Re: [PATCH 4.19 014/110] usb: xhci: Prevent bus suspend if a port connect change or polling state is detected

[Thomas Zeitlhofer]

On Thu, Dec 13, 2018 at 02:24:14PM +0200, Mathias Nyman wrote:
> On 13.12.2018 09:36, Greg Kroah-Hartman wrote:
> > On Wed, Dec 12, 2018 at 11:53:34PM +0100, Thomas Zeitlhofer wrote:
> > > Hello,
> > > 
> > > On Thu, Nov 29, 2018 at 03:11:45PM +0100, Greg Kroah-Hartman wrote:
> > > > 4.19-stable review patch.  If anyone has any objections, please let me
> > > > know.
> > > > 
> > > > ------------------
> > > > 
> > > > From: Mathias Nyman <mathias.nyman@xxxxxxxxxxxxxxx>
> > > > 
> > > > commit 2f31a67f01a8beb22cae754c53522cb61a005750 upstream.
> > > [...]
> > > 
> > > on a current Thinkpad X1 Yoga, this breaks resume from hibernate such
> > > that opening the lid has (in the regular use case, see below) no effect
> > > any more:
> > > 
> > > The system is configured to hibernate when the lid is closed. So, the
> > > expected behavior, which is restored by reverting this patch, is:
> > > 
> > > 	close lid => system hibernates
> > > 	open lid => system resumes
> > > 
> > > With this patch, the following two cases are observed:
> > > 
> > > 	1)
> > > 		close lid => system hibernates
> > > 		open lid => system stays off
> > > 		press power button => system boots and resumes
> > > 
> > > 	2)
> > > 		# systemctl hibernate => system hibernates
> > > 		close lid
> > > 		open lid => system resumes
> > > 
> > 
> > So this is a problem in Linus's tree?  If so, let's work to get this
> > fixed there first.
> > 
> > If not, then we have other issues :)
> > 
> 
> That patch incorrectly reacts to USB2 polling states as well,
> which could cause issues like this.
> 
> Does applying the below code help?
> 
> diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c
> index 94aca1b..01b5818 100644
> --- a/drivers/usb/host/xhci-hub.c
> +++ b/drivers/usb/host/xhci-hub.c
> @@ -1507,7 +1507,8 @@ int xhci_bus_suspend(struct usb_hcd *hcd)
>                 portsc_buf[port_index] = 0;
>                 /* Bail out if a USB3 port has a new device in link training */
> -               if ((t1 & PORT_PLS_MASK) == XDEV_POLLING) {
> +               if ((hcd->speed >= HCD_USB3) &&
> +                   (t1 & PORT_PLS_MASK) == XDEV_POLLING) {
>                         bus_state->bus_suspended = 0;
>                         spin_unlock_irqrestore(&xhci->lock, flags);
>                         xhci_dbg(xhci, "Bus suspend bailout, port in polling\n");

Yes, this fixes the problem.

Thanks,

Thomas