00:00:53 But /which/ conventions? Sending a pull request to linus? 00:00:58 but I can put it in 00:01:54 ah. that. well, push to repo if you're a committer, or mail the exported patch somewhere (or use github's pull request method? I have no idea what host we'll be using, so this varies) 00:02:22 -!- tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has quit [Quit: Leaving.] 00:02:48 *nyef* points out that everyone already pulls from boinkor.net... And probably has a clnet account. 00:03:26 yeah, sure. then it's either push if you're an approved committer, or mail the patch / "here's a branch, see if you want to merge it" to sbcl-devel 00:05:36 (I'm definitely not going to hand out ssh accounts to people though, so boinkor.net is out as a potential host (-:) 00:06:01 Okay, fair enough. 00:06:26 if you setup gerrit on boinkor.net you wouldn't need to. And then there'd be a code-review tool too. :) 00:06:41 haha 00:06:44 (not sure you want to be supporting that, of course) 00:06:47 how about we let drewc handle that (: 00:07:11 I was thinking about slowly decomissioning boinkor.net - all I use it for nowadays is the repo sync and boinkmarks 00:07:16 And I'd rather move away from sourcefarce... But at the same time moving to clnet feels like putting all the CL eggs in one basket. 00:07:23 ...and random websites that could very well live somewhere else 00:07:34 it doesn't really matter, you can move the repo basically anytime 00:07:50 but it's HDDs are way older than I would expect them to live, and I'm nowhere near them to replace. so. 00:08:01 right 00:08:07 -!- gonzojive_ [~red@171.66.85.30] has quit [Quit: gonzojive_] 00:08:08 have repo.or.cz or github mirror wherever the commiter repo lives, and nobody else even needs to care, really. :) 00:09:04 -!- hargettp [~hargettp@pool-71-184-189-141.bstnma.east.verizon.net] has quit [Quit: Linkinus - http://linkinus.com] 00:10:21 foom: true (: 00:10:50 Even the committers don't need to care once they have their local repositories set up. 00:13:35 ...and they have somewhere to push their not-yet-mergeworthy changes to (which is easily solved through a github account or similar) (: 00:14:29 repo.or.cz account for me. 00:14:54 anything 00:43:58 homie` [~levgue@xdsl-78-35-138-102.netcologne.de] has joined #sbcl 00:44:04 wbooze` [~levgue@xdsl-78-35-138-102.netcologne.de] has joined #sbcl 00:46:01 romaeno [~proba@cpe-67-244-127-222.nyc.res.rr.com] has joined #sbcl 00:46:09 hey 00:46:11 anyone around? 00:46:36 wrap.o: In function `sb_mkstemp': /var/tmp/portage/dev-lisp/sbcl-1.0.45/work/sbcl-1.0.45/src/runtime/wrap.c:279: warning: the use of `mktemp' is dangerous, better use `mkstemp' 00:46:48 -!- homie [~levgue@xdsl-78-35-152-244.netcologne.de] has quit [Ping timeout: 250 seconds] 00:46:52 trying to compile dev-lisp/sbcl-1.0.45 00:47:08 I don't know why sbcl has C code on it oO 00:47:15 It's just a warning, ignore it? 00:47:43 And SBCL has less than 10% C code, it's primarily the GC, unix-signal handling, loading the actual lisp image, et cetera. 00:47:45 ah but it fails afterwards 00:47:59 ./linux-nm sbcl | grep -v " [FUw] " > ,sbcl.nm 00:48:08 make: *** [sbcl.nm] Error 1 00:48:10 Perhaps it fails, but not from using mktemp. 00:48:32 ... wait, 1.0.45? 00:49:02 That's... not a good version anyway 00:49:19 arghh 00:49:22 I put in a nasty compiler bug that I didn't manage to track down and fix until 1.0.45.5. 00:49:43 Shouldn't affect the build process that way, though. 00:51:10 You get an Error 1 running nm? 00:51:27 or running grep (: 00:51:38 Yeah, or the little shell script. But at that step? 00:52:27 romaeno: If you're going to build 1.0.45, please apply http://git.boinkor.net/gitweb/sbcl.git/commitdiff/3afdf2de234586523ed94941def9f25a8f7f4906 first. 01:02:07 i figure it out, the problem was my LDFLAGS="-s -Wl,-O1 -Wl,--as-needed -Wl,--sort-common -Wl,-z,now -Wl,--hash-style=gnu" 01:02:30 I changed to CXXFLAGS="${CFLAGS}" and it compiled just fine 01:16:44 -!- pers [~user@174-24-52-78.clsp.qwest.net] has left #sbcl 01:29:23 gonzojive_ [~red@c-71-198-7-84.hsd1.ca.comcast.net] has joined #sbcl 02:12:21 antgreen [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has joined #sbcl 02:23:06 -!- slyrus [~chatzilla@207.189.195.44] has quit [Ping timeout: 272 seconds] 02:39:39 -!- antgreen [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has quit [Remote host closed the connection] 02:47:33 antgreen [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has joined #sbcl 02:48:50 -!- antgreen [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has quit [Remote host closed the connection] 02:50:19 antgreen [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has joined #sbcl 03:39:41 antgreen` [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has joined #sbcl 03:41:00 -!- antgreen [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has quit [Ping timeout: 250 seconds] 03:42:17 cmm [~cmm@109.67.203.146] has joined #sbcl 03:45:20 -!- cmm- [~cmm@bzq-79-181-205-98.red.bezeqint.net] has quit [Ping timeout: 250 seconds] 04:10:15 -!- homie` [~levgue@xdsl-78-35-138-102.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 04:10:31 -!- wbooze` [~levgue@xdsl-78-35-138-102.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 04:21:17 -!- attila_lendvai [~attila_le@catv-80-98-24-21.catv.broadband.hu] has quit [Quit: Leaving.] 04:23:43 homie [~levgue@xdsl-78-35-138-102.netcologne.de] has joined #sbcl 04:32:20 -!- nyef [~nyef@pool-71-161-64-51.cncdnh.east.myfairpoint.net] has quit [Quit: G'night all.] 04:46:51 -!- homie [~levgue@xdsl-78-35-138-102.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 05:05:02 homie [~levgue@xdsl-78-35-138-102.netcologne.de] has joined #sbcl 05:08:16 slyrus [~chatzilla@207.189.195.44] has joined #sbcl 05:35:49 -!- NihilistDandy [~ND@c-75-69-129-101.hsd1.nh.comcast.net] has quit [Remote host closed the connection] 06:46:32 tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has joined #sbcl 06:59:50 -!- tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has quit [Quit: Leaving.] 07:02:43 tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has joined #sbcl 07:17:08 -!- tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has quit [Quit: Leaving.] 07:19:58 tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has joined #sbcl 07:40:32 -!- tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has quit [Quit: Leaving.] 08:45:34 pdponze [~pdponze@144.85.121.191] has joined #sbcl 09:18:25 nikodemus [~nikodemus@cs181199216.pp.htv.fi] has joined #sbcl 09:18:25 -!- ChanServ has set mode +o nikodemus 10:01:26 -!- homie [~levgue@xdsl-78-35-138-102.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 10:08:03 should i delete this? http://www.reddit.com/r/lisp/comments/fhztw/on_reddit_downtime/c1ghrwg 10:08:20 homie [~levgue@xdsl-78-35-138-102.netcologne.de] has joined #sbcl 10:09:03 no, I was quite cross about the comment you're replying to too 10:10:04 but I don't have a reddit account so I was just going to ignore lnostdal more pointedly next time a request came in 10:11:08 it reminded me a bit of the trigger to dan_b's epic "has anyone even thought about making things better" rant 10:12:24 i almost put in something about having considered forking sbcl and keeping my stuff for my customers for 24 months before releasing them to PD, but then i decided that would have been FUD of its own 10:12:45 'cos, *shit* 11:05:39 attila_lendvai [~attila_le@catv-80-98-24-21.catv.broadband.hu] has joined #sbcl 11:21:42 -!- attila_lendvai [~attila_le@catv-80-98-24-21.catv.broadband.hu] has left #sbcl 11:52:52 mega1 [~user@catv4E5CABA2.pool.t-online.hu] has joined #sbcl 12:01:56 bubo [~bubo@178-191-149-171.adsl.highway.telekom.at] has joined #sbcl 12:09:50 -!- antgreen` [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has quit [Ping timeout: 240 seconds] 12:14:53 Blkt [~Blkt@dynamic-adsl-94-34-26-111.clienti.tiscali.it] has joined #sbcl 12:20:57 hargettp [~hargettp@pool-71-184-188-224.bstnma.east.verizon.net] has joined #sbcl 12:24:54 good day everyone 12:25:56 hi 12:33:34 hi there 12:36:51 attila_lendvai [~attila_le@catv-80-98-24-21.catv.broadband.hu] has joined #sbcl 12:46:34 -!- pdponze [~pdponze@144.85.121.191] has left #sbcl 12:51:07 wow, that's a crappy comment 13:00:36 -!- homie [~levgue@xdsl-78-35-138-102.netcologne.de] has quit [Read error: Operation timed out] 13:00:38 homie` [~levgue@xdsl-84-44-153-177.netcologne.de] has joined #sbcl 13:11:09 It is. I wonder who wrote it. 13:14:10 tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has joined #sbcl 13:16:44 -!- gonzojive_ [~red@c-71-198-7-84.hsd1.ca.comcast.net] has quit [Quit: gonzojive_] 13:20:50 -!- homie` [~levgue@xdsl-84-44-153-177.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 13:24:28 homie [~levgue@xdsl-84-44-153-177.netcologne.de] has joined #sbcl 13:25:51 what comment? 13:28:16 stassats [~stassats@wikipedia/stassats] has joined #sbcl 13:29:03 nikodemus: the one you flamed 13:29:34 oh, i already forgot about it. obviously ranting helped :) 13:29:56 :) 13:31:50 apropos, i'm preparing to merge a ton of other people's patches from lp, so unless someone is in a rush, please don't commit any changes to NEWS for a while... 13:32:14 (or put them on the top where they won't conflict with mine...) 13:47:18 -!- hargettp [~hargettp@pool-71-184-188-224.bstnma.east.verizon.net] has quit [Quit: Leaving...] 14:12:22 -!- bubo [~bubo@178-191-149-171.adsl.highway.telekom.at] has quit [Ping timeout: 250 seconds] 14:14:12 bubo [~bubo@62-47-148-101.adsl.highway.telekom.at] has joined #sbcl 14:16:24 oh, _that_ comment. Thought it was the timer one. 14:16:47 heh 14:25:07 mega1_ [~quassel@catv4E5CABA2.pool.t-online.hu] has joined #sbcl 14:25:07 -!- mega1_ [~quassel@catv4E5CABA2.pool.t-online.hu] has quit [Client Quit] 14:25:46 -!- attila_lendvai [~attila_le@catv-80-98-24-21.catv.broadband.hu] has quit [Ping timeout: 265 seconds] 14:49:42 -!- slyrus [~chatzilla@207.189.195.44] has quit [Ping timeout: 276 seconds] 14:53:03 slyrus [~chatzilla@207.189.195.44] has joined #sbcl 15:11:23 mega1: I'm guessing that crappy timer comment was mine. :-/ 15:12:39 mega1: I'm still pretty green especially wrt. reading SBCL code. Gimme time (and/or advice). 15:12:48 I did commit it, but honestly can't remember. 15:13:26 redline6561: you had the patch for the crappy comment 15:13:31 not the original comment 15:13:48 (i'm merging that today sans a minor fix) 15:14:04 Oh, the original. 15:14:05 ah, you are who you are. I see. 15:14:16 nikodemus: Thanks for the review. Sorry I didn't run tests. I'll do better next time. 15:14:21 hargettp [~hargettp@pool-71-184-188-224.bstnma.east.verizon.net] has joined #sbcl 15:15:01 mega1: I was confused by your "did commit it, can't remember" too. Interesting mix up. :) 15:21:52 oh man are you all on ubuntu ? 15:22:10 i see alot of lp stuff pointing to ubuntu 15:22:28 and not only that also github is pointing to ubuntu too 15:22:35 I'm not 15:22:41 hmmm ok 15:22:47 I tolerate launchpad because I can use it using just e-mail 15:30:22 antgreen [~user@CPE00222d6c4710-CM00222d6c470d.cpe.net.cable.rogers.com] has joined #sbcl 15:36:35 i *run* ubuntu on this box, but i'm not *on* ubuntu, personally 15:38:30 is sb-kernel:*handler-clusters* supposed to be unusual? 15:39:11 i get different results when i evaluate in the REPL, evaluate with C-x C-e and inspect in the inspector 15:39:46 the inspector results are most strange: # CAR: -312216098 CDR: 38551601 15:39:46 15:40:08 stassats: it's stack-allocated 15:40:30 so if you look at in the inspector you're likely to be looking at garbage 15:41:30 that's what i see, SBCL cl:inspector gets a memory fault 15:43:13 not surprising, really. inspecting garbage is dangerous 15:43:39 inspector should probably signal a CERROR before inspecting anything not heap-allocated 15:55:40 nikodemus: i'm doing this for slime now, what operations can i do on stack allocated objects? can i safely do type-of? 15:56:26 stassats: if the stack frame in which the object lives has returned, there isn't really anything you ca do safely 15:56:59 if the stack frame is still live, you can do anything you can normally do -- except let a reference outlast the stack frame 15:57:02 so, i how do i avoid doing anything? 15:57:53 (setf ref-to-a-stack-allocated-object nil) ? 15:58:12 how do i know it's stack allocated? 15:58:20 is doing sb-introspect:allocation-information safe? 15:58:33 mm, let me think 15:59:07 it *is* the right tool to answer the question "can i let this escape", but what about after the fact? hm 15:59:45 i now think that support of this in slime is only useful for things like sb-kernel:*handler-clusters* 15:59:50 yeah, should be mostly safe 16:00:30 "mostly" in the sense that i don't see there anything that looks like obviously dangerous 16:00:33 does sbcl spit horrible warnings when stack allocated object escapes? 16:00:38 nope 16:00:46 it doesn't really know when it happens 16:01:25 because dynamic-extent declaration is a sacred promise that things _will_ not escape 16:02:50 in any case, i suppose it's better to use sb-introspect:allocation-information than PRINT 16:04:00 yep :) 16:04:46 alright, is there a way to know if the stack is still alive? 16:04:57 but then again, I regularly inspect stack-allocated objects -- but it's one of the reasons i prefer :fd-handler, because i can keep the state of the stack in my mind much easier, so i know when it is safe 16:04:59 although i don't think that should ever be the case with slime inspector 16:05:39 now that you say it, i think it might be the case sometimes 16:06:25 stassats: not really. you _could_ look at the current stack pointer and compare -- in some cases you can see that the stack doesn't extend to where the object is supposed to be anymore, but it can "look" live, but have been overwritten since by another frame in the same area 16:07:04 i often stick a break or equivalent somewhere, and then go from the debugger backtrace to the inspector -- in which case everything is nice and live 16:07:13 *stassats* abandons the idea 16:07:40 but if you leave the inspector open and eg. abort the debugger ... boom, you just lost the live stack frames 16:10:12 ok, i think i'm done with sbcl for today 16:12:13 I actually hacked up guarding against inspecting unwound objects: http://common-lisp.net/~trittweiler/talks/slime-lightning-talk-eclm-2009.pdf 16:13:50 how? where? 16:14:14 never comitted it because I expected helmut not to like it 16:14:37 too much extra code for too little gain 16:14:57 well... 16:15:39 -!- nikodemus [~nikodemus@cs181199216.pp.htv.fi] has quit [Ping timeout: 255 seconds] 16:15:40 *stassats* waits till he has time to write his own slime, in CL, using Qt 16:47:27 gonzojive_ [~red@c-71-198-7-84.hsd1.ca.comcast.net] has joined #sbcl 17:25:48 -!- bubo [~bubo@62-47-148-101.adsl.highway.telekom.at] has left #sbcl 18:15:53 -!- romaeno [~proba@cpe-67-244-127-222.nyc.res.rr.com] has quit [Ping timeout: 276 seconds] 18:19:46 nikodemus [~nikodemus@cs181199216.pp.htv.fi] has joined #sbcl 18:19:46 -!- ChanServ has set mode +o nikodemus 18:25:42 -!- nikodemus [~nikodemus@cs181199216.pp.htv.fi] has quit [Ping timeout: 255 seconds] 18:46:40 nikodemus [~nikodemus@cs181199216.pp.htv.fi] has joined #sbcl 18:46:40 -!- ChanServ has set mode +o nikodemus 19:33:17 -!- cmm [~cmm@109.67.203.146] has quit [Ping timeout: 240 seconds] 19:34:21 cmm [~cmm@109.67.203.146] has joined #sbcl 20:12:45 -!- tcr1 [~tcr@217-162-131-235.dynamic.hispeed.ch] has quit [Quit: Leaving.] 20:16:19 Hexstream [~hexstream@modemcable075.97-200-24.mc.videotron.ca] has joined #sbcl 20:20:21 I want to get started with SBCL development (trivial patches for now). What source should I be developing against? Simply the 1.0.45 source at http://prdownloads.sourceforge.net/sbcl/sbcl-1.0.45-source.tar.bz2?download ? I think the main repo is in CVS, but I'd rather not touch that. Is there a "canonical" git mirror I can rely on that gets merged in the main repo regularly?... 20:20:54 antifuchs's boinkor mirror 20:23:01 Ok, so that would be somewhere on boinkor.net, but where exactly? 20:23:55 git.boinkor.net 20:24:15 git://sbcl.boinkor.net/sbcl 20:25:36 Thanks. That should really be linked from somewhere on boinkor.net (if it isn't). 20:30:44 rather it should be pointed to from sbcl.org ... 20:46:54 <|3b|> git.sbcl.org ? 20:48:27 i rather ment a pointer to sbcl.boinkoer.net from the webpage 20:49:24 That would be handy! 20:50:56 Is there any planned CVS --> git migration? I don't think that would hurt participation ;) 20:52:46 Hexstream: see recent discussion (and the ones before that) on sbcl-devel 20:53:41 Gotta subscribe to that mailing list... Any other SBCL-related mailing lists or news sources I should be aware of? 20:54:37 sbcl-help, sbcl-commits, launchpad 20:55:46 Thanks. Now I can slowly get an SBCL dev addiction. 21:02:30 git.sbcl.org exists already (points at boinkor.net, but could be redirected). probably a good idea to start using that alias now, since antifuchs did kind of hint that might not run the repo forever :-) 21:11:50 I will run it for as long as sbcl is on cvs (: 21:12:16 but once there's an official git repo, this will either become a mirror or outdate pretty quickly. 21:12:17 that doesn't sound like an ultimatum 21:12:33 haha, not at all right (: 21:13:10 seriously though: I just don't want to make this repo a read/write one 21:13:11 I thought you said something about the disks being old and you being in a different country than the server :-) 21:13:28 readonly is fine by me for as long as the machine can breathe (: 21:13:37 well, yeah that 21:15:35 (sorry about the punctuationlessness of my messages. this netbook's keyboard is just crap) 21:37:07 romaeno [~proba@cpe-67-244-127-222.nyc.res.rr.com] has joined #sbcl 22:10:07 -!- nikodemus [~nikodemus@cs181199216.pp.htv.fi] has quit [Quit: Ex-Chat] 22:34:38 -!- Hexstream [~hexstream@modemcable075.97-200-24.mc.videotron.ca] has left #sbcl 23:30:48 -!- mega1 [~user@catv4E5CABA2.pool.t-online.hu] has quit [Ping timeout: 260 seconds] 23:53:50 -!- Krystof [~csr21@csrhodes.plus.com] has quit [Ping timeout: 240 seconds]