Bug 4236

Summary:	S3 resume: no video, network or keyboard, Fujitsu Lifebook S2020
Product:	ACPI	Reporter:	Per Bjornsson (pgbjornsson)
Component:	Power-Sleep-Wake	Assignee:	Venkatesh Pallipadi (venki)
Status:	REJECTED INSUFFICIENT_DATA
Severity:	normal	CC:	acpi-bugzilla, bunk
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.13	Subsystem:
Regression:	---	Bisected commit-id:
Attachments:	Full dmesg output dmidecode output acpidmp output output from lspci -vv Contents of /proc/interrupts

Description Per Bjornsson 2005-02-21 13:02:02 UTC

Distribution: Fedora Core 3
Hardware Environment: Fujitsu Lifebook S2020 (AMD Athlon XP-M 2100+ on ATi IGP320M)
Software Environment: All 2.6.x kernels to date (currently Fedora
2.6.10-1.766_FC3, the latest Fedora build, which is close to a 2.6.10-ac11).

Problem Description:

My Fujitsu LifeBook S2020 appears to go into S3 sleep mode fine, but on
wakeup it sounds like the drives spin up and the processor fan starts,
but other than that it is completely dead and can only be shut down
with a hard reset. The screen is completely off, I can't get any
response to keyboard commands, network doesn't come back
up. Unfortunately there is no serial port on this machine so I can't
even attempt to hook up a serial console and see where it dies as far
as I can tell.

The distribution I'm running is Fedora Core 3. I haven't ever gotten a
kernel to suspend (tried with a variety of 2.6 flavors), currently
running Red Hat/Fedora 2.6.10-1.766_FC3 ( which is very close to a
current 2.6.10-ac kernel).

Hardware info: ATi Radeon IGP 320M northbridge (integrated UMA video),
AMD Athlon XP-M 2100, ALi (nowadays ULi?) southbridge.

Any tips on how to debug this? I'll attach the files requested on
the Linux-ACPI website below.

Steps to reproduce:
# echo "3" > /proc/acpi/sleep
or
# echo "mem" > /sys/power/state
(computer falls asleep)
Press power button to wake up. Listen to fans (and probably hard disk?) spin up
but fail to do anything productive with the machine.

Comment 1 Per Bjornsson 2005-02-21 13:06:38 UTC

Created attachment 4584 [details]
Full dmesg output

OK, so I'm using the binary-blob-including Atheros driver here; however,
whether the ath_pci module has ever been loaded does not matter for the
symptoms, it is reproducible without.

Comment 2 Per Bjornsson 2005-02-21 13:07:17 UTC

Created attachment 4585 [details]
dmidecode output

Comment 3 Per Bjornsson 2005-02-21 13:07:50 UTC

Created attachment 4586 [details]
acpidmp output

Comment 4 Per Bjornsson 2005-02-21 13:08:27 UTC

Created attachment 4587 [details]
output from lspci -vv

Comment 5 Per Bjornsson 2005-02-21 13:09:07 UTC

Created attachment 4588 [details]
Contents of /proc/interrupts

Comment 6 Len Brown 2005-08-17 10:36:29 UTC

please try 2.6.13
for video, please try Documentation/power/video.txt workarounds.
Also, latest BIOS or latest X video drivers may help.

Comment 7 Per Bjornsson 2005-09-26 12:04:23 UTC

Finally I could try 2.6.13 (actually Fedora 2.6.13-1.1524, which is essentially
2.6.13.2). No dice, but there's a change: not it appears that the kernel panics
on resume instead (at least that's what I think that the blinking CapsLock and
ScrollLock indicators are supposed to mean?) Since the screen is all black I
can't figure out how to get any information out of the computer at this point.
Seems effectively impossible to debug...

Comment 8 Alexey Starikovskiy 2007-06-04 12:17:04 UTC

Please check if kernel 2.6.21 still has this problem.

Comment 9 Adrian Bunk 2007-07-13 20:59:22 UTC

Please reopen this bug if it's still present with kernel 2.6.22.