Distribution: Fedora Core 1 Hardware Environment: Compaq Presario-2100CA Software Environment: Kernel 2.6.9 Problem Description: Suspend to ram using by echoing "mem" to /sys/power/state seems to work fine and puts the laptop into suspend (seems off, with power led flashing). When pressing the power button to awake the machine, it comes back and immediately powers off. Installing a custom DSDT did not change anything. The power-off is pretty much immediate when trying to resume. Same behaviour observed under a number of older kernels (2.4.xx through to 2.6.7). Other ACPI functions, suspend-to-disk and cpu frequency scaling work just fine. Steps to reproduce: echo -n "mem" >/sys/state/power -- wait for suspend -- press power button
Created attachment 3974 [details] Output of lspci -v
Created attachment 3975 [details] Output from dmesg
This is a known issue. System is suspending and resumeing properly. But, after resume acpid Or other such daemon) thinks that the power switch pressed event is for shutdown and shutsdown the system. One workaround is to stop the acpid or powersaved before suspending and start it again after resuming, by using something like /etc/init.d/acpid stop /etc/init.d/acpid start You can also look at bug #3525 for additional comments.
I'm impressed - that was a quick response. Unfortunately, it seems like bug 3525 is a little different as it goes into a shutdown procedure initiated by acpid. In my case there is no shutdown, the laptop comes back and immediately turns itself off, takes a fraction of a second only. I tried again after shutting down acpid and got mixed results: - first time I tried (runlevel 3, X up, working on a text console), it came back alive but with the screen blank and some fairly odd noise (disk/fan spinning up, then shutting down abruptly). Screen remained blank and laptop wouldn't react to anything. - I then tried another few times, without X (either runlevel 2 or booted up in single user mode) and each time got the same behaviour as originally reported - in all these cases no acpid was running. Nor do I have powersaved. Ps does report [kacpid] but I guess that's normal? So, with or without acpid the reaction seems to be the same: disks and fan spin up when I resume and the laptop shuts itself off immediately.
You have Radeon card on your laptop. So, look at bug #3670 and bug #1721. That should get the video back. Please try in runlevel 1 or 3 first. With No X and no framebuffer CONFIGed in kernel.
I did this: - recompiled kernel without framebuffer support - telinit 3 -> so that I have no X - /etc/init.d/acpid stop In that situation, with no X, no acpid, the behaviour is exactly the same as described before: it suspends, but when I press the power button to revive it, it powers up and immediately down. The situation described in the other bugs, where it powers up but the screen is blank, is something that I only got when trying this with X up. From the scenario with no X running I don't even get that far because the laptop switches off the moment it resumes.
Why was this marked as a video issue? In most cases, the laptop directly powers off within a second or so of resuming from suspend. The situation discussed in some of the comments where it would come back but the screen remaining blank only occurs when running X. Without X and without framebuffer, the laptop would power off when resuming from suspend. Given that the power button is used to resume, it seems that somehow the power button press is reacted to immediately after it resumed - thus the laptop switches off. Let me know if there is anything you would like me to try to help resolve this.
I am having the same problem on a presario 2140us. Same series laptop. Immedietly after wakeup from suspend to mem. The system turns off. There is no shutdown, just a hard turn off immedietly after pressing the power button to turn on.
does this patch from Hiroshi Itoh help?: http://marc.theaimsgroup.com/?l=linux-kernel&m=110062740314222&w=2
Unfortunately the patch does not make a difference.
I just tried again but now with 2.6.11 instead of 2.6.10. This time the laptop turns back on without turning off, however it becomes completly unresponsive and the vga doesn't come back on. I did however do something different in that I compiled the kernel with preemptive turned on.
I tried 2.6.11.3 on my Presario 2100 and unfortunately it didn't make a difference. Laptop still powers off immediately after resume from standby.
Is this issue still present in kernel 2.6.16-rc1?
Please reopen this bug if it's still present in kernel 2.6.16.