Skip to contentSkip to conversation

Could not suspend device [VIA UHCI USB controller]: error -22

DISCUSSION Michael Tokarev 3 participants 12 messages Dec 4, 2005 – Dec 7, 2005

lore
Share & tools
lore
#0 lore raw

When I try to "standby" (echo standby > /sys/power/state) a 2.6.14 system running on a VIA C3-based system with VIA chipset (suspend to disk never worked on this system -- as stated on swsusp website it's due to the lack of some CPU instruction on this CPU [but winXP suspends to disk on this system just fine]), it immediately comes back, with the above error message:

Stopping tasks: =====================================================================|
ACPI: PCI interrupt for device 0000:00:07.5 disabled
Could not suspend device 0000:00:07.3: error -22
ACPI: PCI Interrupt 0000:00:07.5[C] -> Link [LNKC] -> GSI 12 (level, low) -> IRQ 12
ACPI: PCI Interrupt 0000:00:0a.0[A] -> Link [LNKC] -> GSI 12 (level, low) -> IRQ 12
eth0: link up, 100Mbps, full-duplex, lpa 0x45E1
Some devices failed to suspend
Restarting tasks... done

Here's the PCI devices:

00:00.0 Host bridge: VIA Technologies, Inc. VT8601 [Apollo ProMedia] (rev 05)
00:01.0 PCI bridge: VIA Technologies, Inc. VT8601 [Apollo ProMedia AGP]
00:07.0 ISA bridge: VIA Technologies, Inc. VT82C686 [Apollo Super South] (rev 40)
00:07.1 IDE interface: VIA Technologies, Inc. VT82C586A/B/VT82C686/A/B/VT823x/A/C PIPC Bus Master IDE (rev 06) 00:07.2 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 1a) 00:07.3 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 1a) 00:07.4 Bridge: VIA Technologies, Inc. VT82C686 [Apollo Super ACPI] (rev 40)
00:07.5 Multimedia audio controller: VIA Technologies, Inc. VT82C686 AC97 Audio Controller (rev 50) 00:0a.0 Multimedia controller: Philips Semiconductors SAA7134 (rev 01)
00:0d.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10) 01:00.0 VGA compatible controller: Trident Microsystems CyberBlade/i1 (rev 6a)

(note it's the second USB interface which fails).
And here's the USB device list:

Bus 002 Device 002: ID 07cc:0301 Carry Computer Eng., Co., Ltd 6-in-1 Card Reader Bus 002 Device 001: ID 0000:0000
Bus 001 Device 003: ID 0458:0036 KYE Systems Corp. (Mouse Systems) Bus 001 Device 001: ID 0000:0000

This machine was never able to go to standby mode completely - before 2.6.14, it performed resume after several secounds of sleeping. Now it does not even try to sleep anymore.

Also, "suspend to mem" does just nothing, -- the same as "suspend to disk" (but for disk, it never worked at all as stated above).

Any hints on where to look at?

Thanks.

/mjt

#1 lore raw
Quoted 7 lines · Michael Tokarev

When I try to "standby" (echo standby > /sys/power/state) a 2.6.14 system running on a VIA C3-based system with VIA chipset (suspend to disk never worked on this system -- as stated on swsusp website it's due to the lack of some CPU instruction on this CPU [but winXP suspends to disk on this system just fine]), it immediately comes back, with the above error message:

Can you try 2.6.15-rc4 or newer to see if that fixes this issue for you?

thanks,

greg k-h

#2 lore raw
Quoted 11 lines · Greg KH
Quoted 7 lines · Michael Tokarev

When I try to "standby" (echo standby > /sys/power/state) a 2.6.14 system running on a VIA C3-based system with VIA chipset (suspend to disk never worked on this system -- as stated on swsusp website it's due to the lack of some CPU instruction on this CPU [but winXP suspends to disk on this system just fine]), it immediately comes back, with the above error message:

Can you try 2.6.15-rc4 or newer to see if that fixes this issue for you?

Yes, 2.6.15-rc4 restores previous functionality - the error in $subject is now gone, and it seems the system goes to standby as it should, without errors and 'standby process interruptions'. Thanks.

With the only problem which was here all the time - it comes "back to C" after less a secound all the disks/monitor/etc are placed into sleep mode.. Ie,

..preparing for standby...
..hdd stops spinning..
..monitor is turned off.. ..less-than-a-secound-pause..
Back to C!
..the system goes back, restoring interrupts etc...

I tried various 'wakeup' settings in bios, incl. turning everything off in that menu - no difference.

The same behaviour is shown by all 2.6 kernels I tried so far (since 2.6.6 or so).

BTW.. It looks like i should change the $subject now.. ;)

/mjt

#3 lore raw
Quoted 29 lines · Michael Tokarev
Quoted 11 lines · Greg KH
Quoted 7 lines · Michael Tokarev

When I try to "standby" (echo standby > /sys/power/state) a 2.6.14 system running on a VIA C3-based system with VIA chipset (suspend to disk never worked on this system -- as stated on swsusp website it's due to the lack of some CPU instruction on this CPU [but winXP suspends to disk on this system just fine]), it immediately comes back, with the above error message:

Can you try 2.6.15-rc4 or newer to see if that fixes this issue for you?

Yes, 2.6.15-rc4 restores previous functionality - the error in $subject is now gone, and it seems the system goes to standby as it should, without errors and 'standby process interruptions'. Thanks.

With the only problem which was here all the time - it comes "back to C" after less a secound all the disks/monitor/etc are placed into sleep mode.. Ie,

..preparing for standby...
..hdd stops spinning..
..monitor is turned off.. ..less-than-a-secound-pause..
Back to C!
..the system goes back, restoring interrupts etc...

Quoted 5 lines

I tried various 'wakeup' settings in bios, incl. turning everything off in that menu - no difference.

The same behaviour is shown by all 2.6 kernels I tried so far (since 2.6.6 or so).

Does normal "suspend" work for you on this machine (echoing "disk" to that sysfs file?)

I'd suggest creating a bugzilla.kernel.org entry for this new problem.

thanks,

greg k-h

#4 lore raw
Quoted 1 line · Greg KH

On Sun, Jan 02, 2000 at 11:28:12AM +0300, Michael Tokarev wrote:

[]

Quoted 11 lines
Quoted 11 lines

With the only problem which was here all the time - it comes "back to C" after less a secound all the disks/monitor/etc are placed into sleep mode.. Ie,

..preparing for standby...
..hdd stops spinning..
..monitor is turned off.. ..less-than-a-secound-pause..
Back to C!
..the system goes back, restoring interrupts etc...

Quoted 8 lines
Quoted 5 lines

I tried various 'wakeup' settings in bios, incl. turning everything off in that menu - no difference.

The same behaviour is shown by all 2.6 kernels I tried so far (since 2.6.6 or so).

Does normal "suspend" work for you on this machine (echoing "disk" to that sysfs file?)

No, it never worked (disk or even mem) because - as i've read - linux needs some CPU instruction (I don't remember which) which isn't implemented on this CPU (neverless, win is pretty happy -- doing suspend/hibernate.. ;)

Quoted 1 line

I'd suggest creating a bugzilla.kernel.org entry for this new problem.

Oh well. Ok, I'll do. Thank you.

/mjt

#8 lore raw
In reply to Michael Tokarev

Hi!

Quoted 5 lines

When I try to "standby" (echo standby > /sys/power/state) a 2.6.14 system running on a VIA C3-based system with VIA chipset (suspend to disk never worked on this system -- as stated on swsusp website it's due to the lack of some CPU instruction on this CPU [but winXP suspends to disk

Lack of 4MB does not mean suspend is impossible, it is only uglier and harder to do than usually. See x86-64 -- it faces similar problem. But I do not have C3 here, and so I can't easily fixes, and there are not enough C3 users to fix it themselves. [Ouch and cc: me on suspend problems.]
Pavel

Signature · 1 line
Thanks, Sharp!
#5 lore raw
In reply to Michael Tokarev

Hi!

Quoted 7 lines

..preparing for standby...
..hdd stops spinning..
..monitor is turned off.. ..less-than-a-secound-pause..
Back to C!
..the system goes back, restoring interrupts etc...

Quoted 5 lines

I tried various 'wakeup' settings in bios, incl. turning everything off in that menu - no difference.

The same behaviour is shown by all 2.6 kernels I tried so far (since 2.6.6 or so).

Try ACPI wakeup settings, and ask on ACPI lists. Unfortunately noone really cares about standby these days.
Pavel

Signature · 1 line
Thanks, Sharp!
#9 lore raw
In reply to Michael Tokarev

Hi!

Quoted 2 lines

Also, "suspend to mem" does just nothing, -- the same as "suspend to disk" (but for disk, it never worked at all as stated above).

Can you quote exact messages? Suspend to mem should not have problems without 4MB pages, as it does not do any pagetables related magic. If it does include same check, it is bug and should be easy to fix.

Pavel

Signature · 1 line
Thanks, Sharp!
#10 lore raw
Quoted 9 lines · Pavel Machek

Hi!

Quoted 2 lines

Also, "suspend to mem" does just nothing, -- the same as "suspend to disk" (but for disk, it never worked at all as stated above).

Can you quote exact messages? Suspend to mem should not have problems without 4MB pages, as it does not do any pagetables related magic. If it does include same check, it is bug and should be easy to fix.

Hmm.. There's no messages, no at all.

echo mem > /sys/power/state

does exactly nothing. When writing 'suspend' to that file, the system at least tries to do something (now with 2.6.15-rc4 it completes the syspend procedure; but it wakes up again in a secound or two), with all the messages et al, but not 'mem' or 'disk' - no messages at all.

/mjt

#6 lore raw
In reply to Pavel Machek
Quoted 10 lines · Pavel Machek

Hi!

Quoted 7 lines

..preparing for standby...
..hdd stops spinning..
..monitor is turned off.. ..less-than-a-secound-pause..
Back to C!
..the system goes back, restoring interrupts etc...

Quoted 8 lines
Quoted 5 lines

I tried various 'wakeup' settings in bios, incl. turning everything off in that menu - no difference.

The same behaviour is shown by all 2.6 kernels I tried so far (since 2.6.6 or so).

Try ACPI wakeup settings, and ask on ACPI lists. Unfortunately noone really cares about standby these days.

Which "ACPI wakeup settings" did you mean? In BIOS or in kernel?

In my BIOS, there's a page called "Power management" (or something of that sort), which, among other things, contains a section "Wakeup devices" or "Wakeup events" - i tried to turn them all on and off, all at once and in alot of different combinations - makes no real difference, the system wakes up in a secound or two regardless.

Too bad no one cares about standby.. :(
I've several of those systems, and I love them for their quiet operation. The only problem for me is the system startup time (about 3 minutes) and applications startup time (due to the empty filesystem cache) -- it'd be very nice to be able to suspend the system somehow instead of turning it off... Now, suspend to disk does not work at all (that 4M pages stuff on a VIA C3 CPU), suspend to mem does not work either, and "normal" standby, while works, triggers a wakeup almost immediately. So, in short, no suspend at all...

But Ok. Lemme see what's going on around that file in the code..

/mjt

#11 lore raw
In reply to Michael Tokarev

Hi!

Quoted 8 lines
Quoted 7 lines
Quoted 2 lines

Also, "suspend to mem" does just nothing, -- the same as "suspend to disk" (but for disk, it never worked at all as stated above).

Can you quote exact messages? Suspend to mem should not have problems without 4MB pages, as it does not do any pagetables related magic. If it does include same check, it is bug and should be easy to fix.

Quoted 6 lines

Hmm.. There's no messages, no at all.

echo mem > /sys/power/state

does exactly nothing. When writing 'suspend' to that file, the system

I think you mean 'standby'?

Quoted 3 lines

at least tries to do something (now with 2.6.15-rc4 it completes the syspend procedure; but it wakes up again in a secound or two), with all the messages et al, but not 'mem' or 'disk' - no messages at all.

You are hitting something else than missing 4MB pages:

static inline int
arch_prepare_suspend(void)
{
/* If you want to make non-PSE machine work, turn off paging
in swsusp_arch_suspend. swsusp_pg_dir should have identity
mapping, so

           it could work...  */
        if (!cpu_has_pse) {
                printk(KERN_ERR "PSE is required for swsusp.\n");
                return -EPERM;
        }
        return 0;
}

...so just insert printks into the code to find out what is going on...

Pavel

Signature · 1 line
Thanks, Sharp!
#7 lore raw
In reply to Michael Tokarev
Quoted 11 lines · Michael Tokarev
Quoted 8 lines · Pavel Machek
Quoted 5 lines

I tried various 'wakeup' settings in bios, incl. turning everything off in that menu - no difference.

The same behaviour is shown by all 2.6 kernels I tried so far (since 2.6.6 or so).

Try ACPI wakeup settings, and ask on ACPI lists. Unfortunately noone really cares about standby these days.

Which "ACPI wakeup settings" did you mean? In BIOS or in kernel?

/proc/acpi/wakeup. And do ask on acpi mailing lists.

Quoted 10 lines

Too bad no one cares about standby.. :(
I've several of those systems, and I love them for their quiet operation. The only problem for me is the system startup time (about 3 minutes) and applications startup time (due to the empty filesystem cache) -- it'd be very nice to be able to suspend the system somehow instead of turning it off... Now, suspend to disk does not work at all (that 4M pages stuff on a VIA C3 CPU), suspend to mem does not work either, and "normal" standby, while works, triggers a wakeup almost immediately. So, in short, no suspend at all...

Well, printk time. Find out what is wrong. Noone else can because lack of hardware.
Pavel

Signature · 1 line
Thanks, Sharp!