IRC log for #brlcad on 20070807

00:33.41*** join/#brlcad iraytrace (n=iraytrac@sdcc-66-78-210-219.smartcity.com)
01:41.03*** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net)
01:49.31*** join/#brlcad IriX64 (n=IriX64@bas2-sudbury98-1177680337.dsl.bell.ca)
02:27.48*** join/#brlcad zap (n=deltazap@pool-72-64-222-114.tampfl.fios.verizon.net)
02:27.55*** part/#brlcad zap (n=deltazap@pool-72-64-222-114.tampfl.fios.verizon.net)
03:32.51*** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net)
05:27.01*** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net)
07:17.23*** join/#brlcad Laniakea (n=clock@zux221-122-143.adsl.green.ch)
08:50.03*** join/#brlcad Elperion (n=Bary@p54875F2F.dip.t-dialin.net)
11:44.17*** join/#brlcad thing0 (n=ric@203-59-168-92.perm.iinet.net.au)
12:51.18*** join/#brlcad ``Erik (i=erik@c-69-250-155-85.hsd1.md.comcast.net)
13:06.17*** join/#brlcad b0ef (n=b0ef@062016141081.customer.alfanett.no)
15:08.38*** join/#brlcad b0ef (n=b0ef@062016141081.customer.alfanett.no)
16:10.55*** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net)
16:11.36pooliorawr. I broke my debian install :\
16:16.41``Erikheh, sweet
16:18.12poolioI've been fixing it since like 2am.
16:18.30poolioand then my LCD went haywire, but then magically started working again
16:18.36poolioso I made sure to image my HD while I slept...
16:19.02``Erikso, uh
16:19.04``Erikwhy'd you break it?
16:19.04``Erik:D
16:19.08poolioI didn't mean to
16:19.22poolioI was trying to run something and it required an older version of gcc
16:19.31poolioand then my whole libc was somehow corrupted
16:19.35poolioand the apt tools wouldn't run
16:20.00poolioso I was planning on reinstalling after I was done working this summer, but I decided to just go ahead and do it being as the system was pretty not functional
16:20.01``Erikneat
16:20.03siggraphgood morning peopls
16:20.06``Erikdpkg stuff was broken, too?
16:20.07pooliomornin siggraph
16:20.09pooliohow's siggraph
16:20.20poolio``Erik: it wasn't broken, it just wouldn't fix the libc issue
16:20.28siggraphabout to listen to what is probably the best paper
16:20.31poolioor I should say I couldn't
16:20.36poolio...jealous
16:20.38poolioand which paper is that?
16:20.44``Erik'ways to make bzflag suck less'? *duck* :D
16:20.54siggraphautomatic cutaway illustrations
16:21.06``Erikhm, sounds interesting
16:21.23siggraphit's *impressive*
16:21.32siggraphs/automatic/interactive/
16:21.52poolioyou don't by any chance have a link to the paper?
16:22.55siggraphthis is related: http://portal.acm.org/citation.cfm?id=1276377.1276416&coll=portal&dl=ACM&CFID=15151515&CFTOKEN=6184618
16:23.21siggraphthe results made the siggraph cover
16:23.58poolioneat
16:25.42pooliosiggraph: you ever figure out where to put the freeing of re_directory_blocks?
16:42.10siggraphpoolio: eh, I've just now touched my laptop since the last time I talked to you, a bit busy :)
16:43.11siggraphthat was a good paper.. we can implement it pretty easily too I think -- very useful
16:46.48pooliosiggraph: alrighty
16:47.22pooliosiggraph: it appears most of the memory leaks are cleared up. the program eats up all my RAM but it never exceeds it and never goes into swap. It hovers right around 990/1024MB
16:47.44pooliothe bad thing is that the tests still aren't promising. even with the two-sphere trivial case it's hardly coming close to converging
17:13.31poolioSHIT
17:13.35poolioMy display is spazing out again
18:10.26*** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net)
18:10.37poolioeeeeeeeek.
18:36.03*** join/#brlcad ertugerata (n=Ertugrul@88.224.38.9)
18:36.09ertugeratasalut
19:03.00ertugeratadans mon system iwidgets and urt exist , si j'utilse parametre --enable-build-urt=no and --enable-iwidgets-build=no configure script corrupt.
19:21.51poolioanother bug bites the dust :)
19:22.48poolioertugerata: qu'est-ce que vous voulez dire avec "corrupt" ?
19:23.18ertugerataconfigure est termine
19:23.37CIA-4BRL-CAD: 03poolio * 10brlcad/src/librt/prep.c: free mro structure
19:24.15poolioertugerata: avec quoi? est-ce qu'il ya un erreur ou quelque chose?
19:24.40pooliolire "config.log" pour plus d'information sur la configuartion
19:25.01ertugerataon ne trove pas iwidgets and urt quand meme existe dans mon system
19:25.10poolioah d'accord
19:25.22ertugerataet aussi http://rafb.net/p/RJy0VK85.html tu peut voire ça
19:25.27pooliomerci
19:26.26pooliohmm, je ne suis pas sure, desole. je pense que c'est un probleme avec la versione de Tk que vous avez, mais je ne sais pas
19:27.25ertugeratapoolio: merci
19:27.34pooliobon chance :)
19:28.05ertugeratamon tk a 8.5alpha6 version
19:28.13poolioje pense que ca c'est d'accord
19:28.28poolioetes-vous utilisant CVS?
19:28.42ertugeratano source 7.10
19:29.13ertugerataCVS correct ou no ?
19:29.39poolioje pense que CVS est meilleur
19:29.58poolioibot: cvs?
19:29.59ibotit has been said that cvs is concurrent versions systems.  more info here http://www.cvshome.org/.
19:29.59ertugeratahmm
19:30.04poolio...
19:30.10poolioertugerata: un moment
19:30.26pooliohttp://www.brlcad.org/build_CVS.html
19:30.52poolioc'est possible que CVS marche meilleur pour vous
19:31.13ertugeratamerci bien
19:33.51pooliosiggraph: only bugs left are the re_directory_block and a tiny chunk for bu_semaphore, bu_semaphore_free exists but it's defined out for some reason
19:47.08``Erikfunny, I've been working on trying to get all that tcl 'junk' to use system shtuff today :) still haven't gotten iwidgets working against it, I think the tests don't grab the right libs? O.o
19:47.37poolioheh, so you guys are further away from release than I thought?
19:48.29poolioertugerata: est-ce que vous comprenez?
19:48.56ertugeratapoolio: download CVs maintenant
19:49.12ertugeratapuis je vais builder
19:49.17poolioertugerata: d'accord, mais ``Erik dit que CVS ne marche pas
19:49.33poolioil est un developer et il travaille sur la meme probleme
19:49.41ertugeratahmm
19:49.50ertugeratamerci bien
19:50.15poolioil dit que maintenant, ce n'est pas possible d'utiliser la versione de iwidgets avec votre systeme
19:50.31poolioil faut qu'on utilise iwidgets en BRL-CAD
19:50.45ertugerataurt ?
19:50.47poolioje ne sais pas avec urt
19:50.54poolio``Erik: you know if system urt works?
19:51.06``Erikyes, that's fine
19:51.45poolioertugerata: ^^ urt doit marcher
19:52.03ertugerataquand source sync vith cvs ?
19:52.08ertugeratapoolio: merci
19:52.24poolioertugerata: hmm...un nouveua release est proche, mais je ne sais pas quand exactement
19:52.41``ErikURT has worked with the last many releases, back to the 7.4.0 days that I know of
19:53.11ertugerataC'est bien. poolio
19:53.13poolioertugerata: URT doit marcher avec les source que vous avez (pas CVS)
19:53.15*** join/#brlcad Elperion (n=Bary@p54875F2F.dip.t-dialin.net)
19:53.27poolio7.10 source doit marcher
19:53.36pooliodoit mais c'est possible qu'il ne marche pas :)
19:53.53ertugeratapoolio: j'essaye mais ne maarche pas
19:54.09ertugeratapeut-etre mon urt a des probleme ?
19:54.13poolioqu'est-ce que c'est l'erreur avec URT?
19:54.15poolioc'est possible
19:54.22``Erikyes, http://www.freebsd.org/cgi/cvsweb.cgi/ports/cad/brlcad/Makefile?annotate=1.9
19:54.50``Erik7.6.0 was when I can confirm that system URT has worked :)
19:55.07poolioest-ce que vous pouvez mettre config.log sur http://rafb.net/paste ?
19:55.22poolioou la partie de configure avec le URT erreur
19:55.43poolio``Erik: fixed that bug with attr_values we were looking at the othe rday
19:55.58ertugeratapoolio: je l'ai perdu
19:56.05``Erikcool, what was it?
19:56.42poolioertugerata: je ne peux pas vous aider si je n'ai pas l'erreur, est-ce que c'est possible de ./configure encore?
19:56.46poolio(pardonner mon franglais)
19:57.30poolio``Erik: it just wasn't being freed properly on rt_clean(). It just ran bu_mro_free which assumes that the mro struct was allocated in the stack, so I had to bu_free() the mro struct along with bu_mro_free()ing it
19:57.47pooliojust commit'ed it a sec ago
19:58.02``Erikah, the cleaning *SHOULD* happen in prep...
19:58.10``Erikif I read the code right
19:58.14ertugeratapoolio: je vais reprodure dans 5 minete
19:58.28ertugerataminute
19:59.41poolioertugerata: merci beaucoup
20:00.12poolio``Erik: hmm? what cleaning? the mro struct should be free'd in rt_prep?
20:01.34``Erikum, I mean the attr stuff, but I gotcha know, missed that :)
20:03.06CIA-4BRL-CAD: 03erikgreenwald * 10brlcad/configure.ac: assume OpenBSD 4+ can deal with dots in the .so name
20:03.46poolio``Erik: any clue on this http://rafb.net/p/nE7AV443.html ... hard to debug :P
20:04.27``Erikdo you have a debugging libc handy? or, um
20:04.35poolionope
20:04.43``Erikuse, uh, nm to get symbol addy's and try to match those addresses to a name
20:04.48``Erikor objdump
20:04.53poolioah k
20:05.19``Erikassuming those aren't munged by relocation
20:05.20``Erik:(
20:07.23pooliohmm it looks like it's shifted or something?
20:07.38pooliolike objdump -x address don't correspond to valgrind output
20:07.47``Erikok, then it's been relocated
20:08.16``Erikif you can shove a debugging libc on and link (or prebind) it, then the stabs info will tell you the function name
20:08.23``Erikuhm
20:08.31pooliohmm, that's a bit over my head
20:08.32``Erikor it'll have it avialable, I dunno if valgrind is that cool
20:08.48poolioI think I'll let it be for now :)
20:08.53``ErikSTABS is a format for debugging data
20:09.00``Erikkinda like COFF and DWARF
20:09.04pooliok
20:09.33ertugeratapoolio: tres long
20:09.43poolioertugerata: je sais :)
20:09.45ertugerataje l'envoie ici
20:10.08poolioou?
20:10.09ertugerataaccept svp
20:10.15ertugeratairc
20:10.24poolioun moment
20:12.42poolioertugerata: ah envoi un autre fois s'il vous plait
20:13.12poolioca dit que vous avez mettre invalid parameters
20:13.18poolioenvoi par e-mail :)
20:13.20ertugeratapoolio: je l'envoie
20:13.38ertugeratamail adress ?
20:14.04``Erikhttp://pastebin.bzflag.bz/dc727094 heh, takes a long time :)
20:15.56poolioeeek I'd imagine
20:18.18``Erik<-- doing that after every change to configure.ac :/ trying to get iwidgets to pass
20:18.29pooliolame.
20:18.39pooliowhat system you on?
20:18.54``Erikquad opteron with 8g ram, it's a pretty beefy footrest
20:19.47``Erikthe nfs hit for my ~ is probably the big sticker
20:22.53``Erikhum, storm moving in, I'll have to roll up my windows soon
20:23.27poolioWhich way is it moving?
20:23.43``Erikeast
20:23.46``Erikhttp://www.weather.com/weather/map/interactive/21005?from=36hour_map_large&zoom=7&interactiveMapLayer=radar
20:25.01pooliohmmm
20:29.48pooliohey ``Erik, I'm trying to help ertugerata with getting his system URT working
20:30.08``Erikok?
20:30.09poolioit fails the test in configure but when he runs the code snippet outside of configure it works fine
20:30.22``Erikwhat's config.log say about it?
20:31.11pooliohttp://rafb.net/p/P7DvzS47.html
20:31.51``Erikthe error will be up, search backwards for lrle or cc or something, or just scroll up
20:31.52poolioHAVE_RLE_H = 1
20:32.06pooliooh dur
20:32.07pooliothanks
20:32.13``Erikthe format is "try to compile, show the error, print out the program I tried o do"
20:32.13``Erik:)
20:33.06pooliohmm
20:33.18poolioso if the proper libray isn't linked then it should be added?
20:33.30``Erikin theory...
20:33.50``Erikis it not named 'librle.so'?
20:34.40pooliothere's no librle.so being linke dto
20:35.11``Erikinteresting
20:35.45``Erikit should try -lutahrle, then -lrle if that fails
20:36.11``Erikaround line 1305 of configure.ac
20:36.33pooliohttp://rafb.net/p/3z6XZ659.html
20:36.42``Erikthere's also some testing around 2350
20:37.04poolioah I see, he needs -L/usr/lib not just -L/usr/local/lib
20:37.35``Erikld should know /usr/lib by default
20:37.35``Erikum
20:38.11``Erikon leenewx, it used to be /etc/ld.so.conf I think?
20:38.11``Erikit's still that on bsd *shrug*
20:39.54poolio``Erik: well the issue is it doesnt look like it's trying to link against rle
20:41.00``Erikyes, you should crack open configure in your favorite editor, go to that line and see why :D
20:41.11pooliorawr.
20:41.49``Erikwhile I prepare to put a boot up tcl's tailpipe for vexing me :)
20:41.50``Erikunless, of course you have a fix for: can't read "tcl_library": no such variable
20:42.27poolioI do. create a tcl_library variable.
20:43.04``Erikhehehe it's supposed to be created by Tcl_CreateInterp(), and that does get called and succeeds
20:43.19``Erik(and I think the variable is in tcl land, not C land)
20:44.45*** join/#brlcad dtidrow (n=dtidrow@host131.objectsciences.com)
21:05.06``Erikwow, that got ugly fast
21:06.46``Erikyeah, brlcad is gonna crap himself when I commit this... MWAHAHAHAHA
21:09.48poolioruh roh.
21:10.01pooliowhile brlcad is out ``Erik will play
21:10.26ertugeratail ya une problem avec enigma en CVS ?
21:10.40poolioavec enigma?
21:11.28ertugeratapas problem je crois
21:11.40pooliod'accord?
21:11.47ertugerataautıogen.sh ne marche pas just
21:11.58ertugerataje vais faire demain
21:12.00poolioquelle est l'erreur?
21:12.03poolioah d'accord :)
21:12.33ertugerataoui
21:12.48ertugerataje vais dormir je suis tres fatigue
21:12.55ertugerataa bientot
21:13.00pooliomais oui, au revoir
21:13.49*** part/#brlcad ertugerata (n=Ertugrul@88.224.38.9)
21:14.59``Erikhah
21:15.02``Erikwhat timing
21:15.35CIA-4BRL-CAD: 03erikgreenwald * 10brlcad/configure.ac:
21:15.35CIA-4BRL-CAD: Several fixes for the Iwidgets test: link itcl, initialize tcl before itcl, and
21:15.35CIA-4BRL-CAD: search the 'package names' even if the 'package require' claims to have failed
21:15.35CIA-4BRL-CAD: (TK version issue caused require to report failure, but actually succeed).
21:21.59pooliohehe
21:22.32poolioit's still not too ugly
21:23.24``Erikthe ternary to pump strtok is the only bit that might skeer newbs, I'd hope
22:17.46*** join/#brlcad b0ef (n=b0ef@062016141081.customer.alfanett.no)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.