[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <0835B3720019904CB8F7AA43166CEEB2F3AC58@RTITMBSV03.realtek.com.tw>
Date: Thu, 23 Jul 2015 09:55:41 +0000
From: Hayes Wang <hayeswang@...ltek.com>
To: Oliver Neukum <oneukum@...e.com>
CC: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
nic_swsd <nic_swsd@...ltek.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>
Subject: RE: [PATCH net 2/3] r8152: fix remote wakeup
> Hi,
>
> this is most likely wrong. Usbcore does check for a device's ability to
> do remote wakeup and will block a runtime suspend if it detects that
> a remote wakeup would be required but the device cannot deliver.
> (static int autosuspend_check())
>
> So by removing the flag in the probe() method means that devices will
> suspend during operations without remote wakeup requested. Thus an
> incoming packet cannot wake them up.
>
> If you remove setting the flag on probe() you need to set it at open()
> [and reset on close()], as devices which cannot do remote wakeup must
> only be suspended when they are down.
Hi,
I don't think I understand your description clearly. My idea is that if the
device doesn't support wakeup, we don't need set " needs_remote_wakeup".
We allow the device could be suspended and couldn't be waked up by
incoming packet. The system could be waked up by other methods except
by the device.
I don't understand why I have to set it at open() and reset it at close(). And
why must the device only be suspended when it is down?
Best Regards,
Hayes
Powered by blists - more mailing lists