[PATCH v2 char-misc 1/1] Drivers: hv: vmbus: Fix ring buffer signaling

[Michael Kelley]

Fix bugs in signaling the Hyper-V host when freeing space in the
host->guest ring buffer:

1. The interrupt_mask must not be used to determine whether to signal
   on the host->guest ring buffer
2. The ring buffer write_index must be read (via hv_get_bytes_to_write)
   *after* pending_send_sz is read in order to avoid a race condition
3. Comparisons with pending_send_sz must treat the "equals" case as
   not-enough-space
4. Don't signal if the pending_send_sz feature is not present. Older
   versions of Hyper-V that don't implement this feature will poll.

Fixes: 03bad714a161 ("vmbus: more host signalling avoidance")
Signed-off-by: Michael Kelley <mhkelley at outlook.com>
---

Changes in v2:
* Restructured to do signaling calcuations only if needed. Makes
  the most common path faster [KY Srinivasan, Steve Hemminger]
* Moved bytes read calculation to new routine hv_pkt_iter_bytes_read
  [Steve Hemminger]

---
 drivers/hv/ring_buffer.c | 52 ++++++++++++++++++++++++++++++++----------------
 1 file changed, 35 insertions(+), 17 deletions(-)

diff --git a/drivers/hv/ring_buffer.c b/drivers/hv/ring_buffer.c
index 50e0714..8699bb9 100644
--- a/drivers/hv/ring_buffer.c
+++ b/drivers/hv/ring_buffer.c
@@ -417,13 +417,24 @@ struct vmpacket_descriptor *
 }
 EXPORT_SYMBOL_GPL(__hv_pkt_iter_next);
 
+/* How many bytes were read in this iterator cycle */
+static u32 hv_pkt_iter_bytes_read(const struct hv_ring_buffer_info *rbi,
+					u32 start_read_index)
+{
+	if (rbi->priv_read_index >= start_read_index)
+		return rbi->priv_read_index - start_read_index;
+	else
+		return rbi->ring_datasize - start_read_index +
+			rbi->priv_read_index;
+}
+
 /*
  * Update host ring buffer after iterating over packets.
  */
 void hv_pkt_iter_close(struct vmbus_channel *channel)
 {
 	struct hv_ring_buffer_info *rbi = &channel->inbound;
-	u32 orig_write_sz = hv_get_bytes_to_write(rbi);
+	u32 curr_write_sz, pending_sz, bytes_read, start_read_index;
 
 	/*
 	 * Make sure all reads are done before we update the read index since
@@ -431,8 +442,12 @@ void hv_pkt_iter_close(struct vmbus_channel *channel)
 	 * is updated.
 	 */
 	virt_rmb();
+	start_read_index = rbi->ring_buffer->read_index;
 	rbi->ring_buffer->read_index = rbi->priv_read_index;
 
+	if (!rbi->ring_buffer->feature_bits.feat_pending_send_sz)
+		return;
+
 	/*
 	 * Issue a full memory barrier before making the signaling decision.
 	 * Here is the reason for having this barrier:
@@ -446,26 +461,29 @@ void hv_pkt_iter_close(struct vmbus_channel *channel)
 	 */
 	virt_mb();
 
-	/* If host has disabled notifications then skip */
-	if (rbi->ring_buffer->interrupt_mask)
+	pending_sz = READ_ONCE(rbi->ring_buffer->pending_send_sz);
+	if (!pending_sz)
 		return;
 
-	if (rbi->ring_buffer->feature_bits.feat_pending_send_sz) {
-		u32 pending_sz = READ_ONCE(rbi->ring_buffer->pending_send_sz);
+	/*
+	 * Ensure the read of write_index in hv_get_bytes_to_write()
+	 * happens after the read of pending_send_sz.
+	 */
+	virt_rmb();
+	curr_write_sz = hv_get_bytes_to_write(rbi);
+	bytes_read = hv_pkt_iter_bytes_read(rbi, start_read_index);
 
-		/*
-		 * If there was space before we began iteration,
-		 * then host was not blocked. Also handles case where
-		 * pending_sz is zero then host has nothing pending
-		 * and does not need to be signaled.
-		 */
-		if (orig_write_sz > pending_sz)
-			return;
+	/*
+	 * If there was space before we began iteration,
+	 * then host was not blocked.
+	 */
 
-		/* If pending write will not fit, don't give false hope. */
-		if (hv_get_bytes_to_write(rbi) < pending_sz)
-			return;
-	}
+	if (curr_write_sz - bytes_read > pending_sz)
+		return;
+
+	/* If pending write will not fit, don't give false hope. */
+	if (curr_write_sz <= pending_sz)
+		return;
 
 	vmbus_setevent(channel);
 }
-- 
1.8.3.1