IRC log for #bzflag on 20130328

00:29.26*** join/#bzflag swigg_ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
01:25.49*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
01:43.20*** join/#bzflag Constitution (~josh@bzflag/developer/Constitution)
01:46.52*** join/#bzflag Detrimental (~james@h245.214.28.71.dynamic.ip.windstream.net)
01:48.20*** join/#bzflag EvilJStoker (jstoker@claire.jcs.me.uk)
02:03.30*** join/#bzflag Guest31852 (jstoker@unaffiliated/jstoker)
03:43.10*** join/#bzflag JeffM (~Jeffery@cpe-76-167-236-199.socal.res.rr.com)
03:43.10*** join/#bzflag JeffM (~Jeffery@unaffiliated/jeffm2501)
03:43.10*** mode/#bzflag [+v JeffM] by ChanServ
04:15.39*** join/#bzflag lahwran- (lahwran@python/site-packages/lahwran)
04:29.42BulletCatcherHuh.
04:29.43BulletCatcherOn the g++ command line, -std=c++11 must come *after* (or replace) -ansi to successfully compile r22713.
04:41.18BulletCatcherThe GCC bundled with Solaris 10 is version 3.4.3, which predates any support for the newfangled C++11 standard.
04:41.20BulletCatcherIt can be used to bootstrap a suitably new version of GCC, however.
04:56.30JeffMheh does anyone host on solaris?
04:56.35BulletCatcherI do.
04:56.42JeffMseriously?
04:56.45BulletCatcherYup.
04:56.52JeffMwhy?
04:57.09BulletCatcherBecause that is the server I have available.
04:57.24BulletCatcherIt is even a SPARC rather than x86.
04:57.51*** join/#bzflag jeffmOSX (~jeffm@cpe-76-167-236-199.socal.res.rr.com)
04:58.29BulletCatcherAnyway, it looks like GCC 4.3 or higher is needed for C++11 support.
04:59.53jeffmOSXshared_ptr may be supported as c++0x in earlyer
05:01.09BulletCatcherhttp://gcc.gnu.org/projects/cxx0x.html indicates that 4.3 introduced c++0x, which was upgraded to c++11 in 4.7.
05:02.11*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
05:09.03BulletCatcherXcode 4.1, which I think is the last version with a real GCC, has gcc-4.2 and does not support c++0x/c++11.
05:09.20jeffmOSXk
05:09.31jeffmOSXso no PPC then
05:09.51jeffmOSXwell if you want to wait on it we can revert it over to the standard pointer version
05:09.51BulletCatcherDefinitely not.
05:10.17BulletCatcherI am willing to leave this one change in for now to see who complains.
05:11.10BulletCatcherIt would be nice if we could stop pretending to support IRIX, BeOS, etc.
05:11.55jeffmOSXyeah
05:12.43jeffmOSXOSs that don't support shared_ptr 'could' use a small boost wraper
05:13.16BulletCatcherThat would work, but ouch!
05:13.29jeffmOSXyeah
05:13.40jeffmOSXit'd be one fancy config.h
05:17.19jeffmOSX4.3 was released in 2008, so that's a decent enough time ago I think
05:18.36BulletCatcherRed Hat Enterprise Linux 5 has gcc 4.1.
05:18.38BulletCatcherRHEL6 has 4.4.
05:22.37*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
05:23.14BulletCatcherRHEL5 has gcc43 and gcc44 packages available, so there is no real problem there.
05:24.05jeffmOSXso sounds like it should be fine for everyone except you and that PPC guy
05:24.19BulletCatcherCould be.
05:24.32BulletCatcherAnd I know how to install a new enough compiler.
05:24.39jeffmOSXor at least it sounds reasonable that people can get the thing that makes it work
05:25.22jeffmOSXapple abandoned the PPC in 2009, it's been a while
05:25.39BulletCatcherI haven't looked at Debian gcc versions.
05:27.36jeffmOSXand they started the intel transion in 2006
05:29.54jeffmOSXmy old debian server has 4.4
05:30.01jeffmOSXsorry ubuntu
05:30.10BulletCatcherI just compiled the entire BZFlag source tree using gcc 4.7 with -std=c++0x instead of -ansi, and it seems to work just the same.
05:30.29jeffmOSXit's on maveric
05:31.19BulletCatcherI'm going to commit the -ansi -> -std=c++0x change in configure.ac.
05:31.24jeffmOSXk
05:31.26BulletCatcherIn for a penny, in for a pound.
05:31.36jeffmOSXit'll be a good way to see who breaks
05:31.57jeffmOSXI believe blast and I_Died_Once have update scripts that will tell us if they puke ore not
05:32.41I_Died_Onceheh I's been pingered
05:32.57jeffmOSXwhat version of linux does your server run? and what GCC does it have?
05:32.58I_Died_Onceyou mean the update script for svn?
05:33.22I_Died_OncePlanet MoFo is running on Debian Squeeze...
05:33.38I_Died_Oncethe svn script works, at least it did for me on the server as well as my home machine
05:34.01I_Died_Oncewhats the command to see what version of GCC ?
05:34.38I_Died_Onceit's whatever version Debian Squeeze has, I haven't installed or taken any jackassy routes with the server
05:35.05BulletCatcher"gcc --version"
05:35.57I_Died_Once$ gcc --version
05:35.57I_Died_Oncegcc (Debian 4.4.5-8) 4.4.5
05:36.02I_Died_Oncety bulletcatcher
05:36.10jeffmOSXyou should be good then, we'll have to see
05:36.21BulletCatcherYou shouldn't have a problem with the commit I will make shortly, then.
05:36.39I_Died_Onceuh oh, sounds interesting
05:37.05jeffmOSXnot really, just a smart pointer and hopefully a fix for a memory leak
05:37.15Notify03BZFlag:bullet_catcher * 22714 trunk/bzflag/configure.ac: Change from compiling c++ code with -ansi (equivalent to c++98) to -std=c++0x, which requires GCC 4.3 or higher. Welcome to 2008.
05:37.46jeffmOSXwonder what version of gcc supports lambdas
05:38.08jeffmOSXand std::function
05:47.53BulletCatcherI have the Sun Studio 12 Update 1 compiler at work.
05:47.54BulletCatcherIt does *not* seem to support any form of c++11.
05:50.15jeffmOSXthow a VM on it :)
05:52.49BulletCatcherI already have gcc 4.5 there, so not a problem.
05:54.00BulletCatcherBut using a completely different compiler can reveal all sorts of portability issues.
05:55.49*** join/#bzflag Gnurdux (~gnurdux@pppoe-68-142-55-30.gmavt.net)
06:11.42DTRemenakBulletCatcher: Solaris Studio 12.3 doesn't support any of c++11 either.  Last year they said "maybe next year."
06:11.52DTRemenaknot holding my breath ;)
06:29.02BulletCatcherMe neither.
06:44.00*** join/#bzflag CRWheelchair (~CRWheelch@c-67-168-181-62.hsd1.wa.comcast.net)
06:49.55*** part/#bzflag CRWheelchair (~CRWheelch@c-67-168-181-62.hsd1.wa.comcast.net)
06:57.30*** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org)
07:59.56*** join/#bzflag Marzipan (~Marzipan@sign-4db6aca6.pool.mediaWays.net)
10:27.00*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
10:37.50*** join/#bzflag Nitroxis (~n@wtwrp.de)
11:24.05*** join/#bzflag Nidhoggr (cj@jormungandr.dragon.org)
11:24.05*** join/#bzflag ahs3- (~ahs3-@adsl-065-005-193-158.sip.rdu.bellsouth.net)
11:24.06*** join/#bzflag mdskpr__ (~quassel@li59-174.members.linode.com)
11:24.06*** join/#bzflag allejo_ (~allejo@li59-174.members.linode.com)
11:24.06*** join/#bzflag apiecux (~apiecux@ui89-892t.21z4-ee.ldti.srv.parano.me)
11:24.07*** join/#bzflag swigg (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
11:24.07*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
11:24.07*** join/#bzflag AAA_awright_ (~aaa@69.72.216.27)
11:24.21*** join/#bzflag Scrumpy_ (~Scrumpy@smurfnet.xs4all.nl)
11:24.22*** join/#bzflag TyroneFHornigh (~dbw@108-197-33-160.lightspeed.sndgca.sbcglobal.net)
11:31.39*** join/#bzflag rathis_ (~rathis@b01bd9ce.bb.sky.com)
11:31.39*** join/#bzflag joevano (~joevano@bzflag/developer/JoeVano)
11:31.39*** join/#bzflag NTH (~bnc@bnc.moooo.org)
11:31.39*** mode/#bzflag [+v joevano] by card.freenode.net
11:54.22*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
11:54.25*** join/#bzflag Gnurdux (~gnurdux@pppoe-68-142-55-30.gmavt.net)
12:32.33*** join/#bzflag Rob5a (~Rob5a@cable-158-181-91-182.cust.telecolumbus.net)
12:33.09*** join/#bzflag swigg_ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
12:46.36*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
13:23.08*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
13:54.07*** join/#bzflag brlcad (~sean@66-118-151-70.static.sagonet.net)
13:58.31*** join/#bzflag Notify (~notify@66-118-151-70.static.sagonet.net)
14:13.57*** join/#bzflag BulletCatcher (~bc@bzflag/developer/BulletCatcher)
15:01.53*** join/#bzflag TimRiker (~TimRiker@bzflag/projectlead/TimRiker)
15:01.54*** mode/#bzflag [+o TimRiker] by ChanServ
15:42.30alezakos7>LINK : fatal error LNK1123: failure during conversion to COFF: file invalid or corrupt
15:42.36alezakos(while building on windows)
15:42.40alezakosWhat is that supposed to mean?
15:45.18blast007do you have Visual C++ 2010 Sevice Pack 1 installed?
15:45.36alezakosNope, should I?
15:45.48blast007yeah, try installing the service pack
15:51.53*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
16:05.04*** join/#bzflag jeffm (~jeffm@cpe-76-167-236-199.socal.res.rr.com)
16:05.05*** join/#bzflag jeffm (~jeffm@unaffiliated/jeffm2501)
16:05.05*** mode/#bzflag [+v jeffm] by ChanServ
16:08.10*** join/#bzflag swigg_ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
16:20.25alezakosThat works, thanks blast007
16:20.46*** join/#bzflag swigg_ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
16:28.10*** join/#bzflag sirquine (~quine@c-76-120-71-113.hsd1.co.comcast.net)
16:46.16*** join/#bzflag jeffm (~jeffm@unaffiliated/jeffm2501)
16:46.16*** mode/#bzflag [+v jeffm] by ChanServ
17:03.50*** join/#bzflag JeffM2501 (~Jeffery@unaffiliated/jeffm2501)
17:03.51*** mode/#bzflag [+v JeffM2501] by ChanServ
17:09.25JeffM2501offta, the PPC guy wants gentoo... that ain't exactly 'simple'
17:11.02blast007yeah... ;)
17:11.29JeffM2501did ether of those changes fix the memory leak?
17:11.51blast007the one you had me do initially fixed that leak (just the delete)
17:11.56blast007haven't tested it with the new change yet
17:12.17JeffM2501ok
17:12.21JeffM2501then the second should as well
17:12.27blast007I have valgrind output from both of my attempts as well.
17:12.38JeffM2501it efectivly does the same thing just using magic faries
17:13.11blast007prior to the delete, it lost around 90MB in a couple hours ;)
17:13.21JeffM2501heh
17:13.32JeffM2501anythign interesting in the report other then that?
17:13.35blast007after it was less than 1KB
17:14.09alezakosBasically
17:14.16alezakosI think the new commit didn't fix the leak
17:14.39JeffM2501have you tested it or are you just looking at the code?
17:14.53alezakosThe submitter of the bug has it on his server
17:14.59blast007hmm.  I'll start a test here with the new code and see.
17:15.01alezakosAnd it's still eating memory like crazy...
17:15.07JeffM2501k
17:15.24JeffM2501perhaps smart pointers aren't as smart as they shouold be on some versions of GCC
17:15.37JeffM2501the delete version is there too, just commented out
17:15.49blast007I'm using 4.6.3 here, so I'll let you know how it goes.
17:15.54JeffM2501k
17:16.35JeffM2501alezakos, this is the kinda stuff I wanted to see, how well these features are supported in various linux compilers
17:17.02JeffM2501I know how they work on windows, but I can't enforce as tight a range of compiler version on linux
17:17.52blast007just include gcc in src/other/  ;)
17:18.28JeffM2501and all the STL
17:18.47blast007I'm sure Debian would just love us to death.  :)
17:20.11JeffM2501hey, they told me that shared_ptr, auto, and std::function were fine to use
17:20.14JeffM2501I even asked
17:20.52blast007I mean if we put gcc in our tree
17:21.17JeffM2501yes that would not be faviorably received :)
17:21.21blast007hehehe
17:21.36blast007seeing as they already don't like the libs we have there
17:21.48blast007they strip them out when they package the tar
17:21.57JeffM2501nobody likes the libs we have
17:22.07blast007not even me
17:26.17blast007hmm, actually I can't test at the moment, since it requires a client
17:26.46blast007I can try it in around 3 hours from now
17:26.52JeffM2501k
17:27.15JeffM2501if the server is running I can probalby join after I finish my homework
17:27.54JeffM2501"write a class that prints the calendar for a given month and year"
17:27.58JeffM2501wheeeee
17:28.30blast007Oh man! Do you have to account for leap years too??
17:28.40JeffM2501yep
17:28.48JeffM2501and compute the day of the week
18:02.21JeffM2501the boring part is that I have to use visio to make a class diagram for it
18:03.23DTRemenakjust remember: "September 1752" is a fun test case ;)
18:04.35JeffM2501our seed date is 1800
18:04.45DTRemenakbah, that's no fun :P
18:04.58JeffM2501none of this class can be defined as fun
18:09.33*** join/#bzflag allejo (~allejo@unaffiliated/allejo)
19:09.23*** join/#bzflag swigg__ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
20:11.59*** join/#bzflag Gabe_G23 (uid557@bzflag/player/GabrielG)
20:57.59*** join/#bzflag swigg__ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
20:59.17*** join/#bzflag swigg_ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
21:04.12*** join/#bzflag swigg__ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
21:16.50blast007JeffM2501: okay, I have the test running now.  I'll let it run for a few hours.  The initial fix (with delete) peaked at around 13% of my RAM, while the broken one climbed up past 32% before I stopped.
21:45.56*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
22:07.29*** join/#bzflag Delusional (~delusiona@pool-108-17-117-49.pitbpa.fios.verizon.net)
22:07.30*** join/#bzflag Delusional (~delusiona@unaffiliated/delusional)
22:09.03*** join/#bzflag swigg_ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
22:09.40*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
22:34.08*** join/#bzflag swigg_ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
22:49.33*** join/#bzflag swigg__ (~swigg@nsc69.38.101-238.newsouth.net)
22:52.47*** join/#bzflag swigg_ (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
22:55.32*** join/#bzflag Quol (Ian@2600:3c03::f03c:91ff:fe93:2c7d)
23:15.43*** join/#bzflag swigg (~swigg@68-119-228-196.dhcp.ahvl.nc.charter.com)
23:15.44*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)
23:16.09*** join/#bzflag jeffm (~jeffm@cpe-76-167-236-199.socal.res.rr.com)
23:16.09*** join/#bzflag jeffm (~jeffm@unaffiliated/jeffm2501)
23:16.09*** mode/#bzflag [+v jeffm] by ChanServ
23:26.17blast007JeffM2501 / jeffm: the fix in SVN seems to work just fine for me.
23:44.18JeffM2501blast007, ok cool
23:50.51*** join/#bzflag swigg (~swigg@bzflag/player/Swigg)

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