#pypy IRC log for Thursday, 2011-11-24

thinker (~thinker@123-194-52-90.dynamic.kbronet.com.tw) left irc: Read error: Connection reset by peer00:02
Nick change: Gulopine -> Gulaway00:31
asmeurer (~asmeurer@dhcp-baca-230.resnet.nmt.edu) joined #pypy.00:33
mat^2 (~mathias@212.130.113.35) joined #pypy.00:37
mfoord (~anonymous@python/psf/voidspace) left irc: Quit: mfoord00:43
thinker (~thinker@123-194-52-90.dynamic.kbronet.com.tw) joined #pypy.01:00
JaredW (~jared@061093200075.ctinets.com) joined #pypy.01:03
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/14801:03
bbot24Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/72201:27
ernop (~ernie@114.250.109.254) left irc: Ping timeout: 240 seconds01:28
jimbaker (~jbaker@canonical/jimbaker) left irc: Ping timeout: 245 seconds01:32
jimbaker (~jbaker@c-75-71-80-146.hsd1.co.comcast.net) joined #pypy.01:32
jimbaker (~jbaker@c-75-71-80-146.hsd1.co.comcast.net) left irc: Changing host01:32
jimbaker (~jbaker@canonical/jimbaker) joined #pypy.01:32
hongminhee (~dahlia@175.125.21.72) left irc: Quit: leaving01:39
hongminhee (u295@gateway/web/irccloud.com/x-fsxllqagmtgxpauo) joined #pypy.01:40
ernop (~ernie@205.164.24.90) joined #pypy.01:40
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/185801:41
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/58501:41
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/28001:41
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/62101:41
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/111901:41
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/80601:41
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/144801:41
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/71701:41
JaredW (~jared@061093200075.ctinets.com) left irc: Ping timeout: 248 seconds01:46
bbot24Failure: 15http://buildbot.pypy.org/builders/own-win-x86-64/builds/53 [12ctismer, win64-stage1]01:52
thinker (~thinker@123-194-52-90.dynamic.kbronet.com.tw) left irc: Read error: Connection reset by peer02:04
pybot joined #pypy.02:10
-ChanServ (ChanServ@services.)- [#pypy] Welcome!02:10
-ChanServ (ChanServ@services.)- [#stackless] Welcome to #stackless. This channel is for the discussion of matters relating to Stackless Python. Remember that others in this channel may not be in the same timezone as you, so stick around for an answer (even if only in your channel history) if you ask a question. The mailing list is also recommended as being a faster place to get help (http://www.stackless.com/mailman/listinfo/stackless).02:11
santagada (~leonardo@201.86.215.236) joined #pypy.02:16
asmeurer (~asmeurer@dhcp-baca-230.resnet.nmt.edu) left irc: Quit: asmeurer02:29
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/95102:40
Kaskuka (~John@osbk-4db16da1.pool.mediaWays.net) left irc: Ping timeout: 240 seconds02:45
Kaskuka (~John@osbk-4d08b0fc.pool.mediaWays.net) joined #pypy.02:45
Surgo_ (~surgo@gw1.miraiyashoten.co.jp) left irc: Quit: Leaving...03:02
ernop (~ernie@205.164.24.90) left irc: Ping timeout: 240 seconds03:22
fprimex (~fprimex@fl-67-233-144-228.dhcp.embarqhsd.net) joined #pypy.03:37
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/185803:39
fprimex (~fprimex@fl-67-233-144-228.dhcp.embarqhsd.net) left irc: Quit: Leaving03:42
santagada (~leonardo@201.86.215.236) left irc: Ping timeout: 240 seconds03:55
santagada (~leonardo@201.86.215.236) joined #pypy.03:55
hakanardo (~hakan@h59ec009e.seluldx.dyn.perspektivbredband.net) left irc: Ping timeout: 245 seconds04:00
jonvk (~jonathan@bas1-montreal29-1242341904.dsl.bell.ca) left irc: Remote host closed the connection04:08
ernop (~ernie@114.250.109.254) joined #pypy.04:14
ernop (~ernie@114.250.109.254) left irc: Ping timeout: 240 seconds04:23
asmeurer (~asmeurer@c-174-56-21-245.hsd1.nm.comcast.net) joined #pypy.04:31
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/144804:35
ernop (~ernie@114.250.109.254) joined #pypy.04:35
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/80604:36
ernop (~ernie@114.250.109.254) left irc: Ping timeout: 248 seconds04:42
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/62104:49
ernop (~ernie@114.250.109.254) joined #pypy.04:50
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/111904:52
ernop (~ernie@114.250.109.254) left irc: Ping timeout: 248 seconds04:55
ernop (~ernie@114.250.109.254) joined #pypy.04:57
ernop (~ernie@114.250.109.254) left irc: Ping timeout: 248 seconds05:03
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/58505:03
ernop (~ernie@69.172.210.254) joined #pypy.05:15
hakanardo (~hakan@h59ec009e.seluldx.dyn.perspektivbredband.net) joined #pypy.05:18
zain (~textual@c-67-160-201-63.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep.05:35
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/28005:36
fijal (~fijal@41.53.91.100) joined #pypy.05:57
fijalhi06:00
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/71706:01
Alex_Gaynorfijal: hey06:04
fijalslightly worrying that ubuntu thinks about porting stuff to python 3, but does not seem to be interested in shipping pypy06:06
Alex_Gaynorfijal: are we ready to merge list-strategies?06:11
fijalI think so06:11
fijalalthough I did not do a very deep review06:11
fijalbut yes, I suppose so06:11
fijalAlex_Gaynor: you know we're ~4x faster on twisted web these days?06:12
fijalthan cpython06:12
Alex_Gaynorcool06:12
fijalwe can't run it nightly though06:12
fijalwe run out of tcp sockets or somesuch06:12
danchr (~danchr@cl-848.chi-02.us.sixxs.net) left irc: Read error: Connection reset by peer06:13
Alex_Gaynorheh06:14
danchr (~danchr@cl-848.chi-02.us.sixxs.net) joined #pypy.06:14
fijalwe're too fast :)06:15
mat^2 (~mathias@212.130.113.35) left irc: Ping timeout: 248 seconds06:34
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 245 seconds06:48
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy.06:54
getxsick (xsx@unaffiliated/getxsick) joined #pypy.06:58
kdas_ (kdas@nat/redhat/x-mzjxsnigxwuglwud) joined #pypy.07:01
kdas_ (kdas@nat/redhat/x-mzjxsnigxwuglwud) left irc: Remote host closed the connection07:01
canta (~canta@77-20-123-240-dynip.superkabel.de) joined #pypy.07:04
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.07:04
ixokai (~ixokai@rrcs-24-199-8-246.west.biz.rr.com) joined #pypy.07:06
ixokai (~ixokai@rrcs-24-199-8-246.west.biz.rr.com) left irc: Changing host07:06
ixokai (~ixokai@pdpc/supporter/student/ixokai) joined #pypy.07:06
lambacck_ (~chris@d24-150-124-118.home.cgocable.net) joined #pypy.07:17
fijalAlex_Gaynor: feel like merging list-strategies07:17
fijal?07:17
lambacck (~chris@d24-150-124-118.home.cgocable.net) left irc: Ping timeout: 260 seconds07:19
zain (~textual@c-24-130-236-153.hsd1.ca.comcast.net) joined #pypy.07:28
Alex_Gaynorfijal: err, I can, is there a reason you didn't though?07:28
mwhudson_ (~mwh@121-73-77-183.cable.telstraclear.net) joined #pypy.07:29
fijalno07:29
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 252 seconds07:29
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.07:33
Alex_Gaynorfijal: ok, merging then07:34
kenaan12alex_gaynor list-strategies 118e6605b379dc 15/: merged default in07:36
kenaan12alex_gaynor list-strategies 112449ac0ea4a1 15/: Closing branch to be merged.07:36
kenaan12alex_gaynor default 1103ac38ea2709 15/pypy/: (lukas, carl) Merged list strategies.  This allows the contents of a list to be unboxed.  See the blog for mo...07:36
antocuni (~antocuni@host142-66-static.94-94-b.business.telecomitalia.it) joined #pypy.07:38
Alex_Gaynorfijal: can you whack the buildbots?07:39
fijalsure07:39
fijalthere is a script for that07:39
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1859 [12fijal, default]07:39
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/718 [12fijal, default]07:39
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1120 [12fijal, default]07:39
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/586 [12fijal, default]07:39
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/281 [12fijal, default]07:40
Alex_Gaynorcool07:40
zain (~textual@c-24-130-236-153.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep.07:40
antocunihi07:40
Alex_Gaynormorning antocuni 07:40
antocuniso, we merged list-strategies?08:06
antocunicool :-)08:06
Action: fijal wonders how benchmarks work08:07
fijallots of spam on the blog post08:08
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.08:09
antocunifijal: what are your doubts about benchmarks?08:11
fijalantocuni: I wonder if list-strategies makes anything faster08:11
fijaland if so, by how much08:11
antocuniI don't see the merge commit on codespeed08:12
fijalthis is probably a bigger event than the release for me08:12
fijalantocuni: because the benchmarks were not run yet?08:12
antocuniah08:12
antocuniI thought it was merged yesterday08:13
Alex_Gaynorno, it was merged 40 minutes ago :)08:13
Action: antocuni has 167 unread commits in his mailbox08:14
fijalafter 3h of pondering I decided - I'll merge numpy-multidim08:15
fijalAlex_Gaynor: after test improvements and perf improvements though08:15
Alex_Gaynorfijal: did you make len(shape) part of isngature?08:16
Alex_Gaynor(and write tests for it)08:16
fijalno, not yet08:16
fijalI made things that don't involve views almost as fast08:16
fijal(modulo one or two int_add)08:16
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds08:17
fijalAlex_Gaynor: why does signature have immutable_fields even?08:22
Alex_Gaynorfijal: what do you mean?08:22
fijalisn't greenkey considered constant anyway08:23
fijal?08:23
Alex_Gaynoryou have a tree of signatures, greenkey isn't recursively constant08:24
fijalright08:28
fijalwhy there is self.signature and self.reduce_signature on W_Ufunc2?08:29
Alex_Gaynorfijal: because they're different things, reduce signature is what gets used for reduce, and signature is for when you call ufunc on an array08:30
canta (~canta@77-20-123-240-dynip.superkabel.de) left irc: Quit: Leaving.08:31
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1859 [12fijal, default]08:40
cfbolzAlex_Gaynor: thanks for merging08:42
cfbolzyay08:42
Alex_Gaynorcfbolz: merging is easy, you and lukas did the work, and fijal did the review :)08:42
cfbolzlukas did most of the work, actually08:42
mwhudson_ (~mwh@121-73-77-183.cable.telstraclear.net) left irc: Ping timeout: 258 seconds08:42
cfbolzI expect nothing at all to change in the nightly benchmarks, and I would call it a good thing08:43
Alex_Gaynorslowspitfire probably gets a bit faster08:43
Alex_Gaynordoesn't it have a big list of strs and then "".join?08:43
cfbolzah, maybe08:43
fijalcfbolz: aren't quite a few affected?08:45
fijallike nbody maybe?08:45
Alex_Gaynornbody has a list of integers?08:45
cfbolzok, possible08:45
fijalfloats I think08:45
Alex_Gaynorwe don't have a float specilization yet08:45
Alex_Gaynoryou should add one :)08:46
fijalah08:46
cfbolzpuh, is that clearly worth it?08:46
fijalcfbolz: lots of benchmarks use lists of floats08:46
Trundle (~andy@p578bfdcf.dip0.t-ipconnect.de) joined #pypy.08:46
fijalI would say yes08:46
Trundle (~andy@p578bfdcf.dip0.t-ipconnect.de) left irc: Changing host08:46
Trundle (~andy@python/site-packages/trundle) joined #pypy.08:46
cfbolzand how much real-world code?08:46
Alex_Gaynoryes, it'd be cool to have them automatically be fast, instead of needing to say, "please use array module or numpy"08:46
fijaldepends what you call real code08:47
fijalnow-i-can-write-this-in-python08:47
fijalsounds like a real use case08:47
fijalfor example hakan's image processing stuff08:47
cfbolzok08:47
fijala bit more advanced case is a list of characters08:48
cfbolzfijal: that sounds useless08:48
fijalwhy?08:48
cfbolzlist of strings is good enough08:48
fijalno08:48
Alex_Gaynorisn't that what bytearray is for?08:48
fijalencryption?08:48
fijalyes ok08:48
Alex_Gaynor(Python has only one way to do things)08:48
fijalwe can make bytearray faster for that08:48
Alex_Gaynorhehe08:48
fijalbut we kinda lack a fast mutable string type08:48
cfbolzfijal: show me some real-world code where list-of-chars is significantly faster than list-of-strings08:49
Alex_Gaynoralways?  avoiding allocating strings sounds like a good win08:49
fijalcfbolz: does that has to be a code that's written already or a code that I would write is fine as well?08:49
fijalI would probably be happy writing say bz2 using list-of-chars08:50
cfbolzfijal: I would say pre-existing code08:50
cfbolzfijal: because if you write it again, you would use bytearray08:50
fijalwould be nicer not to have to use bytearray08:50
cfbolzoh come on08:50
fijalalso except bytearray is super slow in pypy08:50
cfbolzno, only for string-like methods08:50
Alex_Gaynoryes, indexing should still be fast08:51
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) joined #pypy.08:51
cfbolzalso, each new strategy adds to the size of the executable08:51
fijalor passing it to anything else08:51
fijalas an argument08:51
Alex_Gaynor?08:51
Alex_Gaynoroh, to buffer methods you mean08:52
fijal''.join(list-of-bytearrays)08:52
fijalfor example08:52
cfbolzanyway, I won't implement this08:52
fijalcfbolz: would you implement the float strategy?08:52
cfbolzyes08:52
fijalthis one is clearly a win, look at nbody :)08:52
fijalok08:53
fijalI'm fine with saying list of characters is not the best thing to do :)08:53
fijalon a completely unrelated note - how useful will be a hook that dumps a traceback each time you abort jitting?08:53
fijalbecause right now we see abort: force quasi_immutable: 108:54
fijalbut no idea where it comes from08:54
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy.08:56
fijalcfbolz: didn't we discuss that we can have list of ints and floats with some marking?08:57
fijalnan marking08:57
cfbolzyes, that's even more crazy08:57
cfbolzI want to see usecases first :-)08:58
fijalok08:58
fijalI think the only usecase for a mixed one is when someone is not careful08:58
fijaland say initializes the list with [0] * stuff08:58
fijalbut then only uses floats08:58
fijalor slips a . in -1 somewhere08:58
cfbolzok, but nan-marking is slower than just a list of floats08:58
fijalok08:59
Alex_Gaynorbetter to have good logging and performance tools for than, than crazy encoding08:59
fijalAlex_Gaynor: I'm trying to work on that, not easy08:59
cfbolznow that list-strategies are merged, lukas should integrate set-strategies and list-strategies more09:00
fijalyou mean set(list) looking into strategies?09:01
cfbolzyes09:01
fijalis dict.fromkeys(list) doing that?09:01
cfbolzand the other way round09:01
cfbolzfijal: no09:01
fijalsounds like a good thing as well09:01
cfbolzguess so09:01
cfbolzand list(set)09:02
kenaan12cfbolz set-strategies 11074aae06a038 15/: merge default09:02
asmeurer (~asmeurer@c-174-56-21-245.hsd1.nm.comcast.net) left irc: Quit: asmeurer09:14
antocuni (~antocuni@host142-66-static.94-94-b.business.telecomitalia.it) left irc: Ping timeout: 252 seconds09:14
fijal (~fijal@41.53.91.100) left irc: Ping timeout: 252 seconds09:22
G2P (~G2P@fw-asn1.ornis.com) joined #pypy.09:23
CIA-6803jonash 07roundup * 10#733/bz2 decompression is very slow: 09:33
CIA-68Performance of GZip and BZip2 has been improved dramatically in PyPy 1.7 it seems --09:33
CIA-68it's now almost proportional to CPython's performance (PyPy ... * 14https://bugs.pypy.org/issue73309:33
unbit_ (~unbit@93-58-162-185.ip159.fastwebnet.it) joined #pypy.09:34
unbit (~unbit@93-58-162-185.ip159.fastwebnet.it) left irc: Read error: Connection reset by peer09:34
Nick change: unbit_ -> unbit09:34
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/718 [12fijal, default]09:40
amaury_ (amaury_@nat/google/x-xryaxfasmyexfxbd) joined #pypy.09:42
kenaan12alex_gaynor default 115f1a23beccab 15/pypy/: comparing a unicode char to str char no longer requires allocating a string09:52
ojii (~ojii@40-34.60-188.cust.bluewin.ch) joined #pypy.09:54
iori (~iori@110-133-45-54.rev.home.ne.jp) joined #pypy.10:17
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) left irc: Remote host closed the connection10:19
antocuni (~antocuni@host142-66-static.94-94-b.business.telecomitalia.it) joined #pypy.10:21
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.10:23
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.10:23
bivab (~david@fwstups.cs.uni-duesseldorf.de) joined #pypy.10:35
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1120 [12fijal, default]10:36
thinker (~thinker@123-194-52-90.dynamic.kbronet.com.tw) joined #pypy.10:37
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/586 [12fijal, default]10:39
elliottcable (~me@ell.io) joined #pypy.10:51
cfbolzcool, seems like the merge didn't make problems10:58
fijal (~fijal@197.171.24.193) joined #pypy.11:08
canta (~canta@77-20-123-240-dynip.superkabel.de) joined #pypy.11:09
senyai (~senyai@62.173.135.205) joined #pypy.11:09
amaury_ (amaury_@nat/google/x-xryaxfasmyexfxbd) left irc: Ping timeout: 240 seconds11:33
derdon (~derdon@p5DE8AE4D.dip.t-dialin.net) joined #pypy.11:34
asabil (~asabil@131.246.16.62.customer.cdi.no) joined #pypy.11:36
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/281 [12fijal, default]11:40
fijalcrash?11:40
fijalcfbolz: http://buildbot.pypy.org/summary/longrepr?testname=TestWeakReference.%28%29.test_proxy&builder=own-linux-x86-32&build=1859&mod=module.cpyext.test.test_weakref11:41
fijalthis seems like a new failure11:41
arigato (~arigo@fwstups.cs.uni-duesseldorf.de) joined #pypy.11:42
fijalhi armin11:44
arigatohi11:46
lmoura (~lmoura@177.17.97.79) left irc: Ping timeout: 258 seconds11:56
lizardo (~lizardo@189.2.128.130) joined #pypy.11:56
kenaan12arigo jit-targets 11d0c6a5afc72a 15/pypy/jit/backend/x86/test/test_recompilation.py: Fix tests.11:58
cwillu (~cwillu@cwillu.com) joined #pypy.12:05
derdon (~derdon@p5DE8AE4D.dip.t-dialin.net) left irc: Remote host closed the connection12:12
fijal (~fijal@197.171.24.193) left irc: Read error: Connection reset by peer12:15
cfbolzgaaah12:17
JStoker (jstoker@unaffiliated/jstoker) left irc: Ping timeout: 276 seconds12:18
arigatocfbolz: gaah?12:24
kenaan12arigo jit-targets 1110f20488e230 15/pypy/jit/backend/: Subtle fix.  Used to show up in test_loop_unroll occasionally.12:24
cfbolzarigato: http://buildbot.pypy.org/summary/longrepr?testname=TestWeakReference.%28%29.test_proxy&builder=own-linux-x86-32&build=1859&mod=module.cpyext.test.test_weakref12:24
fijal (~fijal@197.172.122.113) joined #pypy.12:24
arigato.-)12:24
JStoker (jstoker@unaffiliated/jstoker) joined #pypy.12:25
amaury_ (amaury_@nat/google/x-tvnlnljboxrdlhnz) joined #pypy.12:25
fijalhakanardo: ping12:26
lmoura (~lmoura@177.17.97.79) joined #pypy.12:26
fijalarigato: would you by chance know details about unroll.py? ;-)12:28
fijalprecisely why there are operations that are forced at the jump that are clearly loop invariant and unused?12:28
senyai (~senyai@62.173.135.205) left irc: Ping timeout: 252 seconds12:29
arigato (~arigo@fwstups.cs.uni-duesseldorf.de) left irc: Ping timeout: 244 seconds12:31
kenaan12cfbolz default 11bedfacfcd12f 15/pypy/module/_weakref/interp__weakref.py: don't use w_None here, it breaks cpyext12:31
fijalI think I scared armin12:32
kenaan12fijal numpy-multidim 11ce322f0bc995 15/pypy/module/micronumpy/: progress on test_zjit. Now test_slice is almost ready, barring few arraylen_gc that are from god knows where...12:34
Alex_Gaynorfijal: I wonder if we should have the "kill unused ops" pass be a part of hte frotnend12:35
fijalthat would not help here12:35
fijalbecause those are leftovers from unroll.py12:35
fijalI guess12:35
Alex_Gaynoralso, why is shapelen a second green, rather than part of a signature12:35
fijalbecause it's not a part of signature12:36
fijalyou only specialize on the length of shape of the result12:36
Alex_Gaynorah, ok12:36
kenaan12fijal numpy-multidim 1120d333f34dfe 15/pypy/module/micronumpy/test/test_zjit.py: leave some comments12:36
fijaltest_zjit for broadcasting would be good (it works but is inefficient)12:37
fijalotherwise it's missing few tests and it's ready to be merged12:37
Action: fijal is running coverage12:37
kenaan12alex_gaynor extradoc 115ca7f8992b0c 15/planning/micronumpy.txt: another TODO item12:38
ThomasWaldmannttp://moinmo.in/EasyToDo/test%20moin2%20with%20pypy%201.7 anything special we should check?12:40
fijalnumpy/numpypy broke -A tests12:42
Alex_Gaynorsigh :/12:43
ThomasWaldmann(that task page is for google code-in)12:43
fijalThomasWaldmann: sounds good12:44
fijaldo you have benchmarks?12:44
ThomasWaldmannyes, but not very good12:44
fijala link to them would be good12:44
ThomasWaldmann(in the tests, but normally skipped)12:45
fijalor how to run them12:46
ThomasWaldmannyup12:46
fijalcause if I were I student I would have no clue what does this mean to check performance12:46
fijalAlex_Gaynor: do you know why our tostr generates 0.0 for float 0 and numpy's 0.12:47
fijal?12:47
Alex_Gaynorfijal: I guess beccause no one has looked up the exact rules of numpy's and implemented them?12:47
fijalok12:47
Alex_Gaynorits not exactly an important priorty for me ATM12:48
fijalright12:48
fijalit should not be a merge blocker either12:48
Alex_Gaynorno, its an issue on trunk as well12:48
kenaan12fijal numpy-multidim 11bb1ed602bd16 15/pypy/module/micronumpy/test/test_numarray.py: Add two tests. One breaks, but it's not exactly a merge blocker12:49
Alex_Gaynorfijal: why commit that test, all our other str tests just test the way we do it (0.0)12:49
fijalwell, ok12:50
Alex_Gaynorwe can't fix that one without fixing all the other ones too12:50
arigato (~arigo@fwstups.cs.uni-duesseldorf.de) joined #pypy.12:50
fijalAlex_Gaynor: do you know how argmax works on multiple items btw?12:53
fijalmultiple dimensions12:53
Alex_Gaynorfijal: it appears to return the integer index into the underlying C array12:54
fijalseems so12:54
fijalit operates on flattened version12:54
Alex_Gaynorright12:54
fijalbut no12:54
fijalif you do a view12:54
fijalit returns index in the view12:55
Alex_Gaynorof course12:55
Alex_Gaynorwhat would you expect?12:55
fijalI don't know12:57
Alex_GaynorI expected a tuple index TBH, but given it returns a flat one, doing the same for a view makes sense12:57
kenaan12fijal numpy-multidim 11195a2e3ca74a 15/pypy/module/micronumpy/: For now multidimarray.argmax is a TypeError. Wait until we implement flatiter12:59
fijalgood12:59
fijalready for the review I guess?12:59
fijalah, broadcasting test_zjit12:59
Alex_Gaynoruh, why TypeError, seems easy to me12:59
Alex_Gaynorjust iterate over the thing and keep your own count13:00
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.13:01
jonathan__ (~jonathan@bas1-montreal29-1242341904.dsl.bell.ca) joined #pypy.13:03
arigatohakanardo: the crashes in jit-targets' test_string look more annoying13:03
arigatohakanardo: I think it's worth adapting test_random to produce LABEL instructions too13:03
fijalAlex_Gaynor: yes, but it also accept an argument13:05
fijalthat does stuff13:05
senyai (~senyai@178.176.20.24) joined #pypy.13:05
fijalAlex_Gaynor: can we keep things separated?13:05
Alex_Gaynorfijal: do you want me to implement it?13:05
fijalyes please13:05
fijalcan you implement .flat at the same time?13:05
Alex_Gaynorthey're kind of seperate IMO13:06
fijalok13:07
fijalgood13:07
fijalAlex_Gaynor: do you feel like reviewing btw?13:07
fijaljust let me commit last changes13:08
Alex_Gaynorfijal: yes, I'll reivew after I do this13:08
fijalso there are possible optimizations13:08
fijalbut it's not too bad as of now13:08
kenaan12fijal numpy-multidim 113d9f6f9f6c19 15/pypy/module/micronumpy/: some work on broadcast iterator. test_zjit requires some support from jit targets13:08
Alex_Gaynorfijal: can you explain to me compute_offset?13:10
fijalyes13:10
Alex_Gaynorerr, compute_index13:10
fijalit probably does not make sense any more :)13:10
fijalto keep it short13:10
fijalI thought we're going to return a tuple of indexes13:10
fijalbut we're not13:11
fijalso it's irrelevant13:11
fijalit's a way to store some sort of result13:11
fijalif we're going to return an index anyway, why bother13:11
fijalAlex_Gaynor: the only things missing tests are in compile()13:13
fijalwhich I would not worry too much for now13:13
senyai (~senyai@178.176.20.24) left irc: Ping timeout: 252 seconds13:18
Alex_Gaynorfijal: http://paste.pocoo.org/show/512348/13:20
Alex_Gaynorlook ok?13:20
hakanardo__ (~hakan@gw.cognimatics.ideon.se) joined #pypy.13:23
kenaan12alex_gaynor numpy-multidim 114f7747b4ff1c 15/pypy/module/micronumpy/: argmax works for multidim now13:24
hakanardo__hi13:24
Alex_Gaynorhakanardo__: fijal has been looking for you13:24
Alex_Gaynor(hi)13:24
hakanardo__fijal: hi13:25
kenaan12alex_gaynor numpy-multidim 117f55ec059153 15/: merged default in13:26
hakanardo__fijal: virtuals are forced at the end of the peel loop if they are not virtuals at the end of the preamble13:26
hakanardo__arigato: thanx for looking into the x86 failures13:28
arigatono problem13:28
arigatoI see that there is already a LABEL in test_random, but only at the start of the loop, so I'm trying to tweak it13:28
hakanardo__great!13:28
arigato(with limited success so far)13:29
hakanardo__ok, I still have some work to do here but hopefully I'll be able get some pypy time tonight...13:31
arigato:-)13:31
ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) joined #pypy.13:44
Rhyolitefijal: Gustavo made his interest in Python 3 very clear at PyCon Brasil13:52
fijal (~fijal@197.172.122.113) left irc: Read error: Connection reset by peer13:52
Rhyoliteah, well13:52
exarkunwhich Gustavo13:52
exarkunNiemeyer?13:53
Rhyoliteyes13:53
exarkunI thought he liked Go now13:53
Rhyolitehe does13:53
Rhyolitebut he also wants to avoid a Python split13:53
Rhyoliteso he is pushing very hard for Python 313:53
exarkunIs he going to use his time machine?13:53
exarkun(Python is split already, how can you avoid it?)13:53
Rhyolitehe is using whatever resources he has13:53
Rhyoliteincluding pushing Canonical to port everything to Python 313:54
`fox` (~fox@95.232.111.141) joined #pypy.13:58
santagadaRhyolite, I talked to him about canonical funding pypy13:58
santagadadidn't go very far13:58
exarkunBarry Warsaw seems to think it will still be 3.5 years before Python 3 is the primary target13:58
santagadaI also talked to kiko... he also wasn't very interested13:59
santagadaubuntu 14.04 will not have python 2.7 by default14:00
santagadathen I think people will migrate14:00
Rhyolitesantagada: Gustavo's primary interest would be Py3k14:00
fijal (~fijal@41.55.94.227) joined #pypy.14:00
fijalhakanardo__: still around?14:00
santagadaRhyolite, it's not like having a fast Py3k would not help14:00
RhyoliteI think Gustavo would be very happy if Canonical could ship Ubuntu with Py3k14:00
Rhyolitefijal: Gustavo made his focus on Python 3 very clear at PyCon Brasil14:01
santagadaRhyolite, they will ship python 3.2 by default with 12.04 and are going to remove 2.7 in 14.0414:01
hakanardo__fijal: yep14:01
fijalhakanardo__: can you run test_zjit -k test_multidim_slice --viewloops on numpy-multidim branch?14:01
fijalin pypy/module/micronumpy/test14:01
fijalRhyolite: that is very unfortunate14:02
hakanardo__ok in a sec14:02
Rhyolitefijal: he wants to use his influence to prevent a Python split14:02
Rhyolitefork14:02
fijalfork into python 2 and python 3?14:03
Rhyoliteyes14:03
fijalAlex_Gaynor: does that even work?14:03
fijalwhat if you have a slice ::214:03
fijal?14:03
asabil_ (~asabil@69-217-15.connect.netcom.no) joined #pypy.14:04
fijalRhyolite: I don't see how this prevents a split at all14:05
fijalbut well14:05
fijalit's more - prevent a split and go for python 314:05
Rhyoliteit doesn't prevent it14:05
Rhyoliteyes, he wants to support Python 3 as much as possible14:06
Rhyolitewhether Python 3 was the right solution and the right time14:06
Rhyolitehe wants to march forward as quickly as possible14:06
fijalI think also that the original blog post is not quite honest14:06
exarkunPython 3: https://en.wikipedia.org/wiki/Sunk_costs14:07
fijal"Even large frameworks such as Twisted are working on Python 3 branches"14:07
Rhyoliteand he can direct Canonical's resources to convert to Python 3 as quickly as possible14:07
fijalit's not entirely true14:07
Rhyolitefijal: you mean FOSS communities use propaganda?!14:07
fijalI guess this is what I said, unbelievable eh?14:08
senyai (~senyai@178.176.20.24) joined #pypy.14:08
asabil (~asabil@131.246.16.62.customer.cdi.no) left irc: Ping timeout: 258 seconds14:08
Nick change: asabil_ -> asabil14:08
Rhyoliteamazing14:08
fijalAlex_Gaynor: ping?14:08
hakanardo__fijal: ok, got it14:08
fijalhakanardo__: so there are two questions14:08
fijalhakanardo__: one is why we have arraylen_gc in the main loop?14:08
fijalthe second one is why the bridge contains so much strange stuff before jump14:09
santagadaexarkun, I think moving to python 3 is nice, but maybe we should never ever attempt a python 3 again14:09
Alex_Gaynorfijal: one sec, someone unplugged my internet, reading backlog14:10
fijalAlex_Gaynor: bastard!14:10
hakanardo__if there is a loop invariant getarrayitem moved out of the loop into the short preamble we need to also insert array bound checks14:10
RhyoliteThey killed Kenny's Internet!14:10
exarkunsantagada: I think moving to Python 3 has already cost much more than it's worth and we're nowhere close to done, and since they did it once they'll probably do it again.14:10
hakanardo__since the array might be too short at the end of some potential future bridge14:11
fijalhakanardo__: but at the end of the day we end up with only arraylen_gc14:11
fijalwe don't have guards!14:11
hakanardo__right, they will be killed by the backend14:11
fijalno, they won't14:11
Alex_Gaynorfijal: ok here14:11
fijalbecause they're jump arguments14:11
kenaan12alex_gaynor numpy-multidim 11fbd6482de5d2 15/pypy/module/micronumpy/: some code-prettyness changes14:11
hakanardo__ah, I see14:11
fijalAlex_Gaynor: so, does it work with a[::2]?14:11
Alex_Gaynorfijal: yes, there's a test for that14:11
santagadaexarkun, unfortunately I think you are right...14:11
fijalAlex_Gaynor: ok14:11
Rhyoliteexarkun: I don't think anyone will try this again. Only Guido would have tried it this time to clean up Python before he retires.14:12
Alex_Gaynorfijal: I just finished my review, http://dpaste.com/661265/ are my items14:12
santagadaexarkun, stop with the logic stuff, you are bringing me down14:12
fijalsantagada: people who push for python 3 are not people who maintain large python codebases14:12
exarkunRhyolite: I think you're pretty naive.14:12
fijalAlex_Gaynor: ok14:12
hakanardo__fijal: the arraylen_gc's at the end of the loop are not jump arguments here?14:12
fijalhakanardo__: yes they're?14:13
Rhyoliteexarkun: I normally don't make discussions personal14:13
Alex_Gaynorexarkun: they did it once a long long time ago, doesn't anyone remember 1.5?14:13
exarkunAlthough if you think Guido's motivation was to make something pretty before he retires, then maybe you're even more pessimistic than me.14:13
Alex_Gaynors/pretty/unbroken/14:13
santagadaAlex_Gaynor, I didn't have any big code base to move, but as a language user 1.5 was not a big jump14:13
Rhyoliteexarkun: I heard that independently14:13
Alex_Gaynorsantagada: err, 1.5->2x I meant14:13
exarkunRhyolite: This entire conversation is personal14:13
exarkunRhyolite: It's all "I think this" and "I think that"14:14
fijalhakanardo__: ah indeed14:14
exarkunRhyolite: I can only comment on your personal opinions.14:14
fijalhakanardo__: ok, so backend kills those, you're right14:14
fijalhakanardo__: would be cooler if it didn't have to but fine14:14
santagadaAlex_Gaynor, yep 1.5->2.0 was much simpler than 2->314:14
fijalhakanardo__: now, to the bridge stuff14:14
kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) left irc: Quit: Textual IRC Client: http://www.textualapp.com/14:14
Alex_Gaynorsantagada: that happened in 2000, I was 10 years old then :)  however I've heard it was still a recent backwards incompatible change14:14
santagadamaybe it was because I was just learning the language14:14
hakanardo__at the end of the bridge the same arraylen's are needed to make sure the bridge did not shrink the array14:15
hakanardo__ie they are followed by guards14:15
exarkunRhyolite: I guess there's that trick, "I think that idea is pretty naive.", some people like that better.14:15
hakanardo__there is also a bunch of guard_nonnull_class which is probably guards that was moved to the preamble14:15
santagadaAlex_Gaynor, I just remembered that you are really young... I was learning python then. Think on the good side, I think you did more for python than I ever will :)14:15
fijalhakanardo__: why do we need the whole chain of guard_class?14:15
exarkunAlex_Gaynor: It was, but all Python releases are backwards incompatible to some degree14:15
fijalhakanardo__: yes, but noone could have changed that right?14:16
pedronislanguage developers and enthusiasts tend to overestimate the importance of languages as such, it's much more relevant which libraries they support (of course that is not independent from the language characteristics but the relationship is nuanced)14:16
Alex_Gaynorexarkun: well, they're all backwards incompatible WRT to observable behavior, but necessarily defined behavior14:16
exarkunAlex_Gaynor: 1.5 -> 2.0 wasn't as bad because free license to break things wasn't taken - useful changes were made, with the realization they'd break things, but not with explicit goals of breaking things.14:16
hakanardo__maybe, but the bridge does not inherit any such information form the loop, so we cant be certan of that14:16
fijalhakanardo__: also it's provable noone changed the array length14:16
fijalok14:16
exarkunAlex_Gaynor: "observable behavior" is all I care about :)14:16
fijalcan we somehow make it so?14:17
exarkunAlex_Gaynor: Python 2.x releases add new keywords, which breaks code14:17
hakanardo__fijal: there is the idea of the jit-fromstart branch14:17
exarkunAlex_Gaynor: They change library behavior, which breaks code14:17
fijalah ok14:17
fijalso it's experimental14:17
Alex_Gaynorexarkun: when I add a new method to a class so that name no longer raises AttributeError, that breaks observable behavior, hell if I'm going to stop doing it though :)14:17
exarkunAlex_Gaynor: They do lots of stuff to break code14:17
hakanardo__what we need is to make the bridge inherit more information from it's parent loop 14:17
exarkunAlex_Gaynor: But it's possible to _deal_ with it14:17
fijalthe thing is, we encode all this stuff in signature14:17
hakanardo__to be able to prove things like this14:17
santagadaexarkun, was there a twisted then?14:17
exarkunAlex_Gaynor: The Python 2.7 port of Twisted took maybe a week14:17
Alex_Gaynorthat long o_O14:17
exarkunAlex_Gaynor: The 3.x port of Twisted is shaping up to take more like 5 years :)14:18
fijalAlex_Gaynor: hey, can you do something productive instead and look for a bit to loops?14:18
Alex_Gaynorfijal: instead of arguing about obscure minutae of python releases that no one but people in #pypy care about?14:18
fijalAlex_Gaynor: yes14:18
Alex_Gaynorwhat loop should I look at?14:18
exarkunAlex_Gaynor: I can deal with a week, but anyone who thinks I'm going to set aside 5 years just because Guido wants a shiny new toy is crazy.14:18
fijalAlex_Gaynor: test_zjit.py -k test_multidim_slice --viewloops14:19
fijaland look at the bridge14:19
exarkunsantagada: Hm.  I wasn't working on the project when Python 1.5 was out.14:19
Alex_Gaynorfijal: only if you tell me how to install pygame without breaking my beautiful system14:19
fijalno14:19
Alex_Gaynortrying to pip install it results in it downloading the OSX installer and exploding14:19
exarkunsantagada: I think things didn't really get going until after 2.0 though.14:19
fijalget yourself an operating system14:19
santagadaAlex_Gaynor, are you using lion?14:20
Alex_GaynorI have one, I just don't use it's package manager for stuff like this.14:20
Alex_Gaynorsantagada: No, I'm using ubuntu!14:20
exarkunAlex_Gaynor: If you want to talk about adding new methods breaking things, then we would have to talk about what "compatiblity" means14:20
santagadaAlex_Gaynor, pygame install sucks big hairy b...14:20
exarkunAlex_Gaynor: which python-dev has always refused to do :)14:20
exarkun(so it would be a short conversation)14:20
Alex_Gaynorexarkun: especially if raymond has his way.  for example apaprently heapq.islice with a negative index has no defined behavior14:21
Alex_GaynorWould it be unreasonable for me to make pypy's implementation spawn nethack?14:22
exarkunWithout even a minimal document about what is considered compatible or not, raymond's games about "no defined behavior" don't even make sense.14:22
exarkunis Python's compatibility policy that only defined behavior will remain the same between releases?  how do I know it's not that some behavior may not be defined, but it will remain the same?14:23
Alex_Gaynorfijal: feel like fixing the items on my TODO, and putting a few comments around the iterator math?14:24
fijalAlex_Gaynor: I'm fixing items on todo14:24
fijalwhat about iterator math?14:24
Alex_Gaynorfijal: a comment explaining it please14:24
fijalok14:25
fijala .reshape would be nice btw14:25
Alex_Gaynorone thing at a time14:25
senyai (~senyai@178.176.20.24) left irc: Ping timeout: 252 seconds14:32
kenaan12fijal numpy-multidim 115da471031a21 15/pypy/module/micronumpy/: Few missing tests, code improvements14:33
kenaan12fijal numpy-multidim 11dc3ff57935ac 15/pypy/module/micronumpy/interp_numarray.py: leave a comment14:33
fijalAlex_Gaynor: ok?14:34
Alex_Gaynorfijal: results for sum() are good, but doing explicit numpy.add.reduce() returns an array with reduction along each axis or something like that14:36
fijalnice14:36
fijalis it ok to say we don't support that yet?14:36
fijalI wonder14:37
Alex_Gaynorfijal: yes it's fine, but raise a NotImplementedError please14:37
fijaldon't we seem to have done most of the basics for numpy by now?14:37
Alex_Gaynorwe've made good progress yet14:37
Alex_Gaynoryes*14:37
Alex_Gaynoronce I merge dtype refactor we'll be well on our way14:37
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) joined #pypy.14:37
fijalI mean, there is tons of stuff to be done14:37
Alex_Gaynoryes14:37
antocuni (~antocuni@host142-66-static.94-94-b.business.telecomitalia.it) left irc: Quit: Leaving14:37
fijalbut important stuff14:38
lotrpy (lotrpy@202.120.58.116) left #pypy.14:40
Rhyolitebivab: ping14:44
fijalAlex_Gaynor: so?14:44
fijalAlex_Gaynor: still reviewing?14:44
Alex_Gaynorfijal: can you do something about the reduce issue, make it raiseNotImplementedError for now?14:45
fijalok14:45
bivabRhyolite: pong14:45
fijalso descr_reduce should raise notimplementederror is len(shape) > 114:45
fijalsimply?14:45
Rhyolitebivab: Did you and Sven want to have a debug session now?14:46
Alex_Gaynorfijal: something like that I suppose, but sum() should still work, somehow14:46
bivabRhyolite: sure, we are currently working on call_assembler, should we meet in #pypy-ppc14:46
Rhyoliteyes14:47
fijalpom pom pom14:48
fijalAlex_Gaynor: maybe I should just implement it :)14:48
Alex_Gaynorfijal: that also works :)14:48
fijalok14:49
witulski (~stupsi@134.99.16.20) joined #pypy.14:49
kenaan12alex_gaynor numpy-multidim 1161e9f8ceeb23 15/pypy/module/micronumpy/: match the percise error message of numpy14:49
fijalcan we merge it and I'll implement it on new branch though?14:49
Alex_Gaynorfijal: that's fine, but weshouldn't have broken features so it really should raise NotImplementedError14:50
fijalyeah, ok14:50
fijalI'll do it just now14:50
fijalit's annoying though :/14:50
kenaan12alex_gaynor numpy-multidim 111b5ac648b2de 15/pypy/module/micronumpy/test/test_numarray.py: better comment14:51
witulski (stupsi@134.99.16.20) left #pypy.14:52
fox__ (~fox@95.232.111.141) joined #pypy.14:53
`fox` (~fox@95.232.111.141) left irc: Read error: Connection reset by peer14:53
kenaan12fijal numpy-multidim 1162ddd31b5a52 15/pypy/module/micronumpy/: raise NotImplementedError in case we call reduce directly on a multidim array14:53
kenaan12fijal numpy-multidim 11005f65935f44 15/pypy/module/micronumpy/: merge14:53
Alex_Gaynorfijal: uh, this parameter really shouldn't be exposed at the python level14:53
santagadaSo you have multidim arrays now? are you guys planing on a 1.7.1 relase just for numpypy?14:55
kenaan12alex_gaynor numpy-multidim 119abe1021b1c6 15/pypy/module/micronumpy/: avoid exposing random parameters to python14:59
fijalAlex_Gaynor: true14:59
Alex_Gaynorfijal: I just fixed it14:59
fijalthanks14:59
fijalfeel like merging then?14:59
Alex_Gaynorsure14:59
asabil_ (~asabil@131.246.16.62.customer.cdi.no) joined #pypy.15:00
Alex_Gaynorfijal: what was the name of the guy who helped out?15:00
asabil (~asabil@69-217-15.connect.netcom.no) left irc: Read error: Connection reset by peer15:00
Nick change: asabil_ -> asabil15:00
fijalAlex_Gaynor: mattip?15:01
fijalthat one?15:01
Alex_Gaynoryes, thanks15:01
kenaan12alex_gaynor numpy-multidim 111bbc91ef6677 15/: close branch for merge15:02
kenaan12alex_gaynor default 11fcfe2115dc2c 15/: (fijal and mattip mostly) Implemented support for multidimensional arrays in numpypy, this support extends to...15:02
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1121 [12Alex]15:03
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1860 [12Alex]15:03
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/719 [12Alex]15:03
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/587 [12Alex]15:03
bbot2Started: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/723 [12Alex]15:03
Nick change: elliottcable -> notelliottcable15:04
fijalcool15:06
fijalAlex_Gaynor: maybe we should not merge list-strategies and numpy-multidim on one day15:06
fijalbut on the other hand should not hurt15:06
Alex_Gaynorfijal: already ran tests, so it should be ok15:06
senyai (~senyai@178.176.20.24) joined #pypy.15:07
chronitis (~chronitis@88-104-116-142.dynamic.dsl.as9105.com) joined #pypy.15:08
fijal (~fijal@41.55.94.227) left irc: Ping timeout: 240 seconds15:11
senyai (~senyai@178.176.20.24) left irc: Ping timeout: 252 seconds15:11
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1121 [12Alex]15:12
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/587 [12Alex]15:13
Sho_ (~EHS1@kde/hein) joined #pypy.15:14
kenaan12alex_gaynor default 11fd61e0e2e288 15/pypy/rpython/lltypesystem/rstr.py: fix15:23
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1122 [12Alex]15:23
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/588 [12Alex]15:23
asabil (~asabil@131.246.16.62.customer.cdi.no) left irc: Read error: Connection reset by peer15:25
asabil (~asabil@131.246.16.62.customer.cdi.no) joined #pypy.15:26
chronitis (~chronitis@88-104-116-142.dynamic.dsl.as9105.com) left irc: Quit: Lost terminal15:30
hakanardo__ (~hakan@gw.cognimatics.ideon.se) left irc: Quit: Leaving15:31
kenaan12alex_gaynor numpy-dtype-refactor 11a98e703d41e2 15/: Merged default in.15:35
fox__ (~fox@95.232.111.141) left irc: Read error: Connection reset by peer15:38
`fox` (~fox@95.232.111.141) joined #pypy.15:42
`fox` (~fox@95.232.111.141) left irc: Ping timeout: 248 seconds15:51
srin (~srin@pool-70-111-141-236.nwrk.east.verizon.net) joined #pypy.15:51
siccegge (~siccegge@faui49p.informatik.uni-erlangen.de) joined #pypy.15:55
Nick change: notelliottcable -> elliottcable15:57
kenaan12alex_gaynor numpy-dtype-refactor 11588e1b235c5c 15/pypy/module/micronumpy/test/test_zjit.py: get these tests passing again15:58
Action: siccegge is wondering why all the pypy benchmarks show it at least slightly better than CPython while the tests I ran take approx. 5 times as long to terminate on pypy compared to cpython15:58
cfbolzsiccegge: because we optimize pypy for the benchmarks. please file a bug15:59
sicceggecfbolz: ok :-)16:00
ixokai (~ixokai@pdpc/supporter/student/ixokai) left irc: Quit: Bye!16:01
ronnysiccegge: are you speaking of unittests?16:02
ronnysiccegge: keep in mind that the jit has some warmup, code thats not repeated enough is being executed with the slow interpreter16:03
sicceggeronny: I'm testing some slow script I have here (takes 20 seconds in cpython at most 1kLOC)16:06
sicceggeand running the same in pypy16:06
sicceggeso there should be plenty of time for the jit to warm up16:06
sicceggesmallish script running nearly 2 minutes in pypy16:06
justinpeel (~justinpee@70.34.6.60) joined #pypy.16:06
Alex_Gaynoryes, please file a bug for this16:07
ixokai (~ixokai@pdpc/supporter/student/ixokai) joined #pypy.16:09
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1860 [12Alex]16:10
CIA-6803siccegge 07roundup * 10#938/pypy 5 times smaller than cpython in smallish script: 16:15
CIA-68[new] rfs2model from16:15
CIA-68git clone http://vamos.informatik.uni-erlangen.de/git/vamos/16:15
CIA-68on http://vamos.cs.fau.de/~siccegge/arm.rsf16:15
CIA-68takes here (2 * ... * 14https://bugs.pypy.org/issue93816:15
justinpeel (~justinpee@70.34.6.60) left irc: Ping timeout: 276 seconds16:16
cfbolzsiccegge: thanks16:16
sicceggehope I put everything in to reproduce it :-)16:19
ixokai (~ixokai@pdpc/supporter/student/ixokai) left irc: Quit: Bye!16:20
Trundle (~andy@python/site-packages/trundle) left irc: Remote host closed the connection16:24
kenaan12alex_gaynor default 11bf59d657e73a 15/pypy/: Comment this out for now, oopspec doesn't appear to play nice with look_inside_iff16:27
arigatomov    -0xaf0100(%rbp),%r1216:27
Alex_Gaynorarigato: erm, ok :)16:30
Alex_Gaynorbtw, I have an idea about how to make str += str fast in the JIT16:30
felipecruz (b1139ec9@gateway/web/freenode/ip.177.19.158.201) joined #pypy.16:35
CIA-6803agaynor 07roundup * 10#938/pypy 5 times smaller than cpython in smallish script: 16:38
CIA-68[wontfix] The problem is you are using += to concatinate strings in a loop. CPython has16:38
CIA-68obscure hacks to make this efficient, but on PyPy this ... * 14https://bugs.pypy.org/issue93816:38
Alex_Gaynorsiccegge: Just left a comment, hope that explains it, if you have any questions let me know.16:38
cfbolzAlex_Gaynor: what's wrong with using withstrbuf?16:38
Alex_Gaynorcfbolz: well, I'm interested in exploring whether it's possible to do it purely within the optimizer of the JIT16:39
Alex_Gaynorstrbuf is basically a simplified trie right?16:39
cfbolzno16:39
cfbolzmuch simpler16:39
Alex_Gaynorah, it's just a builder16:40
cfbolzyes16:40
Alex_Gaynorok, so this is basically what I want to do, but purely within the JIT.16:40
cfbolzwhy?16:40
Alex_Gaynorthat way it keeps the mess out of Python16:40
cfbolzit's 77 lines16:41
amaury__ (amaury_@nat/google/x-pacadzfycalbktzr) joined #pypy.16:41
amaury_ (amaury_@nat/google/x-tvnlnljboxrdlhnz) left irc: Read error: Connection reset by peer16:41
Cheshirc (~Reg@unaffiliated/cheshirc) joined #pypy.16:41
Alex_Gaynorplus little bits of integration all around string{type,object}.py, but yes, I admit it's not much code16:41
Alex_GaynorI think it would be cool to do this purely in the JIT though16:41
Action: Alex_Gaynor -> awayish16:43
kkris (~kris@93-82-33-8.adsl.highway.telekom.at) joined #pypy.16:45
timonator (~timo@neo.setxkbmap.de) left irc: Quit: leaving16:46
asabil (~asabil@131.246.16.62.customer.cdi.no) left irc: Read error: Connection reset by peer16:58
asabil (~asabil@131.246.16.62.customer.cdi.no) joined #pypy.16:58
ixokai (~ixokai@pdpc/supporter/student/ixokai) joined #pypy.16:59
G2P (~G2P@fw-asn1.ornis.com) left irc: Quit: Leaving.16:59
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) left irc: Read error: Operation timed out17:01
arigatoah17:01
bearnard (~bearnard@41.3.127.64) joined #pypy.17:04
stakkars_ (~tismer@dslb-088-072-232-162.pools.arcor-ip.net) joined #pypy.17:05
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/719 [12Alex]17:09
stakkars_ (~tismer@dslb-088-072-232-162.pools.arcor-ip.net) left irc: Quit: schnarch17:14
jedie (~jens@dslb-178-009-128-058.pools.arcor-ip.net) joined #pypy.17:21
jedieI tried to install tkinter-pypy, but get errors...17:22
jedieI deinstall all tk* and tcl* packages and then install only tcl8.4-dev and tk8.4-dev...17:23
jedieon "pip install tkinter-pypy" i get the error, that "tcl.h" not found17:23
jedieBut i found it in: /usr/include/tcl8.4/tcl-private/generic/tcl.h and /usr/include/tcl8.4/tcl.h17:24
jedieany idea?17:24
jedieused pypy-c-jit-49705-0a31d8ef2f8a-linux6417:24
kenaan12arigo jit-targets 1140325623b256 15/pypy/jit/backend/test/test_random.py: Improve randomness: generates a LABEL that is not always the very first operation in the loop, but only when it...17:25
kenaan12arigo jit-targets 11a91b089925be 15/pypy/jit/backend/test/test_random.py: Fix the test.17:25
kenaan12arigo jit-targets 1155729b2290bf 15/pypy/jit/backend/x86/regalloc.py: XXX a reminder, for later17:25
bivab (~david@fwstups.cs.uni-duesseldorf.de) left irc: Quit: bivab17:25
kenaan12arigo jit-targets 11b3737393db2e 15/pypy/jit/backend/: Test and fix, laboriously.17:25
stakkars_ (~tismer@dslb-088-072-232-162.pools.arcor-ip.net) joined #pypy.17:36
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Ping timeout: 240 seconds17:40
iori (~iori@110-133-45-54.rev.home.ne.jp) left irc: Remote host closed the connection17:44
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) joined #pypy.17:49
arigatohow do I grep in a bunch of pdf files?17:50
`fox` (~fox@95.232.111.141) joined #pypy.17:51
cfbolzarigato: convert them to text first?17:53
santagadajedie, no17:53
arigatohow?17:53
jediesantagada: ?17:53
jedieno idea?17:53
santagadajedie, no idea :)17:53
cfbolzarigato: pdftotext17:53
santagadajedie, tkinter-pypy uses distutils?17:53
jediebtw. lubuntu 11.10 64Bit17:53
jedievirtualenv + pip17:54
santagadajedie, maybe pypy-tkinter is not compatible with pip or virtualenv, try setup.py install17:54
jedieok... mom17:54
arigatocfbolz: thanks17:54
santagadajedie, mom? :)17:55
jedieone moment ;)17:55
jedieI used the instrution from http://morepypy.blogspot.com/2011/04/using-tkinter-and-idle-with-pypy.html and they do: $ ./bin/easy_install tkinter-pypy17:55
jedieHm. easy_install tkinter-pypy seems to work.17:56
jedieBut now i get this error: RuntimeError: tk.h version (8.5) doesn't match libtk.a version (8.4)17:57
jedieOK, i have installed the 8.4, because i need it with normal python-tk...17:57
santagadauhmm17:57
jediei try to deinstall it again...17:57
santagadajedie, progress I guess17:57
santagadajedie, do you really want to use idle or do you want tk?17:58
jedieno idle, only tk17:58
santagadajedie, nice... maybe tkinter should be part of pypy. also it should compile with pip (maybe you can help fixing that?)17:59
jediedoes tkinter-pypy only work with tk8.5? Seems that 8.4 is the default17:59
jedieno ;)17:59
santagadajedie, are you sure you don't have tk-8.5 headers lying around? seems like it18:00
jediestep-by-step: I now deinstall tk* and tcl*18:00
jediewhich version should i install?18:00
jedietk-dev and tcl-dev would be install 8.418:00
santagadajedie, dunno, I think the error message is about a mismatch between headers and library18:01
santagadanot about needing a specific version18:01
jedieok, if i install tk-dev and tcl-dev it shoudl work?18:01
jedieOh. it installed 8.5 and not 8.4:18:03
jedie$ sudo aptitude install tcl-dev tk-dev18:03
jedie[sudo] password for jens: 18:03
jedieDie folgenden NEUEN Pakete werden zusätzlich installiert:18:03
jedie  tcl{a} tcl-dev tcl8.5{a} tcl8.5-dev{a} tk{a} tk-dev tk8.5{a} tk8.5-dev{a} 18:03
jediestrange. In the past he has installed 8.4 ...18:04
jedieok, i have done: $ sudo aptitude install python-tk tk-dev tcl-dev18:05
santagadajedie, see. so maybe you have some tcl 8.4 lib lying around18:05
justinpeel (~justinpee@70.34.6.60) joined #pypy.18:05
jedieAh.18:05
jediehave done this:18:05
jediepip uninstall tkinter-pypy18:05
jedie./easy_install tkinter-pypy18:06
jedie/usr/bin/ld: cannot find -ltk8.418:06
jedie/usr/bin/ld: cannot find -ltcl8.418:06
jedieOK, i can now try to deinstall tk stuff and install tk8.4-dev and tcl8.4-dev... Should i?18:07
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1122 [12Alex]18:07
amaury__ (amaury_@nat/google/x-pacadzfycalbktzr) left irc: Ping timeout: 244 seconds18:07
santagadajedie, yes18:07
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/588 [12Alex]18:09
jediedone18:09
jediethen i get this error:18:09
jediesrc/_tkinter.c:74:17: schwerwiegender Fehler: tcl.h: Datei oder Verzeichnis nicht gefunden18:09
jedietcl.h not found18:10
jediebut it exsist in:18:10
jedie/usr/include/tcl8.4/tcl-private/generic/tcl.h18:10
jedie/usr/include/tcl8.4/tcl.h18:10
Action: jedie in a loop ;)18:10
jedieIf i install 8.4 + 8.5 i will get: RuntimeError: tk.h version (8.5) doesn't match libtk.a version (8.4)18:11
jediebut i will try it again18:11
jediedoesn't work. Same error: tcl.h not found.18:12
jediemaybe should i recreate the pypy virtualenv?18:13
jediei send Antonio Cuni a message on bitbucket...18:18
jediein setup.py i find: libraries=["tk8.4", "tcl8.4", "X11"]18:19
jediei try to replace 8.4 with 8.5...18:19
jediesame error: tcl.h not found18:21
jedieok, i deinstall 8.4 and will try again18:21
santagadaman if it was 8.4 and was not finding tcl.h from 8.4 why changing it to 8.5 would work?18:22
santagadajedie, try only one18:22
jedieHm!18:23
jediei see this command:18:23
jediecc -fPIC -Wimplicit -DWITH_APPINIT -I/usr/include/tcl -I/usr/include/tk -I/home/jens/pypy-env/include -c src/_tkinter.c -o build/temp.linux-x86_64-2.7/src/_tkinter.o18:23
jedieBut i haven't the dir /usr/include/tk or /usr/include/tcl18:24
jediei have /usr/include/tcl8.4/ and /usr/include/tcl8.4/18:24
santagadaln -s to the rescue?18:24
santagadajedie, the thing is that debian don't create those links because the default tk is 8.518:25
Nick change: DasIch__ -> DasIch18:26
jedieNo, my error: Haven deinstall tk* tcl* and not install it again ;)18:27
jedieno i installed the default and habe links18:27
jedie...have links18:27
jedieah... now i can install it with:  ~/pypy-env/bin/pypy setup.py install18:28
santagadajedie, the link I mean is from /usr/inlcude/tk to /usr/include/tk8.418:29
jedieno it links to 8.518:29
arigato (~arigo@fwstups.cs.uni-duesseldorf.de) left irc: Quit: See you18:30
jedieHm. seems to work18:30
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) left irc: Quit: Leaving18:30
jediemaybe the patch of the setup.py libraries=["tk8.4", "tcl8.4", "X11"] >> libraries=["tk8.5", "tcl8.5", "X11"] does the job18:31
jediemaybe libraries=["tk", "tcl", "X11"] works better ???18:32
jedieYes, it works.18:33
santagadatk tcl seems better18:35
santagadabut then you have to always be sure that it can link with a bunch of different tcl/tk versions18:36
kenaan12ctismer win64-stage1 11be92adfe2b1b 15/: merge with default18:36
kenaan12ctismer win64-stage1 11df362a54e674 15/pypy/: Starting stage2 a bit by defining maxint in rarithmetic18:36
timo_ (~timo@gallifrey.wybt.net) joined #pypy.18:46
Nick change: timo_ -> Guest8217518:46
Nick change: Guest82175 -> timotimotimo18:47
Kaskuka (~John@osbk-4d08b0fc.pool.mediaWays.net) left irc: Remote host closed the connection18:50
kenaan12hakanardo jit-targets 11a831e44587ae 15/pypy/: fix test18:53
`fox` (~fox@95.232.111.141) left irc: Read error: Connection reset by peer18:53
kenaan12hakanardo jit-targets 11aeb52741cfeb 15/pypy/jit/backend/: hg merge18:53
Kaskuka (~John@osbk-4d08b0fc.pool.mediaWays.net) joined #pypy.18:55
asmeurer (~asmeurer@c-174-56-21-245.hsd1.nm.comcast.net) joined #pypy.18:56
stakkars__ (~tismer@82.113.119.142) joined #pypy.18:56
`fox` (~fox@95.232.111.141) joined #pypy.18:58
justinpeel (~justinpee@70.34.6.60) left irc: Ping timeout: 245 seconds18:59
jedie (~jens@dslb-178-009-128-058.pools.arcor-ip.net) left irc: Quit: Leaving.19:00
stakkars_ (~tismer@dslb-088-072-232-162.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds19:00
santagada (~leonardo@201.86.215.236) left irc: Ping timeout: 240 seconds19:05
kkris (~kris@93-82-33-8.adsl.highway.telekom.at) left irc: Quit: Leaving.19:06
kenaan12hakanardo jit-targets 1124e9e6fb869c 15/pypy/jit/backend/x86/assembler.py: inject debug counter after initial label if present19:07
kkris (~kris@93-82-33-8.adsl.highway.telekom.at) joined #pypy.19:09
kenaan12ctismer win64-stage1 11f6afeeb47557 15/pypy/interpreter/astcompiler/test/test_compiler.py: ignoring L formatting, unification of int and long19:10
kenaan12ctismer win64-stage1 1123ceb8882aee 15/: Merge with default19:10
kenaan12ctismer win64-stage1 11ecd85170665a 15/pypy/: merge19:10
kenaan12ctismer win64-stage1 112819522de5e0 15/pypy/: Merge19:10
fijal (~fijal@41.55.92.32) joined #pypy.19:11
`fox` (~fox@95.232.111.141) left irc: Ping timeout: 248 seconds19:14
amaury__ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.19:19
kenaan12mattip numpy-monkeyaround 112f25f1d0005c 15/: abandon tests, implementation of numpy arange, transpose19:19
mattip (4fb7732a@gateway/web/freenode/ip.79.183.115.42) joined #pypy.19:21
mattipfijal: hi19:21
fijalmattip: hi19:21
fijalmattip: I have few issues19:22
fijalwith your branch19:22
fijallike a) would be cool to support floats in arange19:22
mattipOk, I closed it.19:22
fijal?19:22
fijalwhy?19:22
fijalit's still useful19:23
mattipThe code went kind of stale with all the merges.19:23
fijalno, it's still probably mergeable19:23
mattipThanks, and thanks for the mention on the merges.19:23
fijalwell, you did some work on it no?19:24
mattipOK, the name kind of got old too.19:24
fijalI guess so :)19:24
mattipShould I try again?19:24
fijalplease :)19:24
fijalmost code is usable I think19:24
fijalarange should support floats19:24
fijaland I think None as default argument does not work that well because you'll receive space.w_None19:24
fijalbut I'm not sure, try some tests with all combinations19:25
Action: fijal off to bed19:25
mattipOK. Testing would be alot easier if we had a reshape or GetSet on shape.19:25
fijalok19:25
mattipbut shape is immutable.19:26
fijalthen it makes sense to start with reshape and GetSet on shape19:26
fijalwe can change it19:26
fijalit does not have to be immutable19:26
mattipOK, I will try to do the changes on a new branch.19:26
fijalcool, thanks19:26
mattipFWIW, my personal goal in doing this work is to get a working example of a Kalman filter in pypy as a POC.19:27
mattipSo I think I will call the branch matrixmath, OK?19:27
fijalsounds good19:27
fijalthat's a cool goal :)19:27
fijalI want to get the python-performance benchmark running19:27
fijalI need arange and flatten19:28
fijalso we're on the same page19:28
mattipgood. I was kind of wondering how to benchmark even different changesets of pypy19:29
bbot2Started: 15http://buildbot.pypy.org/builders/own-win-x86-64/builds/54 [12ctismer, win64-stage1]19:32
stakkars__there is something wrong with _pytest19:38
amaury__stakkars__: why?19:39
amaury__or: what?19:39
stakkars__INTERNALERROR>   File "d:\myslave\own-win-x86-64\build\_pytest\resultlog.py", line 84, in pytest_collectreport19:39
stakkars__INTERNALERROR>     longrepr = str(report.longrepr.reprcrash)19:39
stakkars__INTERNALERROR> AttributeError: CollectErrorRepr instance has no attribute 'reprcrash'19:39
stakkars__you can see it in stdio of my current build19:40
amaury__ah, an error during the Collect phase19:42
amaury__do you get the same error interactively?19:43
stakkars__no, this is the first time, anyway19:46
srin (~srin@pool-70-111-141-236.nwrk.east.verizon.net) left irc: Quit: leaving19:46
DasIch_ (~dasich@p3E990C09.dip.t-dialin.net) joined #pypy.19:53
DasIch (~dasich@p4FFDCA21.dip.t-dialin.net) left irc: Ping timeout: 276 seconds19:57
amaury__stakkars__: Instead of all those isinstance(length, (int, long))20:00
amaury__did you consider something like isinstance(length, r_int)20:00
amaury__(or whatever rarithmetic type is ~ to ssize_t)20:00
amaury__and force all functions that return length, hashes and converted pointers20:01
amaury__to return this r_int20:01
srin (~srin@pool-70-111-141-236.nwrk.east.verizon.net) joined #pypy.20:01
srin (~srin@pool-70-111-141-236.nwrk.east.verizon.net) left irc: Client Quit20:02
stakkars__no, I didn't think of it, yet. I was just trying to make it work.20:02
amaury__it's a good start :-)20:02
stakkars__but maybe it makes sense. I am pretty dull today, worked waaaay too long on PyPy20:03
stakkars__and need to have a break for some days20:03
stakkars__yes, this is instance tests could be reviewed and changed into something that20:04
stakkars__really says what we want to test.20:04
srin (~srin@pool-70-111-141-236.nwrk.east.verizon.net) joined #pypy.20:06
stakkars__amaury__: but please remind me, again :-)20:08
amaury__sure20:08
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.20:11
fijal (~fijal@41.55.92.32) left irc: Read error: Connection reset by peer20:14
mat^2 (~mathias@212.130.113.35) joined #pypy.20:19
felipecruz (b1139ec9@gateway/web/freenode/ip.177.19.158.201) left irc: Ping timeout: 265 seconds20:27
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection20:29
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.20:29
chronitis (~chronitis@88-104-116-142.dynamic.dsl.as9105.com) joined #pypy.20:32
stakkars__arghh, somebody used __asm :-(20:33
stakkars__this does not exist on win6420:33
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Ping timeout: 244 seconds20:34
ojii (~ojii@40-34.60-188.cust.bluewin.ch) left irc: Remote host closed the connection20:35
mitchellh (~mitchellh@c-98-245-152-34.hsd1.co.comcast.net) joined #pypy.20:35
__class__ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) left irc: Read error: Operation timed out20:39
amaury__stakkars__: indeed20:40
amaury__why is __asm used for?20:40
amaury__stakkars__: btw, I am playing with a hacked version of CPython2.720:42
stakkars__it is in mem.h20:42
stakkars__#define pypy_asm_keepalive(v)    __asm { }20:42
amaury__On Linux 64bits, but where LONG_MAX = 1<<3120:42
stakkars__amaury__: interesting! Why? Do you want to help me?20:43
amaury__yes :-)20:43
stakkars__it is really much for me, so I can use help :-)20:44
kenaan12hakanardo jit-targets 11f4b8fb7dad06 15/pypy/rlib/rsre/test/test_zjit.py: fix test20:45
stakkars__what kind of keep alive is this ask thing?20:45
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/720 [12hakanardo, jit-targets]20:45
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.20:47
amaury__stakkars__: seems to be a GC operation20:47
amaury__only Boehm and AsmGcRoot gcs need it20:47
stakkars__yeah20:48
lizardo (~lizardo@189.2.128.130) left irc: Quit: Leaving20:49
kkris (~kris@93-82-33-8.adsl.highway.telekom.at) left irc: Quit: Leaving.20:54
JaRoel|4d (~jaroel|4d@2001:980:7133:1:1d90:c622:f1bf:a63) joined #pypy.20:54
stakkars__amaury__: should I call some ask function, then?20:55
tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy.20:55
amaury__stakkars__: ask?20:56
stakkars__oupps, asm function20:56
stakkars__hmm, probably it doesn't need to be assembler, just sty. that cannot be optimized away?20:58
amaury__A call to _ReadWriteBarrier()?21:00
mitchellh (~mitchellh@c-98-245-152-34.hsd1.co.comcast.net) left irc: Quit: Computer has gone to sleep21:01
stakkars__maybe? I'll try that21:02
`fox` (~fox@95.232.111.141) joined #pypy.21:07
bbot24Failure: 15http://buildbot.pypy.org/builders/own-win-x86-64/builds/54 [12ctismer, win64-stage1]21:07
__class__ (~class@99-105-56-217.lightspeed.sntcca.sbcglobal.net) joined #pypy.21:11
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Quit: Leaving21:15
qbproger (~qbproger@c-76-104-159-248.hsd1.wa.comcast.net) joined #pypy.21:19
stakkars__amaury__: works, thanks!21:19
amaury__well, does this ReadWriteBarrier have a real effect?21:20
amaury__if all tests still pass with   #define pypy_asm_keepalive(v) NULL...21:20
stakkars__yes, it costs time ;-)21:20
stakkars__hum, if it works with NULL, then we need better tests21:21
luciani get pypy: error while loading shared libraries: libssl.so.0.9.8: cannot open shared object file: No such file or directory21:26
lucian when starting pypy21:26
kenaan12ctismer win64-stage1 111021c954f50d 15/pypy/translator/c/src/mem.h: boehm on win64 without __asm, of course :-)21:33
kenaan12ctismer win64-stage1 11a3e74b838add 15/pypy/: Merge21:33
kenaan12ctismer win64-stage1 11bd1ef1e7450f 15/pypy/jit/metainterp/history.py: removed a merge glitch21:33
kenaan12ctismer win64-stage1 119a56f7d674e7 15/pypy/rpython/test/test_rptr.py: fixed int tests to use is_valid_int21:33
amaury__lucian: that's a known issue with recent Debian IIRC21:34
lucianit worked a couple of weeks ago21:35
luciani'm on ubuntu 11.0421:35
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) left irc: Quit: Leaving.21:37
kenaan12arigo jit-targets 11b3e7310584d1 15/pypy/jit/backend/x86/regalloc.py: Fix the XXX.21:38
kenaan12arigo jit-targets 1177f8a107311c 15/pypy/jit/: translation fixes21:38
kenaan12arigo jit-targets 11c034ba0e1478 15/pypy/: merge heads21:39
lucianshould I alias openssl 1.0.0?21:39
arigato (~arigo@89.204.153.97) joined #pypy.21:40
arigatohi21:40
mitchellh (~mitchellh@c-98-245-152-34.hsd1.co.comcast.net) joined #pypy.21:40
amaury__lucian: Yes, I heard a solution like this21:42
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.21:43
derdon (~derdon@pD9E1CB8B.dip.t-dialin.net) joined #pypy.21:45
asabil (~asabil@131.246.16.62.customer.cdi.no) left irc: Ping timeout: 244 seconds21:46
cwillu_at_work (~cwillu@cwillu-1-pt.tunnel.tserv13.ash1.ipv6.he.net) left irc: Ping timeout: 244 seconds21:47
tunixman (~tunixman@2001:470:f022:200::1000) left irc: Ping timeout: 244 seconds21:47
cwillu_at_work (~cwillu@cwillu-1-pt.tunnel.tserv13.ash1.ipv6.he.net) joined #pypy.21:48
tunixman (~tunixman@irc.tunixman.com) joined #pypy.21:48
lmoura (~lmoura@177.17.97.79) left irc: Ping timeout: 260 seconds21:49
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.21:49
fox__ (~fox@95.232.111.141) joined #pypy.21:59
`fox` (~fox@95.232.111.141) left irc: Read error: Connection reset by peer21:59
ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) left irc: Quit: Leaving...22:04
bearnard (~bearnard@41.3.127.64) left irc: Ping timeout: 244 seconds22:07
fox__ (~fox@95.232.111.141) left irc: Ping timeout: 276 seconds22:11
lucianwhere would pypy look for .so files?22:14
bearnard (~bearnard@41.3.127.64) joined #pypy.22:14
lmoura (~lmoura@186.215.206.130) joined #pypy.22:16
justinpeel (~justinpee@70.34.6.60) joined #pypy.22:17
ixokai (~ixokai@pdpc/supporter/student/ixokai) left irc: Quit: Bye!22:23
stakkars__ (~tismer@82.113.119.142) left irc: Quit: schnarch22:25
Ademan (~yaaic@79.sub-97-181-24.myvzw.com) joined #pypy.22:26
Ademan (~yaaic@79.sub-97-181-24.myvzw.com) left irc: Client Quit22:27
Nick change: elliottcable -> battlecollie22:29
Nick change: battlecollie -> elliottcable22:30
voidspace (~voidspace@python/psf/voidspace) left irc: Ping timeout: 240 seconds22:33
justinpeel (~justinpee@70.34.6.60) left irc: Ping timeout: 248 seconds22:35
amaury__ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 252 seconds22:37
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/14922:41
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/95222:41
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) left irc: Remote host closed the connection22:46
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/720 [12hakanardo, jit-targets]22:46
maxyz (~maxy@186.23.74.228) left irc: Ping timeout: 244 seconds22:46
Sho_ (~EHS1@kde/hein) got netsplit.22:50
durin42 (~durin@adium/durin42) got netsplit.22:50
stakkars (~tismer@p5DDB7E68.dip.t-dialin.net) got netsplit.22:50
dD0T (~dD0T@unaffiliated/dd0t) got netsplit.22:50
tnelson (~trent@trent.me) got netsplit.22:50
qbproger (~qbproger@c-76-104-159-248.hsd1.wa.comcast.net) got netsplit.22:50
DasIch_ (~dasich@p3E990C09.dip.t-dialin.net) got netsplit.22:50
lambacck_ (~chris@d24-150-124-118.home.cgocable.net) got netsplit.22:50
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) got netsplit.22:50
mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) got netsplit.22:50
oal (u4126@gateway/web/irccloud.com/x-sbljvlxafhkylpzb) got netsplit.22:50
gsnedders (~gsnedders@mail.gsnedders.com) got netsplit.22:50
hakanardo_ (~hakan@gw.cognimatics.ideon.se) got netsplit.22:50
alexsuraci (u1962@gateway/web/irccloud.com/x-uxzloqoejhniynws) got netsplit.22:50
Alex_Gaynor (u1246@gateway/web/irccloud.com/x-dltpajofbilflxlb) got netsplit.22:50
masquerade (~rdeaton@galactica.lug.udel.edu) got netsplit.22:50
hodgestar (~quassel@gluon.za.net) got netsplit.22:50
CIA-68 (~CIA@cia.atheme.org) got netsplit.22:50
jer (~jtregunna@unaffiliated/jer) got netsplit.22:50
henning_ (~henning@unicorn.gnubo.de) got netsplit.22:50
bogner (~bogner@li325-42.members.linode.com) got netsplit.22:50
strtok (~strtok@thales.strtok.net) got netsplit.22:50
bearnard (~bearnard@41.3.127.64) got netsplit.22:50
cwillu (~cwillu@cwillu.com) got netsplit.22:50
bgola (~bgola@c9518082.virtua.com.br) got netsplit.22:50
Arnar (~arnarb@li187-148.members.linode.com) got netsplit.22:50
bbot2 (~bbot2@wyvern.cs.uni-duesseldorf.de) got netsplit.22:50
WildChild (~wildchild@modemcable165.184-57-74.mc.videotron.ca) got netsplit.22:50
chronitis (~chronitis@88-104-116-142.dynamic.dsl.as9105.com) got netsplit.22:50
timotimotimo (~timo@gallifrey.wybt.net) got netsplit.22:50
siccegge (~siccegge@faui49p.informatik.uni-erlangen.de) got netsplit.22:50
Arfrever (~Arfrever@apache/committer/Arfrever) got netsplit.22:50
ericholscher (~eric@173.203.223.63) got netsplit.22:50
dmarkey (~dmarkey@dmarkey.xen.prgmr.com) got netsplit.22:50
Garen (garen@69.76.18.3) got netsplit.22:50
Rhyolite (~dje@pool-108-6-25-71.nycmny.fios.verizon.net) got netsplit.22:50
PqSe (~joe@cafe.rspkg.com) got netsplit.22:50
Seta00 (~Seta00@seta00.com) got netsplit.22:50
GDorn (~gdorn@li223-147.members.linode.com) got netsplit.22:50
ivan` (~ivan@unaffiliated/ivan/x-000001) got netsplit.22:50
mp (maciej@lair.hell.cx) got netsplit.22:50
melinath (~melinath@li341-41.members.linode.com) got netsplit.22:50
carljm (~carljm@spinach.threepines.org) got netsplit.22:50
mitchellh (~mitchellh@c-98-245-152-34.hsd1.co.comcast.net) got netsplit.22:50
srin (~srin@pool-70-111-141-236.nwrk.east.verizon.net) got netsplit.22:50
danchr (~danchr@cl-848.chi-02.us.sixxs.net) got netsplit.22:50
[Arfrever] (~Arfrever@apache/committer/Arfrever) got netsplit.22:50
jacob22_ (~jacob@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) got netsplit.22:50
zeekay (~is@whatit.is) got netsplit.22:50
exarkun (~exarkun@wordeology.com) got netsplit.22:50
kalenz (~kalenz@ping.kalenz.fr) got netsplit.22:50
hpk (~hpk@hq2.merlinux.eu) got netsplit.22:50
asksol (~asksol@sanedefaults.net) got netsplit.22:50
ThomasWaldmann (~twaldmann@shell.moinmo.in) got netsplit.22:50
price (~price@LINERVA.MIT.EDU) got netsplit.22:50
__pv (pav@xen.ihme.org) got netsplit.22:50
getxsick (xsx@unaffiliated/getxsick) got netsplit.22:50
pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) got netsplit.22:50
ronan (~ronan@host86-142-249-87.range86-142.btcentralplus.com) got netsplit.22:50
guibou (~guibou@kraken.insecable.net) got netsplit.22:50
espes__ (~niel@205.185.120.132) got netsplit.22:50
Eventh (evenwiik@caracal.stud.ntnu.no) got netsplit.22:50
jerith (~jerith@chthon.slipgate.za.net) got netsplit.22:50
rndbit (~rndbit@vexar.de) got netsplit.22:50
idnar (~quassel@unaffiliated/idnar) got netsplit.22:50
lahwran (lahwran@unaffiliated/lahwran) got netsplit.22:50
Ormod (~valtha@ohmu.fi) got netsplit.22:50
mattip (4fb7732a@gateway/web/freenode/ip.79.183.115.42) got netsplit.22:50
mat^2 (~mathias@212.130.113.35) got netsplit.22:50
asmeurer (~asmeurer@c-174-56-21-245.hsd1.nm.comcast.net) got netsplit.22:50
hakanardo (~hakan@h59ec009e.seluldx.dyn.perspektivbredband.net) got netsplit.22:50
jml (~jml@canonical/launchpad/jml) got netsplit.22:50
kenaan (~kenaan@cube.twistedmatrix.com) got netsplit.22:50
dibb (~dibb@schonenberg.org) got netsplit.22:50
htoothrot (~mux@unaffiliated/htoothrot) got netsplit.22:50
Husio (husiatyn@oceanic.wsisiz.edu.pl) got netsplit.22:50
shawn_ (~shawn@208-78-98-92.slicehost.net) got netsplit.22:50
nomadium (miguel@regine.miguel.cc) got netsplit.22:50
MostAwesomeDude (~simpson@osuosl/staff/mostawesomedude) got netsplit.22:50
arigato (~arigo@89.204.153.97) left irc: Quit: See you22:50
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds22:52
bearnard (~bearnard@41.3.127.64) returned to #pypy.22:55
mitchellh (~mitchellh@c-98-245-152-34.hsd1.co.comcast.net) returned to #pypy.22:55
chronitis (~chronitis@88-104-116-142.dynamic.dsl.as9105.com) returned to #pypy.22:55
mat^2 (~mathias@212.130.113.35) returned to #pypy.22:55
srin (~srin@pool-70-111-141-236.nwrk.east.verizon.net) returned to #pypy.22:55
mattip (4fb7732a@gateway/web/freenode/ip.79.183.115.42) returned to #pypy.22:55
asmeurer (~asmeurer@c-174-56-21-245.hsd1.nm.comcast.net) returned to #pypy.22:55
timotimotimo (~timo@gallifrey.wybt.net) returned to #pypy.22:55
siccegge (~siccegge@faui49p.informatik.uni-erlangen.de) returned to #pypy.22:55
cwillu (~cwillu@cwillu.com) returned to #pypy.22:55
getxsick (xsx@unaffiliated/getxsick) returned to #pypy.22:55
danchr (~danchr@cl-848.chi-02.us.sixxs.net) returned to #pypy.22:55
hakanardo (~hakan@h59ec009e.seluldx.dyn.perspektivbredband.net) returned to #pypy.22:55
bgola (~bgola@c9518082.virtua.com.br) returned to #pypy.22:55
pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) returned to #pypy.22:55
jml (~jml@canonical/launchpad/jml) returned to #pypy.22:55
[Arfrever] (~Arfrever@apache/committer/Arfrever) returned to #pypy.22:55
jacob22_ (~jacob@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) returned to #pypy.22:55
lahwran (lahwran@unaffiliated/lahwran) returned to #pypy.22:55
zeekay (~is@whatit.is) returned to #pypy.22:55
Arfrever (~Arfrever@apache/committer/Arfrever) returned to #pypy.22:55
ronan (~ronan@host86-142-249-87.range86-142.btcentralplus.com) returned to #pypy.22:55
ericholscher (~eric@173.203.223.63) returned to #pypy.22:55
exarkun (~exarkun@wordeology.com) returned to #pypy.22:55
Arnar (~arnarb@li187-148.members.linode.com) returned to #pypy.22:55
kalenz (~kalenz@ping.kalenz.fr) returned to #pypy.22:55
kenaan (~kenaan@cube.twistedmatrix.com) returned to #pypy.22:55
bbot2 (~bbot2@wyvern.cs.uni-duesseldorf.de) returned to #pypy.22:55
WildChild (~wildchild@modemcable165.184-57-74.mc.videotron.ca) returned to #pypy.22:55
espes__ (~niel@205.185.120.132) returned to #pypy.22:55
dmarkey (~dmarkey@dmarkey.xen.prgmr.com) returned to #pypy.22:55
guibou (~guibou@kraken.insecable.net) returned to #pypy.22:55
Garen (garen@69.76.18.3) returned to #pypy.22:55
Eventh (evenwiik@caracal.stud.ntnu.no) returned to #pypy.22:55
dibb (~dibb@schonenberg.org) returned to #pypy.22:55
Rhyolite (~dje@pool-108-6-25-71.nycmny.fios.verizon.net) returned to #pypy.22:55
hpk (~hpk@hq2.merlinux.eu) returned to #pypy.22:55
htoothrot (~mux@unaffiliated/htoothrot) returned to #pypy.22:55
carljm (~carljm@spinach.threepines.org) returned to #pypy.22:55
melinath (~melinath@li341-41.members.linode.com) returned to #pypy.22:55
ivan` (~ivan@unaffiliated/ivan/x-000001) returned to #pypy.22:55
Ormod (~valtha@ohmu.fi) returned to #pypy.22:55
GDorn (~gdorn@li223-147.members.linode.com) returned to #pypy.22:55
idnar (~quassel@unaffiliated/idnar) returned to #pypy.22:55
mp (maciej@lair.hell.cx) returned to #pypy.22:55
rndbit (~rndbit@vexar.de) returned to #pypy.22:55
__pv (pav@xen.ihme.org) returned to #pypy.22:55
#pypy: mode change '+o exarkun ' by adams.freenode.net22:55
price (~price@LINERVA.MIT.EDU) returned to #pypy.22:55
ThomasWaldmann (~twaldmann@shell.moinmo.in) returned to #pypy.22:55
jerith (~jerith@chthon.slipgate.za.net) returned to #pypy.22:55
PqSe (~joe@cafe.rspkg.com) returned to #pypy.22:55
Seta00 (~Seta00@seta00.com) returned to #pypy.22:55
asksol (~asksol@sanedefaults.net) returned to #pypy.22:55
Husio (husiatyn@oceanic.wsisiz.edu.pl) returned to #pypy.22:55
shawn_ (~shawn@208-78-98-92.slicehost.net) returned to #pypy.22:55
nomadium (miguel@regine.miguel.cc) returned to #pypy.22:55
MostAwesomeDude (~simpson@osuosl/staff/mostawesomedude) returned to #pypy.22:55
qbproger (~qbproger@c-76-104-159-248.hsd1.wa.comcast.net) returned to #pypy.22:56
DasIch_ (~dasich@p3E990C09.dip.t-dialin.net) returned to #pypy.22:56
lambacck_ (~chris@d24-150-124-118.home.cgocable.net) returned to #pypy.22:56
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) returned to #pypy.22:56
mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) returned to #pypy.22:56
oal (u4126@gateway/web/irccloud.com/x-sbljvlxafhkylpzb) returned to #pypy.22:56
jer (~jtregunna@unaffiliated/jer) returned to #pypy.22:56
gsnedders (~gsnedders@mail.gsnedders.com) returned to #pypy.22:56
hakanardo_ (~hakan@gw.cognimatics.ideon.se) returned to #pypy.22:56
alexsuraci (u1962@gateway/web/irccloud.com/x-uxzloqoejhniynws) returned to #pypy.22:56
Alex_Gaynor (u1246@gateway/web/irccloud.com/x-dltpajofbilflxlb) returned to #pypy.22:56
masquerade (~rdeaton@galactica.lug.udel.edu) returned to #pypy.22:56
hodgestar (~quassel@gluon.za.net) returned to #pypy.22:56
CIA-68 (~CIA@cia.atheme.org) returned to #pypy.22:56
strtok (~strtok@thales.strtok.net) returned to #pypy.22:56
henning_ (~henning@unicorn.gnubo.de) returned to #pypy.22:56
bogner (~bogner@li325-42.members.linode.com) returned to #pypy.22:56
#pypy: mode change '+o Alex_Gaynor ' by adams.freenode.net22:56
maxyz (~maxy@186.23.74.228) joined #pypy.22:56
Sho_ (~EHS1@kde/hein) returned to #pypy.22:56
durin42 (~durin@adium/durin42) returned to #pypy.22:56
stakkars (~tismer@p5DDB7E68.dip.t-dialin.net) returned to #pypy.22:56
dD0T (~dD0T@unaffiliated/dd0t) returned to #pypy.22:56
tnelson (~trent@trent.me) returned to #pypy.22:56
stakkars_ (~tismer@p5DDB7E68.dip.t-dialin.net) joined #pypy.22:59
stakkars (~tismer@p5DDB7E68.dip.t-dialin.net) left irc: Read error: Connection reset by peer22:59
Nick change: stakkars_ -> stakkars22:59
ixokai (~ixokai@pdpc/supporter/student/ixokai) joined #pypy.23:01
JaRoel|4d (~jaroel|4d@2001:980:7133:1:1d90:c622:f1bf:a63) left irc: Remote host closed the connection23:03
mitchellh (~mitchellh@c-98-245-152-34.hsd1.co.comcast.net) left irc: Quit: Computer has gone to sleep23:10
lmatos (lmatos@staff.linode.com) joined #pypy.23:10
kenaan12ctismer win64-stage1 117c4368db5b47 15/pypy/: unification of int/long. Extended is_valid_int by parameter 'force_type=True'. This function should be used ...23:12
ixokai (~ixokai@pdpc/supporter/student/ixokai) left irc: Quit: Bye!23:19
voidspace (~voidspace@87-194-212-65.bethere.co.uk) joined #pypy.23:25
voidspace (~voidspace@87-194-212-65.bethere.co.uk) left irc: Changing host23:25
voidspace (~voidspace@python/psf/voidspace) joined #pypy.23:25
pjenveystakkars - 'force_type_type' looks like a typo23:28
stakkarspjenvey: where did you see this?23:30
stakkarsah, f**k23:31
kenaan12ctismer win64-stage1 118b591290caf9 15/pypy/rlib/rarithmetic.py: typo-typopo23:33
mat^2 (~mathias@212.130.113.35) left irc: 23:34
chronitis (~chronitis@88-104-116-142.dynamic.dsl.as9105.com) left irc: Quit: Lost terminal23:37
voidspace (~voidspace@python/psf/voidspace) left irc: Quit: Uhm... gotta go23:42
mat^2 (~mathias@212.130.113.35) joined #pypy.23:51
JStoker (jstoker@unaffiliated/jstoker) left irc: Ping timeout: 240 seconds23:52
bbot24Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/723 [12Alex]23:54
bbot2Started: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/72423:54
JStoker (jstoker@unaffiliated/jstoker) joined #pypy.23:56
voidspace (~anonymous@87-194-212-65.bethere.co.uk) joined #pypy.23:58
voidspace (~anonymous@87-194-212-65.bethere.co.uk) left irc: Changing host23:58
voidspace (~anonymous@python/psf/voidspace) joined #pypy.23:58
zain (~textual@c-24-130-236-153.hsd1.ca.comcast.net) joined #pypy.23:59
--- Fri Nov 25 201100:00

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