antocuni changed the topic of #pypy to: PyPy, the flexible snake (IRC logs: https://botbot.me/freenode/pypy/ ) | use cffi for calling C | "PyPy: the Gradual Reduction of Magic (tm)"
traverseda has quit [Ping timeout: 240 seconds]
lritter has quit [Remote host closed the connection]
traverseda has joined #pypy
oberstet has quit [Ping timeout: 258 seconds]
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tbodt has joined #pypy
marr has quit [Ping timeout: 240 seconds]
traverseda has quit [Ping timeout: 240 seconds]
traverseda has joined #pypy
marky1991 has quit [Ping timeout: 240 seconds]
marky1991 has joined #pypy
marky1991_2 has joined #pypy
marky1991_2 has quit [Remote host closed the connection]
marky1991 has quit [Ping timeout: 240 seconds]
marky1991_2 has joined #pypy
Mepho has joined #pypy
<Mepho> hello, why does the build process need to access '/proc/1/status' (e.g. init)? I have my procfs mounted with hidepid=2 and gentoo doesn't compile under root hence pid1 isn't observable to the build environment.
<Mepho> should I even try to report that on issue tracker or is that way bellow radar and discarded as #justgentoothings ?
<Mepho> ugh, call to detect_pax, sounds like gentoo hardened people (or whoever uses anti-JIT protections in pax/grsecurity) are to blame for this
drolando_ has quit [Remote host closed the connection]
drolando has joined #pypy
marky1991_2 has quit [Read error: Connection reset by peer]
<Mepho> 's#open("/proc/1/status)"#open("/proc/self/status")#'
marky1991 has joined #pypy
marky1991 has quit [Remote host closed the connection]
<Alex_Gaynor> Mepho: send a PR, sounds reasonable to me
<Mepho> Alex_Gaynor: ack
pilne has quit [Remote host closed the connection]
_whitelogger has joined #pypy
ArneBab_ has joined #pypy
ArneBab has quit [Ping timeout: 240 seconds]
jcea has quit [Remote host closed the connection]
jcea has joined #pypy
jcea has quit [Quit: jcea]
<Mepho> damn, I suck at mercurial, now I need to learn to squash commits before anybody notices my PR
forgottenone has joined #pypy
<Alex_Gaynor> Mepho: `hg histedit`
<Mepho> ^^
Mepho has quit [Ping timeout: 258 seconds]
_whitelogger has joined #pypy
bmerry has joined #pypy
<bmerry> Hi. I'm hoping someone can offer a suggestion for a problem I'm having supporting PyPy in an existing C API extension module. I need to run some cleanup code at shutdown to gracefully stop a worker thread, and it currently uses the approach at https://pybind11.readthedocs.io/en/stable/advanced/misc.html?highlight=cleanup#module-destructors (add a capsule to the module with a destructor). However, it doesn't seem to be getting triggered on PyPy2
<njs> bmerry: pypy does not run destructors at shutdown time, so that's not going to work. You might want to look into the atexit module, or see if you can make it a daemon thread.
<bmerry> nm, looks like atexit will work for me.
<njs> there are also low-level API tricks you might be able to use, like https://stackoverflow.com/questions/10390738/calling-a-function-when-thread-is-exiting-in-pthreads-or-windows
<bmerry> I suspect C-level tricks might be a bit too late: I need to run the cleanup before the GIL disappears.
<bmerry> Because the thread makes calls into Python, so if it carries on running for too long, it tries to take the GIL and crashes.
<njs> sounds tricky
<bmerry> So far atexit seems to be working, but I'll need to do some more tests to make sure it doesn't happen too early or too late.
<njs> good luck
<bmerry> Thanks :-)
forgottenone has quit [Ping timeout: 240 seconds]
forgottenone has joined #pypy
Garen has quit [Read error: Connection reset by peer]
Garen has joined #pypy
drolando has quit [Remote host closed the connection]
zmt00 has quit [Quit: Leaving]
slackyy has quit [Ping timeout: 248 seconds]
oberstet has joined #pypy
tormoz has quit [Remote host closed the connection]
bmerry has left #pypy [#pypy]
drolando has joined #pypy
antocuni has joined #pypy
<bbot2> Started: http://buildbot.pypy.org/builders/rpython-linux-x86-64/builds/6 [antocuni: force build, vmprof-0.4.10]
<antocuni> of course I started "own" tests, but what I wanted is "rpython" :(
<bbot2> Exception: http://buildbot.pypy.org/builders/rpython-linux-x86-64/builds/6 [antocuni: force build, vmprof-0.4.10]
<kenaan> antocuni vmprof-0.4.10 4134cbd25c42 /requirements.txt: try to install vmprof and see what happens to test_enable and test_native
<bbot2> Started: http://buildbot.pypy.org/builders/rpython-linux-x86-64/builds/7 [antocuni: force build, vmprof-0.4.10]
antocuni has quit [Ping timeout: 240 seconds]
<bbot2> Failure: http://buildbot.pypy.org/builders/rpython-linux-x86-64/builds/7 [antocuni: force build, vmprof-0.4.10]
raynold has quit [Quit: Connection closed for inactivity]
<kenaan> rlamy assert-rewrite b7b55dee74c6 /rpython/: Fix rpython/memory/ tests
<bbot2> Started: http://buildbot.pypy.org/builders/rpython-linux-x86-64/builds/8 [ronan: force build, assert-rewrite]
<ronan> hmm, looks like kenaan choked on that big merge commit (13a3a37)
Mepho has joined #pypy
Mepho has quit [Changing host]
Mepho has joined #pypy
<bbot2> Failure: http://buildbot.pypy.org/builders/rpython-linux-x86-64/builds/8 [ronan: force build, assert-rewrite]
tormoz has joined #pypy
<kenaan> rlamy assert-rewrite b994fd06043b /rpython/rtyper/: Fix asserts in rpython/rtyper/
pilne has joined #pypy
<kenaan> rlamy assert-rewrite fe04fd3b8632 /rpython/translator/c/test/test_newgc.py: Fix asserts in test_newgc.py
<bbot2> Started: http://buildbot.pypy.org/builders/rpython-linux-x86-64/builds/9 [ronan: force build, assert-rewrite]
kipras`away is now known as kipras
<kenaan> rlamy default 50ba491d0e92 /rpython/jit/metainterp/test/test_del.py: kill test that has been disabled for 6 years
mattip has joined #pypy
<mattip> ronan: commit 0ba20064633a in buildbot needs to be tweaked, the -p argument to virtuaenv should be different for linux + default, win32 + default, linux + py3.5, py3.5 + win32
<mattip> see failing tests
<mattip> the -p arg should probably come from the pypy repo itself rather than the buildbot
<bbot2> Failure: http://buildbot.pypy.org/builders/rpython-linux-x86-64/builds/9 [ronan: force build, assert-rewrite]
dan- has quit [Max SendQ exceeded]
lritter has joined #pypy
dan- has joined #pypy
dan- has quit [Changing host]
dan- has joined #pypy
jcea has joined #pypy
zmt00 has joined #pypy
lritter has quit [Remote host closed the connection]
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
jamesaxl has quit [Read error: Connection reset by peer]
jamesaxl has joined #pypy
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
tbodt has quit [Client Quit]
slackyy has joined #pypy
tbodt has joined #pypy
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
ssbr has quit [Ping timeout: 246 seconds]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
raynold has joined #pypy
tbodt has quit [Client Quit]
ssbr has joined #pypy
tbodt has joined #pypy
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tbodt has joined #pypy
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
tav` has quit [Quit: tav`]
tav has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
tav has quit [Ping timeout: 260 seconds]
tbodt has quit [Client Quit]
MrBismuth has joined #pypy
tbodt has joined #pypy
tbodt has quit [Client Quit]
MrBusiness3 has quit [Ping timeout: 252 seconds]
<kenaan> mattip default bab05da3f317 /lib_pypy/_tkinter/app.py: add method used in matplotlib
<mattip> the good news : this app-level fix makes matplotlib basically work on my branch
<mattip> the bad news: something seems of with threading and the tkinter module,
<mattip> opening the "configure subplot" only sometimes reacts to clicks on the sliders if a pdb.set_trace() is set and the stars align
<mattip> s/of/off/
<LarstiQ> mattip: ooh
<LarstiQ> although I basically only use matplotlib from jupyter these days
tbodt has joined #pypy
<ronan> mattip: I haven't found an easy way to extract the filename of the target, though
<mattip> we have two models both based on running a python script and then calling subprocess.Popen -
gutworth_ has joined #pypy
<mattip> there is package.py which hard codes it into the script, and -A tests that use a config.cfg file
<mattip> then in the py3.5 branch we set the exe name to something different
<mattip> I guess the script that builds the virtualenv could copy the exe to a standard name inside the virtualenv for the next build step
Eyess has joined #pypy
SunDwarf has quit [Disconnected by services]
Eyess is now known as SunDwarf
akash47_ has joined #pypy
forgottenone has quit [Quit: Konversation terminated!]
raynold has quit [*.net *.split]
bbot2 has quit [*.net *.split]
rguillebert has quit [*.net *.split]
akash47 has quit [*.net *.split]
Alex_Gaynor has quit [*.net *.split]
pchiusano has quit [*.net *.split]
gutworth has quit [*.net *.split]
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
jimbaker has quit [Ping timeout: 260 seconds]
tbodt has joined #pypy
tbodt has quit [Client Quit]
jimbaker has joined #pypy
tbodt has joined #pypy
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
<mattip> numpy tests run about 15% faster on the cpyext-avoid-roundtrip branch, but there is one more test failure than on default
<mattip> dt = np.dtype([('a', np.int8), ('b', np.float32, 3)]); dt[-1] raises an IndexError, but works on default
<mattip> type(dt) is numpy.dtype
<mattip> dt.__getitem__ is a wrapper_descriptor
marr has joined #pypy
<mattip> bye for now
mattip has left #pypy ["bye"]
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tbodt has joined #pypy
tbodt has quit [Client Quit]
tbodt has joined #pypy
tbodt has quit [Read error: Connection reset by peer]
tbodt has joined #pypy
<kenaan> rlamy assert-rewrite 7318052f560c /rpython/jit/metainterp/test/: more assert fixes
tbodt has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tbodt has joined #pypy
tbodt has quit [Ping timeout: 240 seconds]
tbodt has joined #pypy
<cfbolz> ronan: I think that branch needs discussion on pypy-dev. Replacing all asserts in rpython tests everywhere looks pretty intrusive
bbot2 has joined #pypy
rguillebert has joined #pypy
pchiusano has joined #pypy
Alex_Gaynor has joined #pypy
raynold has joined #pypy
yuvipanda has quit [Ping timeout: 255 seconds]
abvi[m]1 has quit [Ping timeout: 264 seconds]
pulkitg has quit [Ping timeout: 264 seconds]
bendlas has quit [Ping timeout: 276 seconds]
antocuni has joined #pypy
pulkitg has joined #pypy
_stian has joined #pypy
<_stian> Think the performance improvements is good enough - https://pastebin.com/nYfC24fD
<_stian> Especially for negative numbers stuff, https://pastebin.com/sb21DniC
<_stian> (For some weird reason, having a qrshift used for power of two division makes regular rshift faster).
kcr is now known as _kcr
yuvipanda has joined #pypy
abvi[m]1 has joined #pypy
bendlas has joined #pypy
<simpson> Hi. In RPython GC, I'd like to know when a runtime-level object dies, so that I can see whether it satisfied its duties before being reaped. (Think like a Twisted Deferred, but runtime-level only, with no callbacks to user-level code.)
<simpson> Do I need to do FinalizerQueue?
tav has joined #pypy
squeaky_pl has quit [Ping timeout: 240 seconds]
<raynold> ahh it's a wonderful day