[Sugar-devel] New SoaS-on-XO-1 build done: soasxo57

[Paul Fox]

martin wrote:
 > 
 > - Power management is disabled (XO is configured not to sleep since
 >   rtcwake never wakes up)
 > 

martin -- do you have any ideas on what's going on here?  what
are your symptoms?

the first couple of times i tried it (first verifying that sys and
hwclock times matched, and TZ was UTC, and adjtime contained no LOCAL
line) the system never woke up, and had to be rebooted.

then i had a boot where s/r worked a couple of times, and then i
started getting symptoms i've seen on my own builds of 2.6.30, which
includes these USB errors during resume:

[ 1624.265033] ehci_hcd 0000:00:0f.5: port 1 reset error -110
[ 1624.265066] hub 2-0:1.0: hub_port_status failed (err = -32)
[ 1624.478413] ehci_hcd 0000:00:0f.5: port 1 reset error -110
[ 1624.478445] hub 2-0:1.0: hub_port_status failed (err = -32)
[ 1624.688373] ehci_hcd 0000:00:0f.5: port 1 reset error -110
[ 1624.688404] hub 2-0:1.0: hub_port_status failed (err = -32)
[ 1624.898335] ehci_hcd 0000:00:0f.5: port 1 reset error -110
[ 1624.898366] hub 2-0:1.0: hub_port_status failed (err = -32)
[ 1625.108304] ehci_hcd 0000:00:0f.5: port 1 reset error -110
[ 1625.108338] hub 2-0:1.0: hub_port_status failed (err = -32)
[ 1625.108358] hub 2-0:1.0: Cannot enable port 1.  Maybe the USB cable is

which eventually leads to:
[ 1627.808814] pm_op(): usb_dev_resume+0x0/0xf [usbcore] returns -19
[ 1627.808838] PM: Device 2-1 failed to resume: error -19

after a few repeats of that cycle (each ending with getting control
back in userspace), i get:
[ 1786.436998] PM: Entering mem sleep
[ 1786.437014] Suspending console(s) (use no_console_suspend to debug)
[ 1786.476836] hub 2-0:1.0: suspend error -16
[ 1786.477019] pm_op(): usb_dev_suspend+0x0/0xf [usbcore] returns -16
[ 1786.477043] PM: Device usb2 failed to suspend: error -16
[ 1786.477060] PM: Some devices failed to suspend
[ 1786.609462] pm_op(): usb_dev_resume+0x0/0xf [usbcore] returns -19
[ 1786.609486] PM: Device 2-4 failed to resume: error -19
[ 1786.609934] PM: Finishing wakeup.
[ 1786.609950] Restarting tasks ... done.

at this point rtcwake can't write to /sys/power/state without getting EBUSY
back.

as i say, these last symptoms i've seen before with USB on my laptop.  but
i've never before seen the initial symptom, of not returning from the
suspend at all.

paul
p.s. the serial console, controlled by /etc/event.d/ttyS0, doesn't start
on it's own.  "initctl start ttyS0" gets it going.  not sure what's wrong
with the startup conditions.

=---------------------
 paul fox, pgf at laptop.org