#pypy IRC log for Friday, 2011-11-18

lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds00:04
derdon (~derdon@p5DE8BFF2.dip.t-dialin.net) left irc: Remote host closed the connection00:18
aurojit (~apanda@dhcp-44-233.EECS.Berkeley.EDU) left irc: Remote host closed the connection00:20
fenrrir (~fenrrir@187.40.173.201) left irc: Quit: Saindo00:23
stakkars_ (~tismer@dslb-088-072-225-154.pools.arcor-ip.net) joined #pypy.00:25
stakkars__ (~tismer@p5DDB7E68.dip.t-dialin.net) joined #pypy.00:26
apanda (~apanda@dhcp-44-233.EECS.Berkeley.EDU) joined #pypy.00:26
stakkars_ (~tismer@dslb-088-072-225-154.pools.arcor-ip.net) left irc: Ping timeout: 240 seconds00:29
mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) joined #pypy.00:30
JaredWigmore (~jared@061093200075.ctinets.com) joined #pypy.00:33
bobbyz (~bobbyz@184-229-210-255.pools.spcsdns.net) joined #pypy.00:40
bobbyz_ (~bobbyz@107.24.205.253) joined #pypy.00:43
bobbyz (~bobbyz@184-229-210-255.pools.spcsdns.net) left irc: Ping timeout: 260 seconds00:47
kenaan12mattip numpy-multidim-shards 111be076ca0053 15/pypy/module/micronumpy/: repr and str pass tests00:49
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 244 seconds00:52
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/14201:00
tlynn (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Ping timeout: 240 seconds01:01
dmsk (dmsk@62.61.145.89.generic-hostname.arrownet.dk) left #pypy ("Leaving").01:06
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection01:14
JaredWigmore (~jared@061093200075.ctinets.com) left irc: Ping timeout: 240 seconds01:19
bbot24Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/71501:27
bobbyz_ (~bobbyz@107.24.205.253) left irc: Ping timeout: 260 seconds01:27
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep01:32
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) joined #pypy.01:34
Nick change: Gulaway -> Gulopine01:37
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/184801:43
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/57701:43
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/26901:43
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/61501:43
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/111101:43
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/80001:43
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/144201:43
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/70901:43
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/26701:43
bbot213Exception: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/26701:43
bobbyz_ (~bobbyz@107.24.205.253) joined #pypy.01:59
zain__ (~textual@c-67-160-201-63.hsd1.ca.comcast.net) left irc: Quit: Textual IRC Client: http://www.textualapp.com/02:00
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep02:14
DasIch (~dasich@p4FFDCC7D.dip.t-dialin.net) left irc: Ping timeout: 260 seconds02:14
canta (~canta@77-20-123-240-dynip.superkabel.de) left irc: Quit: Leaving.02:24
DasIch (~dasich@p4FFDFDC2.dip.t-dialin.net) joined #pypy.02:26
Arach (~arach@95-30-210-190.broadband.corbina.ru) left irc: Read error: Connection reset by peer02:28
espes__ (~niel@205.185.120.132) joined #pypy.02:30
bobbyz_ (~bobbyz@107.24.205.253) left irc: Ping timeout: 258 seconds02:31
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/94502:33
Guest_ (~textual@c-71-197-238-154.hsd1.wa.comcast.net) joined #pypy.02:42
Nick change: Guest_ -> rekamso02:42
Arach (~arach@2.95.24.5) joined #pypy.02:43
roadt_ (~roadt@60.168.94.55) joined #pypy.02:45
Shanita (~John@osbk-4d08bcbb.pool.mediaWays.net) left irc: Ping timeout: 240 seconds02:47
Shanita (~John@osbk-4db14933.pool.mediaWays.net) joined #pypy.02:47
roadt_hi, does pypy monitor the changes of original script  and update the cache of native code on the fly ? 02:49
Alex_Gaynorno02:50
roadt_try to understand the pypy jit detail, any manual and link is good for me..02:50
roadt_it just gives a cleaning cache at every startup?02:50
Alex_GaynorYes, pypy's execution model is similar to CPython in that respect, it loads your code and runs it.02:51
Alex_Gaynorit just happens to run it faster with the JIT02:51
roadt_Alex_Gaynor, conciseness and accurate answer, thanks.02:52
roadt_s/conciseness/concise/02:53
Alex_Gaynornp02:55
espes_ (~espes@123-243-221-248.tpgi.com.au) joined #pypy.02:59
Rhyolitehow does one run test_ll_random?03:00
Alex_Gaynor./pytest.py pyp/path/to/test_ll_random.py03:01
espes (~espes@123-243-221-248.tpgi.com.au) left irc: Ping timeout: 244 seconds03:01
Nick change: espes_ -> espes03:01
Rhyolitewhen do tests need a harness like x86/test/test_runner.py for runner_test.py?03:04
rekamso (~textual@c-71-197-238-154.hsd1.wa.comcast.net) left irc: Quit: Textual IRC Client: http://www.textualapp.com/03:30
espes (~espes@123-243-221-248.tpgi.com.au) left irc: Remote host closed the connection03:30
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/184803:32
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/144204:27
Da_Blitz (~Da_Blitz@203.56.250.63) joined #pypy.04:28
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/80004:28
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/61504:35
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/111104:46
d1b (~db@d1b.org) left irc: Remote host closed the connection04:53
apanda (~apanda@dhcp-44-233.EECS.Berkeley.EDU) left irc: Remote host closed the connection04:57
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/57704:57
AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) joined #pypy.04:59
mat^2 (~mathias@212.130.113.35) left irc: Ping timeout: 244 seconds05:00
AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) left irc: Client Quit05:00
kushal (kdas@fedora/kushal) joined #pypy.05:07
xorAxAx (~alexander@moinmoin/coreteam/alexander) left irc: Excess Flood05:07
xorAxAx (~alexander@moinmoin/coreteam/alexander) joined #pypy.05:09
AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) joined #pypy.05:30
Nick change: Gulopine -> Gulaway05:39
Nick change: Gulaway -> Gulopine05:40
mat^2 (~mathias@212.130.113.35) joined #pypy.05:42
AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) left irc: Quit: Leaving.05:51
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/70905:55
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.06:04
bobbyz_ (~bobbyz@107.24.205.253) joined #pypy.06:05
iori (~iori@110-133-45-54.rev.home.ne.jp) left irc: Read error: Connection reset by peer06:24
Nick change: Gulopine -> Gulaway06:24
iori (~iori@110-133-45-54.rev.home.ne.jp) joined #pypy.06:24
mat^2 (~mathias@212.130.113.35) left irc: 06:29
tilgovi (~randall@c-98-210-155-124.hsd1.ca.comcast.net) joined #pypy.06:31
tilgovi (~randall@c-98-210-155-124.hsd1.ca.comcast.net) left irc: Changing host06:31
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.06:31
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep06:47
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.06:50
asmeurer__ (~asmeurer@dhcp-baca-230.resnet.nmt.edu) left irc: Quit: asmeurer__06:53
canta (~canta@77-20-123-240-dynip.superkabel.de) joined #pypy.06:56
fschwidom (~fschwidom@46.115.1.48) joined #pypy.07:02
fijal (~fijal@197.172.244.192) joined #pypy.07:07
fijalhi07:07
antocuni (~antocuni@host200-125-dynamic.25-79-r.retail.telecomitalia.it) joined #pypy.07:10
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.07:10
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep07:31
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 248 seconds07:32
stakkars__ (~tismer@p5DDB7E68.dip.t-dialin.net) left irc: Quit: schnarch07:33
bgola (~bgola@189.100.46.89) left irc: Ping timeout: 255 seconds07:35
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 240 seconds07:50
espes (~espes@123-243-221-248.tpgi.com.au) joined #pypy.07:58
kushal (kdas@fedora/kushal) left irc: Ping timeout: 252 seconds07:59
kushal (kdas@fedora/kushal) joined #pypy.08:04
Trundle (~andy@p578bfdcf.dip0.t-ipconnect.de) joined #pypy.08:06
Trundle (~andy@p578bfdcf.dip0.t-ipconnect.de) left irc: Changing host08:06
Trundle (~andy@python/site-packages/trundle) joined #pypy.08:06
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.08:12
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds08:16
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) joined #pypy.08:22
Vorpal (~AnMaster@unaffiliated/anmaster) joined #pypy.08:22
verte (~verte@python/site-packages/verte) joined #pypy.08:28
arigato (~arigo@82.113.119.33) joined #pypy.08:29
arigatohi08:30
verteArmin!08:31
vertehi.08:31
fijalhi armin08:32
canta (~canta@77-20-123-240-dynip.superkabel.de) left irc: Quit: Leaving.08:32
arigatofijal: the buildbots are green and happy08:34
arigatoso now would be an appropriate time to merge into the release branch08:34
fijalcool08:35
fijalshould the --jit off disabling all jits go there?08:35
fijalit's almost done, I'm running tests08:35
arigatoah, cool too08:36
arigatoyes, I guess so08:36
arigatobah, the Windows buildbot just crashed before even starting08:36
Action: arigato tries to kick it08:36
fijalwe'll run tests on the branch anyway08:36
fijalanother thing we have to do is numpy -> numpypy rename08:36
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/268 [12arigo]08:37
bbot213Exception: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/268 [12arigo]08:37
arigatono08:37
arigatonumpypy: ok08:37
arigatoI'm still a bit confused by the argument, but have no preference for numpy vs numpypy08:38
fijalthe argument is that things that optionally import numpy08:38
fijalwould be confused by semi-working implementation08:38
arigatoah, I see08:38
arigatook08:38
fijalso for now we have to name it something else08:38
fijalalso, we were kindly asked to do so08:38
fijaluntil we have a working version08:39
arigatosubprocess.check_call("TASKKILL /F /PID %s /T" % self.process.pid)08:40
arigatosubprocess.CalledProcessError: Command 'TASKKILL /F /PID None /T' returned non-zero exit status 12808:40
arigatoin other words, self.process.pid was None08:40
fijalnice08:42
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/269 [12arigo]08:45
CIA-6803arigo 07roundup * 10#928/syslog.openlog behavior different from CPython: 08:49
CIA-68Fixed in b6585727903b, where I followed the source of CPython more closely.08:49
CIA-68Unfortunately it seems that there are no tests for syslog. Alexband ... * 14https://bugs.pypy.org/issue92808:49
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy.08:51
fijalarigato: I guess on_compile_bridge and friends should be moved as well08:56
verte (~verte@python/site-packages/verte) left irc: Quit: Yaaic - Yet another Android IRC client - http://www.yaaic.org08:56
fijalbut since noone uses it, it's slightly less relevant08:56
Action: arigato uses "hg rebase" (argh! again), with the goal of rebasing a number of checkins from "default" to a branch08:58
arigatoseems to work08:58
arigatofijal: yes08:58
kenaan12fijal jitdriver-setparam-all 11f121b8ac7363 15/pypy/: Change set_param API. Now we run set_param(driver-or-None, 'name', value) instead of driver.set_para...08:59
fijalarigato: feel free to review the branch08:59
kenaan12arigo op_malloc_gc 1162ea37257eb5 15/: A branch to simplify the backend interface: instead of the various new_xxx resoperations, there is only one ma...09:00
kenaan12arigo op_malloc_gc 11a85153701bb6 15/pypy/jit/: Attempt to collapse several NEW-like operations into a single simpler MALLOC_GC operation: starting...09:00
kenaan12arigo op_malloc_gc 1140d91227555d 15/pypy/jit/backend/llsupport/: In-progress.09:00
kenaan12arigo op_malloc_gc 118b49822025c6 15/pypy/jit/backend/llsupport/: Fix some tests.09:00
kenaan12arigo op_malloc_gc 11b5fdf3f1574e 15/pypy/jit/backend/llsupport/test/test_gc.py: Clean up the tests.09:00
kenaan12arigo op_malloc_gc 11c4bcb5c2b12f 15/pypy/jit/backend/llsupport/: Missing rounding up: tests and fix.09:00
kenaan12arigo op_malloc_gc 11454b0eb8af85 15/pypy/jit/backend/llsupport/: Test and fix for allocating tiny structures --- they are smaller than the minimal_size_in_nursery.09:00
kenaan12arigo op_malloc_gc 11cf80cbdf1c65 15/pypy/jit/backend/llsupport/descr.py: Kill unused attribute.09:00
fijalarigato: tests seem to pass09:04
kenaan12fijal default 11f732c0996244 15/pypy/module/micronumpy/: rename numpy -> numpypy09:09
k_bx (~k_bx@195.20.130.1) joined #pypy.09:14
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/26909:17
fijalarigato: ping?09:20
arigatopong09:20
arigatoreviewing09:21
fijalok, cool09:21
stakkars_ (~tismer@82.113.99.63) joined #pypy.09:23
fijalarigato: do you remember by chance how do I add people to a group of contributors?09:24
fijalgot it09:24
mwhudson (~mwh@linaro/mwhudson) left irc: Ping timeout: 252 seconds09:26
kenaan12arigo jitdriver-setparam-all 1173172e1ee46c 15/pypy/jit/metainterp/warmspot.py: Re-add an assert that was lost.09:26
arigatook, if "translate.py -Ojit" still works then fine09:27
arigatoI'm not absolutely sure it does because of annotator subtleties :-)09:27
CIA-6803alexband 07roundup * 10#928/syslog.openlog behavior different from CPython: 09:27
CIA-68It seems work with me. I just run my application and I haven't seen exception with09:27
CIA-68syslog appear again. * 14https://bugs.pypy.org/issue92809:27
CIA-6803arigo 07roundup * 10#928/syslog.openlog behavior different from CPython: [resolved] Thanks :-) * 14https://bugs.pypy.org/issue92809:28
fijalarigato: should I merge, merge to the release branch and kick buildbots simply?09:31
G2P (~G2P@fw-asn1.ornis.com) joined #pypy.09:31
fijalsince we're going to run the release buildbots anyway09:31
arigatook09:32
kenaan12fijal jitdriver-setparam-all 11310e106460c0 15/: close about to be merged branch09:34
kenaan12fijal default 1142d55fa4ac69 15/pypy/: Merge jitdriver-setparam-all branch that allows setting parameters to all jitdrivers.09:34
fijalhave you seen the new pypy website btw?09:35
kenaan12fijal release-1.7.x 117773f8fc4223 15/: merge in default09:35
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1849 [12fijal, release-1.7.x]09:36
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/710 [12fijal, release-1.7.x]09:36
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-win-x86-32/builds/538 [12fijal, release-1.7.x]09:36
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1112 [12fijal, release-1.7.x]09:36
arigatoYet Another(tm)? :-)09:36
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/578 [12fijal, release-1.7.x]09:36
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/270 [12fijal, release-1.7.x]09:36
fijalno, the same one I believe09:36
fijalhttp://pypy-site.ep.io/09:36
arigatowell yes, Yet Another from my point of view09:37
fijalok :)09:37
fijalyet another is not a very well defined term obviously09:38
fijalwithout full knowledge to what you consider the same09:38
arigatoyou'll have to announce it to pypy-dev, at least, and discuss why you are proposing a redesign of pypy.org09:39
fijalwell, because it's better I suppose09:43
arigatosorry, I'm part of the wrong audiance to argue :-)09:44
bivab (~david@fwstups.cs.uni-duesseldorf.de) joined #pypy.09:44
k_bx (~k_bx@195.20.130.1) left irc: Ping timeout: 252 seconds09:44
Action: fijal sent mail to pypy-dev09:44
arigatoas far as I'm concerned, I've got used to the current pypy.org, so yours looks to me like it's full of strange dynamic html09:45
fijalasking for opinions is usually fine09:45
arigatobut I'm sure that I can get used to the new one too09:45
fijal:-)09:45
fijalyes, you're the wrong kind of audience09:45
k_bx (~k_bx@195.20.130.1) joined #pypy.09:45
ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) joined #pypy.09:46
fijalarigato: so jittargets are supposed to fix my weirdness with new_arrays right?09:47
Action: fijal writes the release announcement then09:47
fijalwe need a codename!09:48
fijalI think we can just copy-paste the previous release announcement09:49
fijalWe're pleased to announce the 1.6 release of PyPy. This release brings a lot09:49
fijalof bugfixes and performance improvements over 1.5, and improves support for09:49
fijalWindows 32bit and OS X 64bit. This version fully implements Python 2.7.1 and09:49
fijalhas beta level support for loading CPython C extensions.  You can download it09:49
fijalhere:09:49
fijal (~fijal@197.172.244.192) left irc: Read error: Connection reset by peer09:53
timonatori still don't see an obvious link to the compatibility wiki :\09:53
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) left irc: Quit: Leaving09:58
fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) left irc: Quit: fzzzy10:04
kenaan12fijal default 117c06e045a439 15/pypy/doc/release-1.7.0.rst: what I remember that goes into the release10:05
fijal (~fijal@41.54.232.97) joined #pypy.10:06
bbot24Finished ??: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/270 [12fijal, release-1.7.x]10:07
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/271 [12fijal, release-1.7.x]10:07
fijalcrash?10:09
fijalslave lost10:10
fijalnice10:10
k_bx (~k_bx@195.20.130.1) left irc: Ping timeout: 240 seconds10:16
derdon (~derdon@p5DE8AE3D.dip.t-dialin.net) joined #pypy.10:22
fijal (~fijal@41.54.232.97) left irc: Read error: Connection reset by peer10:24
EnCuKou (~encukou@a91-156-243-246.elisa-laajakaista.fi) left irc: Remote host closed the connection10:30
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1849 [12fijal, release-1.7.x]10:31
arigatoah, that's only a missing fix in a test10:33
k_bx (~k_bx@195.20.130.1) joined #pypy.10:33
k_bx (~k_bx@195.20.130.1) left irc: Remote host closed the connection10:33
ojii (~ojii@40-34.60-188.cust.bluewin.ch) joined #pypy.10:37
stakkars__ (~tismer@89.204.155.98) joined #pypy.10:37
stakkars__ (~tismer@89.204.155.98) left irc: Remote host closed the connection10:38
stakkars__ (~tismer@89.204.155.98) joined #pypy.10:38
stakkars_ (~tismer@82.113.99.63) left irc: Ping timeout: 240 seconds10:39
voidspace (~anonymous@python/psf/voidspace) joined #pypy.10:46
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.10:50
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.10:50
roadt_ (~roadt@60.168.94.55) left irc: Remote host closed the connection11:00
rokujyouhitoma (~rokujyouh@p5091-ipad03sizuokaden.shizuoka.ocn.ne.jp) joined #pypy.11:03
stakkars__ (~tismer@89.204.155.98) left irc: Quit: schnarch11:11
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/710 [12fijal, release-1.7.x]11:25
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection11:25
`fox` (~fox@host8-107-dynamic.16-79-r.retail.telecomitalia.it) joined #pypy.11:40
arigato (~arigo@82.113.119.33) left irc: Quit: See you11:47
voidspace_ (~voidspace@python/psf/voidspace) joined #pypy.11:50
voidspace_ (~voidspace@python/psf/voidspace) left irc: Client Quit11:50
cfbolz (~cfbolz@dslb-178-003-008-145.pools.arcor-ip.net) joined #pypy.11:53
iori (~iori@110-133-45-54.rev.home.ne.jp) left irc: Remote host closed the connection12:01
lizardo (~lizardo@189.2.128.130) joined #pypy.12:06
`fox` (~fox@host8-107-dynamic.16-79-r.retail.telecomitalia.it) left irc: Ping timeout: 276 seconds12:15
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1112 [12fijal, release-1.7.x]12:21
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.12:21
kenaan12cfbolz set-strategies 11617ce44795c0 15/pypy/objspace/std/test/test_setobject.py: uh? this test is clearly dict order dependent12:23
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/578 [12fijal, release-1.7.x]12:24
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Read error: Operation timed out12:30
rokujyouhitoma (~rokujyouh@p5091-ipad03sizuokaden.shizuoka.ocn.ne.jp) left irc: Remote host closed the connection12:30
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.12:32
PiotrSikora (~none@nginx/adept/piotrsikora) left irc: Excess Flood12:41
PiotrSikora (~none@nginx/adept/piotrsikora) joined #pypy.12:42
JaredWigmore (~jared@061093200075.ctinets.com) joined #pypy.12:45
arigato (~arigo@fwstups.cs.uni-duesseldorf.de) joined #pypy.12:59
mattip (~chatzilla@109.64.118.235) joined #pypy.13:03
mattipfijal: ping13:05
fijal (~fijal@41.52.35.4) joined #pypy.13:08
mattip (chatzilla@109.64.118.235) left #pypy.13:08
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Ping timeout: 248 seconds13:10
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) joined #pypy.13:11
mattip (~chatzilla@bzq-109-64-118-235.red.bezeqint.net) joined #pypy.13:12
fijalmattip: hi13:13
mattipfijal: the zjit test failures ont he numpy-multidim-shards branch will only affect performance, or will they break translation?13:13
fijalthey'll mostly affect performance13:13
fijalbut it's kinda crucial13:13
fijalwe need to fix some of them13:13
mattipGive me a hint of what to play with? What commonly causes loops like that?13:15
fijaloh13:15
fijalI don't think you can help right now13:15
fijalwith those tests13:15
fijalyou can think how to do broadcasting13:15
mattipOr would it be easier to just wait?13:16
fijalwith test_zjit13:17
fijal?13:17
fijalit's missing some other branch jit-targets to work13:17
mattipOK. More "if ..." magic in Call2.eval might do the trick13:18
fijalno13:18
fijalor not really13:18
fijalplease wait a bit with test_zjit13:18
fijalthink about broadcasting13:18
fijaldoes repr/str work?13:18
rokujyouhitoma (~rokujyouh@p5091-ipad03sizuokaden.shizuoka.ocn.ne.jp) joined #pypy.13:19
mattipSorry, too much lag, my comment was how to implement broadcasting.13:19
fijalalso look how it behaves if size > 10000 for larger dimensions13:19
fijalah13:19
mattipDone, test for that situation added.13:19
fijalI don't think more ifs13:19
fijalbut a different iterator maybe?13:19
fijal (~fijal@41.52.35.4) left irc: Read error: Connection reset by peer13:19
rokujyouhitoma (~rokujyouh@p5091-ipad03sizuokaden.shizuoka.ocn.ne.jp) left irc: Read error: Connection reset by peer13:20
rokujyouhitoma (~rokujyouh@p5091-ipad03sizuokaden.shizuoka.ocn.ne.jp) joined #pypy.13:20
mattipOK, I'll try something. We are agreed that the place to put this is Call2.eval ?13:20
mattipAhh, if it's an iterator it would be in the bin_func creation.13:21
Action: mattip letting fingers get ahead of brain13:21
stakkars_ (~tismer@i59F77D6B.versanet.de) joined #pypy.13:24
getxsick (xsx@unaffiliated/getxsick) joined #pypy.13:25
kenaan12l.diekmann type-specialized-instances 11b16fead0c3f3 15/pypy/objspace/std/: introduced IntAttribute13:25
kenaan12l.diekmann type-specialized-instances 1126d4cedd131b 15/pypy/objspace/std/test/test_mapdict.py: removed some old prints13:25
kenaan12l.diekmann type-specialized-instances 1151f420b3af95 15/pypy/objspace/std/mapdict.py: fix: former PlainAttribute is now AbstractStoredAttribute13:25
kenaan12l.diekmann type-specialized-instances 11ca3e806187f6 15/pypy/objspace/std/mapdict.py: fixed read with new selector (still not sure if this is the right fix)13:25
kenaan12l.diekmann type-specialized-instances 116f9cd7e5bdeb 15/pypy/objspace/std/test/test_mapdict.py: started fixing tests to work with new selector13:25
kenaan12l.diekmann type-specialized-instances 11c76cccda3d75 15/pypy/objspace/std/test/test_mapdict.py: removed some old debug prints13:26
mattip (chatzilla@bzq-109-64-118-235.red.bezeqint.net) left #pypy.13:29
Rhy0lite (dje@nat/ibm/x-rhiolehszwlqjzuy) joined #pypy.13:37
voidspace (~anonymous@python/psf/voidspace) left irc: Quit: voidspace13:40
k_bx (~k_bx@195.20.130.1) joined #pypy.13:52
pingveno (adevore@ruby.cat.pdx.edu) left irc: Read error: Operation timed out13:53
pingveno (adevore@ruby.cat.pdx.edu) joined #pypy.13:53
voidspace (~anonymous@87-194-212-65.bethere.co.uk) joined #pypy.13:53
voidspace (~anonymous@87-194-212-65.bethere.co.uk) left irc: Changing host13:53
voidspace (~anonymous@python/psf/voidspace) joined #pypy.13:53
Nick change: Gulaway -> Gulopine13:59
k_bx (~k_bx@195.20.130.1) left irc: Read error: Connection reset by peer14:01
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.14:02
stakkars_arigato: hi Armin!14:02
arigatohi!14:02
stakkars_where can I find the nice script with which you made the stylized video on FSCONS?14:03
arigatohttps://bitbucket.org/pypy/extradoc/raw/extradoc/talk/iwtc11/benchmarks/image/14:04
stakkars_I am preparing something like materials to show to interested companies here14:04
stakkars_ha, thanks!14:04
Sho_ (~EHS1@kde/hein) joined #pypy.14:05
k_bx (~k_bx@195.20.130.1) joined #pypy.14:08
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep14:08
CIA-6803arigo 07roundup * 10#927/Task takes more time on nightly than CPython 2.7: Maybe it makes sense to use in this case a much smaller card ref size? * 14https://bugs.pypy.org/issue92714:08
G2PQuick remarks on the new site: the headings run into each other at my text zoom level; the sprite symbols aren't clear ("external link" looks like "opens in new tab")14:11
G2PPoints for not using the standard twitter widget; that tends to leak a lot14:11
mat^2 (~mathias@212.130.113.35) joined #pypy.14:13
`fox` (~fox@host42-105-dynamic.51-79-r.retail.telecomitalia.it) left irc: Ping timeout: 276 seconds14:16
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.14:17
bivab_ (~david@134.99.38.47) joined #pypy.14:19
bivab_ (~david@134.99.38.47) left irc: Client Quit14:19
gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) joined #pypy.14:21
bivab (~david@fwstups.cs.uni-duesseldorf.de) left irc: Ping timeout: 240 seconds14:22
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep14:22
arigatostakkars_: could you help with https://bugs.pypy.org/issue895 ?14:24
arigatootherwise pypy 1.7 will be released without _continuation support on windows14:24
stakkars_ek14:25
stakkars_s/ek/eek/14:25
stakkars_arigato: why is this problem on windows?14:29
stakkars_and not elsewhere?14:29
arigatoelsewhere, we use a .c file that includes gcc assembler inline14:29
stakkars_ah, I see. Are there tests breaking? Crashing, I presume.14:32
asabil (~asabil@131.246.16.62.customer.cdi.no) joined #pypy.14:32
stakkars_well, will start my virtual machine...14:32
arigatoI presume that the tests in pypy/module/_continuation are all getting skipped with a message like "compiler failed, so I'm guessing that your platform is not supported"14:33
stakkars_grrrmbl14:33
Nick change: Gulopine -> Gulaway14:46
stakkars_arigato: slp_switch_raw: this is what is missing, and I should do that, right?14:57
arigatoit's not missing, but it is in the file switch_x86_msvc.asm14:57
stakkars_(switching context completely, sorry)14:57
arigatowhat is missing is to tell msvc to also compile and link this .asm file14:57
stakkars_arch, yes, and the tool chain does not do asm at all14:58
stakkars_the same thing where I stopped this week, scratching my head.14:58
stakkars_because I didn't want to add ask support just for the one win64 file.14:59
stakkars_but now it is more. yes, I will add that.14:59
arigatothank you14:59
stakkars_when is the deadline for 1.7?14:59
arigatoyou have to ask fijal15:00
stakkars_not online, meaning I have at least time until he shows up15:00
lucianstakkars_: a late windows binary is probably fine15:01
arigato:-)15:01
arigatoyes15:01
stakkars_how does the linux version do it: all with inline assembler in c?15:01
asabil (~asabil@131.246.16.62.customer.cdi.no) left irc: Ping timeout: 240 seconds15:02
stakkars_is it worth it: should I do an extra function for assembler, or hack a bit into the c compiler15:03
stakkars_to treat .asm files a bit different?15:03
stakkars_this is where I was stuck, deciding.15:04
arigatoas you like15:04
asabil (~asabil@131.246.16.62.customer.cdi.no) joined #pypy.15:05
stakkars_ah, exactly three .asm files in the project.15:06
stakkars_ok, will try in in the evening. Fits my win64 porting very well.15:07
bbot24Finished ??: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/271 [12fijal, release-1.7.x]15:07
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/272 [12fijal, release-1.7.x]15:08
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.15:11
cfbolzarigato: I have an optimized trace where the SSA invariant is violated15:13
cfbolzarigato: e.g. the same variable is assigned twice15:13
cfbolzseems to be about virtuals15:14
arigato:-(15:15
cfbolzI bet it is this strange early-force stuff15:15
Rhy0litecfbolz: ping15:17
cfbolzpong15:17
Rhy0litecfbolz: do you have a moment to answer some trivial questions about PyPy testing framework?15:18
cfbolzsure15:18
Rhy0liteI'm trying to understand about cpu-specific tests15:20
Rhy0litewhen one needs the jit/backend/<arch>/test/test_runner.py harness15:21
Rhy0liteand how this interacts with pypy/jit/backend/test/test_ll_random.py15:22
cfbolzRhy0lite: this is all not really a feature of the test framework, but pypy-specific mess15:23
cfbolzRhy0lite: anyway, in theory it's all explicitly coded in these files (not saying that this makes it easy to understand)15:24
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.15:31
kenaan12arigo op_malloc_gc 11211a72d5b8b8 15/pypy/jit/backend/llsupport/test/test_gc.py: Plan which tests are still needed (and the corresponding code).15:32
kenaan12arigo default 11612f7784a228 15/pypy/translator/c/test/test_newgc.py: Test for crashes when we call StringBuilder.build() several times over a growing builder, like W_StringBufferObject...15:32
kenaan12arigo default 115269f0ee1ed2 15/pypy/: Fix for 612f7784a228.15:32
kenaan12arigo default 115019a28f7e6a 15/pypy/rpython/test/test_rbuilder.py: Python 2.5 compatibility.15:34
santagada (~leonardo@189.27.169.205.dynamic.adsl.gvt.net.br) joined #pypy.15:35
arigato (~arigo@fwstups.cs.uni-duesseldorf.de) left irc: Quit: See you15:36
fijal (~fijal@197.174.106.132) joined #pypy.15:37
Rhy0litecfbolz: sorry, was pulled away15:38
cfbolznp15:38
fijalhi15:39
Rhy0liteso how is one suppose to invoke test_ll_random for other architectures?15:39
cfbolzRhy0lite: I think it's supposed to auto-detect your current architecture and use that15:39
Rhy0liteokay15:39
cfbolzlet me read some code15:39
Rhy0liteand why doesn't runner_test.py do that?15:39
Rhy0liteno one got around to updating it?15:39
fijal (~fijal@197.174.106.132) left irc: Read error: Connection reset by peer15:40
fijal (~fijal@197.174.106.132) joined #pypy.15:40
Rhy0litehi, fijal.  wb15:40
cfbolzRhy0lite: runner_test.py is not really a test file, I think. it contains tests that you can subclass per CPU architecture15:41
cfbolzRhy0lite: the auto-detection logic for test_ll_random seems to be in test_random.py, in the function get_cpu()15:42
Rhy0liteIs test_ll_random.py a real test file?15:42
bivab (~david@ip-62-143-59-5.unitymediagroup.de) joined #pypy.15:42
cfbolzyes15:42
Rhy0litecfbolz: yes, I saw test_random.py15:42
cfbolzah, strange15:42
cfbolzok, I am confused now myself15:42
Rhy0litetest_random.py only looks at CPU38615:43
Rhy0liteI was not sure how test_ll_random.py and test_random.py were connected15:43
fijalanyone has any idea why buildbot crashed with no logs?15:43
fijaljust Exception15:43
cfbolzRhy0lite: test_ll_random imports some stuff from test_random15:44
fijalRhy0lite: test_random does not detect15:44
cfbolzthis is all nonsense15:45
fijalyou can pass options to py.test15:45
fijalwe can make it autodetect, there is code for that15:45
cfbolzfijal: yes, but x86/test/test_zll_random *does* autodetect15:45
cfbolzso Rhy0lite could run that15:45
fijalnice15:45
fijalyes, except it's in a wrong directory then15:45
fijalthis doesn't look very good15:46
fijalhttp://buildbot.pypy.org/summary?branch=release-1.7.x15:46
fijaland I have no idea why15:46
fijalRhy0lite: is everything else passing?15:46
fijallike test_runner15:46
Rhy0litefijal: getting there15:47
Rhy0liteAs long as I'm not crazy that this is confusing15:47
fijalthere is absolutely no point in running test_random before everything else passes15:47
fijalbecause errors are just so much more confusing15:47
cfbolzRhy0lite: it is15:48
fijalit is not confusing, it's organically grown15:49
Rhy0liteyes15:49
Rhy0liteand things work for x8615:49
Rhy0liteI'm trying to understand what will be needed for PPC15:49
fijalwe should move test_zll_random to backend/test directory15:50
fijalwell15:50
fijalI can explain why things are the way they're15:50
fijalwant to listen?15:51
[mat^2] (~mathias@212.130.113.35) joined #pypy.15:51
fijal (~fijal@197.174.106.132) left irc: Read error: Connection reset by peer15:51
mat^2 (~mathias@212.130.113.35) left irc: Ping timeout: 244 seconds15:52
fijal (~fijal@196-215-106-182.dynamic.isadsl.co.za) joined #pypy.15:52
fijalRhy0lite: I can explain why the things are the way they're15:52
fijalcfbolz: can you look into doc/release-1.7.0.rst and see what should be added?15:53
fijalI just listed things, it's not in a state to send it of course15:53
ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) left irc: Quit: Leaving...15:55
fijalRhy0lite: I see you're not that interested ;-)16:00
Rhy0liteyou want to write a tutorial16:01
fijalno :)16:01
Rhy0litehigh frequency interrupts this morning16:01
fijalbut I can tell you why tests are the way they're16:02
Rhy0liteyes, I saw that test_zll_random seemed to be the x86 driver16:02
Rhy0liteI'm just trying to understand how these things should be pieced together or accessed for PPC16:02
Rhy0liteI can imagine how they evolved16:02
Rhy0lite:-)16:02
fijalno, it makes sense16:02
fijalyou want to run those tests by hand usually16:02
fijalthe automatic nightly run is a bit useless16:02
fijalbecause it runs too few iterations16:02
Rhy0literun which tests by hand?16:03
Rhy0litetest_ll_random?16:03
fijalso you want to add an option to be able to run test_random/test_zll_random by hand16:03
fijalyes, both of them16:03
Rhy0liteboth what?16:03
fijaland pass some large number of loops and the backend16:03
fijaltest_random & test_ll_random16:03
Rhy0liteokay, yes16:03
fijalwell16:03
fijalit's ok if you have autodetect option as well16:04
asmeurer_ (~asmeurer@dhcp-baca-230.resnet.nmt.edu) joined #pypy.16:04
fijaland it's the default16:04
fijalI can change it if you want16:04
fijalbut you still don't want to run them with default options and be happy, you need more iterations16:04
Rhy0liteMy problem is more high level16:04
Rhy0liteI should say confusion16:04
Rhy0liteTesting is very important16:05
fijalok?16:05
alderz (~alderz@157.Red-83-59-100.dynamicIP.rima-tde.net) joined #pypy.16:05
Rhy0litebut it's not clear what incantations one should use for testing16:05
fijalyou should run test_ll_random with some large number of loops16:05
Rhy0liteother than pytest.py16:05
fijalpreferably the entire night or so16:05
fijal--help is your friend :)16:05
fijalby default is run with llgraph16:06
fijal(which makes sense for testing the tests)16:06
fijalyou can also copy test_zll_random to ppc directory16:06
stakkars_ (~tismer@i59F77D6B.versanet.de) left irc: Ping timeout: 240 seconds16:06
fijal(verbatim copy)16:06
cfbolzRhy0lite: the incantation is something like pytest.py test/test_random.py --backend=ppc16:07
cfbolzoptionally with --repeat=<something large>16:07
cfbolzbut when you are just starting, it will crash quickly16:07
Rhy0liteI will let you guys decide if copying test_zll_random.py or moving it to a common location makes sense16:08
cfbolzit does16:08
cfbolzfijal: could you do it?16:08
fijalno16:08
cfbolzwhy not?16:08
fijalwait a second16:08
Rhy0liteBut I think it would be helpful if there was a pypy webpage, not blog post, about how one runs tests16:09
jterrace (jterrace@aegis.CS.Princeton.EDU) left #pypy.16:09
stakkars_ (~tismer@89.204.153.148) joined #pypy.16:09
fijalRhy0lite: not many people write assembler backends16:09
fijalI think it does not warrant documentation16:09
kenaan12fijal default 11cb316be6a96f 15/pypy/jit/backend/: use autodetect if you don't want to run llgraph16:09
kenaan12fijal default 1154e972f7c8ba 15/pypy/: merge16:09
Rhy0litefijal: you want to encourage that, don't you?16:09
Rhy0lite:-)16:09
fijalI don't think documentation on peculiar issues of testing would help16:10
fijalyou run now test_ll_random.py --backend=cpu16:10
fijaland it'll autodetect the backend16:10
fijalpy.test test_ll_random.py --backend=cpu16:10
fijalnote that default repeat is useless16:10
Rhy0liteI think you are assuming a lot from novice developers16:10
fijalso --repeat=<some large number>16:10
Rhy0litejust because I can look at --help output16:10
Rhy0litedoes not mean that I know which options need to be set16:10
Rhy0litewhich have reasonable defaults16:11
fijalRhy0lite: look, when you want to run those tests, this means you have just written the entire assembler backend16:11
fijalso you know at least where to go and ask for help :)16:11
Rhy0liteokay, okay16:11
Rhy0litetrying to be constructive here16:11
fijalI don't mind explaining it to everyone who writes the entire backend :)16:11
Rhy0liteand not wanting to pester people with seemingly simple questions16:11
fijalthey're not :)16:12
Rhy0litePyPy already seems like magic to a lot of people16:12
fijalbut random testing is a bit special16:12
Rhy0litePieces that can be more approachable are helpful for attracting developers16:12
Rhy0litemaybe not core developers, but people to help with bug triage, etc.16:13
kenaan12fijal default 1181dcafabbba9 15/pypy/jit/backend/: move test_zll_random to llsupport16:13
stakkars_fijal: hi!16:13
fijalstakkars_: hi!16:13
fijalRhy0lite: as I said random tests are a bit special16:13
cfbolzRhy0lite: I agree that pypy is not that approachable in many places16:13
fijalthis really means you pass every other tests16:13
stakkars_fijal: for when is PyPy 1.7 planned?16:13
cfbolzfijal: I don't follow your argument16:13
cfbolzfijal: even *I* was confused by the situation16:14
cfbolzand I'm hardly a novice16:14
fijalcfbolz: I tried to clarify a bit16:14
fijalis it better now?16:14
Rhy0litecfbolz: apparently confusion is a feature! ;-)16:14
fijalI moved test_zll_random to llsupport16:14
fijaland made arguments a bit better16:14
ixokai (~ixokai@rrcs-24-199-8-246.west.biz.rr.com) joined #pypy.16:15
ixokai (~ixokai@rrcs-24-199-8-246.west.biz.rr.com) left irc: Changing host16:15
ixokai (~ixokai@pdpc/supporter/student/ixokai) joined #pypy.16:15
fijalcfbolz: argument is that test_random is a bit special16:15
fijalI agree pypy should be more approachable in places, but this is a bit not the place to make it so16:15
Topic changed on #pypy by cfbolz!cfbolz@dslb-178-003-008-145.pools.arcor-ip.net: PyPy, a fast and compliant python interpreter | IRC logs: http://www.tismer.com/pypy/irc-logs/pypy/latest.log.html | apparently confusion is a feature16:15
fijalhow can you imagine improving the situation?16:15
fijalstakkars_: soon16:15
fijalstakkars_: why?16:15
cfbolzfijal: so special that only you are allowed to understand it?16:15
fijalno :)16:15
cfbolzRhy0lite: congrats, you made the irc channel topic16:16
fijalI said I tried to improve the situation16:16
fijalis it better?16:16
stakkars_fijal: not why. just when ;-)16:16
fijalstakkars_: why do you ask ;-)16:16
Rhy0liteyea!16:16
stakkars_wrong question.16:16
fijalcfbolz: I welcome improvement suggestions :)16:16
fijalhm16:17
fijalok16:17
fijalI think I have an idea actually :)16:17
CIA-6803arigo 07roundup * 10#926/string concatenation using + kills PyPy: 16:25
CIA-68hruske: I think I have fixed the segfault (5269f0ee1ed2). Can you try again?16:25
CIA-68The main point is that we have to decide if we would like strbuf ... * 14https://bugs.pypy.org/issue92616:25
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Ping timeout: 244 seconds16:27
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 240 seconds16:28
fijalRhy0lite: ok, it should be cleaner now16:31
Rhy0litefijal: thanks!16:31
kushal (kdas@fedora/kushal) left irc: Quit: This computer has gone to sleep16:31
kenaan12fijal default 11dcc33426aadc 15/pypy/jit/backend/: Move the stress test directly to backend/test and rename it16:32
fijalyou run test_zll_stress.py from jit/backend/test simply16:32
fijalalthough note that for *real* testing you seriously want to leave it running for a night16:32
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) joined #pypy.16:32
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1850 [12fijal, release-1.7.x]16:34
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/711 [12fijal, release-1.7.x]16:34
bivab (~david@ip-62-143-59-5.unitymediagroup.de) left irc: Read error: Connection reset by peer16:34
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1113 [12fijal, release-1.7.x]16:34
Alex_Gaynorfijal: giving another presentation at my school on numpypy, http://paste.pocoo.org/show/509830/ are the HTML slides if you feel like looking at them16:34
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/579 [12fijal, release-1.7.x]16:34
bivab_ (~david@ip-62-143-59-5.unitymediagroup.de) joined #pypy.16:34
fijalanother attempt16:34
fijalAlex_Gaynor: sure16:34
Alex_Gaynorfijal: I'm trying some HTML slides thing, so the HTML is the original source16:35
fijalit's not africa-friendly16:35
Alex_Gaynorwhat isn't?16:36
Trundle (~andy@python/site-packages/trundle) left irc: Remote host closed the connection16:36
fijalsize of those slides :)16:36
fijalalso it's slow on my chromium16:36
Alex_GaynorWorksforme16:36
santagada (leonardo@189.27.169.205.dynamic.adsl.gvt.net.br) left #pypy.16:39
fijalyou might want to say we call it numpypy these days16:40
fijalsay import numpypy as numpy16:40
fijal?16:40
Alex_Gaynorgood idea, I'll stick it on the first slide16:40
fijalwant to say a word about multi dimensional arrays?16:42
fijalit's work-in-progress16:42
fijalbut it's looking nice16:42
Alex_Gaynoryes16:42
fijalit's the same idea as numpy but it (almost) works nicely in the JIT16:43
fijalso you iterate over the shape16:43
fijalor shards16:43
fijalshards are offsets between consecutive values in dimension N16:43
fijalso &a[...,n + 1,...] - &a[...,n,...]  will be shards[n]16:43
fijaland to iterate over array you iterate over them16:44
__pvshards == strides?16:44
fijal__pv: yeah :)16:44
Alex_Gaynorfijal: any other things on the currently-worked-on for numpy?16:44
fijalI guess those should be named strides16:44
fijalnot that I know about16:44
fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy.16:44
fijalf2pypy?16:44
Alex_GaynorAh yes16:44
fijalanyway, it nicely unrolls16:46
Alex_Gaynorcool16:46
fijalI would show some jit traces from complex operations?16:46
fijallike16:46
fijal(a + b)[::3] * 216:46
fijaland how this looks like16:46
fijalyou explain one piece, which is how we approach data types16:47
fijaland I'm talking about another piece which is chains of operations16:47
fijalhakanardo: ping btw16:47
Alex_Gaynorfijal: well, that's the piece I worked on for class, so yes :)16:47
Alex_Gaynorfijal: I think I'll show off our testing language as well16:47
fijalok :)16:47
fijalthen feel free to ignore16:47
fijaltesting language is cool ;-)16:47
antocuni (~antocuni@host200-125-dynamic.25-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds16:48
fijalAlex_Gaynor: I'm afraid I'm off, I forgot my charger16:50
Alex_Gaynorok see ya16:50
AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) joined #pypy.16:51
k_bx (~k_bx@195.20.130.1) left irc: Ping timeout: 260 seconds16:52
G2P (~G2P@fw-asn1.ornis.com) left irc: Quit: Leaving.16:52
mitchellh (~mitchellh@c-71-202-125-40.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep16:52
k_bx (~k_bx@195.20.130.1) joined #pypy.16:53
rokujyouhitoma (~rokujyouh@p5091-ipad03sizuokaden.shizuoka.ocn.ne.jp) left irc: Remote host closed the connection16:54
AndrewHarrison (~Adium@66-215-59-147.dhcp.atsc.ca.charter.com) left irc: Client Quit16:54
fijalfound charger :)16:56
bivab_ (~david@ip-62-143-59-5.unitymediagroup.de) left irc: Ping timeout: 276 seconds16:57
Alex_Gaynorfijal: nice :)16:57
Alex_Gaynorfijal: btw I think I need multidim to finish some of the scalar stuff in -dtype-refactor16:57
fijalok16:59
fijalcool :)16:59
bivab (~david@ip-62-143-59-5.unitymediagroup.de) joined #pypy.17:00
bivab (~david@ip-62-143-59-5.unitymediagroup.de) left irc: Client Quit17:01
JaredWigmore (~jared@061093200075.ctinets.com) left irc: Ping timeout: 260 seconds17:02
kushal (~kdas@fedora/kushal) joined #pypy.17:04
kushal (~kdas@fedora/kushal) left irc: Client Quit17:04
amaury_ (amaury_@nat/google/x-tpezopmkrctqxhzs) joined #pypy.17:05
k_bx (~k_bx@195.20.130.1) left irc: Remote host closed the connection17:05
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.17:09
fijalRhy0lite: did it help?17:15
arigato (~arigo@89.204.154.65) joined #pypy.17:24
kkris (~kris@93-82-40-48.adsl.highway.telekom.at) joined #pypy.17:27
amaury_ (amaury_@nat/google/x-tpezopmkrctqxhzs) left irc: Ping timeout: 244 seconds17:31
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.17:32
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1850 [12fijal, release-1.7.x]17:33
fijalnot good17:39
Alex_Gaynorfijal: do we have bugs?17:40
fijalapparently17:41
fijalif you look at the buildbot summary it's full of nonsense and buildbot exceptions17:41
Alex_Gaynorbuildbots haven't been running clean in like a week, I think we need some cleanup time before a release17:41
alderz (~alderz@157.Red-83-59-100.dynamicIP.rima-tde.net) left irc: Ping timeout: 240 seconds17:42
kenaan12arigo op_malloc_gc 11feae28c11d92 15/pypy/jit/backend/llsupport/test/: Move the new tests to their own file.17:45
kenaan12arigo op_malloc_gc 11f84b61015be5 15/pypy/jit/metainterp/resoperation.py: The interface that I really want.17:45
arigatofijal: ?17:46
arigatofijal: it looks good to me17:46
fijal (~fijal@196-215-106-182.dynamic.isadsl.co.za) left irc: Ping timeout: 240 seconds17:47
arigatoztranslation fails for obvious reasons of a missed fix17:47
arigatoit seems to me that buildbots have been running clean today, but maybe I'm missing something17:48
Rhy0litefijal: does shard / stride mean a slice of the array?17:55
arigatoa generalized slice, yes; that's numpy terminology17:56
Rhy0liteis it always 1-dimensional?17:56
arigatonot necessarily17:56
Rhy0liteokay, cool17:57
tilgovi (~randall@adsl-67-116-241-49.dsl.pltn13.pacbell.net) joined #pypy.18:01
tilgovi (~randall@adsl-67-116-241-49.dsl.pltn13.pacbell.net) left irc: Changing host18:01
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.18:01
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection18:04
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) joined #pypy.18:04
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Ping timeout: 240 seconds18:06
tilgovi (~randall@adsl-67-116-241-49.dsl.pltn13.pacbell.net) joined #pypy.18:07
tilgovi (~randall@adsl-67-116-241-49.dsl.pltn13.pacbell.net) left irc: Changing host18:07
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.18:07
voidspace (~anonymous@python/psf/voidspace) left irc: Quit: voidspace18:10
ojii (~ojii@40-34.60-188.cust.bluewin.ch) left irc: Quit: Leaving18:11
antocuni (~antocuni@host200-125-dynamic.25-79-r.retail.telecomitalia.it) joined #pypy.18:12
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) joined #pypy.18:15
hruskehi18:15
arigatohi18:16
hruskecompiling.18:17
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/711 [12fijal, release-1.7.x]18:29
fijal (~fijal@41.52.0.11) joined #pypy.18:37
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Ping timeout: 244 seconds18:39
fijalarigato: stride is also something more precise18:41
fijallike I explained to alex n-th stride is a distance between two points in dimension n18:41
fijaldistance in memory18:41
kenaan12fijal numpy-multidim-shards 11ad7839a7ff62 15/pypy/module/micronumpy/interp_numarray.py: remove unused parameter18:43
fijaleh18:46
fijalso mattips changes contain # this should not directly access shards or start18:46
fijalbut it however does access it directly18:46
Alex_Gaynorfijal: is there a TODO file for multidim-shards, and is there a reason we have 2 branches?18:50
fijalthere is no TODO file and there is one item to be done and one item to be decided18:50
fijalwe have 2 branches, because when I started, I didn't want to break everything on numpy-multidim branch18:50
Alex_Gaynorfijal: what are teh 2 things?18:51
fijalone thing is broadcasting (I'm working on it more or less now) the other thing is what we do with test_zjit and multidim views18:51
fijalit seems that we're not doing very well as of now because of the bugs in optimizer18:52
fijalarmin says "wait for jit-targets"18:52
fijalso maybe we merge it anyway18:52
fijal"working"18:52
fijalbut not really that great with multidimensional views18:52
Alex_Gaynorcan we work around them easily, or are we going to merge in large performance regressions?18:52
fijalI think this only affects multidimensional views18:52
fijalbut it might influence normal views as well18:53
Alex_Gaynorwrite a test18:53
fijalthere is an easy workaround18:53
fijalI'm busy with other stuff18:53
fijalbut we have to think what we gonna do18:53
fijalworkaround = special iterators for single-dimensional views18:53
Alex_Gaynoroh, pff -118:53
Alex_Gaynorhave you spoken with hakanardo about the status of his branch?18:53
stakkars_ (~tismer@89.204.153.148) left irc: Quit: schnarch18:57
fijalno19:03
fijalcould not find him19:03
gutworthcan someone remind me how slow jitless pypy is compared to cpython?19:03
gutworth30%, is that right?19:03
Alex_Gaynor.8-4x I think19:03
Alex_Gaynorgutworth: as you well know, benchmark dependent19:03
fijalno it's not19:03
fijalit depends a lot what you do19:03
gutworthwell, yes, but if I wanted to quote one figure...19:04
Alex_Gaynorgutworth: working on your pycon slides already?19:04
gutworthhah, no19:05
gutworthI'm not that optimistic :)19:05
Alex_GaynorThis is the bad part about being on the PC, you can't engage in rampant speculation with everyone else.19:05
arigatogutworth: I think we're aiming for 2x19:06
fijal (~fijal@41.52.0.11) left irc: Read error: Connection reset by peer19:06
arigatobut yes, 0.8x-to-4x is a good range19:06
Alex_GaynorI'm so glad we don't have stuff like http://hg.python.org/cpython/rev/fa2f8dd077e019:06
Action: Alex_Gaynor -> class19:07
arigatosee you19:08
asabil (~asabil@131.246.16.62.customer.cdi.no) left irc: Quit: Ex-Chat19:08
[mat^2] (~mathias@212.130.113.35) left irc: Ping timeout: 245 seconds19:11
redline6561 (~redline65@li69-162.members.linode.com) joined #pypy.19:11
hruske (~Gasper@188-230-156-183.dynamic.t-2.net) left irc: Remote host closed the connection19:15
Topic changed on #pypy by arigato!arigo@89.204.154.65: PyPy, a fast and compliant python interpreter | IRC logs: http://www.tismer.com/pypy/irc-logs/pypy/latest.log.html | apparently confusion is a feature... or is it?19:23
redline6561 (redline65@li69-162.members.linode.com) left #pypy ("ERC Version 5.3 (IRC client for Emacs)").19:24
nickik (~nick@217.150.239.67) joined #pypy.19:25
nickikWhats the best book to learn python? For somebody that allready knows programming just to get started with python quickly.19:28
arigatonickik: you should ask on #python19:29
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1113 [12fijal, release-1.7.x]19:29
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/579 [12fijal, release-1.7.x]19:31
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.19:33
stakkars_ (~tismer@89.204.155.162) joined #pypy.19:35
antocuni (~antocuni@host200-125-dynamic.25-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds19:42
rguilleberthi20:06
asenchi (~asenchi@206.173.142.114) left irc: Ping timeout: 240 seconds20:10
amaury_hi20:11
asenchi (~asenchi@206.173.142.114) joined #pypy.20:13
arigatohi20:14
h0h0 (~k@f38016.upc-f.chello.nl) joined #pypy.20:14
tilgovi (~randall@199.36.158.16) joined #pypy.20:19
tilgovi (~randall@199.36.158.16) left irc: Changing host20:19
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.20:19
ojii (~ojii@80-218-1-60.dclient.hispeed.ch) joined #pypy.20:33
amaury_Huh, I knew CPython was easy to crash, but I did not know it was a feature20:36
amaury_>>> import faulthandler20:37
amaury_>>> faulthandler._sigsegv()20:37
amaury_Segmentation fault20:37
Alex_Gaynoramaury_:I assume that's for testing?20:39
Alex_Gaynoramaury_: it's not like there isn't already a directory of other, more clever ways to crash, courtesy of armin20:39
arigatoos.kill(0, 9)20:41
exarkunor even 11.20:43
arigatoyes20:43
Action: exarkun wonders how _sigsegv differs20:43
arigatowindows support? :-)20:44
exarkunheh20:44
amaury_no, os.kill also works on win3220:44
exarkundoes SIGSEGV work on win32?20:44
amaury_no20:44
rguillebertmaybe it does a real segfault 20:44
rguillebert*NULL = 0; :p20:45
exarkunWhat, rely on undefined behavior?  Who would dare20:45
ojii (~ojii@80-218-1-60.dclient.hispeed.ch) left irc: Read error: Connection reset by peer20:45
amaury_on win32, os.kill has special code for signal.CTRL_C_EVENT and signal.CTRL_BREAK_EVENT20:46
amaury_other signals just call TerminateProcess()20:47
ojii (~ojii@80-218-1-60.dclient.hispeed.ch) joined #pypy.20:48
lizardo (~lizardo@189.2.128.130) left irc: Quit: Leaving20:51
h0h0 (~k@f38016.upc-f.chello.nl) left irc: Quit: leaving20:56
mwhudson (~mwh@linaro/mwhudson) joined #pypy.21:01
stakkars_arigato: with quite some hacking I got continue let etc. to compile. But all tests fail, still :-/21:01
Rhy0lite (dje@nat/ibm/x-rhiolehszwlqjzuy) left irc: Quit: Leaving21:02
stakkars_mams is very picky concerning argument error and ignores everything past the source files21:02
stakkars_masn, sorry21:02
cfbolz (~cfbolz@dslb-178-003-008-145.pools.arcor-ip.net) left irc: Quit: Leaving21:03
mwhudson (~mwh@linaro/mwhudson) left irc: Ping timeout: 258 seconds21:05
legogris_ (~legogris@c80-217-207-157.bredband.comhem.se) joined #pypy.21:07
legogris_ (~legogris@c80-217-207-157.bredband.comhem.se) left irc: Client Quit21:07
mat^2 (~mathias@212.130.113.35) joined #pypy.21:08
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds21:10
stakkars_all: who knows how to test the _continuation module?21:10
stakkars_test_all does not work. It says:21:11
stakkars_AssertionError: stack let: you have to translate with --continuation21:12
stakkars_but I cannot pass that option in!21:12
arigatostakkars_: run "python test_all.py module/_continuation"21:12
Alex_Gaynorstakkars_: ./pytest.py pypy/module/_continuation/ -x is working for me21:12
stakkars_arigato: yes, I did this21:15
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) left irc: Quit: Computer has gone to sleep21:16
stakkars_most of them fail, for instance test_simple lacks thread support ?!21:16
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 245 seconds21:20
stakkars_arigato: stacklet_newthread is missing in many cases21:31
etrepum (~bob@accessnat4.mochimedia.net) joined #pypy.21:32
fschwidom (~fschwidom@46.115.1.48) left irc: Remote host closed the connection21:34
arigatostakkars_: sorry, I can't help with so little information21:37
alderz (~alderz@157.Red-83-59-100.dynamicIP.rima-tde.net) joined #pypy.21:38
DanKluev (~root@83-116-adsl.vntc.ru) left irc: Ping timeout: 276 seconds21:42
stakkars_arigato: what should I tell. I got the stuff to compile fine with windows, but there are almost all tests failing. Maybe we can run the buildbot after I checked in?21:44
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.21:47
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.21:47
arigatoyes21:48
stakkars_stacklet_newthread not found in ...21:49
stakkars_ah, for some reason stack let.c seems to get ignored21:50
stakkars_ok, I will check in first. Irt is at least better than before.21:52
DanKluev (~root@93-88-2-081-xdsl.vntc.ru) joined #pypy.21:56
kenaan12ctismer default 117d9e78a91dce 15/pypy/: made the assembler sources compile so far21:56
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/272 [12fijal, release-1.7.x]21:58
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/273 [12fijal, release-1.7.x]21:58
mitchellh (~mitchellh@c-69-181-107-107.hsd1.ca.comcast.net) joined #pypy.22:10
bbot2Started: 15http://buildbot.pypy.org/builders/own-win-x86-32/builds/8 [12ctismer]22:11
kenaan12ctismer win64-stage1 1130d0ed4adb92 15/: merge22:18
gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) left irc: Remote host closed the connection22:34
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 248 seconds22:36
Zira (~Zeldakits@206.175.34.95.customer.cdi.no) joined #pypy.22:37
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/14322:42
bbot2Started: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/71622:42
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/94622:42
stakkars_ (~tismer@89.204.155.162) left irc: Quit: schnarch22:44
cpets (4f75d76f@gateway/web/freenode/ip.79.117.215.111) joined #pypy.22:50
arigatogood night22:54
arigato (~arigo@89.204.154.65) left irc: Quit: See you22:54
amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 258 seconds22:57
pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) left irc: Quit: Leaving22:57
alderz (~alderz@157.Red-83-59-100.dynamicIP.rima-tde.net) left irc: Ping timeout: 244 seconds23:00
pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) joined #pypy.23:13
jer (~jtregunna@unaffiliated/jer) joined #pypy.23:17
jeranyone happen to know off hand (or can point me to some document) that explains what the JIT does when the exec ... in statement is hit? what i'm interested in is if the JIT sees this and avoids optimizing the code in that trace, or doesn't JIT it at all... or how it performs any optimizations related to it23:19
exarkunI don't think there's anything special about exec.23:22
exarkunIt's just more code.23:22
Alex_Gaynorthere isn't, execing a code object should just trace though it normally, AFAIK23:22
exarkunThe JIT traces what actually gets run23:22
exarkunSo it doesn't matter how dynamic the source is23:22
cpets (4f75d76f@gateway/web/freenode/ip.79.117.215.111) left irc: Ping timeout: 265 seconds23:31
asmeurer_ (~asmeurer@dhcp-baca-230.resnet.nmt.edu) left irc: Quit: asmeurer_23:39
rokujyouhitoma (~rokujyouh@p3023-ipad305sizuokaden.shizuoka.ocn.ne.jp) joined #pypy.23:40
kkris (~kris@93-82-40-48.adsl.highway.telekom.at) left irc: Quit: Leaving.23:42
stakkars_ (~tismer@p5DC46801.dip0.t-ipconnect.de) joined #pypy.23:49
bbot24Failure: 15http://buildbot.pypy.org/builders/own-win-x86-32/builds/8 [12ctismer]23:51
bbot2Started: 15http://buildbot.pypy.org/builders/own-win-x86-32/builds/9 [12ctismer, default]23:51
--- Sat Nov 19 201100:00

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