00:51:57 -!- Blkt [~user@dynamic-adsl-94-37-226-26.clienti.tiscali.it] has quit [Quit: Error: do not makunbound t please!] 01:30:18 -!- nyef [~nyef@pool-64-222-178-106.man.east.myfairpoint.net] has quit [Quit: G'night all.] 02:09:49 The_Jon_Smith [~The_Jon_S@ip24-250-13-137.ri.ri.cox.net] has joined #sbcl 03:43:39 weird, unix-stat result is somehow scrambled 03:49:20 somehow mtime is mixed with atime 04:44:57 -!- The_Jon_Smith [~The_Jon_S@ip24-250-13-137.ri.ri.cox.net] has left #sbcl 05:50:04 ASau [~user@77.246.230.250] has joined #sbcl 05:59:06 -!- drdo [~user@98.192.108.93.rev.vodafone.pt] has quit [Ping timeout: 240 seconds] 06:06:37 mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #sbcl 06:37:30 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 240 seconds] 07:05:28 rbarraud__ [~rbarraud@118-93-187-24.dsl.dyn.ihug.co.nz] has joined #sbcl 07:06:16 -!- rbarraud_ [~rbarraud@118-92-8-84.dsl.dyn.ihug.co.nz] has quit [Ping timeout: 250 seconds] 07:22:45 -!- rbarraud__ [~rbarraud@118-93-187-24.dsl.dyn.ihug.co.nz] has quit [Remote host closed the connection] 08:16:20 ASau` [~user@77.246.230.250] has joined #sbcl 08:20:04 -!- ASau [~user@77.246.230.250] has quit [Ping timeout: 260 seconds] 08:37:17 tcr [~tcr@cpc5-bour5-2-0-cust340.15-1.cable.virginmedia.com] has joined #sbcl 08:44:33 ignotus [~ignotus@unaffiliated/ignotus] has joined #sbcl 08:49:35 rbarraud [~rbarraud@118-93-187-24.dsl.dyn.ihug.co.nz] has joined #sbcl 10:18:31 nyef [~nyef@pool-64-222-178-106.man.east.myfairpoint.net] has joined #sbcl 10:18:40 G'morning all. 10:26:05 nikodemus [~nikodemus@cs181199216.pp.htv.fi] has joined #sbcl 10:26:05 -!- ChanServ has set mode +o nikodemus 10:28:15 afternoon 10:29:24 Hello nikodemus. 10:30:02 *nyef* finally got his closure changes to survive a full build last night. 10:30:52 congrats! 10:32:29 Thanks. 10:33:49 Still a small pile of work to do on it, but it's nice that it's finally working. 10:55:13 mbohun [~mbohun@ppp115-156.static.internode.on.net] has joined #sbcl 11:45:00 -!- nikodemus [~nikodemus@cs181199216.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 13:15:23 jweiss [~weissj@cpe-069-134-025-078.nc.res.rr.com] has joined #sbcl 13:54:05 tcr1 [~tcr@cpc5-bour5-2-0-cust340.15-1.cable.virginmedia.com] has joined #sbcl 13:56:23 -!- tcr [~tcr@cpc5-bour5-2-0-cust340.15-1.cable.virginmedia.com] has quit [Ping timeout: 245 seconds] 14:11:24 Blkt [~user@160.80.132.14] has joined #sbcl 14:15:08 -!- tcr1 is now known as tcr 14:15:38 minion: memo for nikodemus: seems like 1.0.43.66 only changed version.lisp-expr 14:15:38 Remembered. I'll tell nikodemus when he/she/it next speaks. 14:28:21 nikodemus [~nikodemus@cs181058025.pp.htv.fi] has joined #sbcl 14:28:21 -!- ChanServ has set mode +o nikodemus 14:35:19 The_Jon_Smith [~The_Jon_S@ip24-250-13-137.ri.ri.cox.net] has joined #sbcl 14:47:15 nikodemus: I see what you mean about DX on LET-bound closures now. 14:47:57 Just spent about half an hour looking at it. 14:51:12 you came to conclude that _almost_ everything you need is available at ir1-translator for FUNCTION / in WITH-FUN-NAME-LEAF? 14:51:12 nikodemus, memo from tcr: seems like 1.0.43.66 only changed version.lisp-expr 14:51:21 blech 14:54:56 tcr: thanks for the heads-up. actual content went in as 1.0.43.68 14:56:08 will someone cry if i make regular namestrings (as opposed to native ones) on windows print using / as separators instead of the \? 14:58:12 or i should say \\, i guess :) 14:59:54 I certainly wouldn't. 15:00:08 Then again, I haven't used windows for development in ages. 15:00:25 goody. because then we can us a single unparse-physical-directory on both unix and windows 15:00:56 Actually, why on earth would we use backslashes as path separators on windows anyway? 15:01:55 stassats [~stassats@wikipedia/stassats] has joined #sbcl 15:11:06 i think the guy who did the original port decided to use them for some reason :) 15:11:17 no need that I see, windows is happy with forward slashes anyway (only cmd.exe is not, iirc, but who cares) 15:11:21 No, I didn't. :-P 15:11:42 perhaps we didn't have NATIVE-NAMESTRING yet back then? 15:11:48 That's plausible. 15:12:11 I knew very well back then that either slash was acceptable to the various APIs, though. 15:12:12 now, is there a way to get the home directory of another user on windows? 15:12:37 Probably two or three. 15:13:14 That said, it might require loading the user's registry hive and taking a look. 15:13:54 why not punt? :) 15:14:26 True. Go with the 90% solution for the 10% platform! 15:15:01 i think i will punt and signal an error 15:15:11 -!- nikodemus [~nikodemus@cs181058025.pp.htv.fi] has left #sbcl 15:15:18 nikodemus [~nikodemus@cs181058025.pp.htv.fi] has joined #sbcl 15:15:18 -!- ChanServ has set mode +o nikodemus 15:15:22 oops 15:37:47 Oh, wow. I just found a place in the compiler which uses LOOP! 15:42:11 stack analysis? 15:42:34 LOCALL / RECOGNIZE-DYNAMIC-EXTENT-LVARS. 15:42:50 oh, right! 15:43:25 hah, my mind isn't going after all. stack.lisp has a few LOOPs as well :) 15:43:40 So they're not /completely/ forbidden in the compiler. 15:43:45 That's good to know. 15:44:01 no. as long as it reads well and build :) 15:44:07 builds, even 15:48:46 the inherited cmucl code (almost?) never used loop; I think people have not used loop elsewhere just to be consistent 15:49:34 I think people just haven't rewritten enough of the compiler bits yet to make a dent 15:55:26 plus DO plays well with the intrusive lists we tend to have in Python. 15:55:57 for my part, it's just too easy to write a loop which is in fact broken 15:56:03 but which compiles fine until you try to use clisp 15:56:11 true. 15:56:32 which has the opposite interpretation of some unspecified behaviour, at which point things break subtly and then you have to reverse-engineer what the original loop actually meant 15:56:52 I'll happily start using loop in sbcl the day that sbcl's loop prohibits all violations of the loop grammar 15:57:09 Isn't someone working on that last? 16:00:27 just have a portable loop implementation 16:14:27 -!- Blkt [~user@160.80.132.14] has quit [Quit: Error: do not makunbound t please!] 16:26:40 Fare [~Fare@adsl-69-227-148-197.dsl.renocs.nvbell.net] has joined #sbcl 16:28:57 kclifton [~kclifton@64-110-238-43.regn.static.sasknet.sk.ca] has joined #sbcl 16:39:44 -!- vs [~fualo@markov.genomecenter.ucdavis.edu] has quit [Read error: Operation timed out] 16:40:38 vs [~fualo@markov.genomecenter.ucdavis.edu] has joined #sbcl 17:09:53 the error for (string '(#\a)) says (#\a) cannot be coerced to a string., while it clearly can be (coerce '(#\a) 'string) 17:11:18 does that make it a suboptimal error message or nobody will care? 17:11:42 It is perhaps slightly suboptimal. 17:11:54 Perhaps "is not a valid argument for STRING" would be better. 17:12:31 -!- Fare [~Fare@adsl-69-227-148-197.dsl.renocs.nvbell.net] has quit [Ping timeout: 240 seconds] 17:15:07 maybe specifically saying that it's not either "string, character, or symbol" would be more helpful 17:21:04 <_3b``> or it could just take advantage of the 'additional, implementation-defined' bit and call COERCE on it :p 17:21:39 Blkt [~user@dynamic-adsl-94-37-234-2.clienti.tiscali.it] has joined #sbcl 17:22:27 good evening everyone 17:23:26 _3b``: i wouldn't want to be too accustomed to that 17:26:06 "is not a string designator" 17:27:47 because most of the time that error message pops up is not because the person seeing called STRING, but because something else was using it to turn a string designtor into a string. or so i imagine 17:28:34 right, that's how i got to that error message 17:30:22 --> home 17:30:24 -!- nikodemus [~nikodemus@cs181058025.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 18:01:41 Fare [~Fare@adsl-99-68-116-221.dsl.renocs.sbcglobal.net] has joined #sbcl 19:01:18 -!- Fare [~Fare@adsl-99-68-116-221.dsl.renocs.sbcglobal.net] has quit [Ping timeout: 240 seconds] 19:24:46 -!- mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Read error: Connection reset by peer] 19:25:09 mega1 [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #sbcl 19:37:40 nikodemus [~nikodemus@cs181199216.pp.htv.fi] has joined #sbcl 19:37:40 -!- ChanServ has set mode +o nikodemus 19:50:41 -!- mbohun [~mbohun@ppp115-156.static.internode.on.net] has quit [Ping timeout: 252 seconds] 19:50:58 I land into ldb because mprotect returns ENOMEM; what is mprotect used for in the gc? 19:51:12 write barrier 19:51:39 you have a too fragmented memory map, and the kernel has run out of virtual memory mappings 19:51:40 what happens on the sigsegv? 19:52:09 to get rid of the ENOMEM just do one of the two things suggested by the error message 19:52:29 sure I'm trying with a higher limit; still curious about what it's used for 19:52:41 the sigsegv handler for writing to those pages marks the page as having been written to in sbcl's page table, and then unprotects the memory 19:53:33 then when collecting generation X, you only need to consider pages in X or pages marked in older generations that were written to 19:54:10 Ok yes I can see how that's easily the case here 19:54:53 it's not really anything you are doing wrong. it's sbcl being stupid 19:55:09 and arranging the data into memory in the worst possible way from the kernel's perspective 19:55:45 what does it do? what should it be doing instead? 19:56:52 it basically ends up interleaving boxed and unboxed pages. the latter is never mprotected, since unboxed data by definition won't contain pointers, and thus the gc can ignore it 19:57:27 instead it should keep the two kinds of pages segragated (easy given the unlimited address spaces on x86-64) 19:58:51 Hah yeah my thingie seems to be a bad contender in both respects :-) (occassional write to old generations, lots of boxed/unboxed data) 20:06:37 tcr: if you have *lots* of long-lived unboxed data, faking it with malloc can help. 20:11:15 i'm trying our STEP first time in anger 20:11:27 dammit, it actually works really nicely with modern slime 20:11:43 props to everyone involved! 20:13:34 -!- fe[nl]ix [~lacedaemo@pdpc/supporter/professional/fenlix] has quit [Quit: Valete!] 20:14:04 fe[nl]ix [~lacedaemo@pdpc/supporter/professional/fenlix] has joined #sbcl 20:32:59 sneaky sbcl debugging tricks part n: (setf type (type-union type (specifier-type `(eql ,(princ-to-string var))))) 20:37:08 -!- cmm [~cmm@bzq-79-182-202-208.red.bezeqint.net] has quit [Ping timeout: 265 seconds] 20:37:36 cmm [~cmm@bzq-79-182-202-208.red.bezeqint.net] has joined #sbcl 20:43:41 nikodemus: so frustrating when new features work! rollback to 0.9! :) 21:01:23 froydnj: reached for STEP while debugging another new feature, so the cosmic balance remains 21:01:28 see https://bugs.launchpad.net/sbcl/+bug/655203 21:01:48 screamer generated code: a bundle of joy 21:02:24 tmi [~tmi@198.80-202-113.nextgentel.com] has joined #sbcl 21:03:40 oof, yes 21:10:23 ASau`` [~user@77.246.230.250] has joined #sbcl 21:16:48 -!- nyef [~nyef@pool-64-222-178-106.man.east.myfairpoint.net] has quit [*.net *.split] 21:16:48 -!- ASau` [~user@77.246.230.250] has quit [*.net *.split] 21:20:29 Blkt` [~user@dynamic-adsl-94-34-40-93.clienti.tiscali.it] has joined #sbcl 21:24:15 nyef [~nyef@pool-64-222-178-106.man.east.myfairpoint.net] has joined #sbcl 21:24:31 -!- Blkt [~user@dynamic-adsl-94-37-234-2.clienti.tiscali.it] has quit [Ping timeout: 265 seconds] 21:28:35 Fare [~Fare@adsl-69-227-148-197.dsl.renocs.nvbell.net] has joined #sbcl 21:37:07 -!- nikodemus [~nikodemus@cs181199216.pp.htv.fi] has quit [Quit: This computer has gone to sleep] 21:50:15 -!- Blkt` is now known as Blkt 22:06:52 -!- Krystof [~csr21@csrhodes.plus.com] has quit [Ping timeout: 252 seconds] 22:11:48 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 245 seconds] 22:12:06 -!- The_Jon_Smith [~The_Jon_S@ip24-250-13-137.ri.ri.cox.net] has left #sbcl 22:20:15 Krystof [~csr21@csrhodes.plus.com] has joined #sbcl 22:20:15 -!- ChanServ has set mode +o Krystof 22:24:38 -!- Fare [~Fare@adsl-69-227-148-197.dsl.renocs.nvbell.net] has quit [Ping timeout: 240 seconds] 22:55:00 -!- tcr [~tcr@cpc5-bour5-2-0-cust340.15-1.cable.virginmedia.com] has quit [Quit: Leaving.] 23:05:44 -!- kclifton [~kclifton@64-110-238-43.regn.static.sasknet.sk.ca] has quit [Quit: kclifton] 23:19:39 Fare [~Fare@adsl-99-68-116-221.dsl.renocs.sbcglobal.net] has joined #sbcl 23:39:50 stassats [~stassats@wikipedia/stassats] has joined #sbcl 23:58:48 -!- Fare [~Fare@adsl-99-68-116-221.dsl.renocs.sbcglobal.net] has quit [Ping timeout: 240 seconds]