| CIA-55 | 03agaynor 07roundup * 10#797/Deadlock with threads: (log message trimmed) | 00:00 |
|---|---|---|
| CIA-55 | [new] Running this code under PyPy trunk (or 1.5) on Linux 64 results in a deadlock: | 00:00 |
| CIA-55 | import threading | 00:00 |
| CIA-55 | def get_results(): | 00:00 |
| CIA-55 | pass * 14https://bugs.pypy.org/issue797 | 00:00 |
| ramusara (~ramusara@220.156.210.236.user.e-catv.ne.jp) joined #pypy. | 00:01 | |
| stakkars_ (~tismer@91-66-101-126-dynip.superkabel.de) joined #pypy. | 00:06 | |
| kisielk (~kamil@zymeworks.com) left irc: Remote host closed the connection | 00:10 | |
| zain (~zain@69.38.221.130) left irc: Ping timeout: 240 seconds | 00:13 | |
| zain (~zain@69.38.221.130) joined #pypy. | 00:14 | |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 252 seconds | 00:34 | |
| dgl (~dgl@109.86.165.231) left irc: Quit: Leaving... | 00:40 | |
| etrepum (~bob@accessnat4.mochimedia.net) left irc: Ping timeout: 240 seconds | 00:49 | |
| rokujyouhitoma (~rokujyouh@EM114-48-189-222.pool.e-mobile.ne.jp) joined #pypy. | 00:52 | |
| rokujyouhitoma (~rokujyouh@EM114-48-189-222.pool.e-mobile.ne.jp) left irc: Client Quit | 00:53 | |
| Sho_ (~EHS1@kde/hein) joined #pypy. | 01:22 | |
| stakkars (~tismer@91-66-101-126-dynip.superkabel.de) left irc: Quit: schnarch | 01:56 | |
| Nick change: stakkars_ -> stakkars | 01:56 | |
| Hodgestar (~Hodgestar@gluon.za.net) left irc: Remote host closed the connection | 02:22 | |
| iratsu (~alang@cpc2-oxfd21-2-0-cust354.4-3.cable.virginmedia.com) joined #pypy. | 02:23 | |
| Hodgestar (~Hodgestar@gluon.za.net) joined #pypy. | 02:24 | |
| lac_paddle_dream | 02:29 | |
| rhiebert (~rhiebert@24.104.139.205) left irc: Remote host closed the connection | 03:14 | |
| gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) joined #pypy. | 03:22 | |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) joined #pypy. | 03:23 | |
| gtaylor (~gtaylor@108-196-160-120.lightspeed.tukrga.sbcglobal.net) left irc: Quit: gtaylor | 03:30 | |
| cwillu_at_work (~cwillu@cwillu.com) left irc: Ping timeout: 240 seconds | 03:53 | |
| cwillu_at_work (~cwillu@cwillu.com) joined #pypy. | 03:53 | |
| bgola (~bgola@189.121.22.148) left irc: Ping timeout: 240 seconds | 04:05 | |
| gutworth (~benjamin@64-131-13-109.usfamily.net) left irc: Ping timeout: 240 seconds | 04:09 | |
| CIA-55 | 03justinpeel 07roundup * 10#786/Module re, \d with re.U option does not work the same way as with CPython: | 04:14 |
| CIA-55 | It worked for me as linguist said in CPython 2.6 and as dripton said in CPython | 04:14 |
| CIA-55 | 2.7. A version of CPython 3.3 that I have on my computer works j ... * 14https://bugs.pypy.org/issue786 | 04:14 |
| bgola (~bgola@189.121.22.148) joined #pypy. | 04:20 | |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 258 seconds | 04:32 | |
| mkhan (~mkhan@122.172.253.214) left irc: Ping timeout: 260 seconds | 04:41 | |
| mkhan (~mkhan@122.167.122.135) joined #pypy. | 04:46 | |
| jonanin (~jonanin@71-90-76-10.dhcp.ftbg.wi.charter.com) joined #pypy. | 04:50 | |
| donri (~donri@unaffiliated/dagodenhall) left irc: Remote host closed the connection | 04:57 | |
| Nick change: Gulaway -> Gulopine | 04:59 | |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) joined #pypy. | 05:03 | |
| etrepum (~bob@75-101-96-144.dsl.static.sonic.net) joined #pypy. | 05:11 | |
| kenaan | 12alex_gaynor default 11add21ab2b28b 15/pypy/module/: Remove a bunch of dead imports. | 05:15 |
| etrepum (~bob@75-101-96-144.dsl.static.sonic.net) left irc: Quit: etrepum | 05:22 | |
| etrepum (~bob@75-101-96-144.dsl.static.sonic.net) joined #pypy. | 05:24 | |
| ivan (~ivan@unaffiliated/ivan/x-000001) left irc: Excess Flood | 05:29 | |
| ivan (~ivan@unaffiliated/ivan/x-000001) joined #pypy. | 05:30 | |
| CIA-55 | 03justinpeel 07roundup * 10#791/Simple wordcount is significantly slower and fatter than CPython: | 05:41 |
| CIA-55 | I don't know how much help this will be, but I decided to do a break down on | 05:41 |
| CIA-55 | various operations. These were done with a file containing only a m ... * 14https://bugs.pypy.org/issue791 | 05:41 |
| Nick change: Gulopine -> Gulaway | 06:04 | |
| kenaan | 12alex_gaynor default 11cfa96ceb9935 15/pypy/module/_file/interp_stream.py: Fix translation (whoops). | 06:09 |
| kenaan | 12alex_gaynor default 1139ad5205b27a 15/pypy/module/: fix more stuff. | 06:32 |
| justinpeel | so I noticed that app_io (which includes the print function) has # NOT_RPYTHON at the top | 06:46 |
| justinpeel | is it possible to just convert the print function to RPython there without doing the whole file( and keep the print function in the same file)? | 06:46 |
| mkhan | Hi all, what would you suggest for a visual or terminal based python debugger on mac ? | 06:49 |
| sanxiyn (~tinuviel@125.177.240.192) joined #pypy. | 06:51 | |
| lac (~lac@m83-178-21-44.cust.tele2.se) joined #pypy. | 06:51 | |
| Ademan | mkhan: pdb++ also ask in #python not here | 06:58 |
| donri (~donri@unaffiliated/dagodenhall) joined #pypy. | 07:00 | |
| Eruquen (Eruquen@server3.raumopol.de) left irc: Ping timeout: 258 seconds | 07:04 | |
| Eruquen (Eruquen@server3.raumopol.de) joined #pypy. | 07:04 | |
| kans (~kans@c-98-236-30-13.hsd1.pa.comcast.net) left irc: Read error: Connection reset by peer | 07:06 | |
| kans (~kans@c-98-236-30-13.hsd1.pa.comcast.net) joined #pypy. | 07:07 | |
| kans (~kans@c-98-236-30-13.hsd1.pa.comcast.net) left irc: Quit: Leaving. | 07:23 | |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 240 seconds | 07:23 | |
| fijal (~fijal@197.107.80.96) joined #pypy. | 07:29 | |
| Alex_Gaynor | hi fijal | 07:33 |
| fijal | hi alex | 07:33 |
| Alex_Gaynor | I got the deadlock down to a reproducable test case | 07:34 |
| Alex_Gaynor | it only happens with the JIT though | 07:34 |
| fijal | uh | 07:35 |
| fijal | not good | 07:35 |
| fijal | did you look at traces? | 07:35 |
| fijal | well, jit can change race conditions | 07:35 |
| Alex_Gaynor | I tried, it's a bit hard to really look at them without jitviewer, but nothing looked totally crazy | 07:35 |
| Alex_Gaynor | yes, race conditions is my current theory | 07:36 |
| Alex_Gaynor | unfortunately I can't find them, it wouldn't surprise me that threading.py has race conditions that only apepar because pypy is faster | 07:36 |
| fijal | why without a jitviewer? | 07:36 |
| fijal | because it crashes? | 07:36 |
| fijal | if you give me a download link to the log I'll fix it just now | 07:37 |
| Alex_Gaynor | well because it deadlocks, and you have to kill it, and you get an incomplete log | 07:37 |
| Alex_Gaynor | fijal: script is in the issue tracker, you can run it yourself, takes like 30 seconds | 07:37 |
| fijal | ah | 07:37 |
| fijal | well, I wanted to fix the jitviewer issue not the deadlock | 07:38 |
| Alex_Gaynor | oh, the one with bridges you mean? | 07:38 |
| fijal | you can hop to gdb and call flush | 07:38 |
| Alex_Gaynor | yes please :) | 07:38 |
| fijal | so I need a log ;-) | 07:38 |
| fijal | I fixed bridges | 07:38 |
| Alex_Gaynor | one sec, I'll get you the link | 07:38 |
| Alex_Gaynor | http://dl.dropbox.com/u/1015145/django-tests.pypylog.gz | 07:38 |
| fijal | my indicator is lying | 07:39 |
| Alex_Gaynor | fijal: btw jitviewer tests on pypy are broken, I think you forgot to hg add a file | 07:39 |
| fijal | it says "GSM" and I get 300k transfer | 07:39 |
| Alex_Gaynor | 300KB/s? | 07:40 |
| fijal | Kb | 07:40 |
| Alex_Gaynor | :( | 07:40 |
| fijal | which one is bigger B or b? | 07:40 |
| Da_Blitz | B | 07:40 |
| fijal | the bigger one | 07:40 |
| Alex_Gaynor | KB=>kilobytes, Kb=>kilobits | 07:40 |
| fijal | Alex_Gaynor: you removed important imports | 07:40 |
| fijal | so KB | 07:40 |
| Alex_Gaynor | fijal: I fxied them though | 07:41 |
| fijal | bad alex no cookie | 07:41 |
| Alex_Gaynor | :( | 07:41 |
| antocuni (~antocuni@host150-71-dynamic.58-82-r.retail.telecomitalia.it) joined #pypy. | 07:41 | |
| fijal | yes I did forgot | 07:41 |
| Alex_Gaynor | forgot what? | 07:41 |
| fijal | I did forget | 07:43 |
| fijal | to add a file | 07:43 |
| Alex_Gaynor | oh, please add it I guess | 07:43 |
| fijal | I found a problem with our numpy approach | 07:48 |
| fijal | imagine this: | 07:48 |
| fijal | a = b + c + d + e | 07:48 |
| fijal | print (a + 3)[0] | 07:48 |
| fijal | forces a + 3 | 07:48 |
| fijal | print (a + 4)[0] | 07:48 |
| fijal | computes a again | 07:48 |
| Alex_Gaynor | <something with weakrefs and saving partial computations which are alive> | 07:49 |
| Alex_Gaynor | or just have getitem not force the entire array | 07:49 |
| antocuni | bah | 07:49 |
| Alex_Gaynor | that fixes these ones anyways | 07:49 |
| antocuni | benchmarks failed :-/ | 07:49 |
| antocuni | http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/782/steps/shell_7/logs/stdio | 07:49 |
| fijal | antocuni: yeah I've seen that | 07:50 |
| Alex_Gaynor | what does that error even mean | 07:50 |
| fijal | antocuni: you said you'll check if they're the same btw :) | 07:50 |
| antocuni | fijal: I checked the history | 07:50 |
| fijal | aha! | 07:50 |
| antocuni | and the only difference was "importing twisted" | 07:50 |
| antocuni | (instead of having it as svn:external) | 07:51 |
| fijal | Alex_Gaynor: weakrefs are a bad idea, because we don't even compute a) | 07:51 |
| fijal | or we don't store the result | 07:51 |
| fijal | we compute a + 3 and then a + 4 | 07:51 |
| Alex_Gaynor | indeed | 07:51 |
| fijal | I suppose this is not really that important issue | 07:51 |
| fijal | we might need to have a way to officially force though | 07:51 |
| Alex_Gaynor | have getitem force only that index | 07:51 |
| fijal | well, I meant any-forcing-operation | 07:52 |
| fijal | not getitem in particular | 07:52 |
| fijal | I think we should keep getitem forcing and think later :) | 07:52 |
| fijal | because you want to cache the results (maybe) | 07:52 |
| Alex_Gaynor | good point | 07:52 |
| fijal | btw | 07:53 |
| fijal | did you know that pdb.pm() doesn't work on pypy? | 07:53 |
| Alex_Gaynor | no, I did not | 07:53 |
| Alex_Gaynor | did you know that until yesterday we did not free OS resources for locks? | 07:53 |
| fijal | I've seen the commit | 07:54 |
| Alex_Gaynor | unfortunately it's unrelated to my deadlock | 07:55 |
| fijal | antocuni: I've got weird offsets | 07:55 |
| fijal | in assembler | 07:55 |
| hakanardo | fijal: maybe it would be possible to have several forcing/caching strategies that would be choosen from app-level? | 07:56 |
| Alex_Gaynor | hakanardo: well, it ends up a mess because users have to change their code to run it on pypy | 07:56 |
| hakanardo | different applications might need different strategies | 07:56 |
| fijal | hakanardo: yes | 07:56 |
| fijal | Alex_Gaynor: scientists are "I don't give a fuck I want to run it fast" | 07:57 |
| hakanardo | so a defualt one that works in most cases | 07:57 |
| fijal | if they're willing to inline fortran there, common... | 07:57 |
| hakanardo | and more fancy once that can speed up special cases | 07:57 |
| Kaskuka (~John@osbk-4d08b9ef.pool.mediaWays.net) joined #pypy. | 07:59 | |
| sanxiyn | I think there should be explicit force... | 08:01 |
| ronny | reality check - pypy is using the same pyc file format as cpython of the intended version | 08:01 |
| ronny | (someone claimed in a bug to distribute it doesnt) | 08:01 |
| cwillu_at_work (~cwillu@cwillu.com) left irc: Ping timeout: 255 seconds | 08:01 | |
| Shanita (~John@osbk-4d08b9ef.pool.mediaWays.net) left irc: Ping timeout: 255 seconds | 08:01 | |
| cwillu_at_work (~cwillu@cwillu.com) joined #pypy. | 08:01 | |
| fijal_ (~fijal@197.171.46.198) joined #pypy. | 08:02 | |
| fijal_ | Alex_Gaynor: UGH | 08:02 |
| Alex_Gaynor | ? | 08:03 |
| fijal_ | can I do a partial backout? | 08:03 |
| fijal_ | I commited too many files | 08:04 |
| fijal (~fijal@197.107.80.96) left irc: Ping timeout: 258 seconds | 08:04 | |
| fijal_ | Alex_Gaynor: I fixed one issue, but the other issue is that loop addresses are not unique | 08:04 |
| antocuni | fijal_: what happens with pdb.pm on pypy? | 08:04 |
| Alex_Gaynor | you mean if a loop get's deallocated? | 08:04 |
| fijal_ | antocuni: we don't have sys.last_traceback | 08:04 |
| fijal_ | Alex_Gaynor: yes | 08:04 |
| Alex_Gaynor | nice | 08:04 |
| fijal_ | I have to fix this | 08:04 |
| fijal_ | anyway, HG HELP | 08:04 |
| Alex_Gaynor | use a tuple of loop_id, loop_addr everywhere I gues | 08:05 |
| Nick change: fijal_ -> fijal | 08:05 | |
| fijal | Alex_Gaynor: yeah, but I also have an assembler dump | 08:05 |
| fijal | which doesn't have a loop_id | 08:05 |
| Alex_Gaynor | make jit dump loop_id with it | 08:05 |
| fijal | anyway, I know how to fix it but I want to push changes, how do I do this? | 08:05 |
| fijal | EWRONGLEVEL | 08:05 |
| Alex_Gaynor | ETOOMANYLEVELS | 08:05 |
| fijal | they come in order so that's fine | 08:06 |
| Alex_Gaynor | ok | 08:06 |
| fijal | can I merge the reverse of a checkin? | 08:06 |
| antocuni | fijal: for me pdb.pm seems to work | 08:06 |
| Alex_Gaynor | hg backout? | 08:06 |
| fijal | antocuni: http://paste.pocoo.org/show/439965/ | 08:07 |
| fijal | Alex_Gaynor: that backs out the whole thing | 08:07 |
| fijal | I want to review the change by hand | 08:07 |
| fijal | or do specific files | 08:07 |
| Alex_Gaynor | backout the whole thing with --no-commit and then reapply by hand | 08:07 |
| fijal | so I need a diff to look at | 08:07 |
| fijal | ok | 08:07 |
| antocuni | fijal: http://paste.pocoo.org/show/439966/ | 08:08 |
| antocuni | we have last_traceback | 08:08 |
| fijal | nice | 08:08 |
| antocuni | but it seems it's not set inside except: | 08:08 |
| sanxiyn | hg record? | 08:08 |
| fijal | hg backout: option --no-commit not recognized | 08:08 |
| antocuni | fijal: btw, you should use pdb.xpm() (in pdb++), which always work | 08:09 |
| antocuni | works | 08:10 |
| fijal | ok | 08:10 |
| fijal | but we should not break stuff anyway :) | 08:10 |
| mkhan (~mkhan@122.167.122.135) left irc: Disconnected by services | 08:10 | |
| antocuni | """they are set when an exception is not handled and the interpreter prints an error message and a stack traceback""" | 08:10 |
| antocuni | quoting the docs about last_{type,value,traceback} | 08:10 |
| fijal | well ok, fine | 08:10 |
| fijal | but apparently they're not | 08:10 |
| antocuni | I don't know why cpython does something different | 08:11 |
| fijal | should we fix pdb.py that we ship? | 08:11 |
| antocuni | no | 08:11 |
| fijal | no? | 08:11 |
| antocuni | if we want to fix this usecase, we should have last_traceback, IMHO | 08:11 |
| fijal | hm | 08:11 |
| fijal | it's not like it works on cpython | 08:11 |
| antocuni | or fix pdb.py upstream | 08:11 |
| antocuni | (or just ship pdb++ with pypy :-)) | 08:12 |
| fijal | yes, but they'll say "ok, but 3.x only" I'm not that interested | 08:12 |
| fijal | I would rather fix pdb that we ship and provide a bug report | 08:12 |
| antocuni | well, my point is that it's not a "fix" | 08:13 |
| fijal | antocuni: http://paste.pocoo.org/show/439967/ | 08:13 |
| antocuni | it's a new feature | 08:13 |
| fijal | fix as in make pdb.pm use sys.exc_info()[2] | 08:13 |
| antocuni | fijal: yes, I knew that there was a reason why I wrote xpm() :-) | 08:13 |
| fijal | antocuni: btw, flymake that I sent you is awesome | 08:16 |
| fijal | try it | 08:16 |
| antocuni | yes, I suppose I should give it a try at some point | 08:16 |
| fijal | like now :) | 08:16 |
| antocuni | no, now I want to fix benchmarks | 08:16 |
| fijal | if you don't do it now, you won't do it at all until a sprint where we meet | 08:16 |
| Action: fijal off for a bit | 08:16 | |
| antocuni | I won't be online the whole day | 08:16 |
| fijal | Alex_Gaynor: I'll fix the jitviewer later today | 08:17 |
| fijal (~fijal@197.171.46.198) left irc: Quit: Leaving | 08:17 | |
| Ademan | is there a way to create an annotation for a class at import time? | 08:20 |
| antocuni | bah | 08:20 |
| antocuni | the problem with benchmarks is that the bitbucket's svn interface does not support symlinks | 08:21 |
| Ademan | actually that won't work anyways... ugh this is annoying | 08:22 |
| ronny | oO | 08:23 |
| sanxiyn (tinuviel@125.177.240.192) left #pypy (" tÌ Èä."). | 08:25 | |
| rokujyouhitoma (~rokujyouh@conference/rubykaigi/x-snvzqamurvkzjdwj) joined #pypy. | 08:27 | |
| kenaan | 12antocuni benchmarks 118c862863e7fb 15/own/util.py: kill the symlink, and use a python hack to share the content of util.py in both directories | 08:28 |
| antocuni | I'm going to kick a benchmark run on tannit, please don't run translations on it | 08:28 |
| antocuni | (or stop the build if you really need to do so)\ | 08:29 |
| apollo13 (~apollo13@unaffiliated/apollo13) left irc: Read error: Operation timed out | 08:31 | |
| apollo13 (~apollo13@unaffiliated/apollo13) joined #pypy. | 08:35 | |
| tlynn_ (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy. | 08:37 | |
| ousado (~ousado@31.18.108.77) left irc: Ping timeout: 264 seconds | 08:57 | |
| rokujyouhitoma (~rokujyouh@conference/rubykaigi/x-snvzqamurvkzjdwj) left irc: Quit: rokujyouhitoma | 09:12 | |
| rokujyouhitoma (~rokujyouh@conference/rubykaigi/x-mtgxsscshhvpcnts) joined #pypy. | 09:12 | |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) left irc: Excess Flood | 09:15 | |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) joined #pypy. | 09:16 | |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) joined #pypy. | 09:16 | |
| rokujyouhitoma_ (~rokujyouh@conference/rubykaigi/x-omrptfbqqqzlibzh) joined #pypy. | 09:24 | |
| rokujyouhitoma (~rokujyouh@conference/rubykaigi/x-mtgxsscshhvpcnts) left irc: Read error: Connection reset by peer | 09:24 | |
| Nick change: rokujyouhitoma_ -> rokujyouhitoma | 09:24 | |
| cfbolz (~cfbolz@dslb-188-108-231-232.pools.arcor-ip.net) joined #pypy. | 09:26 | |
| cfbolz | hello | 09:26 |
| rokujyouhitoma_ (~rokujyouh@conference/rubykaigi/x-wvklhhaezipshdjv) joined #pypy. | 09:27 | |
| rokujyouhitoma (~rokujyouh@conference/rubykaigi/x-omrptfbqqqzlibzh) left irc: Read error: Connection reset by peer | 09:27 | |
| Nick change: rokujyouhitoma_ -> rokujyouhitoma | 09:27 | |
| xorAxAx | hi cfbolz, still sprinting? :) | 09:29 |
| cfbolz | no | 09:31 |
| cfbolz | finished yesterday | 09:31 |
| rokujyouhitoma (~rokujyouh@conference/rubykaigi/x-wvklhhaezipshdjv) left irc: Quit: rokujyouhitoma | 09:33 | |
| kenaan | 12cfbolz faster-nested-scopes 11c6b24da7b2b9 15/pypy/objspace/flow/: fix flow space. needs a slightly annoying hack | 09:40 |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) left irc: Quit: Off | 09:43 | |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) joined #pypy. | 09:43 | |
| kenaan | 12cfbolz faster-nested-scopes 115994465ca0d0 15/pypy/module/cpyext/frameobject.py: fix cpyext | 09:50 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 09:55 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Read error: Connection reset by peer | 10:02 | |
| lac (~lac@m83-178-21-44.cust.tele2.se) left irc: Read error: Connection reset by peer | 10:03 | |
| Trundle (~andy@dslb-178-002-050-160.pools.arcor-ip.net) joined #pypy. | 10:11 | |
| Trundle (~andy@dslb-178-002-050-160.pools.arcor-ip.net) left irc: Changing host | 10:11 | |
| Trundle (~andy@python/site-packages/trundle) joined #pypy. | 10:11 | |
| Sho_ (~EHS1@kde/hein) left irc: Quit: Stop leaking memory like it's going out of fashion. | 10:15 | |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 10:20 | |
| cfbolz | antocuni: ping? | 10:21 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Read error: Connection reset by peer | 10:21 | |
| antocuni | cfbolz: semi-pong | 10:37 |
| antocuni | (going to lunch very soon) | 10:37 |
| antocuni | like, right now :-) | 10:38 |
| Action: antocuni afk | 10:38 | |
| donri (~donri@unaffiliated/dagodenhall) left irc: Remote host closed the connection | 10:55 | |
| pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) left irc: Quit: Leaving | 11:07 | |
| kenaan | 12cfbolz default 110eee4b30304a 15/pypy/jit/metainterp/test/test_ajit.py: typo | 11:08 |
| mitsuhiko | Alex_Gaynor: found it? | 11:14 |
| antocuni | cfbolz: pull pong, now :-) | 11:15 |
| cfbolz | antocuni: from time to time I have the following bug with pdb++ | 11:15 |
| cfbolz | antocuni: I enter the name of a local var and get a name error | 11:16 |
| cfbolz | if I use interact, I can access the local var | 11:16 |
| antocuni | cfbolz: I think if you have an old version of pdb++ | 11:16 |
| cfbolz | ah | 11:16 |
| antocuni | which version is it? | 11:16 |
| cfbolz | 0.7, I think | 11:17 |
| antocuni | uhm, then it should be fixed | 11:17 |
| cfbolz | hm | 11:17 |
| antocuni | the previous error like the one you describe was only with python 2.7 and if you do "up" | 11:17 |
| cfbolz | ah, I see the problem | 11:17 |
| cfbolz | yes | 11:17 |
| cfbolz | pypy has an older pdb++ installed :-) | 11:17 |
| antocuni | ah :-) | 11:18 |
| antocuni | good | 11:18 |
| cfbolz | that explains why I mostly see it when translating | 11:18 |
| antocuni | pff, benchmarks died again :-( | 11:18 |
| antocuni | http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/784/steps/shell_7/logs/stdio | 11:18 |
| cfbolz | eh | 11:19 |
| cfbolz | time to kill waf? :-) | 11:19 |
| antocuni | maybe | 11:19 |
| antocuni | the concrete problem is probably that bitbucket's svn does not support permissions | 11:19 |
| antocuni | so some file is checked out as non-executable | 11:19 |
| cfbolz | well | 11:19 |
| cfbolz | btw, 72d3b692033c made translate of targetrpystonedalone 10% faster | 11:20 |
| cfbolz | and reduces the size of non-optimized traces by 45% | 11:20 |
| antocuni | nice | 11:21 |
| antocuni | I suppose that the performance improvement comes from the reduced gc pressure? | 11:21 |
| cfbolz | and faster optimization | 11:22 |
| antocuni | right | 11:22 |
| antocuni | uhm, I wonder if we should adjust the trace_limit now | 11:22 |
| antocuni | because if it's actually 45% less, we are going to trace much more than before | 11:22 |
| cfbolz | good? :) | 11:22 |
| antocuni | maybe, I don't know | 11:23 |
| cfbolz | antocuni: I will try another one or two minimally invasive optimizations during tracing | 11:23 |
| antocuni | we should have a buildbot job that regularly tries to play with the various thresholds, to find the optimal ones :-) | 11:23 |
| cfbolz | I think trying that pays off much better than trying the full thing | 11:23 |
| antocuni | cfbolz: instead of doing the full optimization twice? | 11:24 |
| cfbolz | yes | 11:24 |
| antocuni | maybe | 11:24 |
| antocuni | but e.g. you don't get rid of virtuals | 11:24 |
| antocuni | cfbolz: anyway, are you fine if I just kill waf? | 11:24 |
| cfbolz | yes | 11:25 |
| Action: antocuni kills | 11:25 | |
| cfbolz | antocuni: yes, but getting rid of virtuals is a several weeks development job | 11:26 |
| antocuni | because of the mess in resume.py? | 11:26 |
| cfbolz | it's not a mess right now :-) | 11:26 |
| cfbolz | it's just very hard to change | 11:26 |
| antocuni | :-) | 11:26 |
| antocuni | ok | 11:26 |
| antocuni | actually, I didn't look at it | 11:26 |
| antocuni | but if you think it's so hard, I suppose you are right | 11:27 |
| cfbolz | another thing to consider is to that if you look at nonopt traces, the bulk of operations doesn't come from virtuals | 11:28 |
| cfbolz | or, more precisely | 11:28 |
| cfbolz | a huge amount of operations come from *frames* of inlined functions | 11:29 |
| kenaan | 12antocuni benchmarks 11b3ba07245ff9 15/: kill the waf benchmark; it's broken right now, and we decided that it's not interesting | 11:29 |
| antocuni | cfbolz: exactly, but frames of inlined functions becomes virtuals | 11:29 |
| cfbolz | yes, I know | 11:29 |
| cfbolz | but it's not generalized virtuals at all | 11:29 |
| cfbolz | just virtual frames | 11:30 |
| antocuni | so, you are proposing a special trace-time pass which just kills virtual frames? | 11:30 |
| cfbolz | that would be one option, yes | 11:30 |
| Action: antocuni kicks a benchmark run on tannit | 11:30 | |
| antocuni | do you have something precise in mind or it's just a vague idea? | 11:31 |
| cfbolz | antocuni: right now I am trying a different hack | 11:31 |
| cfbolz | antocuni: basically the heap optimization is easier to do during tracing than virtuals | 11:32 |
| cfbolz | because it has no interaction with resume data | 11:32 |
| cfbolz | and it has some of the effects of virtuals | 11:32 |
| cfbolz | (it gets rid of getfields, but not of setfields and news) | 11:32 |
| antocuni | ah, I see | 11:32 |
| cfbolz | so I am trying a very very simple heap cache | 11:32 |
| antocuni | but the drawback of this approach is that we are essentially writing some optimizations twice | 11:33 |
| antocuni | what about using the actual code from optimizeopt/heap.py ? | 11:33 |
| cfbolz | we already do that a lot | 11:34 |
| cfbolz | e.g. pyjitpl does constant folding | 11:34 |
| antocuni | that's not a good excuse to do more :-) | 11:34 |
| ronny | hmm, is there any reason why file considers pyc files by pypy data while pyc files by cpython as compiled python bytecode? | 11:34 |
| antocuni | ronny: yes, because we use a different magic number | 11:35 |
| antocuni | and probably file does not know about it | 11:35 |
| cfbolz | antocuni: yes, I know. but the two optimizations serve very different purposes and are implemented with different trade-offs in mind | 11:36 |
| antocuni | true enough | 11:36 |
| cfbolz | anyway, I'll simply try | 11:36 |
| antocuni | well, I suppose that if the cache is "very very simple", it might be worth | 11:36 |
| cfbolz | that's the plan | 11:36 |
| antocuni | ok | 11:37 |
| ronny | antocuni: is there any other difference? | 11:37 |
| antocuni | ronny: I don't think so | 11:37 |
| cfbolz | ronny: we have slightly different bytecode sets | 11:37 |
| antocuni | yes, but "file" should not care about them | 11:38 |
| cfbolz | right | 11:38 |
| stakkars_ (~tismer@91-66-101-126-dynip.superkabel.de) joined #pypy. | 11:45 | |
| fox__ (~fox@host42-106-dynamic.14-87-r.retail.telecomitalia.it) joined #pypy. | 11:47 | |
| tlynn_ (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Ping timeout: 255 seconds | 11:51 | |
| ousado (~ousado@31.18.130.130) joined #pypy. | 11:55 | |
| lucian (~lucian@78-86-217-168.zone2.bethere.co.uk) joined #pypy. | 11:55 | |
| LoganLK (~LoganLK@202-180-111-127.callplus.net.nz) left irc: Quit: LoganLK | 11:59 | |
| tlynn_ (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy. | 12:02 | |
| santagada (~santagada@c925bcc7.virtua.com.br) joined #pypy. | 12:07 | |
| mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) left irc: Remote host closed the connection | 12:10 | |
| candrea (~andrea@host251-99-dynamic.24-79-r.retail.telecomitalia.it) joined #pypy. | 12:11 | |
| candrea (~andrea@host251-99-dynamic.24-79-r.retail.telecomitalia.it) left irc: Changing host | 12:11 | |
| candrea (~andrea@ubuntu/member/beeseek.developer.andrea-bs) joined #pypy. | 12:11 | |
| mcdonc (~mcdonc@ip72-209-213-54.dc.dc.cox.net) joined #pypy. | 12:13 | |
| zk (~is@adsl-66-140-73-70.dsl.kscymo.swbell.net) got netsplit. | 12:17 | |
| bfirsh (~bfirsh@mini.firshman.co.uk) got netsplit. | 12:17 | |
| Rhyolite (~dje@pool-108-6-25-71.nycmny.fios.verizon.net) got netsplit. | 12:17 | |
| crakdmirror (~crakdmirr@174.127.114.26) got netsplit. | 12:17 | |
| hongminhee (~dahlia@175.125.21.72) got netsplit. | 12:17 | |
| mkhan (~mkhan@122.167.122.135) joined #pypy. | 12:18 | |
| zk (~is@adsl-66-140-73-70.dsl.kscymo.swbell.net) returned to #pypy. | 12:19 | |
| crakdmirror (~crakdmirr@174.127.114.26) returned to #pypy. | 12:19 | |
| bfirsh (~bfirsh@mini.firshman.co.uk) returned to #pypy. | 12:19 | |
| Rhyolite (~dje@pool-108-6-25-71.nycmny.fios.verizon.net) returned to #pypy. | 12:19 | |
| hongminhee (~dahlia@175.125.21.72) returned to #pypy. | 12:19 | |
| derdon (~derdon@p579CC747.dip.t-dialin.net) joined #pypy. | 12:20 | |
| name (~name@chello080108038152.1.11.vie.surfer.at) joined #pypy. | 12:27 | |
| name (~name@chello080108038152.1.11.vie.surfer.at) left irc: Changing host | 12:27 | |
| name (~name@sburn/devel/name) joined #pypy. | 12:27 | |
| name (~name@sburn/devel/name) left irc: Remote host closed the connection | 12:28 | |
| kenaan | 12cfbolz heap-caching-during-tracing 11be79442107df 15/pypy/jit/metainterp/: add a minimal heap cache to be used when tracing | 12:33 |
| antocuni | good, the benchmark run worked well this time | 12:46 |
| antocuni | but now we have very high variations: http://speed.pypy.org/changes/ | 12:46 |
| antocuni | like, ~20% speedup on raytrace-simple and spectral-norm | 12:47 |
| antocuni | and 64% slowdown on go | 12:47 |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) joined #pypy. | 12:49 | |
| neppord | hi pople | 12:50 |
| derdon (~derdon@p579CC747.dip.t-dialin.net) left irc: Remote host closed the connection | 12:51 | |
| stakkars_ | antocuni: is this related to the heap-caching checkin? | 12:53 |
| antocuni | stakkars_: no, it's just that we did not run benchmarks for few days | 12:53 |
| stakkars_ | ah, coincidence. | 12:54 |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) left irc: Quit: This computer has gone to sleep | 12:55 | |
| stakkars_ | antocuni: are you working from Italy, btw.? | 12:55 |
| antocuni | yes | 12:55 |
| stakkars_ | ok. For some reason I thought you were in Sweden. | 12:56 |
| Action: antocuni off | 12:57 | |
| antocuni | bye! | 12:57 |
| stakkars_ | this came from earlier info | 12:57 |
| stakkars_ | bye | 12:57 |
| stakkars_ (tismer@91-66-101-126-dynip.superkabel.de) left #pypy. | 12:59 | |
| kenaan | 12hakanardo jit-short_from_state 11a8eedb9a9b64 15/pypy/jit/metainterp/optimizeopt/: Some extra debug prints. Make optimize_STRLEN keep the original result box if the strlen op is emi... | 12:59 |
| antocuni (~antocuni@host150-71-dynamic.58-82-r.retail.telecomitalia.it) left irc: Ping timeout: 276 seconds | 13:03 | |
| santagada_ (~santagada@c925bcc7.virtua.com.br) joined #pypy. | 13:10 | |
| santagada (~santagada@c925bcc7.virtua.com.br) left irc: Read error: Connection reset by peer | 13:10 | |
| stakkars_ (~tismer@91-66-101-126-dynip.superkabel.de) joined #pypy. | 13:11 | |
| CIA-55 | 03timo 07roundup * 10#706/`import pybrain` fails in 1.5 on OSX: | 13:14 |
| CIA-55 | [chatting] you need to install pybrain using pypy. pypy will not use the site-packages | 13:14 |
| CIA-55 | folder from python (especially not from python2.6, as pyp ... * 14https://bugs.pypy.org/issue706 | 13:14 |
| DasIch_ (~DasIch@p4FFDCB61.dip.t-dialin.net) joined #pypy. | 13:23 | |
| kkris (~kris@80-123-38-183.adsl.highway.telekom.at) joined #pypy. | 13:26 | |
| DasIch (~DasIch@p4FFDF1D0.dip.t-dialin.net) left irc: Ping timeout: 276 seconds | 13:27 | |
| CIA-55 | 03timo 07roundup * 10#706/`import pybrain` fails in 1.5 on OSX: | 13:40 |
| CIA-55 | my apologies, i just tried it myself and you are going to run into another | 13:40 |
| CIA-55 | problem: pybrain needs scipy, which pypy doesn't support yet, due to ... * 14https://bugs.pypy.org/issue706 | 13:40 |
| fijal_ (~fijal@197.104.46.98) joined #pypy. | 13:42 | |
| Nick change: fijal_ -> fijal | 13:42 | |
| fijal | hi | 13:43 |
| neppord | hi | 13:44 |
| tlynn_ (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Ping timeout: 255 seconds | 13:44 | |
| fijal | ok, something went horribly wrong with benchmarks | 13:48 |
| rokujyouhitoma (~rokujyouh@240f:11:fa84:1:21f:f3ff:fed7:c020) joined #pypy. | 14:00 | |
| arigato (~arigo@82.113.98.162) joined #pypy. | 14:04 | |
| rokujyouhitoma (~rokujyouh@240f:11:fa84:1:21f:f3ff:fed7:c020) left irc: Quit: rokujyouhitoma | 14:05 | |
| kenaan | 12arigo default 11eef75a4b072e 15/pypy/translator/c/gcc/trackgcroot.py: Write the progbits section magic marker to make the stack non-executable under (at least) Gen... | 14:07 |
| fijal (~fijal@197.104.46.98) left irc: Ping timeout: 264 seconds | 14:07 | |
| tlynn_ (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) joined #pypy. | 14:11 | |
| kenaan | 12fijal default 11f12c4e733acd 15/pypy/: fix tests | 14:18 |
| kenaan | 12fijal default 11eaa6922449c1 15/pypy/: Backed out changeset f12c4e733acd Ops, too much | 14:18 |
| kenaan | 12fijal default 11a3c110ceadc2 15/pypy/tool/jitlogparser/: fix tests | 14:18 |
| kenaan | 12fijal default 1130cdf33c3942 15/pypy/: merge | 14:18 |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) left irc: Quit: neppord | 14:19 | |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) joined #pypy. | 14:23 | |
| fijal (~fijal@41.53.101.44) joined #pypy. | 14:24 | |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) joined #pypy. | 14:27 | |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) left irc: Ping timeout: 260 seconds | 14:27 | |
| gutworth (~benjamin@64-131-13-109.usfamily.net) joined #pypy. | 14:29 | |
| CIA-55 | 03albert 07roundup * 10#706/`import pybrain` fails in 1.5 on OSX: | 14:29 |
| CIA-55 | Yea... | 14:29 |
| CIA-55 | The SciPy/Numpy functions which are used by PyBrain are also only very basics ones | 14:29 |
| CIA-55 | (matrix multiplications, etc.), so it might be that ... * 14https://bugs.pypy.org/issue706 | 14:29 |
| kenaan | 12fijal default 11402d04acdc09 15/pypy/tool/jitlogparser/: Fix bug with regard to multiple loops at the same assembler piece. Make parsing assembler lazy (a bit ugly) | 14:30 |
| tlynn_ | how's progress towards stackless-pypy-jit, does it depend on tealets and/or are tealets dead? | 14:31 |
| fijal | it's almost done | 14:32 |
| timonator | the last thing i've heard is tealets is the future and very much alive | 14:32 |
| fijal | you can get it if you get a correct branch | 14:32 |
| timonator | hey, fijal, scipy fails installing very early, because numpy.distutils or some such is missing | 14:32 |
| fijal | timonator: you kind of don't want to install scipy | 14:32 |
| fijal | for a varierty of reasons | 14:33 |
| tlynn_ | sweet. alas, my machine would probably take over a day to build it. I suppose I could try though. | 14:33 |
| fijal | how much ram do you have? | 14:34 |
| timonator | fijal: why not? pybrain needs its | 14:34 |
| tlynn_ | a mere 2GB | 14:34 |
| fijal | tlynn_: you should be good on 32bit | 14:35 |
| fijal | timonator: needs what from scipy? | 14:35 |
| tlynn_ | Windows, Linux or both? | 14:35 |
| CIA-55 | 03fijal 07roundup * 10#706/`import pybrain` fails in 1.5 on OSX: On Sat, Jul 16, 2011 at 3:45 PM, Albert Zeyer <tracker@bugs.pypy.org> wrote: > > Albert Zeyer <albzey@gmail.com> added the comment: > > Yea... > > ... * 14https://bugs.pypy.org/issue706 | 14:36 |
| fijal | tlynn_: both | 14:36 |
| timonator | it imports reshape, dot, outer early on | 14:37 |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) joined #pypy. | 14:37 | |
| fijal | yes from numpy | 14:37 |
| fijal | but what from scipy? | 14:37 |
| timonator | no, the line is from scipy import reshape, dot, outer | 14:37 |
| fijal | ok, can you look what they do? | 14:38 |
| timonator | hm, not really that interested in pybrain actually :o | 14:38 |
| fijal | well, really | 14:39 |
| fijal | "using just a few simple functions" | 14:39 |
| fijal | is a claim | 14:39 |
| fijal | and requires evidence | 14:39 |
| fijal | Alex_Gaynor: I fixed the jitviewer | 14:41 |
| fijal | I fear at some point I'll have to get back and refactor it a bit | 14:41 |
| kenaan | 12fijal jitviewer 116159d4330dce 15/bin/jitviewer.py: use the lazy loading of loops | 14:41 |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) left irc: Ping timeout: 252 seconds | 14:41 | |
| fijal | hm | 14:44 |
| fijal | can I get a history of a single branch? | 14:45 |
| rokujyouhitoma (~rokujyouh@p8443d1.tokynt01.ap.so-net.ne.jp) joined #pypy. | 14:45 | |
| tlynn_ | fijal, you can on bitbucket so I presume there's an equivalent cmdline ui | 14:47 |
| tlynn_ | e.g. https://bitbucket.org/pypy/pypy/changesets/tip/branch(tealet) | 14:48 |
| tlynn_ | which I was on since I'm looking for which branch to build for stackless-jit? | 14:49 |
| fijal | yes | 14:49 |
| fijal | --stackless -Ojit | 14:49 |
| fijal | maybe --gcstackrootfinder=shadowstack | 14:49 |
| fijal | er no | 14:49 |
| fijal | no --stackless :) | 14:49 |
| tlynn_ | on branch tealet? | 14:49 |
| fijal | --tealet | 14:49 |
| fijal | yes | 14:49 |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) joined #pypy. | 14:49 | |
| tlynn_ | cool, ta | 14:49 |
| tlynn_ | so just to be clear, "--stackless --tealet -Ojit"? | 14:50 |
| leemp (47d200e8@gateway/web/freenode/ip.71.210.0.232) joined #pypy. | 14:50 | |
| fijal | no, no --stackless | 14:50 |
| fijal | --tealet -Ojit | 14:50 |
| tlynn_ | k, ta | 14:50 |
| timonator | fijal: the "using just a few simple functions" claim wasn't made by me, though | 14:51 |
| timonator | so i suppose i'm off the hook | 14:51 |
| fijal | ok | 14:52 |
| leemp | Quick question. How is pypy's versioning decided? Eg.. right now it's based off of 2.x right? When will that change (from say 2.5 to 2.7, or 2.7 to 3.2)? And as a developer, how will i know which python version my instance of pypy is based off of? | 14:52 |
| fijal | sys.version tells you 2.7 | 14:53 |
| fijal | or sys.version_info | 14:53 |
| fijal | this is a version of Python that it supports | 14:53 |
| fijal | do you also need a version of pypy? | 14:53 |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) left irc: Ping timeout: 252 seconds | 14:54 | |
| leemp | k. So does the pypy crew try to stay up to date with the cpython versioning? Eg, if python 2.8 is released, will they push pypy to 2.8? | 14:54 |
| Rhyolite | leemp: are you asking how to find the version through introspection or how support for a particular version of CPython is chosen? | 14:54 |
| rokujyouhitoma (~rokujyouh@p8443d1.tokynt01.ap.so-net.ne.jp) left irc: Quit: rokujyouhitoma | 14:54 | |
| fijal | leemp: 2.8 will never be released | 14:55 |
| fijal | and we're a bit undecided what to do with python 3 | 14:55 |
| leemp | For example i mean | 14:55 |
| fijal | well, I'm giving you a very precise answer :) | 14:55 |
| fijal | 2.7 is the last of 2.x series | 14:55 |
| leemp | Well 3.x is a huge undertaking, different case imo. | 14:55 |
| fijal | so we're on 2.7 and we're happy | 14:55 |
| timonator | i've been burnt pretty badly by python2's strange unicode support | 14:56 |
| Rhyolite | and other than timonator, most PyPy users are happy ;-) | 14:57 |
| timonator | i'd like to have python3 for string/bytes thingie | 14:57 |
| tlynn_ | fijal: so is there a separate method for getting the pypy version? | 14:57 |
| timonator | well, it was cpython, though, but not because of the implementation, because of the specification | 14:57 |
| fijal | tlynn_: yeah | 14:57 |
| leemp | fijal: True, but it's besides the point :P. I'm asking if pypy trys to stay up to date in that regard :P, but i guess it's a question that will only pop up if they decide to hop into the 3.x line aswell | 14:57 |
| fijal | yes exactly | 14:57 |
| fijal | we stayed pretty much up-to-date module a year or two | 14:58 |
| leemp | Righto :) | 14:58 |
| fijal | tlynn_: sys.pypy_version_info | 14:58 |
| tlynn_ | thanks | 14:58 |
| Action: gutworth can arrange to have pypy 3.x for a small sum :) | 14:58 | |
| fijal | timonator: here we go, you can pay gutworth | 14:58 |
| timonator | haha, a small sum? what's that? a thousand euros? :) | 14:59 |
| fijal | gutworth: can you arrange to have missing codecs? | 14:59 |
| leemp | Also, regarding --sandbox, how much should it be trusted? Is it pretty bleeding edge and possibly has quite a few holes? Or are security issues not currently known? | 14:59 |
| fijal | leemp: it's pretty much secure by design | 15:00 |
| arigato | it's not really well tested, but on the other hand it's completely secure, as far as I know | 15:00 |
| timonator | leemp: it wouldn't appear to me, that it would have problems | 15:00 |
| leemp | --sandbox is my favorite thing about pypy. I was *very* happy to hear you guys did that :D | 15:00 |
| fijal | leemp: I would trust it much more than anything else that's available and has been tested | 15:00 |
| leemp | Yea | 15:00 |
| fijal | because it's designed better | 15:01 |
| arigato | well the option to limit the amount of memory may not be completely secure | 15:01 |
| fijal | arigato: hi | 15:01 |
| tlynn_ | good point - since I'm about to kick off a build, should I expect "--tealet -Ojit --sandbox" to work? | 15:01 |
| leemp | I read a long while back that they (a couple guys) started trying to make a secure cpython but it died quickly. | 15:01 |
| timonator | it even does that linux system call that makes the process unable to do anything apart from accessing the file descriptors it already has open at that time | 15:01 |
| Action: fijal is having fun staring at the assembler of operations | 15:01 | |
| arigato | (I think this option does not limit e.g. the raw memory allocations, which on the other hand are probably rare in a sandboxed process) | 15:01 |
| fijal | tlynn_: --sandbox is not quite what you think it is | 15:01 |
| fijal | tlynn_: I don't know if it works with --tealet but it's less usable than a normal pypy-c | 15:02 |
| tlynn_ | fijal, I understand the multiple processes separation | 15:02 |
| timonator | fijal: is there a way to mechanically test the correctness of the sandbox translation? | 15:02 |
| tlynn_ | to some degree | 15:02 |
| fijal | timonator: it mechanically walks the graphs and replaces all of external calls | 15:02 |
| fijal | I think it is to some extent | 15:03 |
| arigato | timonator: there is no way to mechanically find security bugs :-/ | 15:03 |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) left irc: Quit: This computer has gone to sleep | 15:03 | |
| arigato | (unless the program was badly designed, maybe) | 15:03 |
| fijal | arigato: do you know how to get parents of some revision in hg | 15:03 |
| fijal | ? | 15:03 |
| timonator | right, that's what i feared | 15:04 |
| arigato | fijal: yes, with (guess) an extension | 15:04 |
| arigato | parentrevspec | 15:04 |
| timonator | oh, the sandbox transform should work on any rpython translation, won't it? | 15:04 |
| fijal | aha | 15:04 |
| arigato | fijal: then you can say e.g. "eef75a4b072e^" | 15:04 |
| arigato | which I find useful | 15:04 |
| fijal | hg parentrevspec eef75a4b072e^ | 15:05 |
| fijal | like this? | 15:05 |
| arigato | no | 15:05 |
| fijal | I'm not sure I understand what you mean | 15:05 |
| arigato | parentrevspec is the extension, but it adds syntax to revisions that you specify anywhere | 15:05 |
| arigato | e.g. hg log eef75a4b072e^ | 15:05 |
| arigato | or hg up eef75a4b072e^ | 15:05 |
| fijal | ok | 15:06 |
| fijal | thanks | 15:06 |
| arigato | hg log -r eef75a4b072e^ | 15:06 |
| fijal | so our lib-python tests timeouted threading tests... | 15:06 |
| timonator | arigato: how exactly could a client program in a pypy sandbox get out of it by causing a segfault or similar? wouldn't that just kill the client, allowing the parent to just start up another sandbox? | 15:07 |
| timonator | i don't see how it could cause code to leak from inside the sandbox | 15:07 |
| arigato | timonator: I don't either | 15:08 |
| arigato | I was complaining that the option to limit the amount of RAM the pypy sandbox uses might not be completely reliable | 15:08 |
| arigato | nothing more | 15:09 |
| timonator | oh, so you mean the amount of ram the sandbox could allocate could be much bigger than what the parent process wants? | 15:10 |
| arigato | yes, in theory, although I'm not sure I see exactly how | 15:10 |
| fijal | arigato: can I kick the benchmark run? | 15:10 |
| arigato | fijal: feel free | 15:10 |
| Action: arigato doesn't use tannit today | 15:10 | |
| fijal | ok | 15:11 |
| fijal | I'm fixing the benchmarks | 15:11 |
| fijal | last run was saved as pypy-c-jit although it was pypy-c | 15:11 |
| arigato | :-/ | 15:11 |
| tlynn_ | is pypy memory not measured on speed.pypy or am I missing it? | 15:12 |
| fijal | tlynn_: no | 15:12 |
| fijal | arigato: the goal is to stop benchmarking cpython and instead benchmark pypy-c vs pypy-c-jit | 15:13 |
| kenaan | 12bivab arm-backend-2 1105def4b63047 15/pypy/jit/backend/arm/assembler.py: fix | 15:13 |
| kenaan | 12bivab arm-backend-2 1157c5578f3895 15/pypy/: merge default | 15:13 |
| kenaan | 12bivab arm-backend-2 115690763bfdba 15/pypy/jit/backend/arm/: add a check for the size of the value stored in size | 15:13 |
| kenaan | 12bivab arm-backend-2 118ca230890525 15/pypy/: merge default | 15:13 |
| fijal | and then be able to do the same for 64bity | 15:13 |
| fijal | arigato: can you poke whoever-is-responsible to install pypy on windows machines? | 15:13 |
| arigato | fijal: and put manually once the speed of cpython? | 15:13 |
| fijal | we do that anyway | 15:13 |
| fijal | we don't upload cpython speed nightly | 15:13 |
| arigato | so what do you mean? | 15:14 |
| fijal | we run cpython | 15:14 |
| fijal | and then happily ignore the resutls | 15:14 |
| fijal | (twice a night) | 15:14 |
| arigato | ah | 15:14 |
| arigato | fijal: ENOTENOUGHINFO: what's the issue with pypy on windows machines? | 15:14 |
| fijal | I changed translate benchmark job to use pypy instead of cpython for translation | 15:15 |
| fijal | but now we require all buildbots to have pypy installed | 15:15 |
| arigato | ok | 15:16 |
| kenaan | 12fijal benchmarks 111beaa1484ca2 15/own/chaos.py: remove name errors | 15:16 |
| kenaan | 12fijal benchmarks 11a8836ad0cec7 15/: improve uploading so we can benchmark pypy against pypy no jit | 15:16 |
| arigato | on openend's windows server that's me | 15:16 |
| fijal | I think this is the only one we run | 15:17 |
| fijal | http://buildbot.pypy.org/builders/own-win-x86-32 | 15:18 |
| fijal | this has slave lost | 15:18 |
| fijal | http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32 | 15:18 |
| fijal | this requires pypy | 15:18 |
| fijal | they use the same machine but the first one is not run nightly I think | 15:18 |
| Action: arigato tries to fix | 15:19 | |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) joined #pypy. | 15:21 | |
| neppord | okey.... | 15:21 |
| fijal | arigato: thank you! | 15:21 |
| neppord | is there any one who translates pypy on a mac? | 15:21 |
| fijal | neppord: yeah, a buildbot | 15:22 |
| neppord | ok | 15:22 |
| fijal | arigato: are we also interested in benchmarks on os/x | 15:22 |
| fijal | ? | 15:22 |
| neppord | so translate.py is complaining about that gcc-4.0 is not found | 15:22 |
| fijal | CC=... | 15:22 |
| fijal | ? | 15:22 |
| kenaan | 12fijal buildbot 116717b3a4b6a8 15/bot2/pypybuildbot/builds.py: clean up - now we dont benchmark cpython nightly twice | 15:22 |
| neppord | but i gott gcc from Developer tools(Xcode) | 15:22 |
| neppord | cc=...? | 15:23 |
| zk (~is@adsl-66-140-73-70.dsl.kscymo.swbell.net) left irc: Read error: Operation timed out | 15:23 | |
| neppord | https://gist.github.com/1086444 | 15:25 |
| neppord | if you would be nice and just look at the last rows of the trace | 15:25 |
| arigato | neppord: no, the issue is that your gcc is too recent | 15:26 |
| neppord | okay | 15:26 |
| arigato | if you can't install gcc 4.0, then you have to translate with --gcrootfinder=shadowstack | 15:26 |
| neppord | hmmpf | 15:26 |
| arigato | which gives a 5-10% slowdown in the end | 15:26 |
| neppord | what is that | 15:26 |
| fijal | meh :-/ | 15:26 |
| fijal | codespeed violates first normal form in the DB | 15:26 |
| fijal | or second | 15:26 |
| neppord | doesent help | 15:27 |
| neppord | get the same error | 15:27 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) joined #pypy. | 15:28 | |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) joined #pypy. | 15:29 | |
| fijal | justinpeel: hi | 15:29 |
| justinpeel | fijal: Hello | 15:29 |
| fijal | justinpeel: started cleaning up your branch (separating things from one branch into different places that can be merged) | 15:29 |
| fijal | but got called to do something else | 15:29 |
| justinpeel | that's cool. I was thinking of doing the same | 15:29 |
| justinpeel | I'm having some struggles with setting slices | 15:30 |
| fijal | I guess the approach will be to create new branches | 15:30 |
| fijal | like numpy-ndim-size | 15:30 |
| fijal | and transplant revisions | 15:30 |
| justinpeel | yes, that's what I was thinking too | 15:30 |
| fijal | cool | 15:30 |
| justinpeel | have you created any yet? I can set to it | 15:30 |
| fijal | no, I didn' | 15:30 |
| fijal | t | 15:30 |
| justinpeel | okay, I can work on that | 15:30 |
| fijal | be back in 30min | 15:31 |
| fijal (~fijal@41.53.101.44) left irc: Quit: Leaving | 15:31 | |
| arigato | fijal: started: http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/123/steps/translate/logs/stdio | 15:32 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 250 seconds | 15:36 | |
| tlynn_ (~tlynn@cpc6-cmbg14-2-0-cust121.5-4.cable.virginmedia.com) left irc: Quit: Leaving | 15:39 | |
| santagada_ | neppord: do you have xcode 4? | 15:40 |
| Nick change: santagada_ -> santagada | 15:40 | |
| santagada | neppord: xcode 3.2 comes with gcc-4.0 and gcc-4.2 | 15:41 |
| santagada | I would guess xcode 4 has them also | 15:41 |
| santagada | neppord: either find out where gcc-4.0 is or try to make pypy translate with clang and shadowstack | 15:42 |
| santagada | maybe it even works :) | 15:42 |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) left irc: Ping timeout: 252 seconds | 15:43 | |
| fijal (~fijal@196.24.41.254) joined #pypy. | 15:43 | |
| candrea (~andrea@ubuntu/member/beeseek.developer.andrea-bs) left irc: Quit: killall candrea | 15:44 | |
| Kaskuka (~John@osbk-4d08b9ef.pool.mediaWays.net) left irc: Ping timeout: 250 seconds | 15:45 | |
| santagada (~santagada@c925bcc7.virtua.com.br) left irc: Remote host closed the connection | 15:47 | |
| vokoda (~user@unaffiliated/vokoda) left irc: Remote host closed the connection | 15:48 | |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) joined #pypy. | 15:48 | |
| neppord | ah with --cc=gcc and not --CC=gcc it worked | 15:53 |
| fijal | ah no | 15:54 |
| kans (~kans@c-98-236-30-13.hsd1.pa.comcast.net) joined #pypy. | 15:54 | |
| fijal | I didn't say --CC | 15:54 |
| fijal | I said CC= | 15:54 |
| fijal | as in environmental variable | 15:54 |
| neppord | ah | 15:54 |
| fijal | anyway, you *have to* specify --gcstackrootfinder=shadowstack | 15:54 |
| fijal | otherwise it'll crash later on | 15:54 |
| neppord | okay.. is ther info on why? | 15:54 |
| neppord | cause i would be intrested in trying other compiler then gcc at all.. | 15:55 |
| arigato | --gcrootfinder=shadowstack | 15:58 |
| Moku (~John@osbk-4db14691.pool.mediaWays.net) joined #pypy. | 15:59 | |
| alfa_y_omega | o/, if 2.8 is released then it could be dangerous for the pypy's development, 2.8 could introduce its own bugs. However, pypy is verified for 2.7.x, and many apps work well for 2.7.x. There's not reason for migrating to 2.8 unless that it doesn't break the current pypy's development. | 16:03 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) joined #pypy. | 16:07 | |
| fijal | justinpeel: ok, so you or me? | 16:09 |
| justinpeel | fijal: I'll do it. I've been busy irl the last while | 16:09 |
| fijal | hrhrhr | 16:09 |
| Trundle (~andy@python/site-packages/trundle) left irc: Ping timeout: 246 seconds | 16:10 | |
| Trundle (~andy@dslb-178-002-050-160.pools.arcor-ip.net) joined #pypy. | 16:12 | |
| Trundle (~andy@dslb-178-002-050-160.pools.arcor-ip.net) left irc: Changing host | 16:12 | |
| Trundle (~andy@python/site-packages/trundle) joined #pypy. | 16:12 | |
| stakkars (~tismer@91-66-101-126-dynip.superkabel.de) left irc: Quit: schnarch | 16:13 | |
| Nick change: stakkars_ -> stakkars | 16:13 | |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) left irc: Quit: This computer has gone to sleep | 16:21 | |
| kenaan | 12fijal buildbot 11f1174e9107a5 15/bot2/pypybuildbot/builds.py: grrr, missing coma | 16:22 |
| fijal | arigato: can you install newer pypy as "pypy" on tannit? | 16:24 |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) joined #pypy. | 16:24 | |
| fijal | I mean globally | 16:24 |
| fijal | especially in 32bit | 16:24 |
| arigato | which version do you want? | 16:25 |
| fijal | nightly? | 16:26 |
| fijal | I mean it would cut down our translation time significantly | 16:27 |
| arigato | ok | 16:27 |
| arigato | how significantly? :-) | 16:27 |
| fijal | 2700s -> 1800s I think | 16:27 |
| arigato | uh | 16:27 |
| arigato | ok | 16:27 |
| fijal | wasn't that what cfbolz measured? | 16:27 |
| arigato | I don't know, when? | 16:27 |
| cfbolz | not quite that extreme | 16:27 |
| cfbolz | 30% faster | 16:28 |
| fijal | 30% you said | 16:28 |
| fijal | well | 16:28 |
| irmen (~irmen@razorvine.xs4all.nl) joined #pypy. | 16:28 | |
| fijal | that's 1889 | 16:28 |
| fijal | s | 16:28 |
| fijal | ok | 16:28 |
| arigato | done on tannit32, if you want to try | 16:30 |
| Action: fijal hopes buildbot will try | 16:32 | |
| arigato | done on tannit64 | 16:32 |
| fijal | thx | 16:32 |
| fijal | I'll set up 64bit build and benchmarks if stuff works out | 16:33 |
| arigato | good | 16:33 |
| fijal | do we want os/x 64bit? | 16:33 |
| fijal | I have a machine | 16:33 |
| arigato | if your machine is sure not to have random things running on it during the night, why not | 16:38 |
| kenaan | 12justinpeel numpy-singledim 1129e15adb6915 15/pypy/module/micronumpy/: Added simple repr and str for ViewArrays | 16:38 |
| kenaan | 12justinpeel numpy-ndim-size 11db28a9352aa2 15/pypy/module/micronumpy/: numpy: added ndim and size attributes. This includes work by Timo. | 16:38 |
| justinpeel | fijal: I haven't done anything on the ndim/size stuff since we talked yesterday | 16:39 |
| justinpeel | just going to break everything apart first | 16:39 |
| fijal | yes, sounds good | 16:41 |
| fijal | arigato: yes, it's a laptop laying down in a corner | 16:41 |
| fijal | or so I was told | 16:41 |
| cfbolz | arigato: did you see 72d3b692033c btw? | 16:43 |
| cfbolz | it made translate of targetrpystonedalone 10% faster and reduces the size of non-optimized traces by 45% | 16:44 |
| arigato | baaaaah | 16:45 |
| cfbolz | :-) | 16:45 |
| cfbolz | I learned a lot staring at some non-optimized traces yesterday on the train | 16:46 |
| arigato | and I'm mostly loosing my day changing the API of tealets | 16:48 |
| cfbolz | :-( | 16:49 |
| cfbolz | arigato: trying to make all tealets the same? | 16:49 |
| arigato | yes | 16:49 |
| arigato | I've got a nice API now | 16:49 |
| arigato | but implementing it is work | 16:50 |
| cfbolz | because the world changes | 16:51 |
| arigato | and because I have to really read Kristjan's changes :-) | 16:53 |
| kenaan | 12justinpeel numpy-repr-str 119e83042f6f27 15/pypy/module/micronumpy/: numpy: simple implementations of repr and str. includes work by Timo. | 16:54 |
| cfbolz | arigato: argh | 16:54 |
| timonator | wiiiiiiiii | 16:55 |
| timonator | not that i did much of anything :P | 16:55 |
| fijal | well, we used to run mac benchmarks on os x on minime | 16:55 |
| fijal | I guess it never worked | 16:55 |
| timonator | but i'm glad we have size and ndim now | 16:55 |
| justinpeel | yeah, I've been making a mess of things | 17:00 |
| justinpeel | so I'm cleaning stuff up now | 17:00 |
| Arc (arc@pysoy/developer/ArcRiley) left #pypy. | 17:05 | |
| irmen (~irmen@razorvine.xs4all.nl) left irc: Remote host closed the connection | 17:12 | |
| cfbolz | arghh | 17:21 |
| mvt | fijal, is there documentation explaining how to run the benchmarks on my own machine? | 17:23 |
| mvt | Want to compare shadowstack + clang vs gcc + asmgcc | 17:23 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 258 seconds | 17:30 | |
| danchr | mvt: on what platform? | 17:31 |
| mvt | osx | 17:32 |
| danchr | these two links might make it a bit easier for you: | 17:33 |
| danchr | http://dl.dropbox.com/u/6706818/pypy/pypy-c-jit-45622-87c5f89a4e78-osx64.tar.bz2 http://dl.dropbox.com/u/6706818/pypy/pypy-c-jit-45622-87c5f89a4e78-osx64.tar.bz2 | 17:33 |
| danchr | egh, they were the same, sorry | 17:33 |
| mvt | well i already have them translated, just wanted to run both bechmarks | 17:33 |
| danchr | http://dl.dropbox.com/u/6706818/pypy/pypy-c-jit-45622-87c5f89a4e78-osx64-shadowstack-clang.tar.bz2 | 17:33 |
| mvt | thanks anyway :-) | 17:33 |
| danchr | I tried running the benchmarks on my mac about a week ago; for some reason, many of them failed | 17:33 |
| mvt | any hints on what command i have to run? | 17:34 |
| danchr | some were much slower | 17:34 |
| danchr | lemme check my shell history | 17:34 |
| danchr | I did: | 17:35 |
| danchr | PYTHONPATH='' pypy ./runner.py -b 'richards,slowspitfire,ai,pyflate-fast,fannkuch,telco,meteor-contest,float,chaos,nbody_modified,spectral-norm,crypto_pyaes,raytrace-simple,go,waf' --baseline /tmp/pypy-c-jit-45137-65b1ed60d7da-osx64/bin/pypy -p /tmp/pypy-c-jit-45137-65b1ed60d7da-osx64-shadowstack-clang/bin/pypy | 17:35 |
| mvt | ok | 17:35 |
| danchr | I believe I just cloned the PyPy benchmarks repository | 17:35 |
| leemp (47d200e8@gateway/web/freenode/ip.71.210.0.232) left irc: Quit: Page closed | 17:36 | |
| mkhan (~mkhan@122.167.122.135) left irc: Quit: mkhan | 17:41 | |
| danchr | also, the most macro of all benchmarks, a full self-host: http://dl.dropbox.com/u/6706818/pypy/pypy-timings.txt | 17:42 |
| danchr | (which was ~2% slower) | 17:42 |
| mvt | running now :-) | 17:42 |
| danchr | I should update the bug I filed with those results& | 17:42 |
| mkhan (~mkhan@122.167.122.135) joined #pypy. | 17:42 | |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) left irc: Excess Flood | 17:43 | |
| CIA-55 | 03danchr 07roundup * 10#766/Shadowstack root finder should be competitive with asmgcc in speed: | 17:43 |
| CIA-55 | [chatting] I did another benchmark run a couple of days ago; a full self-hosting translation | 17:43 |
| CIA-55 | of PyPy. Shadowstack did it ~2% slower, but that's ... * 14https://bugs.pypy.org/issue766 | 17:43 |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) joined #pypy. | 17:43 | |
| kenaan | 12cfbolz default 11e5d0029d562d 15/pypy/jit/metainterp/optimizeopt/: bug in lazy setarrayitem. test and fix | 17:45 |
| cfbolz | arigato: could you review this commit please? | 17:45 |
| danchr | mvt: are the results in yet? :) | 17:50 |
| arigato | cfbolz: done... | 17:53 |
| cfbolz | arigato: makes sense? | 17:53 |
| arigato | it seems to be a different way of doing what was already done in emitting_operation | 17:53 |
| arigato | there, force_lazy_setarrayitem is followed by "for ... : cf._cached_fields.clear()" | 17:53 |
| cfbolz | ah, so I should maybe replace this with force_lazy_setarrayitem(..., can_cache=False) there? | 17:54 |
| arigato | yes, it seems to make sense | 17:55 |
| arigato | but the code is not completely equivalent | 17:55 |
| cfbolz | no | 17:55 |
| mvt | danchr, http://paste.pocoo.org/show/DubYLIRXJMFZRetE6aEf/ (shadowstack + clangs against python 2.7) | 17:57 |
| arigato | cfbolz: now I *think* that your patch is not enough | 17:57 |
| cfbolz | :-( | 17:57 |
| danchr | mvt: wouldn't it be the shadowstack+clang vs. asmgcc results that are really interesting? | 17:57 |
| arigato | cfbolz: yes, because if self._lazy_setfield is None, then the self._cached_fields are not cleared | 17:58 |
| mvt | yeah will be there retranslating the asmgcc version (accidently removed it.. ;-) | 17:58 |
| danchr | heh | 17:58 |
| cfbolz | arigato: ah, right | 17:58 |
| danchr | mvt: what sort of machine is this on? | 17:59 |
| arigato | cfbolz: try to write a test where there are only getarrayitem_gc(p0 ,0) before the setarrayitem_gc(p0, i1) | 17:59 |
| cfbolz | yes | 17:59 |
| mvt | danchr, macbookpro i5 2.4ghz / 8gb | 18:00 |
| danchr | heh, want | 18:00 |
| arigato | cfbolz: what about adding to OptHead.force_lazy_set{field,array} a can_cache flag, but not passing it, instead just doing the .clear() afterwards? | 18:01 |
| irmen (~irmen@razorvine.xs4all.nl) joined #pypy. | 18:01 | |
| arigato | cfbolz: it would make sense because right now you can get two KeyErrors one after the other | 18:01 |
| cfbolz | arigato: I don't follow | 18:01 |
| cfbolz | arigato: I guess if you fix it it is faster than telling me :-) | 18:01 |
| arigato | then let me try :-) | 18:02 |
| cfbolz | arigato: I wrote the test | 18:02 |
| danchr | mvt: I think shadowstack is a bit less stable on Mac OS X, though I've seen the odd segfault now and then | 18:02 |
| cfbolz | checkin coming | 18:02 |
| danchr | I get this when running bm_mako on my mac, any idea what it's about? http://paste.pocoo.org/show/440253 | 18:03 |
| arigato | cfbolz: thanks | 18:04 |
| kenaan | 12cfbolz default 118327f6fea2b2 15/pypy/jit/metainterp/optimizeopt/test/test_optimizebasic.py: broken test (armin will try to fix it) | 18:04 |
| Trundle (~andy@python/site-packages/trundle) left irc: Remote host closed the connection | 18:06 | |
| kenaan | 12arigo default 11e911b2205846 15/pypy/jit/metainterp/optimizeopt/heap.py: Fix the test by always clear()ing _cache_fields if can_cache=False. Rewrote some redundant try:except KeyError: co... | 18:08 |
| kenaan | 12cfbolz heap-caching-during-tracing 11afd433b6fc94 15/pypy/jit/metainterp/: setarrayitem does not influence the heap cache | 18:12 |
| kenaan | 12cfbolz heap-caching-during-tracing 11aacbf098fe2c 15/pypy/jit/metainterp/: intermediate checkin: starting an array cache. stopped by a bug on trunk. | 18:12 |
| kenaan | 12cfbolz heap-caching-during-tracing 11b60a9b13da69 15/pypy/: merge default | 18:12 |
| kenaan | 12cfbolz heap-caching-during-tracing 11e617ed3242d8 15/pypy/jit/metainterp/optimizeopt/: merge | 18:12 |
| kenaan | 12cfbolz heap-caching-during-tracing 11feafe98006d1 15/pypy/jit/metainterp/pyjitpl.py: fix second half of the test | 18:12 |
| fox__ (~fox@host42-106-dynamic.14-87-r.retail.telecomitalia.it) left irc: Ping timeout: 258 seconds | 18:12 | |
| cfbolz | arigato: makes sense | 18:12 |
| fox__ (~fox@host132-110-dynamic.8-87-r.retail.telecomitalia.it) joined #pypy. | 18:13 | |
| danchr | cfbolz: do you have an idea what might cause the trace I pasted earlier? | 18:15 |
| cfbolz | danchr: nope, sorry | 18:16 |
| danchr | it's weird that the benchmarks should fail on Mac OS X only& | 18:16 |
| mvt | danchr, all benchmarks worked for me | 18:17 |
| danchr | mvt: really? perhaps it's something about my setup, then& | 18:17 |
| danchr | did you list the benchmarks with -b? | 18:18 |
| mvt | used your cmdline | 18:18 |
| danchr | yeah, I filtered out the ones that crashed | 18:18 |
| danchr | could you try e.g. -b mako? | 18:18 |
| mvt | ok | 18:18 |
| mvt | same error | 18:19 |
| danchr | my results: http://paste.pocoo.org/show/440259/ | 18:20 |
| danchr | interestingly, ai is now faster :) | 18:20 |
| kenaan | 12cfbolz faster-nested-scopes 11623f7c22e667 15/pypy/: merge default | 18:22 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) joined #pypy. | 18:23 | |
| justinpeel | fijal: I got quite a bit of the slicing stuff working, but I can't seem to get things like arr[::2][::-1] = 2 | 18:25 |
| justinpeel | it seems to me that it would first create the slice arr[::2] and call setitem with a slice on it.. | 18:25 |
| justinpeel | arrview = arr[::2][::-1]; arrview[:] = 2 works right now though.. | 18:27 |
| neppord (~neppord@c-92f8e455.239-8-64736c10.cust.bredbandsbolaget.se) left irc: Quit: neppord | 18:43 | |
| lucian (~lucian@78-86-217-168.zone2.bethere.co.uk) left irc: Remote host closed the connection | 18:46 | |
| Alex_Gaynor | arigato: did you see my deadlock bug? | 18:47 |
| arigato | no | 18:47 |
| Alex_Gaynor | ok, it's in the tracker, I have no clue what's causing it, my current theory is race condition that shows up in threading.py because we are faster | 18:48 |
| Alex_Gaynor | mitsuhiko: no I ahven't solved it yet | 18:48 |
| kenaan | 12justinpeel numpy-setslice 119d55cebe8768 15/pypy/module/micronumpy/: numpy: setslice added but doesn't work correctly for setting a slice of a slice yet | 18:49 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 258 seconds | 18:53 | |
| durin42 (~durin@adium/durin42) left irc: Quit: leaving | 18:57 | |
| irmen (irmen@razorvine.xs4all.nl) left #pypy. | 19:07 | |
| Alex_Gaynor | fijal: is there anyway we can have the logs do somthing smarter when you have uint_ge(-1, i0), specifically I want it to show the (unsigned long)-1 | 19:17 |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) left irc: Quit: Off | 19:24 | |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) joined #pypy. | 19:24 | |
| LoganLK (~LoganLK@202-180-111-127.callplus.net.nz) joined #pypy. | 19:29 | |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) left irc: Quit: This computer has gone to sleep | 19:32 | |
| crakdmirror (~crakdmirr@174.127.114.26) left irc: Ping timeout: 252 seconds | 19:33 | |
| crakdmirror (~crakdmirr@174.127.114.26) joined #pypy. | 19:34 | |
| _main_ (~main@adsl-99-173-15-158.dsl.pltn13.sbcglobal.net) left irc: Ping timeout: 246 seconds | 19:35 | |
| __main__ (~main@adsl-99-173-15-158.dsl.pltn13.sbcglobal.net) joined #pypy. | 19:39 | |
| cousteau (~cousteau@80.174.59.133.dyn.user.ono.com) joined #pypy. | 19:43 | |
| cousteau | hi, I ahve some questions about PyPy | 19:44 |
| cousteau | first: why isn't the logo an Auryn? | 19:44 |
| Alex_Gaynor | what as Auryn | 19:44 |
| cousteau | http://en.wikipedia.org/wiki/The_Neverending_Story#AURYN | 19:46 |
| Alex_Gaynor | Because an Ouroboros is more appropriate I think http://en.wikipedia.org/wiki/Ouroboros | 19:46 |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) joined #pypy. | 19:47 | |
| cousteau | oh, so it also exists... well, now I know where the Auryn was inspired on :) | 19:47 |
| arigato | one Auryn == two Ouroboros... | 19:47 |
| cousteau | second, is PyPy supposed to be able to run any Python program out of the box? | 19:48 |
| Alex_Gaynor | any pure python program, that doesn't rely on implementation details of CPython, yes | 19:49 |
| cousteau | even with .so modules? | 19:50 |
| Alex_Gaynor | no, CPython C-Extensions do not fall under such a claim, they may work, but they are sow | 19:50 |
| arigato | ...slow | 19:50 |
| Alex_Gaynor | yes :) | 19:51 |
| cousteau | but they may work... interesting | 19:51 |
| Action: fijal back | 19:52 | |
| fijal | Alex_Gaynor: it' | 19:52 |
| fijal | s all software, you can do them whatever :) | 19:52 |
| Alex_Gaynor | fijal: pff, I meant is it sane | 19:52 |
| cousteau | third, do I need Python in order to run PyPy? | 19:54 |
| Alex_Gaynor | no, you do not need CPython to run a pypy binary | 19:55 |
| cousteau | (cause I doubt PyPy "runs itself") | 19:55 |
| mvt | danchr, see http://paste.pocoo.org/show/rTE982MptcWFWvgwj08N/ | 19:55 |
| Alex_Gaynor | fijal: we need virtual dicts I guess, Django has some hilarious nonsense (not worse than CPython, but it's nonsense all the same) | 19:55 |
| Alex_Gaynor | I still have a bit no clue why this is slower | 19:55 |
| fijal | virtual dicts are coming | 19:57 |
| fijal | stay tuned | 19:57 |
| cfbolz | Alex_Gaynor: for actual applevel dicts? | 19:58 |
| Alex_Gaynor | fijal: well they're easy with a combinatino of what you're doing, and optimizations in the frontend | 19:58 |
| Alex_Gaynor | cfbolz: yes | 19:58 |
| cfbolz | or stuff like **args | 19:58 |
| Alex_Gaynor | **args in this case | 19:58 |
| Alex_Gaynor | as long as they have constant keys @unroll_if_virtual should magically work | 19:58 |
| Alex_Gaynor | if it existed | 19:58 |
| cfbolz | I think then a dict impl for **args would be better | 19:59 |
| Alex_Gaynor | when you say **args do you mean call_func(**random_dict) or call_func(k=1, v=2, c=4) and def call_func(**kwargs) | 20:00 |
| kenaan | 12cfbolz heap-caching-during-tracing 11420fc8c6b8cd 15/pypy/jit/metainterp/: proper invalidation for the array cache | 20:00 |
| kenaan | 12cfbolz heap-caching-during-tracing 11ca2a85b7eef9 15/pypy/jit/metainterp/test/test_tracingopts.py: a passing test | 20:00 |
| kenaan | 12cfbolz heap-caching-during-tracing 1122dc9cdf9e06 15/pypy/jit/metainterp/: make promotion influence the heap cache | 20:00 |
| cfbolz | Alex_Gaynor: yes | 20:00 |
| Alex_Gaynor | which one, or both | 20:00 |
| arigato | phew, I'm getting through the tealet refactoring | 20:00 |
| cfbolz | Alex_Gaynor: ah, I guess I mean the latter mostly | 20:00 |
| Alex_Gaynor | yes, that's the one that is speed-up-able I think | 20:01 |
| Alex_Gaynor | cfbolz: btw any plans to merge your global and closure branches? | 20:01 |
| Action: fijal mumbles about getting benchmarks in order | 20:01 | |
| Alex_Gaynor | fijal: I looked at mako, I have a bit no clue. traces are the same before and after exactly. I don't see how allowing inlining into something which is never called can affect it | 20:02 |
| fijal | touch-the-world... | 20:02 |
| fijal | I don't know | 20:02 |
| Trundle (~andy@python/site-packages/trundle) joined #pypy. | 20:02 | |
| Alex_Gaynor | indeed | 20:03 |
| cfbolz | Alex_Gaynor: about merging: no clue. both are "nearly ready" | 20:03 |
| Alex_Gaynor | ok | 20:04 |
| cousteau | aww, bpython doesn't work... guess I'll need at least some kind of setup to make it work | 20:05 |
| fox__ (~fox@host132-110-dynamic.8-87-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds | 20:05 | |
| arigato | bpython isn't all pure Python? (does it rely on custom .so's?) | 20:06 |
| arigato | or else, I guess the best question is: what error do you get? :-) | 20:06 |
| fox__ (~fox@host145-107-dynamic.8-87-r.retail.telecomitalia.it) joined #pypy. | 20:07 | |
| Alex_Gaynor | cfbolz: these are kind of a lot of code duplication, aren't they? | 20:07 |
| cfbolz | http://darrenjw.wordpress.com/2011/07/16/gibbs-sampler-in-various-languages-revisited/ | 20:07 |
| kenaan | 12fijal benchmarks 110893b0a1eea4 15/saveresults.py: oops | 20:07 |
| cfbolz | Alex_Gaynor: these=? | 20:07 |
| Alex_Gaynor | cfbolz: manual optimizations in frntnend | 20:07 |
| Alex_Gaynor | frontend... | 20:07 |
| cfbolz | kind of | 20:07 |
| cfbolz | the trade-offs are very different | 20:07 |
| Alex_Gaynor | well, I just think it would be useful to somehow share code with optimizeopt, like the paper on automatic LICM for all of them :/ | 20:08 |
| cfbolz | yes but no. I'm repeating myself here, but this would require serious surgery on resume data | 20:09 |
| Alex_Gaynor | i suppose we don't have any doctors on staff | 20:10 |
| cousteau | http://ideone.com/Non4M <- O_o what was that? | 20:11 |
| cousteau | it was like 10x faster! | 20:11 |
| Alex_Gaynor | cfbolz: nice blog post btw | 20:12 |
| arigato | cousteau: :-) | 20:12 |
| arigato | "for example, why doesnt everyone always use PyPy for everything?!" | 20:13 |
| cfbolz | arigato: :-) | 20:14 |
| Arfrever | Some people want to use Python 3 features :) . | 20:14 |
| cousteau | so this project looks interesting... it reminds me of the "micro-controlled processors" idea | 20:15 |
| fijal | Arfrever: more people want other stuff | 20:16 |
| cousteau | (don't know how they were actually called) | 20:16 |
| cousteau | I want to run other apps on pypy | 20:16 |
| Alex_Gaynor | arigato, cfbolz: trunk is 33% faster than 1.5, I'll leavea comment | 20:16 |
| CIA-55 | 03danchr 07roundup * 10#766/Shadowstack root finder should be competitive with asmgcc in speed: | 20:17 |
| CIA-55 | mvt also did a run on Mac OS X, and his results mirror a run I just did: | 20:17 |
| CIA-55 | http://paste.pocoo.org/show/rTE982MptcWFWvgwj08N/ | 20:17 |
| CIA-55 | Shadowstack now va ... * 14https://bugs.pypy.org/issue766 | 20:17 |
| fijal | cfbolz: you should clarify, pypy 1.5 4x slower | 20:18 |
| fijal | we should do 1.6 | 20:19 |
| Arfrever | Is https://bitbucket.org/pypy/compatibility/wiki/Home up-to-date with PyPy 1.5? | 20:19 |
| cfbolz | fijal: oh, come on | 20:19 |
| Alex_Gaynor | fijal: we should fix regressions first, also some parts of ctypes are randomly slow | 20:19 |
| Action: cousteau is going to read what "cool things" can pypy do, other than running code fast | 20:19 | |
| fijal | cfbolz: :) | 20:19 |
| fijal | Alex_Gaynor: yes, we should | 20:19 |
| fijal | they're not "randomly" slow | 20:19 |
| fijal | anto only optimized the fast path | 20:19 |
| fijal | the slow path is still horribly slow | 20:19 |
| Alex_Gaynor | I'm not talking about that | 20:20 |
| Alex_Gaynor | well, although I claim passing str for c_char_p should be fast path | 20:20 |
| Alex_Gaynor | I mean thinks like `if pointer` are slow | 20:20 |
| Alex_Gaynor | they emit calls to _rawffi | 20:20 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) joined #pypy. | 20:21 | |
| fijal | the what? | 20:21 |
| fijal | how do they emit calls to _rawffi that are not inlined? | 20:21 |
| arigato | pypy.module._rawffi is not seen by the JIT | 20:22 |
| fijal | eh :/ | 20:23 |
| fijal | well | 20:23 |
| arigato | this should be fixed, but whoever fixes it needs to be prepared to add tons of jit.dont_look_inside and add support for other cases in the JIT, I fear | 20:23 |
| fijal | I kind of thought the whole idea of jittypes was to remove _rawffi dependency somehow | 20:23 |
| fijal | a lot has been added, but yes, sure | 20:24 |
| arigato | no, _rawffi is still used for data. _ffi only supports calls | 20:24 |
| Alex_Gaynor | IMO ctypes is becoming messier :/ | 20:24 |
| fijal | well ok | 20:24 |
| fijal | but then it's a bit of incomplete part | 20:24 |
| fijal | "take this one piece of ctypes and make it super fast whatever hacks are necessary" | 20:25 |
| arigato | well it's known to be in-progress, I suppose | 20:25 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 240 seconds | 20:25 | |
| fijal | arigato: is it? | 20:25 |
| arigato | besides, it's probably easy, but it needs various extensions in the JIT itself | 20:26 |
| arigato | like we added support for direct_ptradd on raw pointers last week | 20:26 |
| fijal | yes | 20:26 |
| fijal | of course | 20:26 |
| Alex_Gaynor | arigato: only on ccharp though :) | 20:26 |
| fijal | but I'm not too happy with the approach that jittypes took | 20:27 |
| fijal | well, who am I to judge though | 20:27 |
| arigato | Alex_Gaynor: yes | 20:27 |
| arigato | I suppose we need to stare at real traces and find the obvious "oh, this calls something in _rawffi that is a single operation" | 20:27 |
| Alex_Gaynor | cfbolz: we could make this code a bit faster if we inlined into module/random/ so we could avoid allocating flota objects for each ranomd() call | 20:27 |
| gutworth (~benjamin@64-131-13-109.usfamily.net) left irc: Quit: leaving | 20:27 | |
| Alex_Gaynor | arigato: well `if ctypes_ptr` generates serious nonsense ATM (aka it's a call, when it should be a single ASM instruction) | 20:28 |
| cousteau | nothing... I don't seem to be able to run bpython under pypy | 20:28 |
| arigato | Alex_Gaynor: obvious enough :-) | 20:28 |
| cousteau | my bash-fu isn't good enough | 20:28 |
| Alex_Gaynor | arigato: indeed | 20:29 |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) left irc: Excess Flood | 20:29 | |
| Da_Blitz (~Da_Blitz@203.56.250.63) left irc: Ping timeout: 240 seconds | 20:30 | |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) joined #pypy. | 20:30 | |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) joined #pypy. | 20:32 | |
| cfbolz | "We can only see a short distance ahead, but we can see plenty there that needs to be done." - Alan Turing | 20:32 |
| Alex_Gaynor | indeed | 20:33 |
| cfbolz | from the web page of the author of the blog post | 20:33 |
| Ademan | arigato: do you have a second for a stupid annotator error? http://paste.pocoo.org/show/440351/ produces http://paste.pocoo.org/show/440353/ when ll_unbox_integer is called via a direct_call() (if you need to see where I gen the direct call I can paste all of rerased.py or make a branch for this stuff to live in for now) | 20:34 |
| Ademan | it'd be nice if I didn't even have to make that dummy rpython function for ll_unbox_integer in the first place, but meh | 20:35 |
| Ademan | (i'm sure there's a way but I'm also sure I'm ignorant to it) | 20:36 |
| fijal | Ademan: what's the annotation of v30? | 20:36 |
| arigato | SomeOOObject()? that's a low-level annotation, it does not mean SomeInstance() at all | 20:37 |
| fijal | well, I can explain you what happened anyway | 20:37 |
| fijal | even if it was SomeInstance, that would not have attribute x | 20:37 |
| Ademan | fijal: my assert doesn't help? (well obviously it doesn't) | 20:38 |
| fijal | no, because annotator doesn't know that it has an attribute | 20:38 |
| fijal | well, there is more than one thing wrong | 20:38 |
| Ademan | I'm sure there is, I'm in over my head | 20:38 |
| Nick change: Gulaway -> Gulopine | 20:39 | |
| fijal | you can't pass instances there essentially | 20:39 |
| fijal | you need to have some low-level annotation | 20:39 |
| fijal | that has a precise type | 20:40 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 240 seconds | 20:40 | |
| Ademan | fijal: ugh, it will differ for CLI and JVM unless I wrap their respective native boxes in something else... | 20:40 |
| arigato | essentially, SomeOOInstance() is the oo equivalent of SomePtr(), which is not used in rerased.py either | 20:41 |
| arigato | so I don't know how you reached that annotation | 20:41 |
| Ademan | arigato: cluelessness | 20:41 |
| Ademan | ideally I would have an annotation for java.lang.Object and System.Object on the JVM and CLI respectively | 20:42 |
| arigato | look at how GCREF is used | 20:43 |
| fijal | arigato: http://speed.pypy.org/timeline/?exe=1&base=none&ben=nbody_modified&env=tannit&revs=50 | 20:43 |
| fijal | can we say this is just noise? | 20:43 |
| arigato | Ademan: it's not used in annotations, but in the Repr | 20:43 |
| arigato | Ademan: I fear that you need to understand how it works so far, at least partially; so far it looks like you are trying to apply a very different approach (which may make sense too or not, but more likely not :-) | 20:45 |
| Alex_Gaynor | fijal: maybe? it does appear to be trending slower, perhaps for now a perscription of careful observation? | 20:45 |
| Ademan | arigato: I'll take a look, thanks. It'd help if I had a better conceptual idea of what GCREF represented though. Is it an opaque pointer that will be followed by the GC? | 20:45 |
| Ademan | arigato: I thought I roughly did, but obviously not well enough | 20:46 |
| arigato | fijal: agreed with alex | 20:46 |
| arigato | Ademan: yes, exactly | 20:46 |
| arigato | Ademan: in other words it plays the same role as "a java.lang.Object" | 20:46 |
| Alex_Gaynor | fijal: I claim assembly offsets are wrong or something: http://paste.pocoo.org/show/440357/ | 20:46 |
| arigato | Ademan: well I'm unsure why you introduce register_external() in the first place | 20:47 |
| fijal | Alex_Gaynor: did you do hg pull -u on jitviewer? | 20:47 |
| Alex_Gaynor | fijal: yes | 20:47 |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) left irc: Quit: Off | 20:48 | |
| fijal | they might be | 20:48 |
| fijal | I guess pdb time? | 20:48 |
| fijal | to look what's wrong? | 20:48 |
| Alex_Gaynor | fijal: anyway this only seemed to happen at the very end of a trace, before finish() | 20:48 |
| Alex_Gaynor | so I'm not too bothered | 20:48 |
| fijal | and also you get what's there ;-) | 20:48 |
| fijal | there might be some optimization that puts the together | 20:48 |
| fijal | with one write barrier or so | 20:48 |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) joined #pypy. | 20:48 | |
| fijal | so I wouldn't be too concerned | 20:48 |
| Alex_Gaynor | uh, yes indeed I bet it's that | 20:48 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) joined #pypy. | 20:49 | |
| fijal | good :) | 20:49 |
| fijal | assembler is fun | 20:49 |
| Alex_Gaynor | it's amazing, we are fast at this but there is still tons of nonsense | 20:49 |
| Alex_Gaynor | 2 allocations in the inner gammavariant loop, plus a total mess of error checking around log calls | 20:50 |
| fijal | well, sky is the limit :) | 20:50 |
| fijal | I claim we owe a lot to good tools | 20:50 |
| Alex_Gaynor | C is the limit actually | 20:50 |
| fijal | if you ask me | 20:51 |
| Alex_Gaynor | (pun intended) | 20:51 |
| fijal | ;] | 20:51 |
| fijal | justinpeel: shout if some of the branches are up to review | 20:52 |
| fijal | justinpeel: also, did you want some help with something? | 20:52 |
| Alex_Gaynor | fijal: yes, good tools are important | 20:52 |
| Alex_Gaynor | fijal: I'd never notice that W_Random.random is allocating if jit viewr didn't make it very obvious | 20:52 |
| fijal | arigato: what did you use for staring at psyco's assembler? | 20:53 |
| Alex_Gaynor | are there tools which make starting at assembler less painful? | 20:53 |
| fijal | jitviewer? | 20:54 |
| arigato | a web server that serves pages containing a disassembly of one "block" of instructions, together with a summary of the state (equivalent to the virtuals/boxes/consts) | 20:54 |
| arigato | with the jumps being links | 20:54 |
| justinpeel (~nepo@c-76-23-44-72.hsd1.ut.comcast.net) left irc: Ping timeout: 252 seconds | 20:55 | |
| fijal | advanced :) | 20:55 |
| fijal | we still didn't get to jumps being links in jitviewer btw.... | 20:56 |
| arigato | uh? they are | 20:56 |
| arigato | if you click on a jump or on an often-failing guard, you see the target | 20:56 |
| fijal | even at the end of the loop? | 20:57 |
| fijal | ah ok | 20:57 |
| arigato | yes | 20:57 |
| arigato | I don't know, wasn't it you that implemented that? :-) | 20:57 |
| fijal | maaaaybe | 20:58 |
| Alex_Gaynor | we did it at the sprints | 20:58 |
| fijal | if so, then I was completely fried ;-) | 20:58 |
| arigato | :-) | 20:58 |
| fijal | or wasn't me | 20:58 |
| ThantiK (~User@c-71-203-92-115.hsd1.fl.comcast.net) joined #pypy. | 20:58 | |
| fijal | out of random facts, I'm super happy with where pypy is goind | 20:58 |
| arigato | ?-) | 20:58 |
| fijal | going even | 20:59 |
| fijal | at least we didn't create perl 6 | 21:00 |
| arigato | yes, we were going vaguely in the perl 6 direction at the end of the EU project, but then dropped all these half-thought-out attempts | 21:01 |
| arigato | and instead focused on just the core language | 21:02 |
| ThantiK | I'm just hoping PyPy allows someone, anyone, to be a little more serious when it comes to multimedia/graphics, etc. Right now multimedia on python is a steaming pile of joke. | 21:02 |
| Action: Alex_Gaynor personally hopes we drop even more random stuff | 21:02 | |
| arigato | :-) | 21:02 |
| Action: arigato feels almost bad adding tealets | 21:02 | |
| arigato | but only almost | 21:03 |
| fijal | we can drop stackless transform | 21:03 |
| fijal | that's something | 21:03 |
| arigato | yes :-) | 21:03 |
| Alex_Gaynor | the only people have ever used pypy for real stuff have been because of the JIT, why do we bother keeping ootype around | 21:04 |
| cousteau | I haven't managed to figure out how to run bpython yet... using python 2.6's packages, it finds pkg_resources but then it exits with "DistributionNotFound: bpython==0.9.6.2" | 21:04 |
| fijal | Alex_Gaynor: because pypy is designed in a way that it does not cost much ;-) | 21:04 |
| cousteau | (this is an installed version of bpython that I assume is intended to run on the installed Python [CPython]) | 21:05 |
| Alex_Gaynor | fijal: except when you break it because no one has ever tested it. | 21:05 |
| fijal | indeed we would feel bad if stackless were an ongoing maintenance nightmare and we no longer need it before it's actually useful | 21:05 |
| fijal | instead most burden was to set up a buildbot | 21:05 |
| Alex_Gaynor | old stackless will go away when we add tealets right? | 21:06 |
| Alex_Gaynor | (and tealets will be default) | 21:06 |
| Alex_Gaynor | bpython doesn't work because we don't have an _curses module | 21:06 |
| Unhelpful (~quassel@rockbox/developer/Unhelpful) left irc: Ping timeout: 276 seconds | 21:07 | |
| mvt (~mvantelli@53530442.cm-6-4a.dynamic.ziggo.nl) left irc: Quit: This computer has gone to sleep | 21:09 | |
| fijal | Alex_Gaynor: note that this was the first non-noise change in mako since january | 21:12 |
| kans (~kans@c-98-236-30-13.hsd1.pa.comcast.net) left irc: Quit: Leaving. | 21:12 | |
| Alex_Gaynor | fijal: ok so I don't claim it's not real, just that I have bit no clue how to further investigate | 21:12 |
| fijal | well, I'm just sayin for fun | 21:12 |
| Alex_Gaynor | well, but it should be fixed | 21:13 |
| Alex_Gaynor | I just have no idea how to | 21:13 |
| kkris1 (~kris@80-123-34-69.adsl.highway.telekom.at) joined #pypy. | 21:14 | |
| Action: cousteau gives up and goes to dinner | 21:14 | |
| cousteau | (nice project, though) | 21:14 |
| cousteau (~cousteau@80.174.59.133.dyn.user.ono.com) left irc: Quit: Ïa Ïa Cthulhu F'tang F'tang Olé Biscuitbarrel | 21:14 | |
| Alex_Gaynor | btw I'm making a list of possible task ideas for my sprint with pjenvey and brett cannon tomorrow, so far I have: 1) more multibytecodec stuff, 2) clean up multimethods, 3) compose an email about object idenitty for python-dev. Any other suggestions? | 21:14 |
| Trundle (~andy@python/site-packages/trundle) left irc: Ping timeout: 252 seconds | 21:14 | |
| kkris (~kris@80-123-38-183.adsl.highway.telekom.at) left irc: Ping timeout: 252 seconds | 21:15 | |
| Unhelpful (~quassel@rockbox/developer/Unhelpful) joined #pypy. | 21:17 | |
| fijal | Alex_Gaynor: 1) | 21:17 |
| derdon (~derdon@p5DE8A154.dip.t-dialin.net) joined #pypy. | 21:17 | |
| Alex_Gaynor | ? | 21:18 |
| fijal | the first item | 21:18 |
| fijal | from the list | 21:18 |
| Trundle (~andy@python/site-packages/trundle) joined #pypy. | 21:18 | |
| Alex_Gaynor | what about it | 21:18 |
| fijal | do it :) | 21:20 |
| fijal | either os x is super slow or we need to do something | 21:20 |
| cfbolz | Alex_Gaynor: I like them all | 21:20 |
| fijal | I wonder how fast is cpython ./translate.py on os x | 21:20 |
| Alex_Gaynor | the idea is to make a list | 21:20 |
| fijal | I fear multimethods might burn brett | 21:20 |
| Alex_Gaynor | so we can work on whatever interests them | 21:20 |
| kenaan | 12fijal benchmarks 11846fa56a282b 15/runner.py: add a postfix parameter | 21:22 |
| kenaan | 12fijal buildbot 113ca6784afb86 15/bot2/pypybuildbot/master.py: change the URL to the canonical one. will revert if it breaks stuff | 21:23 |
| kenaan | 12fijal buildbot 119f67b389b4fa 15/bot2/pypybuildbot/: enable nightly builds of pypy-c-64bit | 21:23 |
| kenaan | 12fijal buildbot 11ada781d23a2d 15/bot2/pypybuildbot/master.py: typo | 21:24 |
| kenaan | 12fijal buildbot 1100b2db6f1765 15/bot2/pypybuildbot/builds.py: another fix | 21:25 |
| fijal | windows crashed | 21:27 |
| fijal | might have been me restarting the buildbot though | 21:27 |
| fijal | arigato: feel like fixing? | 21:27 |
| magcius (jstpierre@unaffiliated/magcius) left #pypy ("Leaving"). | 21:29 | |
| Action: fijal doesn't know any more if "pypy 4x slower than C" is a good or bad thing | 21:29 | |
| magcius (~jstpierre@unaffiliated/magcius) joined #pypy. | 21:29 | |
| arigato | fijal: it crashed, but I have no clue why | 21:36 |
| arigato | I'm restarting a translation interactively | 21:36 |
| fijal | arigato: restarting the buildmaster might have caused that? | 21:36 |
| arigato | (please don't use the builder in the meantime) | 21:36 |
| fijal | if so I'm sorry | 21:36 |
| arigato | fijal: no, I think it occurred earlier than the restart | 21:37 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy. | 21:37 | |
| arigato | note that it's windows, which means that the process is limited to 2GB exactly | 21:37 |
| arigato | it's not impossible that what "error 0x4000000015" is trying to hint at is "out of address space" | 21:38 |
| fijal | aha :) | 21:40 |
| nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 252 seconds | 21:43 | |
| Ademan | is specialize_call considered an rtyping operation or an annotation operation? (also what calls specialize_call=) | 21:47 |
| mitsuhiko | Whoever pinged me, back now | 21:47 |
| mkhan (~mkhan@122.167.122.135) left irc: Quit: mkhan | 21:58 | |
| Ademan | hrm, I wonder if there's a better way to go about this, I'll have to implement a new opcode though... | 21:58 |
| mkhan (~mkhan@122.167.122.135) joined #pypy. | 21:59 | |
| cfbolz (~cfbolz@dslb-188-108-231-232.pools.arcor-ip.net) left irc: Quit: Leaving | 22:16 | |
| beatpanic (kpanic@gnufunk.org) left irc: Ping timeout: 240 seconds | 22:19 | |
| intertzu (~interrupt@m4.mullvad.net) joined #pypy. | 22:19 | |
| beatpanic (kpanic@gnufunk.org) joined #pypy. | 22:19 | |
| santagada (~santagada@c925bcc7.virtua.com.br) joined #pypy. | 22:23 | |
| zain[] (~textual@c-67-160-201-63.hsd1.ca.comcast.net) joined #pypy. | 22:25 | |
| donri (~donri@unaffiliated/dagodenhall) joined #pypy. | 22:25 | |
| mat^2 (~mathias@87.55.1.34) joined #pypy. | 22:26 | |
| santagada (~santagada@c925bcc7.virtua.com.br) left irc: Remote host closed the connection | 22:27 | |
| kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) joined #pypy. | 22:30 | |
| Alex_Gaynor | fijal: well 4x slower than C is not terrible, but we can do better :) | 22:34 |
| kkris1 (kris@80-123-34-69.adsl.highway.telekom.at) left #pypy ("Leaving."). | 22:34 | |
| kenaan | 12cfbolz heap-caching-during-tracing 11160b69e37ce6 15/pypy/jit/metainterp/: make sure the stuff works with resizable lists too | 22:38 |
| kenaan | 12cfbolz heap-caching-during-tracing 1134414cbab3ef 15/pypy/jit/metainterp/: make new construction go via the normal setfield code | 22:38 |
| kenaan | 12cfbolz heap-caching-during-tracing 11823933ada2fb 15/pypy/jit/metainterp/: make virtualizable code delegate to default implementations to get the right caching effect. | 22:38 |
| arigato (~arigo@82.113.98.162) left irc: Quit: See you | 22:54 | |
| Trundle (~andy@python/site-packages/trundle) left irc: Quit: ... | 22:59 | |
| rhumnzouc (~rhumnzouc@65.202.192.251) joined #pypy. | 23:11 | |
| rhumnzouc (rhumnzouc@65.202.192.251) left #pypy ("Leaving"). | 23:11 | |
| Alex_Gaynor | fijal: eh, is http://paste.pocoo.org/show/440433/ bad? | 23:12 |
| Alex_Gaynor | I suppose yes, my current pypy is broken | 23:13 |
| cousteau (~cousteau@80.174.59.133.dyn.user.ono.com) joined #pypy. | 23:19 | |
| Alex_Gaynor | http://speed.pypy.org/changes/?tre=10&rev=45678%3Ae911b2205846&exe=1&env=tannit | 23:23 |
| fijal | we broke stuff | 23:27 |
| fijal | or were you using tannit? | 23:27 |
| fijal | ### spitfire_cstringio ### | 23:28 |
| fijal | Min: 2.950000 -> 7.510000: 2.5458x slower | 23:28 |
| fijal | Avg: 4.823000 -> 7.572400: 1.5701x slower | 23:28 |
| fijal | that looks like a lot of difference between min and avg | 23:28 |
| Alex_Gaynor | I am not using tannit | 23:29 |
| Alex_Gaynor | was somebody? | 23:29 |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) left irc: Excess Flood | 23:29 | |
| Kami_ (~kami@unaffiliated/kami-/x-9078513) joined #pypy. | 23:30 | |
| dgl (~dgl@109.86.165.231) joined #pypy. | 23:32 | |
| fijal | I don't know | 23:46 |
| fijal | we can rerun the benchmarks maybe | 23:46 |
| ThantiK (User@c-71-203-92-115.hsd1.fl.comcast.net) left #pypy ("Leaving."). | 23:49 | |
| cousteau | can I use a module written in Cython on PyPy? | 23:51 |
| fijal | no | 23:51 |
| cousteau | aww :( | 23:51 |
| fijal | what module? | 23:52 |
| cousteau | one I wrote myself | 23:52 |
| fijal | for what? | 23:52 |
| cousteau | actually, just for testing cython | 23:53 |
| intertzu (~interrupt@m4.mullvad.net) left irc: Quit: Verlassend | 23:53 | |
| cousteau | http://ideone.com/CxXuf <- this one | 23:54 |
| Alex_Gaynor | the first one is plenty fast on pypy | 23:54 |
| fijal | if the only reason is speed, don't even bother | 23:55 |
| cousteau | Alex_Gaynor, I just experienced how it's like 14x faster than CPython, yes | 23:55 |
| cousteau | fijal, also I have a project I did using a lot of C and Cython | 23:56 |
| fijal | I guess time to rethink ;-) | 23:56 |
| Alex_Gaynor | 14x is ok I suppose | 23:57 |
| cousteau | ok... what about installed programs? any chance of running them on pypy instead of cpython? | 23:57 |
| cousteau | Alex_Gaynor, let me see how long it takes with each one | 23:57 |
| fijal | cousteau: what installed programs? | 23:58 |
| cousteau | well, bpython, for example | 23:59 |
| --- Sun Jul 17 2011 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!