IRC log for #brlcad on 20120222

07:20.06*** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol)
08:15.47*** join/#brlcad Stattrav (~Stattrav@61.12.114.82)
08:15.47*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
08:42.42*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
09:35.15*** join/#brlcad Stattrav (~Stattrav@61.12.114.82)
09:35.16*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
11:37.23*** join/#brlcad kaushik_ (~kaushik@122.164.44.178)
11:38.31kaushik_if anyone can help me with reference to this email: http://sourceforge.net/mailarchive/forum.php?thread_name=CACb6h%2Bv7SrRHgQL8AiUpha_HSd%2BcA2m%3Dif%2BM5m4fU3Z0npfGBw%40mail.gmail.com&forum_name=brlcad-devel
11:38.38kaushik_please beep me!
11:38.56kaushik_NURBS fever!
12:08.04*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
16:47.27*** join/#brlcad Stattrav (~Stattrav@61.12.114.82)
16:47.33*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
18:55.56*** join/#brlcad Stattrav (~Stattrav@61.12.114.82)
18:55.57*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
19:30.34*** join/#brlcad kaushik_ (~kaushik@122.164.44.178)
19:40.09CIA-128BRL-CAD: 03starseeker * r49495 10/brlcad/trunk/ (CMakeLists.txt misc/CMake/BRLCAD_CheckFunctions.cmake): bite the bullet - we need to be able to specifiy additional directories in some cases when looking for headers, so add that ability to BRLCAD_INCLUDE_FILE.
19:41.25CIA-128BRL-CAD: 03starseeker * r49496 10/brlcad/branches/STABLE/ (CMakeLists.txt misc/CMake/BRLCAD_CheckFunctions.cmake): Same fix for stable logic - this may necessitate manual intervention for this file in a future merge.
19:55.39brlcadkaushik_: replied to both your e-mails
19:56.17brlcadyou'll want to hang out here in the channel if you want the quickest reqply
20:19.55*** join/#brlcad Stattrav (~Stattrav@223.231.214.136)
20:19.56*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
20:36.03CIA-128BRL-CAD: 03brlcad * r49497 10/brlcad/trunk/src/conv/ply-g.c: revert r49396 since that comparison is always true (and newer versions of gcc are able to figure that out). not a workable fix for tainted data.
20:38.23jordisayolbrlcad: nothing changed -mtune=generic
20:48.58brlcadjordisayol: okay, thanks
20:49.05brlcadso that's not likely the issue
20:49.21jordisayolyes
20:49.45brlcadwhen I first read that log, it looks more like a ibrary incompatibility
20:50.00jordisayolaha
20:50.51brlcadlike if you compiled with opengl and his driver sucks, it could end up in that state
20:51.09brlcadyet he compiles himself and opengl ends up disabled, so doesn't run into the problem
20:51.23brlcadone of many possible scenarios
20:51.31brlcadstarseeker: librtserver failing
20:51.32brlcadLinking C executable ../../bin/rtserverTest
20:51.32brlcad/usr/bin/ld: warning: librt.so.1, needed by /usr/lib/jvm/jre/lib/ppc64/libjawt.so, may conflict with librt.so.19
20:51.35brlcad/lib64/libgio-2.0.so.0: undefined reference to `deflateSetHeader@ZLIB_1.2.2'
20:51.38brlcad/lib64/libgio-2.0.so.0: undefined reference to `inflateGetHeader@ZLIB_1.2.2'
20:52.12brlcadapparently zlib linkage missing from rtserverTest
20:52.47jordisayolbrlcad: if i disable opengl, performance will decrease a lot?
20:53.46brlcadnope
20:53.54brlcadnot for 99% of users
20:54.19jordisayolso what do you think? do i do?
20:54.49brlcadI'd give that a try, easy enough to recompile and retest if only to eliminate or isolate opengl as being involved
20:56.53jordisayolok, i'll create a new deb with opengl disabled
20:57.28CIA-128BRL-CAD: 03brlcad * r49498 10/brlcad/trunk/TODO: ls-dyna importer would be cool.
20:57.56*** join/#brlcad Stattrav (~Stattrav@61.12.114.82)
20:57.56*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
20:58.28CIA-128BRL-CAD: 03brlcad * r49499 10/brlcad/trunk/TODO: it's specifically the 'keyword' file format
21:06.06brlcadstarseeker: so that's a tricky one -- it looks like one of the java libs is linked against libgio (the glib i/o library), which links against libz
21:06.36brlcadwhich is a zlib older than ours, incompatible
21:08.05brlcadthe cmake build needs someway to detect that so the java portions can be disabled
21:09.01brlcadsome recent change apparently caused librtserver to get enabled or linked differently, now halting the build
21:34.39CIA-128BRL-CAD: 03jordisayol * r49500 10/brlcad/trunk/sh/make_deb.sh: revert to erikgreenwald changes
21:49.35jordisayolbrlcad: I'ce built a deb disabling opengl. mged properly works but archer fails:
21:49.35jordisayol$ archer
21:49.35jordisayolERROR:  Requisite display manager is not available.
21:49.35jordisayolBRL-CAD may need to be recompiled with support for:  ogl
21:49.35jordisayolRun 'fbhelp' for a list of available display managers.
21:49.36jordisayolUnexpected error encountered while running Archer.
21:49.36jordisayolAborting.
22:38.07brlcadjordisayol: archer requires opengl
22:38.38brlcadthe point was to test the dependency to see if it's related, but seeing the recent post about libtk, I'm doubtful it has to do with opengl
22:39.13jordisayolok
22:39.14brlcadclearly something drasticly different in his build of tcl compared with yours
22:40.33brlcadsee if he can run "nm libtcl.so > objects.txt" on the libtcl he built then provide you the objects.txt file, you run the same command on the libtcl.so in your .deb .. then we can compare the differences between the two
22:40.53jordisayoljust for your info, I compile these deb's on unbuntu 10.04 LTS
22:41.00CIA-128BRL-CAD: 03bob1961 * r49501 10/brlcad/trunk/src/tclscripts/archer/ArcherCore.tcl: Make ArcherCore::updateTreeDrawLists public.
22:41.16brlcadthey shouldn't be that different
22:41.34brlcadunless they're being compiled in completely different/incompatible ways
22:45.00jordisayolnm /usr/brlcad/lib/libtcl.so
22:45.00jordisayolnm: /usr/brlcad/lib/libtcl.so: no symbols
23:05.46brlcadjordisayol: er, that doesn't sound right
23:05.53brlcadls -la /usr/brlcad/lib/libtcl*
23:08.13jordisayol$ ls -la /usr/brlcad/lib/libtcl*
23:08.13jordisayol-rw-r--r-- 1 root root  404954 2012-02-22 23:56 /usr/brlcad/lib/libtclcad.a
23:08.13jordisayollrwxrwxrwx 1 root root      15 2012-02-23 00:06 /usr/brlcad/lib/libtclcad.so -> libtclcad.so.19
23:08.13jordisayollrwxrwxrwx 1 root root      19 2012-02-23 00:06 /usr/brlcad/lib/libtclcad.so.19 -> libtclcad.so.19.0.1
23:08.13jordisayol-rw-r--r-- 1 root root  237808 2012-02-22 23:56 /usr/brlcad/lib/libtclcad.so.19.0.1
23:08.14jordisayollrwxrwxrwx 1 root root      13 2012-02-23 00:06 /usr/brlcad/lib/libtcl.so -> libtcl.so.8.5
23:08.14jordisayollrwxrwxrwx 1 root root      15 2012-02-23 00:06 /usr/brlcad/lib/libtcl.so.8.5 -> libtcl.so.8.5.9
23:08.14jordisayol-rw-r--r-- 1 root root 1297064 2012-02-22 23:56 /usr/brlcad/lib/libtcl.so.8.5.9
23:08.15jordisayol-rw-r--r-- 1 root root    3994 2012-02-22 23:56 /usr/brlcad/lib/libtclstub.a
23:18.47brlcadmaybe your nm doesn't follow symbolic links, try nm /usr/brlcad/lib/libtcl.so.8.5.9
23:20.24brlcadthen assuming that gives you a lot of stuff, try this:  nm /usr/brlcad/lib/libtcl.so.8.5.9 | sort | uniq | wc
23:20.42brlcadI get 3611 as the first number
23:21.03jordisayol$ nm /usr/brlcad/lib/libtcl.so.8.5.9
23:21.03jordisayolnm: /usr/brlcad/lib/libtcl.so.8.5.9: no symbols
23:21.17brlcader
23:21.45jordisayolit appears tht only show symbols from static libs *.a
23:22.11brlcadreading up on it now, looks like gnu nm wants a -D option
23:22.18jordisayolok
23:22.50brlcadalthrough your stub file is already suspect
23:23.06jordisayolit works now
23:24.12jordisayol$ nm -D /usr/brlcad/lib/libtcl.so.8.5.9 | sort | uniq | wc
23:24.12jordisayol<PROTECTED>
23:28.30*** join/#brlcad Yoshi47 (~jan@d72-39-60-53.home1.cgocable.net)
23:45.44*** join/#brlcad DarkCalff (DC@173.231.40.98)
23:53.12*** join/#brlcad cvds_ (~leila@e255180.upc-e.chello.nl)
23:54.52*** join/#brlcad ``Erik (~erik@BZ.BZFLAG.BZ)
23:55.56jordisayolbrlcad: I've to go. as soon as he send me symbols.txt, I'll send both to you
23:59.50*** join/#brlcad tofu (~Adium@c-68-34-98-23.hsd1.md.comcast.net)

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