| Hamled|Erp (n=Charles@c-24-18-185-118.hsd1.mn.comcast.net) left irc: Read error: 110 (Connection timed out) | 00:02 | |
| psycoman_ (n=waldecir@20151109048.user.veloxzone.com.br) joined #pypy. | 00:22 | |
| mwhudson__ (n=mwh@118-92-26-35.dsl.dyn.ihug.co.nz) left irc: Read error: 104 (Connection reset by peer) | 00:32 | |
| jcea2 (n=jcea@jabber.hst.ru) left irc: Remote closed the connection | 00:32 | |
| jcea2 (n=jcea@jabber.hst.ru) joined #pypy. | 00:34 | |
| asabil (n=asabil@acces1249.res.insa-lyon.fr) left irc: "Ex-Chat" | 00:35 | |
| jkogut (n=jkogut@gw1.genesilico.pl) left irc: Remote closed the connection | 00:36 | |
| mwhudson__ (n=mwh@118-92-26-35.dsl.dyn.ihug.co.nz) joined #pypy. | 00:37 | |
| CIA-3 (n=CIA@208.69.182.149) left irc: Remote closed the connection | 00:38 | |
| CIA-3 (n=CIA@208.69.182.149) joined #pypy. | 00:42 | |
| Nick change: psycoman_ -> Psycoman | 00:50 | |
| jacob22 (n=jacob@c-22c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Read error: 110 (Connection timed out) | 00:58 | |
| jacob22 (n=jacob@c-22c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy. | 00:58 | |
| cfbolz (n=cfbolz@p4FDA936C.dip0.t-ipconnect.de) left irc: "Leaving" | 01:00 | |
| Hamled (n=Charles@c-24-18-185-118.hsd1.mn.comcast.net) joined #pypy. | 01:28 | |
| Hamled|Erp| (n=Charles@c-24-18-185-118.hsd1.wa.comcast.net) left irc: Read error: 110 (Connection timed out) | 01:46 | |
| nilton (n=nilton@201.82.128.100) joined #pypy. | 02:05 | |
| nilton (n=nilton@201.82.128.100) left irc: Remote closed the connection | 02:23 | |
| Hamled|Erp (n=Charles@c-24-18-185-118.hsd1.mn.comcast.net) joined #pypy. | 03:19 | |
| Hamled (n=Charles@c-24-18-185-118.hsd1.mn.comcast.net) left irc: Connection timed out | 03:28 | |
| Misto_ (n=Misto@72.37.205.4) left irc: | 04:11 | |
| fzzzy_ (n=donovan@c-76-126-86-69.hsd1.ca.comcast.net) joined #pypy. | 04:18 | |
| psycoman_ (n=waldecir@20151126129.user.veloxzone.com.br) joined #pypy. | 04:26 | |
| fzzzy__ (n=donovan@c-76-126-86-69.hsd1.ca.comcast.net) joined #pypy. | 04:26 | |
| fzzzy_ (n=donovan@c-76-126-86-69.hsd1.ca.comcast.net) left irc: Read error: 104 (Connection reset by peer) | 04:27 | |
| fzzzy_ (n=donovan@c-76-126-86-69.hsd1.ca.comcast.net) joined #pypy. | 04:43 | |
| Psycoman (n=waldecir@20151109048.user.veloxzone.com.br) left irc: Read error: 110 (Connection timed out) | 04:43 | |
| fzzzy__ (n=donovan@c-76-126-86-69.hsd1.ca.comcast.net) left irc: Read error: 113 (No route to host) | 04:43 | |
| fzzzy_ (n=donovan@c-76-126-86-69.hsd1.ca.comcast.net) left irc: Client Quit | 04:45 | |
| verte (n=verte@CPE-124-176-168-151.vic.bigpond.net.au) joined #pypy. | 04:47 | |
| panni__ (i=hannes@92.227.193.203) left irc: "( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de )" | 05:11 | |
| psycoman_ (n=waldecir@20151126129.user.veloxzone.com.br) left irc: Read error: 113 (No route to host) | 05:43 | |
| mwhudson_ (n=mwh@203-173-194-157.dsl.dyn.ihug.co.nz) joined #pypy. | 06:59 | |
| mwhudson__ (n=mwh@118-92-26-35.dsl.dyn.ihug.co.nz) left irc: Read error: 110 (Connection timed out) | 07:00 | |
| mwhudson_ (n=mwh@203-173-194-157.dsl.dyn.ihug.co.nz) left irc: Read error: 104 (Connection reset by peer) | 07:07 | |
| mwhudson_ (n=mwh@203-173-194-157.dsl.dyn.ihug.co.nz) joined #pypy. | 07:08 | |
| Rhamphoryncus (n=rhamph@unaffiliated/rhamphoryncus) left irc: "Moo. ISO 8601, yada yada. http://www.cl.cam.ac.uk/~mgk25/iso-time.html" | 07:11 | |
| mwhudson_ (n=mwh@203-173-194-157.dsl.dyn.ihug.co.nz) left irc: Read error: 110 (Connection timed out) | 07:28 | |
| mwhudson_ (n=mwh@118-92-18-178.dsl.dyn.ihug.co.nz) joined #pypy. | 07:30 | |
| mwhudson_ (n=mwh@118-92-18-178.dsl.dyn.ihug.co.nz) left irc: Read error: 104 (Connection reset by peer) | 07:37 | |
| mwhudson_ (n=mwh@118-92-18-178.dsl.dyn.ihug.co.nz) joined #pypy. | 07:38 | |
| joe_mp_ (i=joe@p549C7BBA.dip.t-dialin.net) joined #pypy. | 07:46 | |
| stakkars (n=tismer@72.37.205.4) left irc: | 07:53 | |
| verte (n=verte@CPE-124-176-168-151.vic.bigpond.net.au) left irc: "http://coyotos.org" | 08:05 | |
| joe_mp__ (i=joe@p549C640C.dip.t-dialin.net) left irc: Read error: 110 (Connection timed out) | 08:06 | |
| headius_ (n=headius@216.160.3.79) joined #pypy. | 08:19 | |
| headius (n=headius@216.160.3.79) left irc: Read error: 104 (Connection reset by peer) | 08:19 | |
| mwhudson__ (n=mwh@118-92-28-55.dsl.dyn.ihug.co.nz) joined #pypy. | 08:32 | |
| mwhudson_ (n=mwh@118-92-18-178.dsl.dyn.ihug.co.nz) left irc: Read error: 110 (Connection timed out) | 08:34 | |
| mwhudson__ (n=mwh@118-92-28-55.dsl.dyn.ihug.co.nz) left irc: Read error: 104 (Connection reset by peer) | 08:37 | |
| mwhudson__ (n=mwh@118-92-28-55.dsl.dyn.ihug.co.nz) joined #pypy. | 08:38 | |
| asabil (n=asabil@acces1249.res.insa-lyon.fr) joined #pypy. | 08:44 | |
| arigato (n=arigo@d212-152-11-57.cust.tele2.ch) joined #pypy. | 08:54 | |
| xorAxAx | arigato: i just got some feedback by somebody evaluating all ordered dict classes: about your impl, he says "this perfect odict fails 12 tests, and triggers exceptions in 2 others" (This message has been postponed.) | 08:54 |
|---|---|---|
| fzzzy_ (n=donovan@c-71-198-81-84.hsd1.ca.comcast.net) joined #pypy. | 08:58 | |
| arigato | xorAxAx: interesting but useless information so far :-) | 08:58 |
| jkogut (n=jkogut@host-53-19.compi.net.pl) joined #pypy. | 09:02 | |
| fijal (n=fijal@apn-77-115-38-29.gprs.plus.pl) joined #pypy. | 09:14 | |
| fijal | morning | 09:14 |
| CIA-3 | 03fijal * r53898 10pypy/pypy/doc/extending.txt: Document that was laying down in my wc forever. | 09:16 |
| arigato | morning | 09:16 |
| CIA-3 | 03fijal * r53899 10pypy/pypy/doc/discussion/gc.txt: first attempt to document new experimental gc features | 09:17 |
| fijal_ (n=fijal@apn-77-112-132-247.gprs.plus.pl) joined #pypy. | 09:25 | |
| CIA-3 | 03cami 07gameboy-emulator * r53900 10pypy/pypy/lang/gameboy/ (cartridge.py test/test_cartridge.py): introducing cartride classe which merges some functions from cartridgeStoreDriver and CartridgeFactory | 09:31 |
| xorAxAx | arigato: indeed, he used the ABC tests and they seem to be fairly broken itself :-) | 09:37 |
| arigato | isn't that testing the py3k dict api? | 09:44 |
| xorAxAx | arigato: he took the tests from 2.6 | 09:45 |
| xorAxAx | i think they backported that | 09:45 |
| fijal (n=fijal@apn-77-115-38-29.gprs.plus.pl) left irc: Read error: 110 (Connection timed out) | 09:47 | |
| arigato | can't find any such test | 09:50 |
| Action: xorAxAx checks | 09:50 | |
| xorAxAx | arigato: mapping_tests.py | 09:52 |
| arigato | ah, not related to abc then | 09:52 |
| xorAxAx | ok | 09:53 |
| arigato | this file is here almost unmodified since 2. | 09:53 |
| arigato | 2.4 | 09:53 |
| pjenvey_ (n=pjenvey@astound-69-42-4-166.ca.astound.net) left irc: | 09:55 | |
| Action: xorAxAx asks him | 09:57 | |
| arigato | many tests fail just because UserDict don't define __eq__ | 09:58 |
| fijal_ (n=fijal@apn-77-112-132-247.gprs.plus.pl) left irc: Read error: 110 (Connection timed out) | 09:59 | |
| arigato | wrong, sorry | 09:59 |
| arigato | many tests fail because OrederedDict is explicitly not == to a regular dict | 10:00 |
| arigato | it's only == to another OrderedDict in which the order is the same | 10:01 |
| xorAxAx | yep | 10:03 |
| cfbolz (n=cfbolz@p4FDA8870.dip0.t-ipconnect.de) joined #pypy. | 10:36 | |
| cfbolz | hi! | 10:36 |
| mwhudson__ | hello cf | 10:37 |
| Nick change: mwhudson__ -> mwhudson | 10:37 | |
| cfbolz | hi michael! | 10:38 |
| Action: cfbolz found a bug in SWI-Prolog today | 10:40 | |
| cfbolz | (yesterday night, more) | 10:40 |
| arigato | hi! | 10:45 |
| arigato | cfbolz: cool, we can add SWI-Prolog to the list of projects that PyPy found bugs in | 10:45 |
| cfbolz | yes | 10:46 |
| cfbolz | arigato: basically atom_concat(X, X, abcabc) gives wrong solutions | 10:46 |
| stakkars (n=tismer@adsl-71-133-157-82.dsl.irvnca.pacbell.net) joined #pypy. | 10:46 | |
| arigato | what is it? | 10:46 |
| stakkars | arigato: still up??? | 10:47 |
| arigato | ouch | 10:47 |
| xorAxAx | stakkars: its 11:46 | 10:48 |
| Action: stakkars got really weird timing problems with psyco, btw. | 10:48 | |
| arigato | waa | 10:48 |
| arigato | stakkars: you're getting the timezone correction algorithm wrong | 10:48 |
| arigato | it's almost lunch time around here | 10:48 |
| xorAxAx | cfbolz: i get No. | 10:48 |
| stakkars | arigato: ok, this hanntz again. NP :-) | 10:49 |
| xorAxAx | hmm | 10:49 |
| stakkars | when I'm going2bed late-nite, then it is the opposite 4u | 10:49 |
| xorAxAx | stakkars: how is psyco stuff working out? | 10:50 |
| stakkars | happns | 10:50 |
| stakkars | ktbrckboard... | 10:50 |
| stakkars | OK I will sleep NOW | 10:51 |
| stakkars | but something about psyco timing is completely insane. | 10:52 |
| stakkars | arigo: do you patch any of the timing functions? | 10:53 |
| stakkars | I got such a weird timing behavior that I doubt in myself. Which is within reason... | 10:54 |
| stakkars | ok, ok, I will find it alone, like so many times. | 10:56 |
| arigato | stakkars: I can't answer this without knowing more details | 10:56 |
| arigato | what timing functions are you referring to? time.time()? | 10:57 |
| stakkars | and I don't want to spoil you with details | 10:57 |
| stakkars | yes. | 10:58 |
| arigato | no, nothing special about the time module | 10:58 |
| stakkars | I found alarm, but that doesn't apply. | 10:58 |
| stakkars | There are some quite innocent timing functions from Raymond, got save his soul. | 10:59 |
| cfbolz | xorAxAx: isn't that wrong too? | 10:59 |
| antocuni (n=antocuni@host74-56-dynamic.14-87-r.retail.telecomitalia.it) joined #pypy. | 10:59 | |
| arigato | cfbolz: there is custom code in atom_concat for exactly this case... | 10:59 |
| antocuni | hi! | 10:59 |
| arigato | hi! | 10:59 |
| xorAxAx | cfbolz: hmm, the docs say "non-deterministically" but i guess one might reasonably expect X to be abc | 10:59 |
| antocuni | nice channel topic :-) | 11:00 |
| cfbolz | arigato: in the swi source? | 11:00 |
| arigato | yes | 11:00 |
| stakkars | they are repeating some timing of compiled psyco functions. | 11:00 |
| cfbolz | arigato: where? I tried to fix it but couldn't. anyway, it's fixed in the repo now | 11:00 |
| arigato | stakkars: are you talking about the code that I can't even see? | 11:01 |
| stakkars | an d from time to time, during some timing code that tests a few functions of a module, | 11:01 |
| arigato | cfbolz: grep atom_concat in src | 11:01 |
| arigato | cfbolz: ah, it's fixed in the swi repo? | 11:01 |
| stakkars | yes I am. forget about it. | 11:01 |
| cfbolz | arigato: I reported the bug yesterday and they fixed it | 11:02 |
| arigato | great | 11:02 |
| stakkars | I am trying a release, with Fabrizio, but we fall into deep traps. | 11:02 |
| cfbolz | arigato: which file are you looking at? | 11:02 |
| arigato | cfbolz: don't remember the file name, the function name was pl_atom_concat() | 11:02 |
| arigato | in a .c file | 11:02 |
| stakkars | I'm talking about code that you don't want to see. | 11:03 |
| arigato | the custom code I found was in the function concat() | 11:03 |
| stakkars | didn't | 11:03 |
| cfbolz | arigato: right, I looked at that too yesterday | 11:03 |
| cfbolz | arigato: but it didn't seem to account for sharing of the two first arguments | 11:04 |
| arigato | stakkars: that's right, I wouldn't have agreed to any condition about not redistributing the code, so I was not allowed to see it | 11:04 |
| xorAxAx | which is a bit weird because one might assume that people would have used that idiom before :) | 11:04 |
| arigato | cfbolz: then it was indeed fixed in the meantime :-) | 11:04 |
| cfbolz | arigato: when did you get your checkout? | 11:05 |
| arigato | 20 minutes ago | 11:05 |
| stakkars | confused ?!!? | 11:05 |
| cfbolz | arigato: right | 11:05 |
| stakkars | there are too many negations in "stakkars: that's right, I wouldn't have agreed to any condition about not redistributing the code, so I was not allowed to see it" for me to respond, properly | 11:06 |
| stakkars | I will try again, more positively. :-) | 11:07 |
| stakkars | anyway, I'm lost in translation, right now. | 11:08 |
| xorAxAx | would be a nice title for a movie about psyco or the pypy jit :) | 11:08 |
| cfbolz | :-) | 11:09 |
| stakkars | not that new, but still an opportunity | 11:09 |
| cfbolz | stakkars: are the hurdles for making a release technical right now? or political? | 11:09 |
| stakkars | arigato: just asking in a general way... | 11:09 |
| stakkars | technical. | 11:10 |
| stakkars | arigato: If I'm probing a compiled function, 7 times, and it happens to use a hundred times more time on the | 11:11 |
| cfbolz | well, maybe then you could make available the current version somewhere already, because helping is much easier if you can look at it | 11:11 |
| stakkars | first two calls, does that ring a bell with you? | 11:11 |
| stakkars | and this happens reproducably, on windows and osx. | 11:12 |
| xorAxAx | why do you think that it is "compiled"? | 11:12 |
| cfbolz | arigato: should we do a bit of prolog pairing? | 11:12 |
| stakkars | same code, just repeated. | 11:12 |
| stakkars | cfbolz: I need to ask for admission, of course. | 11:12 |
| arigato | stakkars: same comment as cfbolz | 11:13 |
| arigato | I've nothing against closed-source forks (I wouldn't use the MIT Licence) but don't count on me for blind-debugging it | 11:13 |
| fzzzy_ (n=donovan@c-71-198-81-84.hsd1.ca.comcast.net) left irc: | 11:13 | |
| arigato | cfbolz: I'm off to lunch | 11:13 |
| arigato | at around 1:30? | 11:13 |
| cfbolz | arigato: ok, sounds good | 11:14 |
| arigato | stakkars: sorry, can't help now | 11:14 |
| stakkars | I am somehow between the tree and the bork | 11:15 |
| xorAxAx | wuah: | 11:15 |
| xorAxAx | - | 11:15 |
| xorAxAx | +import os; | 11:15 |
| xorAxAx | .oO(no, semicolons are not needed to make code rpythonic) | 11:15 |
| stakkars | arigato: yes, I didn't expect this, either. just wanted to disturb. | 11:15 |
| cfbolz | xorAxAx: where is this? | 11:16 |
| xorAxAx | cfbolz: r53900 | 11:16 |
| xorAxAx | gbx branch | 11:16 |
| cfbolz | stakkars: sure, I understand it's not a fun situation for you. however, you are the only one who can effectively do anything against this | 11:16 |
| stakkars | cfbolz: I bugged arigo about three times, till now, and it always saved me a lot. that's why. :-) | 11:18 |
| cfbolz | stakkars: maybe you should bug whoever is responsible for allowing to release the code three times :-) | 11:18 |
| CIA-3 | 03cfbolz 07gameboy-emulator * r53901 10pypy/pypy/lang/gameboy/cartridge.py: kill some semicolons you don't need them in Python | 11:19 |
| stakkars | cfbolz: non comprehende. cause I announced it three times? | 11:19 |
| cfbolz | stakkars: I think you could go to your supperiors and say "I want to make it available now" | 11:20 |
| cfbolz | not as a release, just put it into a public repo | 11:21 |
| stakkars | this no longer a fun project; it turned out to be a PITA | 11:21 |
| cfbolz | :-( | 11:22 |
| stakkars | unfortunately there is more about it. But I will try on Monday. | 11:22 |
| cfbolz | well, ok | 11:22 |
| cfbolz | but I wouldn't expect armin to think about this as long as it is not available | 11:23 |
| stakkars | I just hoped for an "aha, yes, this is a common effect" effect. | 11:23 |
| stakkars | because the timing behavior defeats all my logic so far. | 11:24 |
| stakkars | but sure, I can come up with more dirty trousers to shudder about :-) | 11:25 |
| stakkars | g'nite | 11:25 |
| Action: cfbolz walk | 11:28 | |
| stakkars | arigato: is there any cheating about time.time/clock ? | 11:38 |
| stakkars | that's all I'm asking, today. | 11:38 |
| arigato | stakkars: maybe I found a way to solve the chicken-and-egg issue | 11:40 |
| stakkars | ????? | 11:40 |
| xorAxAx | stakkars: he means the non-available code :) | 11:40 |
| arigato | yes | 11:40 |
| arigato | stakkars: I can agree to help debugging the code if I receive a copy with a licence that says: | 11:41 |
| arigato | "this code can be freely redistributed under the MIT Licence after the 1st of July 2008" | 11:41 |
| arigato | or something like that | 11:41 |
| arigato | in this way, I have a guarantee that I'm helping an open source project, and I'm happy | 11:41 |
| stakkars | yes, but I see by my best wilt no way how this can be related to the code. It is just insane. | 11:42 |
| stakkars | Well, I will try it, as you said. | 11:42 |
| arigato | choose the date to be well after your expected release date | 11:42 |
| arigato | so that if the release occurs, this clause doesn't matter to you at all | 11:43 |
| stakkars | but how can already compiled psyco code behave such differently, or is it a mac issue? | 11:43 |
| arigato | I have no clue, it doesn't ring any bell to me | 11:44 |
| stakkars | ok, that's something to me. | 11:44 |
| stakkars | I'm refactoring like hell to see where speed goes, and things are completely counter-intuitive. | 11:46 |
| stakkars | yes I will ask them on Monday. | 11:46 |
| ikkei (n=teamikl@cap004-155.kcn.ne.jp) joined #pypy. | 11:46 | |
| stakkars | btw., compact object gave me a had time, again. I wished I had the time to refactor it to no longer abuse propagation. | 11:47 |
| stakkars | (tamed it, anyway, by another hack) | 11:48 |
| Action: stakkars swares to stop quibbling, now | 11:49 | |
| cfbolz | back | 12:38 |
| antocuni (n=antocuni@host74-56-dynamic.14-87-r.retail.telecomitalia.it) left irc: "Leaving" | 14:01 | |
| bigdo1 (n=scmikes@72-197-8-8-arpa.cust.cinci.current.net) joined #pypy. | 14:18 | |
| bigdog (n=scmikes@72-197-8-8-arpa.cust.cinci.current.net) left irc: Read error: 110 (Connection timed out) | 14:19 | |
| mwhudson (n=mwh@118-92-28-55.dsl.dyn.ihug.co.nz) left irc: Read error: 104 (Connection reset by peer) | 14:33 | |
| mwhudson (n=mwh@118-92-28-55.dsl.dyn.ihug.co.nz) joined #pypy. | 14:38 | |
| cfbolz (n=cfbolz@p4FDA8870.dip0.t-ipconnect.de) left irc: Remote closed the connection | 15:58 | |
| cfbolz (n=cfbolz@p4FDA8870.dip0.t-ipconnect.de) joined #pypy. | 15:59 | |
| xorAxAx | interesting, http://gcc.gnu.org/wiki/whopr | 16:03 |
| xorAxAx | http://gcc.gnu.org/wiki/DeadlySins of a compiler writer :) | 16:04 |
| Nick change: bigdo1 -> bigdog | 16:13 | |
| fijal (n=fijal@apn-77-115-22-157.gprs.plus.pl) joined #pypy. | 16:25 | |
| CIA-3 | 03fijal * r53910 10pypy/pypy/translator/c/genc.py: don't prevent profopt from happening | 16:27 |
| CIA-3 | 03fijal * r53911 10pypy/pypy/objspace/descroperation.py: hotpath for comparisons | 16:28 |
| xorAxAx | fijal: did you benchmark 53911? | 16:28 |
| fijal | yop | 16:30 |
| xorAxAx | cool, how much faster? | 16:30 |
| CIA-3 | 03fijal * r53912 10pypy/pypy/interpreter/pyframe.py: pre-alloc 20 items here. | 16:30 |
| fijal_ (n=fijal@apn-77-115-35-99.gprs.plus.pl) joined #pypy. | 16:33 | |
| CIA-3 | 03fijal * r53913 10pypy/pypy/interpreter/pyopcode.py: use one class instead of many. | 16:33 |
| Rhamphoryncus (n=rhamph@unaffiliated/rhamphoryncus) joined #pypy. | 16:43 | |
| xorAxAx | fijal: whats the advantage of 53919? | 16:53 |
| fijal (n=fijal@apn-77-115-22-157.gprs.plus.pl) left irc: Read error: 110 (Connection timed out) | 16:57 | |
| fijal_ (n=fijal@apn-77-115-35-99.gprs.plus.pl) left irc: Read error: 110 (Connection timed out) | 16:58 | |
| jacob22 (n=jacob@c-22c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Read error: 110 (Connection timed out) | 17:01 | |
| jacob22 (n=jacob@c-22c5e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy. | 17:02 | |
| codingmaster (n=georgy@srbk-590fbdb6.pool.einsundeins.de) joined #pypy. | 17:22 | |
| -NickServ (NickServ@services.)- This nickname is owned by someone else | 17:25 | |
| -NickServ (NickServ@services.)- If this is your nickname, type /msg NickServ IDENTIFY <password> | 17:25 | |
| Nick change: headius_ -> headius | 17:26 | |
| jkogut (n=jkogut@host-53-19.compi.net.pl) left irc: Remote closed the connection | 17:42 | |
| ikkei (n=teamikl@cap004-155.kcn.ne.jp) left irc: "Leaving..." | 18:02 | |
| Dekcarki (i=rico@port-ip-213-211-242-176.sta.reverse.mdcc-fun.de) left irc: "Verlassend" | 18:02 | |
| antocuni (n=antocuni@host242-122-dynamic.0-79-r.retail.telecomitalia.it) joined #pypy. | 18:19 | |
| tverwaes | xorAxAx: we - won't - tell - you - we - won't - tell - you | 18:25 |
| tverwaes | :) | 18:25 |
| xorAxAx | tverwaes: yeah, seems to be a secret checkin :) | 18:26 |
| cfbolz | it basically prevents a virtual method call | 18:26 |
| xorAxAx | cfbolz: yeah, but he does dispatching with if | 18:27 |
| xorAxAx | i guess that he will start pooling | 18:27 |
| xorAxAx | and needs polymorphic class instances :-) | 18:27 |
| cfbolz | stupid micro-optimizations | 18:28 |
| xorAxAx | :-) | 18:28 |
| xorAxAx | well, if you could change the type of instances in rpython, it might be easier :-) | 18:28 |
| pjenvey_ (n=pjenvey@astound-69-42-4-166.ca.astound.net) joined #pypy. | 18:31 | |
| panni_ (i=hannes@92.227.193.203) joined #pypy. | 18:41 | |
| pedronis (n=user@c-91d570d5.022-54-67626719.cust.bredbandsbolaget.se) joined #pypy. | 18:50 | |
| cfbolz | hi samuele! | 18:51 |
| pedronis | hi | 18:51 |
| arigato | hi | 18:52 |
| pedronis | 53913 unless is producing a really great speed up seems a bit overboard | 18:52 |
| arigato | argh | 18:53 |
| xorAxAx | pedronis: i dont think that it produces a speedup :) | 18:53 |
| arigato | it makes little sense in a smallfunc build IMHO | 18:53 |
| pedronis | that too | 18:53 |
| xorAxAx | 19:25:46 < xorAxAx> i guess that he will start pooling | 18:53 |
| xorAxAx | 19:25:57 < xorAxAx> and needs polymorphic class instances :-) | 18:53 |
| xorAxAx | but because he hasnt replied yet, we can only guess | 18:53 |
| cfbolz | but pooling makes no sense with a good gc | 18:54 |
| xorAxAx | cfbolz: depends, even the jvm does pooling and has a good gc :-) | 18:54 |
| pedronis | but this objects are really shortlived | 18:54 |
| tverwaes | stupid question: what does "pooling" mean? :) | 18:54 |
| pedronis | is not like they are frames | 18:54 |
| xorAxAx | tverwaes: manually keeping a freelist | 18:55 |
| tverwaes | or is it like a constants pool | 18:55 |
| tverwaes | ok | 18:55 |
| headius (n=headius@216.160.3.79) left irc: | 18:57 | |
| pedronis | xorAxAx: I would really like fijal to understand that logs messages are indeed there because we cannot read people brains | 19:03 |
| xorAxAx | hehe :) | 19:03 |
| Misto_ (n=Misto@dsl093-065-034.lax1.dsl.speakeasy.net) joined #pypy. | 19:04 | |
| pedronis | considering the tangle that calls are already also anything that gives speedups less than 20% should be left alone until we take a fresh look at the whole thing | 19:05 |
| xorAxAx | you mean indirect calls? | 19:05 |
| Hamled|Erp| (n=Charles@c-24-18-185-118.hsd1.wa.comcast.net) joined #pypy. | 19:05 | |
| pedronis | I mean python level calls | 19:06 |
| xorAxAx | or interpreter calls? (then what gives less than 20%?) | 19:06 |
| pedronis | I'm saying that I want checkins that says I got this speedup | 19:06 |
| xorAxAx | ah! | 19:06 |
| pedronis | and is larger equal thatn 20% | 19:06 |
| xorAxAx | yes, lets see what he presents | 19:07 |
| pedronis | instead of being left puzzling about what is going on in his brain | 19:07 |
| arigato | pedronis: I'll ask fijal when he comes around to please hack on a branch | 19:07 |
| pedronis | thanks | 19:08 |
| Hamled|Erp (n=Charles@c-24-18-185-118.hsd1.mn.comcast.net) left irc: Connection timed out | 19:25 | |
| bigdog (n=scmikes@72-197-8-8-arpa.cust.cinci.current.net) left irc: Read error: 110 (Connection timed out) | 20:15 | |
| bigdog (n=scmikes@72-197-8-8-arpa.cust.cinci.current.net) joined #pypy. | 20:15 | |
| xorAxAx | arigato: hopefully if space.py3k is not the only way to get it working ... | 20:15 |
| xorAxAx | (in a perfect world at least :)) | 20:16 |
| tav_ (n=tav@91.84.68.0) joined #pypy. | 20:38 | |
| panni__ (i=hannes@f049172126.adsl.alicedsl.de) joined #pypy. | 20:40 | |
| tav (n=tav@91.84.125.39) left irc: Read error: 110 (Connection timed out) | 20:49 | |
| panni_ (i=hannes@92.227.193.203) left irc: Read error: 110 (Connection timed out) | 20:50 | |
| headius (n=headius@fw.g5.dunn.pcspeed.com) joined #pypy. | 21:18 | |
| headius_ (n=headius@fw.g5.dunn.pcspeed.com) joined #pypy. | 21:19 | |
| headius (n=headius@fw.g5.dunn.pcspeed.com) left irc: Read error: 104 (Connection reset by peer) | 21:19 | |
| headius_ (n=headius@fw.g5.dunn.pcspeed.com) left irc: Read error: 104 (Connection reset by peer) | 21:21 | |
| headius (n=headius@fw.g5.dunn.pcspeed.com) joined #pypy. | 21:21 | |
| headius_ (n=headius@fw.g5.dunn.pcspeed.com) joined #pypy. | 21:24 | |
| headius (n=headius@fw.g5.dunn.pcspeed.com) left irc: Read error: 104 (Connection reset by peer) | 21:24 | |
| Nick change: headius_ -> headius | 21:32 | |
| fwierzbicki (n=fwierzbi@cpe-075-182-091-179.nc.res.rr.com) left irc: Read error: 104 (Connection reset by peer) | 21:43 | |
| fwierzbicki (n=fwierzbi@cpe-075-182-091-179.nc.res.rr.com) joined #pypy. | 21:44 | |
| pjenvey_ (n=pjenvey@astound-69-42-4-166.ca.astound.net) left irc: | 21:57 | |
| pjenvey_ (n=pjenvey@astound-69-42-4-166.ca.astound.net) joined #pypy. | 22:00 | |
| Misto_ (n=Misto@dsl093-065-034.lax1.dsl.speakeasy.net) left irc: | 22:04 | |
| headius (n=headius@fw.g5.dunn.pcspeed.com) left irc: | 22:08 | |
| santagada (n=santagad@201.21.248.4) joined #pypy. | 22:22 | |
| Misto_ (n=Misto@dsl093-065-034.lax1.dsl.speakeasy.net) joined #pypy. | 22:37 | |
| Action: pedronis goes to bed | 22:38 | |
| pedronis (n=user@c-91d570d5.022-54-67626719.cust.bredbandsbolaget.se) left irc: "night" | 22:38 | |
| santagada (n=santagad@201.21.248.4) left #pypy. | 22:38 | |
| Nick change: WAMtone -> eyetone | 22:42 | |
| panni_ (i=hannes@92.227.182.185) joined #pypy. | 22:46 | |
| antocuni (n=antocuni@host242-122-dynamic.0-79-r.retail.telecomitalia.it) left irc: "Leaving" | 22:47 | |
| arigato (n=arigo@d212-152-11-57.cust.tele2.ch) left irc: Read error: 110 (Connection timed out) | 22:55 | |
| panni__ (i=hannes@f049172126.adsl.alicedsl.de) left irc: Read error: 110 (Connection timed out) | 22:57 | |
| xorAxAx | hmm, py3k has __instancecheck__ and __subclasscheck__ ... that would finally mean that one could write completly transparent proxies, right? | 22:58 |
| xorAxAx | ah, hmm, type() remains | 22:58 |
| exarkun | type can already be fooled, at least on cpython | 23:08 |
| exarkun | hmm can it? maybe I'm thinking of something else. | 23:09 |
| exarkun | maybe isinstance is the thing that can already be fooled, so those two methods don't do anything | 23:09 |
| fzzzy (n=donovan@c-71-198-81-84.hsd1.ca.comcast.net) left irc: | 23:13 | |
| xorAxAx | exarkun: how can you fool it? :) by patching some __class__ attribute/descriptor somewhere? | 23:18 |
| bgola (n=Bruno@189.100.196.47) joined #pypy. | 23:19 | |
| exarkun | xorAxAx: yea | 23:21 |
| xorAxAx | hmm, a __type__ slot would be nice | 23:22 |
| xorAxAx | with some sys.real_type function :) | 23:22 |
| codingmaster (n=georgy@srbk-590fbdb6.pool.einsundeins.de) left irc: "Leaving" | 23:35 | |
| psycoman_ (n=waldecir@20151126129.user.veloxzone.com.br) joined #pypy. | 23:46 | |
| --- Mon Apr 21 2008 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!