Merge branch 'ocelot-ptp'

Yangbo Lu says:

====================
Support Ocelot PTP Sync one-step timestamping

This patch-set is to support Ocelot PTP Sync one-step timestamping.
Actually before that, this patch-set cleans up and optimizes the
DSA slave tx timestamp request handling process.

Changes for v2:
	- Split tx timestamp optimization patch.
	- Updated doc patch.
	- Freed skb->cb usage in dsa core driver, and moved to device
	  drivers.
	- Other minor fixes.
Changes for v3:
	- Switched sequence of patch #3 and #4 with rebasing to fix build.
	- Replaced hard coded 48 of memset(skb->cb, 0, 48) with sizeof().
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

Documentation/networking/timestamping.rst

@@ -630,30 +630,45 @@ hardware timestamping on it. This is because the SO_TIMESTAMPING API does not
 allow the delivery of multiple hardware timestamps for the same packet, so
 anybody else except for the DSA switch port must be prevented from doing so.
 
-In code, DSA provides for most of the infrastructure for timestamping already,
-in generic code: a BPF classifier (``ptp_classify_raw``) is used to identify
-PTP event messages (any other packets, including PTP general messages, are not
-timestamped), and provides two hooks to drivers:
-
-- ``.port_txtstamp()``: The driver is passed a clone of the timestampable skb
-  to be transmitted, before actually transmitting it. Typically, a switch will
-  have a PTP TX timestamp register (or sometimes a FIFO) where the timestamp
-  becomes available. There may be an IRQ that is raised upon this timestamp's
-  availability, or the driver might have to poll after invoking
-  ``dev_queue_xmit()`` towards the host interface. Either way, in the
-  ``.port_txtstamp()`` method, the driver only needs to save the clone for
-  later use (when the timestamp becomes available). Each skb is annotated with
-  a pointer to its clone, in ``DSA_SKB_CB(skb)->clone``, to ease the driver's
-  job of keeping track of which clone belongs to which skb.
-
-- ``.port_rxtstamp()``: The original (and only) timestampable skb is provided
-  to the driver, for it to annotate it with a timestamp, if that is immediately
-  available, or defer to later. On reception, timestamps might either be
-  available in-band (through metadata in the DSA header, or attached in other
-  ways to the packet), or out-of-band (through another RX timestamping FIFO).
-  Deferral on RX is typically necessary when retrieving the timestamp needs a
-  sleepable context. In that case, it is the responsibility of the DSA driver
-  to call ``netif_rx_ni()`` on the freshly timestamped skb.
+In the generic layer, DSA provides the following infrastructure for PTP
+timestamping:
+
+- ``.port_txtstamp()``: a hook called prior to the transmission of
+  packets with a hardware TX timestamping request from user space.
+  This is required for two-step timestamping, since the hardware
+  timestamp becomes available after the actual MAC transmission, so the
+  driver must be prepared to correlate the timestamp with the original
+  packet so that it can re-enqueue the packet back into the socket's
+  error queue. To save the packet for when the timestamp becomes
+  available, the driver can call ``skb_clone_sk`` , save the clone pointer
+  in skb->cb and enqueue a tx skb queue. Typically, a switch will have a
+  PTP TX timestamp register (or sometimes a FIFO) where the timestamp
+  becomes available. In case of a FIFO, the hardware might store
+  key-value pairs of PTP sequence ID/message type/domain number and the
+  actual timestamp. To perform the correlation correctly between the
+  packets in a queue waiting for timestamping and the actual timestamps,
+  drivers can use a BPF classifier (``ptp_classify_raw``) to identify
+  the PTP transport type, and ``ptp_parse_header`` to interpret the PTP
+  header fields. There may be an IRQ that is raised upon this
+  timestamp's availability, or the driver might have to poll after
+  invoking ``dev_queue_xmit()`` towards the host interface.
+  One-step TX timestamping do not require packet cloning, since there is
+  no follow-up message required by the PTP protocol (because the
+  TX timestamp is embedded into the packet by the MAC), and therefore
+  user space does not expect the packet annotated with the TX timestamp
+  to be re-enqueued into its socket's error queue.
+
+- ``.port_rxtstamp()``: On RX, the BPF classifier is run by DSA to
+  identify PTP event messages (any other packets, including PTP general
+  messages, are not timestamped). The original (and only) timestampable
+  skb is provided to the driver, for it to annotate it with a timestamp,
+  if that is immediately available, or defer to later. On reception,
+  timestamps might either be available in-band (through metadata in the
+  DSA header, or attached in other ways to the packet), or out-of-band
+  (through another RX timestamping FIFO). Deferral on RX is typically
+  necessary when retrieving the timestamp needs a sleepable context. In
+  that case, it is the responsibility of the DSA driver to call
+  ``netif_rx_ni()`` on the freshly timestamped skb.
 
 3.2.2 Ethernet PHYs
 ^^^^^^^^^^^^^^^^^^^

drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c

@@ -373,30 +373,38 @@ long hellcreek_hwtstamp_work(struct ptp_clock_info *ptp)
 	return restart ? 1 : -1;
 }
 
-bool hellcreek_port_txtstamp(struct dsa_switch *ds, int port,
-			     struct sk_buff *clone, unsigned int type)
+void hellcreek_port_txtstamp(struct dsa_switch *ds, int port,
+			     struct sk_buff *skb)
 {
 	struct hellcreek *hellcreek = ds->priv;
 	struct hellcreek_port_hwtstamp *ps;
 	struct ptp_header *hdr;
+	struct sk_buff *clone;
+	unsigned int type;
 
 	ps = &hellcreek->ports[port].port_hwtstamp;
 
-	/* Check if the driver is expected to do HW timestamping */
-	if (!(skb_shinfo(clone)->tx_flags & SKBTX_HW_TSTAMP))
-		return false;
+	type = ptp_classify_raw(skb);
+	if (type == PTP_CLASS_NONE)
+		return;
 
 	/* Make sure the message is a PTP message that needs to be timestamped
 	 * and the interaction with the HW timestamping is enabled. If not, stop
 	 * here
 	 */
-	hdr = hellcreek_should_tstamp(hellcreek, port, clone, type);
+	hdr = hellcreek_should_tstamp(hellcreek, port, skb, type);
 	if (!hdr)
-		return false;
+		return;
+
+	clone = skb_clone_sk(skb);
+	if (!clone)
+		return;
 
 	if (test_and_set_bit_lock(HELLCREEK_HWTSTAMP_TX_IN_PROGRESS,
-				  &ps->state))
-		return false;
+				  &ps->state)) {
+		kfree_skb(clone);
+		return;
+	}
 
 	ps->tx_skb = clone;
 
@@ -406,8 +414,6 @@ bool hellcreek_port_txtstamp(struct dsa_switch *ds, int port,
 	ps->tx_tstamp_start = jiffies;
 
 	ptp_schedule_worker(hellcreek->ptp_clock, 0);
-
-	return true;
 }
 
 bool hellcreek_port_rxtstamp(struct dsa_switch *ds, int port,

drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h

@@ -44,8 +44,8 @@ int hellcreek_port_hwtstamp_get(struct dsa_switch *ds, int port,
 
 bool hellcreek_port_rxtstamp(struct dsa_switch *ds, int port,
 			     struct sk_buff *clone, unsigned int type);
-bool hellcreek_port_txtstamp(struct dsa_switch *ds, int port,
-			     struct sk_buff *clone, unsigned int type);
+void hellcreek_port_txtstamp(struct dsa_switch *ds, int port,
+			     struct sk_buff *skb);
 
 int hellcreek_get_ts_info(struct dsa_switch *ds, int port,
 			  struct ethtool_ts_info *info);

drivers/net/dsa/mv88e6xxx/hwtstamp.c

@@ -468,30 +468,38 @@ long mv88e6xxx_hwtstamp_work(struct ptp_clock_info *ptp)
 	return restart ? 1 : -1;
 }
 
-bool mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
-			     struct sk_buff *clone, unsigned int type)
+void mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
+			     struct sk_buff *skb)
 {
 	struct mv88e6xxx_chip *chip = ds->priv;
 	struct mv88e6xxx_port_hwtstamp *ps = &chip->port_hwtstamp[port];
 	struct ptp_header *hdr;
+	struct sk_buff *clone;
+	unsigned int type;
 
-	if (!(skb_shinfo(clone)->tx_flags & SKBTX_HW_TSTAMP))
-		return false;
+	type = ptp_classify_raw(skb);
+	if (type == PTP_CLASS_NONE)
+		return;
 
-	hdr = mv88e6xxx_should_tstamp(chip, port, clone, type);
+	hdr = mv88e6xxx_should_tstamp(chip, port, skb, type);
 	if (!hdr)
-		return false;
+		return;
+
+	clone = skb_clone_sk(skb);
+	if (!clone)
+		return;
 
 	if (test_and_set_bit_lock(MV88E6XXX_HWTSTAMP_TX_IN_PROGRESS,
-				  &ps->state))
-		return false;
+				  &ps->state)) {
+		kfree_skb(clone);
+		return;
+	}
 
 	ps->tx_skb = clone;
 	ps->tx_tstamp_start = jiffies;
 	ps->tx_seq_id = be16_to_cpu(hdr->sequence_id);
 
 	ptp_schedule_worker(chip->ptp_clock, 0);
-	return true;
 }
 
 int mv88e6165_global_disable(struct mv88e6xxx_chip *chip)

drivers/net/dsa/mv88e6xxx/hwtstamp.h

@@ -117,8 +117,8 @@ int mv88e6xxx_port_hwtstamp_get(struct dsa_switch *ds, int port,
 
 bool mv88e6xxx_port_rxtstamp(struct dsa_switch *ds, int port,
 			     struct sk_buff *clone, unsigned int type);
-bool mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
-			     struct sk_buff *clone, unsigned int type);
+void mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
+			     struct sk_buff *skb);
 
 int mv88e6xxx_get_ts_info(struct dsa_switch *ds, int port,
 			  struct ethtool_ts_info *info);
@@ -151,11 +151,9 @@ static inline bool mv88e6xxx_port_rxtstamp(struct dsa_switch *ds, int port,
 	return false;
 }
 
-static inline bool mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
-					   struct sk_buff *clone,
-					   unsigned int type)
+static inline void mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
+					   struct sk_buff *skb)
 {
-	return false;
 }
 
 static inline int mv88e6xxx_get_ts_info(struct dsa_switch *ds, int port,

drivers/net/dsa/ocelot/felix.c

@@ -1395,19 +1395,20 @@ static bool felix_rxtstamp(struct dsa_switch *ds, int port,
 	return false;
 }
 
-static bool felix_txtstamp(struct dsa_switch *ds, int port,
-			   struct sk_buff *clone, unsigned int type)
+static void felix_txtstamp(struct dsa_switch *ds, int port,
+			   struct sk_buff *skb)
 {
 	struct ocelot *ocelot = ds->priv;
-	struct ocelot_port *ocelot_port = ocelot->ports[port];
+	struct sk_buff *clone = NULL;
 
-	if (ocelot->ptp && (skb_shinfo(clone)->tx_flags & SKBTX_HW_TSTAMP) &&
-	    ocelot_port->ptp_cmd == IFH_REW_OP_TWO_STEP_PTP) {
-		ocelot_port_add_txtstamp_skb(ocelot, port, clone);
-		return true;
-	}
+	if (!ocelot->ptp)
+		return;
 
-	return false;
+	if (ocelot_port_txtstamp_request(ocelot, port, skb, &clone))
+		return;
+
+	if (clone)
+		OCELOT_SKB_CB(skb)->clone = clone;
 }
 
 static int felix_change_mtu(struct dsa_switch *ds, int port, int new_mtu)

drivers/net/dsa/sja1105/sja1105_main.c

@@ -3137,7 +3137,7 @@ static void sja1105_port_deferred_xmit(struct kthread_work *work)
 	struct sk_buff *skb;
 
 	while ((skb = skb_dequeue(&sp->xmit_queue)) != NULL) {
-		struct sk_buff *clone = DSA_SKB_CB(skb)->clone;
+		struct sk_buff *clone = SJA1105_SKB_CB(skb)->clone;
 
 		mutex_lock(&priv->mgmt_lock);
 

drivers/net/dsa/sja1105/sja1105_ptp.c

@@ -431,20 +431,24 @@ bool sja1105_port_rxtstamp(struct dsa_switch *ds, int port,
 	return true;
 }
 
-/* Called from dsa_skb_tx_timestamp. This callback is just to make DSA clone
- * the skb and have it available in DSA_SKB_CB in the .port_deferred_xmit
+/* Called from dsa_skb_tx_timestamp. This callback is just to clone
+ * the skb and have it available in SJA1105_SKB_CB in the .port_deferred_xmit
  * callback, where we will timestamp it synchronously.
  */
-bool sja1105_port_txtstamp(struct dsa_switch *ds, int port,
-			   struct sk_buff *skb, unsigned int type)
+void sja1105_port_txtstamp(struct dsa_switch *ds, int port, struct sk_buff *skb)
 {
 	struct sja1105_private *priv = ds->priv;
 	struct sja1105_port *sp = &priv->ports[port];
+	struct sk_buff *clone;
 
 	if (!sp->hwts_tx_en)
-		return false;
+		return;
 
-	return true;
+	clone = skb_clone_sk(skb);
+	if (!clone)
+		return;
+
+	SJA1105_SKB_CB(skb)->clone = clone;
 }
 
 static int sja1105_ptp_reset(struct dsa_switch *ds)

drivers/net/dsa/sja1105/sja1105_ptp.h

@@ -104,8 +104,8 @@ void sja1105_ptp_txtstamp_skb(struct dsa_switch *ds, int slot,
 bool sja1105_port_rxtstamp(struct dsa_switch *ds, int port,
 			   struct sk_buff *skb, unsigned int type);
 
-bool sja1105_port_txtstamp(struct dsa_switch *ds, int port,
-			   struct sk_buff *skb, unsigned int type);
+void sja1105_port_txtstamp(struct dsa_switch *ds, int port,
+			   struct sk_buff *skb);
 
 int sja1105_hwtstamp_get(struct dsa_switch *ds, int port, struct ifreq *ifr);