USB: EHCI update VIA workaround

This revamps handling of the hardware "async advance" IRQ, and its watchdog
timer.  Basically it dis-entangles that important timeout from the others,
simplifying the associated state and code to make it more robust.

This reportedly improves behavior of EHCI on some systems with VIA chips,
and AFAIK won't affect non-VIA hardware.  VIA systems need this code to
recover from silcon bugs whereby the "async advance" IRQ isn't issued.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

drivers/usb/host/ehci-dbg.c

@@ -754,9 +754,7 @@ show_registers (struct class_device *class_dev, char *buf)
 	}
 
 	if (ehci->reclaim) {
-		temp = scnprintf (next, size, "reclaim qh %p%s\n",
-				ehci->reclaim,
-				ehci->reclaim_ready ? " ready" : "");
+		temp = scnprintf (next, size, "reclaim qh %p\n", ehci->reclaim);
 		size -= temp;
 		next += temp;
 	}

drivers/usb/host/ehci-hcd.c

@@ -111,7 +111,7 @@ static const char	hcd_name [] = "ehci_hcd";
 #define	EHCI_TUNE_MULT_TT	1
 #define	EHCI_TUNE_FLS		2	/* (small) 256 frame schedule */
 
-#define EHCI_IAA_JIFFIES	(HZ/100)	/* arbitrary; ~10 msec */
+#define EHCI_IAA_MSECS		10		/* arbitrary */
 #define EHCI_IO_JIFFIES		(HZ/10)		/* io watchdog > irq_thresh */
 #define EHCI_ASYNC_JIFFIES	(HZ/20)		/* async idle timeout */
 #define EHCI_SHRINK_JIFFIES	(HZ/200)	/* async qh unlink delay */
@@ -254,6 +254,7 @@ static void ehci_quiesce (struct ehci_hcd *ehci)
 
 /*-------------------------------------------------------------------------*/
 
+static void end_unlink_async (struct ehci_hcd *ehci, struct pt_regs *regs);
 static void ehci_work(struct ehci_hcd *ehci, struct pt_regs *regs);
 
 #include "ehci-hub.c"
@@ -263,25 +264,36 @@ static void ehci_work(struct ehci_hcd *ehci, struct pt_regs *regs);
 
 /*-------------------------------------------------------------------------*/
 
-static void ehci_watchdog (unsigned long param)
+static void ehci_iaa_watchdog (unsigned long param)
 {
 	struct ehci_hcd		*ehci = (struct ehci_hcd *) param;
 	unsigned long		flags;
+	u32			status;
 
 	spin_lock_irqsave (&ehci->lock, flags);
+	WARN_ON(!ehci->reclaim);
 
-	/* lost IAA irqs wedge things badly; seen with a vt8235 */
+	/* lost IAA irqs wedge things badly; seen first with a vt8235 */
 	if (ehci->reclaim) {
-		u32		status = readl (&ehci->regs->status);
-
+		status = readl (&ehci->regs->status);
 		if (status & STS_IAA) {
 			ehci_vdbg (ehci, "lost IAA\n");
 			COUNT (ehci->stats.lost_iaa);
 			writel (STS_IAA, &ehci->regs->status);
-			ehci->reclaim_ready = 1;
+			end_unlink_async (ehci, NULL);
 		}
 	}
 
+	spin_unlock_irqrestore (&ehci->lock, flags);
+}
+
+static void ehci_watchdog (unsigned long param)
+{
+	struct ehci_hcd		*ehci = (struct ehci_hcd *) param;
+	unsigned long		flags;
+
+	spin_lock_irqsave (&ehci->lock, flags);
+
 	/* stop async processing after it's idled a bit */
 	if (test_bit (TIMER_ASYNC_OFF, &ehci->actions))
 		start_unlink_async (ehci, ehci->async);
@@ -333,8 +345,6 @@ static void ehci_port_power (struct ehci_hcd *ehci, int is_on)
 static void ehci_work (struct ehci_hcd *ehci, struct pt_regs *regs)
 {
 	timer_action_done (ehci, TIMER_IO_WATCHDOG);
-	if (ehci->reclaim_ready)
-		end_unlink_async (ehci, regs);
 
 	/* another CPU may drop ehci->lock during a schedule scan while
 	 * it reports urb completions.  this flag guards against bogus
@@ -369,6 +379,7 @@ static void ehci_stop (struct usb_hcd *hcd)
 
 	/* no more interrupts ... */
 	del_timer_sync (&ehci->watchdog);
+	del_timer_sync (&ehci->iaa_watchdog);
 
 	spin_lock_irq(&ehci->lock);
 	if (HC_IS_RUNNING (hcd->state))
@@ -415,6 +426,10 @@ static int ehci_init(struct usb_hcd *hcd)
 	ehci->watchdog.function = ehci_watchdog;
 	ehci->watchdog.data = (unsigned long) ehci;
 
+	init_timer(&ehci->iaa_watchdog);
+	ehci->iaa_watchdog.function = ehci_iaa_watchdog;
+	ehci->iaa_watchdog.data = (unsigned long) ehci;
+
 	/*
 	 * hw default: 1K periodic list heads, one per frame.
 	 * periodic_size can shrink by USBCMD update if hcc_params allows.
@@ -431,7 +446,6 @@ static int ehci_init(struct usb_hcd *hcd)
 		ehci->i_thresh = 2 + HCC_ISOC_THRES(hcc_params);
 
 	ehci->reclaim = NULL;
-	ehci->reclaim_ready = 0;
 	ehci->next_uframe = -1;
 
 	/*
@@ -605,7 +619,7 @@ static irqreturn_t ehci_irq (struct usb_hcd *hcd, struct pt_regs *regs)
 	/* complete the unlinking of some qh [4.15.2.3] */
 	if (status & STS_IAA) {
 		COUNT (ehci->stats.reclaim);
-		ehci->reclaim_ready = 1;
+		end_unlink_async (ehci, regs);
 		bh = 1;
 	}
 
@@ -709,10 +723,14 @@ static int ehci_urb_enqueue (
 
 static void unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh)
 {
-	/* if we need to use IAA and it's busy, defer */
-	if (qh->qh_state == QH_STATE_LINKED
-			&& ehci->reclaim
-			&& HC_IS_RUNNING (ehci_to_hcd(ehci)->state)) {
+	// BUG_ON(qh->qh_state != QH_STATE_LINKED);
+
+	/* failfast */
+	if (!HC_IS_RUNNING (ehci_to_hcd(ehci)->state))
+		end_unlink_async (ehci, NULL);
+
+	/* defer till later if busy */
+	else if (ehci->reclaim) {
 		struct ehci_qh		*last;
 
 		for (last = ehci->reclaim;
@@ -722,12 +740,8 @@ static void unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh)
 		qh->qh_state = QH_STATE_UNLINK_WAIT;
 		last->reclaim = qh;
 
-	/* bypass IAA if the hc can't care */
-	} else if (!HC_IS_RUNNING (ehci_to_hcd(ehci)->state) && ehci->reclaim)
-		end_unlink_async (ehci, NULL);
-
-	/* something else might have unlinked the qh by now */
-	if (qh->qh_state == QH_STATE_LINKED)
+	/* start IAA cycle */
+	} else
 		start_unlink_async (ehci, qh);
 }
 
@@ -749,8 +763,20 @@ static int ehci_urb_dequeue (struct usb_hcd *hcd, struct urb *urb)
 		qh = (struct ehci_qh *) urb->hcpriv;
 		if (!qh)
 			break;
+		switch (qh->qh_state) {
+		case QH_STATE_LINKED:
+		case QH_STATE_COMPLETING:
 			unlink_async (ehci, qh);
 			break;
+		case QH_STATE_UNLINK:
+		case QH_STATE_UNLINK_WAIT:
+			/* already started */
+			break;
+		case QH_STATE_IDLE:
+			WARN_ON(1);
+			break;
+		}
+		break;
 
 	case PIPE_INTERRUPT:
 		qh = (struct ehci_qh *) urb->hcpriv;
@@ -841,6 +867,7 @@ ehci_endpoint_disable (struct usb_hcd *hcd, struct usb_host_endpoint *ep)
 		unlink_async (ehci, qh);
 		/* FALL THROUGH */
 	case QH_STATE_UNLINK:		/* wait for hw to finish? */
+	case QH_STATE_UNLINK_WAIT:
 idle_timeout:
 		spin_unlock_irqrestore (&ehci->lock, flags);
 		schedule_timeout_uninterruptible(1);

drivers/usb/host/ehci-hub.c

@@ -48,7 +48,7 @@ static int ehci_bus_suspend (struct usb_hcd *hcd)
 	}
 	ehci->command = readl (&ehci->regs->command);
 	if (ehci->reclaim)
-		ehci->reclaim_ready = 1;
+		end_unlink_async (ehci, NULL);
 	ehci_work(ehci, NULL);
 
 	/* suspend any active/unsuspended ports, maybe allow wakeup */

drivers/usb/host/ehci-pci.c

@@ -303,7 +303,7 @@ static int ehci_pci_resume(struct usb_hcd *hcd)
 	/* emptying the schedule aborts any urbs */
 	spin_lock_irq(&ehci->lock);
 	if (ehci->reclaim)
-		ehci->reclaim_ready = 1;
+		end_unlink_async (ehci, NULL);
 	ehci_work(ehci, NULL);
 	spin_unlock_irq(&ehci->lock);
 

drivers/usb/host/ehci-q.c

@@ -967,7 +967,7 @@ static void end_unlink_async (struct ehci_hcd *ehci, struct pt_regs *regs)
 	struct ehci_qh		*qh = ehci->reclaim;
 	struct ehci_qh		*next;
 
-	timer_action_done (ehci, TIMER_IAA_WATCHDOG);
+	iaa_watchdog_done (ehci);
 
 	// qh->hw_next = cpu_to_le32 (qh->qh_dma);
 	qh->qh_state = QH_STATE_IDLE;
@@ -977,7 +977,6 @@ static void end_unlink_async (struct ehci_hcd *ehci, struct pt_regs *regs)
 	/* other unlink(s) may be pending (in QH_STATE_UNLINK_WAIT) */
 	next = qh->reclaim;
 	ehci->reclaim = next;
-	ehci->reclaim_ready = 0;
 	qh->reclaim = NULL;
 
 	qh_completions (ehci, qh, regs);
@@ -1052,11 +1051,10 @@ static void start_unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh)
 		return;
 	}
 
-	ehci->reclaim_ready = 0;
 	cmd |= CMD_IAAD;
 	writel (cmd, &ehci->regs->command);
 	(void) readl (&ehci->regs->command);
-	timer_action (ehci, TIMER_IAA_WATCHDOG);
+	iaa_watchdog_start (ehci);
 }
 
 /*-------------------------------------------------------------------------*/

drivers/usb/host/ehci.h

@@ -58,7 +58,6 @@ struct ehci_hcd {			/* one per controller */
 	/* async schedule support */
 	struct ehci_qh		*async;
 	struct ehci_qh		*reclaim;
-	unsigned		reclaim_ready : 1;
 	unsigned		scanning : 1;
 
 	/* periodic schedule support */
@@ -81,6 +80,7 @@ struct ehci_hcd {			/* one per controller */
 	struct dma_pool		*itd_pool;	/* itd per iso urb */
 	struct dma_pool		*sitd_pool;	/* sitd per split iso urb */
 
+	struct timer_list	iaa_watchdog;
 	struct timer_list	watchdog;
 	unsigned long		actions;
 	unsigned		stamp;
@@ -114,9 +114,21 @@ static inline struct usb_hcd *ehci_to_hcd (struct ehci_hcd *ehci)
 }
 
 
+static inline void
+iaa_watchdog_start (struct ehci_hcd *ehci)
+{
+	WARN_ON(timer_pending(&ehci->iaa_watchdog));
+	mod_timer (&ehci->iaa_watchdog,
+			jiffies + msecs_to_jiffies(EHCI_IAA_MSECS));
+}
+
+static inline void iaa_watchdog_done (struct ehci_hcd *ehci)
+{
+	del_timer (&ehci->iaa_watchdog);
+}
+
 enum ehci_timer_action {
 	TIMER_IO_WATCHDOG,
-	TIMER_IAA_WATCHDOG,
 	TIMER_ASYNC_SHRINK,
 	TIMER_ASYNC_OFF,
 };
@@ -134,9 +146,6 @@ timer_action (struct ehci_hcd *ehci, enum ehci_timer_action action)
 		unsigned long t;
 
 		switch (action) {
-		case TIMER_IAA_WATCHDOG:
-			t = EHCI_IAA_JIFFIES;
-			break;
 		case TIMER_IO_WATCHDOG:
 			t = EHCI_IO_JIFFIES;
 			break;
@@ -153,8 +162,7 @@ timer_action (struct ehci_hcd *ehci, enum ehci_timer_action action)
 		// async queue SHRINK often precedes IAA.  while it's ready
 		// to go OFF neither can matter, and afterwards the IO
 		// watchdog stops unless there's still periodic traffic.
-		if (action != TIMER_IAA_WATCHDOG
-				&& t > ehci->watchdog.expires
+		if (time_before_eq(t, ehci->watchdog.expires)
 				&& timer_pending (&ehci->watchdog))
 			return;
 		mod_timer (&ehci->watchdog, t);