Currently, shadow_power is just used to determine if the device is
active or suspended. Is there a function is_usb_device_suspended() or
something along the lines of that? I see that the usb_device has a
'state' variable which could be 'USB_STATE_SUSPENDED' which is what
shadow_power is reflecting. I didn't know if that was considered
"private" and should not use it. Can I?
When the suspend(), reset_resume(), and resume() functions (in the
usb_driver struct) execute the actual device is currently non
suspended right? So the usb_device->state != USB_STATE_SUSPENDED ?
If that is the case, it should eliminate the use of shadow_power.
Harry Metzger
On Tue, Mar 9, 2010 at 10:21 AM, Oliver Neukum <oliver@xxxxxxxxxx> wrote:
> Am Montag, 8. März 2010 16:15:14 schrieb Harrison Metzger:
>> Also, the old logic would call usb_autopm_put_interface() if
>> shadow_power == 1 and the user requested the display to be off ("echo
>> 0 > /sys/$DEVICE/powered"). If usb autosuspend was disabled the
>> suspend() function would never be called, and the shadow_power
>> variable would still be 1. So the user could "echo 0" multiple times
>> (even though the display is already off) and this would really screw
>> up the reference count (it would be negative from calling
>> usb_autopm_put_interface() multiple times).
>>
>> This is why there is a separate variable to track if we have called
>> usb_autopm_get/put_interface().
>
> OK, why not remove shadow_power?
>
> Regards
> Oliver
>
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
