#pypy IRC log for Wednesday, 2012-01-25

aboudreault (~alanb@osgeo/member/aboudreault) left irc: Quit: Leaving00:04
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) left irc: Read error: Connection reset by peer00:06
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) joined #pypy.00:07
CIA-1703papercrane 07roundup * 10#1020/segfault when running test_greenlet.py through pypy pytest.py: 00:07
CIA-17[chatting] Not sure if this is the same segfault but if I try to run my jevent both.py00:07
CIA-17example (see Issue #1016 for tarball) through py.py now t ... * 14https://bugs.pypy.org/issue102000:07
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) left irc: Read error: Connection reset by peer00:09
mattip (~mattip@bzq-79-177-109-132.red.bezeqint.net) left irc: Quit: bye00:11
azanella (~azanella@189.6.80.131) left irc: Ping timeout: 240 seconds00:11
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) joined #pypy.00:13
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) left irc: Read error: Connection reset by peer00:13
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) joined #pypy.00:14
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) left irc: Read error: Connection reset by peer00:27
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) joined #pypy.00:27
csenger (~user@dslb-084-058-161-075.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds00:39
nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 248 seconds00:45
nedbat (~nedbat@python/psf/nedbat) joined #pypy.00:50
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) left irc: Ping timeout: 248 seconds00:51
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) joined #pypy.00:53
dmalcolm_ (david@nat/redhat/x-yuvxhvolunasrtsm) left irc: Read error: Operation timed out00:54
kenaan12mattip numppy-flatitter 11d73d63090af7 15/pypy/module/micronumpy/: setitem, getitem work, add failing tests for comparison ops01:02
kenaan12mattip numppy-flatitter 11d1a6dc38d5b5 15/pypy/module/micronumpy/interp_numarray.py: remove debug cruft01:02
bbot24Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/78601:04
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection01:30
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/4201:33
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/103101:35
voidspace (~voidspace@python/psf/voidspace) left irc: Quit: Goodbye...01:47
kvda (~kvda@124-169-134-221.dyn.iinet.net.au) joined #pypy.01:51
pnguyen (~pnguyen@cpe-70-112-214-119.austin.res.rr.com) joined #pypy.01:54
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/193902:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/68902:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/34902:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/33902:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/68602:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/121102:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/87102:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/153002:00
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/80002:00
whyking (~quassel@ip68-14-13-208.ri.ri.cox.net) left irc: Ping timeout: 252 seconds02:18
whyking (~quassel@ip68-14-13-208.ri.ri.cox.net) joined #pypy.02:22
DasIch (~DasIch@p4FFDE139.dip.t-dialin.net) left irc: Quit: DasIch02:37
papercrane (~papercran@75.101.111.82) left irc: Ping timeout: 252 seconds02:39
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Quit: Computer has gone to sleep.02:47
Kaskuka (~John@osbk-4db17b2e.pool.mediaWays.net) left irc: Ping timeout: 240 seconds03:04
Shanita (~John@osbk-4db15e46.pool.mediaWays.net) joined #pypy.03:04
pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) left irc: Ping timeout: 244 seconds03:10
dmalcolm_ (~david@c-24-61-12-82.hsd1.ma.comcast.net) joined #pypy.03:12
ericflo (~ericflo@75.103.8.110) left irc: Quit: ericflo03:17
nettok (~quassel@190.148.246.190) joined #pypy.03:19
Tobu (~Tobu@unaffiliated/tobu) left irc: Ping timeout: 252 seconds03:23
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/21503:24
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/33903:27
nirbheek_ (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy.03:30
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 252 seconds03:32
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.03:35
papercrane (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) joined #pypy.03:42
papercrane (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) left irc: Client Quit03:47
papercrane (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) joined #pypy.03:50
JustinP (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) joined #pypy.03:53
papercrane (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) left irc: Client Quit03:54
Nick change: JustinP -> papercrane03:55
papercrane (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) left irc: Quit: leaving03:58
JustinP (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) joined #pypy.03:58
JustinP (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) left irc: Client Quit03:59
papercrane (~papercran@c-76-103-172-115.hsd1.ca.comcast.net) joined #pypy.03:59
nedbat (~nedbat@python/psf/nedbat) left irc: Disconnected by services04:00
nedbat_ (~nedbat@python/psf/nedbat) joined #pypy.04:00
papercrane.04:03
Last message repeated 1 time(s).04:26
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/193904:26
whyking (~quassel@ip68-14-13-208.ri.ri.cox.net) left irc: Ping timeout: 244 seconds04:48
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.04:48
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Ping timeout: 244 seconds04:53
nettok (~quassel@190.148.246.190) left irc: Remote host closed the connection05:02
price (~price@BUZZWORD-BINGO.MIT.EDU) joined #pypy.05:04
anish_ (anish@nat/redhat/x-pjvidzznzhafrpvu) joined #pypy.05:16
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.05:22
nirbheek_ (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 240 seconds05:23
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/153005:31
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/87105:34
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/121105:49
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/68605:50
pnguyen (~pnguyen@cpe-70-112-214-119.austin.res.rr.com) left irc: Read error: Connection reset by peer05:57
pnguyen (~pnguyen@cpe-70-112-214-119.austin.res.rr.com) joined #pypy.05:57
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/68906:00
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Read error: Connection reset by peer06:06
nirbheek (~nirbheek@117.197.144.49) joined #pypy.06:27
nirbheek (~nirbheek@117.197.144.49) left irc: Changing host06:27
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy.06:27
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Read error: Connection reset by peer06:27
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/80007:00
asmeurer (~asmeurer@dhcp-altamirano-166.resnet.nmt.edu) left irc: Quit: asmeurer07:10
PiotrSikora (~none@nginx/adept/piotrsikora) left irc: Excess Flood07:13
PiotrSikora (~none@nginx/adept/piotrsikora) joined #pypy.07:14
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/34907:15
Tobu (~Tobu@unaffiliated/tobu) left irc: Read error: Operation timed out07:16
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) joined #pypy.07:22
JaRoel (~jaroel|4d@2001:980:7133:1:d865:c606:2463:4919) left irc: Remote host closed the connection07:35
pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) joined #pypy.07:45
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) left irc: Quit: DasIch07:47
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) joined #pypy.07:49
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.07:50
zain (~textual@bzq-79-182-196-50.red.bezeqint.net) joined #pypy.08:03
GeneralMaximus (~general@122.162.38.115) joined #pypy.08:04
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.08:08
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) left irc: Quit: DasIch08:09
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.08:09
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.08:10
fijal (~fijal@196-210-150-33.dynamic.isadsl.co.za) joined #pypy.08:12
fijalhi08:12
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection08:12
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.08:13
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Ping timeout: 245 seconds08:18
antocuni (~antocuni@host155-125-dynamic.25-79-r.retail.telecomitalia.it) joined #pypy.08:20
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.08:20
Varriount (~Varriount@hnng.public-craft.com) left irc: Ping timeout: 252 seconds08:26
dmitrey (~quassel@77.121.195.200) joined #pypy.08:27
Varriount (~Varriount@hnng.public-craft.com) joined #pypy.08:27
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Ping timeout: 244 seconds08:30
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.08:35
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds08:38
pnguyen (~pnguyen@cpe-70-112-214-119.austin.res.rr.com) left irc: Quit: pnguyen08:38
arigato (~arigo@adsl-89-217-207-222.adslplus.ch) joined #pypy.08:41
GeneralMaximus (~general@122.162.38.115) left irc: Quit: [self dealloc];08:42
fijal (~fijal@196-210-150-33.dynamic.isadsl.co.za) left irc: Ping timeout: 240 seconds08:49
arkx (~aku@eagleflow.fi) left irc: Quit: Lost terminal08:54
arkx (~aku@eagleflow.fi) joined #pypy.08:55
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Ping timeout: 244 seconds08:55
squiddy (~squiddy@g224193065.adsl.alicedsl.de) joined #pypy.08:58
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.09:01
dmitrey (~quassel@77.121.195.200) left irc: Ping timeout: 240 seconds09:04
fijal (~fijal@197.168.65.36) joined #pypy.09:11
fijal (~fijal@197.168.65.36) left irc: Read error: Connection reset by peer09:14
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Ping timeout: 244 seconds09:15
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy.09:16
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.09:23
kvda (~kvda@124-169-134-221.dyn.iinet.net.au) left irc: Quit: x___x09:30
TheNoodle (~noodle@sarcasticsupport.com) left irc: Ping timeout: 245 seconds09:35
JaredW (~JaredW@118.142.17.238) left irc: Remote host closed the connection09:35
JaredW (~JaredW@118.142.17.238) joined #pypy.09:36
mitsuhiko (~mitsuhiko@ubuntu/member/mitsuhiko) left irc: Quit: ZNC - http://znc.sourceforge.net09:36
__pv (pav@xen.ihme.org) left irc: Ping timeout: 245 seconds09:37
exnihilo (~nathan@descartes.its.unc.edu) left irc: Ping timeout: 245 seconds09:37
__pv (pav@xen.ihme.org) joined #pypy.09:38
TheNoodle (~noodle@sarcasticsupport.com) joined #pypy.09:38
mitsuhiko_ (~mitsuhiko@hammett.srv.pocoo.org) joined #pypy.09:38
exnihilo (~nathan@descartes.its.unc.edu) joined #pypy.09:43
espes (~espes@123-243-221-248.tpgi.com.au) left irc: Read error: Connection reset by peer09:46
espes (~espes@123-243-221-248.tpgi.com.au) joined #pypy.09:47
dmitrey (~quassel@77.121.195.200) joined #pypy.09:48
G2P (~G2P@fw-asn1.ornis.com) joined #pypy.09:51
apoirier (~apr@210.168.3.109.rev.sfr.net) joined #pypy.09:54
bivab (~david@ip-62-143-59-5.unitymediagroup.de) joined #pypy.09:57
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Ping timeout: 244 seconds10:02
amaury (amaury_@nat/google/x-udhyahsncecvgqam) joined #pypy.10:05
tgirod (~thomas@4ad54-1-88-172-90-158.fbx.proxad.net) joined #pypy.10:19
umgeher (~umgeher@unaffiliated/umgeher) left irc: Ping timeout: 240 seconds10:19
umgeher (~umgeher@bd21ce7a.virtua.com.br) joined #pypy.10:20
tgirodhello there.10:20
tgirodI have been working for quite some time on a project based on CPython and C++, interfaced with boost-python (https://github.com/tgirod/pyck)10:22
tgirodbut right now I'm facing a problem10:22
tgirodthe C++ part has to run some python code, and this callback is too slow.10:23
Action: mikefc_ waits for the question10:24
tgirodSo I'm wondering if I could gain something by using pypy.10:25
tgirodFrom what I've read on the website, interfacing pypy and kind of prevents some optimizations to take place10:25
tgirodso the net gain might not be that great10:25
tgirodbut I don't know much about that, that's why I'm asking for guidance there.10:26
mikefc_tgirod: I don't know enough to answer your question.  If the act of doing the callback is taking so long (rather than executing the python code itself) you might want to rethink the approach 10:28
mattip (c1a9466d@gateway/web/freenode/ip.193.169.70.109) joined #pypy.10:28
mattipflatiter is missing functioning comparison operators:10:29
mattipa.flat == a10:30
mattipand friends (gt, lt, ge, le ne)10:30
tgirodmikefc_: indeed. unfortunately, this callback designed is non-negociable. it's an audio synthesis app, the pace of the execution is ultimately driven by the hardware.10:31
mattipon the numppy-flatitter I added tests for them all, but of course the tests fail since simply linearizing the base array is not good enough10:32
mattipI'm open to suggestions: add a transformation, add a "target shape" so that the binary operators won't complain about shape mismatch, etc.10:33
setmeaway2 (setmeaway3@118.45.149.58) joined #pypy.10:34
mattip (c1a9466d@gateway/web/freenode/ip.193.169.70.109) left irc: Quit: Page closed10:35
arigatotgirod: in theory pypy can compile assembler for the Python function from the callback, and call that directly10:36
apoirier (~apr@210.168.3.109.rev.sfr.net) left irc: Ping timeout: 245 seconds10:36
stargaming (stargaming@unaffiliated/stargaming) left irc: Ping timeout: 245 seconds10:36
arigatobut that's all depending on the availibility of a direct callback interface10:37
setmeaway (~setmeaway@118.45.149.58) left irc: Ping timeout: 245 seconds10:37
arigatoboost is not supported so far, and generally the other solutions like ctypes don't support fast callbacks in pypy so far10:37
stargaming (stargaming@server3.raumopol.de) joined #pypy.10:40
tgirodarigato: that's interesting. can you elaborate on "direct callback interface" ? is it a way to make low-level calls between C and pypy ?10:41
arigatoyes10:41
arigatoI'm basically saying we don't have a fast solution right now, but it could be done10:42
stefanop (~stefanop@77.72.196.122) joined #pypy.10:42
arigatowe didn't focus on callbacks so far10:42
lizardo (lizardo@nat/indt/x-ekfghlxmnewlvluj) joined #pypy.10:42
fijal (~fijal@196-210-150-33.dynamic.isadsl.co.za) joined #pypy.10:44
tgirodarigato: ok. so it might be a bit early for me to follow this path10:45
fijalof mattip is gone10:48
lizardo (lizardo@nat/indt/x-ekfghlxmnewlvluj) left irc: Remote host closed the connection10:49
mikefc_fijal: if you can access a log for the last half hour, he had a lot to say about: "flatiter is missing functioning comparison operators:" eq, gt, lt, ge etc10:49
mattip (c1a9466d@gateway/web/freenode/ip.193.169.70.109) joined #pypy.10:51
mikefc_speak o the devil10:51
mattipsorry, kind of busy today.10:51
kenaan12antocuni buildbot 1119bba7abc55c 15/bot2/pypybuildbot/master.py: try to run at least the own tests in the py3k branch nightly. The locks should take care of scheduling them aft...10:52
fijalmattip: why it does not work?10:52
fijalmattip: I can give a look at elast10:52
mattipOK. The problem is I defined class W_FlatIter(ViewArray)10:53
mattipbut couldn't make a ViewArray that both is one-dimensional and10:53
mattipiterates properly across the values in the parent array withot copying10:53
fijalyou need a new signature I think10:54
fijaland that's pretty much it10:54
mattipyes.10:54
fijalit should probably inherit from BaseArray IMO10:54
bbot2 (~bbot2@wyvern.cs.uni-duesseldorf.de) left irc: Quit: buildmaster reconfigured: bot disconnecting10:54
bbot2 (~bbot2@wyvern.cs.uni-duesseldorf.de) joined #pypy.10:54
mattipsounds good.10:54
mattipfijal: good. I will do the signature, it still needs a jit_merge_point in setitem and getitem10:56
fijalyeah, don't worry about the merge point now ok?10:57
fijalwe can still merge it without one10:57
mattipgood!10:57
mattipwriting it is no big deal, but adding the test to zjit means messing with compile again.10:57
Action: mattip being rude, sorry.11:00
mattipttfn11:00
mattip (c1a9466d@gateway/web/freenode/ip.193.169.70.109) left irc: Quit: Page closed11:00
antocuniamaury: is there any official way to use a py3k-only syntax inside applevel tests in the py3k branch?11:03
fijalarigato: FYI pypy py.test <some greenlet tests> segfault11:04
fijalthere is a bug on the bug tracker11:04
amaury__ (~amaury_@74.125.57.34) joined #pypy.11:10
amaury (amaury_@nat/google/x-udhyahsncecvgqam) left irc: Ping timeout: 248 seconds11:11
arigato (~arigo@adsl-89-217-207-222.adslplus.ch) left irc: Ping timeout: 252 seconds11:12
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.11:18
tgirod (thomas@4ad54-1-88-172-90-158.fbx.proxad.net) left #pypy.11:18
voidspace (~voidspace@87-194-212-65.bethere.co.uk) joined #pypy.11:20
voidspace (~voidspace@87-194-212-65.bethere.co.uk) left irc: Changing host11:20
voidspace (~voidspace@python/psf/voidspace) joined #pypy.11:20
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Ping timeout: 244 seconds11:24
amaury__antocuni: for py3k-only syntax, you can still use exec() :-)11:24
antocunih11:24
antocunieh11:24
amaury__like in test_compiler11:24
antocuniI'm trying to hack conftest.py so that you can write the code directly in the compiler11:25
verte-wleslie (~verte@python/site-packages/verte) joined #pypy.11:25
antocunibah11:25
antocuniin the docstring11:25
Nick change: verte-wleslie -> verte11:25
Nick change: verte -> wleslie11:25
antocunibut indeed, I'm not sure, maybe it's better to just use exec11:26
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.11:26
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.11:26
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.11:29
dmitreyfijal: does anyone work on concanate? I could try to implement it (it seems to be in c code, so I'll try to write my own implementation). hstack, vstack and dstack depends on it.11:31
fijalno11:31
fijaldmitrey: how about giving a go at rpython level?11:31
dmitreyfijal: what do you mean?11:32
fijalpypy/module/micronumpy/ is RPython11:32
fijalit's not python, it's ours equivalent of C11:32
dmitreyso you propose to test are my funcs work with RPython?11:33
fijalto try implementing it in RPython yes11:34
dmitreydo you mean concatenate() or those ones_like etc?11:35
fijalconcatenate11:35
dmitreyI observe strange behaviour of cpython numpy concatenate:11:36
dmitrey>>> np.concatenate(((1.0, 2.0), (3, 4)), 1) 11:36
dmitreyarray([ 1.,  2.,  3.,  4.])11:36
dmitrey>>> np.concatenate(((1.0, 2.0), (3, 4)), 2) 11:36
dmitreyarray([ 1.,  2.,  3.,  4.])11:36
dmitrey>>> np.concatenate(((1.0, 2.0), (3, 4)), 10) 11:36
dmitreyarray([ 1.,  2.,  3.,  4.])11:36
dmitreyisn't it a bug?11:36
dmitreyfijal: what should yeild our implementation of concatenate with np.concatenate(((1.0, 2.0), (3, 4)), 1)?11:39
dmitreyand with higher values of axis parameter?11:40
lizardo (lizardo@nat/indt/x-azskxoozzffmtjhs) joined #pypy.11:50
voidspac_ (~voidspace@87-194-212-65.bethere.co.uk) joined #pypy.11:52
voidspace (~voidspace@python/psf/voidspace) left irc: Quit: Bye...11:59
Nick change: voidspac_ -> voidspace11:59
mikefc_dmitrey: i would have thought that that would have thrown an 'axis out of bounds' error. but for 1d arrays there is only one way to concat them, so I guess axis is ignored.12:03
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Ping timeout: 252 seconds12:03
dmitreyso what numpypy implementation should have: throw a bug or yeild output same to numpy version?12:04
mikefc_it's not a bug. it's a feature.12:05
mikefc_1d arrays only concat in one way, so axis is ignored.12:05
mikefc_so that's what numpypy should do.12:05
azanella (~azanella@189.6.80.131) joined #pypy.12:06
mikefc_I would check the actual behaviour on numpy2.x though.12:06
mikefc_i'm still on 1.6.1 here.12:06
Nick change: nedbat_ -> nedbat12:08
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.12:12
whyking (~quassel@ip68-14-13-208.ri.ri.cox.net) joined #pypy.12:18
CIA-1703Dmitrey 07roundup * 10#1021/numpypy: incorrect dtype comparison (vs original numpy): (log message trimmed)12:19
CIA-17[new] >>> a=np.array([1,2,3],'int64')12:19
CIA-17>>> a.dtype12:19
CIA-17dtype('int64')12:19
CIA-17>>> a.dtype=='int64' * 14https://bugs.pypy.org/issue102112:19
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) left irc: Remote host closed the connection12:22
AlekSi (~AlekSi@85.235.191.82) joined #pypy.12:23
fijaldmitrey: I think it wraps around by default?12:31
fijaldmitrey: I don't know12:31
fijaldmitrey: I would say raising an error is fine12:31
dmitreyok, I will raise a error12:31
mikefc_fijal: it fails properly if the arrays have more than 1d and you use an axis>ndim12:31
mikefc_it just doesn't fail if it's 1d.  it just concats them in the only way possible.12:32
Alex_Gaynorfijal: Justin or someone started a concat impl IIRC.12:32
fijalyeah, not sure what's the status of it though12:32
Alex_Gaynorme neither12:33
mikefc_ok. i'm out. good night and good luck.12:36
mikefc_ (~mike@60-242-240-196.static.tpgi.com.au) left irc: Quit: mikefc_12:36
anish_ (anish@nat/redhat/x-pjvidzznzhafrpvu) left irc: Quit: Leaving12:36
pjdelport (~pjdelport@c-69-251-199-136.hsd1.md.comcast.net) left irc: Remote host closed the connection12:37
nedbatAlex_Gaynor: on the train?12:40
Alex_Gaynornedbat: yup12:40
nedbatAlex_Gaynor: sweet!  I'm making a list of questions.12:40
Alex_Gaynornedbat: Cool, google moderator or just a list?12:41
nedbatAlex_Gaynor: just a list.  I guess I could load up a moderator and let people ruminate on it today.12:43
Alex_Gaynorno preference on my part, just curious12:44
nedbatno, it's a good idea to get people's participation.12:44
umgeher (~umgeher@bd21ce7a.virtua.com.br) left irc: Ping timeout: 240 seconds12:48
fijalwhat are you guys up to?12:49
umgeher (~umgeher@bd21ce7a.virtua.com.br) joined #pypy.12:50
mitsuhiko_ (~mitsuhiko@hammett.srv.pocoo.org) left irc: Changing host12:50
mitsuhiko_ (~mitsuhiko@ubuntu/member/mitsuhiko) joined #pypy.12:50
Nick change: mitsuhiko_ -> mitsuhiko12:50
Alex_Gaynorfijal: http://meetup.bostonpython.com/events/48542762/12:50
nedbatbeat me to it12:51
fijalcool12:51
nedbatAlex_Gaynor: http://www.google.com/moderator/#16/e=1d669f12:53
umgeher (~umgeher@bd21ce7a.virtua.com.br) left irc: Ping timeout: 240 seconds12:54
apoirier (~apr@210.168.3.109.rev.sfr.net) joined #pypy.12:55
pjdelport (~pjdelport@c-69-251-199-136.hsd1.md.comcast.net) joined #pypy.12:58
pnguyen (~pnguyen@cpe-70-112-214-119.austin.res.rr.com) joined #pypy.13:00
CIA-1703masklinn 07roundup * 10#1022/dict.update significantly slower than series of dict.__setitem__: 13:10
CIA-17[new] In CPython, as the number of keys grow dict.update settles at roughly half the13:10
CIA-17speed of an13:10
CIA-17equivalent bunch of dict[key]=value sequence: ... * 14https://bugs.pypy.org/issue102213:10
mat^2 (~mathias@212.130.113.35) joined #pypy.13:10
nedbatI can see the voting in the Moderator is now heavily PyPy-skewed.. :)13:10
nedbatj/k, not badly.13:10
fijalI think we have a reasonably good PR for python world13:11
Alex_Gaynornedbat: all the django people are west coast hippies who sleep late :)13:11
fijalwe completely suck compared to say rubinius though13:11
nedbatAlex_Gaynor: also, this is the only place I've posted the URL!13:11
AlekSi (~AlekSi@85.235.191.82) left irc: Ping timeout: 240 seconds13:11
CIA-1703Dmitrey 07roundup * 10#1023/numpypy: empty() cannot handle dtype='object': (log message trimmed)13:11
CIA-17[new] >>> import numpy as np13:11
CIA-17>>> np.empty((2,3),object)13:11
CIA-17array([[None, None, None],13:11
CIA-17[None, None, None]], dtype=object) * 14https://bugs.pypy.org/issue102313:11
fijaldmitrey: we don't support object dtype13:12
fijalcan you close this issue?13:12
dmitreyok13:12
fijalI have your paste at hand I'm trying to figure out what empty_like actually does13:13
CIA-1703Dmitrey 07roundup * 10#1023/numpypy: empty() cannot handle dtype='object': [wontfix] dtype 'object' seems won't be supported according to dev team plans * 14https://bugs.pypy.org/issue102313:13
fijalAlex_Gaynor: do you know what github runs on?13:14
lucianfijal: ruby/rails i think13:14
lucianand according to wikipedia, also some erlang13:15
nedbatAlex_Gaynor: thanks for the tweet.  Now I see where the votes are coming from.13:20
pnguyen (~pnguyen@cpe-70-112-214-119.austin.res.rr.com) left irc: Quit: pnguyen13:22
AlekSi (~AlekSi@85.235.191.82) joined #pypy.13:23
AlekSi (~AlekSi@85.235.191.82) left irc: Read error: Connection reset by peer13:27
AlekSi_ (~AlekSi@85.235.191.82) joined #pypy.13:27
AlekSi_ (~AlekSi@85.235.191.82) left irc: Read error: Connection reset by peer13:28
AlekSi (~AlekSi@85.235.191.82) joined #pypy.13:30
AlekSi (~AlekSi@85.235.191.82) left irc: Remote host closed the connection13:32
arigato (~arigo@adsl-89-217-207-222.adslplus.ch) joined #pypy.13:35
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) left irc: Ping timeout: 245 seconds13:42
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) joined #pypy.13:44
CIA-1703csenger 07roundup * 10#1015/Add build to run the benchmarks with python 2.7 tip: 13:45
CIA-17I went on and removed the guessing about what is the executable name, revision13:45
CIA-17and project from the interpreter path and turned that into explici ... * 14https://bugs.pypy.org/issue101513:45
csenger (~user@dslb-084-058-185-001.pools.arcor-ip.net) joined #pypy.13:47
dmitrey (~quassel@77.121.195.200) left irc: Remote host closed the connection13:53
rinu (~quassel@117.202.119.205) joined #pypy.13:54
fijalAlex_Gaynor: int64 is a subclass of int right?13:58
CIA-1703arigo 07roundup * 10#1020/segfault when running test_greenlet.py through pypy pytest.py: 14:02
CIA-17Sorry, I tried to debug it but I'm blocked by not remembering exactly how14:02
CIA-17asmgcroot interacts with continuations. I may spend more time, but I m ... * 14https://bugs.pypy.org/issue102014:02
fijalarigato: ping?14:03
arigatopong14:03
wleslie (~verte@python/site-packages/verte) left irc: Quit: ~~~ Crash in JIT!14:04
fijalif I create an int subclass space.int_w(a_subclass_of_int_instance) should work right?14:04
arigatoyes14:05
rinu (~quassel@117.202.119.205) left irc: Read error: Connection reset by peer14:06
fijaldoes not work :(14:06
fijalalthough the subclass is created on interp-level14:06
fijalarigato: you can try if you want numpypy.int64 is a subclass of int14:07
fijalor it's not supposed to work that way?14:08
dmalcolm_ (~david@c-24-61-12-82.hsd1.ma.comcast.net) left irc: Quit: Leaving14:08
arigatoit is14:09
fijalthen it does not :/14:09
fijalalthough admitedly I don't see how it would work14:10
fijalsince it has int in __mro__ but does not seem to use the attribute at all14:10
fijalI'm not completely sure what to do now14:11
fijalbut obviously list[array[item]] would not work14:11
arigatoah, it's a "subclass" of int in the sense of having int in its mro --- then no chance indeed14:11
rinu (~quassel@117.217.133.161) joined #pypy.14:12
fijal:/14:12
fijalhow do I make it work14:12
fijal?14:12
fijalarigato: it works in cpython :/14:15
fijalarigato: should I provide an int_w method simply?14:15
rinu (~quassel@117.217.133.161) left irc: Remote host closed the connection14:16
arigatoin theory, by writing custom interp-level classes with the same set of multimethods as regular ints14:16
fijalseems so14:16
fijalhaha!14:16
fijalI did my homework and int_w is no longer a multimethod14:16
fijal<win>14:16
arigatoI'm still confused by the numpy types like int6414:17
arigatofirst, how can int64 be a subclass of int on 32-bit platforms?14:17
fijalI think it's only on 64bit platforms14:17
fijalyes14:17
fijalseems mro does depend on the platform :/14:18
arigatook, then what is the difference between int and int64 on a 64bit platform?14:18
fijalint64 is also a subclass of numpy.integer14:18
fijalnumpy.signedinteger etc.14:18
fijalit also has methods like an array14:18
fijallike int64(3).real == 314:19
arigatobut all regular "__xyz__" methods of ints also apply to int64, or are there some overridden?14:19
fijala bit no clue14:19
arigatocan't you just create the class int64 at app-level?14:20
arigatoe.g. what does int64(3) + int64(4) return?14:20
exnihilo python translate.py --opt=1 targetpypystandalone.py --allworkingmodules should produce a pypy-c with translation.continuation: True right?  Is there a translation log somewhere I can check to see if there was a problem?14:21
fijalexnihilo: why are you using -O1 btw?14:22
fijalarigato: I can in theory create in64 in applevel14:22
exnihilobecause my linux box has 2 gb of memory and is slow... I just want to start testing my code14:22
fijalbut I would need to create all others in interp-level14:22
fijalexnihilo: what have you written?14:23
fijalexnihilo: what have you to test so pypy-c -O1 is useful14:23
inhahe (~inhahe@c-50-128-212-190.hsd1.fl.comcast.net) left irc: Ping timeout: 248 seconds14:23
exnihiloI want to test converting some of the interpreter loops to continulets so I can capture frames14:23
arigatoexnihilo: run "import _continuation" in the pypy-c14:23
fijalexnihilo: can't you just test it on uncompiled pypy?14:24
exnihilofijal: I would love to, but I get the "must be translated with --continuation" issue14:24
arigatoor on the latest nightly build of pypy, for that matter14:24
exnihiloI'm very excited to start hacking w/ pypy as soon as I can get it built with all the options I need :D14:25
fijalexnihilo: I fixed that recently, didn't I?14:27
fijalpy.py --withmod-continuation should work now14:27
exnihilohrm, I'll try pulling again, the copy I'm trying to translate is only ~2 days old14:27
fijalexnihilo: I fixed it yesterday14:28
exnihilok14:28
fijalarigato: meh, multimethods are such a mess :/14:28
fijalI'll kill some more14:28
Rhy0lite (dje@nat/ibm/x-hlxrgwqsszrxytuu) joined #pypy.14:35
kenaan12l.diekmann type-specialized-instances 117636a491c009 15/: merge default14:36
kenaan12l.diekmann type-specialized-instances 11b15e618e3d82 15/pypy/: (cfbolz, l.diekmann): merge int-tag-untag-as-operationus to get the necessary optimizations...14:36
exnihilofijal: working now, thanks!14:36
gtaylor (~gtaylor@108-196-160-156.lightspeed.tukrga.sbcglobal.net) joined #pypy.14:39
whitelynx (~whitelynx@63.241.75.144) joined #pypy.14:44
kenaan12fijal numpy-back-to-applevel 1196b51eda2096 15/pypy/module/micronumpy/: progress on take14:46
kenaan12fijal kill-more-multimethods 11ef9ea1a64a8e 15/pypy/objspace/std/tupleobject.py: a branch to perform some experiments on killing some multimethods. a performance experiment (Complet...14:50
inhahe (~inhahe@c-50-128-212-190.hsd1.fl.comcast.net) joined #pypy.14:50
AlekSi (~AlekSi@176.57.74.133) joined #pypy.14:51
ciphergoth_Are test failures logged anywhere?14:57
fijalciphergoth_: yop14:57
fijalbuildbot.pypy.org/summary14:57
lizardo (lizardo@nat/indt/x-azskxoozzffmtjhs) left #pypy ("Leaving").14:57
ciphergoth_I'm in pypy/translator/sandbox and I've run ../../../pytest.py - I get lots of output and at the end it says my tests have failed14:58
ciphergoth_fijal: sorry I mean locally14:58
fijalno :/14:58
ciphergoth_but I'm not sure what to do to find out more14:58
fijalyou have to run with --tb=short or so14:58
fijalI usually run -x14:58
ciphergoth_OK what's the recommended way to run the tests?14:58
fijalthat shows just one failure14:58
fijal-s as well if you want to disable capturing output14:58
rguillebertor --tb=none you just see the failures14:58
rguillebertbut no traceback14:59
ciphergoth_OK so -x will help me work on one failure at a time14:59
Nick change: ciphergoth_ -> ciphergoth14:59
ciphergothbut what should I do to capture as much info as possible to a file so I can investigate what went wrong?14:59
fijalI have no idea15:00
rguilleberta redirection15:00
fijalfirst you complain there is too much output15:00
fijaland you can't find relevant part15:00
fijaland now that you want it all, but readable?15:00
rguillebert./py.test -s > myfile15:00
fijal--pdb would put you in a debugger15:00
ciphergothI'm coming from the Java world, where I'd find the verbose info about each failed test in target/surefire/file-per-test15:01
ciphergothI'm complaining about too much on the console and too little in files.  Sounds like just redirecting to a file is the way to go.15:01
fijalyeah15:01
rguillebertor use tee if you want to have it on stdin as well15:01
ciphergothrguillebert: sounds wise15:01
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) joined #pypy.15:02
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) joined #pypy.15:06
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.15:07
antocuniciphergoth: I suggest to use --pdb, so you can inspect everything you need about what went wrong15:07
antocuni(the value of local variables, etc.)15:07
exarkunHow does py.magic.greenlets relate to PyPy?  At all?  Does it work on PyPy?  Is there a different library for that kind of thing that works on CPython and and PyPy?  Is py.magic.greenlets even what people use on CPython anymore?15:08
fijalexarkun: I think the official one is "greenlet" these days15:09
fijalit comes in "greenlet" package on pypi and should work on pypy as well15:09
exarkunThanks15:10
exarkunDarn.  It's packaged in Debian, but it's also a .so.15:11
exarkunsiiigh and it requires setuptools15:12
exarkunOuch.  And it doesn't want to build against my PyPy, although maybe my PyPy is sort of old.15:13
fijalah no15:13
fijalyou don't build it against pypy15:13
fijalpypy comes with a builtin one15:13
exarkunAh15:13
exarkunGreat15:13
fijalexarkun: if we pressure armin enough, he'll make jit neatly work with greenlets15:14
fijalright now every loop that contains switching code will be kinda slow15:14
ronnyexarkun: what are you trying to do with greenlet?15:17
Tobu (~Tobu@unaffiliated/tobu) left irc: Ping timeout: 272 seconds15:18
exarkunPedagogy15:18
fijalexarkun: uh then it does not work, forget everything I said15:19
exarkunfijal: It doesn't work for pedagogy?  Or it doesn't work at all on PyPy?15:20
fijalI don't know :)15:22
fijalwas trying to trick you not to do pedagogy15:22
fprimex (~fprimex@brent-macbook.sc.fsu.edu) joined #pypy.15:22
exarkunheh15:22
whyking (~quassel@ip68-14-13-208.ri.ri.cox.net) left irc: Read error: Operation timed out15:24
Taggnostr2_ (~quassel@dyn57-362.yok.fi) joined #pypy.15:28
Taggnostr2_ (~quassel@dyn57-362.yok.fi) left irc: Remote host closed the connection15:28
dmitrey (~quassel@77.121.195.200) joined #pypy.15:28
dmalcolm (david@nat/redhat/session) joined #pypy.15:30
ericholscher_ (~eric@173.203.223.63) joined #pypy.15:32
fijal (~fijal@196-210-150-33.dynamic.isadsl.co.za) left irc: Ping timeout: 240 seconds15:33
mikefc (9665bcc5@gateway/web/freenode/ip.150.101.188.197) got netsplit.15:34
bbot2 (~bbot2@wyvern.cs.uni-duesseldorf.de) got netsplit.15:34
setmeaway2 (setmeaway3@118.45.149.58) got netsplit.15:34
jell (~jell@host-134-23.dmtec.eu) got netsplit.15:34
ciphergoth (~paul@host238.lshift.net) got netsplit.15:34
gutworth (~benjamin@dhcp-128-36-73-23.central.yale.edu) got netsplit.15:34
jml (~jml@canonical/launchpad/jml) got netsplit.15:34
dimazest (~dimazest@balticom-202-59.balticom.lv) got netsplit.15:34
ericholscher (~eric@173.203.223.63) got netsplit.15:34
htoothrot (~mux@unaffiliated/htoothrot) got netsplit.15:34
CIA-17 (~CIA@cia.atheme.org) got netsplit.15:34
MostAwesomeDude (JuTorzalS0@osuosl/staff/mostawesomedude) got netsplit.15:34
timotimo (~timo@gallifrey.wybt.net) got netsplit.15:34
azanella (~azanella@189.6.80.131) got netsplit.15:34
panni_ (smuxi@descent.fragstore.net) got netsplit.15:34
ojii (~ojii@50.57.189.190) got netsplit.15:34
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) got netsplit.15:34
JaredW (~JaredW@118.142.17.238) got netsplit.15:34
guibou (~guibou@kraken.insecable.net) got netsplit.15:34
mvt (~mvt@53530442.cm-6-4a.dynamic.ziggo.nl) got netsplit.15:34
EnCuKou (~encukou@ip-94-113-220-25.net.upcbroadband.cz) got netsplit.15:34
legogris (~legogris@c80-217-207-157.bredband.comhem.se) got netsplit.15:34
nopf (~froschif@147.172.224.4) got netsplit.15:34
Ormod_ (~valtha@ohmu.fi) got netsplit.15:34
alexsuraci (~alexsurac@li175-235.members.linode.com) got netsplit.15:34
idnar (~quassel@unaffiliated/idnar) got netsplit.15:34
exarkun (~exarkun@wordeology.com) got netsplit.15:34
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) got netsplit.15:34
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) got netsplit.15:34
inhahe (~inhahe@c-50-128-212-190.hsd1.fl.comcast.net) got netsplit.15:34
voidspace (~voidspace@87-194-212-65.bethere.co.uk) got netsplit.15:34
zain (~textual@bzq-79-182-196-50.red.bezeqint.net) got netsplit.15:34
stakkars (~tismer@p5DDB5C07.dip.t-dialin.net) got netsplit.15:34
Taggnostr (~quassel@dyn57-362.yok.fi) got netsplit.15:34
Dulak (~michael@unaffiliated/dulak) got netsplit.15:34
JStoker (jstoker@unaffiliated/jstoker) got netsplit.15:34
zk (~is@whatit.is) got netsplit.15:34
getxsick (xsx@unaffiliated/getxsick) got netsplit.15:34
nigelb (~nigel@ubuntu/member/nigelb) got netsplit.15:34
espes (~espes@123-243-221-248.tpgi.com.au) got netsplit.15:34
__pv (pav@xen.ihme.org) got netsplit.15:34
csenger (~user@dslb-084-058-185-001.pools.arcor-ip.net) got netsplit.15:34
Varriount (~Varriount@hnng.public-craft.com) got netsplit.15:34
nedbat (~nedbat@python/psf/nedbat) got netsplit.15:34
Telugodu (~rambaho@202.3.77.231) got netsplit.15:34
tos9 (~tos9@unaffiliated/tos9) got netsplit.15:34
__class__ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) got netsplit.15:34
Husio (husiatyn@oceanic.wsisiz.edu.pl) got netsplit.15:34
jperras (~jperras@li245-112.members.linode.com) got netsplit.15:34
carljm (~carljm@spinach.threepines.org) got netsplit.15:34
myf (~myf@li245-175.members.linode.com) got netsplit.15:34
inhahe (~inhahe@c-50-128-212-190.hsd1.fl.comcast.net) returned to #pypy.15:37
-mquin (~mquin@freenode/staff/mquin) to $$*- [Global Notice] We are experiencing some connectivity problems, please bear with us while we investigate (and run around with duct tape)15:39
getxsick (xsx@tramwaj.asi.pwr.wroc.pl) joined #pypy.15:39
timotimo (~timo@gallifrey.wybt.net) returned to #pypy.15:39
dimazest (~dimazest@balticom-202-59.balticom.lv) returned to #pypy.15:39
CIA-6 (~CIA@cia.atheme.org) joined #pypy.15:41
setmeaway2 (setmeaway3@118.45.149.58) returned to #pypy.15:42
Tobu (~Tobu@2a01:e35:8a2c:b230:213:d3ff:fe85:5614) joined #pypy.15:43
JStoker (jstoker@atlas.jcs.me.uk) joined #pypy.15:44
stakkars (~tismer@p5DDB5C07.dip.t-dialin.net) returned to #pypy.15:44
jell (~jell@host-134-23.dmtec.eu) returned to #pypy.15:44
ericholscher (~eric@173.203.223.63) got lost in the net-split.15:45
panni_ (smuxi@descent.fragstore.net) got lost in the net-split.15:45
ojii (~ojii@50.57.189.190) got lost in the net-split.15:45
__class__ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) got lost in the net-split.15:45
Ormod_ (~valtha@ohmu.fi) got lost in the net-split.15:45
htoothrot (~mux@unaffiliated/htoothrot) got lost in the net-split.15:45
zk (~is@whatit.is) got lost in the net-split.15:45
alexsuraci (~alexsurac@li175-235.members.linode.com) got lost in the net-split.15:45
guibou (~guibou@kraken.insecable.net) got lost in the net-split.15:45
jperras (~jperras@li245-112.members.linode.com) got lost in the net-split.15:45
nigelb (~nigel@ubuntu/member/nigelb) got lost in the net-split.15:45
carljm (~carljm@spinach.threepines.org) got lost in the net-split.15:45
myf (~myf@li245-175.members.linode.com) got lost in the net-split.15:45
exarkun (~exarkun@wordeology.com) got lost in the net-split.15:45
idnar (~quassel@unaffiliated/idnar) got lost in the net-split.15:45
Husio (husiatyn@oceanic.wsisiz.edu.pl) got lost in the net-split.15:45
MostAwesomeDude (JuTorzalS0@osuosl/staff/mostawesomedude) got lost in the net-split.15:45
CIA-17 (~CIA@cia.atheme.org) got lost in the net-split.15:45
tos9 (~tos9@unaffiliated/tos9) got lost in the net-split.15:45
nopf (~froschif@147.172.224.4) got lost in the net-split.15:45
jml (~jml@canonical/launchpad/jml) got lost in the net-split.15:45
legogris (~legogris@c80-217-207-157.bredband.comhem.se) got lost in the net-split.15:45
gutworth (~benjamin@dhcp-128-36-73-23.central.yale.edu) got lost in the net-split.15:45
mikefc (9665bcc5@gateway/web/freenode/ip.150.101.188.197) got lost in the net-split.15:45
Dulak (~michael@unaffiliated/dulak) got lost in the net-split.15:45
EnCuKou (~encukou@ip-94-113-220-25.net.upcbroadband.cz) got lost in the net-split.15:45
Taggnostr (~quassel@dyn57-362.yok.fi) got lost in the net-split.15:45
mvt (~mvt@53530442.cm-6-4a.dynamic.ziggo.nl) got lost in the net-split.15:45
ciphergoth (~paul@host238.lshift.net) got lost in the net-split.15:45
Telugodu (~rambaho@202.3.77.231) got lost in the net-split.15:45
nedbat (~nedbat@python/psf/nedbat) got lost in the net-split.15:45
zain (~textual@bzq-79-182-196-50.red.bezeqint.net) got lost in the net-split.15:45
Varriount (~Varriount@hnng.public-craft.com) got lost in the net-split.15:45
JaredW (~JaredW@118.142.17.238) got lost in the net-split.15:45
__pv (pav@xen.ihme.org) got lost in the net-split.15:45
espes (~espes@123-243-221-248.tpgi.com.au) got lost in the net-split.15:45
bbot2 (~bbot2@wyvern.cs.uni-duesseldorf.de) got lost in the net-split.15:45
voidspace (~voidspace@87-194-212-65.bethere.co.uk) got lost in the net-split.15:45
azanella (~azanella@189.6.80.131) got lost in the net-split.15:45
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) got lost in the net-split.15:45
csenger (~user@dslb-084-058-185-001.pools.arcor-ip.net) got lost in the net-split.15:45
jaffachief (~textual@host86-159-232-5.range86-159.btcentralplus.com) got lost in the net-split.15:45
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) got lost in the net-split.15:45
fijal (~fijal@196-210-150-33.dynamic.isadsl.co.za) joined #pypy.15:45
Dulak (~michael@ip98-167-237-72.ph.ph.cox.net) joined #pypy.15:46
jml (~jml@mumak.net) joined #pypy.15:50
Nick change: jml -> jml`15:50
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) joined #pypy.15:50
Nick change: jml` -> jml15:50
nigelb_ (~nigel@nigelb.me) joined #pypy.15:53
azanella (~azanella@189.6.80.131) joined #pypy.15:53
panni_ (smuxi@descent.fragstore.net) joined #pypy.15:53
ojii (~ojii@50.57.189.190) joined #pypy.15:53
bivab (~david@ip-62-143-59-5.unitymediagroup.de) left irc: Read error: Connection reset by peer15:53
voidspace (~voidspace@87-194-212-65.bethere.co.uk) joined #pypy.15:53
voidspace (~voidspace@87-194-212-65.bethere.co.uk) left irc: Client Quit15:53
panni_ (smuxi@descent.fragstore.net) left irc: Excess Flood15:53
smuxi (smuxi@descent.fragstore.net) joined #pypy.15:53
gutworth (~benjamin@dhcp-128-36-73-23.central.yale.edu) joined #pypy.15:54
zk (~is@whatit.is) joined #pypy.15:54
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) joined #pypy.15:54
JaredW (~JaredW@118.142.17.238) joined #pypy.15:54
guibou (~guibou@kraken.insecable.net) joined #pypy.15:54
mvt (~mvt@53530442.cm-6-4a.dynamic.ziggo.nl) joined #pypy.15:54
EnCuKou (~encukou@ip-94-113-220-25.net.upcbroadband.cz) joined #pypy.15:54
legogris (~legogris@c80-217-207-157.bredband.comhem.se) joined #pypy.15:54
nopf (~froschif@147.172.224.4) joined #pypy.15:54
Ormod_ (~valtha@ohmu.fi) joined #pypy.15:54
alexsuraci (~alexsurac@li175-235.members.linode.com) joined #pypy.15:54
idnar (~quassel@unaffiliated/idnar) joined #pypy.15:54
exarkun (~exarkun@wordeology.com) joined #pypy.15:54
#pypy: mode change '+o exarkun ' by adams.freenode.net15:54
pjdelport (~pjdelport@c-69-251-199-136.hsd1.md.comcast.net) left irc: Excess Flood15:55
zain (~textual@bzq-79-182-196-50.red.bezeqint.net) joined #pypy.15:55
pjdelport (~pjdelport@c-69-251-199-136.hsd1.md.comcast.net) joined #pypy.15:55
Nick change: smuxi -> panni_15:55
voidspace (~voidspace@87-194-212-65.bethere.co.uk) joined #pypy.15:55
Taggnostr2_ (~quassel@dyn57-362.yok.fi) joined #pypy.15:55
csenger (~user@dslb-084-058-185-001.pools.arcor-ip.net) joined #pypy.15:55
espes (~espes@123-243-221-248.tpgi.com.au) joined #pypy.15:55
__pv (pav@xen.ihme.org) joined #pypy.15:55
Varriount (~Varriount@hnng.public-craft.com) joined #pypy.15:55
nedbat (~nedbat@python/psf/nedbat) joined #pypy.15:55
Telugodu (~rambaho@202.3.77.231) joined #pypy.15:55
tos9 (~tos9@unaffiliated/tos9) joined #pypy.15:55
__class__ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) joined #pypy.15:55
Husio (husiatyn@oceanic.wsisiz.edu.pl) joined #pypy.15:55
jperras (~jperras@li245-112.members.linode.com) joined #pypy.15:55
carljm (~carljm@spinach.threepines.org) joined #pypy.15:55
myf (~myf@li245-175.members.linode.com) joined #pypy.15:55
ciphergoth (~paul@host238.lshift.net) joined #pypy.16:01
AlekSi (~AlekSi@176.57.74.133) left irc: Remote host closed the connection16:03
16SAALZS3 (~tismer@i59F714CF.versanet.de) joined #pypy.16:06
mikefc (9665bcc5@gateway/web/freenode/ip.150.101.188.197) joined #pypy.16:06
MostAwesomeDude (JuTorzalS0@osuosl/staff/mostawesomedude) joined #pypy.16:06
bivab (~david@ip-62-143-59-5.unitymediagroup.de) joined #pypy.16:06
dmalcolm (david@nat/redhat/session) left irc: Changing host16:06
dmalcolm (david@nat/redhat/x-ozeixvtlexqcdgwx) joined #pypy.16:06
JStoker (jstoker@atlas.jcs.me.uk) left irc: Changing host16:06
JStoker (jstoker@unaffiliated/jstoker) joined #pypy.16:06
bivab (~david@ip-62-143-59-5.unitymediagroup.de) left irc: Quit: bivab16:06
Nick change: jell -> Guest185916:07
Nick change: zk -> Guest3234216:07
htoothrot (~mux@unaffiliated/htoothrot) joined #pypy.16:08
CIA-603Dmitrey 07roundup * 10#1024/numpypy: bug with type/dtype: 16:10
CIA-6[new] This bug prevents porting of numpy funcs concatenate, hstack, vstack, dstack and16:10
CIA-6may affect lots of other code (unrevealed yet).16:10
CIA-6code:16:10
CIA-6a ... * 14https://bugs.pypy.org/issue102416:10
fijaldmitrey: it's more complex than that16:11
fijaldmitrey: isisntance(numpy.array([1])[0], numpy.int64) will fail for example16:12
kenaan12antocuni py3k 11b83de607e7cc 15/pypy/interpreter/astcompiler/test/test_symtable.py: fix the test, which was broken by the implementation of nonlocal in d6d9ea7ba30616:12
kenaan12antocuni py3k 114df3b47d0b13 15/pypy/interpreter/astcompiler/test/test_symtable.py: add more tests about the nonlocal declaration introduced by d6d9ea7ba30616:12
kenaan12antocuni py3k 11f1dde39ee4a3 15/pypy/interpreter/astcompiler/: make 'from x import *' always an error if it's not at the module level instead of a warning. Adapt or kill the old ...16:12
kenaan12antocuni py3k 11def902951d4f 15/pypy/: add the possibility to write applevel tests inside the docstring (only if the body of the method is empty. This way...16:12
kenaan12antocuni py3k 11b34f81eba785 15/pypy/tool/pytest/test/test_conftest1.py: test docstrings separately from the other applevel tests16:12
dmitreyI haven't noticed it in active bugs in bugtracker16:12
kenaan12antocuni py3k 11443723ebd5e0 15/pypy/tool/pytest/test/test_conftest1.py: fix these tests which were broken by the previous commit (because now the docstring tests are unselected)16:12
kenaan12antocuni py3k 11d3e2c1dbfd08 15/pypy/: support docstring-tests also in functions16:12
kenaan12antocuni py3k 111b6b08120524 15/pypy/: skip docstring-tests when running with -A16:12
kenaan12antocuni py3k 119868585f9314 15/pypy/: correctly run docstring-tests when using --appdirect=...16:12
fijaldmitrey: hm, wait a sec16:12
antocuniamaury__: you can now write applevel tests which use py3k-only syntax inside docstrings :-)16:13
Action: antocuni fails to pop what he was doing from his todo stack16:13
fijaldmitrey: I don't remember but it's annoying :/16:14
dmitreyso why it's not fixed yet? bug hasn't been locayed yet?16:14
dmitreylocated16:14
exarkunWhy would you care about the "syntax" of text in a docstring?16:14
fijalexarkun: because python 3 is insane :/16:15
kenaan12fijal kill-more-multimethods 115ab71bed7f64 15/pypy/objspace/std/objspace.py: remove performance hack16:15
exarkunokay nevermind16:15
dmitreyfijal:I think I can create a walkaround for concatenate, but it will be a little bit slower (operate with string names of types)16:15
antocuniexarkun: there are tests which are written as plain python function, but internally we fish their source code and feed it to the pypy python interpreter16:16
antocunibut sometimes we need to use a py3k-only syntax, so we need to write the source code for those tests in a docstring16:17
Action: exarkun wishes he hadn't asked16:18
antocuni:-)16:18
antocuniaargh16:18
amaury__antocuni: appdoctests?16:19
antocuniamaury__: they are not really doctests16:19
antocunijust a nicer way to execute your string with py.py16:19
stefanop (~stefanop@77.72.196.122) left irc: Remote host closed the connection16:19
fijalhm16:20
fijalis space.gettypefor supposed to return a singleton?16:20
dmitrey (~quassel@77.121.195.200) left irc: Read error: Operation timed out16:24
kenaan12fijal default 113890a6e149fa 15/pypy/module/micronumpy/: a test and a fix for boxes identity16:28
amaury__antocuni: in py3k branch, the -A option should select a python316:29
antocuniamaury__: but only if you pass --appdirect, right?16:29
CIA-603fijal 07roundup * 10#1024/numpypy: bug with type/dtype: [resolved] Fixed in 3890a6e149fa * 14https://bugs.pypy.org/issue102416:29
amaury__-A <=> --appdirect16:30
fijalisn't -A and --appdirect the same thing?16:30
antocuniok, then what is this runappdirect vs appdirect in the conftest?16:31
antocuniyes, indeed16:31
antocuni-A is the same as --runappdirect16:31
antocuniand uses the hosting python interpreter16:31
antocuni--appdirect lets you to specifty the desired interpreter to run16:32
antocuniI suppose the sanest thing to do in the py3k branch is to kill --runappdirect, and select a python3 executable by default when using --appdirect16:32
amaury__yes16:33
amaury__sorry, I forgot that I added this option16:34
antocuninp16:34
voidspace (~voidspace@87-194-212-65.bethere.co.uk) left irc: Quit: Goodbye...16:34
AlekSi (~AlekSi@176.57.74.133) joined #pypy.16:37
dmalcolm (david@nat/redhat/x-ozeixvtlexqcdgwx) left irc: Read error: Connection reset by peer16:40
dmalcolm (david@nat/redhat/x-qppgkiamgbezjuau) joined #pypy.16:40
Action: antocuni is having "fun" with str vs unicode issues16:43
antocunihow could all of this have ever worked?16:43
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection16:44
lahwranthere is currently no work in progress to move the translation toolchain from python 2 to python 3, correct?16:50
lahwranor is that being worked on as part of the python 3 interpreter initiative16:50
antocunilahwran: the former16:50
antocuniit's even unclear if we'll ever move it to python 316:50
antocunibecause it'd be a lot of work16:51
Action: fijal votes "no"16:51
Action: antocuni too16:51
fijalantocuni: note that we don't generally mind "lots of work"16:51
fijalit's "a lot of work that brings absolutely no benefit"16:51
antocuniwell, it depends on the benefits16:51
antocuniyes, exactly16:51
antocuniactually, the only one benefit would be that once we port the translator toolchain to python3, we can discontinue the python2 interpreter16:52
antocunibut we are not interested in doing that, of course16:52
lahwranthat's quite a ways in the future though16:52
fijalwell16:52
fijalI would not discontinue python 2 interpreter until benchmarks are ported :)16:52
fijallahwran: how do you know?16:52
fijallahwran: future is hard to predict16:52
lahwranthat wouldn't be feasible until python 2 reaches EOL16:53
exarkunas if "EOL" were some well-defined point in time :)16:53
lahwranis it not?16:53
fijallahwran: no16:53
lahwranah16:53
exarkunPython is not Oracle16:53
exarkunA King sitting on top of a mountain of gold cannot announce one day "You may not longer use Python 2!"16:54
antocuniwell, it can16:54
fijalexarkun: have you seen the secret picture of apple HQ by chance?16:54
antocuniit's just that people would not respect the order :-)16:54
lahwranEOL would pretty much be defined by "we won't provide support for cpython 2.7 anymore"16:54
fijalexarkun: http://t.co/YzJDlmg216:54
fijaler16:54
exarkunlahwran: It's open source.  Anyone can provide support for it.16:54
fijalhttp://yfrog.com/nxypjrsj16:54
exarkunfijal: Har16:55
lahwranexarkun: this is true, but gnome 2 is "EOL" despite that people have continued it, same deal with KDE 316:55
exarkunlahwran: Sounds like "EOL" doesn't mean anything, just as I said :)16:55
fijallahwran: that's really up to discussion :)16:55
lahwrananyway16:55
antocunianyway, pypy will support python2 at least as long as the translator toolchain is written in python216:56
dmitrey (~quassel@77.121.195.200) joined #pypy.16:56
dmitrey (~quassel@77.121.195.200) left irc: Remote host closed the connection16:56
antocuniand the TT will be written in python2 as long as pypy supports it :-)16:56
G2P (~G2P@fw-asn1.ornis.com) left irc: Quit: Leaving.17:13
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Read error: Operation timed out17:16
Nick change: Taggnostr2_ -> Taggnostr17:20
voidspace (~voidspace@python/psf/voidspace) joined #pypy.17:33
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) left irc: Quit: DasIch17:40
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) joined #pypy.17:41
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) left irc: Read error: Operation timed out17:43
htoothrot (~mux@unaffiliated/htoothrot) left irc: Quit: here we are17:55
htoothrot (~mux@unaffiliated/htoothrot) joined #pypy.17:55
cocoatomo (~cocoatomo@p2118-ipbf901souka.saitama.ocn.ne.jp) joined #pypy.17:57
cocoatomo (~cocoatomo@p2118-ipbf901souka.saitama.ocn.ne.jp) left irc: Remote host closed the connection17:57
whyking (~quassel@128.148.231.16) joined #pypy.18:02
apoirier (apr@210.168.3.109.rev.sfr.net) left #pypy ("Leaving...").18:13
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.18:18
Telugodu (~rambaho@202.3.77.231) left irc: Ping timeout: 255 seconds18:19
Telugodu (~rambaho@202.3.77.231) joined #pypy.18:20
jacob22 (~jacob@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Ping timeout: 240 seconds18:22
jacob22 (~jacob@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy.18:22
ciphergothright, have gone the whole hog and made my own public mirror of the pypy sources18:28
ciphergothso I can easily share my changes http://hg.opensource.lshift.net/pypy18:29
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds18:29
sdfasdklj (8e3a5eb7@gateway/web/freenode/ip.142.58.94.183) joined #pypy.18:31
Nick change: Guest32342 -> zk18:32
sdfasdkljSuper-fast question: I had to translate PyPy with a bunch of modules deactivated; now, when I try to `import tarfile`, I get the following error: "OSError: libpython2.7.dll: No such file or directory". My question is: should pypy have generated this library?18:33
Telugodu (~rambaho@202.3.77.231) left irc: Ping timeout: 240 seconds18:33
Telugodu (~rambaho@202.3.77.231) joined #pypy.18:34
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) joined #pypy.18:37
arigatosdfasdklj: no, pypy doesn't generate libpython2.7.dll18:40
sdfasdkljAah. Thank you!18:44
sdfasdklj (8e3a5eb7@gateway/web/freenode/ip.142.58.94.183) left #pypy.18:44
Tobu (~Tobu@2a01:e35:8a2c:b230:213:d3ff:fe85:5614) left irc: Remote host closed the connection18:45
AlekSi (~AlekSi@176.57.74.133) left irc: Ping timeout: 252 seconds18:52
AlekSi (~AlekSi@176.57.74.133) joined #pypy.18:52
AlekSi (~AlekSi@176.57.74.133) left irc: Remote host closed the connection18:53
voidspace (~voidspace@python/psf/voidspace) left irc: Quit: Goodbye...18:55
Nick change: ericholscher_ -> ericholscher18:56
Tobu (~Tobu@2a01:e35:8a2c:b230:213:d3ff:fe85:5614) joined #pypy.19:02
Tobu (~Tobu@2a01:e35:8a2c:b230:213:d3ff:fe85:5614) left irc: Changing host19:02
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.19:02
16SAALZS3 (~tismer@i59F714CF.versanet.de) left irc: Quit: schnarch19:03
mattip (~mattip@bzq-79-177-109-132.red.bezeqint.net) joined #pypy.19:08
arigatostm progress report:19:08
arigatoit went down from 10.6x to 8.2x slower than regular pypy, by making the accesses to the frame fields direct --- and that's without all reads and writes to fastlocals_w, which are still indirect for now19:09
arigatoat some point, maybe now, I should start thinking about whether it's really a good idea to use the current stm library19:09
arigato(which I wrote myself, but still)19:09
Rhy0litearigato: great!19:09
arigato(based on rstm)19:09
arigato:-)19:10
Alex_Gaynorarigato: how many- cores is it at 8.2x slower?19:10
Alex_Gaynor(and what's the scaling look like)19:10
arigatoAlex_Gaynor: 1 core.  the scaling looks very good19:10
Alex_Gaynormeaning? :)19:10
arigatobut tannit has really only 4 cores (without counting hyperthreading) so I cannot scale it past that19:11
Alex_Gaynorwhat's the time at 2 or 4?19:11
arigatoI suppose it means that on a 16-cores you would see it going faster :-)19:11
antocunigood, then it means that now we can go at half the speed by using 4 cores instead of 119:11
antocuni(sorry :-))19:11
arigatowith 2 or 4 cores, the run-time is almost divided by 2 or 419:11
Rhy0litearigato: more motivation to get PyPy running on POWER architecture :-)19:11
arigatoyes :-)19:11
arigatoantocuni: precisely :-)19:12
Alex_Gaynorgreat!19:12
Rhy0lite(POWER7 in GCC Farm has 16 cores and 64 hardware threads)19:13
arigatoRhy0lite: for run, I should try to get access and run pypy-tm there19:14
arigato(it doesn't use the JIT at all, so far)19:14
Rhy0litearigato: definitely19:15
arigatothe next GC question is the level of logged stm accesses: either word-based (like now) or object-based19:15
arigatoI'm thinking about ideas for object-level accesses that would fit nicely in a simple GC19:16
Rhy0litearigato: you should get your own account, but if I can get your public SSH key, I can add it to Sven's account19:16
Action: arigato Rhy0lite ah, great19:16
Alex_Gaynorarigato: do you mind if I tweet this status?19:16
arigatono, sure19:17
antocuniarigato: object level access means that you still need to stm-write every word but that if you read N fields out of an object you just need to stm-read one word and then do N regular reads?19:17
arigatoantocuni: that's the idea, but no :-)19:19
arigatoit would mean (in the current way it's implemented, called a "lazy/lazy" implementation):19:19
arigatoobjects are either global or local; for simplicity, we can have a gc where global objects are non-movable, mark-sweep,19:20
arigatoand local objects are part of the thread-local nursery19:20
arigatoall accesses to local objects are free19:20
arigatowhen you read from a global object you need to record that fact (probably all reads, to detect if another thread committed in the meantime)19:21
arigatoupon the first write to a global object you make a local copy of the object19:21
arigatoand then all reads and writes are done in this local copy19:21
arigatoat transaction commit time, we perform the equivalent of a minor collection, to move the surviving objects out of the local part19:22
arigato(which would mean either allocate them globally, or if they are already a copy of a global object, just update the global object)19:23
antocunihow do you detect if there were conflicts? Using a timestamp on the object?19:23
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.19:23
arigatoyes, in this model, a timestamp field on every object seems a good compromize for now19:23
antocuniyes, this has the big advantage that you get stm-free read/writes for all the objects which die young19:25
antocuniif I understood it correctly19:25
arigatoyes19:26
arigatoexactly19:26
antocuniuhm, but the other problem is that you might detect conflicts only at commit-time19:27
antocuniwhen are conflicts detected in the current model?19:27
arigatothe current model works pretty much like I described above, except with "words" of memory instead of objects19:27
arigatoso all writes go to a local pool19:28
arigatoand indeed they don't cause detection of immediate conflicts19:28
arigatobut the reads cause such detection19:28
arigatoto talk about the object model, you wouldn't be able to read from a global object that has been modified "too far in the future"19:29
arigato(in fact, if you do, the stm checks if all the previous reads that you already did would still be valid if done now; and if they would, then the transaction's time is updated to "now" and you don't have a conflict)19:30
antocuniI see19:31
arigato(this comment in parenthesis is only one of many variants, but I thought it made sense, and it should also make sense in the object version)19:31
dmalcolm (david@nat/redhat/x-qppgkiamgbezjuau) left irc: Read error: Connection reset by peer19:32
dmalcolm (david@nat/redhat/x-cczpobzcxleeawky) joined #pypy.19:33
ericflo (~ericflo@75.103.8.110) joined #pypy.19:35
amaury__ (~amaury_@74.125.57.34) left irc: Ping timeout: 276 seconds19:38
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy.19:39
asmeurer__ (~asmeurer@dhcp-altamirano-166.resnet.nmt.edu) joined #pypy.19:52
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.19:53
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Read error: Operation timed out20:05
CIA-603papercrane 07roundup * 10#1020/segfault when running test_greenlet.py through pypy pytest.py: 20:09
CIA-6Are you saying that greenlets may be set as a non-working module in the future?20:09
CIA-6That would be a shame.... It's odd that running the greenlet tes ... * 14https://bugs.pypy.org/issue102020:09
Tobu (~Tobu@unaffiliated/tobu) left irc: Ping timeout: 252 seconds20:15
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.20:17
rguillebertarigato, do you compare pypy-stm with a pypy with jit or without ?20:19
arigatorguillebert: without jit20:26
CIA-603arigo 07roundup * 10#1020/segfault when running test_greenlet.py through pypy pytest.py: 20:31
CIA-6I think it works fine with "shadowstack", which is the default on OS/X and20:31
CIA-6Windows but not on Linux. Maybe you can try with a version of pypy tr ... * 14https://bugs.pypy.org/issue102020:31
Guest (~textual@67.51.82.66) joined #pypy.20:36
Nick change: Guest -> rekamso20:36
antocuni (~antocuni@host155-125-dynamic.25-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds20:36
kenaan12arigo stm 11de5b100916b6 15/pypy/translator/stm/transform.py: Comment out this, as it doesn't work reliably.20:37
kenaan12arigo stm 11f0ee46ba0dfb 15/: merge heads20:37
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection20:39
arigatoah, no way to have stm working on a POWER machine right now20:40
arigatothe library is based on the guarantees that Intel chips give, which are stronger than PowerPC20:40
amaury__ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.20:41
inhahe (~inhahe@c-50-128-212-190.hsd1.fl.comcast.net) left irc: Ping timeout: 252 seconds20:42
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.20:47
kenaan12mattip numppy-flatitter 11dd14fbe5d0a4 15/pypy/module/micronumpy/: added signature, all tests now pass20:50
kenaan12mattip numppy-flatitter 1135b17e8dce50 15/pypy/module/micronumpy/interp_numarray.py: translation passes20:50
mattipfijal: ping20:50
mikefc_ (~mike@60-242-240-196.static.tpgi.com.au) joined #pypy.20:52
mattipnumppy-flatitter (sp) passes tests, performance is probably non-optimal.20:54
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.20:54
squiddy (~squiddy@g224193065.adsl.alicedsl.de) left irc: Quit: Leaving20:58
azanella (~azanella@189.6.80.131) left irc: Ping timeout: 245 seconds21:01
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) left irc: Ping timeout: 240 seconds21:07
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) joined #pypy.21:09
mattip (~mattip@bzq-79-177-109-132.red.bezeqint.net) left irc: Quit: bye21:18
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) left irc: Ping timeout: 252 seconds21:19
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) joined #pypy.21:21
fprimex (~fprimex@brent-macbook.sc.fsu.edu) left irc: Quit: http://www.fprimex.com21:22
Rhy0litewoohoo.  translating PyPy on PPC64 doesn't completely fail early21:22
Rhy0liteAre these timings considered good or bad?21:23
Rhy0lite[Timer] Timings:21:23
Rhy0lite[Timer] annotate                       --- 2378.4 s21:23
Rhy0lite[Timer] rtype_lltype                   --- 2267.3 s21:23
Rhy0lite[Timer] pyjitpl_lltype                 ---    6.2 s21:23
Rhy0lite[Timer] ===========================================21:23
Rhy0lite[Timer] Total:                         --- 4651.9 s21:23
arigatouh, that's horrible21:23
fijalRhy0lite: what are you using as your host?21:23
Rhy0litea POWER7 system21:23
fijaler, host python21:24
arigatotannit is around 4 or 5 times faster21:24
arigatoah21:24
Rhy0liteThis is using CPython-2.6.421:24
fijalRhy0lite: so if you use python 2.7, it's about 3x faster21:24
Rhy0liteokay21:24
fijalthat's still horrible21:25
arigatoand if you use pypy, it's still faster (don't know by how much)21:25
fijalarigato: if you have a working JIT on power :)21:25
fijalarigato: about 30% I think21:25
fijalbut [citation needed]21:25
arigatoobviously, I was missing this detail :-)21:25
fijalarigato: does STM have gc support?21:26
Rhy0liteI'll build CPython-2.7 and check21:26
arigatofijal: not yet, at all21:26
arigatoit runs with "--gc=none" only21:26
fijalarigato: ok, so the richards numbers or not very relevant21:26
fijalit might be faster/slower21:26
fijalRhy0lite: still fairly slow21:26
fijalRhy0lite: where does it crash?21:26
arigatoyes, but to a reasonable extend I think21:26
fijalRhy0lite: does it pass all the tests in jit/ directory?21:27
Rhy0litefijal: there still is one stress test failure21:28
Rhy0litebut I just wanted to see what would happen if I tried to translate21:28
dmalcolm (david@nat/redhat/x-cczpobzcxleeawky) left irc: Read error: Connection reset by peer21:28
arigatoRhy0lite: for this purpose, you can also try21:29
dmalcolm (david@nat/redhat/x-jfzpblvlrpemiito) joined #pypy.21:29
Rhy0liteand I don't have the prolog target21:29
Rhy0liteI actually wanted to try translating prolog21:29
arigato"translate.py targetpypystandalone --no-allworkingmodules"21:29
arigatohttps://bitbucket.org/cfbolz/pyrolog21:29
arigatohg clone ^^^21:29
Rhy0liteis it obvious how to merge that into the pypy checkout?21:30
fijalyou don't merge it21:30
fijalyou just run translate from directory where prolog is21:30
fijalat worst you have to setup PYTHONPATH somehow21:30
arigatoyes, something like "/path/to/pypy/translator/goal/translate.py targetprologstandalone"21:30
arigatohum21:31
arigatoyou need to either play with PYTHONPATH, or create a symlink with "ln -s . prolog"21:32
rekamso (~textual@67.51.82.66) left irc: Ping timeout: 245 seconds21:33
arigatoor to move the file targetprologstandalone.py into the parent directory21:33
dmalcolm (david@nat/redhat/x-jfzpblvlrpemiito) left irc: Quit: Leaving21:40
nedbat (~nedbat@python/psf/nedbat) left irc: Disconnected by services21:45
nedbat_ (~nedbat@python/psf/nedbat) joined #pypy.21:46
Tobu (~Tobu@unaffiliated/tobu) left irc: Remote host closed the connection21:48
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.21:49
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.21:54
nickik (~nick@217.150.233.216) joined #pypy.21:56
amaury (~amaury_@74.125.122.49) joined #pypy.22:01
Rhy0litebut mainly I wanted to see if there were any *unexpected* problems with translation on POWER that should be debugged22:02
gutworth (~benjamin@dhcp-128-36-73-23.central.yale.edu) left irc: Quit: leaving22:03
tumbleweedthere are certainly some worrying test failures on power (well, big endian in general) but it seems mostly usable22:03
Rhy0litewell, CPython-2.7.2 is much better22:03
Rhy0lite[Timer] Timings:22:04
Rhy0lite[Timer] annotate                       ---  590.1 s22:04
Rhy0lite[Timer] rtype_lltype                   ---  734.3 s22:04
Rhy0lite[Timer] pyjitpl_lltype                 ---    1.8 s22:04
Rhy0lite[Timer] ===========================================22:04
Rhy0lite[Timer] Total:                         --- 1326.2 s22:04
fijalthat's about tannit speed22:05
fijalI think22:05
amaury__ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 248 seconds22:05
fijalRhy0lite: can you paste the traceback?22:05
Rhy0litehttp://paste.pocoo.org/show/540780/22:06
fijalthis is pypy or prolog btw?22:07
Rhy0litepypy22:09
JaRoel|4d (~jaroel|4d@sink.jaroel.nl) joined #pypy.22:09
fijalok22:09
fijalprolog will be faster22:09
fijalwhy you have there such errors btw?22:09
Rhy0liteI need to figure out what to link22:09
fijal"link"?22:09
Rhy0liteto link to prolog22:11
fijalit's easy22:11
fijalit's PYTHONPATH only22:11
Nick change: nedbat_ -> nedbat22:11
fijalit's PYTHONPATH=.. <path to translate.py> targetprologstandalone.py22:12
fijalI think22:12
EnCuKou (~encukou@ip-94-113-220-25.net.upcbroadband.cz) left irc: Read error: Connection timed out22:15
Rhy0liteI don't know what causes the error22:16
EnCuKou (~encukou@ip-94-113-220-25.net.upcbroadband.cz) joined #pypy.22:16
lucian_ (~lucian@94.197.127.111.threembb.co.uk) joined #pypy.22:16
arigatoRhy0lite: ah, well22:16
arigatoclearly you need first to make at least one transation test pass in the jit22:17
fijalRhy0lite: you have a translation error there22:17
fijalthat was introduced on this branch22:17
fijal(it's not on trunk)22:17
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Quit: pnguyen22:17
Rhy0liteof course22:17
fijalit looks deliberate to me22:18
arigatosomething like: copy test_stuff_translates() from test_ztranslation.py22:18
fijalRhy0lite: does even test_ztranslation in metainterp pass?22:18
fijalprobably22:18
fijalthis is llsupport22:18
fijalbut yes, you totally need ztranslation tests22:18
arigatoyes, sorry, I meant x86/test/test_ztranslation22:18
fijal    def get_on_leave_jitted_hook(self):22:19
fijal        # this function needs to be overridden for things to work with22:19
fijal        # our framework GCs22:19
fijal        translation_time_error22:19
fijalthis is obviously not going to work22:19
fijalI have no idea why this is changed from trunk?22:19
fijalhow is get_on_leave_jitted_hook GC relevant even?22:20
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.22:20
arigatoprobably just an old version22:20
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds22:20
fijalRhy0lite: merge default then?22:20
arigatohah22:20
arigatono22:20
fijalarigato: on default it's lambda : None or so22:20
fijalI don't think it's old version22:20
Rhy0liteokay, test_ztranslation fails 22:21
arigatoonly do that if you're ready to perform the updates needed to the backend, which may take 1-2 days22:21
arigatofijal: yes, it is22:21
fijalRhy0lite: have you seen my mails btw?22:21
fijalDavid removed test from test_runner and put it to x8622:21
fijalthat was a wrong choice, it should work there22:21
Rhy0liteokay22:22
fijalarigato: I suspect you btw ;-)22:23
Action: arigato too, but I remember writing translation_time_error but not removing it22:24
fijalarigato: no, I suspect you about the other failing test :)22:24
fijalthat tests if the backend can get assembler correctly22:25
Rhy0litefijal: I have not seen your emails22:26
Rhy0liteI only see your sprint email22:27
fijalok I wanted to send it apparently :/22:27
Rhy0liteunless you sent it to my personal account22:27
fijalI thought I sent it to pypy-dev22:28
Rhy0liteI eagerly await to see what the overlords at Google do22:28
arigatoah, the translation_time_error was introduced on the arm branch only22:28
Rhy0liteyes22:28
fijal839659291f0322:29
Rhy0liteDavid merges default to ARM22:29
Rhy0liteand then ARM to PPC22:29
fijalRhy0lite: what overlords?22:29
arigatoyes22:29
arigatowell, the comment in that function should explain the whole problem22:30
fijalit totally does not :)22:30
arigatouh?22:30
Rhy0liteI don't know if I want to try to fix it and diverge or work with David to fix it upstream22:30
arigatofijal: you need to override that method in the backend22:30
fijalwell maybe it does :)22:30
Rhy0litefijal: Google's change in privacy policy22:30
fijalraise Exception(comment)22:30
fijalwould be better ;-)22:30
Rhy0liteand merging more across Google platforms22:31
fijalRhy0lite: I thought they already do it22:31
fijalI was surprised to find out they;'re not22:31
Rhy0litefijal: no, of course not!22:31
Rhy0litethey're not evil!22:31
arigatofijal: well, no, we want a failure at translation time22:31
Rhy0liteand they will be even less evil after the change 22:32
Rhy0lite;-)22:32
arigatofijal: not just a program that crashes with "fatal RPython error: Exception"22:32
fijalarigato: ah ok :)22:32
fijalarigato: good :)22:32
Rhy0litefijal: I should ask Google: "when did you stop beating your evil users?"22:32
lucian_ (~lucian@94.197.127.111.threembb.co.uk) left irc: Read error: Connection reset by peer22:32
lucian (~lucian@94.197.127.111.threembb.co.uk) joined #pypy.22:33
amauryHi, I played with the idea of disallowing NUL bytes in filenames and most posix functions22:33
fijalRhy0lite: anyway offtopic!22:33
fijalarigato: yes ok ok, I see :)22:33
Rhy0litefijal: :-P22:33
fijalamaury: can we merge 2.7.2 to default before that btw?22:34
amauryyes, it's a separate thing22:34
Rhy0liteI'll go home.  If there is somethings specific that you want me to change to make further progress translating, I can do it when I'm home22:35
fijalRhy0lite: write the function on ppc backend :)22:35
Rhy0liteokay22:36
Rhy0lite (dje@nat/ibm/x-hlxrgwqsszrxytuu) left irc: Quit: Leaving22:36
fijalI'm surprised there is no direct test22:36
nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 276 seconds22:37
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.22:42
Turicas (~alvaro@mail.justen.eng.br) joined #pypy.22:43
arigatowe should add a test to runner_test22:44
arigatosomething that tries to translate only the backend22:45
whyking (~quassel@128.148.231.16) left irc: Read error: Operation timed out22:49
fijalwith all the features preferably22:54
lucian (~lucian@94.197.127.111.threembb.co.uk) left irc: Quit: Bye23:01
gtaylor (~gtaylor@108-196-160-156.lightspeed.tukrga.sbcglobal.net) left irc: Quit: Konversation terminated!23:06
Arfrever (~Arfrever@apache/committer/Arfrever) left irc: Ping timeout: 260 seconds23:15
bbot2 (~bbot2@wyvern.cs.uni-duesseldorf.de) joined #pypy.23:17
Nick change: Guest1859 -> jell23:19
arigato (~arigo@adsl-89-217-207-222.adslplus.ch) left irc: Quit: See you23:30
Rhyolitefijal: I don't see translation_time_error in the changeset that you referenced23:31
whitelynx (~whitelynx@63.241.75.144) left irc: Quit: Ex-Chat23:42
mikefc (9665bcc5@gateway/web/freenode/ip.150.101.188.197) got netsplit.23:46
htoothrot (~mux@unaffiliated/htoothrot) got netsplit.23:46
MostAwesomeDude (JuTorzalS0@osuosl/staff/mostawesomedude) got netsplit.23:46
Nick change: mikefc_ -> mikefc23:46
Possible future nick collision: mikefc23:46
-tomaw (tom@freenode/staff/tomaw) to $$*- [Global Notice] Hi, we're experiencing an outage at once of our sponsors, causing NickServ/ChanServ and friends to have left as well as some server being unavailable. We're looking into the cause now. More updates by wallops, /mode yournick +w to see them!23:49
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: Ping timeout: 276 seconds23:51
voidspace (~voidspace@87-194-212-65.bethere.co.uk) joined #pypy.23:51
Dulak (~michael@ip98-167-237-72.ph.ph.cox.net) left irc: Quit: Leaving23:52
tito (~tito@pdpc/supporter/active/tito) left irc: Ping timeout: 240 seconds23:55
tito (~tito@178.73.216.222) joined #pypy.23:55
[Arfrever] (~Arfrever@apache/committer/Arfrever) left irc: Ping timeout: 252 seconds23:56
bbot24Finished ??: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/4323:56
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/4423:56
MostAwesomeDude (JuTorzalS0@osuosl/staff/mostawesomedude) got lost in the net-split.23:57
htoothrot (~mux@unaffiliated/htoothrot) got lost in the net-split.23:57
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.23:58
MostAwesomeDude (JuTorzalS0@osuosl/staff/mostawesomedude) joined #pypy.23:58
--- Thu Jan 26 201200:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!