#pypy IRC log for Monday, 2011-11-21

lambacck (~chris@d24-150-124-118.home.cgocable.net) joined #pypy.00:05
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) joined #pypy.00:06
kvda (~kvda@124-168-163-66.dyn.iinet.net.au) joined #pypy.00:07
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) left irc: Client Quit00:07
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.00:08
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 252 seconds00:16
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer00:19
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.00:21
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.00:24
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer00:24
Nick change: Gulaway -> Gulopine00:26
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.00:30
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer00:30
JaredW (~jared@061093200075.ctinets.com) joined #pypy.00:39
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.00:42
stakkars___ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.00:44
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer00:46
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Ping timeout: 248 seconds00:47
tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Ping timeout: 240 seconds00:50
kvda (~kvda@124-168-163-66.dyn.iinet.net.au) left irc: Quit: x___x00:50
Nick change: Gulopine -> Gulaway00:53
maxyz (~maxy@186.23.74.228) left irc: Ping timeout: 252 seconds00:56
Ademan (~dan@adsl-71-141-224-79.dsl.snfc21.pacbell.net) joined #pypy.00:58
maxyz (~maxy@186.23.74.228) joined #pypy.00:58
berkerpeksag (~berkerpek@46.197.8.139) left irc: Quit: Instantbird 1.101:01
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/14501:02
JaredW (~jared@061093200075.ctinets.com) left irc: Ping timeout: 252 seconds01:07
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.01:10
stakkars___ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer01:11
Surgo_ (~surgo@gw1.miraiyashoten.co.jp) joined #pypy.01:13
ronan (~ronan@host86-142-249-87.range86-142.btcentralplus.com) joined #pypy.01:18
bbot24Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/71901:27
tilgovi (~randall@99.13.242.166) joined #pypy.01:28
tilgovi (~randall@99.13.242.166) left irc: Changing host01:28
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.01:28
Nick change: Gulaway -> Gulopine01:32
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/185301:42
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/58201:42
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/27701:42
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/61801:42
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/111601:42
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/80301:42
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/144501:42
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/71401:42
qbproger (~qbproger@c-76-104-159-248.hsd1.wa.comcast.net) joined #pypy.01:42
mfoord (~anonymous@python/psf/voidspace) left irc: Quit: mfoord01:45
qbproger_ (~qbproger@c-76-104-159-248.hsd1.wa.comcast.net) joined #pypy.01:46
qbproger (~qbproger@c-76-104-159-248.hsd1.wa.comcast.net) left irc: Ping timeout: 244 seconds01:49
Nick change: qbproger_ -> qbproger01:50
qbprogerAlex_Gaynor: ping01:50
Alex_Gaynorpong-ish, in the middle of a pycon-pc meeting01:50
qbprogerah ok01:51
qbprogeri got a script that's about 2.5x slower on the last pypy release01:51
Alex_Gaynor:( can you try trunk?01:52
Alex_Gaynor2.5x slower than CPython or a previous pypy?01:52
qbprogerCPython 2.601:52
Alex_GaynorThat's no good.01:52
qbprogeri was doing some project euler...01:53
qbprogerAlex_Gaynor: https://gist.github.com/138141301:56
Alex_Gaynorqbproger: you said you tried with 1.6?01:56
qbprogeryea01:57
qbprogeron windows01:57
Ademan (~dan@adsl-71-141-224-79.dsl.snfc21.pacbell.net) left irc: Ping timeout: 240 seconds01:57
Alex_Gaynorcan you try trunk?01:57
qbprogeris there an exe to download somewhere?01:57
Alex_Gaynorbuildbot.pypy.org/nightly should01:58
Ademan (~dan@adsl-71-141-224-79.dsl.snfc21.pacbell.net) joined #pypy.01:59
qbprogerthat page gives me: Firefox can't establish a connection to the server at localhost:8099.02:02
Alex_Gaynorhttp://buildbot.pypy.org/nightly/trunk/02:02
Alex_Gaynorepic redirect fail02:02
qbprogerAlex_Gaynor: pypy-c-jit-latest-win32.zip 16M 2011-11-14 <--- that one good?02:03
Alex_Gaynorqbproger: yeah, should be02:04
asmeurer_ (~asmeurer@dhcp-baca-230.resnet.nmt.edu) left irc: Quit: asmeurer_02:07
DasIch_ (~dasich@p3E9915CF.dip.t-dialin.net) joined #pypy.02:10
DasIch (~dasich@p3E990375.dip.t-dialin.net) left irc: Ping timeout: 258 seconds02:14
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection02:20
Ademan (~dan@adsl-71-141-224-79.dsl.snfc21.pacbell.net) left irc: Quit: leaving02:30
Shanita (~John@osbk-4db16952.pool.mediaWays.net) left irc: Remote host closed the connection02:33
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/94802:41
asmeurer__ (~asmeurer@70.56.139.215) joined #pypy.02:43
qbprogerAlex_Gaynor: it's actually slower on trunk02:49
Alex_Gaynorqbproger: epic fail.  can you compare it with and without the caching?02:49
qbprogerI can't do the recursion as deep without it02:50
mattip (~chatzilla@bzq-79-178-115-92.red.bezeqint.net) joined #pypy.02:52
qbprogerfaster than cpython, but still slower on trunk02:52
mattiphi02:54
mattipIs there a "pypy way" to write a unittest-like timing test02:55
mattipin order to compare two changesets?02:55
Alex_Gaynorqbproger: ok, interested in trying to debug the issue yourself, or want to just file an issue on it?02:55
qbprogerAlex_Gaynor: i'll file an issue02:56
Alex_Gaynorqbproger: k, if you feel like debuging it http://lostinjit.blogspot.com/2011/11/analysing-pythons-performance-under.html should be a good start02:57
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.02:58
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer02:58
qbprogerAlex_Gaynor: is there a link to the bug tracker on this page: http://pypy.org/03:02
Alex_Gaynorqbproger: its linked on the contact one03:02
mattip (chatzilla@bzq-79-178-115-92.red.bezeqint.net) left #pypy.03:06
Sho_ (~EHS1@kde/hein) left irc: Quit: Stop leaking memory like it's going out of fashion.03:25
CIA-6803qbproger 07roundup * 10#930/Recursion with cache slower on pypy than CPython 2.6: (log message trimmed)03:27
CIA-68[new] This runs slower on Pypy 1.6 (and trunk) than CPython 2.6.5. CPython 2.7 runs03:27
CIA-68this even faster03:27
CIA-68OR_VAL = [0] * 1003:27
CIA-68FINAL_CHECK = 0 * 14https://bugs.pypy.org/issue93003:27
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/185303:40
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds03:42
tilgovi (~randall@c-76-102-28-39.hsd1.ca.comcast.net) joined #pypy.03:57
tilgovi (~randall@c-76-102-28-39.hsd1.ca.comcast.net) left irc: Changing host03:57
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.03:57
kenaan12mattip numpy-monkeyaround 11425ecbd26101 15/: branch for experimental implementations04:00
kenaan12mattip numpy-monkeyaround 111c02422a8f96 15/pypy/module/micronumpy/test/test_numarray.py: Add test for arange04:00
kenaan12mattip numpy-monkeyaround 111ae3fd3bf105 15/pypy/module/micronumpy/: test, implement arange04:00
asmeurer__ (~asmeurer@70.56.139.215) left irc: Quit: asmeurer__04:05
DasIch_ (~dasich@p3E9915CF.dip.t-dialin.net) left irc: Ping timeout: 260 seconds04:17
asmeurer (~asmeurer@70.56.139.215) joined #pypy.04:17
[Arfrever] (~Arfrever@apache/committer/Arfrever) left irc: Remote host closed the connection04:18
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/144504:32
DasIch (~dasich@p4FFDF354.dip.t-dialin.net) joined #pypy.04:33
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/80304:34
[Arfrever] (~Arfrever@apache/committer/Arfrever) joined #pypy.04:35
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/61804:39
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Read error: Operation timed out04:41
tilgovi (~randall@c-76-102-28-39.hsd1.ca.comcast.net) joined #pypy.04:41
tilgovi (~randall@c-76-102-28-39.hsd1.ca.comcast.net) left irc: Changing host04:41
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.04:41
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.04:43
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer04:43
Arfrever (~Arfrever@apache/committer/Arfrever) left irc: Quit: Ex+re04:48
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/111604:51
asmeurer (~asmeurer@70.56.139.215) left irc: Quit: asmeurer04:56
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/58205:04
zk (~is@whatit.is) left irc: Quit: leaving05:08
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.05:12
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Client Quit05:12
asmeurer_ (~asmeurer@70.56.139.215) joined #pypy.05:12
zk (~is@whatit.is) joined #pypy.05:15
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.05:21
xcombelle (~chatzilla@AToulouse-551-1-14-118.w86-196.abo.wanadoo.fr) joined #pypy.05:22
Vorpal (~AnMaster@unaffiliated/anmaster) left irc: Quit: ZNC - http://znc.sourceforge.net05:29
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/27705:37
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy.05:45
fijal (~fijal@197.174.45.1) joined #pypy.06:00
ericholscher (~eric@173.203.223.63) left irc: Excess Flood06:02
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/71406:02
ericholscher (~eric@173.203.223.63) joined #pypy.06:02
CIA-6803fijal 07roundup * 10#930/Recursion with cache slower on pypy than CPython 2.6: 06:07
CIA-68[chatting] This program finishes in 0.05s for me. This is definitely not enough for the JIT06:07
CIA-68to warm up. * 14https://bugs.pypy.org/issue93006:07
fijalAlex_Gaynor: we should ask people not to submit stuff like that06:08
Alex_Gaynorfijal: I didn't look at the total tim06:08
Alex_Gaynorfijal: we should write a document on how to file performance bugs06:09
fijalto be honest, if you run it 100 times it's 2.6s vs 3s06:09
fijalwhere pypy loses06:09
fijalso "ok"06:09
Alex_Gaynornot good though06:09
Alex_Gaynorwe should win :)06:09
Alex_Gaynor1000 times?06:09
Action: fijal does not have 40s06:11
fijalbut also - we should clear the caches each time06:11
fijalso I don't feel like doing something there06:11
fijal (~fijal@197.174.45.1) left irc: Read error: Connection reset by peer06:13
fijal (~fijal@41.53.143.199) joined #pypy.06:15
fijalback06:15
fijalAlex_Gaynor: http://arewefastyet.com/06:16
fijaland JS vendors are reporting 0.68% speedup on something06:16
fijalhow the fuck do they measure that?06:16
Alex_Gaynorfijal: if you read the firefox dev tracker they count the number of ms improvements thing get06:16
Alex_Gaynorfijal: holger's jenkins runs on their servers as well I see :D06:16
Action: fijal fails to understand06:17
Alex_Gaynorunderstand what?06:17
Alex_Gaynorfijal: I fail to understand how anyone can write a JIT by hand, so maybe I'm not one to comment :)06:18
qbprogerfijal: it was 0.5s for me, so I thought that was long enough cause I thought there were benchmarks on speed.pypy.org that run in that time06:18
fijalqbproger: are you sure not 0.05s?06:18
fijalAlex_Gaynor: JS is simpler06:18
qbprogermy laptop is a bit weak :\06:19
qbprogerthat might be it06:19
Alex_Gaynorfijal: not that simple, every obj can have valueOf which is used on every operation, I think you can even monkey patch this onto integer06:19
Alex_Gaynorfijal: plus you have to do crazy stuff like try to detect when a float is really an int to be fast06:19
fijalright sure06:20
fijalbut it's smallert06:20
fijalyou don't have to worry about generator expressions06:20
fijallongs06:20
Alex_Gaynorstill though06:20
fijaltons and tons of other stuff06:20
fijalbut you still can't measure ms improvements06:20
Alex_Gaynorevery time standards comittee adds a new type you have to add new stuff to the JIT06:20
Alex_Gaynorwhen they add a language feature to chrome they have to update machine code stuff, that's crazy IMO06:21
mattip (c1a9466d@gateway/web/freenode/ip.193.169.70.109) joined #pypy.06:21
qbprogerfijal: that was on windows too, not sure if that has anything to do with it06:22
fijalno no no06:22
fijalstandards commitee works way slower than python-dev06:22
fijalthere were no new JS features in years!06:22
fijalqbproger: windows is probably slightly slower06:22
fijalbut not by much06:22
Alex_Gaynorfijal: yes, but JIT keeps up nicely wiht python-dev06:22
fijalqbproger: anyway, can you provide a version that lets the JIT warm up06:23
fijalAlex_Gaynor: yes? we're years behind06:23
qbprogerfor me it was pypy 0.416 vs cpython (2.6) 0.189s06:23
Alex_Gaynorfijal: too bad we didn't merge 2.7.2 for 1.7 release06:23
fijalqbproger: it's 0.3 vs 0.05 for me06:24
fijalbut 0.07 with --jit off06:25
fijalso it's tracing06:25
fijalwe kinda know tracing in recursion takes tons of time06:25
qbprogerhmmm06:25
qbprogerso just wrap 'main' in a for loop?06:25
fijalkinda06:25
fijalyeah06:25
qbprogeralright, I can make a post with that modification06:26
fijalwait a sec :)06:26
fijalso for a 100, I get 3.0s pypy 2.6s cpython 2.6s pypy --jit threshold=-1 (only compiling functions)06:27
fijalthat's bad but not *too* bad06:27
fijalAlex_Gaynor: feel like helping me with the release?06:27
qbprogerif you take out the cache pypy wins06:27
Alex_Gaynorfijal: what would I need to do?  I'm rwiting papers ATM and tomorrow I'm travelling06:28
fijalsomeone has to download nightly builds, rename directories and put them on bitbucket06:28
Alex_Gaynorprobably no time for that, sorry06:28
fijalI have internet limit/month so would be cool if someone else does this :)06:28
Alex_Gaynorcan't you do it form tannit?06:28
fijalI'll get someone else to do it then06:28
fijalannoying 06:28
fijalbut I guess I can06:28
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.06:30
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer06:30
mattipfijal: hi06:30
qbprogerfijal: anything you want me to do with my bug report? (btw, perf bug guidelines are a good idea)06:31
mattipFWIW, commit 1109220a3f29a0 15/pypy/module/micronumpy/interp_numarray.py: fix signatures should not have been necessary,06:32
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 244 seconds06:32
xcombelle (~chatzilla@AToulouse-551-1-14-118.w86-196.abo.wanadoo.fr) left irc: Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243]06:32
mattipbut it's hard for me to beleive hg is misbehaving.06:32
mattipI don't understand.06:33
mattipSo I have created a branch and you are welcome to merge from/to it.06:33
mattipOr to ignore it.06:33
fijalmattip: hi06:34
fijalqbproger: paste your sessions maybe?06:34
fijalmattip: cool, thanks!06:34
qbprogermy runtimes?06:34
fijalqbproger: yeah, when you run it 100 times?06:34
qbprogerok06:34
fijalthere is a blog post feel free to look into that06:35
fijalqbproger: btw, a 4-character indent would be cool :)06:35
fijalmattip: I'll look at it just now06:35
qbprogerfijal: it is 4 character, isn't it?06:36
fijalI count 306:36
qbprogerwhere?06:37
fijalin place_numbers06:37
qbprogerhmmm, I count 406:38
fijalwhatever06:39
mat^2 (~mathias@212.130.113.35) left irc: Ping timeout: 240 seconds06:39
asmeurer_ (~asmeurer@70.56.139.215) left irc: Quit: asmeurer_06:39
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.06:40
mattip (c1a9466d@gateway/web/freenode/ip.193.169.70.109) left irc: Quit: Page closed06:41
CIA-6803qbproger 07roundup * 10#930/Recursion with cache slower on pypy than CPython 2.6: (log message trimmed)06:42
CIA-68This was on windows 7 6406:42
CIA-68I put a for loop around main to run it 100 times06:42
CIA-68python 2.6:06:42
CIA-68$ time python 178.py * 14https://bugs.pypy.org/issue93006:42
qbprogeri found a change that makes it a bit worse for pypy, made the cache into a function decorator06:44
asmeurer_ (~asmeurer@70.56.139.215) joined #pypy.06:46
fijal (~fijal@41.53.143.199) left irc: Read error: Connection reset by peer06:51
fijal (~fijal@197.171.4.234) joined #pypy.06:54
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 252 seconds06:54
iori (~iori@110-133-45-54.rev.home.ne.jp) joined #pypy.06:55
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.06:56
Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Quit: leaving07:14
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 240 seconds07:20
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy.07:22
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Ping timeout: 260 seconds07:33
kenaan12fijal default 11f0938cd9a6f1 15/pypy/doc/release-1.7.0.rst: work out the release announcement07:34
asmeurer_ (~asmeurer@70.56.139.215) left irc: Quit: asmeurer_07:36
asmeurer (~asmeurer@70.56.139.215) joined #pypy.07:37
asmeurer (~asmeurer@70.56.139.215) left irc: Client Quit07:41
qbproger (~qbproger@c-76-104-159-248.hsd1.wa.comcast.net) left irc: Quit: Leaving07:50
asmeurer (~asmeurer@71.216.240.168) joined #pypy.07:53
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.07:55
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep07:56
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.07:56
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Client Quit07:56
roelbruggink (~jaroel|4d@office.fourdigits.nl) joined #pypy.07:58
asmeurer (~asmeurer@71.216.240.168) left irc: Ping timeout: 248 seconds07:58
fijal (~fijal@197.171.4.234) left irc: Ping timeout: 248 seconds08:00
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.08:00
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Ping timeout: 244 seconds08:00
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Read error: Connection reset by peer08:02
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.08:02
whitelynx (~whitelynx@li117-47.members.linode.com) left irc: Quit: Ex-Chat08:05
asmeurer__ (~asmeurer@71.216.240.168) joined #pypy.08:06
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 260 seconds08:06
asmeurer__ (~asmeurer@71.216.240.168) left irc: Client Quit08:08
canta (~canta@77-20-123-240-dynip.superkabel.de) left irc: Quit: Leaving.08:08
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) joined #pypy.08:17
stakkars_ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Ping timeout: 276 seconds08:19
kvda (~kvda@124-168-163-66.dyn.iinet.net.au) joined #pypy.08:20
fijal (~fijal@197.169.130.1) joined #pypy.08:21
stakkarsfijal: what about the release:08:21
stakkarsfijal: can I get my change for continue let/asm support on windows in?08:22
Kaskuka (~John@osbk-4db1653f.pool.mediaWays.net) joined #pypy.08:23
fijalstakkars: how about we do 1.7.1 next week?08:23
fijalinstead?08:23
asmeurer_ (~asmeurer@71.216.240.168) joined #pypy.08:24
stakkarsok. Armin asked me to do it, but probably assumed it would be ready on Friday.08:25
stakkarsbut took me the Saturday, too08:25
stakkars__ (~tismer@dslb-088-072-196-160.pools.arcor-ip.net) left irc: Ping timeout: 258 seconds08:25
stakkarsand actually I don't trust my skills enough to modify a tested release...08:26
stakkars_ (~tismer@dslb-088-074-039-231.pools.arcor-ip.net) joined #pypy.08:28
bivab (~david@fwstups.cs.uni-duesseldorf.de) joined #pypy.08:28
fijalI'm not a huge fan of doing last-minute changes08:29
fijalgiven that we release at worst every 3 months08:29
fijalit's not like you have to wait 2 years for your changes to get in :)08:29
Alex_Gaynorfijal: we really should merge 2.7.2 in for next release08:31
fijales08:31
fijalyes08:31
fijalit's a mess usually though08:31
Alex_Gaynorright08:31
fijalstakkars: note that until someone fixes it, stackless is still borderline usable08:31
fijalbecause it disables the jit08:31
Alex_Gaynorit only disables the JIT in the function which calls switch08:32
fijalwell08:32
fijalyes, bad enough08:32
fijalhow can you use stackless without switching?08:32
Alex_Gaynoryou can't :)08:36
Alex_Gaynorbut if your CPU time is in a loo pwithout switch, it's maybe ok08:36
stakkars_ (~tismer@dslb-088-074-039-231.pools.arcor-ip.net) left irc: Read error: Connection reset by peer08:37
stakkars_ (~tismer@dslb-088-074-039-231.pools.arcor-ip.net) joined #pypy.08:37
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep08:45
CIA-6803arigo 07roundup * 10#292/try using const keyword for rpython strings with C backend: 08:46
CIA-68[wontfix] No, that doesn't work. No preallocated structure is really constant, because08:46
CIA-68some GC flags may be set and cleared in the header. Closing. * 14https://bugs.pypy.org/issue29208:46
arigato (~arigo@1.106.113.82.net.de.o2.com) joined #pypy.08:47
fijalarigato: hi armin08:47
arigatohi08:47
fijalarigato: can you repackage files and upload them to bitbucet?08:47
fijalbitbucket08:47
fijalI don't want to waste 100M of my bandwidth :/08:47
Alex_Gaynorarigato: what about const on the individual fields, or is that a C++ thing?08:47
arigatoer, yes, but I don't have good bandwidth here either08:48
arigatoI'm fine with doing it on tannit08:48
fijalcan you upload files to bitbucket on tannit?08:48
fijalanyway, please do08:48
arigatofijal: the answer is "yes", even though it might be a bit convoluted08:48
fijalit's the matter of changing directories and renaming it I presume08:49
fijalsomeone can also read/fix the release announcement08:49
arigatoAlex_Gaynor: I think that's a C++ thing, because in C I don't see how you would set exceptions to the rule of const-ness for when you really need to initialize the fields08:50
Alex_Gaynorarigato: indeed08:50
fijalarigato, Alex_Gaynor: I wrote the release announcement08:51
fijalcodename is silly, maybe we can have something better?08:51
arigato:-)08:52
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy.08:54
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Remote host closed the connection08:56
arigatopypy/tool/release/make_release.py is either completely outdated or needs more trickery08:56
fijalI think both08:57
arigato:-)08:57
Alex_Gaynorboth is a very pypy way08:57
arigatonowadays, the releases are just what is in nightly/release-1.7.x/, right?08:57
fijalyes08:58
asmeurer_ (~asmeurer@71.216.240.168) left irc: Quit: asmeurer_08:58
fijalbut you need to change directory name and the archive name08:58
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy.08:58
arigatook, if that's all, I can do it manually08:58
fijalthat's precisely why I never bothered...08:59
fijalbecause it's always manual08:59
fijaler08:59
arigatoexcept that there isn't a win3208:59
fijalbecause it's always simple :)08:59
arigatook08:59
fijaluh?08:59
fijalthere is?08:59
fijalhttp://buildbot.pypy.org/nightly/release-1.7.x/pypy-c-jit-48924-a01f9701efa7-win32.zip08:59
arigatowin32 is anyway waiting08:59
arigatothat's from the 14th of november08:59
arigatoI assume it's too old08:59
fijalah08:59
fijalok08:59
fijalso no08:59
fijalas usual08:59
fijalyes, way too old09:00
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Client Quit09:00
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy.09:00
fijalit's waiting for a button press or so09:00
arigatoand we don't care that these archives contain two failing tests?09:00
arigatoonly the "own" tests that hadn't been fixed so far09:01
arigato...ah, they don't contain the "own" tests anyway09:01
fijaluh09:01
fijalthey have been fixed09:01
fijalyou mean the ztranslation ones on x86 backend?09:01
arigatoyes, but I forgot that the tests are not in the archive themselves09:02
fijalyes09:02
fijalthe builds don't contain fixes09:02
fijalbut it does not matter09:02
arigatoso I rename the dirs to "pypy-1.7"...09:03
arigato(pff, bad connectivity)09:03
Action: fijal glosses through how-to-release09:06
fijalI wonder if I can kill "a meeting or meetings should be organized"09:06
fijalfrom how-to-release09:06
kenaan12fijal default 113cde514a9a92 15/pypy/doc/how-to-release.rst: kill the meeting talk09:08
arigatohum09:10
fijalyes?09:10
arigatogoing to be a real mess for me to upload without using my connection09:11
fijalyes :/09:11
fijalprecisely why I wanted someone with a better connection09:11
arigatolynx or links both fail to work09:11
arigatoin the worst case I'll be at the university in 3 hours, but it's still an interesting challenge :-)09:13
stakkars_ (~tismer@dslb-088-074-039-231.pools.arcor-ip.net) left irc: Read error: Connection timed out09:16
stakkars_ (~tismer@dslb-088-074-039-231.pools.arcor-ip.net) joined #pypy.09:17
G2P (~G2P@fw-asn1.ornis.com) joined #pypy.09:18
fijalstakkars: for what is worth, the windows won't be ready today09:18
fijalso I suppose it's probably fine to merge your changes09:18
stakkarsfijal: if Armin is ok with that, fine. I was just trying to get it in time because he asked for it.09:19
stakkars__ (~tismer@dslb-088-074-039-231.pools.arcor-ip.net) joined #pypy.09:20
stakkarsbut that's good, I will probably merge win664-stage1 before 1.7.1 09:20
stakkars_ (~tismer@dslb-088-074-039-231.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds09:23
fijal (~fijal@197.169.130.1) left irc: Ping timeout: 240 seconds09:26
kkris (~kris@93-82-44-6.adsl.highway.telekom.at) joined #pypy.09:26
fijal (~fijal@197.104.70.70) joined #pypy.09:28
ojii (~ojii@40-34.60-188.cust.bluewin.ch) joined #pypy.09:30
stakkars__ (~tismer@dslb-088-074-039-231.pools.arcor-ip.net) left irc: Ping timeout: 244 seconds09:46
arigatoah09:48
arigatogetting somewhere09:48
arigatofijal: upload done09:51
ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) joined #pypy.09:51
arigatofor reference: "links" works fine after (1) we change some slightly-crazy options, most importantly "No referer" -> "Send real referer only to the same server"09:52
arigatoand (2) after I figured out that we have to type the file name twice in html09:52
arigato(I guess in firefox there is special javascript that copies the file name automatically)09:52
fijalpfff09:53
Last message repeated 2 time(s).09:53
fijalok cool;09:53
fijalI'll update the pypy.org website09:53
fijalarigato: feel like skimming the release announcement once more?09:53
arigatoyes09:53
fijalwould be cool if someone downloads the binaries and checks if they run09:54
fijalany volunteers?09:54
jml (~jml@mumak.net) joined #pypy.09:54
jml (~jml@mumak.net) left irc: Changing host09:54
jml (~jml@canonical/launchpad/jml) joined #pypy.09:54
fijalshould I (as usual) update download.html to say "windows not ready yet"?09:55
fijalah09:56
fijalchecksums and source exports09:56
kvda (~kvda@124-168-163-66.dyn.iinet.net.au) left irc: Read error: Connection reset by peer09:56
arigatowindows not ready yet: yes09:57
fijalchecksums?09:58
arigatomd5 + sha?09:58
fijalyes09:58
fijalfor 32/64/osx09:58
kenaan12fijal release-1.7.x 117a1d5bd05243 15/.hgtags: Added tag release-1.7 for changeset ff4af8f3188209:59
kenaan12arigo default 113c59b9c3ba47 15/pypy/doc/release-1.7.0.rst: Fix.10:01
kenaan12arigo default 11aed533007ccf 15/pypy/doc/how-to-release.rst: merge heads10:01
arigatosource export...10:01
fijalno, we don't need source export10:01
fijalbitbucket does that automatically10:01
arigatoah, from the .hgtag10:02
arigatogood10:02
arigatoshouldn't the .hgtags addition be done on trunk?10:04
fijalno?10:04
fijalI don't think so10:05
fijalbecause you tag the state of the branch no?10:05
fijalif I'm wrong then maybe merge this changeset?10:05
arigatobut .hgtags on trunk lists release-1.X10:05
fijalok10:05
kenaan12fijal pypy.org[extradoc] 110a1d901e60ef 15/: bump the release number10:06
kenaan12fijal pypy.org[extradoc] 11e2d26191fd54 15/source/download.txt: checksums10:06
kenaan12fijal pypy.org[extradoc] 11a92d5719d9dc 15/source/download.txt: merge10:06
kenaan12fijal pypy.org[extradoc] 113b0fd88061dc 15/: regen html10:06
voidspace (~voidspace@python/psf/voidspace) joined #pypy.10:06
fijalok10:06
fijalwe seem to be good to go10:06
arigatogood10:06
fijalanything that can stop us?10:06
fijalbuhahahaha10:06
arigato:-)10:07
fijalvoidspace: hey10:07
arigatoworld domination is there, or something10:07
fijalmy internet is super slow10:07
kenaan12fijal default 115c2c8665692c 15/.hgtags: Added tag release-1.7 for changeset ff4af8f3188210:08
arigato"hg transplant" has the strange result of me pushing a single commit with "fijal" as the author10:08
fijal (~fijal@197.104.70.70) left irc: Read error: Connection reset by peer10:08
arigatoI wonder: if two people transplant the same checkin to the same new branch10:10
fijal (~fijal@197.105.215.160) joined #pypy.10:10
arigatodo they end up with an identical checkin?10:10
fijalyes10:10
fijalif you use transplant10:10
arigatook, good10:10
fijalanything else?10:11
fijalor do I just push it to the blog and send mails?10:11
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.10:12
arigatogood to go, I suppose10:13
kenaan12fijal default 114330437caa7d 15/.hgtags: Added tag release-1.7 for changeset ff4af8f3188210:15
kenaan12fijal default 1110cb0adcfdd9 15/pypy/doc/release-1.7.0.rst: merge10:15
arigato...no10:17
kenaan12fijal default 11a8545592fa3e 15/pypy/doc/release-1.7.0.rst: minor fixes10:18
arigatowe both hg-transplant-ed the same revision to "default", and it ends up as two different changesetes10:18
fijalpfff10:18
fijalnice10:18
xorAxAxwell, of course, because every transplanted changeset has a different head10:18
arigatoah, it's because the ancestor is different10:18
xorAxAxancestor i mean :)10:18
arigatoyes10:19
fijaleh10:19
fijalI forgot to add signature I suppose10:19
fijalbut well, whatever10:19
arigatosignature?10:19
cfbolz (~cfbolz@dslb-178-003-061-166.pools.arcor-ip.net) joined #pypy.10:20
cfbolzhello10:20
fijalhi carl10:20
cfbolzfijal: I can read the release announcement10:20
stakkars__ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.10:20
fijalcfbolz: it's gone10:20
cfbolzah, damn10:20
Topic changed on #pypy by fijal!fijal@197.105.215.160: PyPy, a fast and compliant python interpreter | IRC logs: http://www.tismer.com/pypy/irc-logs/pypy/latest.log.html | PyPy 1.7 released10:20
fijalsorry10:21
cfbolznp10:21
fijalcan we merge list-strategies/specialized-tuples etc now?10:21
Alex_Gaynorspecialized tuple isn't really AFAIK10:21
Alex_Gaynorlist-strategies was, I think10:21
fijalready?10:21
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds10:21
Alex_Gaynoryes10:21
cfbolzplease go ahead with list-strategies10:22
Alex_Gaynorfijal: feel like giving it a final review and merging?10:23
fijalI can 10:23
fijalI'm a bit busy with other stuff though10:23
Alex_Gaynorcool, I reviewed a few weeks ago, but another person wouldn't hurt10:23
kenaan12l.diekmann type-specialized-instances 11118048845a8d 15/pypy/objspace/std/: reverted last fix, found a new (and better) one10:24
cfbolzAlex_Gaynor: if you are in a reviewing mood, set-strategies can be reviewed too10:26
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.10:26
Alex_Gaynorcfbolz: remind me in 12-24 hours.  I've got a paper to finish, and a flight home.  Then I have a week of relaxing, branch reviewing, and anything else I feel like10:26
cfbolzcool10:26
arigato:-)10:26
cfbolzAlex_Gaynor: it's very similar to list-strategies10:27
Alex_GaynorI assume so10:27
stakkars__ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Ping timeout: 245 seconds10:28
stakkarswell, _merging_ sometimes has strange effects.10:29
cfbolzarigato: the guy who started sets during the sprint, will he continue?10:29
stakkarsI lost something of win64 support.10:29
cfbolzs/sets/tuples10:29
stakkarswill check a few things into default which are just compiler-specific.10:30
arigatocfbolz: I can't tell10:31
cfbolzarigato: ok10:31
cfbolzarigato: very cool that you started generators in rpython :-)10:32
fijalI can pick up specialized tuples probably10:32
cfbolzfijal: otherwise lukas probably won't mind10:32
fijalok10:32
fijalI definitely wouldn't mind lukas doing that :)10:32
fijalhe seems to be an expert in specializing stuff10:32
arigatoI'm also ready to hand out generators in rpython to anyone who minds10:32
Alex_Gaynorfijal, cfbolz: I'm not super pleased with the approach in that branch, the tuple shapes should be dynamic IMO10:32
cfbolzAlex_Gaynor: why?10:32
fijalAlex_Gaynor: write your paper10:33
cfbolzfijal: he's struggling quite a bit with mapdicts, atm10:33
fijalok10:33
Alex_Gaynorcfbolz: because there's no reason a 3 item tuple of (bool, str, int) should randomly be slow because we didn't think of it10:33
stakkarsarigato: generators?10:33
Alex_Gaynorwe don't think of most things, let's be fast even then10:33
arigatoAlex_Gaynor: well the question is: do we want an unlimited number of specializations, or just a few10:33
Alex_Gaynorarigato: I hink unlimited, just like mapdicts10:34
arigatoduring the sprint I tried hard to convince him to go for very limited instead10:34
cfbolzthe tradeoff isn't that clear10:34
fijalarigato: now you have a chance to convince alex as well ;-)10:34
arigatoI wanted him to avoid supporting anything but 2-tuples, but he insisted on keeping the unroll_iterable stuff10:34
lucianit seems to be that (bool, str, int) is odd enough to not bother specialising for10:36
lucians/to be/to me/10:36
Alex_Gaynorit's odd to you, the guy who uses it in his code wants to know why you're telling him "don't do that"10:36
lucianpeople would notice having their int lists or coordinate tuples faster, though10:36
lucianAlex_Gaynor: do that, but don't expect magic, i guess10:36
stakkarswhat is the risk when supporting too much?10:36
stakkarstoo many specialisations/code bloat?10:37
arigatostakkars: yes10:37
lucianof course, i'm the one not writing PyPy code, so who am I to speak10:37
ltratt (~ltratt@ntratt.plus.com) joined #pypy.10:37
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds10:38
stakkars__ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.10:38
Mepho (~null@gentoo/contributor/mepho) joined #pypy.10:38
Mephohi, dous anyone know whether it's possible to compile 64bit pypy on linux (gentoo) using icc compiler with agressive flags? or does pypy makefile forces gcc?10:40
arigatolucian: please do speak :-)10:40
arigatoMepho: it should work, yes10:41
Alex_Gaynorarigato: probably requires --gcrootfinder=shadowstack though right?10:41
arigatotrue10:41
fijalMepho: also, it's very unlikely to give you significant speedups10:42
fijalbecause we compile a lot of assembler ourselves10:42
fijalbut there are cases where it might help10:42
fijallike longs10:42
Alex_Gaynorfijal: unless your code is bottlenecked by our gcc bug :)10:42
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.10:43
fijalyes10:43
stakkars__ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Read error: Connection reset by peer10:43
Mephoill do some timings with some of my scripts and if i'll do some blogpost in future if my procrastination will get any worse10:44
arigatopff10:46
fijalarigato: ???10:47
arigatoI've spent the morning at my house instead of going to uni to wait for a fedex package I don't even know where from, and result: fedex didn't show up10:47
fijalarigato: yes, expected10:47
arigatoheh10:48
lucianarigato: they all suck. what I don't get is why they don't just send you an email/sms when the van leaves for town10:48
arigatothey showed up the minute I complained about them :-)10:49
fijalarigato: you know it works that way at least on twitter?10:49
arigatolucian: well, the whole model is still archaic I suppose --- e.g. I had to really phone them, no email contact10:50
arigatofijal: how so? :-)10:50
fijalarigato: I still want to break in their servers and send all trucks to one spot10:50
lucianarigato: sure, but SMS is pretty archaic :)10:50
fijalif you complain loudly on twitter, customer service usually gets back to you10:50
arigatoah :-)10:51
fijal(happened to me)10:52
arigatolucian: I know it costs more per letter to send an SMS than any other communication method, including sending a Mars probe and listening to it10:52
arigatoso yes, I have actually never written a SMS10:53
Action: cfbolz , as usual, is deeply confused by some code in optimizeopt10:54
arigato:-/10:54
Action: arigato -> uni10:55
arigato (~arigo@1.106.113.82.net.de.o2.com) left irc: Quit: See you10:55
senyai (~senyai@62.173.135.205) joined #pypy.10:55
lucianarigato: heh. i wrote a few, if only because they're free on my plan10:58
kenaan12cfbolz default 1166bdc5c4e59c 15/pypy/jit/metainterp/optimizeopt/test/test_optimizeopt.py: there was really no optimizeopt test for circular virtuals??!11:11
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Ping timeout: 244 seconds11:13
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.11:13
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.11:13
mat^2 (~mathias@212.130.113.35) joined #pypy.11:15
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.11:18
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Ping timeout: 260 seconds11:26
arigato (~arigo@fwstups.cs.uni-duesseldorf.de) joined #pypy.11:28
arigatouh11:28
arigatofijal: you sent out the announcements before updating the web site11:28
arigatothat's kind of bad11:28
Topic changed on #pypy by arigato!arigo@fwstups.cs.uni-duesseldorf.de: PyPy, a fast and compliant python interpreter | IRC logs: http://www.tismer.com/pypy/irc-logs/pypy/latest.log.html | PyPy 1.7 released --- wait a sec11:29
fijaluh what?11:30
kenaan12ctismer win64_gborg 114a3f7bf69952 15/: Merge with default11:30
kenaan12ctismer default 114dcb9ca387a6 15/pypy/: 'merge'11:30
kenaan12ctismer default 11015d05dac523 15/: merge11:30
fijalI did update the website no?11:30
kenaan12ctismer win64-stage1 11ab9d4a11c14d 15/: merge11:30
kenaan12ctismer win64-stage1 110e90cd957a7d 15/pypy/jit/metainterp/optimizeopt/test/test_optimizeopt.py: merge11:30
arigatohttp://pypy.org is not updated11:30
arigatouh?11:30
arigatosorry11:30
arigatoyes it is11:30
arigatoobscure caching issue in my browser?11:30
Topic changed on #pypy by arigato!arigo@fwstups.cs.uni-duesseldorf.de: PyPy, a fast and compliant python interpreter | IRC logs: http://www.tismer.com/pypy/irc-logs/pypy/latest.log.html | PyPy 1.7 released11:30
fijaliyes11:30
fijaljust normal caching I fear11:31
arigatoright11:32
arigatosomehow I don't know how web caching works, and I assumed it would be based on a checksum of the page, but it's not I guess11:32
fijalno11:34
fijalI think it just caches for the period that server gives you11:34
fijalno questions asked11:34
fijalchecksums...11:34
arigatook :-)11:35
ronnyinstead of sending expires, it might make sense to use etag/If-Modified11:35
fijalronny: feel free to set it up somehow11:35
arigatomay I complain again about the fact that the website's "py3k/numpy donation" pages and the "Donate towards..." link are not related?11:35
fijalbut I suppose we don't change it often enough for it to matter11:35
fijalarigato: how they're not related?11:35
fijalthey're related as in "read proposal"11:36
arigatoe.g. if I go:11:36
arigatohome page -> click "donate to py3k" -> click "read proposal"11:36
fijalit says "donate to numpy"11:36
fijalno, that's a bug11:36
arigatoyes11:37
fijalI thought I fixed it11:37
fijalif you don't mind, I won't look there now, my internet sucks a lot11:37
arigatojust reporting it as a bug --- makes sense for me to add it to the bug tracker, maybe?11:38
fijalok11:38
lucianarigato: web caching is fun fun FUN! especially when you mix in cache.manifest ...11:39
CIA-6803arigo 07roundup * 10#931/web site bugs: 11:40
CIA-68[new] On the web site http://pypy.org:11:40
CIA-68home page -> click "donate to py3k" -> click "read proposal"11:40
CIA-68it ends up selecting "donate to numpy" a ... * 14https://bugs.pypy.org/issue93111:40
kenaan12ctismer win64_gborg 119d59a8501c00 15/: close old branch11:42
kenaan12ctismer win64 test 1103816b1a901c 15/: close old branch11:42
kenaan12bivab arm-backend-2 112bddb9109f82 15/pypy/jit/backend/arm/opassembler.py: typo11:42
kenaan12bivab arm-backend-2 111ff3890a1ce8 15/pypy/jit/backend/: add ARM to conftest and test_random11:42
kenaan12bivab arm-backend-2 112771f3c2a471 15/pypy/jit/backend/arm/: Updating the position of the regalloc to soon when merging operations with the following guard leads to some ...11:42
CIA-6803arigo 07roundup * 10#930/Recursion with cache slower on pypy than CPython 2.6: 11:45
CIA-68I agree that we have to look at it, but I just want to note that (for me)11:45
CIA-68running 500 iterations ends up being faster with pypy, while 100 iterat ... * 14https://bugs.pypy.org/issue93011:45
fijalarigato: #930 is a bit obscure11:45
fijalwhy-we-are-so-slow11:46
fijalkinda issue11:46
fijalwith --jit threshold=-1 we're the same speed as cpython11:46
fijalexcept we do a lot more optimizations11:46
fijalit seems that packing/unpacking tuple/objects might be an issue11:46
Mepho (~null@gentoo/contributor/mepho) left irc: Quit: "irssi, client of the future... waiting for future since '06"11:46
Action: fijal off11:47
arigatodo you mean "--jit off"?11:47
fijalno11:47
fijalI specifically mean leave only the function jit11:47
fijalso we don't end up tracing a lot11:48
arigatoI'm unsure what you're comparing against, given that with 100 iterations it's a bit slower than cpython, and with 500 iterations it's a bit faster11:48
cfbolzclearly we need specialized tuples :-)11:48
fijalarigato: with --jit threshold=-1 and normal pypy-c we're 2.6 vs 3.0s11:49
fijalso 0.4s is probably spent tracing and doing other stuff11:50
fijalthat's a bit too much IMO11:50
arigatofijal: doing 100 iterations or 500?11:50
fijal10011:50
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.11:51
lotrpy (~lotrpy@202.120.58.116) joined #pypy.11:57
stakkars__ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.11:58
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Read error: Connection reset by peer11:58
stakkars__ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Read error: Connection reset by peer12:01
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.12:01
fijal (~fijal@197.105.215.160) left irc: Read error: Connection reset by peer12:08
Vorpal (~AnMaster@unaffiliated/anmaster) joined #pypy.12:12
antocuni (~antocuni@h10-ipv4-80-68-184.mynet.it) joined #pypy.12:19
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds12:19
antocunihi12:20
antocunicongrats for the release :-)12:20
canta (~canta@77-20-123-240-dynip.superkabel.de) joined #pypy.12:23
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.12:24
senyaiHi. When will windows version be ready?12:27
tav (~tav@host-92-20-3-37.as13285.net) left irc: Ping timeout: 240 seconds12:30
ronan (~ronan@host86-142-249-87.range86-142.btcentralplus.com) left irc: Ping timeout: 240 seconds12:30
ronan (~ronan@host86-142-249-87.range86-142.btcentralplus.com) joined #pypy.12:30
cfbolzgaaaargh12:31
Action: cfbolz hates subtle annoying bugs12:31
tav (~tav@host-92-20-3-37.as13285.net) joined #pypy.12:32
nedbat (~nedbat@python/psf/nedbat) joined #pypy.12:32
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Read error: Connection reset by peer12:34
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) joined #pypy.12:35
cfbolzAlex_Gaynor: your optimization in the JIT that makes Consts when you force a virtual filled with Consts is nonsense12:37
cfbolzAlex_Gaynor: again, it does not deal with recursive virtuals at all12:38
cfbolzit took me a week to hunt down this12:38
arigatouh?12:39
arigatoI thought I already did that long ago12:39
arigato0db17e3bbb6412:41
cfbolzarigato: still in the process of writing a test12:44
cfbolzbut this code is definitely not dealing with recursive immutable virtuals12:44
cfbolzAlex_Gaynor: sorry, for some reason I thought it was you who wrote that code12:45
cfbolzarigato: AbstractVirtualStructValue._really_force can call itself via the force_box in _is_immutable_and_filled_with_constants12:46
cfbolzwhich means you emit the same new_with_vtable twice12:47
cfbolzarigato: I guess this cannot happen in the Python interpreter, but it can in Prolog12:48
tos932bit binaries haven12:49
tos9't been built for OSX since 1.4.1 correct?12:49
cfbolzarigato: see what I mean?12:49
Trundle (~andy@p578bfdcf.dip0.t-ipconnect.de) joined #pypy.12:52
Trundle (~andy@p578bfdcf.dip0.t-ipconnect.de) left irc: Changing host12:52
Trundle (~andy@python/site-packages/trundle) joined #pypy.12:52
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Quit: Leaving12:55
kenaan12cfbolz default 1124e8fd936c50 15/pypy/jit/metainterp/optimizeopt/test/test_optimizeopt.py: also a test where the circular virtual is forced12:59
kenaan12cfbolz default 11f6d1a45766ce 15/pypy/jit/metainterp/optimizeopt/test/: a failing test that shows a problem around forcing of virtual immutable recursive structs12:59
kenaan12cfbolz default 115225def4e322 15/: merge12:59
cfbolzarigato: ok, there is a failing test now13:00
cfbolzarigato: this is not the only problem, as pyrolog crashes yet differently13:00
kenaan12ctismer win64-stage1 11fbf211aee071 15/pypy/: _continuelet on win64 _works_ ! There was an old direct reference to windows in rwin32.py, changed it to hos...13:02
stakkars_ (~tismer@dslb-088-074-042-148.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds13:09
Rhy0lite (dje@nat/ibm/x-beaawflpnsgzazew) joined #pypy.13:10
merwok (c134d1e6@gateway/web/freenode/ip.193.52.209.230) joined #pypy.13:13
merwokCongrats on the release!13:13
arigatocfbolz: sorry, back13:16
cfbolzarigato: np13:17
arigatobaaah, ok13:17
cfbolz:-(13:17
cfbolzarigato: what do you think of this simple fix: http://paste.pocoo.org/show/510904/13:18
cfbolzthat means the whole thing works only for immutables of primitive values13:18
merwok_ (c134d1e6@gateway/web/freenode/ip.193.52.209.230) joined #pypy.13:19
cfbolzmerwok_: thanks!13:19
arigatotos9: yes, that's correct13:20
merwok (c134d1e6@gateway/web/freenode/ip.193.52.209.230) left irc: Ping timeout: 265 seconds13:20
tos9arigato: Thanks.13:20
Nick change: merwok_ -> merwij13:21
Nick change: merwij -> merwok13:21
arigatodo you have an idea for how to do better?13:23
arigatocfbolz: ?13:23
cfbolzarigato: better than the patch I posted?13:24
arigatoyes13:25
arigatoin theory we want to check if the graph of all the optvalues reachable from the current one contains only primitive consts and further immutable ones13:25
cfbolzarigato: yes, well13:25
cfbolzarigato: you would have to keep a dict of things you already checked13:25
arigatoyes13:25
cfbolzarigato: I still think you should never call force_box13:25
cfbolzto not mix up these two concerns13:26
arigatoI agree13:26
cfbolzarigato: feel like implementing it?13:26
arigatoI can try13:26
cfbolzI can review13:26
arigatook :-)13:26
arigatobut first, "rpython/test/test_rclass -k test_hash_preservation", which works on Python 2.6 but fails on 2.713:27
cfbolzpfff13:27
cfbolzarigato: I find it funny that the bug is in a part of virtuals that we didn't formalize in the paper :-)13:27
arigato"everything is either formalized or buggy", some would say13:28
JaredW (~jared@061093200075.ctinets.com) joined #pypy.13:28
cfbolzyes13:28
kkris (~kris@93-82-44-6.adsl.highway.telekom.at) left irc: Quit: Leaving.13:29
merwok:°)13:30
guibouDoes a "pypy1.7" is 3 times slower than 1.6 is considered as a major regression and should I open a bug repport ?13:34
cfbolz_ (~cfbolz@dslb-178-003-052-126.pools.arcor-ip.net) joined #pypy.13:35
stakkars_ (~tismer@dslb-088-072-229-176.pools.arcor-ip.net) joined #pypy.13:35
cfbolz (~cfbolz@dslb-178-003-061-166.pools.arcor-ip.net) left irc: Ping timeout: 258 seconds13:36
lucianwow, json is blazing fast. thanks fijal13:36
lucianoh, no there13:37
arigatoguibou: yes13:39
jonathan__ (~jonathan@bas1-montreal29-1242341904.dsl.bell.ca) joined #pypy.13:40
arigatostakkars_: you merged win64 to default?  it crashes now13:41
arigatoI'm not sure you meant to merge that much13:43
arigatoalso, your merge checkins are not reported as merge checkins at all,13:44
arigatowhich means (1) I don't know how you did it, and (2) it's impossible to track13:44
antocuniarigato: in case we need to revert stakkars's change, we might prefer to create a new head which starts before the merge instead of using hg backout13:44
antocuniit will preserve more history13:44
arigatoantocuni: or just cancel cfbolz's latest checkin, which is a default "merge heads"13:45
antocuniright13:45
antocuniwell, everything but hg backout is fine :-)13:46
stakkarsarigato: oups? not that I'm aware of it!13:46
CIA-6803peterjc 07roundup * 10#919/implement multi-dimension arrays in micronumpy: 13:46
CIA-68[chatting] Fixed typo in title.13:46
CIA-68Note PyPy 1.7 release notes suggest this is being worked on,13:46
CIA-68http://morepypy.blogspot.com/2011/11/pypy-17-wide ... * 14https://bugs.pypy.org/issue91913:46
arigatook, so let's revert cleanly13:46
stakkars_ (~tismer@dslb-088-072-229-176.pools.arcor-ip.net) left irc: Read error: Connection reset by peer13:46
stakkarsI was merging default into my branch, I thought13:46
Nick change: Gulopine -> Gulaway13:46
arigatoantocuni: how do you plan to do it?13:46
arigatowe'd like basically f6d1a45766ce to become the "default" head13:47
merwokThe blog link above is not working13:47
antocuniarigato: I didn't look at the history, but if there are not many changes, we could transplant all children of the "wrong" head to the "good" one 13:47
antocuniand the merge the wrong head to dead-branches13:47
antocunis/the/then13:47
antocuni(sorry, I'm at a customer's place and cannot do it by myself right now)13:48
arigatofine :-)13:48
arigatothe history looks like this:13:48
stakkars_ (~tismer@dslb-088-072-229-176.pools.arcor-ip.net) joined #pypy.13:49
stakkarsarch, maybe I messed things up at 11:47 after a wrong update :-(13:49
arigatohttp://paste.pocoo.org/show/510913/13:49
antocunithen we can probably just merge 5225def4e322 to dead-branches13:49
arigatoand make a dummy checkin from f6d1a45766ce13:50
antocuniand do a dummy cheking as a child of f6d1a45766ce13:50
arigatook13:50
antocuniyes :-)13:50
merwok (c134d1e6@gateway/web/freenode/ip.193.52.209.230) left irc: Quit: Page closed13:51
stakkarsoh, very sorry about that, absolutely not my intent.13:52
stakkarsarigato: what should I do?13:53
Action: arigato fixes13:54
stakkarsI forgot that I switched computers and did not update correctly :-(13:55
arigatoantocuni: hum13:55
arigatoantocuni: do I need to "hg ci --close-branch" the wrong head on default?13:55
arigatoI guess so13:55
lucianarigato: yeah, --close-branch is "the right thing"13:56
antocuniuhm, I don't think it's the "right thing"13:56
lucianantocuni: how so?13:56
antocuniI think that close-branch just label the branch "default" as closed13:56
lucianwell, if it's a dummy you can strip13:56
lucianantocuni: ah, default. stupid me13:57
antocunis/label/labels13:57
lucianbut you can close default too, in fact i did it a few times13:57
arigatobasically even after merging the wrong branch to "closed-branches", I still have it in "hg heads"13:57
lucianand you can fork from a parent and keep going with default13:57
arigatoso that's not right either13:57
lucianthat's odd. try strip13:57
antocunilucian: we cannot strip, it's already on bitbucket13:58
lucianif it's still local, that is13:58
lucianif no one's pulled yet, you can still strip :)13:58
antocuniarigato: that's weird13:58
arigatoand "hg up" still complains that there are two heads13:58
antocuniarigato: do you still see it as a head if you do hg heads --topo?13:58
cfbolz_lucian: it was pushed13:58
arigatono13:59
Nick change: cfbolz_ -> cfbolz13:59
arigatobut it doesn't help "hg up" 13:59
antocuniarigato: ah, maybe you can also try to do a "dummy merge" of the wrong head into the good one13:59
cfbolzantocuni: that sounds weird13:59
luciancfbolz: i meant using bitbucket's strip, if no one has pulled the dummy yet13:59
antocunihttp://mercurial.selenic.com/wiki/TipsAndTricks13:59
antocunipoint 2213:59
CIA-6803guibou 07roundup * 10#932/Python 1.7 slower on home maid path tracing than 1.6: 14:00
CIA-68[new] I just tested pypy 1.7 on very poor implementation of path tracing I wrote for14:00
CIA-68fun. (it is mainly a copy/paste/rewrite version of the C sma ... * 14https://bugs.pypy.org/issue93214:00
arigatoit does sound weird, but indeed, it looks like the only way to "delete" some checkins14:01
iori (~iori@110-133-45-54.rev.home.ne.jp) left irc: Remote host closed the connection14:01
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.14:01
arigatolucian: that's a bit too late, I'd rather not have random repos out there suddenly break14:02
antocuniarigato: I still don't understand why merging it to closed-branches doesn't work, but ok14:03
lucianarigato: sure. but they wouldn't break anyway. the only risk is that one of them would push back to bitbucket and undo the strip14:03
antocuniarigato: also, stripping from the bitbucket repo would not break anything. The only risk is that if someone already pulled and the pushes again, the checkin will "resurrect"14:04
stakkarshmm, I really cannot determine exactly what happened.14:04
arigatowell if someone already pulled, he definitely gets two heads and is in a confused state14:05
arigatohttp://mercurial.selenic.com/wiki/TipsAndTricks point 22 doesn't seem to work14:06
Action: arigato tries to do it manually14:07
derdon (~derdon@p5DE8A92B.dip.t-dialin.net) joined #pypy.14:10
arigatomaybe14:12
kenaan12arigo default 115b572a486ead 15/pypy/rpython/test/test_rclass.py: No-op clarification.14:13
kenaan12arigo default 119e1ad8713760 15/: Attempt to kill 5225def4e322, which contains the broken merges 015d05dac523 and 4dcb9ca387a6.14:13
arigatook, that worked14:13
arigatostakkars: I've no clue how you did the merge, but in any case, it was broken: your two merges don't have two ancestors but one14:14
stakkars_ (~tismer@dslb-088-072-229-176.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds14:15
Nick change: Gulaway -> Gulopine14:15
CIA-6803peterjc 07roundup * 10#933/AttributeError: 'pyexpat.XMLParserType' object has no attribute 'StartElementHandler': 14:15
CIA-68[new] I didn't report this earlier as our code was blocked by Issue 914, which is now14:15
CIA-68fixed in PyPy 1.7.14:15
CIA-68Expected behaviour to match C Python: ... * 14https://bugs.pypy.org/issue93314:15
arigatoeveryone: done, please update to the current head before checking in stuff14:15
stakkarsI'm very sorry, did not recognize it.14:16
arigato(or check in with an older head and "merge heads", but just please not with the broken 5225def4e322)14:16
antocuniarigato: how did you fix it at the end?14:17
arigatoa dummy merge14:17
antocuniah, good to know that it works14:18
arigatoyes, it seems to.  I was confused for a while14:18
arigatobut the instructions of http://mercurial.selenic.com/wiki/TipsAndTricks are correct14:19
Action: arigato tries to pop items off his stack14:20
fprimex (~fprimex@fl-67-233-144-228.dhcp.embarqhsd.net) joined #pypy.14:23
rokujyouhitoma (~rokujyouh@KD121107219133.ppp-bb.dion.ne.jp) joined #pypy.14:23
rokujyouhitoma (~rokujyouh@KD121107219133.ppp-bb.dion.ne.jp) left irc: Read error: Connection reset by peer14:23
rokujyouhitoma (~rokujyouh@240f:11:f952:1:855c:c2ed:7325:2783) joined #pypy.14:24
stakkarsbtw, which buildbot showed the crashes that I caused?14:25
arigatostakkars: none, we would have seen them tomorrow14:25
arigatostakkars: but on heads, we couldn't run "translate.py" any more14:26
arigato(it was found out by bivab first, more precisely)14:26
stakkarswhich platform? I need to fix thet14:27
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.14:28
whitelynx (~whitelynx@li117-47.members.linode.com) joined #pypy.14:32
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.14:39
ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) left irc: Quit: Leaving...14:40
stakkars_ (~tismer@dslb-088-074-044-208.pools.arcor-ip.net) joined #pypy.14:41
gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) joined #pypy.14:41
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.14:41
voidspace (~voidspace@python/psf/voidspace) left irc: Quit: Uhm... gotta go14:45
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Quit: Leaving14:47
rokujyouhitoma (~rokujyouh@240f:11:f952:1:855c:c2ed:7325:2783) left irc: Remote host closed the connection14:47
Nick change: Gulopine -> Gulaway14:48
stakkars_ (~tismer@dslb-088-074-044-208.pools.arcor-ip.net) left irc: Ping timeout: 245 seconds14:49
PiotrSikora (~none@nginx/adept/piotrsikora) left irc: Excess Flood14:49
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.14:49
sven_hager (~sven@fwstups.cs.uni-duesseldorf.de) joined #pypy.14:49
PiotrSikora (~none@nginx/adept/piotrsikora) joined #pypy.14:50
arigatostakkars: you don't need to fix anything14:53
stakkarsI'm so sorry that I caused this mess.14:54
arigatono problem, it occurs14:54
stakkarsthere are hundreds of files flagged with "M U".14:58
stakkarsis that important?14:58
kenaan12arigo default 11ab5c477f2f34 15/pypy/rpython/test/test_rclass.py: Disable a half-broken test that fails on Python 2.7 (see comments).14:59
arigatoI don't know TortoiseHG14:59
arigatobut you should be working from fbf211aee071 with no big number of local modified files15:00
stakkarsah, hundreds unresolved conflicts :-(15:01
jmr_mp (~jmr@14-200-200-212.static.tpgi.com.au) joined #pypy.15:01
arigatodoes _continuation work on win32 on "default"?15:01
stakkarscan't check right now, machine busy...15:02
stakkarsidt did, Saturday15:03
jmr_mpis gcc-4.0 still the only apple compiler that builds pypy correctly?15:04
whitelynx (~whitelynx@li117-47.members.linode.com) left irc: Remote host closed the connection15:09
whitelynx (~whitelynx@63.241.75.144) joined #pypy.15:10
voidspace (~voidspace@python/psf/voidspace) joined #pypy.15:11
stakkars_ (~tismer@dslb-088-074-038-199.pools.arcor-ip.net) joined #pypy.15:16
stakkarstesting win32 default _continuation15:18
arigatothanks :-)15:19
stakkars__ (~tismer@dslb-088-074-038-199.pools.arcor-ip.net) joined #pypy.15:19
arigatojmr_mp: no, we gave up on using the special feature on OS/X, so now any ANSI C compiler should be ok15:20
stakkarsyes, works fine!15:20
arigatostakkars: :-)15:20
jmr_mparigato: ok, cool15:20
stakkars_ (~tismer@dslb-088-074-038-199.pools.arcor-ip.net) left irc: Ping timeout: 244 seconds15:21
jmr_mparigato: it wasn't clear because the comments in darwin.py still say otherwise15:21
stakkarsarigato: ah, but crashes in win64-stage1 !!!15:21
stakkars__ (~tismer@dslb-088-074-038-199.pools.arcor-ip.net) left irc: Read error: Connection reset by peer15:28
stakkars_ (~tismer@dslb-088-074-038-199.pools.arcor-ip.net) joined #pypy.15:29
arigatostakkars_: fails for me:15:30
arigatoNMAKE : fatal error U1073: don't know how to make 'switch_x86_msvc.a.obj'15:30
arigatorunning pypy/rlib/test/test_rstacklet.py15:30
kenaan12l.diekmann type-specialized-instances 11510e6bc4ef58 15/pypy/objspace/std/mapdict.py: None should be erased by PlainAttribute15:31
kenaan12l.diekmann type-specialized-instances 1190348184732d 15/pypy/objspace/std/test/test_mapdict.py: fixed some tests where storage items are erased. there is no general erasing pair anymore s...15:31
gabriel_ (~gabriel@bootp-217.zbh.uni-hamburg.de) joined #pypy.15:31
gabriel_hi!15:32
arigatobut indeed, the tests in pypy/module/_continuation work15:32
arigatogabriel_: hi15:32
td123 (~tom@archlinux/trusteduser/td123) joined #pypy.15:32
gabriel_i just checked out the latest version of pypy from bitbucket and started the translation15:32
gabriel_it takes huuge amounts of memory15:33
td123I was wondering, if pypy is compatible with the C interface for python, why do you have to rewrite numpy?15:33
gabriel_i didn't have the problem with older versions15:33
arigatogabriel_: http://pypy.org/download.html#building-from-source : does it take more than explained in this paragraph?15:34
stakkars_ (~tismer@dslb-088-074-038-199.pools.arcor-ip.net) left irc: Ping timeout: 244 seconds15:34
arigatotd123: numpy is using a lot of tricks that go beyond the limited compatibility we provide,15:35
gabriel_i've got 8 gigs of ram and it's swapping....15:35
arigatotd123: and fixing it to support it is pointless because you would get a very slow result15:35
gabriel_output from top:  8580 gabriel   20   0 9241m 6.4g 1064 D    9 83.7  24:43.87 pypy15:37
arigatogabriel_: that looks like far too much15:37
gabriel_that's what i thought15:37
gabriel_it's not the first time, that i translate pypy myself15:38
arigatogabriel_: can you tell us more precisely which platform, and which revision?15:38
gabriel_never had that problem before15:38
gabriel_linux x6415:38
gabriel_this version of pypy is what i'm using for translation:15:39
gabriel_Python 2.7.1 (d5cc360c2060, Nov 08 2011, 14:29:28)15:39
gabriel_[PyPy 1.7.1-dev0 with GCC 4.6.1] on linux215:39
arigatogabriel_: ah, there is a known issue15:39
arigatoyes, it was fixed on the 11th of November15:40
arigatoif that's the problem, it's caught in an infinite loop consuming more and more RAM15:40
arigatocan you try again by using the official pypy-1.7-linux64 release for translation?15:41
gabriel_ok, i will15:41
arigatothank you15:41
gabriel_thanks for the help15:41
gabriel_oh, great! the current official build works out of the box on ubuntu 11.10. with 1.6 I had library-issues15:46
td123 (tom@archlinux/trusteduser/td123) left #pypy ("WeeChat 0.3.6").15:47
stakkars_ (~tismer@dslb-088-072-219-152.pools.arcor-ip.net) joined #pypy.15:53
Unhelpfulhuh, gcc native transactional memory in the pipeline. but i'm assuming you don't want to be tied to a compiler implementation?15:54
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) left irc: Remote host closed the connection15:54
mattip (~chatzilla@bzq-109-65-106-250.red.bezeqint.net) joined #pypy.15:54
kenaan12bivab arm-backend-2 115f60920a9b51 15/pypy/jit/backend/arm/regalloc.py: (arigo, bivab) remove a reference to boxes15:55
kenaan12bivab arm-backend-2 1132cc8083b6f1 15/: merge default15:55
kenaan12bivab arm-backend-2 112c3c610bb2de 15/pypy/jit/backend/arm/runner.py: (arigo, bivab) remove gcremovetypeptr for the arm backend for now15:55
kenaan12bivab arm-backend-2 11b9c81b1b4af5 15/: merge default15:55
kenaan12bivab arm-backend-2 11406a32f88290 15/pypy/jit/backend/arm/regalloc.py: remove another reference to boxes15:55
kenaan12bivab arm-backend-2 11955f8735ff0f 15/: merge default15:55
kenaan12bivab arm-backend-2 11b14409bab3b1 15/pypy/jit/backend/arm/regalloc.py: more fixes15:55
kenaan12bivab arm-backend-2 1181bac4da2b46 15/: merge heads15:55
kenaan12bivab arm-backend-2 112ec66beccfff 15/pypy/rpython/test/test_rclass.py: merge default (again)15:55
asmeurer (~asmeurer@dhcp-baca-230.resnet.nmt.edu) joined #pypy.15:59
stakkars__ (~tismer@dslb-088-072-219-152.pools.arcor-ip.net) joined #pypy.15:59
stakkars_ (~tismer@dslb-088-072-219-152.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds15:59
jterrace (~jterrace@aegis.CS.Princeton.EDU) joined #pypy.16:00
jterraceis the PPA (https://launchpad.net/~pypy/+archive/ppa) going to get updated with 1.7?16:01
xorAxAxjterrace: ask https://launchpad.net/~pypy16:01
rguillebertAKA getxsick apparently16:03
Action: arigato wonders how much fixing we'd need if we strengthen the rpython rule about Unsigned:16:05
arigatoat annotation-time, union(Signed, Unsigned) crashes, unless you do union(constant-Signed, Unsigned)16:05
arigato(and then even the constant signed must not be negative)16:06
stakkars__ (~tismer@dslb-088-072-219-152.pools.arcor-ip.net) left irc: Ping timeout: 260 seconds16:07
nedbat (~nedbat@python/psf/nedbat) left irc: Quit: Leaving16:09
kenaan12l.diekmann type-specialized-instances 111ff52ab49127 15/pypy/objspace/std/mapdict.py: fallback to PlainAttribute on AttributeError. this fix is needed for modules, since they do...16:17
_class_ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) joined #pypy.16:18
jterrace (jterrace@aegis.CS.Princeton.EDU) left #pypy.16:20
__class__ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) left irc: Ping timeout: 260 seconds16:22
Nick change: _class_ -> __class__16:23
DasIch (~dasich@p4FFDF354.dip.t-dialin.net) left irc: Remote host closed the connection16:24
bivaba char is also considered as signed?16:30
arigatoyes16:32
arigatowell no16:32
stakkarsarigato: fixed, works on (default, win32), (win64-stage1, win32), (win64-stage1, win64)16:32
arigatoa char is its own type, and you have to use ord(), which returns a signed16:32
stakkarsarigato: sorry, this message is wrong again after the last merge :-(16:32
stakkarshow do I get back changes which were overwritten in the merge?16:33
stakkarsa quite substantial amount of changes16:34
jnoller (~jnoller@vpn.nasuni.com) joined #pypy.16:39
roelbruggink (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection16:42
arigatostakkars: basically, you want to get rid of the broken merge attempt you did, and do a clean merge?16:42
stakkarsarigato: yes, I did this, but then some modules were overwritten, the version that was checked it seemed to win over my now older changes.16:44
arigatofwiw what you checked in is not a merge at all16:44
arigatoa merge is a changeset with two ancestors16:44
arigatoI have no clue how you made the changesets which you checked in, but they have only one ancestor16:45
stakkarswhich checking, the mess you mean?16:45
arigato015d05dac523 and 4dcb9ca387a616:45
JaredW (~jared@061093200075.ctinets.com) left irc: Read error: Operation timed out16:46
stakkarswell, I reverted and then merged with the new default tip16:46
arigato015d05dac523 and 4dcb9ca387a6 are not merges at all16:46
stakkarsand then many patches were gone16:46
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.16:46
ronnyarigato: i just stumbled uppon http://gcc.gnu.org/wiki/TransactionalMemory (starting with gcc 4.7)16:47
stakkars_ (~tismer@dslb-088-074-052-067.pools.arcor-ip.net) joined #pypy.16:49
arigatoronny: interesting16:49
gabriel_arigato: translation worked like a charm using the official release16:49
arigato:-)16:49
sven_hager (~sven@fwstups.cs.uni-duesseldorf.de) left irc: Quit: Leaving16:50
stakkars_ (~tismer@dslb-088-074-052-067.pools.arcor-ip.net) left irc: Ping timeout: 244 seconds16:53
bivab (~david@fwstups.cs.uni-duesseldorf.de) left irc: Quit: bivab16:54
arigato (~arigo@fwstups.cs.uni-duesseldorf.de) left irc: Quit: See you16:54
ojii (~ojii@40-34.60-188.cust.bluewin.ch) left irc: Quit: Leaving16:55
stakkars_ (~tismer@dslb-088-074-052-067.pools.arcor-ip.net) joined #pypy.16:57
stakkars_ (~tismer@dslb-088-074-052-067.pools.arcor-ip.net) left irc: Ping timeout: 260 seconds17:03
G2P (~G2P@fw-asn1.ornis.com) left irc: Quit: Leaving.17:03
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Ping timeout: 244 seconds17:03
antocuni (~antocuni@h10-ipv4-80-68-184.mynet.it) left irc: Ping timeout: 252 seconds17:06
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy.17:08
dmalcolm (david@nat/redhat/x-ljyssrjjysovdhau) joined #pypy.17:14
Alex_Gaynorcfbolz: Can you update the version number in docs/conf.py?17:14
stakkars_ (~tismer@dslb-088-074-047-144.pools.arcor-ip.net) joined #pypy.17:16
chronitis (~chronitis@88-104-120-223.dynamic.dsl.as9105.com) joined #pypy.17:17
mattip (chatzilla@bzq-109-65-106-250.red.bezeqint.net) left #pypy.17:19
JaRoel|4d (~jaroel|4d@2001:980:7133:1:3990:eeda:7dab:978d) joined #pypy.17:19
fox__ (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.17:32
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer17:32
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.17:38
fox__ (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer17:38
hruske (~Gasper@89.143.223.18) joined #pypy.17:39
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer17:39
hruske (~Gasper@89.143.223.18) left irc: Read error: Connection reset by peer17:39
stakkars__ (~tismer@dslb-088-074-047-144.pools.arcor-ip.net) joined #pypy.17:41
itazzzz (~ita@kde/developer/tnagy) joined #pypy.17:41
Nick change: itazzzz -> ita17:42
stakkars_ (~tismer@dslb-088-074-047-144.pools.arcor-ip.net) left irc: Ping timeout: 245 seconds17:42
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.17:43
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.17:44
kenaan12ctismer win64-stage1 11b5a4bc7caf97 15/: more unification of int and long17:45
kenaan12ctismer win64-stage1 1144db5b5f850f 15/: Merge with default17:45
kenaan12ctismer win64-stage1 110c8590e158a9 15/pypy/: cleanup of merge conflicts17:45
kenaan12ctismer win64-stage1 115030492cfa10 15/pypy/rpython/lltypesystem/rffi.py: merge heads17:45
kenaan12ctismer win64-stage1 11ba80ac53c5fb 15/: bad bad17:45
kenaan12ctismer win64-stage1 11413ea7aa3594 15/: reverted merge with default17:46
itapypy 1.7 hangs with subprocess (pypy 1.6 does not)17:47
jnoller (~jnoller@vpn.nasuni.com) left irc: 17:50
santagada (~leonardo@177.18.66.26) joined #pypy.17:51
senyai (~senyai@62.173.135.205) left irc: Quit: senyai17:54
amaury_ (~amaury_@74.125.57.34) joined #pypy.17:55
stakkars__ (~tismer@dslb-088-074-047-144.pools.arcor-ip.net) left irc: Read error: Connection timed out17:57
CIA-6803ita 07roundup * 10#934/subprocess hangs in 1.7 (but not in 1.6 or 1.5): 17:58
CIA-68[new] The problem is new in 1.7, after launching processes pypy can just hang. Try the17:58
CIA-68testcase in attachment17:58
CIA-68It is probably unrelated but it ... * 14https://bugs.pypy.org/issue93417:58
stakkars_ (~tismer@dslb-088-074-047-144.pools.arcor-ip.net) joined #pypy.17:59
itathat's my report ^^ :-) ^^17:59
berkerpeksag (~berkerpek@46.197.8.139) joined #pypy.18:00
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) joined #pypy.18:00
arigato (~arigo@82.113.106.35) joined #pypy.18:02
amaury_ (~amaury_@74.125.57.34) left irc: Ping timeout: 240 seconds18:04
stakkars_ (~tismer@dslb-088-074-047-144.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds18:05
chronitisamaury_: http://pastebin.com/Bdv1Yd3n < convert "string" to "bytes" in py3k18:11
chronitisah, too late18:11
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.18:16
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection18:16
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer18:19
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.18:22
arigatocfbolz: I'm working on the recursive immutable problem18:24
kenaan12ctismer win64-stage1 118cd49e98e5a5 15/pypy/translator/platform/windows.py: relevant_environ updated for windows18:29
kenaan12ctismer default 11d4d7276882d4 15/pypy/translator/platform/windows.py: relevant_environ updated for windows18:29
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds18:29
kenaan12ctismer win64-stage1 11d3cf57e55482 15/pypy/translator/platform/windows.py: merged with default18:33
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer18:34
jmr_mparigato: seems gcc-4.2 still fails: http://pastebin.com/y1Q1XiHG18:34
[mat^2] (~mathias@212.130.113.35) joined #pypy.18:34
mat^2 (~mathias@212.130.113.35) left irc: Ping timeout: 252 seconds18:35
arigatojmr_mp: this doesn't look like a gcc version issue18:38
arigatoit's a typing error in micronumpy18:38
jmr_mparigato: nonetheless it builds ok if I use gcc-4.018:38
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.18:38
arigatojmr_mp: that's rather strange18:38
stakkars_ (~tismer@dslb-088-072-199-141.pools.arcor-ip.net) joined #pypy.18:38
arigatomaybe it is a random order-dependent failure?18:38
jmr_mpcould be, I don't really feel like building several more times though :)18:39
arigatoyes, I understand18:39
arigatoah, btw, I don't know which version of gcc our mac buildslave uses18:40
arigatoI guess it's just the one out of the box18:40
jmr_mpdefault in darwin.py is gcc-4.018:40
arigatoah, right, ok18:41
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.18:41
arigatowell we should make sure that the paste above ends up being seen by the micronumpy guys18:41
arigato(!= me :-)18:41
stakkars_ (~tismer@dslb-088-072-199-141.pools.arcor-ip.net) left irc: Read error: Connection reset by peer18:41
stakkars_ (~tismer@dslb-088-072-199-141.pools.arcor-ip.net) joined #pypy.18:42
arigatoplease add it as a bug to bugs.pypy.org18:43
jmr_mparigato: sure18:43
arigatothank you :-)18:43
kenaan12arigo default 117744697c53dd 15/pypy/jit/metainterp/optimizeopt/test/test_optimizeopt.py: Improve the tests, testing a fully recursive solution.18:43
kenaan12arigo default 1126947e25881a 15/pypy/jit/metainterp/optimizeopt/: Write the recursive possibly-looping case of this logic.18:43
kenaan12arigo default 119106a03c8a60 15/pypy/translator/platform/windows.py: merge heads18:43
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.18:44
santagada (~leonardo@177.18.66.26) left irc: Ping timeout: 244 seconds18:44
ltratt (~ltratt@ntratt.plus.com) left irc: Read error: Connection reset by peer18:44
ltratt (~ltratt@ntratt.plus.com) joined #pypy.18:45
ltratt (~ltratt@ntratt.plus.com) left irc: Read error: Connection reset by peer18:46
stakkars_ (~tismer@dslb-088-072-199-141.pools.arcor-ip.net) left irc: Read error: Connection reset by peer18:46
stakkars_ (~tismer@dslb-088-072-199-141.pools.arcor-ip.net) joined #pypy.18:47
CIA-6803arigo 07roundup * 10#933/AttributeError: 'pyexpat.XMLParserType' object has no attribute 'StartElementHandler': 18:52
CIA-68[chatting] Ah, it seems to be that W_XMLParserType has a __setattr__ method but no18:52
CIA-68__getattr__, which would mean that you can set attributes like ... * 14https://bugs.pypy.org/issue93318:52
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds18:54
ltratt (~ltratt@ntratt.plus.com) joined #pypy.18:56
iori (~iori@110-133-45-54.rev.home.ne.jp) joined #pypy.18:56
stakkars_ (~tismer@dslb-088-072-199-141.pools.arcor-ip.net) left irc: Ping timeout: 244 seconds18:57
ericflo (~ericflo@75.103.8.110) joined #pypy.18:57
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.19:03
CIA-6803jmr 07roundup * 10#935/Possible typing error in micronumpy: 19:05
CIA-68[new] Platform: OS X 10.6.8 x86_64, Xcode 3.2.619:05
CIA-68See attached traceback that occurred during translation. I was going to chalk19:05
CIA-68this up to gcc- ... * 14https://bugs.pypy.org/issue93519:05
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.19:06
kenaan12arigo default 11ddbc82ef4d8f 15/pypy/module/pyexpat/: Test and fix: instead of using a custom __setattr__ handler (and no __getattr__, creating confusion), create a bunc...19:13
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.19:16
gabriel_ (~gabriel@bootp-217.zbh.uni-hamburg.de) left irc: Ping timeout: 248 seconds19:17
chronitisamaury_: http://pastebin.com/Bdv1Yd3n < convert "string" to "bytes" in py3k (with some comments at the top)19:18
fenrrir (~fenrrir@189.70.228.110) joined #pypy.19:22
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.19:23
CIA-6803arigo 07roundup * 10#933/AttributeError: 'pyexpat.XMLParserType' object has no attribute 'StartElementHandler': [resolved] Fixed in ddbc82ef4d8f. * 14https://bugs.pypy.org/issue93319:24
tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy.19:29
canta (~canta@77-20-123-240-dynip.superkabel.de) left irc: Quit: Leaving.19:30
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer19:34
kkris (~kris@80-123-45-98.adsl.highway.telekom.at) joined #pypy.19:34
fenrrir (~fenrrir@189.70.228.110) left irc: Quit: Saindo19:37
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.19:38
fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy.19:40
cfbolz (~cfbolz@dslb-178-003-052-126.pools.arcor-ip.net) left irc: Quit: Leaving19:42
fox___ (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.19:52
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer19:52
Trundle (~andy@python/site-packages/trundle) left irc: Quit: &19:53
berkerpeksag_ (~berkerpek@46.197.8.139) joined #pypy.20:11
berkerpeksag_ (~berkerpek@46.197.8.139) left irc: Client Quit20:11
fox___ (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer20:17
Rhy0lite (dje@nat/ibm/x-beaawflpnsgzazew) left irc: Quit: Leaving20:20
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Read error: Connection reset by peer20:21
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.20:21
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.20:22
fijal (~fijal@197.104.111.99) joined #pypy.20:24
nickik (~nick@217.150.239.67) joined #pypy.20:24
fijalseems pypy is mature enough most comments on the blog are nonsense/insults :(20:28
timonatorgreat!20:28
itafijal: :P20:29
fijaltimonator: I'm sorry your changes didn't make it20:29
timonatorno problem, it's mostly my fault for being a lazy bum20:29
amaury_fijal: :-( indeed20:29
timonatorFWIW, i'm not too sure why you decided to call it "numpypy" instead of "numpy"20:30
fijaltimonator: because it's not ready20:30
fijaland people complained20:30
Alex_Gaynorbecause numpy people asked.20:30
timonatorwas it to signal to the users/developers, that the compatibility is ... yeah, ok20:30
fijal (~fijal@197.104.111.99) left irc: Read error: Connection reset by peer20:31
timonatorhm, i believed there were already nightly windows builds. is the lack of a windows build due to the calling convention mixup still?20:32
fijal (~fijal@41.54.161.132) joined #pypy.20:34
timonatoranyway, don't let the "politics" get to you. you are all doing fantastic jobs and pypy is an amazing project20:34
fijaltimonator: it's hard to live without politics20:35
timonatorthat was referring mostly to the entitled people who say "well, you should have waited with the release until you have windows binaries!"20:35
fijalyeah well20:36
fijalI was given a good quote today20:36
fijalthose people don't contribute anything to the world20:36
fijalso why bother20:36
kenaan12mattip numpy-monkeyaround 11a5ced36cd80a 15/pypy/module/micronumpy/: add test and implementation for transpose, add test for set_shape20:37
fijalok, I'll look into issues tomorrow20:38
timonatorthat is true. better to concentrate on being awesome20:39
fijal (~fijal@41.54.161.132) left irc: Read error: Connection reset by peer20:39
fijal (~fijal@197.111.212.164) joined #pypy.20:40
fijalgrumble20:41
fijaltimonator: windows build was underway20:42
fijalbut probably stopped because some tests crashed and had dialog boxes with "ok"20:42
fijalto be clicked20:42
timonatorgaah, that problem still :(20:42
fijalI think it's a problem family20:43
timonatori thought someone found some kind of hack which would disable all kinds of boxes through some registry entry or something similar?20:43
fijalI don't think there is a hack that disables all20:43
fijalbut you can disable some20:43
fijalI don't know, I don't follow the windows development, I don't care20:43
arigatothere is a different hack20:44
arigatostakkars has a program that actively notices such dialog boxes and click Ok for you20:44
fijalarigato: did you install the thing that disables dr watson btw?20:44
arigatofijal: that was long ago, I happily forgot all details20:44
fijalok20:45
timonatorall right20:45
fijalanyway, nto that I care20:46
fijalarigato: cool, few issues reported , but nothing major20:46
arigatoyes20:46
fijalgood I suppose20:46
arigatothank you for your job as release manager20:46
fijalnp20:47
arigatofwiw, I still agree with the comment from "D" on the blog, missing the point of "numpypy"20:47
arigatovs "numpy"20:47
fijalarigato: http://gcc.gnu.org/wiki/TransactionalMemory20:48
fijalhave you seen this?20:48
arigatoit has been posted earlier today20:49
arigatobut I didn't look in detail so far20:49
Alex_Gaynorfijal: it looks like jeremy siek worked on the BOOST version20:49
fijalAlex_Gaynor: ?20:50
Alex_Gaynorfijal: of the STM stuff20:50
fijalreally?20:50
fijalcool20:50
arigatofijal: re numpypy: we'll *probably* renaming it back once it's finished?20:52
fijalyes?20:52
arigatoonly "probably"?20:52
fijalI don't want anyone to quote me on this20:53
Alex_Gaynorthat's the future, who knows!20:53
fijalas far as I know now, we're renaming it20:53
arigatoI say that if, when we think we are finished, numpy people still object, then it's their problem, not ours20:53
fijalyes20:53
timonatormaking the interpreter faster is quite low on the priority list, is that correct?20:53
fijaltimonator: it's not low, it's just a bit unclear how20:53
fijalI mean if you provide a way to make interpreter faster, I'm all ears20:54
fijalother than "just hack at like million places"20:54
timonatorthat's what cpython does, correct?20:54
fijalarigato: I agree20:54
Alex_Gaynorfijal: we're a bit fond of "faster, for free, everywhere" :)20:54
fijalarigato: I don't want the discussion to start now20:54
arigatotimonator: yes :-)20:54
timonatorright. i'm sorry, but i don't have any ideas :\20:54
arigatofijal: ok :-)20:54
fijaltimonator:  :)20:55
Ademan (~dan@54.sub-97-182-116.myvzw.com) joined #pypy.20:55
Ademancongrats on 1.720:55
fijalAdeman: thanks20:57
tilgovi (~randall@69.111.59.168) joined #pypy.21:01
tilgovi (~randall@69.111.59.168) left irc: Changing host21:01
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.21:01
fox__ (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.21:03
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer21:03
timonatorwhat's the status of memory usage benchmarks? are benchmarks missing or is benchmark infrastructure missing?21:05
fijalboth21:07
timonatori've started work on a benchmark "framework", that forks itself, executes a python callable and measures the maxrss of the child - does that sound like something i could flesh out for you to use?21:09
timonatorit needs a lot more work before it's actually useful. it doesn't store the results of benchmark runs yet, for instance21:09
_class_ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) joined #pypy.21:10
_class_0 (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) joined #pypy.21:10
__class__ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) left irc: Ping timeout: 240 seconds21:11
kkris (~kris@80-123-45-98.adsl.highway.telekom.at) left irc: Quit: Leaving.21:12
fijalmaxrss might not be an answer21:13
fijalmight as well be, a bit depends21:13
timonatorright, but i haven't found anything more useful that's still easy to get21:13
fijaltimonator: storing can be done as simply as json.dumps(results)21:14
timonatorof course. in fact, the child already sends its relevant data to the parent as a json dumped object21:14
_class_ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) left irc: Ping timeout: 244 seconds21:14
fijalarmin had some code that did some analysis21:14
fijalwe can also upload it to a codespeed instance21:14
timonatorbut there's more to it than just storing it into one big file. there should be some kind of organisation and analysis21:14
fijalwell, ok :)21:14
timonatordoes codespeed have a way to display memory usage yet?21:15
fijalcan't codespeed do that?21:15
fijalit's all graphs21:15
fijalit doesn'21:15
fijalt care21:15
timonatorheh. i know next to nothing about codespeed, so one of you would have to answer that21:15
timonatorah, so the file would just say "memory usage" somewhere instead of "execution time" or something?21:15
fijalI don't think there is an issue with that21:15
fijalI don't think the file says execution time anywhere21:15
fijalbut whatever21:15
Action: fijal off to bed though21:16
timonatoroh, that's true, it only says "result"21:16
timonatorgood night!21:16
fijal (~fijal@197.111.212.164) left irc: Read error: Connection reset by peer21:18
fox__ (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer21:18
timonatorwow, holy hell, the json_bench is ridiculously slow with non-jit-pypy compared to cpython - not that it's of any real interest, just stumbled upon it21:20
Ademan (~dan@54.sub-97-182-116.myvzw.com) left irc: Quit: leaving21:22
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.21:23
Alex_GaynorFor anyone who wants a debugging challenge, see if you can find out why this test fails: http://ci.djangoproject.com/job/Django/database=sqlite3,python=pypy1.7/lastCompletedBuild/testReport/django.contrib.humanize.tests/HumanizeTests/test_naturaltime/21:23
Alex_Gaynorit's a fun one~!21:23
jmr_mp (jmr@14-200-200-212.static.tpgi.com.au) left #pypy.21:24
Arfrever (~Arfrever@apache/committer/Arfrever) joined #pypy.21:25
fox__ (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.21:28
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer21:28
timonatorAlex_Gaynor: sqlite brings its own datetime class?21:29
ltratt (~ltratt@ntratt.plus.com) left irc: Quit: using sirc version 2.211+KSIRC/1.3.1221:36
derdon (~derdon@p5DE8A92B.dip.t-dialin.net) left irc: Remote host closed the connection21:47
fox__ (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Connection reset by peer21:49
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.21:53
stakkarsfijal: contact me if you need such a proggy.21:54
stakkarsarigato: I would like to add a win32 build bot, anyway. Better than using the existing one.21:55
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Ping timeout: 276 seconds22:01
stakkars__ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.22:01
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds22:02
interrupt (~interrupt@175.197.245.105) joined #pypy.22:03
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.22:03
gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) left irc: Remote host closed the connection22:04
stakkars__ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds22:06
berkerpeksag (~berkerpek@46.197.8.139) left irc: Quit: Instantbird 1.122:06
antocuni (~antocuni@host41-181-static.95-94-b.business.telecomitalia.it) joined #pypy.22:07
berkerpeksag (~berkerpek@46.197.8.139) joined #pypy.22:07
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Ping timeout: 258 seconds22:08
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Ping timeout: 245 seconds22:14
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.22:15
arigato (~arigo@82.113.106.35) left irc: Quit: See you22:17
stakkars__ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.22:22
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Ping timeout: 258 seconds22:23
dmalcolm (david@nat/redhat/x-ljyssrjjysovdhau) left irc: Quit: Leaving22:26
stakkars__ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Ping timeout: 248 seconds22:27
whitelynx (~whitelynx@63.241.75.144) left irc: Quit: Ex-Chat22:28
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.22:29
tilgovi (~randall@c-98-210-155-124.hsd1.ca.comcast.net) joined #pypy.22:31
tilgovi (~randall@c-98-210-155-124.hsd1.ca.comcast.net) left irc: Changing host22:31
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.22:31
whitelynx (~whitelynx@63.241.75.144) joined #pypy.22:34
kenaan12ctismer win64-stage1 113e561e90ee5a 15/lib_pypy/pyrepl/unix_console.py: manual merge22:35
kenaan12ctismer win64-stage1 11b3b8a737d506 15/pypy/: Merge with default22:35
bbot2Started: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/72022:42
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/14622:42
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/94922:42
nickik (~nick@217.150.239.67) left irc: Quit: Verlassend22:45
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Read error: Connection timed out22:47
JaRoel|4d (~jaroel|4d@2001:980:7133:1:3990:eeda:7dab:978d) left irc: Remote host closed the connection22:53
JaRoel|4d (~jaroel|4d@sink.jaroel.nl) joined #pypy.22:53
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.22:55
JaRoel|4d (~jaroel|4d@sink.jaroel.nl) left irc: Ping timeout: 260 seconds22:57
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Ping timeout: 252 seconds23:01
_class_ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) joined #pypy.23:02
berkerpeksag (~berkerpek@46.197.8.139) left irc: Quit: Leaving23:02
_class_0 (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) left irc: Ping timeout: 244 seconds23:03
whitelynx (~whitelynx@63.241.75.144) left irc: Quit: Ex-Chat23:05
Nick change: _class_ -> __class__23:07
fzzzy_ (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy.23:12
fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) left irc: Ping timeout: 240 seconds23:14
Nick change: fzzzy_ -> fzzzy23:14
oal (u4126@gateway/web/irccloud.com/x-qboyegsrdlnmqjyd) left irc: Excess Flood23:14
oal (u4126@gateway/web/irccloud.com/x-sbljvlxafhkylpzb) joined #pypy.23:14
__class__ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) left irc: Ping timeout: 240 seconds23:15
interrupt (~interrupt@175.197.245.105) left irc: Remote host closed the connection23:16
legogris (~legogris@c80-217-207-157.bredband.comhem.se) left irc: Read error: Connection reset by peer23:19
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds23:21
legogris (~legogris@c80-217-207-157.bredband.comhem.se) joined #pypy.23:23
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep23:24
antocuni (~antocuni@host41-181-static.95-94-b.business.telecomitalia.it) left irc: Quit: Leaving23:27
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) joined #pypy.23:27
mfoord (~anonymous@python/psf/voidspace) joined #pypy.23:27
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) joined #pypy.23:35
tlynn_ (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy.23:39
tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Ping timeout: 240 seconds23:43
stakkars_ (~tismer@dslb-088-072-224-185.pools.arcor-ip.net) left irc: Ping timeout: 260 seconds23:50
--- Tue Nov 22 201100:00

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