Distribution: debian, unstable Hardware Environment: fujitsu-siemens lifebook (notebook) E-6624 (intel 830, pIII-933, radeon m6) Software Environment: debian. windows (98,2000,xp) is o.k. with acpi. Problem Description: first, thank you, the kernel oops from acpi are eliminated at 2.6.0testX, but: doing an "echo 3>/proc/acpi/sleep" sleeps the notebook well, the small lcd panel shows the blinking power icon to indicate this. If I press the power button to wake up, it starts to use the disk, and pccards (as shown on this small lcd display), and it is waken up (as the power icon does not blink any more), but braindead: no display, no keyboard, no network, so it hangs. the same happened (happening) in apm, my only goal is to solve somehow to be able to do suspend-to-ram... but currently both acpi and apm do not work. I tried the patch mentioned in a previous bug (1488?) (which enumerates devices in /proc/acpi/wakeup_devices or something, and put echo "LID" >/proc/acpi/wakeup_devices, no success. how to debug ? Steps to reproduce: load the kernel, echo "3" >/proc/acpi/sleep. press the power button.
Created attachment 1709 [details] dmesg from 2.6.0 loading... hope you find any idea..
is there possibly a relation to http://bugzilla.kernel.org/show_bug.cgi?id=1877 ? The i830 agp code is also used for the Intel 855GM chipset, apparently.
Is the problem still there with latest 2.6 kernel? If yes, can you make sure whether the problem is only with restore of the video (try connecting through network or serial after resume). If it is only the video problem, - Please try the options in Documentation/power/video.txt - Try the workaround in bug #3670
Yes, 2.6.8.1 is resuming without video (network access from the outside is ok. after resume). I've also tested this on older kernels and there were no such 'good' results. After all your notes and http://www.doesi.gmxhome.de/linux/tm800s3/s3.html is really good, http://www.doesi.gmxhome.de/linux/tm800s3/resources/boot-radeon.c almost does the job but it leaves the video in 80x24 but linux thinks it is 80x50 or something, so a sleep 2;SVGATextMode -a -o refines the resume process. I did not check it in X, but a dream already came true...
Good :).. You can push your luck further with the vgapost utility in bug #3670. If the user utility works, also try the kernel patch there and you should have X working too..
*** This bug has been marked as a duplicate of 3670 ***
X is also o.k. with http://www.doesi.gmxhome.de/linux/tm800s3/s3.html (script somewhat modified) and http://www.doesi.gmxhome.de/linux/tm800s3/resources/boot-radeon.c and an SVGATextMode -a -o. Now my problem is revitalizing ethernet card - after suspend in X it cannot reach the network 'some times' - it means the first impessions showed that netrok is o.k. from outside but now it feels that it is not available from inside-out. Anyhow. 2.8.1 + boot-radeon rocks.