Ubuntu
linux package

  1. Bug #367134
  2. Comment #3

Comment 3 for bug 367134

Revision history for this message
Jeremy Sequoia (jeremyhu) wrote :

The title needs to be changed since this is not an oops, but I don't see where to do that...

I used the pm trace functionality described here:
https://wiki.ubuntu.com/DebuggingKernelSuspend

but I did not get anywhere with that. On the next boot, dmesg showed:
Magic number: 5:435:547

but there was no hash match.

With X11 running using nvidia drivers and without X11 running, S1/standby works successfully (I did not test standby/nv)

With X11 running using nv or nvidia and without X11 running, S3/mem fails. I have periodically been able to get a network connection after the resume, and each time that I have, I've noticed that either Xorg was consuming 100% CPU (see https://bugs.launchpad.net/ubuntu/+source/linux/+bug/363672) or vbetool post was consuming 100% CPU (see https://bugs.launchpad.net/ubuntu/+source/vbetool/+bug/130979).

Using test_suspend=mem on boot (to eliminate possible bugs in modules), my video does not come back, and I can successfully interact blindly with the machine or connect via network.

My BIOS has an option to re-POST the vide card after S3 resume. Having this option on, I have never been able to connect via network or successfully interact blindly with the machine, so I assume the machine is hung with that option, so I've been debugging with it off.