[2.6.9-rc3] suspend-to-disk oddities

[2.6.9-rc3] suspend-to-disk oddities

[Jan De Luyck]

Hello list,

Just tried swsusp, works great, besides a few strange things:

- The suspend routine is unable to shutdown the mysqld process:

Oct  4 10:19:43 precious kernel: Stopping tasks: =================================================
Oct  4 10:19:43 precious kernel:  stopping tasks failed (1 tasks remaining)
Oct  4 10:19:43 precious kernel: Restarting tasks...<6> Strange, mysqld not stopped
Oct  4 10:19:43 precious kernel:  done

- USB subsystem is totally unworking until I reinitialise it (using /etc/init.d/hotplug restart)
- modem does not work (uses ALSA module snd_intel8x0m with a 'modem daemon'). Restarting the daemon makes things work.
- ALSA mixer volume is set to 0 for all channels (can be fixed by running /etc/init.d/alsa start at resume)

Otherwise than these small problems it works fine, from within X. Using the radeon driver.

Jan
-- 
Beauty and harmony are as necessary to you as the very breath of life.

Re: [2.6.9-rc3] suspend-to-disk oddities

[Oliver Neukum]

Am Montag, 4. Oktober 2004 11:07 schrieb Jan De Luyck:
> Just tried swsusp, works great, besides a few strange things:
> 
> - The suspend routine is unable to shutdown the mysqld process:
> 
> Oct  4 10:19:43 precious kernel: Stopping tasks: =================================================
> Oct  4 10:19:43 precious kernel:  stopping tasks failed (1 tasks remaining)
> Oct  4 10:19:43 precious kernel: Restarting tasks...<6> Strange, mysqld not stopped
> Oct  4 10:19:43 precious kernel:  done
> 
> - USB subsystem is totally unworking until I reinitialise it (using /etc/init.d/hotplug restart)

Precisely how does it fail?
Is taht after a failed suspend or after suspend?

	Regards
		Oliver

Re: [2.6.9-rc3] suspend-to-disk oddities

[Jan De Luyck]

On Monday 04 October 2004 13:31, Oliver Neukum wrote:
> Am Montag, 4. Oktober 2004 11:07 schrieb Jan De Luyck:
> > Just tried swsusp, works great, besides a few strange things:
> >
> > - The suspend routine is unable to shutdown the mysqld process:
> >
> > Oct  4 10:19:43 precious kernel: Stopping tasks:
> > ================================================= Oct  4 10:19:43
> > precious kernel:  stopping tasks failed (1 tasks remaining) Oct  4
> > 10:19:43 precious kernel: Restarting tasks...<6> Strange, mysqld not
> > stopped Oct  4 10:19:43 precious kernel:  done
> >
> > - USB subsystem is totally unworking until I reinitialise it (using
> > /etc/init.d/hotplug restart)
>
> Precisely how does it fail?

This is after a successfull suspend-resume.

It doesn't work, period. No messages in the logs, anything I plug in isn't 
reacted to, lsusb gives nothing. It's just 'not there'.

Jan

-- 
BOFH excuse #136:

Daemons loose in system.

Re: [2.6.9-rc3] suspend-to-disk oddities

[Oliver Neukum]

Am Montag, 4. Oktober 2004 13:59 schrieb Jan De Luyck:
> On Monday 04 October 2004 13:31, Oliver Neukum wrote:
> > Am Montag, 4. Oktober 2004 11:07 schrieb Jan De Luyck:
> > > Just tried swsusp, works great, besides a few strange things:
> > >
> > > - The suspend routine is unable to shutdown the mysqld process:
> > >
> > > Oct  4 10:19:43 precious kernel: Stopping tasks:
> > > ================================================= Oct  4 10:19:43
> > > precious kernel:  stopping tasks failed (1 tasks remaining) Oct  4
> > > 10:19:43 precious kernel: Restarting tasks...<6> Strange, mysqld not
> > > stopped Oct  4 10:19:43 precious kernel:  done
> > >
> > > - USB subsystem is totally unworking until I reinitialise it (using
> > > /etc/init.d/hotplug restart)
> >
> > Precisely how does it fail?
> 
> This is after a successfull suspend-resume.
> 
> It doesn't work, period. No messages in the logs, anything I plug in isn't 
> reacted to, lsusb gives nothing. It's just 'not there'.

Does "cat /proc/bus/usb/devices" give you an empty file or does it hang?
Is that modular USB or is it compiled into the kernel? OHCI or UHCI?

	Regards
		Oliver

Re: [2.6.9-rc3] suspend-to-disk oddities

[Jan De Luyck]

On Monday 04 October 2004 14:06, Oliver Neukum wrote:
> Am Montag, 4. Oktober 2004 13:59 schrieb Jan De Luyck:
> > On Monday 04 October 2004 13:31, Oliver Neukum wrote:
> > > Am Montag, 4. Oktober 2004 11:07 schrieb Jan De Luyck:
> > > > Just tried swsusp, works great, besides a few strange things:
> > > >
> > > > - The suspend routine is unable to shutdown the mysqld process:
> > > >
> > > > Oct  4 10:19:43 precious kernel: Stopping tasks:
> > > > ================================================= Oct  4 10:19:43
> > > > precious kernel:  stopping tasks failed (1 tasks remaining) Oct  4
> > > > 10:19:43 precious kernel: Restarting tasks...<6> Strange, mysqld not
> > > > stopped Oct  4 10:19:43 precious kernel:  done
> > > >
> > > > - USB subsystem is totally unworking until I reinitialise it (using
> > > > /etc/init.d/hotplug restart)
> > >
> > > Precisely how does it fail?
> >
> > This is after a successfull suspend-resume.
> >
> > It doesn't work, period. No messages in the logs, anything I plug in
> > isn't reacted to, lsusb gives nothing. It's just 'not there'.
>
> Does "cat /proc/bus/usb/devices" give you an empty file or does it hang?
> Is that modular USB or is it compiled into the kernel? OHCI or UHCI?

UHCI. I just did a test-suspend-resume, currently plugged USB devices don't work, but it does show up in the devices file. It also responds to replugging.... I don't get it.
 I had no response whatsoever earlier. Mouse doesn't work until replugged, lots of messages like this in dmesg:

Oct  4 14:16:49 precious kernel: drivers/usb/input/hid-core.c: input irq status -84 received
Oct  4 14:16:54 precious last message repeated 209 times

Jan
-- 
BOFH excuse #285:

Telecommunications is upgrading. 

Re: [2.6.9-rc3] suspend-to-disk oddities

[Rafael J. Wysocki]

On Monday 04 of October 2004 14:22, Jan De Luyck wrote:
[-- snip --]
> > Does "cat /proc/bus/usb/devices" give you an empty file or does it hang?
> > Is that modular USB or is it compiled into the kernel? OHCI or UHCI?
> 
> UHCI. I just did a test-suspend-resume, currently plugged USB devices don't 
work, but it does show up in the devices file. It also responds to 
replugging.... I don't get it.
>  I had no response whatsoever earlier. Mouse doesn't work until replugged, 
lots of messages like this in dmesg:
> 
> Oct  4 14:16:49 precious kernel: drivers/usb/input/hid-core.c: input irq 
status -84 received
> Oct  4 14:16:54 precious last message repeated 209 times

Have you tried booting with pci=routeirq?  It may help.

Greets,
RJW

-- 
- Would you tell me, please, which way I ought to go from here?
- That depends a good deal on where you want to get to.
		-- Lewis Carroll "Alice's Adventures in Wonderland"

Re: [linux-usb-devel] Re: [2.6.9-rc3] suspend-to-disk oddities

[David Brownell]

On Monday 04 October 2004 5:56 am, Rafael J. Wysocki wrote:
> On Monday 04 of October 2004 14:22, Jan De Luyck wrote:
> > 
> > UHCI. I just did a test-suspend-resume, currently plugged
> > USB devices don't  work, but it does show up in the devices
> > file. It also responds to  replugging.... I don't get it.
> >  I had no response whatsoever earlier. ...
> 
> Have you tried booting with pci=routeirq?  It may help.

I've seen strangeness on the resume path too, with devices
not responding but still being known to the kernel.  What I
found odd was that I know the OHCI hardware was reinitialized
correctly, and it was just khubd that wasn't responding.  So
it wouldn't  even respond to unplug/replug.  And this is with
known-OK IRQ settings (driver did get IRQs after resume).

There's some other stuff misbehaving still.  I'm hoping to see
the simpler stuff work first -- standby and STR behaving, wakeup
by USB keyboards not ACPI power buttons -- and at that point
the remaining problems will be specific to STD.  There are
enough wierd BIOS and power states with STD that making
that work first is very likely to bork the other stuff.

- Dave