On Sat, Sep 22, 2012 at 11:48 PM, Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote:
>
> This is basically okay. I have a couple of suggestions.
Thanks for your suggestion.
>
>> @@ -3192,6 +3214,15 @@ static int hub_suspend(struct usb_interface *intf, pm_message_t msg)
>>
>> /* stop khubd and related activity */
>> hub_quiesce(hub, HUB_SUSPEND);
>> +
>> + if (PMSG_IS_AUTO(msg) && hub->quirk_check_port_auto_suspend) {
>
> The correct condition is hdev->do_remote_wakeup != 0, not PMSG_IS_AUTO.
Yes, but the returned failure will be ignored for non-PMSG_IS_AUTO
cases, so it is better to just check ports for auto suspend situation.
>
>> + /* check if there are changes pending on hub ports */
>> + if (check_ports_changed(hub)) {
>> + hub_activate(hub, HUB_RESUME);
>> + return -EBUSY;
>> + }
>> + }
>
> If the new code is moved up before the hub_quiesce() call then you
> won't need to call hub_activate().
Looks fine, will do it.
Thanks,
--
Ming Lei
--
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
