#pypy on 2018-04-01 — irc logs at freenode.irclog.whitequark.org

2018-02-26 15:52 cfbolz changed the topic of #pypy to: PyPy, the flexible snake (IRC logs: https://botbot.me/freenode/pypy/ ) | use cffi for calling C | the secret reason for us trying to get PyPy users: to test the JIT well enough that we're somewhat confident about it

00:00 <bbot2_> Started: http://buildbot.pypy.org/builders/pypy-c-jit-linux-s390x/builds/824 [default]

00:30 marr has quit [Ping timeout: 264 seconds]

00:35 <bbot2_> Success: http://buildbot.pypy.org/builders/build-pypy-c-jit-linux-armhf-raspbian/builds/1953 [default]

00:35 <bbot2_> Started: http://buildbot.pypy.org/builders/pypy-c-jit-linux-armhf-raspbian/builds/1674 [default]

01:00 <bbot2_> Started: http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/5376 [py3.5]

01:00 <bbot2_> Started: http://buildbot.pypy.org/builders/own-linux-x86-64/builds/6655 [py3.5]

01:00 <bbot2_> Started: http://buildbot.pypy.org/builders/own-linux-x86-32/builds/5776 [py3.5]

01:00 <bbot2_> Started: http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/4542 [py3.5]

01:00 <bbot2_> Started: http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/3813 [py3.5]

01:00 <bbot2_> Started: http://buildbot.pypy.org/builders/build-pypy-c-jit-linux-armhf-raspbian/builds/1954 [py3.5]

01:00 <bbot2_> Failure: http://buildbot.pypy.org/builders/own-linux-x86-32/builds/5776 [py3.5]

01:05 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

01:13 <bbot2_> Success: http://buildbot.pypy.org/builders/build-pypy-c-jit-linux-armel/builds/2195 [default]

01:13 <bbot2_> Started: http://buildbot.pypy.org/builders/build-pypy-c-jit-linux-armel/builds/2196 [py3.5]

01:22 pf_moore has quit [Quit: Connection closed for inactivity]

01:56 <bbot2_> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/3679 [default]

01:56 <bbot2_> Started: http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/3680 [py3.5]

02:09 asmeurer_ has quit [Quit: asmeurer_]

02:10 <bbot2_> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/5376 [py3.5]

02:14 tbodt has joined #pypy

02:15 dddddd has quit [Remote host closed the connection]

02:19 <bbot2_> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/4542 [py3.5]

02:24 <bbot2_> Failure: http://buildbot.pypy.org/builders/own-linux-x86-64/builds/6655 [py3.5]

02:42 <bbot2_> Success: http://buildbot.pypy.org/builders/build-pypy-c-jit-linux-armel/builds/2196 [py3.5]

02:58 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

03:00 <bbot2_> Started: http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/2493

03:00 <bbot2_> Started: http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/3291

03:12 <bbot2_> Success: http://buildbot.pypy.org/builders/build-pypy-c-jit-linux-armhf-raspbian/builds/1954 [py3.5]

03:12 <bbot2_> Started: http://buildbot.pypy.org/builders/pypy-c-jit-linux-armhf-raspbian/builds/1675 [py3.5]

03:24 energizer has quit [Remote host closed the connection]

03:28 asmeurer__ has joined #pypy

03:41 jacob22_ has joined #pypy

03:42 jacob22__ has quit [Ping timeout: 264 seconds]

04:04 tbodt has joined #pypy

04:35 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

04:36 tbodt has joined #pypy

04:40 tbodt has quit [Client Quit]

04:41 tbodt has joined #pypy

04:50 <bbot2_> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/3812 [default]

04:58 forgottenone has joined #pypy

04:59 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

05:00 tbodt has joined #pypy

05:02 tbodt has quit [Client Quit]

05:03 energizer has joined #pypy

05:03 tbodt has joined #pypy

05:23 <bbot2_> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/3680 [py3.5]

05:25 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

05:25 tbodt has joined #pypy

05:26 tbodt has quit [Client Quit]

05:39 tbodt has joined #pypy

05:46 <kenaan> mattip jit-hooks-can-be-disabled bc36eabcc716 /pypy/module/cpyext/pyobject.py: close merged branch

05:46 <kenaan> mattip default 50bc92bdf69a /pypy/doc/whatsnew-head.rst: document merged branch

05:48 <cfbolz> mattip: bc36eabcc716 contains code changes

05:48 <cfbolz> (it doesn't matter, but in case you are wondering where your local modifications went)

05:48 <mattip> thanks

05:49 <cfbolz> Thank you for documenting my branch, sorry for forgetting

05:49 <mattip> :)

05:49 <cfbolz> So, I have one branch left to merge, which I will do today

05:50 <mattip> cool, armin found the source of the bug I was investigating, now I need to find an efficient fix

05:51 <mattip> I guess unicode-utf8 will not make this release :(

05:53 <kenaan> mattip default eb7ebd3110b0 /pypy/module/cpyext/pyobject.py: remove redundant code, this is done in track_reference

05:54 <cfbolz> mattip: is the bug related to my optimization?

05:55 <mattip> nope, related to the difference bewteen rawrefcount.create_link_pyobj and rawrefcount.create_link_pypy

05:55 <cfbolz> Ugh

05:55 <cfbolz> That sounds subtle

05:56 jamesaxl has joined #pypy

05:56 <mattip> for subclasses of CPyObject (c-api class) created in app-level python

05:57 <mattip> most likely if the construction does not call __new__, but I am still investigating the subclass creation pathway

05:58 <mattip> and how it avoids __new__

06:00 <mattip> maybe calling __new__ of the base class with the subclass typedef ?

06:00 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

06:08 <bbot2_> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/3813 [py3.5]

06:08 commandoline has quit [Read error: Connection reset by peer]

06:12 inhahe__ has joined #pypy

06:13 commandoline has joined #pypy

06:13 tbodt has joined #pypy

06:16 inhahe_ has quit [Ping timeout: 264 seconds]

06:19 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

06:21 tbodt has joined #pypy

06:48 <mattip> the object creation goes through tp_alloc, which is PyType_GenericAlloc, and at that point has no ob_pypy_link

06:54 <mattip> then it calls the python level __array_finalize__, which ends up creating a w_obj which is linked to ob_pypy_link

07:10 <bbot2_> Success: http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/3291

07:41 marr has joined #pypy

08:02 forgottenone has quit [Ping timeout: 260 seconds]

08:05 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

08:09 tbodt has joined #pypy

08:12 tbodt has quit [Client Quit]

08:51 energizer has quit [Quit: Leaving]

09:26 forgottenone has joined #pypy

09:33 oberstet has joined #pypy

09:36 <kenaan> mattip cpyext-subclass-setattr 90146fea6553 /pypy/module/cpyext/test/: add a test to show how GC inappropriately kills connection btwn pyobj and w_obj Passes on CPython...

09:55 asmeurer__ has quit [Quit: asmeurer__]

10:08 <bbot2_> Success: http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/2493

10:20 xorAxAx has quit [Read error: Connection reset by peer]

10:37 <kenaan> cfbolz default 045483a22d66 /: merge pyparser-improvements: - speed up python parser - create slightly more helpful SyntaxError message (only in...

10:54 hniksic has joined #pypy

10:59 hniksic has quit [Client Quit]

11:04 pf_moore has joined #pypy

11:15 dddddd has joined #pypy

11:49 marr has quit [Ping timeout: 248 seconds]

12:23 irclogs_io_bot has joined #pypy

12:26 <kenaan> mattip cpyext-subclass-setattr 2315922104c7 /pypy/module/cpyext/: restore rpython Py_IncRef, Py_DecRef for PYPY_DEBUG_REFCOUNT, old versions segfaulted

12:36 <kenaan> mattip cpyext-subclass-setattr 866aa67a514f /pypy/module/cpyext/test/array.c: remove excess Py_XDECREF, test now passes untranslated unfortunately

12:37 <mattip> still fails when run -A on default

13:58 <bbot2_> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-linux-s390x/builds/824 [default]

14:10 antocuni has joined #pypy

14:16 <kenaan> mattip cpyext-subclass-setattr 3f81ccde1932 /pypy/module/cpyext/pyobject.py: fix for NULL obj in Py_IncRef

14:47 forgottenone has quit [Ping timeout: 260 seconds]

14:51 forgottenone has joined #pypy

14:55 antocuni has quit [Ping timeout: 276 seconds]

15:00 <kenaan> mattip cpyext-subclass-setattr 36ff2c150968 /pypy/: be more selective in setting w_obj.flag_cpytype, fixes translated test

15:05 <mattip> arigato: could you take a look when you get a chance, I think this solves the problem and only hurts performance for user-defined classes

15:15 tbodt has joined #pypy

15:16 tbodt has quit [Client Quit]

15:18 tbodt has joined #pypy

15:19 tbodt has quit [Client Quit]

15:22 tbodt has joined #pypy

15:29 Hasimir has quit [Read error: Connection reset by peer]

15:31 Hasimir has joined #pypy

15:31 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

15:37 tbodt has joined #pypy

15:44 Hasimir has quit [Read error: Connection reset by peer]

15:51 tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]

15:57 Hasimir has joined #pypy