Bug 2319

Summary:	resume cause reboot - via chipset
Product:	ACPI	Reporter:	Enrico Manzini (wiselynx.naima)
Component:	Power-Sleep-Wake	Assignee:	Shaohua (shaohua.li)
Status:	REJECTED DUPLICATE
Severity:	normal	CC:	acpi-bugzilla, amilo
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.4-gentoo-r1	Subsystem:
Regression:	---	Bisected commit-id:
Attachments:	output of `lspci -v` output of `lsmod` Hiro's patch

Description Enrico Manzini 2004-03-17 02:24:33 UTC

Distribution: Gentoo

Hardware Environment: Fujutsu Siemens Amilo M6800 (PIV 1700). Output of `lspci
-v` is in attachment.

Software Environment: Output of `lsmod` is in attachment. The kernel is loaded
with the following parameters: 
root=/dev/hda5 vga=791 acpi_sleep=s3_mode

Problem Description: after suspending to ram and succesfully getting the system
up, general failures happen. The HD activity LED is stays fixed on, and the
system hangs when I do anything. I could experience various behaviours:
sometimes I get a kernel Oops, once the system beginned a shutdown with the
command `lsmod` (?!), more often the systems just hangs with the first command I
give.

Steps to reproduce: send the system in suspend-to-ram with `echo -n "mem" >
/sys/power/state` (but things are the same with `echo 3 > /proc/acpi/sleep`).
The system goes in suspend mode. Pressing the power button wakes up the system
within a few seconds, but when I try anything, bad stuff happen.

I had the same results with both kernel 2.6.1 and 2.6.3-r1

Comment 1 Enrico Manzini 2004-03-17 02:25:22 UTC

Created attachment 2353 [details]
output of `lspci -v`

Comment 2 Enrico Manzini 2004-03-17 02:26:03 UTC

Created attachment 2354 [details]
output of `lsmod`

Comment 3 Len Brown 2004-11-03 18:53:07 UTC

same with 2.6.9?

Comment 4 Len Brown 2005-01-03 20:14:33 UTC

please re-open if this issue exists in 2.6.10 or later.

Comment 5 Enrico Manzini 2005-01-16 07:05:42 UTC

sorry for the long delay, I had my laptop down for repairs for a couple of months

now that I have it back working, I've tested a 2.6.10-r4 kernel from gentoo. the
problem now is that system goes in suspend mode, but when I resume, after asking
for BIOS passwd, reboots instead of resuming.

Comment 6 Enrico Manzini 2005-01-18 02:45:24 UTC

Some other news. I've tried also a gentoo-dev 2.6.10-r5 kernel (which is in "~"
mode, which should mean "unstable"). 

here the effect is different again, I can suspend but after resume, just after
typing the BIOS passwd, I get a yellow "Linu" writing, and the system hags badly
(no response, I have to hard reboot).

Comment 7 Shaohua 2005-01-18 17:11:56 UTC

Created attachment 4432 [details]
Hiro's patch

Great to know you can see 'Linu' on the screen.
Could you please try attached patch? It's from Hiroshi Itoh <hiroit AT
mcn.ne.jp>

Comment 8 Enrico Manzini 2005-01-19 04:14:53 UTC

tried the patch (note: one hunk did not succed, I had to add one line manually,
but I think it's not a problem)

the system now wakes up, and immediately reboots :P

Comment 9 Shaohua 2005-08-18 00:10:35 UTC

We find several similar issues in VIA chipset based system, unknown the 
reason :(.

Comment 10 Shaohua 2005-10-18 23:41:13 UTC


*** This bug has been marked as a duplicate of 2347 ***