usb: force warm reset to break link re-connect livelock

Resuming a powered down port sometimes results in the port state being
stuck in the training sequence.

 hub 3-0:1.0: debounce: port 1: total 2000ms stable 0ms status 0x2e0
 port1: can't get reconnection after setting port  power on, status -110
 hub 3-0:1.0: port 1 status 0000.02e0 after resume, -19
 usb 3-1: can't resume, status -19
 hub 3-0:1.0: logical disconnect on port 1

In the case above we wait for the port re-connect timeout of 2 seconds
and observe that the port status is USB_SS_PORT_LS_POLLING (although it
is likely toggling between this state and USB_SS_PORT_LS_RX_DETECT).
This is indicative of a case where the device is failing to progress the
link training state machine.

It is resolved by issuing a warm reset to get the hub and device link
state machines back in sync.

 hub 3-0:1.0: debounce: port 1: total 2000ms stable 0ms status 0x2e0
 usb usb3: port1 usb_port_runtime_resume requires warm reset
 hub 3-0:1.0: port 1 not warm reset yet, waiting 50ms
 usb 3-1: reset SuperSpeed USB device number 2 using xhci_hcd

After a reconnect timeout when we expect the device to be present, force
a warm reset of the device.  Note that we can not simply look at the
link status to determine if a warm reset is required as any of the
training states USB_SS_PORT_LS_POLLING, USB_SS_PORT_LS_RX_DETECT, or
USB_SS_PORT_LS_COMP_MOD are valid states that do not indicate the need
for warm reset by themselves.

Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Cc: Vincent Palatin <vpalatin@chromium.org>
Cc: Lan Tianyu <tianyu.lan@intel.com>
Cc: Ksenia Ragiadakou <burzalodowa@gmail.com>
Cc: Vivek Gautam <gautam.vivek@samsung.com>
Cc: Douglas Anderson <dianders@chromium.org>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Sunil Joshi <joshi@samsung.com>
Cc: Hans de Goede <hdegoede@redhat.com>
Acked-by: Julius Werner <jwerner@chromium.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/usb/core/hub.c

@@ -2587,13 +2587,20 @@ static int hub_port_reset(struct usb_hub *hub, int port1,
 /* Is a USB 3.0 port in the Inactive or Compliance Mode state?
  * Port worm reset is required to recover
  */
-static bool hub_port_warm_reset_required(struct usb_hub *hub, u16 portstatus)
+static bool hub_port_warm_reset_required(struct usb_hub *hub, int port1,
+		u16 portstatus)
 {
-	return hub_is_superspeed(hub->hdev) &&
-		(((portstatus & USB_PORT_STAT_LINK_STATE) ==
-		  USB_SS_PORT_LS_SS_INACTIVE) ||
-		 ((portstatus & USB_PORT_STAT_LINK_STATE) ==
-		  USB_SS_PORT_LS_COMP_MOD)) ;
+	u16 link_state;
+
+	if (!hub_is_superspeed(hub->hdev))
+		return false;
+
+	if (test_bit(port1, hub->warm_reset_bits))
+		return true;
+
+	link_state = portstatus & USB_PORT_STAT_LINK_STATE;
+	return link_state == USB_SS_PORT_LS_SS_INACTIVE
+		|| link_state == USB_SS_PORT_LS_COMP_MOD;
 }
 
 static int hub_port_wait_reset(struct usb_hub *hub, int port1,
@@ -2630,7 +2637,7 @@ static int hub_port_wait_reset(struct usb_hub *hub, int port1,
 	if ((portstatus & USB_PORT_STAT_RESET))
 		return -EBUSY;
 
-	if (hub_port_warm_reset_required(hub, portstatus))
+	if (hub_port_warm_reset_required(hub, port1, portstatus))
 		return -ENOTCONN;
 
 	/* Device went away? */
@@ -2730,9 +2737,10 @@ static int hub_port_reset(struct usb_hub *hub, int port1,
 		if (status < 0)
 			goto done;
 
-		if (hub_port_warm_reset_required(hub, portstatus))
+		if (hub_port_warm_reset_required(hub, port1, portstatus))
 			warm = true;
 	}
+	clear_bit(port1, hub->warm_reset_bits);
 
 	/* Reset the port */
 	for (i = 0; i < PORT_RESET_TRIES; i++) {
@@ -2769,7 +2777,8 @@ static int hub_port_reset(struct usb_hub *hub, int port1,
 					&portstatus, &portchange) < 0)
 				goto done;
 
-			if (!hub_port_warm_reset_required(hub, portstatus))
+			if (!hub_port_warm_reset_required(hub, port1,
+					portstatus))
 				goto done;
 
 			/*
@@ -2856,8 +2865,13 @@ static int check_port_resume_type(struct usb_device *udev,
 {
 	struct usb_port *port_dev = hub->ports[port1 - 1];
 
+	/* Is a warm reset needed to recover the connection? */
+	if (status == 0 && udev->reset_resume
+		&& hub_port_warm_reset_required(hub, port1, portstatus)) {
+		/* pass */;
+	}
 	/* Is the device still present? */
-	if (status || port_is_suspended(hub, portstatus) ||
+	else if (status || port_is_suspended(hub, portstatus) ||
 			!port_is_power_on(hub, portstatus) ||
 			!(portstatus & USB_PORT_STAT_CONNECTION)) {
 		if (status >= 0)
@@ -4872,7 +4886,7 @@ static void port_event(struct usb_hub *hub, int port1)
 	 * Warm reset a USB3 protocol port if it's in
 	 * SS.Inactive state.
 	 */
-	if (hub_port_warm_reset_required(hub, portstatus)) {
+	if (hub_port_warm_reset_required(hub, port1, portstatus)) {
 		dev_dbg(&port_dev->dev, "do warm reset\n");
 		if (!udev || !(portstatus & USB_PORT_STAT_CONNECTION)
 				|| udev->state == USB_STATE_NOTATTACHED) {

drivers/usb/core/hub.h

@@ -52,6 +52,8 @@ struct usb_hub {
 	unsigned long		power_bits[1]; /* ports that are powered */
 	unsigned long		child_usage_bits[1]; /* ports powered on for
 							children */
+	unsigned long		warm_reset_bits[1]; /* ports requesting warm
+							reset recovery */
 #if USB_MAXCHILDREN > 31 /* 8*sizeof(unsigned long) - 1 */
 #error event_bits[] is too short!
 #endif

drivers/usb/core/port.c

@@ -103,16 +103,19 @@ static int usb_port_runtime_resume(struct device *dev)
 	msleep(hub_power_on_good_delay(hub));
 	if (udev && !retval) {
 		/*
-		 * Attempt to wait for usb hub port to be reconnected in order
-		 * to make the resume procedure successful.  The device may have
-		 * disconnected while the port was powered off, so ignore the
-		 * return status.
+		 * Our preference is to simply wait for the port to reconnect,
+		 * as that is the lowest latency method to restart the port.
+		 * However, there are cases where toggling port power results in
+		 * the host port and the device port getting out of sync causing
+		 * a link training live lock.  Upon timeout, flag the port as
+		 * needing warm reset recovery (to be performed later by
+		 * usb_port_resume() as requested via usb_wakeup_notification())
 		 */
-		retval = hub_port_debounce_be_connected(hub, port1);
-		if (retval < 0)
-			dev_dbg(&port_dev->dev, "can't get reconnection after setting port  power on, status %d\n",
-					retval);
-		retval = 0;
+		if (hub_port_debounce_be_connected(hub, port1) < 0) {
+			dev_dbg(&port_dev->dev, "reconnect timeout\n");
+			if (hub_is_superspeed(hdev))
+				set_bit(port1, hub->warm_reset_bits);
+		}
 
 		/* Force the child awake to revalidate after the power loss. */
 		if (!test_and_set_bit(port1, hub->child_usage_bits)) {