Am 04.08.2013 15:17, schrieb Andreas Färber:
> Am 04.08.2013 15:05, schrieb Andreas Färber:
>> pxa2xx_i2c_init() creates a pxa2xx-i2c-slave device on a second i2c-bus,
>> which has a NULL parent device. This causes an assertion in
>> object_get_canonical_path() when accessing pxa2xx-i2c-slave's
>> "parent_bus" link<bus> property in tosa and likely other PXA2xx machines.
>>
>> Fix this by using the pxa2xx_i2c device, created just before, as parent.
>>
>> Signed-off-by: Andreas Färber <afaerber@suse.de>
> 
> CC'ing Anthony as this may be lurking elsewhere, too.
> 
> Unfortunately qtest can still only send QMP command but does not return
> the response, so we can't generically test walking the QOM composition
> tree in my proposed qom-test.

Came up with another way to test: The attached patch complementing my
earlier qom-test series (and based on some more -kernel/-bios error
avoidance for qtest) asserts that all buses except SysBus in the machine
default configurations have a parent. SysBus is known to sit on
/machine/unattached/sysbus and thus is fine, too.

Andreas

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg