Bug 13735
| Summary: | system fails to suspend ram | ||
|---|---|---|---|
| Product: | Power Management | Reporter: | Brent Millare (puttanginyournose) |
| Component: | Hibernation/Suspend | Assignee: | Zhang Rui (rui.zhang) |
| Status: | CLOSED INSUFFICIENT_DATA | ||
| Severity: | normal | CC: | rui.zhang, tedtheologian, yakui.zhao |
| Priority: | P1 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.30 | Subsystem: | |
| Regression: | No | Bisected commit-id: | |
| Attachments: |
dmesg after failure
lspci -vv output Log after console mode acpidump |
||
Created attachment 22243 [details]
lspci -vv output
Hardware is a Dell Vostro 1400 laptop
Also suspend worked for sure in 2.6.26, I had the same problem in >2.6.29, may 28 too but I can't remember. Hi, Brent
Will you please boot the system into the console mode and see whether the box can enter the suspend state?
From the log it seems that one user space application can't be freezed.
Thanks.
Created attachment 22328 [details]
Log after console mode
Hello,
Should I try single user mode? It still fails after closing X, xdm, and removing nvidia module.
Actually, I just tried when in single user mode and it still fails with the same error. hi, Brent
Sorry for the late response.
Will you please try the boot option of "maxcpus=1" and see whether it still fails?
Thanks.
Hi ykzhao, IT WORKS! (With those args) Hopefully that will be the key to fixing it. Best, Brent Hi, Brent
Thanks for your test.
will you please attach the output of acpidump?
Thanks.
Created attachment 23123 [details]
acpidump
does the problem still exist in kernel 2.6.31? we have two acpi patches shipped recently, and now all the SMI calls invoked in AML code are scheduled on CPU0. The problem still exists for me using 2.6.31-r2 and 2.6.31-r4. Now the system hangs with the flashing underscore instead of resuming back into X please just run "echo disk > /sys/power/state" as root user to hibernate instead of "sudo hibernate-ram --force". If the problem still exist, please re-try latest upstream kernel, say 2.6.32 on 2.6.31 echo mem > /sys/power/state works now although I need to be shutdown xdm and upon resuming I lose the backlight and need to blind restart xdm to resume normal operation will try 2.6.32 when have time Any idea on potential solutions or pointers to suspending and resuming a full X session? Im on fc12 64bit Linux laptop 2.6.31.12-174.2.3.fc12.x86_64 #1 SMP Mon Jan 18 19:52:07 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux Vostro 1520. I am having this same problem, can't suspend to ram. if I do maxcpus=1, since I have core duo 2, won't that cause performance problems? Changing to maxcpus=1 was used to help identify the problem, not to be confused with a solution. Changing it to 1 will deactivate the other core and result in less multithreaded performance. hey, all, It's great that kernel bugzilla is back. can anyone of you verify if the problem still exists in the latest upstream kernel please? bug closed as there is no response from the bug reporter. |
Created attachment 22242 [details] dmesg after failure To reproduce: 1. Boot computer 2. sudo hibernate-ram --force Computer attempts to suspend but fails and resumes as normal. An error message to console reads "/bin/echo: write error: Resource temporarily unavailable"