diff mbox series

[net,3/3] ice: avoid premature Rx buffer reuse

Message ID 20200825091629.12949-4-bjorn.topel@gmail.com
State Awaiting Upstream
Delegated to: David Miller
Headers show
Series Avoid premature Rx buffer reuse for XDP_REDIRECT | expand

Commit Message

Björn Töpel Aug. 25, 2020, 9:16 a.m. UTC 
From: Björn Töpel <bjorn.topel@intel.com>

The page recycle code, incorrectly, relied on that a page fragment
could not be freed inside xdp_do_redirect(). This assumption leads to
that page fragments that are used by the stack/XDP redirect can be
reused and overwritten.

To avoid this, store the page count prior invoking xdp_do_redirect().

Fixes: efc2214b6047 ("ice: Add support for XDP")
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_txrx.c | 31 ++++++++++++++++-------
 1 file changed, 22 insertions(+), 9 deletions(-)

Comments

Li RongQing Aug. 25, 2020, 9:55 a.m. UTC | #1 
> -----Original Message-----
> From: Björn Töpel [mailto:bjorn.topel@gmail.com]
> Sent: Tuesday, August 25, 2020 5:16 PM
> To: jeffrey.t.kirsher@intel.com; intel-wired-lan@lists.osuosl.org
> Cc: Björn Töpel <bjorn.topel@intel.com>; magnus.karlsson@intel.com;
> magnus.karlsson@gmail.com; netdev@vger.kernel.org;
> maciej.fijalkowski@intel.com; piotr.raczynski@intel.com;
> maciej.machnikowski@intel.com; Li,Rongqing <lirongqing@baidu.com>
> Subject: [PATCH net 3/3] ice: avoid premature Rx buffer reuse
> 
> From: Björn Töpel <bjorn.topel@intel.com>
> 
> The page recycle code, incorrectly, relied on that a page fragment could not be
> freed inside xdp_do_redirect(). This assumption leads to that page fragments
> that are used by the stack/XDP redirect can be reused and overwritten.
> 
> To avoid this, store the page count prior invoking xdp_do_redirect().
> 
> Fixes: efc2214b6047 ("ice: Add support for XDP")
> Signed-off-by: Björn Töpel <bjorn.topel@intel.com>


Reported-and-analyzed-by: Li RongQing <lirongqing@baidu.com>

Thanks

-Li
diff mbox series

Patch

diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index 9d0d6b0025cf..03a88c8f17b7 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -768,7 +768,8 @@  ice_rx_buf_adjust_pg_offset(struct ice_rx_buf *rx_buf, unsigned int size)
  * pointing to; otherwise, the DMA mapping needs to be destroyed and
  * page freed
  */
-static bool ice_can_reuse_rx_page(struct ice_rx_buf *rx_buf)
+static bool ice_can_reuse_rx_page(struct ice_rx_buf *rx_buf,
+				  int rx_buf_pgcnt)
 {
 	unsigned int pagecnt_bias = rx_buf->pagecnt_bias;
 	struct page *page = rx_buf->page;
@@ -779,7 +780,7 @@  static bool ice_can_reuse_rx_page(struct ice_rx_buf *rx_buf)
 
 #if (PAGE_SIZE < 8192)
 	/* if we are only owner of page we can reuse it */
-	if (unlikely((page_count(page) - pagecnt_bias) > 1))
+	if (unlikely((rx_buf_pgcnt - pagecnt_bias) > 1))
 		return false;
 #else
 #define ICE_LAST_OFFSET \
@@ -859,6 +860,15 @@  ice_reuse_rx_page(struct ice_ring *rx_ring, struct ice_rx_buf *old_buf)
 	new_buf->pagecnt_bias = old_buf->pagecnt_bias;
 }
 
+static int ice_rx_buf_page_count(struct ice_rx_buf *rx_buf)
+{
+#if (PAGE_SIZE < 8192)
+	return page_count(rx_buf->page);
+#else
+	return 0;
+#endif
+}
+
 /**
  * ice_get_rx_buf - Fetch Rx buffer and synchronize data for use
  * @rx_ring: Rx descriptor ring to transact packets on
@@ -870,11 +880,13 @@  ice_reuse_rx_page(struct ice_ring *rx_ring, struct ice_rx_buf *old_buf)
  */
 static struct ice_rx_buf *
 ice_get_rx_buf(struct ice_ring *rx_ring, struct sk_buff **skb,
-	       const unsigned int size)
+	       const unsigned int size,
+	       int *rx_buf_pgcnt)
 {
 	struct ice_rx_buf *rx_buf;
 
 	rx_buf = &rx_ring->rx_buf[rx_ring->next_to_clean];
+	*rx_buf_pgcnt = ice_rx_buf_page_count(rx_buf);
 	prefetchw(rx_buf->page);
 	*skb = rx_buf->skb;
 
@@ -1017,7 +1029,7 @@  ice_construct_skb(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf,
  * of the rx_buf. It will either recycle the buffer or unmap it and free
  * the associated resources.
  */
-static void ice_put_rx_buf(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf)
+static void ice_put_rx_buf(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, int rx_buf_pgcnt)
 {
 	u16 ntc = rx_ring->next_to_clean + 1;
 
@@ -1028,7 +1040,7 @@  static void ice_put_rx_buf(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf)
 	if (!rx_buf)
 		return;
 
-	if (ice_can_reuse_rx_page(rx_buf)) {
+	if (ice_can_reuse_rx_page(rx_buf, rx_buf_pgcnt)) {
 		/* hand second half of page back to the ring */
 		ice_reuse_rx_page(rx_ring, rx_buf);
 	} else {
@@ -1088,6 +1100,7 @@  int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
 	unsigned int xdp_res, xdp_xmit = 0;
 	struct bpf_prog *xdp_prog = NULL;
 	struct xdp_buff xdp;
+	int rx_buf_pgcnt;
 	bool failure;
 
 	xdp.rxq = &rx_ring->xdp_rxq;
@@ -1125,7 +1138,7 @@  int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
 		dma_rmb();
 
 		if (rx_desc->wb.rxdid == FDIR_DESC_RXDID || !rx_ring->netdev) {
-			ice_put_rx_buf(rx_ring, NULL);
+			ice_put_rx_buf(rx_ring, NULL, 0);
 			cleaned_count++;
 			continue;
 		}
@@ -1134,7 +1147,7 @@  int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
 			ICE_RX_FLX_DESC_PKT_LEN_M;
 
 		/* retrieve a buffer from the ring */
-		rx_buf = ice_get_rx_buf(rx_ring, &skb, size);
+		rx_buf = ice_get_rx_buf(rx_ring, &skb, size, &rx_buf_pgcnt);
 
 		if (!size) {
 			xdp.data = NULL;
@@ -1174,7 +1187,7 @@  int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
 		total_rx_pkts++;
 
 		cleaned_count++;
-		ice_put_rx_buf(rx_ring, rx_buf);
+		ice_put_rx_buf(rx_ring, rx_buf, rx_buf_pgcnt);
 		continue;
 construct_skb:
 		if (skb) {
@@ -1193,7 +1206,7 @@  int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
 			break;
 		}
 
-		ice_put_rx_buf(rx_ring, rx_buf);
+		ice_put_rx_buf(rx_ring, rx_buf, rx_buf_pgcnt);
 		cleaned_count++;
 
 		/* skip if it is NOP desc */