| nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 255 seconds | 00:00 | |
| mat^2 (~mathias@212.130.113.35) left irc: Ping timeout: 240 seconds | 00:01 | |
| rekamso | does pypy keep local variables on the stack? | 00:01 |
|---|---|---|
| _stian_ (~stian@137.241.189.109.customer.cdi.no) left irc: Quit: Leaving. | 00:05 | |
| nedbat (~nedbat@python/psf/nedbat) joined #pypy. | 00:12 | |
| gutworth | rekamso: more context needed | 00:17 |
| gutworth | jitted or interpretered, c stack or python stack? | 00:17 |
| aleksi (~aleksi@109-184-97-156.dynamic.mts-nn.ru) left irc: Remote host closed the connection | 00:21 | |
| rekamso | jitted and interpreted; not entirely sure how the python stack works | 00:22 |
| rekamso | is it just a chunk of allocated memory seperate from the c stack? | 00:22 |
| rekamso | my recollection is cpython puts python variables on the c stack the same as it puts python call frames on the c stack | 00:23 |
| rekamso | but that actually could be wrong :) | 00:23 |
| gutworth | the answer is no | 00:28 |
| fermianyon (~lane@c-71-229-21-197.hsd1.al.comcast.net) left irc: Ping timeout: 240 seconds | 00:36 | |
| ronan (~ronan@host86-143-209-182.range86-143.btcentralplus.com) left irc: Quit: Ex-Chat | 00:36 | |
| kenaan | 12hager ppc-jit-backend 11b3a8f9928a2e 15/pypy/jit/backend/ppc/ppcgen/: fix basic field operations | 00:38 |
| gtaylor (~gtaylor@108-196-160-156.lightspeed.tukrga.sbcglobal.net) left irc: Quit: Konversation terminated! | 00:45 | |
| gutworth (~benjamin@64-131-9-52.usfamily.net) left irc: Ping timeout: 240 seconds | 00:47 | |
| gutworth (~benjamin@64-131-9-52.usfamily.net) joined #pypy. | 00:47 | |
| GDorn (gdorn@li223-147.members.linode.com) left #pypy. | 01:03 | |
| mtigas (~mtigas@users.spokesman.com) left irc: Quit: mtigas | 01:03 | |
| tilgovi (~randall@ool-4357ca60.dyn.optonline.net) joined #pypy. | 01:09 | |
| tilgovi (~randall@ool-4357ca60.dyn.optonline.net) left irc: Changing host | 01:09 | |
| tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy. | 01:09 | |
| cpher (~chris@216-197-175-156.sktn.hsdb.sasknet.sk.ca) joined #pypy. | 01:13 | |
| nettok (~quassel@200.119.181.16) joined #pypy. | 01:16 | |
| nettok_ (~quassel@200.119.187.69) joined #pypy. | 01:20 | |
| fzzzy (~donovan@nat/mozilla/x-egdecpsujerlpihz) left irc: Read error: Connection reset by peer | 01:20 | |
| fzzzy (~donovan@2620:101:8003:200:c2f:dea6:589c:bd1a) joined #pypy. | 01:20 | |
| nettok (~quassel@200.119.181.16) left irc: Ping timeout: 252 seconds | 01:22 | |
| nettok_ (~quassel@200.119.187.69) left irc: Ping timeout: 252 seconds | 01:27 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/194 | 01:30 |
| jterrace (~Adium@dynamic-oit-vapornet-b-18.Princeton.EDU) joined #pypy. | 01:30 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/22 | 01:30 |
| halgari (475a45cc@gateway/web/freenode/ip.71.90.69.204) joined #pypy. | 01:33 | |
| nettok (~quassel@200.119.187.69) joined #pypy. | 01:39 | |
| halgari (475a45cc@gateway/web/freenode/ip.71.90.69.204) left irc: Quit: Page closed | 01:40 | |
| fzzzy (~donovan@2620:101:8003:200:c2f:dea6:589c:bd1a) left irc: Quit: fzzzy | 01:41 | |
| ericflo (~ericflo@75.103.8.110) left irc: Quit: ericflo | 01:44 | |
| cpher (~chris@216-197-175-156.sktn.hsdb.sasknet.sk.ca) left irc: Quit: cpher | 01:44 | |
| nettok_ (~quassel@200.119.185.251) joined #pypy. | 01:44 | |
| nettok (~quassel@200.119.187.69) left irc: Ping timeout: 240 seconds | 01:45 | |
| kenaan | 12hager ppc-jit-backend 1181584461b2f7 15/pypy/jit/backend/ppc/ppcgen/: made test_array_basic pass | 01:46 |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/766 | 01:47 |
| davisagli (davisagli@davisagli.com) left #pypy. | 01:47 | |
| zain (~textual@12.130.118.22) joined #pypy. | 01:49 | |
| nettok (~quassel@190.143.178.2) joined #pypy. | 01:50 | |
| nettok_ (~quassel@200.119.185.251) left irc: Ping timeout: 244 seconds | 01:51 | |
| mat^2 (~mathias@212.130.113.35) joined #pypy. | 01:59 | |
| rekamso (~textual@63-228-127-70.dia.static.qwest.net) left irc: Quit: bus | 01:59 | |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1911 | 02:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/328 | 02:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/664 | 02:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/317 | 02:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/666 | 02:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1182 | 02:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/849 | 02:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/1501 | 02:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/776 | 02:00 |
| nettok (~quassel@190.143.178.2) left irc: Ping timeout: 248 seconds | 02:03 | |
| nettok (~quassel@190.143.190.233) joined #pypy. | 02:04 | |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy. | 02:06 | |
| nettok (~quassel@190.143.190.233) left irc: Ping timeout: 255 seconds | 02:08 | |
| nettok (~quassel@190.143.178.170) joined #pypy. | 02:09 | |
| fermianyon (~lane@c-71-229-21-197.hsd1.al.comcast.net) joined #pypy. | 02:13 | |
| nettok (~quassel@190.143.178.170) left irc: Ping timeout: 252 seconds | 02:14 | |
| nettok_ (~quassel@190.143.179.75) joined #pypy. | 02:14 | |
| tilgovi (~randall@couchdb/developer/tilgovi) left irc: Ping timeout: 240 seconds | 02:26 | |
| nettok_ (~quassel@190.143.179.75) left irc: Read error: Connection reset by peer | 02:29 | |
| nettok (~quassel@190.143.180.36) joined #pypy. | 02:30 | |
| durin42 (~durin@adium/durin42) left irc: Ping timeout: 252 seconds | 02:35 | |
| derdon (~derdon@p5DE8A9FE.dip.t-dialin.net) left irc: Remote host closed the connection | 02:36 | |
| durin42 (~durin@adium/durin42) joined #pypy. | 02:37 | |
| tilgovi (~randall@ool-4357ca60.dyn.optonline.net) joined #pypy. | 02:39 | |
| tilgovi (~randall@ool-4357ca60.dyn.optonline.net) left irc: Changing host | 02:39 | |
| tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy. | 02:39 | |
| mat^2 (~mathias@212.130.113.35) left irc: Ping timeout: 252 seconds | 02:49 | |
| Ademan (~dan@adsl-71-141-224-79.dsl.snfc21.pacbell.net) left irc: Quit: leaving | 03:01 | |
| Kaskuka (~John@osbk-4db14907.pool.mediaWays.net) left irc: Ping timeout: 240 seconds | 03:03 | |
| Kaskuka (~John@osbk-4db06dd0.pool.mediaWays.net) joined #pypy. | 03:03 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/1003 | 03:11 |
| stakkars_ (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 03:13 | |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 03:13 | |
| Nick change: stakkars_ -> stakkars | 03:13 | |
| stakkars_ (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 03:16 | |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 03:16 | |
| Nick change: stakkars_ -> stakkars | 03:16 | |
| stakkars_ (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 03:17 | |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 03:17 | |
| Nick change: stakkars_ -> stakkars | 03:17 | |
| stakkars_ (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 03:18 | |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 03:18 | |
| Nick change: stakkars_ -> stakkars | 03:18 | |
| dracman (~draco@212.255.47.158) left irc: Ping timeout: 252 seconds | 03:19 | |
| stakkars_ (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 03:23 | |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 03:23 | |
| Nick change: stakkars_ -> stakkars | 03:23 | |
| stakkars_ (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 03:24 | |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 03:24 | |
| Nick change: stakkars_ -> stakkars | 03:24 | |
| dracman (~draco@212.255.38.129) joined #pypy. | 03:25 | |
| zain (~textual@12.130.118.22) left irc: Quit: Textual IRC Client: http://www.textualapp.com/ | 03:43 | |
| nedbat (~nedbat@python/psf/nedbat) left irc: Disconnected by services | 04:13 | |
| nedbat_ (~nedbat@python/psf/nedbat) joined #pypy. | 04:13 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1911 | 04:13 |
| anish_ (anish@nat/redhat/x-fazxbtihfhegzbcb) joined #pypy. | 04:14 | |
| thinker (~thinker@123-194-52-90.dynamic.kbronet.com.tw) left irc: Remote host closed the connection | 04:17 | |
| Nick change: nedbat_ -> nedbat | 04:26 | |
| nettok_ (~quassel@190.143.182.96) joined #pypy. | 04:30 | |
| nettok (~quassel@190.143.180.36) left irc: Ping timeout: 252 seconds | 04:33 | |
| nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 248 seconds | 04:34 | |
| nettok (~quassel@190.143.176.89) joined #pypy. | 04:35 | |
| nettok_ (~quassel@190.143.182.96) left irc: Ping timeout: 252 seconds | 04:37 | |
| stakkars_ (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 04:41 | |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 04:41 | |
| Nick change: stakkars_ -> stakkars | 04:41 | |
| pjenvey (~pjenvey@underboss.org) left irc: Ping timeout: 255 seconds | 04:44 | |
| ericholscher (~eric@173.203.223.63) left irc: Read error: Operation timed out | 04:44 | |
| mat^2 (~mathias@212.130.113.35) joined #pypy. | 04:45 | |
| pjenvey (~pjenvey@underboss.org) joined #pypy. | 04:48 | |
| ericholscher (~eric@173.203.223.63) joined #pypy. | 04:48 | |
| elliottcable (~me@ell.io) joined #pypy. | 04:57 | |
| jterrace (~Adium@dynamic-oit-vapornet-b-18.Princeton.EDU) left irc: Quit: Leaving. | 04:57 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 248 seconds | 05:02 | |
| mwhudson (~mwh@linaro/mwhudson) left irc: Ping timeout: 252 seconds | 05:09 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/1501 | 05:17 |
| ernop (~ernie@199.48.225.86) joined #pypy. | 05:19 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/849 | 05:19 |
| fijal (~fijal@196-215-5-11.dynamic.isadsl.co.za) left irc: Ping timeout: 240 seconds | 05:26 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/666 | 05:27 |
| thinker (~thinker@218-167-101-92.dynamic.hinet.net) joined #pypy. | 05:27 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/1182 | 05:31 |
| PiotrSikora (~none@nginx/adept/piotrsikora) left irc: Excess Flood | 05:32 | |
| pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) left irc: Quit: Leaving | 05:33 | |
| PiotrSikora (~none@nginx/adept/piotrsikora) joined #pypy. | 05:34 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 05:40 | |
| thinker (~thinker@218-167-101-92.dynamic.hinet.net) left irc: Remote host closed the connection | 05:43 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/664 | 05:49 |
| tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection | 06:01 | |
| fijal (~fijal@197.105.98.164) joined #pypy. | 06:03 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 240 seconds | 06:17 | |
| fijal (~fijal@197.105.98.164) left irc: Read error: Connection reset by peer | 06:18 | |
| nettok (~quassel@190.143.176.89) left irc: Remote host closed the connection | 06:31 | |
| ernop (~ernie@199.48.225.86) left irc: Ping timeout: 240 seconds | 06:36 | |
| ernop (~ernie@199.48.225.86) joined #pypy. | 06:36 | |
| ernopp (~ernie@199.48.225.86) joined #pypy. | 06:38 | |
| ernopp (~ernie@199.48.225.86) left irc: Client Quit | 06:41 | |
| ernop (~ernie@199.48.225.86) left irc: Quit: Leaving | 06:41 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/328 | 06:44 |
| gutworth (~benjamin@64-131-9-52.usfamily.net) left irc: Ping timeout: 240 seconds | 06:45 | |
| mat^2 (~mathias@212.130.113.35) left irc: | 06:47 | |
| bbot2 | 3Success: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/776 | 06:49 |
| fermianyon (~lane@c-71-229-21-197.hsd1.al.comcast.net) left irc: Ping timeout: 240 seconds | 06:50 | |
| zain (~textual@c-67-160-201-63.hsd1.ca.comcast.net) joined #pypy. | 06:54 | |
| anish_ (anish@nat/redhat/x-fazxbtihfhegzbcb) left irc: Remote host closed the connection | 06:55 | |
| ernop (~ernie@114.250.129.15) joined #pypy. | 06:59 | |
| zain (~textual@c-67-160-201-63.hsd1.ca.comcast.net) left irc: Quit: Textual IRC Client: http://www.textualapp.com/ | 07:00 | |
| zain (~textual@c-67-160-201-63.hsd1.ca.comcast.net) joined #pypy. | 07:05 | |
| ernop (~ernie@114.250.129.15) left irc: Ping timeout: 240 seconds | 07:17 | |
| fijal (~fijal@197.169.202.78) joined #pypy. | 07:18 | |
| thinker (~thinker@218-167-101-92.dynamic.hinet.net) joined #pypy. | 07:19 | |
| fijal | Alex_Gaynor: ping | 07:20 |
| lmoura (lmoura@gateway/indt/session) left irc: Ping timeout: 240 seconds | 07:25 | |
| fijal (~fijal@197.169.202.78) left irc: Read error: Connection reset by peer | 07:25 | |
| JaRoel|4d (~jaroel|4d@2001:980:7133:1:15c:aba2:a04c:4a8e) left irc: Remote host closed the connection | 07:25 | |
| lmoura (lmoura@gateway/indt/session) joined #pypy. | 07:25 | |
| anish_ (anish@nat/redhat/x-lvwqhbhniajllvkq) joined #pypy. | 07:27 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/317 | 07:41 |
| anish_ (anish@nat/redhat/x-lvwqhbhniajllvkq) left irc: Remote host closed the connection | 07:45 | |
| fijal (~fijal@197.168.235.192) joined #pypy. | 07:47 | |
| fijal (~fijal@197.168.235.192) left irc: Read error: Connection reset by peer | 07:50 | |
| anish_ (anish@nat/redhat/x-fxrbbeswiajqahie) joined #pypy. | 07:51 | |
| anish_ (anish@nat/redhat/x-fxrbbeswiajqahie) left irc: Read error: Connection reset by peer | 07:51 | |
| anish_ (anish@nat/redhat/x-rqvjmxjgjeuholoa) joined #pypy. | 07:53 | |
| anish_ (anish@nat/redhat/x-rqvjmxjgjeuholoa) left irc: Read error: Connection reset by peer | 07:55 | |
| cpets (c358b60a@gateway/web/freenode/ip.195.88.182.10) joined #pypy. | 08:04 | |
| anish_ (anish@nat/redhat/x-xllcckxbdkwozgmz) joined #pypy. | 08:08 | |
| anish_ (anish@nat/redhat/x-xllcckxbdkwozgmz) left irc: Remote host closed the connection | 08:09 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy. | 08:10 | |
| bogner (~bogner@2600:3c03::f03c:91ff:fedf:7ef4) left irc: Quit: Coyote finally caught me | 08:12 | |
| ernop (~ernie@114.250.129.15) joined #pypy. | 08:13 | |
| bogner (~bogner@d50-99-181-221.abhsia.telus.net) joined #pypy. | 08:13 | |
| int3__ (~int3__@220.232.235.85) joined #pypy. | 08:14 | |
| anish_ (anish@nat/redhat/x-taytszxloyuyrlur) joined #pypy. | 08:16 | |
| anish_ (anish@nat/redhat/x-taytszxloyuyrlur) left irc: Read error: Connection reset by peer | 08:17 | |
| bogner` (~bogner@2600:3c03::f03c:91ff:fedf:7ef4) joined #pypy. | 08:18 | |
| anish_ (anish@nat/redhat/x-wbtqdbzgtfxsveos) joined #pypy. | 08:22 | |
| zk (~is@whatit.is) left irc: Read error: Operation timed out | 08:22 | |
| zk (~is@whatit.is) joined #pypy. | 08:23 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 08:24 | |
| anish_ (anish@nat/redhat/x-wbtqdbzgtfxsveos) left irc: Read error: Connection reset by peer | 08:24 | |
| Dulak (~michael@unaffiliated/dulak) left irc: Ping timeout: 240 seconds | 08:25 | |
| amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 252 seconds | 08:25 | |
| anish_ (anish@nat/redhat/x-exsfhvlrwvvqiwap) joined #pypy. | 08:28 | |
| bogner (~bogner@d50-99-181-221.abhsia.telus.net) left irc: Quit: Ex-Chat | 08:29 | |
| int3__ (~int3__@220.232.235.85) left irc: Remote host closed the connection | 08:32 | |
| fijal (~fijal@196-215-121-137.dynamic.isadsl.co.za) joined #pypy. | 08:35 | |
| anish_ (anish@nat/redhat/x-exsfhvlrwvvqiwap) left irc: Read error: Connection reset by peer | 08:36 | |
| anish_ (anish@nat/redhat/x-ofeoymebxwwzrhiv) joined #pypy. | 08:43 | |
| anish_ (anish@nat/redhat/x-ofeoymebxwwzrhiv) left irc: Read error: Connection reset by peer | 08:44 | |
| anish_ (anish@nat/redhat/x-xckeljyowyiyipjr) joined #pypy. | 08:47 | |
| fijal (~fijal@196-215-121-137.dynamic.isadsl.co.za) left irc: Ping timeout: 240 seconds | 08:49 | |
| kenaan | 12cfbolz extradoc 11165f8a7523d2 15/talk/dagstuhl2012/: add the talk I gave at dagstuhl | 08:50 |
| anish_ (anish@nat/redhat/x-xckeljyowyiyipjr) left irc: Read error: Connection reset by peer | 08:50 | |
| anish_ (anish@nat/redhat/x-ybdbhhiixyvtqbyz) joined #pypy. | 08:54 | |
| mikefc | 0.25 minutes of pypy hacking | 08:57 |
| anish_ (anish@nat/redhat/x-ybdbhhiixyvtqbyz) left irc: Read error: Connection reset by peer | 09:08 | |
| hruske (~Gasper@188-230-156-183.dynamic.t-2.net) joined #pypy. | 09:10 | |
| squiddy (~squiddy@g224193131.adsl.alicedsl.de) joined #pypy. | 09:12 | |
| anish_ (anish@nat/redhat/x-mgqoxmmhvhgjjyag) joined #pypy. | 09:21 | |
| thinker (~thinker@218-167-101-92.dynamic.hinet.net) left irc: Remote host closed the connection | 09:23 | |
| antocuni_ (~antocuni@host208-85-dynamic.7-79-r.retail.telecomitalia.it) joined #pypy. | 09:25 | |
| lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy. | 09:30 | |
| mimi (~mimi@97-88-28-179.dhcp.roch.mn.charter.com) left irc: Read error: Connection reset by peer | 09:35 | |
| Husio (husiatyn@oceanic.wsisiz.edu.pl) left irc: Ping timeout: 248 seconds | 09:37 | |
| Husio (husiatyn@oceanic.wsisiz.edu.pl) joined #pypy. | 09:37 | |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) left irc: Quit: fzzzy | 09:38 | |
| G2P (~G2P@fw-asn1.ornis.com) joined #pypy. | 09:41 | |
| lesshaste (~lesshaste@87-194-206-189.bethere.co.uk) left irc: Quit: Leaving | 09:43 | |
| JaRoel (~jaroel|4d@office.fourdigits.nl) joined #pypy. | 09:55 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Ping timeout: 252 seconds | 09:55 | |
| leshaste (~lesshaste@87-194-206-189.bethere.co.uk) joined #pypy. | 10:00 | |
| leshaste | hg clone https://bitbucket.org/pypy/pypy | 10:10 |
| leshaste | abort: HTTP Error 503: Service Unavailable | 10:10 |
| leshaste | is that bitbucket which is down? | 10:10 |
| hakanardo | probably, I cant access it either: | 10:11 |
| leshaste | :( | 10:11 |
| hakanardo | remote: ssh_exchange_identification: Connection closed by remote host | 10:11 |
| hakanardo | abort: no suitable response from remote hg! | 10:11 |
| leshaste | is there another way to get the source? | 10:11 |
| leshaste | a non bitbucket way that is | 10:12 |
| hakanardo | dont know, I have a fairly recent checkout localy | 10:12 |
| leshaste | can I get yours somehow? | 10:12 |
| leshaste | using git I suppose | 10:12 |
| leshaste | or any other method | 10:13 |
| hakanardo | maybe, give me a sec... | 10:13 |
| leshaste | thanks | 10:13 |
| leshaste | hmm.. http://wyvern.cs.uni-duesseldorf.de:5000/ from http://morepypy.blogspot.com/2011/08/visualization-of-jitted-code.html is also down | 10:15 |
| leshaste | or doesn't exist | 10:15 |
| hakanardo | http://morepypy.blogspot.com/2011/08/visualization-of-jitted-code.html works fine here | 10:17 |
| leshaste | sure but does the http://wyvern.cs.uni-duesseldorf.de:5000/ link work for you? | 10:18 |
| leshaste | the "online demo" link | 10:19 |
| hakanardo | ah, no indeed | 10:19 |
| MostAwesomeDude | The demo jitviewer was hammered heavily ever since that post; I bet whoever owns it took it down. | 10:20 |
| leshaste | ok | 10:20 |
| leshaste | I would just like to be able to download pypy at this point :) | 10:20 |
| leshaste | I am trying to get jitviewer to work | 10:20 |
| leshaste | which seems to need the pypy source | 10:20 |
| hakanardo | would the source release on http://pypy.org/download.html do? | 10:23 |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 10:23 | |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 10:23 | |
| hakanardo | or you need more recent? | 10:23 |
| leshaste | that source is fine. it's all at bitbucket isn't it? | 10:24 |
| leshaste | is there a non-bitbucket source link there? | 10:25 |
| hakanardo | hmm, no, sorry | 10:25 |
| leshaste | ok | 10:26 |
| hakanardo | try http://hakan.ardoe.net/~hakan/pypy.tgz | 10:27 |
| leshaste | thanks! | 10:29 |
| hakanardo | it's a .tar.bz2 file dispite it's name... | 10:29 |
| leshaste | :) | 10:29 |
| leshaste | hmm...ImportError: No module named jinja2 | 10:32 |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) left irc: Read error: Connection reset by peer | 10:32 | |
| leshaste | where is that? | 10:32 |
| stakkars (~tismer@p5DDB7A14.dip.t-dialin.net) joined #pypy. | 10:32 | |
| JaRoel (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection | 10:33 | |
| squiddy | leshaste: check the requirements.txt, you'll need those packages | 10:33 |
| leshaste | hmm | 10:33 |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy. | 10:34 | |
| unbit (~unbit@93-58-162-185.ip159.fastwebnet.it) left irc: Remote host closed the connection | 10:35 | |
| hakanardo | I think you can run jitviewer/setup.py develop to install the packages needed | 10:36 |
| christophler (~web49_an_@84.45.87.204) left irc: Remote host closed the connection | 10:36 | |
| leshaste | right I did that | 10:36 |
| hakanardo | also you probably need to mach the pypy source with the pypy binary your using | 10:36 |
| leshaste | I still get ImportError: No module named jinja2 | 10:36 |
| hakanardo | what I sent you should be close to latest nightly | 10:37 |
| leshaste | and I see Installed /usr/local/lib/python2.6/dist-packages/Jinja2-2.6-py2.6.egg | 10:37 |
| leshaste | so I am not sure what is going on | 10:37 |
| leshaste | except maybe a path error | 10:37 |
| hakanardo | you should run the jitviewer under pypy | 10:38 |
| hakanardo | and install the packages under pypy | 10:39 |
| leshaste | http://paste.pocoo.org/show/530299/ | 10:39 |
| leshaste | what do you make of that? | 10:39 |
| leshaste | hakanardo, oh...err.. can you just walk me through that? | 10:39 |
| leshaste | I just am following the README | 10:40 |
| leshaste | where it says python setup.py develop | 10:40 |
| squiddy | have you activated the virtualenv before? | 10:40 |
| leshaste | no I can't as my version is too old it seems | 10:40 |
| hakanardo | right, I think it refers to the "python" within the virtualenv which it sets up to be pypy | 10:41 |
| hakanardo | so you are not using virtualenv? | 10:41 |
| squiddy | virtualenv is a single python file, you can download it in case you can't install a more recent one | 10:41 |
| lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Ping timeout: 240 seconds | 10:42 | |
| hakanardo | the alternative to virtualenv is to manuallt download the source of python-setuptools and then install it under pypy using "pypy setup.py install" | 10:43 |
| hakanardo | and then use it to install the rest of the dependencies | 10:44 |
| leshaste | hmm | 10:44 |
| leshaste | virtualenv comes in a tar ball with lots of stuff | 10:44 |
| squiddy | https://raw.github.com/pypa/virtualenv/master/virtualenv.py, python virtualenv.py envname | 10:45 |
| squiddy | as described ehre http://pypi.python.org/pypi/virtualenv | 10:45 |
| leshaste | ok | 10:46 |
| leshaste | got new virtualenv | 10:46 |
| leshaste | it ran without problem | 10:46 |
| leshaste | went back into the jitviewer directory and ran | 10:46 |
| leshaste | sudo python setup.py develop | 10:46 |
| leshaste | it said it already had everything | 10:46 |
| leshaste | then ran | 10:46 |
| leshaste | jitviewer/bin/jitviewer.py log.pypylog | 10:47 |
| leshaste | ImportError: No module named jinja2 | 10:47 |
| lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy. | 10:47 | |
| hakanardo | the virtualenv call from the readme creates a dir called pypy-viewer | 10:48 |
| hakanardo | use the pythin in there: | 10:48 |
| hakanardo | ./pypy-viewer/bin/python setup.py develop | 10:48 |
| anish_ (anish@nat/redhat/x-mgqoxmmhvhgjjyag) left irc: Quit: Leaving | 10:48 | |
| leshaste | ok did that | 10:49 |
| leshaste | jitviewer/bin/jitviewer.py log.pypylog | 10:49 |
| leshaste | ImportError: No module named jinja2 | 10:49 |
| hakanardo | hmm, what about: pypy-viewer/bin/pypy jitviewer/bin/jitviewer.py log.pypylog | 10:50 |
| leshaste | http://paste.pocoo.org/show/530309/ | 10:51 |
| leshaste | progress? :) | 10:51 |
| hakanardo | ok, that's progress :) | 10:51 |
| voidspace (~voidspace@python/psf/voidspace) joined #pypy. | 10:52 | |
| hakanardo | now we need to mach your binary with your pypy srource code | 10:52 |
| hakanardo | which pypy binary are you using | 10:52 |
| ernop (~ernie@114.250.129.15) left irc: Remote host closed the connection | 10:52 | |
| leshaste | 1.7 | 10:53 |
| hakanardo | ok, then the source is too recent | 10:53 |
| antocuni_ | pff, is it just me or bitbutcket is down? | 10:54 |
| antocuni_ | I get HTTP Error 503 when trying to pull | 10:54 |
| leshaste | antocuni_, it is down :( | 10:54 |
| arigato (~arigo@adsl-89-217-185-214.adslplus.ch) joined #pypy. | 10:54 | |
| leshaste | I don't know if they know | 10:54 |
| Action: antocuni_ tries with ssh | 10:54 | |
| Action: hakanardo already did :) | 10:54 | |
| antocuni_ | hakanardo: and it doesn't work? | 10:55 |
| hakanardo | no | 10:55 |
| antocuni_ | pff | 10:55 |
| anish_ (anish@nat/redhat/x-lwrwmimgrwppqrtm) joined #pypy. | 10:55 | |
| hakanardo | leshaste: ok, try http://hakan.ardoe.net/~hakan/pypy-1.7.tar.bz2 | 10:57 |
| squiddy (~squiddy@g224193131.adsl.alicedsl.de) left irc: Quit: Leaving | 11:00 | |
| leshaste | ImportError: cannot import name 'parse_log_counts' | 11:03 |
| leshaste | :) | 11:03 |
| leshaste | I just copied it to the same place | 11:03 |
| leshaste | http://paste.pocoo.org/show/530317/ | 11:04 |
| barnash (~barnash@89-139-199-136.bb.netvision.net.il) joined #pypy. | 11:04 | |
| hakanardo | argh! | 11:05 |
| leshaste | oh yes... :) | 11:05 |
| hakanardo | maybe you need an older version of the jitviewer aswell | 11:05 |
| leshaste | marvellous | 11:05 |
| leshaste | as an alternative.. maybe you could look at the trace for me? :) | 11:06 |
| leshaste | it might be a *lot* quicker | 11:06 |
| hakanardo | no I would get the same issues as the log was produced with pypy 1.7 and not latest night | 11:07 |
| hakanardo | *nightly* | 11:08 |
| cpets (c358b60a@gateway/web/freenode/ip.195.88.182.10) left irc: Ping timeout: 258 seconds | 11:08 | |
| mfoord (~anonymous@python/psf/voidspace) joined #pypy. | 11:11 | |
| hakanardo | leshaste: do you have a clone of the jitviewer repository? | 11:13 |
| voidspace (~voidspace@python/psf/voidspace) left irc: Quit: Uhm... gotta go | 11:14 | |
| Nick change: mfoord -> voidspace | 11:14 | |
| hakanardo | if so, try version 8b993824df41 | 11:15 |
| leshaste | hakanardo, I meant you could make the log too | 11:16 |
| leshaste | hakanardo, or I could remake the log using the latest pypy? | 11:16 |
| hakanardo | yes that should work | 11:17 |
| hakanardo | grab the latest nightly build here: http://buildbot.pypy.org/nightly/trunk/ | 11:17 |
| unbit (~unbit@93-58-162-185.ip159.fastwebnet.it) joined #pypy. | 11:18 | |
| hakanardo | rm the pypy-viewer dir from virtualenv and recreate it using virtualenv and "setup.py develop" | 11:18 |
| leshaste | thanks.. I am going to give up for a bit | 11:18 |
| leshaste | it's all taken too many hours so far | 11:19 |
| leshaste | I'll try again another day | 11:19 |
| hakanardo | ok, good luck :) | 11:19 |
| leshaste | thanks | 11:19 |
| lmoura (lmoura@gateway/indt/session) left irc: Read error: Operation timed out | 11:28 | |
| lmoura (lmoura@gateway/indt/session) joined #pypy. | 11:29 | |
| leshaste | http://status.bitbucket.org/ | 11:40 |
| arigato (~arigo@adsl-89-217-185-214.adslplus.ch) left irc: Ping timeout: 252 seconds | 11:54 | |
| mikefc | status.bitbucket.org isn't responding for me :( | 11:57 |
| mikefc | nginx gateway error | 11:57 |
| mikefc | ugh. | 11:57 |
| ronny | mikefc: their main nfs died | 12:01 |
| nedbat (~nedbat@python/psf/nedbat) joined #pypy. | 12:02 | |
| leshaste | ronny, how do you know that? | 12:11 |
| witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy. | 12:24 | |
| witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy. | 12:24 | |
| fijal (~fijal@196-215-5-11.dynamic.isadsl.co.za) joined #pypy. | 12:30 | |
| ronny | leshaste: a bird told me | 12:33 |
| leshaste | :) | 12:33 |
| fijal | hey | 12:36 |
| ronny | sup fijal | 12:41 |
| leshaste | hi | 12:45 |
| amaury_ (~amaury_@74.125.57.34) joined #pypy. | 12:45 | |
| fijal | leshaste, MostAwesomeDude: for what is worth, the BaseHTTPServer takes itself down on it's own | 12:49 |
| leshaste | :) | 12:49 |
| fijal | antocuni_: ping | 12:50 |
| antocuni_ | fijal: pong | 12:50 |
| fijal | antocuni_: I'm a bit unhappy with ops_offset being a dict | 12:51 |
| fijal | why not store an offset on ops? | 12:51 |
| antocuni_ | I don't remember the details | 12:51 |
| antocuni_ | but I think it was not to have an extra attribute on all the ops | 12:51 |
| antocuni_ | because we are interested in the offset of only a small subset of those (the ones which are not killed by optimizeopt) | 12:52 |
| fijal | why not a list then? | 12:53 |
| antocuni_ | good point | 12:54 |
| antocuni_ | as I said, I don't remember the details | 12:54 |
| fijal | my slight worry is that we would like to get rid of dict lookups for the backend | 12:55 |
| antocuni_ | yes, it sounds like a good idea | 12:55 |
| fijal | while in general I agree "one more would not make a difference" eventually, I hope to get rid of that :/ | 12:55 |
| Nick change: antocuni_ -> antocuni | 12:55 | |
| antocuni | although we spend relatively little time in the backend, compared to e.g. pyjitpl | 12:56 |
| fijal | yes | 12:57 |
| fijal | but this is also an artifact of say tracing a lot | 12:57 |
| fijal | for example for translate.py or pypy py.test test_numarray.py we do trace quite a lot without compiling | 12:57 |
| fijal | so yes, it's all true and whatnot, but we might not care :) | 12:59 |
| fijal | also there are floating ideas about light-weight tracing | 13:00 |
| fijal | that would stress assembler more | 13:00 |
| nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 245 seconds | 13:13 | |
| antocuni | what are the ideas for lightweight tracing? | 13:14 |
| jnoller (~jnoller@vpn.nasuni.com) joined #pypy. | 13:14 | |
| fijal | jnoller: hi jesse | 13:15 |
| fijal | antocuni: to say trace only pyopcode.py | 13:15 |
| fijal | and leave everything else as calls | 13:15 |
| jnoller | fijal: hola | 13:15 |
| fijal | and don't run optimizations | 13:15 |
| fijal | or so | 13:15 |
| antocuni | fijal: ah, that would be roughly equivalent to a "normal" jit | 13:15 |
| antocuni | and then run full tracing when the code is hot? | 13:15 |
| fijal | yes | 13:16 |
| antocuni | an alternative could be to do "abstract tracing", in which we can actually compile the whole functions before executing | 13:16 |
| antocuni | unsure whether it's a good idea or not | 13:16 |
| fijal | we can experiment | 13:19 |
| aleksi (~aleksi@95-37-48-82.dynamic.mts-nn.ru) joined #pypy. | 13:21 | |
| aleksi (~aleksi@95-37-48-82.dynamic.mts-nn.ru) left irc: Remote host closed the connection | 13:23 | |
| leshaste | I have a(nother) dim question... if you do "if '1' == '2' in python or pypy, what size of the representations of 1 and 2? | 13:44 |
| leshaste | what size are the ... | 13:44 |
| exarkun | which representations? | 13:44 |
| exarkun | eg, there's the code objects, which are basically marshal strings, and roughly compatible with CPython | 13:45 |
| exarkun | small integers are 5 bytes in the marshal format | 13:45 |
| exarkun | (shoulda used banana - 2 bytes ftw) | 13:46 |
| exarkun | But code objects don't do a whole lot by themselves | 13:46 |
| exarkun | if you actually _run_ it, then you've got other representations of the code and data | 13:46 |
| exarkun | and I imagine that's where it gets complicated :) | 13:47 |
| ronny | leshaste: in memory, the '1' and '2' are both 1 character strings, so they will be as big as a 1 character string is | 13:51 |
| leshaste | ok.. the point is to do with type inference and speed :) | 13:52 |
| leshaste | isn't everything | 13:52 |
| leshaste | in C you could imagine they would be one byte each | 13:52 |
| leshaste | and a comparison would be a single instruction | 13:53 |
| ronny | in python you got object headers and indirection | 13:53 |
| leshaste | right...err.. I am no expert | 13:53 |
| leshaste | even in this case? | 13:54 |
| ronny | python has no primitives | 13:54 |
| ronny | only objects | 13:54 |
| leshaste | is it plausible to optimise this out? | 13:54 |
| leshaste | in the sort of case above | 13:54 |
| ronny | yup, if the types in the trace fit, stuff can be unboxed in the assembler | 13:55 |
| leshaste | so in theory you could get a huge space saving? | 13:55 |
| leshaste | as well as a speed up | 13:55 |
| ronny | speed up sure, space saving probably not | 13:56 |
| nirbheek_ (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 13:56 | |
| ronny | the strings are stored on the code object in the above case | 13:56 |
| leshaste | hmm.. so I don't really get it then | 13:56 |
| leshaste | an int could be 32 bits say. Presumably all the headers take up space | 13:57 |
| mat^2 (~mathias@212.130.113.35) joined #pypy. | 13:57 | |
| ronny | leshaste: intermediate objects can be allocated on the stack without any headers for example | 13:58 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 252 seconds | 13:58 | |
| ronny | leshaste: but anything that is in some sense global (like constants or any durect you use inside a functon), its on a code object anyway | 13:58 |
| voidspace (~anonymous@python/psf/voidspace) left irc: Quit: voidspace | 13:59 | |
| leshaste | the more I learn about it the more I realise that the right answer to the question "how do you optimise python code" is " I wouldn't start from here" :) | 13:59 |
| ronny | you cant afford to think like in c | 14:00 |
| leshaste | or even c++ | 14:00 |
| leshaste | or even java it seems | 14:00 |
| anish_ (anish@nat/redhat/x-lwrwmimgrwppqrtm) left irc: Quit: Leaving | 14:00 | |
| ronny | python is truely object oriented, and has much more metadata in the runtime and s more dynamic | 14:00 |
| leshaste | right.. so my naive fantasy was that code that doesn't take advantage of these advanced features could be treated like c++ or java code in some sense | 14:01 |
| leshaste | and the slowdown would only occur "where necessary" | 14:01 |
| leshaste | I don't really mean treated like c++ or java.. just more that a number that is just used as a number can be treated as a number ;) | 14:02 |
| ronny | actualy you can just do that | 14:03 |
| ronny | in part it already works perfect (like intermediate values in code that gets jited) | 14:03 |
| ronny | in other parts is starting to be fine (like the collection strategies that store a list of numeric values in a compact memory repressentation | 14:04 |
| leshaste | don't you need some clever type inference in the interpreter/translator for this to be work well? | 14:04 |
| ronny | and for the rest there is still no knowdegde on whether or not inlining data for example will make sense | 14:04 |
| ronny | leshaste: why inference? the types are known to the objects, so if you create an list and only store a certain type in, that can be optimized | 14:05 |
| asmeurer (~asmeurer@c-174-56-21-245.hsd1.nm.comcast.net) left irc: Quit: asmeurer | 14:05 | |
| ronny | (it will just be painfull if one actually breaks one of the implicit contracts, and the layout needs to be changed | 14:06 |
| leshaste | ok... you don't know if something is constant or not though right? | 14:06 |
| leshaste | for example | 14:06 |
| ronny | all instances of strings and numbers are constant (cause they are immutable) | 14:06 |
| leshaste | ok. I suppose I was thinking also of something a bit different.. like https://bugs.pypy.org/issue980 | 14:07 |
| leshaste | this is not the same I realise | 14:07 |
| leshaste | but if the system knew the number of iterations of the loop I feel it could do something with that information | 14:07 |
| leshaste | (see testham.py and testham-unrolled.py) | 14:07 |
| leshaste | but what I meant above was.. if you know a variable will only ever take the value 5, that seems to allow you represent it more compactly | 14:08 |
| leshaste | than if it might suddenly take some much larger value | 14:08 |
| ronny | i have no idea why you are concerned about +- a few hundret bytes in thatt example, when the real question is, why is a slower than b | 14:10 |
| leshaste | ronny, oh sorry I have mixed topics.. in that example I am only interested in the speed | 14:11 |
| leshaste | ronny, and in that example I wondered if knowing that l = 5 might help the optimiser | 14:12 |
| leshaste | which it could infer by seeing that the variable is unchanged after it is set | 14:12 |
| aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy. | 14:12 | |
| leshaste | fijal says he knows why it is slower | 14:12 |
| leshaste | or as it's a jit, I suppose it can see the value of l at runtime | 14:13 |
| leshaste | it's a pretty amazing speed up from a trivial change to the code | 14:14 |
| leshaste | ronny, what do you think about that speedup? | 14:17 |
| ronny | its interesting | 14:18 |
| gtaylor (~gtaylor@108-196-160-156.lightspeed.tukrga.sbcglobal.net) joined #pypy. | 14:19 | |
| ronny | if i was you, i'd take a look at the differences between the 2 in jitviewer | 14:19 |
| leshaste | right | 14:19 |
| leshaste | I spent several hours trying to get the jitviewer to work this morning! | 14:19 |
| leshaste | I gave up in the end | 14:19 |
| leshaste | I'll try again tomorrow | 14:19 |
| leshaste | it seemed to be permanent version clashes | 14:19 |
| ronny | basically grab a nightly of pypy, grab a recent version of the hg checkout and a recent version of jitviewer, and it should work | 14:19 |
| leshaste | bitbucket was down which didn't help | 14:20 |
| ronny | yeah | 14:20 |
| leshaste | errr. is down | 14:21 |
| leshaste | still not up :( | 14:21 |
| nedbat (~nedbat@python/psf/nedbat) joined #pypy. | 14:21 | |
| ronny | reminds me, we should want mirrors anyway | 14:22 |
| Rhy0lite (dje@nat/ibm/x-adteunbjilwndrdx) joined #pypy. | 14:22 | |
| leshaste | yes! | 14:22 |
| leshaste | or put it on bittorrent :) | 14:22 |
| leshaste | free mirrors | 14:23 |
| leshaste | not so good for dailies maybe | 14:23 |
| ronny | the ngihtly builds are on our server anyway | 14:23 |
| leshaste | how can I get them? | 14:23 |
| leshaste | oh.. the binaries | 14:24 |
| leshaste | I need the source | 14:24 |
| ronny | wrt source, that i still need to fix | 14:25 |
| leshaste | k | 14:25 |
| voidspace (~anonymous@python/psf/voidspace) joined #pypy. | 14:50 | |
| whitelynx (~whitelynx@63.241.75.144) joined #pypy. | 14:52 | |
| anish_ (anish@nat/redhat/x-bashmpdonrkxwsmr) joined #pypy. | 14:53 | |
| fprimex (~fprimex@brent-macbook.sc.fsu.edu) joined #pypy. | 15:02 | |
| anish_ (anish@nat/redhat/x-bashmpdonrkxwsmr) left irc: Quit: Leaving | 15:07 | |
| tos9 (~tos9@unaffiliated/tos9) left irc: Quit: leaving | 15:11 | |
| gutworth (~benjamin@64-131-9-52.usfamily.net) joined #pypy. | 15:11 | |
| tos9 (~tos9@unaffiliated/tos9) joined #pypy. | 15:11 | |
| arigato (~arigo@adsl-89-217-185-214.adslplus.ch) joined #pypy. | 15:17 | |
| kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) joined #pypy. | 15:20 | |
| nirbheek_ (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 252 seconds | 15:33 | |
| espes (espes@2001:470:1f05:ca1::) left irc: Ping timeout: 240 seconds | 15:34 | |
| espes (~espes@123-243-221-248.tpgi.com.au) joined #pypy. | 15:37 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 15:39 | |
| squiddy (~squiddy@g224193131.adsl.alicedsl.de) joined #pypy. | 15:43 | |
| tos9 (~tos9@unaffiliated/tos9) left irc: Quit: leaving | 15:46 | |
| tos9 (~tos9@unaffiliated/tos9) joined #pypy. | 15:46 | |
| thinker (~thinker@123-194-52-90.dynamic.kbronet.com.tw) joined #pypy. | 15:50 | |
| lambacck_ (~chris@d24-150-124-118.home.cgocable.net) joined #pypy. | 15:58 | |
| lahwran | why does the pypy interactive console have four arrows instead of three? | 15:59 |
| arigato (~arigo@adsl-89-217-185-214.adslplus.ch) left irc: Ping timeout: 252 seconds | 16:00 | |
| lucian | lahwran: because it's not CPython | 16:00 |
| lahwran | how specific | 16:00 |
| lambacck (~chris@d24-150-124-118.home.cgocable.net) left irc: Ping timeout: 248 seconds | 16:01 | |
| MostAwesomeDude | IIRC the original reasoning is from back when people ran it as py.py, and it was supposed to signify a Python-in-Python environment. | 16:01 |
| tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy. | 16:01 | |
| MostAwesomeDude | In py.py you could drop down to a Python shell which would let you examine the current PyPy space, and that prompt would have >>> instead. | 16:01 |
| fermianyon (~lane@c-71-229-21-197.hsd1.al.comcast.net) joined #pypy. | 16:02 | |
| dmalcolm (david@nat/redhat/x-gjaowlnqngorprom) joined #pypy. | 16:03 | |
| lahwran | ah | 16:03 |
| lahwran | that makes sense | 16:03 |
| lucian | i don't know if it's been kept out of inertia or because it's still deemed useful | 16:03 |
| lahwran | or maybe because it's just cool :P | 16:04 |
| lahwran | I thought it was something related to speed | 16:04 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 240 seconds | 16:04 | |
| lucian | it'd need 15 > then, wouldn't it? :) | 16:04 |
| lahwran | haha | 16:04 |
| exarkun | There are some optimizations that can be made | 16:06 |
| exarkun | eg perhaps four >s can be replaced by one £ | 16:07 |
| lucian | exarkun: maybe we could involve the JIT in that somehow :) | 16:08 |
| exarkun | (and then you've got about a thousand more arrow code points to choose from when you get tired of that) | 16:08 |
| christophler (~web49_an_@84.45.87.204) joined #pypy. | 16:17 | |
| gutworth (~benjamin@64-131-9-52.usfamily.net) left irc: Ping timeout: 252 seconds | 16:23 | |
| gutworth (~benjamin@64-131-9-52.usfamily.net) joined #pypy. | 16:30 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection | 16:42 | |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy. | 16:45 | |
| dmalcolm (david@nat/redhat/x-gjaowlnqngorprom) left irc: Ping timeout: 240 seconds | 17:00 | |
| nedbat (~nedbat@python/psf/nedbat) left irc: Disconnected by services | 17:05 | |
| nedbat_ (~nedbat@python/psf/nedbat) joined #pypy. | 17:06 | |
| arigato (~arigo@adsl-89-217-185-214.adslplus.ch) joined #pypy. | 17:09 | |
| Action: arigato is in Leysin in the middle of a blizzard (but safely indoor) | 17:10 | |
| fijal | :] | 17:11 |
| fijal | arigato: did you get some good skiing? | 17:12 |
| arigato | yes | 17:12 |
| arigato | the snow is excellent | 17:12 |
| arigato | (and abundant) | 17:12 |
| fijal | cool! | 17:12 |
| arigato | :-) | 17:12 |
| fijal | one thing I won't get here :) | 17:12 |
| Action: fijal has been busy with admin | 17:13 | |
| fijal | moving, visa, etc :/ | 17:13 |
| arigato | getting anywhere close to finish? :-/ | 17:13 |
| fijal | I think I can measure some sort of progress at least | 17:14 |
| fijal | I moved | 17:14 |
| fijal | and I have a visa | 17:14 |
| fijal | now I have to do some numpy branch review at some point | 17:14 |
| arigato | :-) | 17:15 |
| arigato | "good" | 17:15 |
| fijal | arigato: did you follow the hash-randomization discussion on python-dev btw? | 17:15 |
| arigato | ah, no | 17:15 |
| fijal | do you know the entire story? | 17:15 |
| fijal | or should I brief you quickly | 17:15 |
| arigato | ah, that's about the potential security issue? | 17:16 |
| barnash (barnash@89-139-199-136.bb.netvision.net.il) left #pypy. | 17:16 | |
| arigato | indeed, I started to read that thread | 17:16 |
| fijal | so essentially people found out that dictionaries are O(n**2) worst case scenario | 17:16 |
| arigato | yes, that's known | 17:17 |
| fijal | well, apparently not enough | 17:17 |
| fijal | so someone said "now you can have many colliding hashes in post request and CPU will keep spinning" | 17:18 |
| arigato | yes | 17:18 |
| fijal | and as a result python-dev wants to introduce randomization into dict hashes as a security fix | 17:18 |
| arigato | yes | 17:18 |
| fijal | ok, this is pretty much it | 17:18 |
| arigato | :-) | 17:19 |
| Da_Blitz | does pypy intend to do anything about it? | 17:19 |
| fijal | arigato: do you happen to have an opinion? | 17:20 |
| arigato | no | 17:20 |
| kenaan | 12hager ppc-jit-backend 118690df3243d0 15/pypy/jit/backend/ppc/ppcgen/: add INSTANCE_PTR_EQ, INSTANCE_PTR_NE | 17:20 |
| kenaan | 12hager ppc-jit-backend 11e6f908f85e4c 15/pypy/jit/backend/ppc/ppcgen/: implemented CALL_MALLOC_GC and removed old new_XXX ops | 17:20 |
| arigato | I suppose I can be convinced that it would be a good idea to have it | 17:21 |
| arigato | but there are issues in pypy because prebuilt strings have prebuilt hashes | 17:21 |
| fijal | arigato: I can be convinced that having a randomized hash is a good idea to make your tests not rely on dict order :) | 17:21 |
| arigato | fijal: yes :-) | 17:21 |
| fijal | but I'm not convinced myself this can be treated as "security issue" | 17:21 |
| fijal | really | 17:21 |
| Action: Da_Blitz thought part of the issue wasnet just o(n**2) but that the sender also may have control over the ammount (n) of items sent | 17:22 | |
| fijal | Da_Blitz: that is an application problem isn't it? | 17:22 |
| Da_Blitz | heh, this is reminding me os slow loris | 17:22 |
| Da_Blitz | of | 17:23 |
| fijal | Da_Blitz: I think patching previous python versions, like 2.6 is a bad idea | 17:23 |
| fijal | hell, half of the stdlib tests will break | 17:23 |
| gutworth | only 1/64 actually | 17:23 |
| Da_Blitz | yes i have seen that | 17:23 |
| nedbat_ (~nedbat@python/psf/nedbat) left irc: Ping timeout: 255 seconds | 17:24 | |
| Da_Blitz | i am just hthinking there are cases where it may be beyond your control or you are forced to accept a large ammount of keys | 17:24 |
| Da_Blitz | JSON comes to mind | 17:24 |
| fijal | Da_Blitz: if you can accept an arbitrary data, you're screwed | 17:25 |
| fijal | it won't fit in memory | 17:25 |
| exarkun | You can always stop accepting it. | 17:25 |
| fijal | or kill the rogue process | 17:25 |
| lahwran | is there a "make install" equivalent in pypy? or, how would I install pypy 1.7 from the release archive to my syste | 17:26 |
| lahwran | +m | 17:26 |
| fijal | lahwran: use virtualenv, seriously | 17:26 |
| kenaan | 12hakanardo jit-usable_retrace_2 1175ff444ada2d 15/pypy/jit/metainterp/: Make kill_consts separate from force_at_end_of_preamble to be able to call it earlier and thereby ... | 17:26 |
| kenaan | 12hakanardo jit-usable_retrace_2 114d40b1df8e90 15/pypy/jit/metainterp/optimizeopt/unroll.py: Dont rename boxes in the label, only their bidnings to values | 17:26 |
| lahwran | fijal: ... | 17:26 |
| kenaan | 12hakanardo jit-usable_retrace_2 1176eace813ed4 15/pypy/jit/metainterp/test/test_ajit.py: fix tests, we no longer specialice retraces on constants | 17:26 |
| kenaan | 12hakanardo jit-usable_retrace_2 11ed9ad0e9eacc 15/pypy/jit/metainterp/test/test_send.py: fix test (see comment) | 17:26 |
| kenaan | 12hakanardo jit-usable_retrace_2 11ba02b282da2e 15/pypy/: extract loops at end of bridges too | 17:26 |
| fijal | lahwran: yes? | 17:26 |
| lahwran | that's not helpful | 17:27 |
| exarkun | fijal: It'd be nice to have a dict method describing how collide-y the data is | 17:27 |
| exarkun | It's hard to make a decision about what to do now, because it's hard to recognize you're under attack. | 17:27 |
| lahwran | I can use virtualenv, but I want pypy system installed. how do I go about this? | 17:27 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/665 [12hakanardo, jit-usable_retrace_2] | 17:27 |
| fijal | lahwran: I seriously don't know, never did it | 17:28 |
| lahwran | specifically, I want typing `pypy` to start the pypy interpreter :| | 17:28 |
| Action: Da_Blitz dosent expect everyone to know to put such safeguards in | 17:28 | |
| fijal | lahwran: put in on your PATH | 17:28 |
| fijal | (I do that) | 17:28 |
| lahwran | bah, why does there have to be an easy way out of everything | 17:28 |
| Action: lahwran grudgingly puts it on the path | 17:28 | |
| fijal | exarkun: I guess write this to python-dev? | 17:29 |
| Action: lucian puts pypy in ~/.local/pypy with a symlink in ~/.local/pypy/bin/pypy | 17:29 | |
| lessshaste (~lesshaste@87-194-206-189.bethere.co.uk) joined #pypy. | 17:29 | |
| fijal | I'm sure *this* thing will be 3.4 only | 17:29 |
| fijal | or so | 17:29 |
| exarkun | fijal: no, I don't talk to them | 17:29 |
| exarkun | it's a waste of time, and no fun on top of that | 17:29 |
| Da_Blitz | does pypy use the same hash algo as cpython? | 17:30 |
| fijal | Da_Blitz: it does not really matter | 17:30 |
| fijal | it's deterministic | 17:30 |
| Da_Blitz | i would have thought it would affect dict order | 17:30 |
| fijal | it does | 17:30 |
| fijal | but you can do DDOS independently of algo | 17:31 |
| leshaste (~lesshaste@87-194-206-189.bethere.co.uk) left irc: Ping timeout: 255 seconds | 17:31 | |
| antocuni | lahwran: the recommended way to put install pypy is to put it e.g. in /opt/pypy-1.7 | 17:31 |
| lahwran | antocuni: ah | 17:31 |
| antocuni | and then make a symlink in /usr/bin/pypy pointing to the binary | 17:31 |
| Da_Blitz | actually i was wondering how pypy handles the diffrent dict ordering for doctest | 17:31 |
| Da_Blitz | not anything to do with a DOS | 17:31 |
| fijal | Da_Blitz: it doesn't I think | 17:31 |
| lahwran | Da_Blitz: that's a bug in doctest | 17:31 |
| Action: Da_Blitz is aware | 17:31 | |
| fijal | Da_Blitz: if your tests rely on dict ordering, it is your problem | 17:31 |
| Da_Blitz | i was refering to the stdlib doctests or other testing | 17:33 |
| fijal | Da_Blitz: we definitely wont do a new 1.7 release just because dicts are O(n^2) | 17:34 |
| fijal | well | 17:35 |
| fijal | dicts were always O(n^2), just because people got accepted to CCC to be precise | 17:35 |
| fijal | exarkun: note that python-dev solution is unideal | 17:36 |
| fijal | I'm sure you can get the seed by a timing attack | 17:37 |
| fijal | so seed per-run is not good enough | 17:37 |
| nedbat (~nedbat@python/psf/nedbat) joined #pypy. | 17:37 | |
| lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Ping timeout: 252 seconds | 17:37 | |
| espes (~espes@123-243-221-248.tpgi.com.au) left irc: Ping timeout: 240 seconds | 17:38 | |
| G2P (~G2P@fw-asn1.ornis.com) left irc: Quit: Leaving. | 17:39 | |
| Alex_Gaynor | fijal: pong | 17:41 |
| Alex_Gaynor | fijal, arigato: FWIW I strongly beleive if CPython fixes this, PyPy must as well. I'm aware of the prebuilt string issue, that's "too bad" IMO, it's not like it's a huge optimization | 17:42 |
| arigato (~arigo@adsl-89-217-185-214.adslplus.ch) left irc: Ping timeout: 252 seconds | 17:43 | |
| fijal | yeah, I fear so | 17:45 |
| fzzzy (~donovan@76-198-130-19.lightspeed.mtvwca.sbcglobal.net) joined #pypy. | 17:45 | |
| espes (~espes@123-243-221-248.tpgi.com.au) joined #pypy. | 17:45 | |
| fijal | Alex_Gaynor: do you have *any* clue what's going on with our 18 numpy branches? | 17:45 |
| Alex_Gaynor | fijal: review axisops :) | 17:46 |
| fijal | I did | 17:47 |
| Alex_Gaynor | did matti handle your review? | 17:47 |
| Alex_Gaynor | if soplease review again :) | 17:47 |
| fijal | no | 17:48 |
| Alex_Gaynor | there's a patch for ndarray.{var,std} I think | 17:48 |
| fijal | yes | 17:48 |
| fijal | it's on issue tracker | 17:48 |
| Alex_Gaynor | that should be easy to review, it needs some stylistic help though I think | 17:48 |
| fijal | I'm speaking about branches | 17:48 |
| Alex_Gaynor | numpy-dtype-strings is closed | 17:49 |
| Alex_Gaynor | numpy-ndim-size can be closed I think? | 17:49 |
| Alex_Gaynor | numpy-sort was never done correctly and is all out of date anyways | 17:49 |
| Alex_Gaynor | I don't know about -random, -data-buffer, or -indexing-by-arrays | 17:50 |
| Alex_Gaynor | numpy-dtype can be closed | 17:50 |
| Alex_Gaynor | I don't know about numpy-comparison | 17:50 |
| Alex_Gaynor | numpy-complex can be closed | 17:50 |
| fijal | can you close all that you say can be close :) | 17:50 |
| Alex_Gaynor | sure | 17:50 |
| JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection | 17:53 | |
| kenaan | 12alex_gaynor numpy-dtype-strings 112adf19881a7c 15/: close merged branch | 17:54 |
| kenaan | 12alex_gaynor numpy-ndim-size 1151e67e28230a 15/: close this branch, all its features were added elsewhere as far as I can tell | 17:54 |
| kenaan | 12alex_gaynor numpy-sort 11aaab53d723c0 15/: close branch, it's been totally invalidated, create a fresh branch to work on it | 17:55 |
| kenaan | 12alex_gaynor numpy-dtype 11e1b50a7fd007 15/: close branch, a different approach was taken | 17:55 |
| kenaan | 12alex_gaynor numpy-complex 111436740d3b9b 15/: close branch, different approach taken | 17:55 |
| fprimex (~fprimex@brent-macbook.sc.fsu.edu) left irc: Read error: Connection reset by peer | 17:55 | |
| kenaan | 12alex_gaynor jit-raw-array-of-struct 11c260a0d96e73 15/: close, never went anywhere and not needed anymore | 17:55 |
| rekamso (~textual@67.51.82.66) joined #pypy. | 17:55 | |
| arigato (~arigo@adsl-84-227-97-253.adslplus.ch) joined #pypy. | 18:00 | |
| dmalcolm (david@nat/redhat/x-gqvurvbbnyyzentj) joined #pypy. | 18:03 | |
| idnar (~quassel@unaffiliated/idnar) left irc: Remote host closed the connection | 18:07 | |
| idnar (~quassel@unaffiliated/idnar) joined #pypy. | 18:10 | |
| fijal | Alex_Gaynor: thanks | 18:17 |
| Alex_Gaynor | fijal: np | 18:18 |
| fijal | meh | 18:19 |
| fijal | waking up at 6 didn't make me want to work :/ | 18:19 |
| arigato (~arigo@adsl-84-227-97-253.adslplus.ch) left irc: Ping timeout: 252 seconds | 18:22 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/665 [12hakanardo, jit-usable_retrace_2] | 18:29 |
| rekamso (~textual@67.51.82.66) left irc: Ping timeout: 252 seconds | 18:33 | |
| Alex_Gaynor | fijal: does https://gist.github.com/1566551 look correct to you? | 18:36 |
| fijal | it's not silly btw | 18:37 |
| fijal | with minimark at least | 18:37 |
| Alex_Gaynor | fijal: it's not silly *with minimark* | 18:37 |
| amaury_ (~amaury_@74.125.57.34) left irc: Ping timeout: 240 seconds | 18:39 | |
| fijal | yes | 18:39 |
| tilgovi (~randall@couchdb/developer/tilgovi) left irc: Read error: Connection reset by peer | 18:50 | |
| rekamso (~textual@63-228-127-70.dia.static.qwest.net) joined #pypy. | 18:56 | |
| voidspace (~anonymous@python/psf/voidspace) left irc: Quit: voidspace | 18:59 | |
| andrewfr (~andrew@modemcable211.101-70-69.static.videotron.ca) joined #pypy. | 19:11 | |
| fijal | meh | 19:15 |
| fijal | I don't seem to be able to work :/ | 19:15 |
| hongminhee (u295@gateway/web/irccloud.com/x-mfozvlkzreawkyvp) left irc: | 19:16 | |
| hongminhee (~dahlia@175.125.21.72) joined #pypy. | 19:19 | |
| ericflo (~ericflo@75.103.8.110) joined #pypy. | 19:21 | |
| mattip (~chatzilla@109.65.107.23) joined #pypy. | 19:23 | |
| rekamso | is the stackless (enabling infinite recursion) part of pypy still on hold? I know greenlets are available | 19:25 |
| fijal | yes, it's there | 19:26 |
| rekamso | and enabled? | 19:26 |
| rekamso | wait, nvm I'm using 1.5 | 19:26 |
| amaury_ (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy. | 19:29 | |
| ronny | fijal: wasnt stackless transform deprecated for continulets? | 19:37 |
| ronny | rekamso: what exactly do you mean by "infinite recursion"? | 19:38 |
| fijal | ronny: yes | 19:38 |
| fijal | but stackless features work | 19:38 |
| fijal | hakanardo: ping | 19:39 |
| fijal | Alex_Gaynor: ping | 19:40 |
| arigato (~arigo@adsl-84-227-97-253.adslplus.ch) joined #pypy. | 19:40 | |
| Alex_Gaynor | fijal: pong | 19:40 |
| fijal | Alex_Gaynor: you know why I can no longer do this: http://paste.pocoo.org/show/530562/ | 19:41 |
| fijal | ? | 19:41 |
| andrewfr | @arigato thanks for the STM answer | 19:41 |
| fijal | arigato: or you might know? | 19:41 |
| mwhudson (~mwh@linaro/mwhudson) joined #pypy. | 19:41 | |
| arigato | hi | 19:41 |
| andrewfr | still have compilation problems | 19:41 |
| arigato | andrewfr: you're welcome, I hope this helps | 19:41 |
| andrewfr | I have been looking over the targetdemo programme | 19:42 |
| arigato | fijal: uh? obscure | 19:42 |
| fijal | arigato: it complains with (this is test_runner) http://paste.pocoo.org/show/530566/ | 19:43 |
| fijal | I can put the entire TB | 19:43 |
| fijal | if you want | 19:43 |
| andrewfr | i see transaction_boundary but no begin_transaction, is begin_transaction implicit with the start of the thread? | 19:44 |
| arigato | fijal: ah, well, the descr of the final JUMP needs to be an instance of TargetToken | 19:44 |
| fijal | arigato: then how does all the test_optimizeopt work? | 19:45 |
| fijal | and everything else? | 19:45 |
| fijal | without passing descr= everywhere | 19:45 |
| arigato | fijal: thinking about it :-)... | 19:45 |
| Alex_Gaynor | fijal: the tests auto add them probably | 19:45 |
| fijal | Alex_Gaynor: yes, how? | 19:46 |
| arigato | fijal: there is a convert_old_style_to_targets() call in test_optimizeopt | 19:46 |
| fijal | uh | 19:46 |
| fijal | so I need to invent a jump descrt | 19:46 |
| fijal | and put it in the label | 19:47 |
| fijal | ok | 19:47 |
| antocuni (~antocuni@host208-85-dynamic.7-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds | 19:47 | |
| arigato | yes, seems so | 19:47 |
| fijal | arigato: do you have an example in tests that uses that? | 19:47 |
| arigato | grep for TargetToken(token) in test_optimizebasic.py | 19:47 |
| arigato | or e.g. in test/runner_test.py | 19:48 |
| arigato | (if that really helps, which I doubt) | 19:48 |
| fijal | meh :/ | 19:48 |
| fijal | well yes | 19:48 |
| fijal | I'm sure you can patch a loop | 19:48 |
| fijal | but having support in parser even better | 19:48 |
| fijal | pfff | 19:49 |
| fijal | few-yaks-away | 19:49 |
| arigato | I missed convert_old_style_to_targets() | 19:49 |
| arigato | I would also have preferred proper support in the parser | 19:49 |
| Action: fijal just does it | 19:49 | |
| fijal | what does label accept? | 19:49 |
| fijal | how about | 19:49 |
| fijal | label(1) | 19:49 |
| fijal | jump(descr=1) | 19:49 |
| fijal | ? | 19:49 |
| arigato | label(i1, i2, descr=t1) | 19:50 |
| arigato | jump(descr=t1) | 19:50 |
| arigato | ? | 19:50 |
| fijal | and I have to write t1 by hand? | 19:50 |
| fijal | which would be invariably boring? | 19:50 |
| andrewfr (andrew@modemcable211.101-70-69.static.videotron.ca) left #pypy. | 19:50 | |
| arigato | ok, then "descr=1" and special-case it in the parser to make the TargetToken automatically | 19:50 |
| fijal | yes | 19:50 |
| fijal | AssertionError: You should not import module pypy.jit.metainterp.history | 19:56 |
| fijal | ok what is that? | 19:56 |
| arigato | ? | 19:57 |
| Alex_Gaynor | hehe | 20:00 |
| kenaan | 12fijal better-jit-hooks 116ee1610cf4a4 15/pypy/jit/tool/: Implement inventing descrs for jumps and labels (and everything else) | 20:03 |
| kenaan | 12fijal better-jit-hooks 1197c7263a2d0d 15/pypy/jit/tool/: invent new descrs only for labels | 20:03 |
| fijal | arigato: http://dpaste.com/682269/ | 20:04 |
| fijal | guess what sort of code generated that assembler? | 20:05 |
| fijal | or am I missing something? | 20:05 |
| fijal | IMO the code looks like this: | 20:05 |
| fijal | label(i0) | 20:05 |
| fijal | i1 = int_add(i0, i0) | 20:06 |
| fijal | jump(i1) | 20:06 |
| fijal | arigato: what are those adds doing? | 20:08 |
| arigato | they are adding to counter of how many times the code was run | 20:08 |
| arigato | we have two of them: one at the start of the loop, and one just after the label() | 20:09 |
| arigato | (it's the stuff that you added, I think :-) | 20:09 |
| arigato | (it's only generated in some debugging case) | 20:10 |
| fijal | oh right :) | 20:10 |
| fijal | waking up at 6 does not help | 20:10 |
| fijal | why is debugging on? | 20:10 |
| fijal | ah uh | 20:10 |
| fijal | there was another issue that I never got to fix :/ | 20:11 |
| arigato | I think it's turned on automatically if we use PYPYLOG=jit-backend | 20:11 |
| mtigas (~mtigas@users.spokesman.com) joined #pypy. | 20:12 | |
| fijal | arigato: untranslated, PYPYLOG is meaningless | 20:12 |
| fijal | it's always on | 20:13 |
| fijal | have_debug_prints will return you True | 20:13 |
| fijal | and you can't do anything about it | 20:13 |
| fijal | would be cool if you could | 20:13 |
| arigato | you can say "we_are_translated() and have_debug_prints()" | 20:13 |
| mtigas (~mtigas@users.spokesman.com) left irc: Client Quit | 20:14 | |
| fijal | no | 20:14 |
| fijal | I'm fine with calling set_debug by hand | 20:14 |
| fijal | but maybe indeed they should not be enabled by default untranslated | 20:14 |
| fijal | nobody complained so far :) | 20:14 |
| fijal | arigato: uh, do we have *any* test involving a bridge in test_optimizeopt? | 20:22 |
| arigato | I doubt it | 20:22 |
| fijal | :/ | 20:22 |
| fijal | not good :/ | 20:22 |
| Alex_Gaynor | fijal: there are somewhere, I think. | 20:31 |
| jnoller (~jnoller@vpn.nasuni.com) left irc: Quit: Leaving... | 20:32 | |
| Dulak (~michael@unaffiliated/dulak) joined #pypy. | 20:39 | |
| Rhy0lite (dje@nat/ibm/x-adteunbjilwndrdx) left irc: Quit: Leaving | 20:45 | |
| kenaan | 12fijal better-jit-hooks 1162b1ef8e5cd7 15/pypy/jit/backend/: add extra return values from assemble_loop/assemble_bridge | 20:50 |
| kenaan | 12fijal better-jit-hooks 1174cc4b1b667e 15/pypy/: finish refactoring - move on_compile/on_compile hooks to jitportal, probably breaks pypy | 20:50 |
| fijal | wuhu | 20:51 |
| kenaan | 12fijal better-jit-hooks 11404a51debbaa 15/pypy/jit/backend/test/runner_test.py: skip pointless test on llgraph | 20:54 |
| Alex_Gaynor | fijal: cool, I've almost got the data model finished off | 20:57 |
| fijal | uh, my cpu is boiling | 20:58 |
| fijal | I'll just kick the buildbot | 21:00 |
| kenaan | 12fijal better-jit-hooks 11d329360b0a1c 15/pypy/jit/backend/llsupport/test/test_runner.py: fix a test | 21:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1912 [12fijal, better-jit-hooks] | 21:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/777 [12fijal, better-jit-hooks] | 21:00 |
| fijal | Alex_Gaynor: we now even get assembler :) | 21:01 |
| fijal | "good" I guess | 21:02 |
| fijal | so python-dev consensus seems to be to have an option | 21:03 |
| Alex_Gaynor | fijal: neat, is there enough info to interleave it with the sourceode and resops? | 21:03 |
| fijal | off by default | 21:03 |
| Alex_Gaynor | fijal: off by default is totally wrong, I'll argue against that | 21:03 |
| fijal | Alex_Gaynor: for 2.6/2.7? | 21:03 |
| Alex_Gaynor | yes | 21:03 |
| fijal | Alex_Gaynor: yesish, not quite though | 21:03 |
| fijal | because ops_offsets are separate | 21:03 |
| fijal | :/ | 21:03 |
| fijal | I guess I can just pass it along | 21:03 |
| fijal | break, coffee and back to work | 21:04 |
| fijal | Alex_Gaynor: the applevel part is not done yet | 21:04 |
| fijal | I want to improve on that as well | 21:04 |
| Alex_Gaynor | fijal: well, that's fine, I'm just working on backend stuff ATM, preparing to write tests :) | 21:04 |
| fijal | note that the current logparse will totally not work | 21:05 |
| fijal | which does quite a lot of work, like recombination and stuff | 21:05 |
| Alex_Gaynor | I don't need a log parser since i"m doing it in memory :) | 21:05 |
| Alex_Gaynor | in memory of the host process that is | 21:05 |
| lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy. | 21:07 | |
| lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 252 seconds | 21:13 | |
| fijal | well | 21:18 |
| fijal | you need to reconnect stuff at some point | 21:18 |
| fijal | you can do it on the server side obviously | 21:19 |
| fijal | anyone has anything against naming jitdrivers | 21:20 |
| fijal | ? | 21:20 |
| Alex_Gaynor | fijal: ideally I want the client to do the interleaving and send the server ASM, RESOP, and SOURCE chunks for each label | 21:21 |
| Alex_Gaynor | seems way easier that way | 21:21 |
| fijal | well | 21:22 |
| fijal | the more computations you do on the client side the less likely you're able to run this together with whatever performance assesment | 21:22 |
| fijal | unless you do it post-factum | 21:22 |
| fijal | probably a good idea | 21:22 |
| fijal | my idea was to open a file and just dump the information there | 21:22 |
| Alex_Gaynor | fijal: yeah, I'll do it afterwords, store data in the hook and transmit on completion | 21:22 |
| Alex_Gaynor | fijal: another cool idea I have (for much much later) is a --live flag, where it does live transmission and the site auto updates as it gets new data | 21:23 |
| fijal | yeah | 21:24 |
| lesshaste (~lesshaste@87-194-206-189.bethere.co.uk) joined #pypy. | 21:26 | |
| fijal | coffee coffee | 21:28 |
| fijal | arigato: ping | 21:29 |
| kenaan | 12fijal better-jit-hooks 11032bbe1b32c3 15/pypy/: add name attribute to jitdrivers. start shifting code around in module/pypyjit | 21:30 |
| voidspace (~anonymous@87-194-212-65.bethere.co.uk) joined #pypy. | 21:30 | |
| voidspace (~anonymous@87-194-212-65.bethere.co.uk) left irc: Changing host | 21:30 | |
| voidspace (~anonymous@python/psf/voidspace) joined #pypy. | 21:30 | |
| mattip (chatzilla@109.65.107.23) left #pypy. | 21:33 | |
| lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy. | 21:45 | |
| arigato | fijal: pong | 21:45 |
| fijal | arigato: so, I was thinking if it's possible to somehow merge optimizing and backend | 21:45 |
| fijal | but I guess not really, since backend needs an extra pass | 21:45 |
| arigato | yes, and optimize may need several passes (for the portions between labels), can cancel everything, etc. | 21:48 |
| DanielHolth (~dholth@ip98-180-34-112.ga.at.cox.net) left irc: Read error: Operation timed out | 21:49 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1912 [12fijal, better-jit-hooks] | 21:53 |
| lesshaste | given a list, how would you make a new list consisting of all consecutive pairs (including overlaps)? | 21:55 |
| lesshaste | oops.. wrong channel | 21:55 |
| lesshaste | sorry | 21:55 |
| kenaan | 12fijal better-jit-hooks 119f4f6c879538 15/pypy/jit/metainterp/: pass also ops_offset, for good measure | 21:58 |
| kenaan | 12fijal better-jit-hooks 119e3906158e08 15/pypy/rlib/jit.py: oops | 21:58 |
| rekamso (~textual@63-228-127-70.dia.static.qwest.net) left irc: Ping timeout: 252 seconds | 22:05 | |
| derdon (~derdon@pD9E1CB12.dip.t-dialin.net) joined #pypy. | 22:13 | |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1913 [12fijal, better-jit-hooks] | 22:13 |
| arigato | /usr/bin/ld: cannot find -lc | 22:17 |
| Action: arigato gives up again on his attempt to compile gcc trunk | 22:17 | |
| arigato | (on a perfectly standard ubuntu 64, with "configure; make") | 22:17 |
| mwhudson | make bootstrap? | 22:19 |
| Rhyolite | doesn't matter for trunk | 22:19 |
| mwhudson | ah right | 22:19 |
| Rhyolite | but ubuntu moves libraries around | 22:19 |
| Rhyolite | tsk tsk | 22:19 |
| mwhudson | i guess i last built gcc in the 90s... | 22:20 |
| Alex_Gaynor | arigato: if you install enough 3rd party libraries and mess with enough things it'll eventually work. no one documented this process :/ | 22:20 |
| arigato | Alex_Gaynor: I'm getting closer every evening that I spend half an hour hacking at it before running away | 22:21 |
| arigato | but of course I have no clue how far the goal is | 22:22 |
| Rhyolite | Alex_Gaynor: are you referring to gmp/mpfr/mpc or cloog? | 22:22 |
| Rhyolite | cloog is not *required* | 22:22 |
| Rhyolite | arigato: did the link line include an explicit search path? | 22:23 |
| Alex_Gaynor | Rhyolite: I don't know, it was like 2 months ago and I blokced it from my mind | 22:25 |
| Rhyolite | yes, better to worry about Django tuple choices | 22:25 |
| Alex_Gaynor | Rhyolite: heh, I'm working on the performance command center for pypy | 22:26 |
| arigato | work on anything, it sounds more productive | 22:26 |
| Alex_Gaynor | http://tracebin.ep.io/trace/1000/compiled/1/ is the preview :) | 22:26 |
| Rhyolite | neat! | 22:28 |
| aboudreault (~alanb@osgeo/member/aboudreault) left irc: Quit: Leaving | 22:28 | |
| fijal | Alex_Gaynor: I still hate the CSS | 22:29 |
| fijal | I think jitviewer's is better | 22:29 |
| aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy. | 22:30 | |
| Rhyolite | fijal: How I Learned to Stop Worrying and Love CSS/JavaScript/AJAX | 22:30 |
| fijal | Alex_Gaynor: not to mention that on my tiny screen I have things that have scrollbars that occypy roughly half of the screen | 22:31 |
| fijal | while the rest is perfectly usable white | 22:31 |
| Rhyolite | everyone's a critic! ;-) | 22:32 |
| aboudreault (~alanb@osgeo/member/aboudreault) left irc: Client Quit | 22:33 | |
| fijal | grumble :/ | 22:36 |
| fijal | arigato: how can I cast llmemory.GCREF to PyCode instance? | 22:37 |
| Action: arigato gives gcc another try on a Debian 32 | 22:37 | |
| arigato | fijal: annlowlevel.cast_ptr_to_instance() | 22:37 |
| fijal | cast_base_ptr_to_instance(PyCode, cast_opaque_ptr(robject.OBJECT, gcref))? | 22:37 |
| arigato | yes | 22:37 |
| fijal | yeah, but I have a GCREF | 22:37 |
| Action: fijal has fun casting greenkey boxes to some python objects to expose them at applevel | 22:38 | |
| arigato | robject.OBJECTPTR | 22:38 |
| arigato | rclass.OBJECTPTR, even | 22:39 |
| voidspace (~anonymous@python/psf/voidspace) left irc: Quit: voidspace | 22:42 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/777 [12fijal, better-jit-hooks] | 22:42 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/778 [12fijal, better-jit-hooks] | 22:42 |
| fijal | cool | 22:44 |
| fijal | arigato: it seems to work | 22:44 |
| arigato | :-) | 22:44 |
| kenaan | 12fijal better-jit-hooks 11ad94daae774d 15/pypy/jit/metainterp/compile.py: fix tests | 22:44 |
| fijal | arigato: now you can actually have a hook that would point you to a just compiled assembler | 22:44 |
| kenaan | 12fijal better-jit-hooks 11e7ce390271a7 15/pypy/module/pypyjit/: fix pypyjit module, more work tomorrow | 22:44 |
| fijal | you can use ctypes to disassemble that part (from memory) | 22:44 |
| fijal | `if you're smart, you can even have a go at modifying it by hand :) | 22:45 |
| fijal | and see if this improves anything :) | 22:45 |
| fijal | at least that's the idea, a bit impossible to test without full compilation though | 22:46 |
| fijal | heh | 22:49 |
| fijal | I gonna write a blog post modifying assembler on the fly ;-) | 22:49 |
| fijal | how cool is that | 22:49 |
| arigato | :-) | 22:50 |
| fijal | we already do it for tests, but tss.... | 22:51 |
| arigato | ...nope, gcc on Debian 32 just fails to build differently | 22:51 |
| Rhyolite | arigato: really? | 22:52 |
| Rhyolite | debian / ubuntu really has weird directory layout | 22:52 |
| arigato | that doesn't look like a directory layout issue | 22:52 |
| Rhyolite | do you have the libraries installed? | 22:53 |
| Rhyolite | the initial configure normally finds missing dependencies | 22:53 |
| arigato | I suppose | 22:53 |
| arigato | extract: http://paste.pocoo.org/show/530676/ | 22:53 |
| oal (u4126@gateway/web/irccloud.com/x-icsmsdfcxuezfibg) left irc: Read error: Connection reset by peer | 22:54 | |
| bfirsh (u1308@gateway/web/irccloud.com/x-qgqdwwoyjivsucqk) left irc: Remote host closed the connection | 22:54 | |
| alexsuraci (u1962@gateway/web/irccloud.com/x-mlfztzslyrlfzbko) left irc: Remote host closed the connection | 22:54 | |
| fijal | arigato: and people complain building pypy is too hard ;-) | 22:54 |
| Alex_Gaynor (u1246@gateway/web/irccloud.com/x-mqtrohrcnigifexe) left irc: Remote host closed the connection | 22:54 | |
| arigato | :-) | 22:54 |
| fijal | uh | 22:54 |
| fijal | I guess at 10 arguments to a function would be cool to have a container object :/ | 22:55 |
| Action: arigato tries with "configure --disable-lto" | 22:55 | |
| arigato | nope | 22:56 |
| oal (u4126@gateway/web/irccloud.com/x-xmvfknsrhzjydqmi) joined #pypy. | 22:57 | |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/195 | 23:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/767 | 23:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/1004 | 23:00 |
| bbot2 | Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/23 | 23:00 |
| alexsuraci (u1962@gateway/web/irccloud.com/x-iaebohqulkqcxjvh) joined #pypy. | 23:02 | |
| arigato (~arigo@adsl-84-227-97-253.adslplus.ch) left irc: Quit: See you | 23:06 | |
| Alex_Gaynor (u1246@gateway/web/irccloud.com/x-tirznnnobywzpxgd) joined #pypy. | 23:10 | |
| #pypy: mode change '+o Alex_Gaynor' by ChanServ!ChanServ@services. | 23:10 | |
| whitelynx (~whitelynx@63.241.75.144) left irc: Quit: Ex-Chat | 23:17 | |
| gtaylor (~gtaylor@108-196-160-156.lightspeed.tukrga.sbcglobal.net) left irc: Ping timeout: 240 seconds | 23:24 | |
| mwhudson (~mwh@linaro/mwhudson) left irc: Ping timeout: 252 seconds | 23:27 | |
| nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 276 seconds | 23:32 | |
| bbot2 | 4Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1913 [12fijal, better-jit-hooks] | 23:36 |
| arkx (arkx@hilla.kapsi.fi) left #pypy. | 23:40 | |
| --- Fri Jan 6 2012 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!