| 00:29.46 | *** join/#brlcad Twingy (n=justin@74.92.144.217) | |
| 00:32.10 | *** join/#brlcad Twingy (n=justin@74.92.144.217) | |
| 01:42.52 | *** join/#brlcad andrecastelo (n=chatzill@189.71.7.20) | |
| 01:43.17 | andrecastelo | good evening |
| 01:43.18 | andrecastelo | hi brlcad |
| 01:54.03 | brlcad | howdy andrecastelo |
| 02:05.04 | andrecastelo | good night guys hehehe |
| 02:05.16 | andrecastelo | tomorrow i'll be here the whole day :b hehe |
| 02:05.18 | andrecastelo | cya |
| 02:34.41 | starseeker | yukonbob: Yes and yes :-) |
| 03:34.00 | yukonbob | starseeker: heh :) |
| 04:41.18 | *** join/#brlcad pacman87 (i=127@resnet-45-192.dorm.utexas.edu) | |
| 04:44.31 | CIA-20 | BRL-CAD: 03starseeker * r30651 10/brlcad/trunk/src/proc-db/tire.c: Fix subtractions for ell intersections so wireframe does not extend so far beyond actual tire surface. |
| 05:28.46 | *** join/#brlcad illethal (n=oden@c-71-200-221-159.hsd1.fl.comcast.net) | |
| 05:29.39 | *** part/#brlcad illethal (n=oden@c-71-200-221-159.hsd1.fl.comcast.net) | |
| 05:41.47 | *** join/#brlcad clock_ (n=clock@77-56-95-244.dclient.hispeed.ch) | |
| 07:03.37 | *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch) | |
| 07:21.55 | *** join/#brlcad d_rossberg (n=rossberg@bz.bzflag.bz) | |
| 07:43.50 | brlcad | yay brep csg reply |
| 07:44.06 | brlcad | waves g'morning/g'night to d_rossberg |
| 08:35.23 | d_rossberg | wishes a well sleeping to brlcad! |
| 09:56.23 | *** join/#brlcad mafm (n=mafm@elnet-111.lip.pt) | |
| 09:56.58 | mafm | hallo |
| 10:50.36 | *** join/#brlcad SJ (n=surya@203.200.95.130) | |
| 10:50.44 | SJ | hey guys |
| 11:00.24 | mafm | hey |
| 11:01.52 | SJ | this is hippieindamakin8 btw :) |
| 11:06.00 | mafm | :D |
| 11:18.49 | SJ | sean can u gimme a link to Mr.Mike Muuss's papers on ray tracing |
| 13:07.56 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) | |
| 13:11.21 | *** join/#brlcad andrecastelo (n=chatzill@189.71.41.1) | |
| 13:12.19 | andrecastelo | hi everyone |
| 13:13.47 | mafm | hi |
| 14:40.58 | brlcad | yawns |
| 15:00.27 | prasad_ | arrr |
| 15:06.48 | ``Erik | yaaarrrgggghhhhhh, matey |
| 15:08.45 | poolio | brlcad: shouldn't you bet at work already? :) |
| 15:17.07 | mafm | nobody should bet at work, specially with the bosses |
| 15:19.32 | poolio | mafm: yeah yeah...I'm still getting used to this keyboard |
| 15:24.09 | mafm | which keyboard is that? |
| 15:27.56 | brlcad | poolio: yep |
| 15:37.51 | *** join/#brlcad Elperion (n=Bary@p54873B47.dip.t-dialin.net) | |
| 15:42.20 | *** join/#brlcad andre|away (n=chatzill@189.71.41.1) | |
| 15:45.53 | *** part/#brlcad prasad_ (n=psilva@static-70-108-244-218.res.east.verizon.net) | |
| 16:37.25 | *** join/#brlcad andrecastelo (n=chatzill@189.71.41.1) | |
| 16:38.51 | yukonbob | morning cadheads |
| 16:39.43 | andrecastelo | yukonbob: good morning / afternoon :) |
| 16:59.17 | andrecastelo | is there a point list structure already implemented in brlcad ? |
| 17:01.57 | *** join/#brlcad prasad_ (n=psilva@static-70-108-244-218.res.east.verizon.net) | |
| 17:27.17 | CIA-20 | BRL-CAD: 03bob1961 * r30652 10/brlcad/trunk/misc/nsis/brlcad.nsi: ReadEnvStr must be called within a function or a section. Also, PROGRAMFILES cannot be set (i.e. its a constant defined by nsis). Put the ReadEnvStr inside the .onInit callback function. |
| 18:01.19 | *** join/#brlcad clock_ (n=clock@77-56-82-18.dclient.hispeed.ch) | |
| 18:37.46 | *** join/#brlcad micges (n=michu@public-gprs66780.centertel.pl) | |
| 18:38.19 | *** part/#brlcad micges (n=michu@public-gprs66780.centertel.pl) | |
| 18:40.26 | mafm | heading off, take care |
| 18:51.47 | *** join/#brlcad docelic (n=docelic@78.134.198.227) | |
| 19:19.34 | *** join/#brlcad fi-dschi (n=philipp@p5498A033.dip0.t-ipconnect.de) | |
| 19:53.58 | fi-dschi | hello, is there somebody listening? |
| 19:54.53 | fi-dschi | I need some help as I don't succeed in compiling brlcad for arch linux i686 |
| 19:56.01 | fi-dschi | You may look at the last 263 lines of output at http://pastebin.com/m55261921 |
| 19:57.16 | fi-dschi | It seems to be a problem connected to asc2g |
| 19:57.49 | fi-dschi | My CFLAGS and CXXFLAGS are -march=athlon-xp -O2 -pipe |
| 20:48.21 | brlcad | fi-dschi: hello |
| 20:48.49 | fi-dschi | hello |
| 20:49.09 | brlcad | fi-dschi: I can't get to pastebin.com from current location, could you repost to pastebin.bzflag.bz ? |
| 20:49.16 | fi-dschi | ok |
| 20:50.43 | fi-dschi | brlcad:http://pastebin.bzflag.bz/m42158c73 |
| 20:51.27 | prasad_ | who has vmware expreience |
| 20:51.34 | brlcad | fi-dschi: well the good news is that the build completed |
| 20:51.57 | brlcad | ~translate de en Ungültiger Maschinenbefehl |
| 20:52.26 | fi-dschi | in en it is invalid machine code |
| 20:52.44 | prasad_ | ~translate jp en konichiwa |
| 20:52.49 | fi-dschi | hello |
| 20:52.51 | prasad_ | boo |
| 20:53.02 | prasad_ | ~translate en jp hello |
| 20:53.10 | prasad_ | ah nice :) |
| 20:53.55 | brlcad | fi-dschi: if you run "src/conv/asc2g", what does it report? |
| 20:54.00 | brlcad | same error? |
| 20:54.03 | brlcad | or a usage statement? |
| 20:54.52 | fi-dschi | tells me how to use |
| 20:55.02 | fi-dschi | Usage: asc2g file.asc file.g .... |
| 20:55.20 | brlcad | try: cd db && ../src/conv/asc2g operators.asc operators.g |
| 20:55.38 | fi-dschi | from inside src/conv? |
| 20:55.44 | fi-dschi | ah, ok |
| 20:56.02 | brlcad | no, from inside db/ |
| 20:56.09 | fi-dschi | this gives me "invalid machine instruction" |
| 20:56.12 | brlcad | rather .. from the top level |
| 20:56.13 | brlcad | hm |
| 20:56.43 | brlcad | then from db/ try this: ../libtool --mode=execute gdb --args ../src/conv/asc2g operators.asc operators.g |
| 20:57.39 | fi-dschi | reports that there is no lobtool in ../ |
| 20:58.16 | brlcad | hopefully no libtool? |
| 20:58.24 | fi-dschi | yes, of course |
| 20:58.24 | brlcad | not lobtool :) |
| 20:58.27 | brlcad | ok |
| 20:58.53 | brlcad | that's odd .. are you building out of dir? |
| 20:59.16 | fi-dschi | no, this is done by some tools |
| 20:59.40 | brlcad | find /home/philipp/Software/abs/brlcad/src/brlcad-7.12.0 -name libtool |
| 20:59.40 | fi-dschi | but the files should be from the src archive |
| 21:00.08 | fi-dschi | ok, it is therr |
| 21:00.09 | brlcad | hello andrecastelo :) |
| 21:00.15 | fi-dschi | this was my mistake |
| 21:00.18 | brlcad | then ../libtool should have worked.. ;) |
| 21:00.22 | fi-dschi | no |
| 21:00.41 | fi-dschi | ah ok |
| 21:00.42 | brlcad | "then from db/ try this" |
| 21:00.49 | fi-dschi | i didn't understand something else |
| 21:00.52 | brlcad | k |
| 21:00.55 | fi-dschi | ../libtool: line 6566: exec: gdb: not found |
| 21:01.02 | brlcad | eep |
| 21:01.03 | fi-dschi | that's the output |
| 21:01.05 | brlcad | no gdb installed? |
| 21:01.24 | fi-dschi | thanks |
| 21:01.35 | fi-dschi | this is missing as a deendency |
| 21:01.55 | fi-dschi | ok, so I wil try again with gdb, thank you very much |
| 21:02.10 | fi-dschi | this will need some time |
| 21:02.53 | brlcad | what platform type is this? |
| 21:03.03 | fi-dschi | platform type? |
| 21:03.06 | fi-dschi | i686? |
| 21:03.06 | brlcad | hardware |
| 21:03.12 | fi-dschi | athlon xp |
| 21:04.42 | brlcad | hmm, an easy/quick test would be to take off the -march=athlon-xp option and see if the instruction error goes away |
| 21:05.01 | fi-dschi | does it compile faster then? |
| 21:05.12 | prasad_ | heh |
| 21:05.25 | brlcad | if you want a fast test compile, use the default options |
| 21:05.38 | brlcad | maybe just ./configure --enable-all |
| 21:05.43 | brlcad | no flags |
| 21:06.25 | fi-dschi | tis is all done by scripts, I normally type "makepkg" which will download the sources, compile and create a package for the pkgmanager of archlinux |
| 21:06.41 | fi-dschi | but I will try manually |
| 21:11.23 | brlcad | ahh, the archlinux guy hangs out here |
| 21:11.38 | fi-dschi | yes, thats me |
| 21:11.44 | brlcad | oh heh |
| 21:11.58 | fi-dschi | did you read the package comments? |
| 21:12.25 | brlcad | are you louipc ? |
| 21:12.31 | fi-dschi | no, this is the maintainer |
| 21:12.34 | fi-dschi | i am fi-dschi |
| 21:27.34 | brlcad | okay, fi-dschi -- sorry about that :) |
| 21:27.57 | brlcad | from our perspective, we've only dealt with lou to date |
| 21:28.13 | brlcad | he commits the archlinux changes directly to our repository misc/archlinux in your sources |
| 21:31.55 | fi-dschi | he told me to ask for help in this channel |
| 21:39.05 | brlcad | i highly suspect the problem has to do with the compilation settings and math computations, but it's really hard to say without seeing a stack trace |
| 21:39.15 | brlcad | that error just usually indicates bad compiler ju ju |
| 21:40.15 | fi-dschi | is there still a problem? I thought there was simply gdb missing? |
| 21:40.48 | brlcad | hm? |
| 21:41.34 | brlcad | gdb was just to try to figure out why it was reporting a bad machine instruction |
| 21:42.28 | fi-dschi | oh, so it should now give a more detailed error message? |
| 21:45.31 | brlcad | right |
| 21:45.36 | brlcad | to see if it was code or compiler |
| 21:45.39 | brlcad | or both |
| 21:46.01 | fi-dschi | ok, compilation has just finished |
| 21:46.10 | fi-dschi | and there are no errors as far as i can see |
| 21:46.35 | brlcad | ah, cool |
| 21:46.54 | brlcad | that would 'likely' be a compiler bug for that -march |
| 21:47.09 | brlcad | or something being provoked with non-ieee floats or something similar |
| 21:47.29 | fi-dschi | ok, i will try again with my cflags or what else can i doà |
| 21:47.33 | fi-dschi | ? |
| 21:47.53 | brlcad | try again with --enable-optimized later |
| 21:48.23 | brlcad | that turns on a lot more optimization flags, still hardware generic, but should do much better than O2 |
| 21:48.50 | *** join/#brlcad louipc (n=louipc@bas8-toronto63-1096669639.dsl.bell.ca) | |
| 21:48.53 | brlcad | have you seen the PKGBUILD file loui wrote up? |
| 21:48.56 | brlcad | speak of the devil |
| 21:48.58 | louipc | :D |
| 21:49.01 | fi-dschi | :D |
| 21:49.07 | fi-dschi | yes |
| 21:49.08 | louipc | yeah I was coming in for that exact purpose |
| 21:49.17 | brlcad | howdy louipc |
| 21:49.28 | fi-dschi | did you get the post? |
| 21:49.42 | louipc | yep |
| 21:51.13 | louipc | yeah I took out any special configure flags, only --prefix is left because I had a problem with asc2g recently and was going to debug. but it worked fine with the flags removed hehe |
| 21:52.55 | louipc | brlcad: greetings indeed |
| 21:54.18 | louipc | fi-dschi: maybe the ./configure output might give a clue... particularly the summary |
| 21:54.49 | fi-dschi | for the default ./configure? one moment |
| 21:55.16 | louipc | ooh nevermind you found the problem? |
| 21:55.36 | fi-dschi | me? no. i donÃt even understand what is going on! |
| 21:55.56 | brlcad | louipc: the error is a run-time machine code error .. which is usually compilation settings or floating point data |
| 21:55.57 | louipc | does it work if you have gdb installed? |
| 21:56.16 | fi-dschi | with ./configure --all-enabled, yes |
| 21:56.43 | louipc | brlcad: yeah I was thinking something along those lines |
| 21:56.49 | brlcad | it worked if he removed his custom CFLAGS, which includes the -march optimization flag for a particular generation of athlons |
| 21:57.20 | brlcad | I suspect it's just that -march flag causing the problem |
| 21:58.37 | louipc | ah |
| 21:58.39 | louipc | nods |
| 21:59.08 | brlcad | athlon-xp is either buggy or not the right architecture |
| 21:59.16 | louipc | fi-dschi: it should work with just generic x86_64 flags |
| 21:59.44 | fi-dschi | is this used by ./configure --all-enabled? |
| 21:59.46 | louipc | well another user told me it works :D |
| 21:59.56 | brlcad | --enable-optimized should do the trick if performance is what you're going for |
| 22:00.16 | brlcad | fi-dschi: run "uname -a" |
| 22:00.26 | fi-dschi | but athlon-xp should be the optimized flag for me? |
| 22:00.27 | louipc | fi-dschi: ./configure by itself |
| 22:00.43 | fi-dschi | Linux PHPC_ARCH 2.6.24-ARCH #1 SMP PREEMPT Sun Mar 30 11:40:06 CEST 2008 i686 AMD Athlon(tm) AuthenticAMD GNU/Linux |
| 22:00.55 | louipc | :D |
| 22:01.27 | fi-dschi | some problems in the past made me think that athlon-xp was wrong |
| 22:01.39 | louipc | ya |
| 22:01.46 | fi-dschi | but my cpu is called athlon-xp |
| 22:01.55 | fi-dschi | it is even printed on it |
| 22:02.16 | louipc | oh you're running i686 cool |
| 22:02.47 | brlcad | ./configure --enable-all --enable-optimized CFLAGS="-mtune=i686" |
| 22:02.52 | fi-dschi | this is the archlinux default isn't it? |
| 22:03.40 | louipc | archlinux default flags are CFLAGS="-march=i686 -mtune=generic -O2 -pipe" |
| 22:03.48 | louipc | if you go with that it should work |
| 22:03.50 | fi-dschi | ok, i see |
| 22:03.52 | brlcad | that doesn't necessarily mean that gcc thinks you have what it calls athlon-xp nor does it mean that the optimization loop in gcc isn't buggy ;) |
| 22:04.43 | fi-dschi | but shouldn't i be looking for the best fit? |
| 22:04.43 | brlcad | notes that --enable-optimized is more agressive than "-march=i686 -mtune=generic -O2 -pipe" |
| 22:04.49 | fi-dschi | athlon, athlon-xp, i686? |
| 22:05.14 | fi-dschi | it has a higher priority? |
| 22:05.20 | brlcad | fi-dschi: it really won't matter -- the difference is going to be minimal |
| 22:06.29 | fi-dschi | there are just so many sites telling you which cflags you should use :) |
| 22:06.40 | louipc | hehe |
| 22:08.06 | fi-dschi | another question to the above: what does "enable-optimized" do? does it e.g. use athlon-xp if this is the best fitting? |
| 22:08.44 | louipc | it takes twice as long ;D |
| 22:09.07 | brlcad | fi-dschi: in the summary at the end of configure, it tells ou the flags it'll use |
| 22:09.14 | brlcad | s/ou/you/ |
| 22:10.50 | brlcad | it basically turns on several optimization flags, almost as many as it can without modifying ieee behavior or becoming platform/hardware dependent |
| 22:10.50 | fi-dschi | so what are the settings in makepkg.conf for if the configure script can find the best flags on its own? |
| 22:11.31 | fi-dschi | does this create platform depedency? |
| 22:11.37 | louipc | it's for packages which don't have such cool configure options :D |
| 22:12.03 | brlcad | archlinux users may also just want to override the flags with their own, or make their build platform dependent (for that extra 2% performance) |
| 22:12.31 | louipc | that too, but we're generally not ricers |
| 22:12.32 | fi-dschi | ok, so let's set the flags back to default :) |
| 22:12.44 | louipc | fi-dschi: good idea |
| 22:13.49 | louipc | CXXFLAGS are the same as well |
| 22:14.39 | fi-dschi | yes, i know that |
| 22:19.04 | fi-dschi | i will try to recompile with makepkg tomorrow. If there are problems you will hear from me again :). Thank ou so far and good night. |
| 22:19.17 | fi-dschi | s/ou/you/ ;) |
| 22:19.23 | brlcad | cheers |
| 22:19.25 | louipc | alright take care |
| 22:45.49 | *** join/#brlcad spike_ (i=[U2FsdGV@centaur.acm.jhu.edu) | |
| 22:46.19 | spike_ | is running off to the gym, brb |
| 23:06.53 | brlcad | cya |