| 02:00.53 | *** join/#brlcad tofu1 (~Adium@c-68-34-98-23.hsd1.md.comcast.net) |
| 03:57.16 | CIA-128 | BRL-CAD: 03starseeker * r49528 10/brlcad/trunk/ (6 files in 6 dirs): Take a different approach to per-target defines management - use global properties instead of cached values. This should have better behavior when configure options are changed. |
| 04:48.32 | CIA-128 | BRL-CAD: 03brlcad * r49529 10/brlcad/trunk/misc/CMake/CheckCInline.cmake: (log message trimmed) |
| 04:48.32 | CIA-128 | BRL-CAD: add an initial implementation of a new CHECK_C_INLINE() macro for detecting |
| 04:48.32 | CIA-128 | BRL-CAD: legacy forms of specifying inlining. this impl is loosely based off a snippet |
| 04:48.32 | CIA-128 | BRL-CAD: from Jack Kelly on the cmake mailing list except it avoids writing out a source |
| 04:48.32 | CIA-128 | BRL-CAD: file and only defines inline to something else if we need to. alas, this |
| 04:48.33 | CIA-128 | BRL-CAD: implementation does cop-out and sets a global definition instead of letting the |
| 04:48.33 | CIA-128 | BRL-CAD: caller decide by passing in a variable, but it should work well enough for those |
| 04:49.04 | CIA-128 | BRL-CAD: 03brlcad * r49530 10/brlcad/trunk/CMakeLists.txt: call the new CHECK_C_INLINE macro for improved portability |
| 05:36.58 | CIA-128 | BRL-CAD: 03brlcad * r49531 10/brlcad/trunk/src/other/lemon/lempar.c: apparently yyRuleName is only used during debug compilation according to icc |
| 05:41.28 | CIA-128 | BRL-CAD: 03brlcad * r49532 10/brlcad/trunk/src/util/bwmod.c: make sure the index doesn't exceep the mapbuf buffer |
| 05:47.15 | CIA-128 | BRL-CAD: 03brlcad * r49533 10/brlcad/trunk/NEWS: slew of code fixes and slight build system tweaks in support of the intel compiler, significant enough to note since it's fundamental portability infrastructure. |
| 05:49.30 | CIA-128 | BRL-CAD: 03brlcad * r49534 10/brlcad/trunk/NEWS: Apply alphabetical sorting to the toplevel objects list in rtwizard. (edit due to inconsistent case) |
| 05:51.57 | CIA-128 | BRL-CAD: 03brlcad * r49535 10/brlcad/trunk/NEWS: |
| 05:51.57 | CIA-128 | BRL-CAD: Make sure the attributes and comb are always on the same page, and that we don't |
| 05:51.57 | CIA-128 | BRL-CAD: miss color if someone works with rgb. (expanding name so it'll be easier to |
| 05:51.57 | CIA-128 | BRL-CAD: tabulate and group down the road, need to fix others after next PRB) |
| 05:54.43 | CIA-128 | BRL-CAD: 03brlcad * r49536 10/brlcad/trunk/NEWS: |
| 05:54.44 | CIA-128 | BRL-CAD: Bob fixed a bug in Archer's tree view - the tree view wasn't highlighting the |
| 05:54.44 | CIA-128 | BRL-CAD: correct objects when the option was set to highlight objects that would be |
| 05:54.44 | CIA-128 | BRL-CAD: impacted by editing the currently selected object. (reword to expand name) |
| 06:00.47 | CIA-128 | BRL-CAD: 03brlcad * r49537 10/brlcad/trunk/NEWS: |
| 06:00.47 | CIA-128 | BRL-CAD: reword for prb tracking since the commit messages are extracted. tom browder |
| 06:00.47 | CIA-128 | BRL-CAD: fixed a bug in mged where the relatively new BotEditor GUI was no longer |
| 06:00.47 | CIA-128 | BRL-CAD: functional. breakage was reported in sf bug report 3392650, fixed in r48073 by |
| 06:00.47 | CIA-128 | BRL-CAD: tom. gui now displays correctly. |
| 06:04.11 | CIA-128 | BRL-CAD: 03brlcad * r49538 10/brlcad/trunk/NEWS: reword for prb. tom browder fixed a bug affecting the regions and solids commands where they were dumping core if run due to bogus type coercion. reported and fixed by tom in sf bug report 3392558, fixed in r48037. |
| 08:08.32 | *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol) |
| 09:51.50 | *** join/#brlcad merzo (~merzo@193.254.217.44) |
| 10:07.53 | *** join/#brlcad kaushik_ (~kaushik@122.164.129.124) |
| 10:08.27 | *** join/#brlcad merzo (~merzo@193.254.217.44) |
| 10:08.59 | kaushik_ | brlcad: Replied back! I'm looking through the NURBS code now... Anything specific I should be looking for when talking about the Implicit-Nurbs conversion project? |
| 10:46.49 | *** join/#brlcad kaushik_ (~kaushik@122.164.129.124) |
| 10:51.16 | *** join/#brlcad merzo_ (~merzo@193.254.217.44) |
| 12:17.11 | *** join/#brlcad kaushik_ (~kaushik@122.164.129.124) |
| 12:28.48 | *** join/#brlcad kaushik__ (~kaushik@122.164.212.105) |
| 13:46.44 | *** join/#brlcad kaushik__ (~kaushik@122.164.212.105) |
| 14:00.52 | *** join/#brlcad tofu (~Adium@c-68-34-98-23.hsd1.md.comcast.net) |
| 14:00.52 | *** mode/#brlcad [+o tofu] by ChanServ |
| 15:24.30 | *** join/#brlcad jarray52 (~solargrin@unaffiliated/jarray52) |
| 18:08.07 | CIA-128 | BRL-CAD: 03starseeker * r49539 10/brlcad/trunk/src/other/lemon/lempar.c: Take a different approach to the issue seen with r49253 |
| 19:39.14 | *** join/#brlcad Stattrav (~Stattrav@61.12.114.82) |
| 19:39.15 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 19:47.31 | *** join/#brlcad Stattrav (~Stattrav@61.12.114.82) |
| 19:47.35 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 19:53.29 | CIA-128 | BRL-CAD: 03brlcad * r49540 10/brlcad/trunk/misc/CMake/CompilerFlags.cmake: |
| 19:53.29 | CIA-128 | BRL-CAD: GAH, revert the msse3 addition from r46520 as it invariably caused days of |
| 19:53.29 | CIA-128 | BRL-CAD: mysterious debugging for the exact same reason it was disabled in the first |
| 19:53.29 | CIA-128 | BRL-CAD: place. you can't just turn on sse for binaries that are distributed since |
| 19:53.29 | CIA-128 | BRL-CAD: you'll end up with illegal instructions (crashes) at runtime. debugged issue |
| 19:53.29 | CIA-128 | BRL-CAD: encountered on debian with jordi sayol and user notius, who was running .deb |
| 19:53.30 | CIA-128 | BRL-CAD: installs on a Pentium M system. |
| 19:57.52 | CIA-128 | BRL-CAD: 03brlcad * r49541 10/brlcad/trunk/misc/CMake/CompilerFlags.cmake: |
| 19:57.52 | CIA-128 | BRL-CAD: go ahead and enable sse for debug builds, but make an emphatic plea to not |
| 19:57.52 | CIA-128 | BRL-CAD: enable sse for release builds since we don't have the right run-time |
| 19:57.52 | CIA-128 | BRL-CAD: infrastructure in place to avoid run-time illegal instructions on systems that |
| 19:57.52 | CIA-128 | BRL-CAD: don't have sse. |
| 20:21.03 | jordisayol | brlcad: very good! I see you solved the problem! |
| 20:22.12 | jordisayol | brlcad: now there's -mtune=generic flag. is ok or is better to remove it? |
| 20:23.02 | jordisayol | brlcad: just to know before to build a new deb package for testing |
| 20:36.31 | CIA-128 | BRL-CAD: 03starseeker * r49542 10/brlcad/trunk/misc/CMake/CheckCInline.cmake: The MSVC build fails when trying to link openNURBS when we run this macro - avoid it on MSVC until the correct solution is determined. |
| 20:36.52 | CIA-128 | BRL-CAD: 03brlcad * r49543 10/brlcad/trunk/TODO: need to improve sse support big-time. been bitten by distribution bugs at least twice now. also metaball adjusting seems to be hosed. |
| 20:39.55 | CIA-128 | BRL-CAD: 03brlcad * r49544 10/brlcad/trunk/misc/CMake/CompilerFlags.cmake: mtune=generic wasn't the cause of the illegal instruction problem, so remove it. cause was sse. |
| 20:40.17 | brlcad | jordisayol: it's okay and better to remove it ;) |
| 20:41.10 | jordisayol | brlcad: ok, i'll create a new deb and send for testing, to assure that everything is fine now. many thanks! :-) |