USB: EHCI: use hrtimer for (s)iTD deallocation

This patch (as1579) adds an hrtimer event to handle deallocation of
iTDs and siTDs in ehci-hcd.

Because of the frame-oriented approach used by the EHCI periodic
schedule, the hardware can continue to access the Transfer Descriptor
for isochronous (or split-isochronous) transactions for up to a
millisecond after the transaction completes.  The iTD (or siTD) must
not be reused before then.

The strategy currently used involves putting completed iTDs on a list
of cached entries and every so often returning them to the endpoint's
free list.  The new strategy reduces overhead by putting completed
iTDs back on the free list immediately, although they are not reused
until it is safe to do so.

When the isochronous endpoint stops (its queue becomes empty), the
iTDs on its free list get moved to a global list, from which they will
be deallocated after a minimum of 2 ms.  This delay is what the new
hrtimer event is for.

Overall this may not be a tremendous improvement over the current
code, but to me it seems a lot more clear and logical.  In addition,
it removes the need for each iTD to keep a reference to the
ehci_iso_stream it belongs to, since the iTD never needs to be moved
back to the stream's free list from the global list.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/usb/host/ehci-hcd.c

@@ -509,6 +509,7 @@ static void ehci_stop (struct usb_hcd *hcd)
 	spin_lock_irq (&ehci->lock);
 	if (ehci->async)
 		ehci_work (ehci);
+	end_free_itds(ehci);
 	spin_unlock_irq (&ehci->lock);
 	ehci_mem_cleanup (ehci);
 

drivers/usb/host/ehci-hub.c

@@ -303,6 +303,7 @@ static int ehci_bus_suspend (struct usb_hcd *hcd)
 	if (ehci->async_unlink)
 		end_unlink_async(ehci);
 	ehci_handle_intr_unlinks(ehci);
+	end_free_itds(ehci);
 
 	/* allow remote wakeup */
 	mask = INTR_MASK;

drivers/usb/host/ehci-mem.c

@@ -118,7 +118,6 @@ fail:
 
 static void ehci_mem_cleanup (struct ehci_hcd *ehci)
 {
-	free_cached_lists(ehci);
 	if (ehci->async)
 		qh_destroy(ehci, ehci->async);
 	ehci->async = NULL;

drivers/usb/host/ehci-sched.c

@@ -1045,31 +1045,6 @@ iso_stream_put(struct ehci_hcd *ehci, struct ehci_iso_stream *stream)
 	if (stream->refcount == 1) {
 		// BUG_ON (!list_empty(&stream->td_list));
 
-		while (!list_empty (&stream->free_list)) {
-			struct list_head	*entry;
-
-			entry = stream->free_list.next;
-			list_del (entry);
-
-			/* knows about ITD vs SITD */
-			if (stream->highspeed) {
-				struct ehci_itd		*itd;
-
-				itd = list_entry (entry, struct ehci_itd,
-						itd_list);
-				dma_pool_free (ehci->itd_pool, itd,
-						itd->itd_dma);
-			} else {
-				struct ehci_sitd	*sitd;
-
-				sitd = list_entry (entry, struct ehci_sitd,
-						sitd_list);
-				dma_pool_free (ehci->sitd_pool, sitd,
-						sitd->sitd_dma);
-			}
-		}
-
-		stream->bEndpointAddress &= 0x0f;
 		if (stream->ep)
 			stream->ep->hcpriv = NULL;
 
@@ -1230,17 +1205,19 @@ itd_urb_transaction (
 	spin_lock_irqsave (&ehci->lock, flags);
 	for (i = 0; i < num_itds; i++) {
 
-		/* free_list.next might be cache-hot ... but maybe
-		 * the HC caches it too. avoid that issue for now.
+		/*
+		 * Use iTDs from the free list, but not iTDs that may
+		 * still be in use by the hardware.
 		 */
-
-		/* prefer previously-allocated itds */
-		if (likely (!list_empty(&stream->free_list))) {
-			itd = list_entry (stream->free_list.prev,
+		if (likely(!list_empty(&stream->free_list))) {
+			itd = list_first_entry(&stream->free_list,
 					struct ehci_itd, itd_list);
+			if (itd->frame == ehci->clock_frame)
+				goto alloc_itd;
 			list_del (&itd->itd_list);
 			itd_dma = itd->itd_dma;
 		} else {
+ alloc_itd:
 			spin_unlock_irqrestore (&ehci->lock, flags);
 			itd = dma_pool_alloc (ehci->itd_pool, mem_flags,
 					&itd_dma);
@@ -1762,24 +1739,18 @@ itd_complete (
 
 done:
 	itd->urb = NULL;
-	if (ehci->clock_frame != itd->frame || itd->index[7] != -1) {
-		/* OK to recycle this ITD now. */
-		itd->stream = NULL;
-		list_move(&itd->itd_list, &stream->free_list);
-		iso_stream_put(ehci, stream);
-	} else {
-		/* HW might remember this ITD, so we can't recycle it yet.
-		 * Move it to a safe place until a new frame starts.
-		 */
-		list_move(&itd->itd_list, &ehci->cached_itd_list);
-		if (stream->refcount == 2) {
-			/* If iso_stream_put() were called here, stream
-			 * would be freed.  Instead, just prevent reuse.
-			 */
-			stream->ep->hcpriv = NULL;
-			stream->ep = NULL;
-		}
+
+	/* Add to the end of the free list for later reuse */
+	list_move_tail(&itd->itd_list, &stream->free_list);
+
+	/* Recycle the iTDs when the pipeline is empty (ep no longer in use) */
+	if (list_empty(&stream->td_list)) {
+		list_splice_tail_init(&stream->free_list,
+				&ehci->cached_itd_list);
+		start_free_itds(ehci);
 	}
+
+	iso_stream_put(ehci, stream);
 	return retval;
 }
 
@@ -1930,17 +1901,19 @@ sitd_urb_transaction (
 		 * means we never need two sitds for full speed packets.
 		 */
 
-		/* free_list.next might be cache-hot ... but maybe
-		 * the HC caches it too. avoid that issue for now.
+		/*
+		 * Use siTDs from the free list, but not siTDs that may
+		 * still be in use by the hardware.
 		 */
-
-		/* prefer previously-allocated sitds */
-		if (!list_empty(&stream->free_list)) {
-			sitd = list_entry (stream->free_list.prev,
+		if (likely(!list_empty(&stream->free_list))) {
+			sitd = list_first_entry(&stream->free_list,
 					 struct ehci_sitd, sitd_list);
+			if (sitd->frame == ehci->clock_frame)
+				goto alloc_sitd;
 			list_del (&sitd->sitd_list);
 			sitd_dma = sitd->sitd_dma;
 		} else {
+ alloc_sitd:
 			spin_unlock_irqrestore (&ehci->lock, flags);
 			sitd = dma_pool_alloc (ehci->sitd_pool, mem_flags,
 					&sitd_dma);
@@ -2157,24 +2130,18 @@ sitd_complete (
 
 done:
 	sitd->urb = NULL;
-	if (ehci->clock_frame != sitd->frame) {
-		/* OK to recycle this SITD now. */
-		sitd->stream = NULL;
-		list_move(&sitd->sitd_list, &stream->free_list);
-		iso_stream_put(ehci, stream);
-	} else {
-		/* HW might remember this SITD, so we can't recycle it yet.
-		 * Move it to a safe place until a new frame starts.
-		 */
-		list_move(&sitd->sitd_list, &ehci->cached_sitd_list);
-		if (stream->refcount == 2) {
-			/* If iso_stream_put() were called here, stream
-			 * would be freed.  Instead, just prevent reuse.
-			 */
-			stream->ep->hcpriv = NULL;
-			stream->ep = NULL;
-		}
+
+	/* Add to the end of the free list for later reuse */
+	list_move_tail(&sitd->sitd_list, &stream->free_list);
+
+	/* Recycle the siTDs when the pipeline is empty (ep no longer in use) */
+	if (list_empty(&stream->td_list)) {
+		list_splice_tail_init(&stream->free_list,
+				&ehci->cached_sitd_list);
+		start_free_itds(ehci);
 	}
+
+	iso_stream_put(ehci, stream);
 	return retval;
 }
 
@@ -2239,28 +2206,6 @@ done:
 
 /*-------------------------------------------------------------------------*/
 
-static void free_cached_lists(struct ehci_hcd *ehci)
-{
-	struct ehci_itd *itd, *n;
-	struct ehci_sitd *sitd, *sn;
-
-	list_for_each_entry_safe(itd, n, &ehci->cached_itd_list, itd_list) {
-		struct ehci_iso_stream	*stream = itd->stream;
-		itd->stream = NULL;
-		list_move(&itd->itd_list, &stream->free_list);
-		iso_stream_put(ehci, stream);
-	}
-
-	list_for_each_entry_safe(sitd, sn, &ehci->cached_sitd_list, sitd_list) {
-		struct ehci_iso_stream	*stream = sitd->stream;
-		sitd->stream = NULL;
-		list_move(&sitd->sitd_list, &stream->free_list);
-		iso_stream_put(ehci, stream);
-	}
-}
-
-/*-------------------------------------------------------------------------*/
-
 static void
 scan_periodic (struct ehci_hcd *ehci)
 {
@@ -2282,10 +2227,7 @@ scan_periodic (struct ehci_hcd *ehci)
 		clock = now_uframe + mod - 1;
 		clock_frame = -1;
 	}
-	if (ehci->clock_frame != clock_frame) {
-		free_cached_lists(ehci);
-		ehci->clock_frame = clock_frame;
-	}
+	ehci->clock_frame = clock_frame;
 	clock &= mod - 1;
 	clock_frame = clock >> 3;
 	++ehci->periodic_stamp;
@@ -2463,7 +2405,6 @@ restart:
 			clock = now;
 			clock_frame = clock >> 3;
 			if (ehci->clock_frame != clock_frame) {
-				free_cached_lists(ehci);
 				ehci->clock_frame = clock_frame;
 				++ehci->periodic_stamp;
 			}

drivers/usb/host/ehci-timer.c

@@ -71,6 +71,7 @@ static unsigned event_delays_ns[] = {
 	1 * NSEC_PER_MSEC,	/* EHCI_HRTIMER_POLL_PSS */
 	1 * NSEC_PER_MSEC,	/* EHCI_HRTIMER_POLL_DEAD */
 	1125 * NSEC_PER_USEC,	/* EHCI_HRTIMER_UNLINK_INTR */
+	2 * NSEC_PER_MSEC,	/* EHCI_HRTIMER_FREE_ITDS */
 	10 * NSEC_PER_MSEC,	/* EHCI_HRTIMER_DISABLE_PERIODIC */
 	15 * NSEC_PER_MSEC,	/* EHCI_HRTIMER_DISABLE_ASYNC */
 };
@@ -165,7 +166,6 @@ static void ehci_poll_PSS(struct ehci_hcd *ehci)
 
 	/* The status is up-to-date; restart or stop the schedule as needed */
 	if (want == 0) {	/* Stopped */
-		free_cached_lists(ehci);
 		if (ehci->periodic_count > 0) {
 
 			/* make sure ehci_work scans these */
@@ -188,9 +188,6 @@ static void ehci_poll_PSS(struct ehci_hcd *ehci)
 static void ehci_disable_PSE(struct ehci_hcd *ehci)
 {
 	ehci_clear_command_bit(ehci, CMD_PSE);
-
-	/* Poll to see when it actually stops */
-	ehci_enable_event(ehci, EHCI_HRTIMER_POLL_PSS, true);
 }
 
 
@@ -250,6 +247,50 @@ static void ehci_handle_intr_unlinks(struct ehci_hcd *ehci)
 }
 
 
+/* Start another free-iTDs/siTDs cycle */
+static void start_free_itds(struct ehci_hcd *ehci)
+{
+	if (!(ehci->enabled_hrtimer_events & BIT(EHCI_HRTIMER_FREE_ITDS))) {
+		ehci->last_itd_to_free = list_entry(
+				ehci->cached_itd_list.prev,
+				struct ehci_itd, itd_list);
+		ehci->last_sitd_to_free = list_entry(
+				ehci->cached_sitd_list.prev,
+				struct ehci_sitd, sitd_list);
+		ehci_enable_event(ehci, EHCI_HRTIMER_FREE_ITDS, true);
+	}
+}
+
+/* Wait for controller to stop using old iTDs and siTDs */
+static void end_free_itds(struct ehci_hcd *ehci)
+{
+	struct ehci_itd		*itd, *n;
+	struct ehci_sitd	*sitd, *sn;
+
+	if (ehci->rh_state < EHCI_RH_RUNNING) {
+		ehci->last_itd_to_free = NULL;
+		ehci->last_sitd_to_free = NULL;
+	}
+
+	list_for_each_entry_safe(itd, n, &ehci->cached_itd_list, itd_list) {
+		list_del(&itd->itd_list);
+		dma_pool_free(ehci->itd_pool, itd, itd->itd_dma);
+		if (itd == ehci->last_itd_to_free)
+			break;
+	}
+	list_for_each_entry_safe(sitd, sn, &ehci->cached_sitd_list, sitd_list) {
+		list_del(&sitd->sitd_list);
+		dma_pool_free(ehci->sitd_pool, sitd, sitd->sitd_dma);
+		if (sitd == ehci->last_sitd_to_free)
+			break;
+	}
+
+	if (!list_empty(&ehci->cached_itd_list) ||
+			!list_empty(&ehci->cached_sitd_list))
+		start_free_itds(ehci);
+}
+
+
 /*
  * Handler functions for the hrtimer event types.
  * Keep this array in the same order as the event types indexed by
@@ -260,6 +301,7 @@ static void (*event_handlers[])(struct ehci_hcd *) = {
 	ehci_poll_PSS,			/* EHCI_HRTIMER_POLL_PSS */
 	ehci_handle_controller_death,	/* EHCI_HRTIMER_POLL_DEAD */
 	ehci_handle_intr_unlinks,	/* EHCI_HRTIMER_UNLINK_INTR */
+	end_free_itds,			/* EHCI_HRTIMER_FREE_ITDS */
 	ehci_disable_PSE,		/* EHCI_HRTIMER_DISABLE_PERIODIC */
 	ehci_disable_ASE,		/* EHCI_HRTIMER_DISABLE_ASYNC */
 };

drivers/usb/host/ehci.h

@@ -83,6 +83,7 @@ enum ehci_hrtimer_event {
 	EHCI_HRTIMER_POLL_PSS,		/* Poll for periodic schedule off */
 	EHCI_HRTIMER_POLL_DEAD,		/* Wait for dead controller to stop */
 	EHCI_HRTIMER_UNLINK_INTR,	/* Wait for interrupt QH unlink */
+	EHCI_HRTIMER_FREE_ITDS,		/* Wait for unused iTDs and siTDs */
 	EHCI_HRTIMER_DISABLE_PERIODIC,	/* Wait to disable periodic sched */
 	EHCI_HRTIMER_DISABLE_ASYNC,	/* Wait to disable async sched */
 	EHCI_HRTIMER_NUM_EVENTS		/* Must come last */
@@ -139,7 +140,9 @@ struct ehci_hcd {			/* one per controller */
 
 	/* list of itds & sitds completed while clock_frame was still active */
 	struct list_head	cached_itd_list;
+	struct ehci_itd		*last_itd_to_free;
 	struct list_head	cached_sitd_list;
+	struct ehci_sitd	*last_sitd_to_free;
 	unsigned		clock_frame;
 
 	/* per root hub port */
@@ -250,8 +253,6 @@ timer_action_done (struct ehci_hcd *ehci, enum ehci_timer_action action)
 	clear_bit (action, &ehci->actions);
 }
 
-static void free_cached_lists(struct ehci_hcd *ehci);
-
 /*-------------------------------------------------------------------------*/
 
 #include <linux/usb/ehci_def.h>