for trying to simplify the C API
<vstinner> mattip: "trying" well, it's a work in progress :)
<vstinner> mattip: i "removed" *a lot* of private functions: moved to the internal API
<vstinner> mattip: i'm not brave, so i chose to start with the private API which is safe to modify (we don't provide an backward compatibility warranty)
<mattip> I would think performance PRs would have some kind of benchmarking attached
<mattip> I am sure their work is motivated by problems they see, but it would be nice to share it
<vstinner> mattip: i requested benchmark for the new public function :)
jacob22_ has joined #pypy
jacob22 has quit [Ping timeout: 252 seconds]
jacob22_ has quit [Ping timeout: 246 seconds]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
themsay has quit [Read error: Connection reset by peer]
themsay has joined #pypy
brett-soric has joined #pypy
antocuni has quit [Ping timeout: 258 seconds]
<kenaan> mattip setitem2d 06767d845ae4 /pypy/objspace/std/: test for 2d memoryview.__setitem__, passes with -A (issue bb-3028)
themsay has quit [Ping timeout: 245 seconds]
<mattip> bb-3028 is bitbucket issue 3028, in the hopes that soon we will move off bitbucket to something else
themsay has joined #pypy
xcm has quit [Remote host closed the connection]
brett-soric has left #pypy [#pypy]
xcm has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
<LarstiQ> vstinner: you seem to be presenting a viewpoint keeping pypy in mind in those c-api performance discussions, thank you!
themsay has quit [Read error: Connection reset by peer]
jacob22_ has joined #pypy
themsay has joined #pypy
marky1991 has quit [Ping timeout: 244 seconds]
jacob22_ has quit [Ping timeout: 248 seconds]
speeder39_ has joined #pypy
<mattip> memoryview indexing is not very advanced, multiple dimensons support only integer indexing m[2, 3] and not slices m[:, 3]
<mattip> even on cpython 3.8
Arfrever has quit [Ping timeout: 245 seconds]
jacob22_ has joined #pypy
Rhy0lite has quit [Quit: Leaving]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
<vstinner> LarstiQ: my long term goal would be to be able the same binary on Python 3.8, Python 3.9 *and* PyPy :)
<kenaan> mattip setitem2d 0f9d6d72dfe9 /pypy/objspace/std/: fix for for 2d memoryview.__setitem__, (issue bb-3028)
<mattip> we would need to rethink our header generation
<mattip> We puposefully rename the functions exported from libpypy.so to prevent reuse by c-extension modules
<mattip> since some cpython macros are functions on pypy, PyListObject is opaque, PyDatetime is implemented in python not C, and others
<bbot2> Started: http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/6280 [mattip: force build, setitem2d]
<tos9> how is the lib-python directory physically assembled
<tos9> it seems probably like more than just `rm -rf /pypy/lib-python/3/ && cp -R /cpython/Lib/ /pypy/lib-python/3/`
<tos9> or is it not? There's e.g. plat- directories that don't appear in my cpython checkout
<cfbolz> tos9: there are modifications we do to some files
<mattip> we copy it on to a branch, then merge that branch into default. So you get the fixes we have re-applied to the cpython update
<mattip> ahh, cross-post, sorry
<cfbolz> but if we have too many files, it's possible that we forget to delete them
<tos9> mattip: a ha, ok
<tos9> cfbolz: yeah there are some files that aren't in the cpython tree
<tos9> ok, maybe lemme see if I do this roughly correctly then
<mattip> tos9: look at the stdlib-2.7.16 branch commit 51ed8b3f8235 from about three weeks ago
xcm has quit [Ping timeout: 245 seconds]
* tos9 checks
<tos9> ah right, I was going to ask
<tos9> mattip: so you merge default first into that branch
<tos9> whee and I've borked my checkout already
<tos9> ⊙ hg merge --abort -> abort: no merge in progress
<tos9> ⊙ hg merge py3.6 -> abort: outstanding merge conflicts
xcm has joined #pypy
<tos9> :/
<tos9> (ok hg update -C apparently.)
<mattip> take a look at lib-python/stdlib-upgrade.txt
<tos9> aha, thanks
speeder39_ has quit [Quit: Connection closed for inactivity]
kipras has joined #pypy
dansan has quit [Remote host closed the connection]
dansan has joined #pypy
dansan has quit [Remote host closed the connection]
dansan has joined #pypy
dansan has quit [Excess Flood]
dansan has joined #pypy
themsay has quit [Ping timeout: 272 seconds]
<bbot2> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/6280 [mattip: force build, setitem2d]
themsay has joined #pypy
lritter has quit [Quit: Leaving]
dansan has quit [Remote host closed the connection]
dansan has joined #pypy
speeder39_ has joined #pypy
<kenaan> mattip freelibrary ec338ff9ef03 /: export FreeLibrary, non-closing LoadLibrary from rwin32
<bbot2> Started: http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/4678 [mattip: force build, freelibrary]
<bbot2> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/4678 [mattip: force build, freelibrary]
tsaka__ has quit [Ping timeout: 245 seconds]
fryguybob has quit [Quit: Lost terminal]