Public bug reported:

I am using an Intel D945GCLF2 motherboard with the dualcore
hyperthreading Atom 330 processor.

Everything seems to be working fine so far, except suspend-to-RAM.

Suspend-to-RAM (ACPI S3) is explicitly supported by that motherboard, I
just upgraded to the latest BIOS version (0122), power-management-option
in BIOS "ACPI Suspend" is set to "S3 State" (the other selectable option
is "S1 State").

Testing with Ubuntu 8.10, once I initiate pm-suspend the computer is
gone to sleep within a few seconds, and it's definitely correctly
sleeping (and not just off), because it can be woken up by pressing any
button on a connected PS/2 keyboard (in addition to using the power
button), which is not the case, if the computer is powered off (then it
can only be switched on with the power button).

The problem is now, that it can not be woken up correctly. Once a button
on the keyboard or the power button is pressed, I can see the LED on the
CDROM drive flash several times (as if it's searching for something on a
CDROM), and after a few seconds the computer starts to boot normally, as
if it was not sleeping, but just switched off.

It kind of seems, as if the kernel segfaults on resume and then resets
the machine, which results in a normal boot, but I can't really tell,
because during the failed resume I have no output on screen, and then it
is just followed by a normal boot process.

I understand, that such a problem is very hard to debug, because it just
can not be analyzed, what's going on during such a failing resume. I
already searched the web, and found the Sleep Quirk Debugger Page at

http://people.freedesktop.org/~hughsient/quirk/quirk-suspend-
advanced.html.

I followed the shown instructions:

$ echo 1 > /sys/power/pm_trace
$ pm-suspend

but checking dmesg after a failed resume and following reboot just gives
this:

$ dmesg | grep -i matches
[    2.836980] bdi 1:15: hash matches

and I have no idea, what that means (it does not look like the example
shown on the mentioned webpage).

Can anybody with an Intel D945GCLF or D945GCLF2 motherboard please
check, if suspend-to-RAM results in the same problem on resume, as
described, and report back?

I am confident, that this is a general issue, because I tried it with a
stock Ubuntu 8.10 installation (minimal-install using the alternate CD
with "install a commandline-system"), and the Ubuntu 8.10 Desktop LiveCD
(both 32bit). It's just strange, that I could not find any info on
resume problems for that motherboard searching the internet.

** Affects: ubuntu
     Importance: Undecided
         Status: New

-- 
Intel D945GCLF2 Motherboard: Unable to use suspend-to-RAM, resume broken
https://bugs.launchpad.net/bugs/304457
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to