| 02:06.57 | *** join/#brlcad Pimpi (~frank@p5481A5DB.dip0.t-ipconnect.de) |
| 04:27.13 | brlcad | wonder where that jano fella is too |
| 05:30.30 | EricWilhelm | I thought surely it would open when the time hit 1111111111 |
| 05:47.40 | *** join/#brlcad ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc) |
| 05:47.40 | *** topic/#brlcad is http://brlcad.org/ || BRL-CAD is now Open Source! || Screenshots: http://sourceforge.net/project/screenshots.php?group_id=105292 || http://brlcad.org/images/mged.jpg || Release 7.2.0 is posted (20050305) |
| 14:16.30 | CIA-8 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: fix up the header line to be consistent with the others, s/\t/ /g for the itemized section list, add a paragraph on the difference between using enable and with arguments |
| 16:21.48 | CIA-8 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: removed obsolete comment |
| 18:13.11 | *** join/#brlcad ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc) |
| 18:13.11 | *** topic/#brlcad is http://brlcad.org/ || BRL-CAD is now Open Source! || Screenshots: http://sourceforge.net/project/screenshots.php?group_id=105292 || http://brlcad.org/images/mged.jpg || Release 7.2.0 is posted (20050305) |
| 18:40.50 | CIA-8 | BRL-CAD: 03brlcad * 10brlcad/sh/footer.sh: add support for m4 files too -- m4 files don't seem to have an emacs m4-mode variable that controls indentation, so just make sure the standard indent is set. |
| 19:04.10 | CIA-8 | BRL-CAD: 03brlcad * 10brlcad/ (configure.ac m4/cache.m4): separate out the configure cache check from configure.ac to it's own m4 file, m4/cache.m4. takes one optional arg for the cache file name. call it BC_CONFIG_CACHE. |
| 19:12.54 | CIA-8 | BRL-CAD: 03brlcad * 10brlcad/m4/Makefile.am: add cache.m4 as one of the first brlcad m4 files |
| 20:20.52 | narnia | brlcad, any word on when part23 may arrive? |
| 21:03.00 | brlcad | not yet -- should be pretty soon |
| 21:13.18 | narnia | okay |
| 21:21.46 | narnia | i am coming to the conclusion that when reading large part21 files the problem is with the way entities and their inheritance works in 10303 and not with the hash table/symbol table/any other tables implementation. the x-refs require alot of memory and processing. it is better than opencascade but the 40mbyte yanmar diesel engine still is painful slow. |
| 21:22.24 | brlcad | have you profiled it? |
| 21:23.36 | narnia | brlcad, still working on it. the preliminary profiles show memory usage going up rather dramatically. processing then gets bogged down in the x-refs. |
| 21:27.09 | *** join/#brlcad thiemo ([HtyDkjUZ2@helibox.thiemo.net) |
| 21:27.13 | thiemo | hi |
| 21:27.23 | brlcad | howdy |
| 21:27.36 | thiemo | all fine here, thanks |
| 21:28.10 | thiemo | just a problem with -- guess -- brlcad |
| 21:28.18 | brlcad | what's up? |
| 21:29.16 | thiemo | starting mged (on FreeBSD running some X.org), all is fine when i start it from a shell (e.g. in an xterm) |
| 21:29.40 | brlcad | that's good |
| 21:29.47 | thiemo | when i put it into the menu of my window manager (fvwm), it doesn't start up, seems to say "backgrounding", but then terminates |
| 21:29.59 | narnia | the other area it bogs down is entity instance validation in the data section. part21 requires that each entity instance be validated. the yanmar diesel engine has 695998 individual entity instance. |
| 21:30.30 | brlcad | narnia: is that a model you can share? sounds interesting to test it here too |
| 21:30.55 | narnia | even with 1gbyte of ram that is swapping to disk. |
| 21:31.05 | narnia | brlcad, yes i can share it. |
| 21:31.14 | brlcad | thiemo: hrm.. it does detach and kick to the background on startup |
| 21:31.29 | narnia | i will put it on my web server and you can pull it from there. |
| 21:31.33 | thiemo | yup, seen that in the source... i cannot imagine what's going wrong there |
| 21:31.48 | thiemo | also did some ktrace (syscall tracing that is), but found no errors |
| 21:32.01 | brlcad | thiemo: perhaps fvwm is killing it because the parent died? |
| 21:32.09 | thiemo | hm. |
| 21:32.25 | thiemo | let's see... :) |
| 21:32.25 | brlcad | not something i've heard of, but i dont' think anyone's tried that yet |
| 21:33.36 | thiemo | classic_mged (starting with -c) should lead to no backgrounding, shouldn't it? |
| 21:34.00 | brlcad | right |
| 21:34.08 | brlcad | as should -n |
| 21:34.14 | thiemo | :) |
| 21:34.32 | thiemo | weird, it seems to ignore the options. |
| 21:34.33 | brlcad | if it does, try typing "gui" at the mged command prompt |
| 21:34.37 | brlcad | ahh |
| 21:35.21 | thiemo | starting with -c prints "Backgrounding " etcetera |
| 21:35.37 | thiemo | which should not be if i can trust my source code reading. |
| 21:35.54 | brlcad | right |
| 21:36.03 | brlcad | it's not getting passed the argument |
| 21:37.48 | thiemo | Yowsa! There is a /usr/local/bin/mged script which finally does "exec /usr/local/brlcad/bin/mged" |
| 21:38.07 | brlcad | huh?? |
| 21:38.22 | brlcad | ahh |
| 21:38.45 | brlcad | hehe |
| 21:38.56 | brlcad | hrm.. I didn't notice that they did that |
| 21:39.06 | thiemo | neither did i :-} |
| 21:40.02 | brlcad | there are several "merge" issues of adding a suite of 400 tools and a few dozen libraries .. |
| 21:40.18 | brlcad | especially what to do about naming collisions |
| 21:41.06 | thiemo | well, they seem to have put anything but that script into /usr/local/brlcad, and set an appropriate environment... |
| 21:41.35 | brlcad | yes, there was talk about using the single root (/usr/local/brlcad) |
| 21:41.42 | brlcad | just didn't know about the wrapper scripts |
| 21:42.03 | narnia | brlcad, wget http://204.238.34.206/3tnv70-asa.stp the file is 45081315 bytes. |
| 21:42.06 | brlcad | brl-cad really "prefers" to have a single root |
| 21:42.30 | brlcad | i've been working on massaging the organization and installation assumptions so that is no longer required |
| 21:42.53 | brlcad | that should help all of the packaging systems |
| 21:43.08 | narnia | brlcad, do you have a cad system that reads part21 files? |
| 21:43.21 | thiemo | indeed |
| 21:43.23 | brlcad | narnia: does pro/e and unigraphics count? |
| 21:43.35 | narnia | brlcad, that is cheating. ;-) |
| 21:44.06 | brlcad | :-) |
| 21:45.27 | narnia | brlcad, would you do me a favor? e-mail jano and ask him to come back and let me out of the cage of doom. ;-) |
| 21:45.35 | brlcad | thiemo: brl-cad was started around the same time and in a similar style as X11 was to give you an idea of the single-root preference |
| 21:45.50 | thiemo | :) |
| 21:46.17 | brlcad | narnia: I'll see if I can hunt him down, he'll return eventually regardless though :) |
| 21:48.02 | narnia | brlcad, when you have the time i would really like to know how long pro/e takes to read the yanmar diesel engine. |
| 21:48.57 | brlcad | narnia: pro/e is not known for it's speed or memory use |
| 21:48.58 | narnia | brlcad, pro/e sells for around $30k usd to start right? |
| 21:49.01 | brlcad | absolutely horrible actually |
| 21:49.09 | narnia | ummm. |
| 21:49.20 | brlcad | i'll give it a try and see how long it takes |
| 21:49.54 | brlcad | yeah, proe and unig both start around 10k and go up depending on options |
| 21:51.40 | narnia | brlcad, would you know what pro/e is written in? c or c++ or other? |
| 21:54.12 | brlcad | holly cow |
| 21:54.17 | brlcad | they finally released |
| 21:54.20 | brlcad | incrTcl |
| 21:55.41 | brlcad | 3.3-RC |
| 22:19.23 | thiemo | now i'm using a seperate wrapper script in the window manager that executes an xterm and within that starts mged -- that's working. |
| 22:19.30 | thiemo | not elegant, but functional :) |
| 22:19.56 | brlcad | I'll mention it to the fbsd folks if you don't, but I suspect the problem will fix itself shortly |
| 22:19.59 | thiemo | i just cannot find the real cause for mged terminating without terminal |
| 22:20.24 | thiemo | well, then i'll leave all that to you... |
| 22:20.45 | brlcad | that's true.. the real cause is still mysterious ;) |
| 22:20.46 | thiemo | i guess you have more clue if and when that will be really fixed |
| 22:21.12 | thiemo | whatever -- it's working now, so i can start exploring the program itself |
| 22:21.25 | thiemo | from a user, not debugger point of view :) |
| 22:21.37 | brlcad | :) |
| 22:21.44 | thiemo | (installed it just a few days ago) |
| 22:22.26 | thiemo | so, many thanks! |
| 22:22.52 | brlcad | no problem |
| 22:23.03 | brlcad | feel free to drop back in if you have any questions |
| 22:23.15 | brlcad | or just hang too |
| 22:24.00 | thiemo | i will |
| 22:24.29 | thiemo | though i suppose this channel's timezone is US oriented |
| 22:24.37 | thiemo | (and i'm in europe) |
| 22:25.04 | thiemo | (i've never seen the .pdpc TLD before, though :) |
| 22:25.44 | brlcad | it's a cloak |
| 22:26.10 | brlcad | i.e. my domain/ip is hidden because I'm registered |
| 22:26.35 | brlcad | no worries about the timezone, we overlap enough -- I work with lots of folks around the world ;) |
| 22:26.50 | thiemo | ah, i see |
| 22:26.58 | thiemo | well then - good night for now :) |
| 22:27.07 | brlcad | there's only a few hours of the day that I'm not listening, and even when not -- it's logged and I'll respond ;) |
| 22:27.10 | brlcad | have a good one |
| 23:25.09 | CIA-8 | BRL-CAD: 03brlcad * 10brlcad/ (configure.ac m4/restore.m4): |
| 23:25.09 | CIA-8 | BRL-CAD: separate out the tests for a clobbered INSTALL/COPYING file into their own |
| 23:25.09 | CIA-8 | BRL-CAD: m4/restore.m4 using a BC_RESTORE_CLOBBERED macro function. it takes two |
| 23:25.09 | CIA-8 | BRL-CAD: optional arguments for the files to check and the directory off of the srcdir to |
| 23:25.09 | CIA-8 | BRL-CAD: find them in. if the files have a revision id, those take precedence over a |
| 23:25.09 | CIA-8 | BRL-CAD: content difference in the backup. |
| 23:25.51 | CIA-8 | BRL-CAD: 03brlcad * 10brlcad/m4/Makefile.am: add restore.m4 |
| 23:38.44 | CIA-8 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: missing var init and output a label of output instead of Makefiles |
