[PATCH 2/2] ath10k: sanitize tx ring index access properly

[Michal Kazior]

The tx ring index was immediately trimmed with a
bitmask. This discarded the 0xFFFFFFFF error case
(which theoretically can happen when a device is
abruptly disconnected) and led to using an invalid
tx ring index. This could lead to memory
corruption.

Signed-off-by: Michal Kazior <michal.kazior at tieto.com>
---
 drivers/net/wireless/ath/ath10k/ce.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/ce.c b/drivers/net/wireless/ath/ath10k/ce.c
index d185dc0..7c6c7d5 100644
--- a/drivers/net/wireless/ath/ath10k/ce.c
+++ b/drivers/net/wireless/ath/ath10k/ce.c
@@ -603,16 +603,19 @@ static int ath10k_ce_completed_send_next_nolock(struct ath10k_ce_pipe *ce_state,
 		if (ret)
 			return ret;
 
-		src_ring->hw_index =
-			ath10k_ce_src_ring_read_index_get(ar, ctrl_addr);
-		src_ring->hw_index &= nentries_mask;
+		read_index = ath10k_ce_src_ring_read_index_get(ar, ctrl_addr);
+		if (read_index == 0xFFFFFFFF)
+			return -ENODEV;
+
+		read_index &= nentries_mask;
+		src_ring->hw_index = read_index;
 
 		ath10k_pci_sleep(ar);
 	}
 
 	read_index = src_ring->hw_index;
 
-	if ((read_index == sw_index) || (read_index == 0xffffffff))
+	if (read_index == sw_index)
 		return -EIO;
 
 	sbase = src_ring->shadow_base;
-- 
1.8.5.3