Comment 7 for bug 1385761

It is reproducible. I switched back to the nVidia-provided driver, but can go back to Nouveau if to reproduce it again if needed. It happens when I leave VirtualBox overnight with my Windows 8.1 64-bit guest running. I _think_ it happens when the Windows guest decides to blank the screen, but I'm not 100% sure as I've always been asleep when it happens...

Along with the above message regarding libdrm_nouveau.so.2.0, the messages below from nouveau were also what led me to believe it to be a nouveau problem, although clearly it could be something VirtualBox is messing up when calling nouveau that causes the trap.

Oct 25 16:36:34 rlinux kernel: [ 138.214092] nouveau E[ PGRAPH][0000:01:00.0] TRAP_M2MF IN
Oct 25 16:36:34 rlinux kernel: [ 138.214122] nouveau E[ PGRAPH][0000:01:00.0] TRAP_M2MF 00320251 203cbf80 00000000 04000432
Oct 25 16:36:34 rlinux kernel: [ 138.214130] nouveau E[ PGRAPH][0000:01:00.0] TRAP
Oct 25 16:36:34 rlinux kernel: [ 138.214143] nouveau E[ PGRAPH][0000:01:00.0] ch 6 [0x003f800000 VBoxCrWinCmd[3045]] subc 5 class 0x5039 mthd 0x0328 data 0x00000000
Oct 25 16:36:34 rlinux kernel: [ 138.214162] nouveau E[ PFB][0000:01:00.0] trapped read at 0x0020377000 on channel 0x0003f800 [VBoxCrWinCmd[3045]] PGRAPH/DISPATCH/M2M_IN reason: PAGE_NOT_PRESENT
Oct 25 16:36:41 rlinux kernel: [ 145.175686] nouveau E[VBoxCrWinCmd[3045]] push 0 buffer not in list
Oct 25 16:36:41 rlinux kernel: [ 145.175734] VBoxCrWinCmd[3045]: segfault at 7f7800000000 ip 00007f615168b503 sp 00007f61524ffae0 error 4 in libdrm_nouveau.so.2.0.0[7f6151689000+5000]