IRC log for #brlcad on 20110713

00:28.57 *** join/#brlcad DarkCalf (DC@173.231.40.98)
00:50.42 CIA-62 BRL-CAD: 03kunigami * r45445 10/brlcad/trunk/src/liboptical/ (liboslrend.h sh_osl.cpp):
00:50.42 CIA-62 BRL-CAD: trying to allow multi-threading. I surrounded OSLRender access with semaphores
00:50.42 CIA-62 BRL-CAD: lock, but with 2 processors it is still crashing. I first tried using
00:50.42 CIA-62 BRL-CAD: RT_SEM_LAST, but I get a run-time error because the number of semaphores
00:50.42 CIA-62 BRL-CAD: exceeded 12, so I added a random identifier 8
01:59.46 *** join/#brlcad ibot (~ibot@rikers.org)
01:59.46 *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || #brlcad logs: http://ibot.rikers.org/%23brlcad/ || BRL-CAD release 7.20.2 being tagged today (20110701) || BRL-CAD has applied to participate in the ESA Summer of Code in Space!
03:06.12 starseeker ``Erik: do you know a good desktop PC maker these days?
03:06.43 starseeker may have to look into a new system, and isn't sure he can afford the time to collect components and do it "right"
03:09.28 *** join/#brlcad epileg (~epileg@unaffiliated/epileg)
03:28.56 *** join/#brlcad epileg (~epileg@unaffiliated/epileg)
03:42.53 CIA-62 BRL-CAD: 03bhinesley * r45446 10/brlcad/trunk/src/libged/alter.c: update argument flags, add several argument node helper functions
05:39.30 *** join/#brlcad alex_joni (~alex_joni@emc/board-of-directors/alexjoni)
07:30.00 *** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ)
07:34.45 *** join/#brlcad Stattrav (~Stattrav@117.202.18.108)
07:34.45 *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
08:25.14 *** join/#brlcad mafm (~mafm@90.Red-83-42-153.dynamicIP.rima-tde.net)
09:14.56 *** join/#brlcad Stattrav (~Stattrav@122.178.195.64)
09:14.56 *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
09:54.15 *** join/#brlcad epileg (~epileg@unaffiliated/epileg)
12:08.38 ``Erik starseeker: been starting to research that myself, think I'm about due for 2 new machines
13:21.29 kunigami_ Which tool do you use to debug multi-thread applications?
13:30.31 brlcad kunigami_: gdb can be used to debug multithreaded apps, you can switch threads and processes while debugging, set separate breakpoints, etc
13:30.45 *** join/#brlcad Stattrav (~Stattrav@122.178.195.64)
13:30.45 *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
13:37.02 CIA-62 BRL-CAD: 03Sean 07http://brlcad.org * r3007 10/wiki/Celestial_mechanics_particle_system:
13:39.05 *** join/#brlcad abhi2011 (~chatzilla@ip170-79-211-87.adsl2.static.versatel.nl)
13:42.15 *** join/#brlcad mafm (~mafm@90.Red-83-42-153.dynamicIP.rima-tde.net)
13:47.34 brlcad yawns
14:42.00 *** join/#brlcad epileg (~epileg@unaffiliated/epileg)
17:25.06 *** join/#brlcad merzo (~merzo@252-7-132-95.pool.ukrtel.net)
18:20.44 *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
18:24.36 *** join/#brlcad dli (~dli@66.49.188.111)
18:49.40 CIA-62 BRL-CAD: 03erikgreenwald * r45447 10/brlcad/trunk/src/proc-db/ringworld.c: add some body to make this do something (shows render issues at large scale)
18:52.34 CIA-62 BRL-CAD: 03erikgreenwald * r45448 10/brlcad/trunk/NEWS: list ringworld proc-db
18:53.48 CIA-62 BRL-CAD: 03brlcad * r45449 10/brlcad/trunk/include/bu.h: add a BU_LIST_INIT_MAGIC() macro so we can init and set a magic as a one-liner
18:59.13 CIA-62 BRL-CAD: 03brlcad * r45450 10/brlcad/trunk/include/bu.h: use the new BU_LIST_INIT_MAGIC() macro, reduce
19:01.02 CIA-62 BRL-CAD: 03brlcad * r45451 10/brlcad/trunk/include/ged.h: initial stubs for supporting a ged command container within the ged struct. a ged_cmd object describes a single named command providing the name, a brief description, the name of its manual page, and callback hooks.
19:01.37 CIA-62 BRL-CAD: 03brlcad * r45452 10/brlcad/trunk/include/magic.h: define a magic number for ged_cmd objects. using 'exec' since that's the heart of a ged command.
19:08.34 CIA-62 BRL-CAD: 03brlcad * r45453 10/brlcad/trunk/include/ (bu.h magic.h):
19:08.34 CIA-62 BRL-CAD: fix a FIXME, define a magic number for bu_observer objects. if there was bad
19:08.34 CIA-62 BRL-CAD: code that relied on zero-initialization (e.g., via calloc), that will no longer
19:08.34 CIA-62 BRL-CAD: result in proper initialization and will fail until fixed with a
19:08.34 CIA-62 BRL-CAD: BU_OBSERVER_INIT OR BU_OBSERVER_INIT_ZERO call.
19:16.53 CIA-62 BRL-CAD: 03brlcad * r45454 10/brlcad/trunk/src/liboptical/sh_prj.c: set the magic as soon as we begin as part of initialization.
19:18.13 CIA-62 BRL-CAD: 03brlcad * r45455 10/brlcad/trunk/src/ (5 files in 3 dirs): use the new BU_LIST_INIT_MAGIC() macro, reducing a few lines
19:25.40 CIA-62 BRL-CAD: 03brlcad * r45456 10/brlcad/trunk/src/libged/zoom/zoom.c: work in progress. define the zoom command including load/unload functions that add/remove it from a gedp.
19:29.17 bhinesley grabs popcorn
19:32.01 kunigami_ Even adding semaphores around OSLRender sh_osl crashes for two processors. I tried increasing the size of the stack at bu_parallel. without sucess
19:33.18 kunigami_ The problem only occurs when I do recursive calls (through rt_shootray)
19:34.07 kunigami_ I took a look at the implementation of rr_render and it does not seem to take any extra care before calling rt_shootray
19:57.30 CIA-62 BRL-CAD: 03brlcad * r45457 10/brlcad/trunk/include/magic.h: expand all of the magic numbers with what they represent in ascii, using '?' for non-printable characters.
20:13.37 CIA-62 BRL-CAD: 03starseeker * r45458 10/brlcad/trunk/CMakeLists.txt:
20:13.37 CIA-62 BRL-CAD: Start trying to get the point where we can do 64 bit builds on 32 bit systems
20:13.37 CIA-62 BRL-CAD: and vice versa with the flick of an option. Start with a more verbose warning
20:13.37 CIA-62 BRL-CAD: and option correction for MSVC, where one must pick one generator or the other
20:13.37 CIA-62 BRL-CAD: at CMake configure time.
21:25.10 CIA-62 BRL-CAD: 03starseeker * r45459 10/brlcad/trunk/src/libfft/CMakeLists.txt: Whoops - adding --no-undefined to linker flags caught libfft needing M_LIBRARY
21:26.08 CIA-62 BRL-CAD: 03starseeker * r45460 10/brlcad/trunk/CMakeLists.txt: Add --no-undefined to linker, fix printing of linker settings.
21:27.10 CIA-62 BRL-CAD: 03starseeker * r45461 10/brlcad/trunk/misc/CMake/CompilerFlags.cmake: Untested, but add in some flags intended to help handle the 'build 32 on 64bit' situation.
21:31.13 CIA-62 BRL-CAD: 03starseeker * r45462 10/brlcad/trunk/misc/CMake/CompilerFlags.cmake: Add linker settings for 32 as well. Ideally these should be tests to see if the linker can handle the flags, but not sure how to set that up yet.
21:31.37 CIA-62 BRL-CAD: 03brlcad * r45463 10/brlcad/trunk/include/nurb.h: don't conditionally include the headers we need. they already have built-in inclusion protection.
21:34.51 CIA-62 BRL-CAD: 03brlcad * r45464 10/brlcad/trunk/ (3 files in 3 dirs): replace RT_CNURB_MAGIC and RT_SNURB_MAGIC with the corresponding NMG magic defines that were identical values.
22:40.14 CIA-62 BRL-CAD: 03bhinesley * r45465 10/brlcad/trunk/src/libged/alter.c: rename several silly struct/union/members, more helper function work
22:46.33 kunigami_ Running with GDB, I always get one of these two errors: http://paste.ubuntu.com/643588/ (the cause seems to be invalid memory access)
22:52.43 CIA-62 BRL-CAD: 03bhinesley * r45466 10/brlcad/trunk/src/libged/ (alter.c edit.c): renaming alter to edit... let's see if I can rename the file first
23:01.07 CIA-62 BRL-CAD: 03bhinesley * r45467 10/brlcad/trunk/ (6 files in 4 dirs): other instances of alter that needed to be renamed to edit, and add edit command to archer
23:16.44 *** join/#brlcad mafm (~mafm@90.Red-83-42-153.dynamicIP.rima-tde.net)
23:28.38 CIA-62 BRL-CAD: 03bhinesley * r45468 10/brlcad/trunk/src/mged/setup.c: rename alter -> edit for mged
23:29.42 CIA-62 BRL-CAD: 03bhinesley * r45469 10/brlcad/trunk/src/libged/edit.c: make most functions HIDDEN, at least for now

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