| 01:06.47 | *** join/#brlcad merzo (~merzo@68-104-133-95.pool.ukrtel.net) | |
| 03:58.26 | *** join/#brlcad milamber (~devlin@d118-75-244-176.try.wideopenwest.com) | |
| 04:40.49 | Notify | 03BRL-CAD:brlcad * 54929 brlcad/trunk/src/other/boost/CMakeLists.txt: smart_ptr doesn't actually have a CMakeLists.txt file to describe the build |
| 04:42.43 | *** join/#brlcad merzo (~merzo@103-97-133-95.pool.ukrtel.net) | |
| 04:58.01 | *** join/#brlcad hsrai (~hsrai@202.164.53.116) | |
| 05:16.45 | *** join/#brlcad hsrai (~hsrai@202.164.53.116) | |
| 05:20.30 | Notify | 03BRL-CAD:brlcad * 54930 brlcad/trunk/src/other/boost/libs/thread/CMakeLists.txt: the library name is supposed to be boost_thread |
| 05:22.10 | Notify | 03BRL-CAD:brlcad * 54931 brlcad/trunk/src/other/CMakeLists.txt: build the boost_thread compiled library (although this is probably temporary) |
| 05:23.22 | Notify | 03BRL-CAD:brlcad * 54932 brlcad/trunk/src/libpc/CMakeLists.txt: don't turn off all warnings, just make them not be errors |
| 05:28.01 | Notify | 03BRL-CAD:brlcad * 54933 brlcad/trunk/include/bu.h: declare the new bu_parallel_id() function for returning the current bu_parallel's cpu number (1 to ncpu) |
| 05:32.49 | Notify | 03BRL-CAD:brlcad * 54934 brlcad/trunk/src/libbu/parallel.h: provide an initial implementation of two functions to get/set the current cpu number in thread-local-storage (TLS). this preliminarily uses boost threads, but will probably change to minimize the complexity where we have support for intrinsic compiler TLS. |
| 05:33.16 | Notify | 03BRL-CAD:brlcad * 54935 brlcad/trunk/src/libbu/thread.cpp: fix header, thread.cpp |
| 05:36.11 | Notify | 03BRL-CAD:brlcad * 54936 brlcad/trunk/src/libbu/CMakeLists.txt: we're using it now, so compile thread.cpp and link against boost_thread |
| 05:39.17 | Notify | 03BRL-CAD:brlcad * 54937 brlcad/trunk/src/libbu/heap.c: we now have a bu_parallel_id() that works, shift from 1-to-ncpu to 0-to-ncpu-1 |
| 05:44.45 | Notify | 03BRL-CAD:brlcad * 54938 (brlcad/trunk/src/libbu/ispar.c =================================================================== and 81 others): remove the no longer compiled ispar.c, merged into parallel.c so we could eliminate a global |
| 05:53.52 | *** join/#brlcad hsrai (~hsrai@202.164.53.116) | |
| 07:55.38 | *** join/#brlcad luca79 (~luca@net-188-216-230-48.cust.dsl.vodafone.it) | |
| 11:24.05 | *** join/#brlcad ``Erik_ (~erik@pool-74-103-121-45.bltmmd.fios.verizon.net) | |
| 12:00.32 | *** join/#brlcad milamber (~devlin@d118-75-244-176.try.wideopenwest.com) | |
| 12:06.44 | *** join/#brlcad luca79 (~luca@net-188-216-230-48.cust.dsl.vodafone.it) | |
| 12:40.07 | *** join/#brlcad witness (uid10044@gateway/web/irccloud.com/x-kpvazlaidedjpimh) | |
| 13:54.06 | *** join/#brlcad brlcad (~sean@66-118-151-70.static.sagonet.net) | |
| 13:56.29 | ``Erik | netmask looks good now, what was the critical issue? |
| 13:56.49 | *** join/#brlcad starseeker (~starseeke@66-118-151-70.static.sagonet.net) | |
| 13:57.29 | *** join/#brlcad n_reed (~molto_cre@66-118-151-70.static.sagonet.net) | |
| 13:58.31 | *** join/#brlcad Notify (~notify@66-118-151-70.static.sagonet.net) | |
| 14:45.26 | Notify | 03BRL-CAD:bob1961 * 54939 brlcad/trunk/src/tclscripts/archer/ArcherCore.tcl: Added ArcherCore::freezeGUI. |
| 15:39.43 | *** join/#brlcad luca79 (~luca@net-188-216-230-48.cust.dsl.vodafone.it) | |
| 15:48.04 | Notify | 03BRL-CAD:bob1961 * 54940 brlcad/trunk/src/tclscripts/archer/ArcherCore.tcl: This fixes an issue where geometry doesn't show up in the newly visible display windows that are exposed when going to multi-pane mode AND display-lists are enabled. |
| 17:07.35 | starseeker | brlcad: do you want to go ahead and update to libpng 1.6.1? |
| 17:14.32 | brlcad | starseeker: I don't really want to do anything that doesn't get us closer to wrapping things up for release |
| 17:14.44 | brlcad | so unless that fixes autoconf, not really |
| 17:15.20 | brlcad | i'm busy trying to wrap up my heap work myself, then I have 350 commits to get through, all hopefully before this weekend |
| 17:34.19 | Notify | 03BRL-CAD:brlcad * 54941 brlcad/trunk/src/libbu/parallel.c: hook up to the thread local storage so bu_parallel_id() can track our thread number, fix record keeping that keeps track of how many started/finished |
| 17:34.32 | Notify | 03BRL-CAD:bob1961 * 54942 brlcad/trunk/src/libged/mater.c: Get rid of "" and [] from the mater commands prompt array. This breaks things when sending back to Tcl land. |
| 17:37.55 | Notify | 03BRL-CAD:brlcad * 54943 brlcad/trunk/src/libbu/parallel.c: hindsight consideration, don't pretend we'd ever be able to talk to 'thread' 0 since it's going to be sitting there waiting for the invoked threads to finish. kinda like execl. so number them 0 to ncpu-1 |
| 17:45.32 | Notify | 03BRL-CAD:brlcad * 54944 brlcad/trunk/src/libpc/CMakeLists.txt: since we can't seem to do a damn thing about the inline warnings, turn them off outright |
| 17:49.21 | Notify | 03BRL-CAD:brlcad * 54945 (brlcad/trunk/src/libbu/CMakeLists.txt brlcad/trunk/src/libbu/thread.cpp): upon closer inspection, it looks like boost just did exactly what I had originally planned to do anyways, namely calling the pthread and win32 TLS api, so strip out the boost hooks. it's not worth having the additional compilation complexity to get boost_thread compiling cleanly and the major third-party dep it would |
| 17:49.23 | Notify | be adding to libbu. we can just do our own thing here in far far fewer lines of code, far less integration complexity. |
| 17:50.14 | Notify | 03BRL-CAD:brlcad * 54946 brlcad/trunk/src/other/CMakeLists.txt: revert back to ignoring the boost subdir |
| 17:51.19 | brlcad | named[1149]: creating IPv4 interface em0 failed; interface ignored |
| 17:51.43 | brlcad | wonders why named is trying to create ethernet interfaces |
| 18:04.34 | Notify | 03BRL-CAD:brlcad * 54947 brlcad/trunk/misc/CMake/BRLCAD_CompilerFlags.cmake: there are far too many unquellable false-positive inline warnings with more recent versions of gcc (4.7ish) where it warns about not being able to inline default constructors and desctructors that it created and that we cannot quell with the given inline hooks. so turn them all off (for C++) for now |
| 19:33.50 | *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol) | |
| 19:46.13 | Notify | 03BRL-CAD:carlmoore * 54948 brlcad/trunk/src/anim/chan_permute.c: modify messages for can't open |
| 19:49.10 | Notify | 03BRL-CAD:carlmoore * 54949 brlcad/trunk/src/anim/chan_permute.c: oops, also put in the correct command name in these error messages |
| 19:51.27 | *** join/#brlcad witness (uid10044@gateway/web/irccloud.com/x-fwwohzaaroynixtd) | |
| 20:00.00 | Notify | 03BRL-CAD:carlmoore * 54950 brlcad/trunk/src/mged/cad_boundp.c: provide Usage based on the man page, but the program is allowed to continue running |
| 20:20.59 | Notify | 03BRL-CAD:carlmoore * 54951 brlcad/trunk/src/librt/primitives/bot/tie_kdtree.c: fix warning about format designator |
| 20:38.58 | ``Erik | yuh oh, word on the street is that a serious patch to postgresql is going to be released next week with some kinda security aspect |
| 20:45.16 | ``Erik | hm, is http://trepan.bzflag.bz/bzdistort.jpg displacement mapping, or real geometry? O.o |
| 21:34.51 | ``Erik | http://www.postgresql.org/message-id/14040.1364490185@sss.pgh.pa.us |
