| 00:37.49 | CIA-65 | BRL-CAD: 03starseeker * r50546 10/brlcad/trunk/src/librt/primitives/ (ehy/ehy_brep.cpp ell/ell_brep.cpp hyp/hyp_brep.cpp): |
| 00:37.50 | CIA-65 | BRL-CAD: Apply brep patches from 3513288 - slight alteration to the hyp patch which seems |
| 00:37.50 | CIA-65 | BRL-CAD: to result in correct raytracing of the hyp, although it probably needs a bit |
| 00:37.50 | CIA-65 | BRL-CAD: more checking to make sure all the surfaces are facing the right way. |
| 02:56.22 | *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol) | |
| 04:32.41 | CIA-65 | BRL-CAD: 03brlcad * r50547 10/brlcad/trunk/NEWS: |
| 04:32.41 | CIA-65 | BRL-CAD: add a line item for the conversion of implicit geometry into boundary |
| 04:32.41 | CIA-65 | BRL-CAD: representation format now that they are predominantly all implemented and seem |
| 04:32.41 | CIA-65 | BRL-CAD: to be working well (as seen by csgbrep conversion testing and brep mged |
| 04:32.41 | CIA-65 | BRL-CAD: command). 95+% all cliff with some support from wu on fixing up ehy, ell, hyp |
| 04:32.42 | CIA-65 | BRL-CAD: and 'brep' command. also, a smidgen from yours truly fixing various memory |
| 04:32.43 | CIA-65 | BRL-CAD: management bugs and stabbing in the initial solid bspline->brep conversion. |
| 04:37.29 | CIA-65 | BRL-CAD: 03Sean 07http://brlcad.org * r3660 10/wiki/Main_Page: link to tom's new mged faq too |
| 04:39.14 | brlcad | crdueck: undefined references are linker errors, not preprocessor or compilation errors |
| 04:39.49 | brlcad | that means you're missing a link to libbu ("-Lpath/to/libbu/dir -lbu" for *nix-style systems) |
| 04:40.36 | brlcad | crdueck: that said, why are you compiling test_density by hand? it should be compiled already during 'make', just not installed |
| 04:42.07 | brlcad | "cd brlcad-builddir/src/libanalyze && make clean && make" should rebuild build it |
| 04:52.46 | CIA-65 | BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/rights: changed group membership for User:Tbrowder from (none) to sysop: promoted so he can edit the main page |
| 05:19.11 | *** join/#brlcad KimK (~Kim__@2001:470:1f0f:1042:4261:86ff:fe43:bcad) | |
| 05:32.57 | *** join/#brlcad cristina (~cristina@188.24.76.101) | |
| 05:35.10 | *** join/#brlcad Stattrav_ (~Stattrav@117.192.157.79) | |
| 06:39.42 | *** join/#brlcad Stattrav_ (~Stattrav@117.192.133.52) | |
| 07:11.25 | *** join/#brlcad stas (~stas@82.208.133.12) | |
| 07:17.54 | *** join/#brlcad merzo (~merzo@233-204-132-95.pool.ukrtel.net) | |
| 07:18.24 | *** join/#brlcad KimK (~Kim__@209.248.147.2.nw.nuvox.net) | |
| 08:07.57 | *** join/#brlcad ksuzee (~ksu@193.151.105.83) | |
| 08:08.05 | *** join/#brlcad ksuzee_ (~ksu@193.151.105.83) | |
| 08:12.47 | *** part/#brlcad ksuzee_ (~ksu@193.151.105.83) | |
| 08:41.08 | *** join/#brlcad Stattrav_ (~Stattrav@117.192.133.52) | |
| 09:02.07 | *** join/#brlcad Jak_o_Shadows (~Fake@unaffiliated/jak-o-shadows/x-0479135) | |
| 09:22.11 | *** join/#brlcad cristina (~cristina@188.24.81.180) | |
| 10:16.17 | *** join/#brlcad kane_ (~Mesut@g226124248.adsl.alicedsl.de) | |
| 10:36.06 | *** join/#brlcad Jak_o_Shadows (~Fake@unaffiliated/jak-o-shadows/x-0479135) | |
| 10:43.54 | *** join/#brlcad merzo (~merzo@9-217-200-46.pool.ukrtel.net) | |
| 11:27.17 | *** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ) | |
| 11:27.36 | starseeker | yawns |
| 11:33.16 | *** join/#brlcad Stattrav_ (~Stattrav@223.234.140.220) | |
| 11:47.01 | *** join/#brlcad Stattrav_ (~Stattrav@223.235.130.40) | |
| 11:50.30 | brlcad | tosses a cheerio in |
| 11:52.13 | CIA-65 | BRL-CAD: 03erikgreenwald * r50548 10/brlcad/trunk/TODO: mention VLS %lld regress failure |
| 11:58.50 | *** join/#brlcad Stattrav_ (~Stattrav@223.234.14.15) | |
| 12:19.06 | brlcad | ``Erik: thoughts on http://sourceforge.net/tracker/?func=detail&atid=640804&aid=3526862&group_id=105292 |
| 12:52.39 | CIA-65 | BRL-CAD: 03brlcad * r50549 10/brlcad/trunk/src/adrt/isst_tcltk.c: apply Ksenija Slivko's ( ksuzee ) sf patch 3526862 removing the duplicitous Isst_Init() function. Issttcltk_Init() is identical and actually called. |
| 12:56.50 | ``Erik | brlcad: looks like different platforms had issues with that, starseeker dup'd the function with the name change for linux on may 11, 2011 |
| 12:56.57 | CIA-65 | BRL-CAD: 03brlcad * r50550 10/brlcad/trunk/AUTHORS: chris and wu made code contributions, attribute correctly |
| 12:57.10 | ``Erik | r44604 |
| 12:58.17 | ``Erik | don't recall exactly why I changed the name with 44590, probably trying to get it to work with the mac |
| 12:59.43 | CIA-65 | BRL-CAD: 03brlcad * r50551 10/brlcad/trunk/AUTHORS: credit ksenija silvko (ksuzee) with her sf code reduction patches, one recently applied to isst/adrt with others pending. |
| 12:59.47 | *** join/#brlcad Stattrav_ (~Stattrav@223.234.14.15) | |
| 13:01.13 | brlcad | I guess we'll have to revisit since the commit doesn't say why it's needed (and that claim seems dubious) |
| 13:01.51 | brlcad | maybe he just had an unclean build since your previous commit changed the name |
| 13:02.08 | brlcad | either way, shouldn't have been a wholesale copy of the other |
| 13:02.35 | ``Erik | yeah, if it is a name assembly issue, one should call the other |
| 13:04.12 | ``Erik | fbsd and mac are failing on launching isst_tcltk now, building on linux.. . (might be a cmake issue, putting the lib in the wrong place) |
| 13:10.36 | starseeker | brlcad: I have a vague memory that I needed one init function name on Windows and another init name on Linux... |
| 13:11.57 | starseeker | ``Erik: yeah, I may not have the lib in the right place (or the package require foo looking in the right place) in the installed location |
| 13:12.22 | starseeker | been a while since I did anything with the isst stuff, so there may be some early hacks in there |
| 13:13.20 | *** join/#brlcad andrei (~andrei@5-12-70-105.residential.rdsnet.ro) | |
| 13:13.27 | ``Erik | TCL_PKGINDEX(issttcltk isst "0.1") <-- is that requesting both issttcltk and isst packages defined? |
| 13:20.51 | CIA-65 | BRL-CAD: 03erikgreenwald * r50552 10/brlcad/trunk/src/adrt/CMakeLists.txt: put library in lib/ instead of lib/isst0.1/ so require can find it |
| 13:22.39 | CIA-65 | BRL-CAD: 03Tbrowder 07http://brlcad.org * r3661 10/wiki/MgedFAQ: |
| 13:34.17 | CIA-65 | BRL-CAD: 03erikgreenwald * r50553 10/brlcad/trunk/src/adrt/isst_tcltk.c: Isst_Init() is the name used by FreeBSD, Linux and MacOSX. Add a comment breadcrumb mentioning that Issttcltk_Init() might be required for certain platforms. |
| 13:37.03 | *** join/#brlcad Stattrav_ (~Stattrav@223.234.14.15) | |
| 13:54.37 | brlcad | I presume that's so package require can load an "Isst" package, so it makes sense it'd need to be named specifically that (unless you change the name of the package) |
| 14:07.04 | Stattrav_ | brlcad: I was wondering if the paths of executables used for the benchmarking be stored seperately ? |
| 14:17.20 | *** join/#brlcad Stattrav_ (~Stattrav@223.235.133.204) | |
| 14:25.17 | CIA-65 | BRL-CAD: 03Tbrowder 07http://brlcad.org * r3662 10/wiki/MgedFAQ: added "titles" to questions |
| 14:38.44 | *** join/#brlcad kane__ (~Mesut@g226123012.adsl.alicedsl.de) | |
| 14:43.10 | *** join/#brlcad ksuzee (~ksu@193.151.107.42) | |
| 14:43.10 | CIA-65 | BRL-CAD: 03Tbrowder 07http://brlcad.org * r3663 10/wiki/Msgid1010480: extract of the note from MgedFAQ to here to serve as a note to that page |
| 14:43.36 | *** join/#brlcad Stattrav_ (~Stattrav@223.234.119.200) | |
| 14:50.04 | CIA-65 | BRL-CAD: 03Tbrowder 07http://brlcad.org * r3664 10/wiki/MgedFAQ: removed top note and made a reference to it |
| 15:08.39 | *** join/#brlcad RedTide (~myself@host221-244-dynamic.51-79-r.retail.telecomitalia.it) | |
| 15:09.03 | *** part/#brlcad RedTide (~myself@host221-244-dynamic.51-79-r.retail.telecomitalia.it) | |
| 15:34.15 | brlcad | Stattrav: not sure what you mean, there's the brlcad-config utility? |
| 15:34.42 | brlcad | the benchmark itself has no awareness of the utilities other than their names and standard places to look for them |
| 15:36.59 | CIA-65 | BRL-CAD: 03Phoenix 07http://brlcad.org * r3665 10/wiki/User:Phoenix/GSoc2012/Reports: /* Community Bonding */ |
| 16:08.34 | *** part/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol) | |
| 17:45.39 | *** join/#brlcad DarkCalf (DC@173.231.40.98) | |
| 18:01.20 | *** join/#brlcad Stattrav_ (~Stattrav@223.236.1.226) | |
| 18:07.40 | CIA-65 | BRL-CAD: 03Ksuzee 07http://brlcad.org * r3666 10/wiki/User:Ksuzee/Reports: |
| 18:48.58 | *** join/#brlcad Stattrav_ (~Stattrav@117.192.139.161) | |
| 18:57.33 | CIA-65 | BRL-CAD: 03Stattrav 07http://brlcad.org * r3667 10/wiki/User:Stattrav/GSoC2012_log: Log of the day 15th May |
| 19:00.59 | *** join/#brlcad ibot (~ibot@rikers.org) | |
| 19:00.59 | *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || logs: http://ibot.rikers.org/%23brlcad/ || Congratulations to our 11 accepted GSoC students!! | |
| 19:03.10 | CIA-65 | BRL-CAD: 03starseeker * r50555 10/brlcad/trunk/src/other/lemon/lempar.c: if yyRuleName isn't defined, don't try to sizeof it. |
| 19:05.37 | CIA-65 | BRL-CAD: 03starseeker * r50556 10/brlcad/trunk/src/other/lemon/lemon.c: While we're at it, pull the latest lemon.c file |
| 19:40.46 | CIA-65 | BRL-CAD: 03n_reed * r50557 10/brlcad/trunk/src/other/step/src/ (7 files in 4 dirs): Apply changes from SCL git 1016b07. Use std::string more frequently and move ErrorDescriptor message strings to stack to simplify things. |
| 19:41.47 | *** join/#brlcad Stattrav (u3131@unaffiliated/stattrav) | |
| 19:41.47 | *** join/#brlcad Stattrav (u3131@gateway/web/irccloud.com/x-btcdotlynjvynyrw) | |
| 19:51.55 | *** join/#brlcad Stattrav_ (~Stattrav@117.192.139.161) | |
| 20:32.36 | andrei | hello |
| 20:50.19 | ``Erik | howdy, andrei |
| 20:58.53 | CIA-65 | BRL-CAD: 03erikgreenwald * r50558 10/brlcad/trunk/src/libbu/booleanize.c: Apply fix to '0' test from sf tracker patch 3519874 (cprecup's test_booleanize). |
| 21:01.25 | andrei | Erik, did you have time to see my mailing-list post? |
| 21:05.20 | ``Erik | yesterdays? I skimmed it. I'll see if I can re-read it later tonight or tomorrow, but I'm heading home right now... if you have questions or more comments, I'm much better about reading/responding to irc than mailing lists... (for the 24k issue, what about moving a few megs? maybe test speed across a range of sizes and graph results?) |
| 21:05.32 | ``Erik | heads home :) hasta |
| 21:07.45 | andrei | ah, that would be a good idea. |
| 21:08.27 | brlcad | andrei: also, the ttcp program doesn't send packets the size of the inputs, does it? |
| 21:09.13 | brlcad | er, tpkg |
| 21:11.11 | andrei | the file I sent from is a 50mb /dev/random stuff. |
| 21:11.19 | andrei | I have been using it for purposes like this. |
| 21:12.05 | andrei | I ll try with a computer from a different network |
| 21:12.16 | brlcad | that's not my point :) |
| 21:12.18 | andrei | I know |
| 21:12.27 | brlcad | you could send 50TB |
| 21:12.49 | brlcad | and it could still be sending that 50TB in 1 byte packets |
| 21:13.13 | brlcad | i'd bet tpkg is sending the data in something like 2048 byte packets |
| 21:13.33 | brlcad | the 24k issue wasn't sending 24k of data but sending a single packet larger than 24k |
| 21:15.36 | brlcad | indeed: bytes = fread(buffer, 1, 2048, fp); |
| 21:16.51 | CIA-65 | BRL-CAD: 03brlcad * r50559 10/brlcad/trunk/src/libpkg/tpkg.c: use TPKG_BUFSIZE here too |
| 21:23.34 | brlcad | andrei: that'd be a fine simple patch, make that package size an option to tpkg instead of just being hard-coded |
| 21:27.29 | andrei | thanks for the hint. |
| 21:28.10 | *** join/#brlcad stas (~stas@188.24.35.114) | |
| 21:34.38 | CIA-65 | BRL-CAD: 03brlcad * r50560 10/brlcad/trunk/AUTHORS: credit cristina precup for her booleanize patch #3519874 (unit test for the file src/libbu/booleanize.c) partially applied in r50558 |
| 21:36.15 | brlcad | andrei: do you know shell scripting? |
| 21:37.16 | brlcad | it might be beneficial to write a shell script that invokes tpkg for various data input sizes and various buffer sizes, so we can see a visual 3D graph of performance |
| 21:37.35 | andrei | ah, I did learn some |
| 21:38.05 | andrei | will do that |
| 21:38.41 | brlcad | something like http://www.sims2000.com/images/demo3.gif or http://www.ni.com/images/labview/2010/neutral/ui_3d_surface_graph.jpg |
| 21:38.47 | andrei | I don t manage to recompile tpkg and I don t know why |
| 21:39.00 | brlcad | "make tpkg" should do it |
| 21:39.22 | brlcad | you could also run: cd src/libpkg |
| 21:39.30 | brlcad | then "make clean && make |
| 21:39.46 | brlcad | it'll only clean/delete and rebuild the libpkg products |
| 21:40.34 | andrei | my brlcad install is located in /usr/brlcad/dev-7.21.0/ |
| 21:41.23 | brlcad | if you only run make, it'll not reinstall |
| 21:41.31 | brlcad | you have to run make install for that |
| 21:41.52 | brlcad | still, best to be in path_to_cmake_build_dir/src/libpkg |
| 21:42.02 | brlcad | so you can run make clean && make && make install |
| 21:42.06 | brlcad | you don't need to install, though |
| 21:42.06 | andrei | yes, I tried to run make tpkg |
| 21:42.10 | andrei | in the build directory. |
| 21:42.14 | brlcad | and? |
| 21:42.28 | andrei | [andrei@Andrei-Desktop brlcad-build]$ make tpkg |
| 21:42.29 | andrei | make: *** No rule to make target `tpkg'. Stop. |
| 21:42.31 | andrei | [andrei@Andrei-Desktop brlcad-build]$ pwd |
| 21:42.32 | andrei | /home/andrei/brlcad-build |
| 21:42.38 | andrei | maybe I m doing something wrong, I initially compiled it using gcc and paths |
| 21:42.51 | andrei | the gcc command that s written inside it. |
| 21:43.25 | brlcad | oh! |
| 21:43.28 | brlcad | that's not your fault |
| 21:43.36 | brlcad | the cmake build is lacking a target for tpkg apparently |
| 21:43.54 | brlcad | the autotools build had one, apparently one needs to be added for cmake |
| 21:44.16 | andrei | I still have the binary, so I m sure I managed to compile it |
| 21:44.23 | brlcad | probably by hand |
| 21:44.38 | andrei | I used gcc -I/usr/brlcad/include -L/usr/brlcad/lib -o tpkg tpkg.c -lpkg -lbu |
| 21:44.47 | andrei | well it has two minor issues |
| 21:44.51 | brlcad | sure, that would be by hand ;) |
| 21:45.08 | andrei | it needs spaces between options and path. and my include is /brlcad/$version/include |
| 21:45.39 | brlcad | you can add the cmake build logic that's missing |
| 21:46.18 | andrei | now It seems to miss common.h |
| 21:46.20 | andrei | gcc -I /usr/brlcad/dev-7.21.0/include -L /usr/brlcad/dev-7.21.0/lib -o tpkg tpkg.c -lpkg -lbu |
| 21:46.22 | andrei | tpkg.c:30:20: fatal error: common.h: No such file or directory |
| 21:46.43 | brlcad | nods |
| 21:46.55 | brlcad | so it's basically up to you what you want to learn next |
| 21:47.23 | brlcad | you can and should figure out how to build applications manually like that -- it's a good skill regardless |
| 21:47.48 | brlcad | and/or you can figure out the cmake logic to get cmake to build it for you -- less prevalent but simple skill to learn |
| 21:47.55 | brlcad | example in src/libbu/CMakeLists.txt for that |
| 21:48.49 | brlcad | or you could use the autotools build where it's already defined and will build/rebuild for you (but is less useful down the road, you don't learn anything) |
| 21:49.04 | brlcad | suggest the first two options ;) |
| 21:50.21 | brlcad | ask a question when you're stuck for more than a couple minutes |
| 21:53.02 | andrei | it claims it can't find common.h( as I linked earlier) |
| 21:53.15 | andrei | but the file is present on the system |
| 21:53.16 | andrei | ah |
| 21:53.18 | andrei | I found the issue |
| 21:53.34 | andrei | <PROTECTED> |
| 22:04.23 | CIA-65 | BRL-CAD: 03n_reed * r50561 10/brlcad/trunk/src/other/step/src/ (4 files in 2 dirs): consolidate definitions of STRING_DELIM |
| 22:05.53 | brlcad | there ya go, so you need at least two include dirs |
| 22:27.06 | *** join/#brlcad andrei (~andrei@5-12-70-105.residential.rdsnet.ro) | |
| 23:29.37 | cristina | ``Erik: I fixed the test_booleanize.c patch and I've also found an identical bug as the previous one a few lines below. |
| 23:29.58 | cristina | I've updated the patch. You can have a look here: https://sourceforge.net/tracker/index.php?func=detail&aid=3519874&group_id=105292&atid=640804 |
| 23:30.05 | cristina | thanks for your advice |
| 23:33.24 | CIA-65 | BRL-CAD: 03Cprecup 07http://brlcad.org * r3668 10/wiki/User:Cprecup/GSoC2012_progress: unit tests patch for src/libbu/booleanize.c updated + small bug found |
