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
dddddd has quit [Remote host closed the connection]
jcea has quit [Quit: jcea]
adamholmberg has joined #pypy
adamholmberg has quit [Ping timeout: 265 seconds]
lweber1 has joined #pypy
lweber1 has quit [Ping timeout: 256 seconds]
zmt00 has quit [Read error: Connection reset by peer]
zmt00 has joined #pypy
Garen has joined #pypy
mcyprian has joined #pypy
mcyprian has quit [Client Quit]
lweber1 has joined #pypy
twb has joined #pypy
<twb>
In Debian 9, I can see pypy=2.4.0+dfsg-3 but no pypy3 package --- does that mean I can only run pypy in "python2 mode" ?
<twb>
When I run that pypy without arguments it says "Python 2.7.8" and I can't see a CLI option to make it be Python 3, so I guess I can't have Python3-style pypy on Debian 9, unless I build it myself or use a PPA
<kenaan>
mattip py3.5 bcbefe2f5e14 /: merge default into branch
<kenaan>
mattip release-pypy2.7-6.x cda3e08c8c81 /: merge default into release
<kenaan>
mattip release-pypy3.5-6.x 4b3a0384cf33 /: merge py3.5 into release
<fijal>
mattip: armin said at some stage he will not be much online for the next week or so
<fijal>
that's usually a lie, but he might have longer pings
adamholmberg has joined #pypy
antocuni has joined #pypy
<kenaan>
antocuni gc-hooks df2490d5d814 /pypy/: argh, this was a bad bug: make sure to clear action._next after perform(), else the next time we put it in the ...
jcea has joined #pypy
jcea has quit [Remote host closed the connection]
jcea has joined #pypy
adamholmberg has quit [Remote host closed the connection]
<antocuni>
cool, the gc-hooks branch seems to work well; and, most importantly, enabling the hooks does not seem to cause any slowdown (even if I make a print on every minor collection)
adamholmberg has joined #pypy
raynold has quit [Quit: Connection closed for inactivity]
adamholmberg has quit [Ping timeout: 260 seconds]
forgottenone has joined #pypy
mcyprian has joined #pypy
[Arfreve1] has quit [Quit: leaving]
mcyprian has quit [Ping timeout: 240 seconds]
pf_moore has joined #pypy
[Arfrever] has joined #pypy
oberstet has quit [Ping timeout: 264 seconds]
<mattip>
antocuni: nice. when you merge it, please merge to py3.5 as well. Will be a nice addition to PyPy6
<antocuni>
isn't it better to merge it automatically the next time we merge default to py3.5?
mcyprian has joined #pypy
<mattip>
"automatic" means "whoever wants to update py3.5", and it is sometimes unclear how to resolve merge failures
<antocuni>
if I am not mistaken, this is the first release which contains the big cpyext speedups which we did in cpyext-avoid-roundtrip. Maybe it is worth highlighting them more?
<antocuni>
ah no
<antocuni>
they were already in 5.10
<antocuni>
uh no, maybe not. I'm confused :)
<mattip>
I think they made it into 5.10.1, which was only released for py3.5 so anyhow we can emphasize it more
<mattip>
antocuni: could you summarize the GC hook options in the docs somewhere (docstrings? rpython/docs/garbage_collection.rst ? ) and the expected output ?
<mattip>
maybe also another blog post?
<antocuni>
yes sure, both are on my todo list
<antocuni>
but I need to finalize the branch first
<mattip>
cool, it is a really nice addition
<antocuni>
uh, don't we have a way to do space.getattr_xxx(w_obj, 'foo')?
<antocuni>
i.e., to pass 'foo' as an unwrapped string?
<kenaan>
antocuni gc-hooks 47b121de440b /pypy/module/gc/: chage the API: instead of using the ugly set_hook (and having to implement get_on_gc_*), we expose a hooks obje...
<kenaan>
antocuni gc-hooks 1298b7800106 /pypy/module/gc/hook.py: this is a wrapped class now, better to use the w_* naming convention
<kenaan>
antocuni gc-hooks db58a930e0f9 /pypy/module/gc/: add two convenience methods: hooks.set to pass an instance which contains the appropriate on_gc_* methods, and ...
<mattip>
there is also space.findattr
adamholmberg has joined #pypy
<antocuni>
too late, I used space.getattr(... space.newtext()) :)