Suspend and Resume traces
zecke at openmoko.org
Thu Jun 12 12:43:00 CEST 2008
On Thursday 12 June 2008 09:29:12 Andy Green wrote:
> Somebody in the thread at some point said:
> | Things are starting to look way better with suspend/resume. I have a
> | frew traces for you.
> Good! It's been stable with defconfig for many weeks but I know it
> isn't for other configs. Maybe we can backtrace our way to
> understanding it.
> | <4>[<c021fc7c>] (pcf50633_usb_curlim_set+0x0/0xdc) from [<c006caa0>]
> | (__gta02_udc_vbus_draw+0x34/0x44)
> Great thanks, this is the async notification that host PC has enumerated
> us at 500mA power, it's randomly coming in before we resumed pcf50633...
> that is definitely something that would make trouble with non-charging
> GTA02 after suspend / resume.
> The trace spew was all from the one event which tries to go on the bus
> multiple times. I never see this symptom here but this is definitely a
> genuine issue for all configs to be fixed.
Wow, that might explain why I sometimes thought that pluggin in the usb
cable/removing it, or the charger had something to do with my device not
More information about the openmoko-kernel