[PATCH 14/25] aio: Make aio_read_evt() more efficient

[Kent Overstreet]

Previously, aio_read_event() pulled a single completion off the
ringbuffer at a time, locking and unlocking each time.

Changed it to pull off as many events as it can at a time, and copy them
directly to userspace. The easiest way to do this was by making it
lockless, since we can't copy_to_user() with a spinlock held.

We can't use cmpxchg() on the ring buffer's head pointer directly, since
it's modded to nr_events and would be susceptible to ABA. So instead we
maintain a shadow head that uses the full 32 bits, and cmpxchg() that
and then updated the real head pointer.

This also fixes a bug where if copying the event to userspace failed,
we'd lose the event.

Signed-off-by: Kent Overstreet <koverstreet@xxxxxxxxxx>
---
 fs/aio.c | 173 +++++++++++++++++++++++++++++----------------------------------
 1 file changed, 80 insertions(+), 93 deletions(-)

diff --git a/fs/aio.c b/fs/aio.c
index 7961f60..de255d8 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -63,7 +63,7 @@ struct aio_ring_info {
 	unsigned long		mmap_size;
 
 	struct page		**ring_pages;
-	spinlock_t		ring_lock;
+	struct mutex		ring_lock;
 	long			nr_pages;
 
 	unsigned		nr, tail;
@@ -324,7 +324,7 @@ static struct kioctx *ioctx_alloc(unsigned nr_events)
 
 	atomic_set(&ctx->users, 2);
 	spin_lock_init(&ctx->ctx_lock);
-	spin_lock_init(&ctx->ring_info.ring_lock);
+	mutex_init(&ctx->ring_info.ring_lock);
 	init_waitqueue_head(&ctx->wait);
 
 	INIT_LIST_HEAD(&ctx->active_reqs);
@@ -704,86 +704,86 @@ put_rq:
 }
 EXPORT_SYMBOL(aio_complete);
 
-/* aio_read_evt
- *	Pull an event off of the ioctx's event ring.  Returns the number of 
- *	events fetched (0 or 1 ;-)
- *	FIXME: make this use cmpxchg.
- *	TODO: make the ringbuffer user mmap()able (requires FIXME).
+/* aio_read_events
+ *	Pull an event off of the ioctx's event ring.  Returns the number of
+ *	events fetched
  */
-static int aio_read_evt(struct kioctx *ioctx, struct io_event *ent)
+static int aio_read_events(struct kioctx *ctx, struct io_event __user *event,
+			   long nr)
 {
-	struct aio_ring_info *info = &ioctx->ring_info;
+	struct aio_ring_info *info = &ctx->ring_info;
 	struct aio_ring *ring;
-	unsigned long head;
-	int ret = 0;
+	unsigned head, pos;
+	int ret = 0, copy_ret;
+
+	mutex_lock(&info->ring_lock);
 
 	ring = kmap_atomic(info->ring_pages[0]);
-	pr_debug("h%u t%u m%u\n", ring->head, ring->tail, ring->nr);
+	head = ring->head;
+	kunmap_atomic(ring);
 
-	if (ring->head == ring->tail)
-		goto out;
+	pr_debug("h%u t%u m%u\n", head, info->tail, info->nr);
+
+	while (ret < nr) {
+		unsigned i = (head < info->tail ? info->tail : info->nr) - head;
+		struct io_event *ev;
+		struct page *page;
+
+		if (head == info->tail)
+			break;
+
+		i = min_t(int, i, nr - ret);
+		i = min_t(int, i, AIO_EVENTS_PER_PAGE -
+			  ((head + AIO_EVENTS_OFFSET) % AIO_EVENTS_PER_PAGE));
+
+		pos = head + AIO_EVENTS_OFFSET;
+		page = info->ring_pages[pos / AIO_EVENTS_PER_PAGE];
+		pos %= AIO_EVENTS_PER_PAGE;
+
+		ev = kmap(page);
+		copy_ret = copy_to_user(event + ret, ev + pos, sizeof(*ev) * i);
+		kunmap(page);
+
+		if (unlikely(copy_ret)) {
+			ret = -EFAULT;
+			goto out;
+		}
 
-	spin_lock(&info->ring_lock);
-
-	head = ring->head % info->nr;
-	if (head != ring->tail) {
-		struct io_event *evp = aio_ring_event(info, head);
-		*ent = *evp;
-		head = (head + 1) % info->nr;
-		smp_mb(); /* finish reading the event before updatng the head */
-		ring->head = head;
-		ret = 1;
-		put_aio_ring_event(evp);
+		ret += i;
+		head += i;
+		head %= info->nr;
 	}
-	spin_unlock(&info->ring_lock);
 
-out:
+	smp_mb(); /* finish reading the event before updatng the head */
+
+	ring = kmap_atomic(info->ring_pages[0]);
+	ring->head = head;
 	kunmap_atomic(ring);
-	pr_debug("%d  h%u t%u\n", ret, ring->head, ring->tail);
+
+	pr_debug("%d  h%u t%u\n", ret, head, info->tail);
+out:
+	mutex_unlock(&info->ring_lock);
+
 	return ret;
 }
 
 static int read_events(struct kioctx *ctx,
-			long min_nr, long nr,
-			struct io_event __user *event,
-			struct timespec __user *timeout)
+		       long min_nr, long nr,
+		       struct io_event __user *event,
+		       struct timespec __user *timeout)
 {
 	DEFINE_WAIT(wait);
 	struct hrtimer_sleeper t;
 	size_t i = 0;
 	int ret;
-	struct io_event		ent;
-
-	/* needed to zero any padding within an entry (there shouldn't be 
-	 * any, but C is fun!
-	 */
-	memset(&ent, 0, sizeof(ent));
-	ret = 0;
-	while (likely(i < nr)) {
-		ret = aio_read_evt(ctx, &ent);
-		if (unlikely(ret <= 0))
-			break;
 
-		pr_debug("%Lx %Lx %Lx %Lx\n",
-			 ent.data, ent.obj, ent.res, ent.res2);
-
-		/* Could we split the check in two? */
-		ret = -EFAULT;
-		if (unlikely(copy_to_user(event, &ent, sizeof(ent)))) {
-			pr_debug("lost an event due to EFAULT.\n");
-			break;
-		}
-		ret = 0;
-
-		/* Good, event copied to userland, update counts. */
-		event ++;
-		i ++;
-	}
+	ret = aio_read_events(ctx, event + i, nr - i);
+	if (ret < 0)
+		return ret;
 
-	if (min_nr <= i)
+	i += ret;
+	if (i >= min_nr)
 		return i;
-	if (ret)
-		return ret;
 
 	/* End fast path */
 
@@ -802,47 +802,34 @@ static int read_events(struct kioctx *ctx,
 				       current->timer_slack_ns, HRTIMER_MODE_REL);
 	}
 
-	while (likely(i < nr)) {
+	while (i < nr) {
 		prepare_to_wait_exclusive(&ctx->wait, &wait,
 					  TASK_INTERRUPTIBLE);
 
-		do {
-			ret = aio_read_evt(ctx, &ent);
-			if (ret)
-				break;
-			if (min_nr <= i)
-				break;
-			if (unlikely(atomic_read(&ctx->dead))) {
-				ret = -EINVAL;
-				break;
-			}
-			if (!t.task)	/* Only check after read evt */
-				break;
-			/* Try to only show up in io wait if there are ops
-			 *  in flight */
-			if (atomic_read(&ctx->reqs_active))
-				io_schedule();
-			else
-				schedule();
-			if (signal_pending(current)) {
-				ret = -EINTR;
-				break;
-			}
-			/*ret = aio_read_evt(ctx, &ent);*/
-		} while (1) ;
-
-		if (unlikely(ret <= 0))
+		ret = aio_read_events(ctx, event + i, nr - i);
+		if (ret < 0)
 			break;
 
-		ret = -EFAULT;
-		if (unlikely(copy_to_user(event, &ent, sizeof(ent)))) {
-			pr_debug("lost an event due to EFAULT.\n");
+		i += ret;
+		if (i >= min_nr)
+			break;
+		if (unlikely(atomic_read(&ctx->dead))) {
+			ret = -EINVAL;
 			break;
 		}
+		if (!t.task)	/* Only check after read evt */
+			break;
+
+		/* Try to only show up in io wait if there are ops in flight */
+		if (atomic_read(&ctx->reqs_active))
+			io_schedule();
+		else
+			schedule();
 
-		/* Good, event copied to userland, update counts. */
-		event ++;
-		i ++;
+		if (signal_pending(current)) {
+			ret = -EINTR;
+			break;
+		}
 	}
 out:
 	hrtimer_cancel(&t.timer);
-- 
1.7.12

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/