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
asmeurer__ has joined #pypy
ludicrux has joined #pypy
jcea has quit [Remote host closed the connection]
jcea has joined #pypy
_whitelogger has joined #pypy
speeder39_ has quit [Quit: Connection closed for inactivity]
jcea has quit [Remote host closed the connection]
antocuni has quit [Ping timeout: 252 seconds]
ludicrux has quit [Quit: This computer has gone to sleep]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
energizer has quit [Quit: ZNC 1.7.0+deb0+xenial1 - https://znc.in]
asmeurer__ has quit [Quit: asmeurer__]
ludicrux has joined #pypy
Garen_ has joined #pypy
Garen has quit [Ping timeout: 245 seconds]
ludicrux has quit [Quit: This computer has gone to sleep]
lritter has joined #pypy
Garen_ has quit [Read error: Connection reset by peer]
Garen has joined #pypy
xcm has quit [Remote host closed the connection]
irclogs_io_bot has quit [Remote host closed the connection]
xcm has joined #pypy
kostia has joined #pypy
energizer has joined #pypy
dddddd has joined #pypy
<mattip> rerunning the package.pypy.org builder after updating to pypy3 7.0, https://github.com/pypy/pypy.packages
<mattip> it turns out pypy3 7.0 cannot install zmq, which is a requirement for other packages like pandas
<mattip> I would like to release 7.1 (which can install zmq) soon, once the newmemoryview-app-level branch lands
antocuni has joined #pypy
rubdos has quit [Remote host closed the connection]
rubdos has joined #pypy
dddddd_ has joined #pypy
dddddd has quit [Ping timeout: 252 seconds]
dddddd_ is now known as dddddd
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
<kenaan> fijal arm64 173ed1a7572d /rpython/jit/backend/: enough scaffolding to actually run the first loop test. It does not pass yet, but it does run!
<fijal> arigato: maaaaybe getting somewhere
darkman66 has joined #pypy
antocuni has quit [Ping timeout: 255 seconds]
xcm has quit [Read error: Connection reset by peer]
xcm has joined #pypy
zaytsev has quit [Quit: leaving]
irclogs_io_bot has joined #pypy
<kenaan> fijal arm64 91919b103009 /rpython/jit/backend/aarch64/: enough to successfully exit the loop
lritter has quit [Ping timeout: 245 seconds]
kostia has quit [Remote host closed the connection]
lritter has joined #pypy
Zaab1t has joined #pypy
Zaab1t has quit [Quit: bye bye friends]
kostia has joined #pypy
kostia has quit [Ping timeout: 240 seconds]
kostia has joined #pypy
lesshaste has joined #pypy
<lesshaste> mattip, what's the current status with numpy and cffi? What I mean is, there were suggestions at one point for how to make it fast
<mattip> and thinking about adapting some type of cython-to-IR-to-X where X will be C or PyPy
<lesshaste> mattip, sorry I always get cffi and cpyext mixed up
<lesshaste> thanks.. that looks very interesting
<mattip> there is a third way, but it will not play well with the ecosystem that deeply embed numpy (pandas),
<lesshaste> when you say deeply embed, do you mean they don't just call the functions ?
<mattip> which is to go back to micronumpy via the __array_function__ and __array_ufunc__ numpy protocols
<mattip> pandas uses the numpy c-api and alot of cython
<lesshaste> aha
<lesshaste> was micronumpy where you were reimplementing numpy in rpython?
<mattip> +1
<mattip> cffi is still the best answer
<lesshaste> reimplementing numpy seemed a little crazy to me. Isn't that just too much work?
<lesshaste> like one thousand pieces of university coursework
<mattip> the problem is more that it is a moving target. See the way dask, numba, cupy approach numpy
<mattip> they only reimplement the bits they have to, and inherit the rest
<lesshaste> got you
<mattip> cupy, since it came last, has the cleanest interface
<mattip> and is very similar to micronumpy (except it only has limited dtypes)
<kenaan> fijal arm64 144deb45999c /rpython/jit/backend/: pass the first loop test!
<mattip> fijal: yay!
<fijal> mattip: beginning of the journey, but maaaybe?
kostia has quit [Remote host closed the connection]
kostia has joined #pypy
kostia has quit [Remote host closed the connection]
Rhy0lite has joined #pypy
<lesshaste> mattip, just looked up cupy.. I am learning :)
kostia has joined #pypy
<mattip> the one that does numpy on the gpu with cuda
<lesshaste> yes, looks intriguing
<lesshaste> but I wish people wouldn't use cuda and would use opencl and openmp instead
kostia has quit [Remote host closed the connection]
<mattip> lesshaste: +1, although practicality beats purity
<lesshaste> mattip, ... but coding for CUDA is harder so for me it's less practical
<lesshaste> I mean if you are coding in C/C++ . I have never tried doing it from a higher level language
<mattip> harder than opencl? Or do you use tools to convert your code
<mattip> the tooling around cuda is really nice. The profiler shows loads of information
<lesshaste> (no tools used)
<mattip> haven;t tried opencl yet
<lesshaste> CUDA doesn't even have vectors does it?
<lesshaste> it's just a pain
<mattip> to each their own. I don't like c++, and like the std library even less
<lesshaste> ah ok :)
<lesshaste> what libraries do you use instead?
<lesshaste> I can't say I love the std libraries but it seems silly to reimplement everything yourself
<mattip> maybe our goals are different. I haven't really found a good use case for std::vectors personally
<mattip> but thrust works well for cuda
<mattip> at least that's what they tell me
<lesshaste> yes I think that's right
<lesshaste> I think I am just old fashioned and want the same code to compileable on my laptop as on a GPU
<lesshaste> silly question but say you want to a 2d matrix of user defined size. You can make an array of arrays but then you are dereferencing all the time
<mattip> to each their own
<lesshaste> that's what I ended up using vectors for.. but maybe it's foolish
<lesshaste> I don't claim C++ expertise!
<mattip> the error message when using templated c++ drive me crazy
<lesshaste> good point!
<lesshaste> In theory gcc is improving those
<mattip> anyhow, back to pypy
<lesshaste> https://gcc.gnu.org/gcc-8/changes.html (search for template)
<lesshaste> yes :)
kostia has joined #pypy
<kenaan> fijal arm64 13d01ec04e4e /rpython/jit/backend/aarch64/: basic IMM support in add
<mattip> ronan: perhaps we should add a c implementation of PyTuple_Size to cpyext/src/tupleobject.c, and use it for tp_len
<kenaan> fijal arm64 88c9f1531c0d /rpython/jit/backend/aarch64/: int_sub
jcea has joined #pypy
<mattip> although it seems we are not yet using c code for tp_* calls
<mattip> antocuni (for the logs) ^^^
antocuni has joined #pypy
<kenaan> fijal arm64 a50f3a71ef1d /rpython/jit/backend/aarch64/: bridge basics
<kenaan> fijal arm64 d301852919b9 /rpython/jit/backend/aarch64/: in progress
darkman66 has quit [Remote host closed the connection]
NemeXis has joined #pypy
Ai9zO5AP has joined #pypy
i9zO5AP has joined #pypy
Ai9zO5AP has quit [Ping timeout: 255 seconds]
<phlebas> fijal: you around?
<fijal> phlebas: yeah
<phlebas> sorry, I seem to suck at email threading. what did we decide? should we have a phone call? just discuss here?
<fijal> phlebas: you were supposed to decide with lukasz, do you need me to organize that?
<phlebas> ah, ok, I understood you wanted to be part :) then I'll just ping lukasz
<kenaan> antocuni default c42a88cbac10 /pypy/doc/: update the docs about GC hooks: the unit of the duration fields was changed by the expose-gc-time branch, but th...
Taggnostr has quit [Ping timeout: 264 seconds]
i9zO5AP has quit [Quit: WeeChat 2.4]
darkman66 has joined #pypy
jcea has quit [Remote host closed the connection]
jcea has joined #pypy
darkman66 has quit [Ping timeout: 240 seconds]
NemeXis has quit [Remote host closed the connection]
Taggnostr has joined #pypy
<kenaan> arigo default ddbc46f09114 /pypy/module/_cffi_backend/test/_backend_test_c.py: update to cffi/6d435908617d
<kenaan> arigo default dc08ce14d28b /: update to cffi/6d435908617d (really this time)
<kenaan> arigo default 28c9c798ba05 /lib_pypy/cffi.egg-info/PKG-INFO: update to cffi/6d435908617d (3rd try)
Ai9zO5AP has joined #pypy
ludicrux has joined #pypy
marky1991 has joined #pypy
antocuni has quit [Ping timeout: 240 seconds]
marky1991 has quit [Ping timeout: 245 seconds]
ludicrux has quit [Quit: This computer has gone to sleep]
ludicrux has joined #pypy
kostia has quit [Remote host closed the connection]
kostia has joined #pypy
jacob22 has quit [Read error: Connection reset by peer]
slam_V has joined #pypy
jacob22 has joined #pypy
Rhy0lite has quit [Quit: Leaving]
slam_V has quit [Quit: Mutter: www.mutterirc.com]
slam_V has joined #pypy
slam_V has quit [Client Quit]
kostia has quit [Remote host closed the connection]
kostia has joined #pypy
kostia has quit [Ping timeout: 240 seconds]
darkman66 has joined #pypy
marky1991 has joined #pypy
darkman66 has quit [Remote host closed the connection]
slam_V has joined #pypy
darkman66 has joined #pypy
marvin has quit [Remote host closed the connection]
marvin has joined #pypy
slam_V has quit [Remote host closed the connection]
marky1991 has quit [Ping timeout: 240 seconds]
antocuni has joined #pypy
ludicrux has quit [Quit: This computer has gone to sleep]
darkman66 has quit [Remote host closed the connection]
kostia has joined #pypy
ludicrux has joined #pypy
darkman66 has joined #pypy
kostia has quit [Ping timeout: 240 seconds]
zmt01 has joined #pypy
zmt00 has quit [Ping timeout: 240 seconds]
kostia has joined #pypy
kostia has quit [Remote host closed the connection]
darkman66 has quit [Remote host closed the connection]
<mattip> fijal: buildbot.pypy.org is saying "PTY allocation request failed on channel 0", so I can't log in
<mattip> doing ssh buildbot.pypy.org "/bin/bash -i" seems to let me in, but since there is no tty sudo fails
<fijal> mattip: I forwarded you instructions how to fix it
<mattip> +1, thanks
darkman66 has joined #pypy
<kenaan> mattip buildbot 4cc524e68457 /bot2/pypybuildbot/master.py: add temporary 32bit buildbot, hopefully bencher4 will return
energizer` has joined #pypy
bbot2 has quit [Quit: buildmaster reconfigured: bot disconnecting]
bbot2 has joined #pypy
kostia has joined #pypy
kostia has quit [Ping timeout: 252 seconds]
Zaab1t has joined #pypy
Zaab1t has quit [Quit: bye bye friends]
Ai9zO5AP has quit [Quit: WeeChat 2.4]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
darkman66 has quit [Remote host closed the connection]
lritter has quit [Ping timeout: 245 seconds]
jacob22 has quit [Read error: Connection reset by peer]
jacob22 has joined #pypy