![]() |
Etch/Feisty Segmentation fault - Printable Version +- Forums (https://www.vdrift.net/Forum) +-- Forum: Community (https://www.vdrift.net/Forum/forumdisplay.php?fid=3) +--- Forum: Bugs (https://www.vdrift.net/Forum/forumdisplay.php?fid=7) +--- Thread: Etch/Feisty Segmentation fault (/showthread.php?tid=566) |
- thelusiv - 03-25-2007 Alright, thanks again for the testing. Here's yet another package to test. This time the combination is g++3.3 and g++-4.1. Same link: http://clemsonlinux.org/~thelusiv/VDrift-2007-03-23-minimal-2.package - Nigo - 03-26-2007 same :oops: if we don't have the same libc version, it would be better to compile it on Etch directly. I will see if I have time to try it this week - rookie1 - 03-26-2007 segfault in Feisty too. - thelusiv - 03-26-2007 Well I'm not sure what to do at this point. I think the autopackage devs may even be stumped, though I'll keep asking questions. I guess the best solution would be to make a .deb that will work on Feisty and Etch, and direct users of those operating systems to that. - Nigo - 03-26-2007 I just compiled VDrift (it's sooooo easier and faster on Linux :roll: ) and replaced the binary and it now works also, in the SConstruct file, when using old_openal=1, "-lalut" is missing - thelusiv - 03-26-2007 Yes, everything should work from a compile. -lalut is missing on purpose, that option is meant to do that, because older versions of OpenAL had ALUT built in to libopenal. - Nigo - 03-26-2007 ok. Also, is it just me or there is something wrong with the AI ? the race begin at 40fps and the framerate drops over time, after 30s I got less than 1fps ??!! everything is fine on Practice Game - thelusiv - 03-26-2007 It's not the AI that's the problem, but the collision detection routines running on many cars does not perform well. This is mentioned in the known issues section of the release notes. ![]() - thelusiv - 03-27-2007 Alright some more help from the ever wonderful autopackage devs (thanks Taj!) I learned that the package was not being double-compiled as I thought it was, and for a subtle reason...because the two different compiled binaries were not going into the right place. I have put up yet another package, this time it is also compiled with debug symbols so you can use gdb to obtain a bactrace of the crash. The new link: http://clemsonlinux.org/~thelusiv/VDrift%202007-03-27-debug-2.package - joevenzon - 03-27-2007 gdb backtrace: Code: Program received signal SIGSEGV, Segmentation fault. seems like there's some funkiness with libstdc++ - thelusiv - 03-27-2007 Alright last package for tonight: http://clemsonlinux.org/~thelusiv/VDrift%202007-03-27-debug-2.package If you test this please do the following Code: gdb vdrift Code: backtrace Also, if you could paste the output of the command Code: ldd vdrift Also the Visual Dependency Walker could provide some useful info. Get it here: http://ftp.sunsite.dk/projects/autopackage/packages/Visual%20Dependency%20Walker%201.2.package The problem is getting narrower, and it seems like it might be a problem with apbuild. It is including libstdc++.so.5 when it shouldn't. I'll update on the status of this problem... - rookie1 - 03-27-2007 The latest package works in Feisty. - thelusiv - 03-27-2007 That's good to hear. Here's one more package, this time the difference is that I built the static libraries myself using apbuild, so it can do its magic on the entire thing. http://clemsonlinux.org/~thelusiv/VDrift-2007-03-27-debug-2-all-apbuild.package - joevenzon - 03-27-2007 With the package in your 6:26 PM post: Code: Program received signal SIGSEGV, Segmentation fault. as soon as I click on practice game. - thelusiv - 03-28-2007 Alright here's yet another package. This one was built with a patch to apbuild that should prevent it from statically linking X... http://clemsonlinux.org/~thelusiv/VDrift-2007-03-27-debug-2-no-static-x.package |