[PATCH net] ixgbe: reserve TX timestamp after skb preparation

[Eric Dumazet <edumazet@kernel.org>]

ixgbe_xmit_frame_ring() takes an extra skb reference with skb_get()
for the pending TX timestamp before preparing the skb for transmit.
Several of the following steps may need to reallocate the skb head:

- ixgbe_tso() calls skb_cow_head(), which reaches pskb_expand_head()
  when the header is cloned (e.g. a packet tap holds a clone).
- ixgbe_tx_csum() falls back to skb_checksum_help() for unsupported
  checksum offsets, which can call pskb_expand_head() through
  skb_ensure_writable().
- With DCB and a software VLAN tag, skb_cow_head() is called to
  rewrite the VLAN priority.

pskb_expand_head() used to BUG() on shared skbs. Since commit
c7d8bc3dd511 ("net: skbuff: don't BUG() on bad arguments to
pskb_expand_head()") it returns -EINVAL instead, so these packets
are dropped (TSO, DCB) or sent with an unfinished checksum
(skb_checksum_help() errors are ignored by ixgbe_tx_csum()).

Reserve the TX timestamp right before ixgbe_tx_map(), after all skb
preparation, as i40e and ice already do. The timestamp request
remains covered on the FCoE path, and the cleanup on ixgbe_tx_map()
failure is unchanged.

Fixes: 891dc0821d20 ("ixgbe: Add ptp work item to poll for the Tx timestamp")
Cc: stable@vger.kernel.org
Signed-off-by: Eric Dumazet <edumazet@kernel.org>
---
Cc: Tony Nguyen <anthony.l.nguyen@intel.com>
Cc: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Cc: Andrew Lunn <andrew+netdev@lunn.ch>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 40 +++++++++++--------
 1 file changed, 23 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index f91856498eb2d9e0e4173153c8c00bc2248bb708..b9496f0bee1312499f28c2f3c24fcff9a1e66e0d 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -9444,23 +9444,6 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb,
 	}
 	protocol = vlan_get_protocol(skb);
 
-	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
-	    adapter->ptp_clock) {
-		if (adapter->tstamp_config.tx_type == HWTSTAMP_TX_ON &&
-		    !test_and_set_bit_lock(__IXGBE_PTP_TX_IN_PROGRESS,
-					   &adapter->state)) {
-			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
-			tx_flags |= IXGBE_TX_FLAGS_TSTAMP;
-
-			/* schedule check for Tx timestamp */
-			adapter->ptp_tx_skb = skb_get(skb);
-			adapter->ptp_tx_start = jiffies;
-			schedule_work(&adapter->ptp_tx_work);
-		} else {
-			adapter->tx_hwtstamp_skipped++;
-		}
-	}
-
 #ifdef CONFIG_PCI_IOV
 	/*
 	 * Use the l2switch_enable flag - would be false if the DMA
@@ -9525,6 +9508,29 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb,
 #ifdef IXGBE_FCOE
 xmit_fcoe:
 #endif /* IXGBE_FCOE */
+	/* Reserve the Tx timestamp only after all skb preparation:
+	 * skb_get() makes the skb shared, and pskb_expand_head()
+	 * (reached from skb_cow_head() or skb_checksum_help())
+	 * refuses shared skbs.
+	 */
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+	    adapter->ptp_clock) {
+		if (adapter->tstamp_config.tx_type == HWTSTAMP_TX_ON &&
+		    !test_and_set_bit_lock(__IXGBE_PTP_TX_IN_PROGRESS,
+					   &adapter->state)) {
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+			tx_flags |= IXGBE_TX_FLAGS_TSTAMP;
+			first->tx_flags |= IXGBE_TX_FLAGS_TSTAMP;
+
+			/* schedule check for Tx timestamp */
+			adapter->ptp_tx_skb = skb_get(skb);
+			adapter->ptp_tx_start = jiffies;
+			schedule_work(&adapter->ptp_tx_work);
+		} else {
+			adapter->tx_hwtstamp_skipped++;
+		}
+	}
+
 	if (ixgbe_tx_map(tx_ring, first, hdr_len))
 		goto cleanup_tx_timestamp;
 
-- 
2.53.0