Skip to contentSkip to conversation

problem with suspending linux-2.6.12-rc1

DISCUSSION Pavel Machek 2 participants 2 messages Mar 28, 2005 – Mar 31, 2005 2 messages not in the archive Input drivers

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

#2 lore raw
Quoted 13 lines · Ulrich Lauther
Quoted 13 lines

Hi!

Quoted 7 lines

since upgrading from 2.6.11 to 2.6.12-rc1 software suspend doesn't work anymore for me:
The last I see when suspending (echo 4 > /proc/acpi/sleep) is a message refering to eth0, then when "writing to swap space" should appear, the system stops.

Please let me know if you need more information.

Standard debugging: try again with init=/bin/bash and minimal set of drivers, and get me last few messages...

Quoted 7 lines
Quoted 1 line

Pavel

with init=/bin/bash it works.

Further experimentation showed that the problem occurs only when X11 is running. I then tried to work without the evdev driver (and without the alps module in X11, that needs evdev), but that didn't make a difference).

Do you have 3D acceleration enabled in X? Does it work if you have X running but on the background console? Does it work if you quit X then suspend from console?
Pavel

Signature · 2 lines
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!
#3 lore raw
Quoted 6 lines · Pavel Machek
Quoted 5 lines · Ulrich Lauther
Quoted 5 lines
Quoted 5 lines

since upgrading from 2.6.11 to 2.6.12-rc1 software suspend doesn't work anymore for me:
The last I see when suspending (echo 4 > /proc/acpi/sleep) is a message refering to eth0, then when "writing to swap space" should appear, the system stops.

This sounds similar to the swsusp hang I started seeing when I upgraded from 2.6.11 to 2.6.12-rc1; my touchpad was failing to reinitialize, so input was calling hotplug which tried to exec /sbin/hotplug, and the exec() blocked because userland was stopped. If you do Sysrq-T and see a process with a stack trace like
wait_for_completion
call_usermodehelper
kobject_hotplug
kobject_del
class_device_del
class_device_unregister
mousedev_disconnect
input_unregister_device
then you're seeing the same kind of bug.

Try the patch below from Dmitry.

-andy

===================================================================

Input: serio - do not attempt to immediately disconnect port if
resume failed, let kseriod take care of it. Otherwise we may attempt to unregister associated input devices which will generate hotplug events which are not handled well during swsusp.

 serio.c |    1 -
 1 files changed, 1 deletion(-)
drivers/input/serio/serio.c +0 -1
Index: dtor/drivers/input/serio/serio.c
===================================================================
@@ -779,7 +779,6 @@ static int serio_resume(struct device *d
 	struct serio *serio = to_serio_port(dev);
 
 	if (!serio->drv || !serio->drv->reconnect || serio->drv->reconnect(serio)) {
-		serio_disconnect_port(serio);
 		/*
 		 * Driver re-probing can take a while, so better let kseriod
 		 * deal with it.