XBMC + nvidia problems
#1
This is my first time setting up X, and it's on a fresh install of Gentoo. I'm having problems getting it all to play nice.
As some background information, I'm setting this box up to run XBMC.
After following the Gentoo x86 handbook, I carried on to the article on setting up X. From there, to the nvidia setup guide. After a bit back and forth from there, I ran into some minor issues, but I figured there were all resolved when I emerged twm and xterm, which worked without a hitch.
I also tried "glxinfo | grep direct" which came back with "direct rendering: Yes". When I tried launching glxgears, a window for it would open, then close with the message "Running synchronized to the vertical refresh. The framerate should be approximately the same as the monitor refresh rate. Aborted". I thought nothing of this.
From here, I did some digging around, and ended up finding media-tv/xbmc-9.11-r3, which I masked and emerged.
From here, I did a simple echo xbmc > ~/.xinitrc and ran startx. The first two times, X had errors. The next time I tried it, XBMC launched fine. At some point while looking around, it crashed.
After more restarting, googling, and trying to launch XBMC, I'm still at this point.
Advice from IRC said to try emerging extreme-tuxracer and seeing if that would run. It seemed to have the exact same issue(s) as XBMC. Sometimes X seemed to segfault, but after a reboot, then startx (with ~/.xinitrc having the new tuxracer comamnd), tuxracer would seem to start, then crash. After another startx, X seemed to segfault again.
From here, I emerged xterm and twm, rebooted for good measure, did a 'mv ~/.xinitrc ~/.xinitrc.old' to make sure that wouldn't affect it, and ran startx. xterm/twm work perfectly from here.
After running "xbmc" from xwm, I get a core dump, as attached.
From here I killed X using ctrl + alt + backspace. After running another startx, xwm doesn't start, see attached log file.
I've also attached several logs from XBMC, if anyone thinks they could have a look at them and determine what might be the problem.
Also, I've attached any files I would think to be relevant to my computers configuration, including my kernel configuration.

If anyone needs any more info, I'm more than happy to try and get it. Hopefully I can get XBMC working .

I realized that theres a log of files, so I've just made a .tar.bz2 and uploaded it.
http://www.mediafire.com/?ndtm2wnotmd

Edit: Also added output from 'emerge --info' here - http://pastie.org/904193 and' rc-status' here - http://pastie.org/904194

I've just copied accross from the thread found at http://forums.gentoo.org/viewtopic-t-822239.html so I can bring it to a wider range of people who may be able to help me get XBMC up and running. Feel free to post either here or over there.

Edit2: And a lspci - http://pastie.org/904767
Edit3: At the suggest of someone, I have upgraded my graphics drivers to version 195.36.15 (Masked as ~x86). X started fine, XBMC crashed, this is the new coredump + log - http://uppit.com/v/B1F32X98
Edit4: As per instruction from #xorg, I have recompiled xorg-server with the debug useflag and without the -fomit-frame-pointer cflag. When trying to start twm, I got the following updated X log file - http://pastie.org/904894. If I could get more debugging information out of X or anything, please let me know. Recompiling isn't too much of an issue (It's not working anyway)
Reply
#2
Although I wouldn't consider this "solved" as such, it seems everything starts working perfectly when I'm using NvAGP rather than agpgart. Not sure if it was a misconfiguration of agpgart, but I'll just use NvAGP now that it's working.

I guess this was X/nvidia that was causing the XMBC crashes too. Sorry guys
Reply

Logout Mark Read Team Forum Stats Members Help
XBMC + nvidia problems0