Merge branch 'ethtool-fec'

Roopa Prabhu says:

====================
ethtool: support for forward error correction mode setting on a link

Forward Error Correction (FEC) modes i.e Base-R
and Reed-Solomon modes are introduced in 25G/40G/100G standards
for providing good BER at high speeds. Various networking devices
which support 25G/40G/100G provides ability to manage supported FEC
modes and the lack of FEC encoding control and reporting today is a
source for interoperability issues for many vendors.
FEC capability as well as specific FEC mode i.e. Base-R
or RS modes can be requested or advertised through bits D44:47 of base link
codeword.

This patch set intends to provide option under ethtool to manage and
report FEC encoding settings for networking devices as per IEEE 802.3
bj, bm and by specs.

v2 :
        - minor patch format fixes and typos pointed out by Andrew
        - there was a pending discussion on the use of 'auto' vs
          'automatic' for fec settings. I have left it as 'auto'
          because in most cases today auto is used in place of
          automatic to represent automatically generated values.
          We use it in other networking config too. I would prefer
          leaving it as auto.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c

@@ -801,6 +801,104 @@ static int set_link_ksettings(struct net_device *dev,
 	return ret;
 }
 
+/* Translate the Firmware FEC value into the ethtool value. */
+static inline unsigned int fwcap_to_eth_fec(unsigned int fw_fec)
+{
+	unsigned int eth_fec = 0;
+
+	if (fw_fec & FW_PORT_CAP_FEC_RS)
+		eth_fec |= ETHTOOL_FEC_RS;
+	if (fw_fec & FW_PORT_CAP_FEC_BASER_RS)
+		eth_fec |= ETHTOOL_FEC_BASER;
+
+	/* if nothing is set, then FEC is off */
+	if (!eth_fec)
+		eth_fec = ETHTOOL_FEC_OFF;
+
+	return eth_fec;
+}
+
+/* Translate Common Code FEC value into ethtool value. */
+static inline unsigned int cc_to_eth_fec(unsigned int cc_fec)
+{
+	unsigned int eth_fec = 0;
+
+	if (cc_fec & FEC_AUTO)
+		eth_fec |= ETHTOOL_FEC_AUTO;
+	if (cc_fec & FEC_RS)
+		eth_fec |= ETHTOOL_FEC_RS;
+	if (cc_fec & FEC_BASER_RS)
+		eth_fec |= ETHTOOL_FEC_BASER;
+
+	/* if nothing is set, then FEC is off */
+	if (!eth_fec)
+		eth_fec = ETHTOOL_FEC_OFF;
+
+	return eth_fec;
+}
+
+/* Translate ethtool FEC value into Common Code value. */
+static inline unsigned int eth_to_cc_fec(unsigned int eth_fec)
+{
+	unsigned int cc_fec = 0;
+
+	if (eth_fec & ETHTOOL_FEC_OFF)
+		return cc_fec;
+
+	if (eth_fec & ETHTOOL_FEC_AUTO)
+		cc_fec |= FEC_AUTO;
+	if (eth_fec & ETHTOOL_FEC_RS)
+		cc_fec |= FEC_RS;
+	if (eth_fec & ETHTOOL_FEC_BASER)
+		cc_fec |= FEC_BASER_RS;
+
+	return cc_fec;
+}
+
+static int get_fecparam(struct net_device *dev, struct ethtool_fecparam *fec)
+{
+	const struct port_info *pi = netdev_priv(dev);
+	const struct link_config *lc = &pi->link_cfg;
+
+	/* Translate the Firmware FEC Support into the ethtool value.  We
+	 * always support IEEE 802.3 "automatic" selection of Link FEC type if
+	 * any FEC is supported.
+	 */
+	fec->fec = fwcap_to_eth_fec(lc->supported);
+	if (fec->fec != ETHTOOL_FEC_OFF)
+		fec->fec |= ETHTOOL_FEC_AUTO;
+
+	/* Translate the current internal FEC parameters into the
+	 * ethtool values.
+	 */
+	fec->active_fec = cc_to_eth_fec(lc->fec);
+
+	return 0;
+}
+
+static int set_fecparam(struct net_device *dev, struct ethtool_fecparam *fec)
+{
+	struct port_info *pi = netdev_priv(dev);
+	struct link_config *lc = &pi->link_cfg;
+	struct link_config old_lc;
+	int ret;
+
+	/* Save old Link Configuration in case the L1 Configure below
+	 * fails.
+	 */
+	old_lc = *lc;
+
+	/* Try to perform the L1 Configure and return the result of that
+	 * effort.  If it fails, revert the attempted change.
+	 */
+	lc->requested_fec = eth_to_cc_fec(fec->fec);
+	ret = t4_link_l1cfg(pi->adapter, pi->adapter->mbox,
+			    pi->tx_chan, lc);
+	if (ret)
+		*lc = old_lc;
+	return ret;
+}
+
 static void get_pauseparam(struct net_device *dev,
 			   struct ethtool_pauseparam *epause)
 {
@@ -1255,6 +1353,8 @@ static int get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info,
 static const struct ethtool_ops cxgb_ethtool_ops = {
 	.get_link_ksettings = get_link_ksettings,
 	.set_link_ksettings = set_link_ksettings,
+	.get_fecparam      = get_fecparam,
+	.set_fecparam      = set_fecparam,
 	.get_drvinfo       = get_drvinfo,
 	.get_msglevel      = get_msglevel,
 	.set_msglevel      = set_msglevel,

drivers/net/ethernet/chelsio/cxgb4/t4_hw.c

@@ -3840,11 +3840,64 @@ void t4_ulprx_read_la(struct adapter *adap, u32 *la_buf)
 		     FW_PORT_CAP_SPEED_40G | FW_PORT_CAP_SPEED_100G | \
 		     FW_PORT_CAP_ANEG)
 
+/* Translate Firmware Port Capabilities Pause specification to Common Code */
+static inline unsigned int fwcap_to_cc_pause(unsigned int fw_pause)
+{
+	unsigned int cc_pause = 0;
+
+	if (fw_pause & FW_PORT_CAP_FC_RX)
+		cc_pause |= PAUSE_RX;
+	if (fw_pause & FW_PORT_CAP_FC_TX)
+		cc_pause |= PAUSE_TX;
+
+	return cc_pause;
+}
+
+/* Translate Common Code Pause specification into Firmware Port Capabilities */
+static inline unsigned int cc_to_fwcap_pause(unsigned int cc_pause)
+{
+	unsigned int fw_pause = 0;
+
+	if (cc_pause & PAUSE_RX)
+		fw_pause |= FW_PORT_CAP_FC_RX;
+	if (cc_pause & PAUSE_TX)
+		fw_pause |= FW_PORT_CAP_FC_TX;
+
+	return fw_pause;
+}
+
+/* Translate Firmware Forward Error Correction specification to Common Code */
+static inline unsigned int fwcap_to_cc_fec(unsigned int fw_fec)
+{
+	unsigned int cc_fec = 0;
+
+	if (fw_fec & FW_PORT_CAP_FEC_RS)
+		cc_fec |= FEC_RS;
+	if (fw_fec & FW_PORT_CAP_FEC_BASER_RS)
+		cc_fec |= FEC_BASER_RS;
+
+	return cc_fec;
+}
+
+/* Translate Common Code Forward Error Correction specification to Firmware */
+static inline unsigned int cc_to_fwcap_fec(unsigned int cc_fec)
+{
+	unsigned int fw_fec = 0;
+
+	if (cc_fec & FEC_RS)
+		fw_fec |= FW_PORT_CAP_FEC_RS;
+	if (cc_fec & FEC_BASER_RS)
+		fw_fec |= FW_PORT_CAP_FEC_BASER_RS;
+
+	return fw_fec;
+}
+
 /**
  *	t4_link_l1cfg - apply link configuration to MAC/PHY
- *	@phy: the PHY to setup
- *	@mac: the MAC to setup
- *	@lc: the requested link configuration
+ *	@adapter: the adapter
+ *	@mbox: the Firmware Mailbox to use
+ *	@port: the Port ID
+ *	@lc: the Port's Link Configuration
  *
  *	Set up a port's MAC and PHY according to a desired link configuration.
  *	- If the PHY can auto-negotiate first decide what to advertise, then
@@ -3857,22 +3910,46 @@ int t4_link_l1cfg(struct adapter *adap, unsigned int mbox, unsigned int port,
 		  struct link_config *lc)
 {
 	struct fw_port_cmd c;
-	unsigned int mdi = FW_PORT_CAP_MDI_V(FW_PORT_CAP_MDI_AUTO);
-	unsigned int fc = 0, fec = 0, fw_fec = 0;
+	unsigned int fw_mdi = FW_PORT_CAP_MDI_V(FW_PORT_CAP_MDI_AUTO);
+	unsigned int fw_fc, cc_fec, fw_fec;
+	unsigned int rcap;
 
 	lc->link_ok = 0;
-	if (lc->requested_fc & PAUSE_RX)
-		fc |= FW_PORT_CAP_FC_RX;
-	if (lc->requested_fc & PAUSE_TX)
-		fc |= FW_PORT_CAP_FC_TX;
 
-	fec = lc->requested_fec & FEC_AUTO ? lc->auto_fec : lc->requested_fec;
+	/* Convert driver coding of Pause Frame Flow Control settings into the
+	 * Firmware's API.
+	 */
+	fw_fc = cc_to_fwcap_pause(lc->requested_fc);
+
+	/* Convert Common Code Forward Error Control settings into the
+	 * Firmware's API.  If the current Requested FEC has "Automatic"
+	 * (IEEE 802.3) specified, then we use whatever the Firmware
+	 * sent us as part of it's IEEE 802.3-based interpratation of
+	 * the Transceiver Module EPROM FEC parameters.  Otherwise we
+	 * use whatever is in the current Requested FEC settings.
+	 */
+	if (lc->requested_fec & FEC_AUTO)
+		cc_fec = lc->auto_fec;
+	else
+		cc_fec = lc->requested_fec;
+	fw_fec = cc_to_fwcap_fec(cc_fec);
 
-	if (fec & FEC_RS)
-		fw_fec |= FW_PORT_CAP_FEC_RS;
-	if (fec & FEC_BASER_RS)
-		fw_fec |= FW_PORT_CAP_FEC_BASER_RS;
+	/* Figure out what our Requested Port Capabilities are going to be.
+	 */
+	if (!(lc->supported & FW_PORT_CAP_ANEG)) {
+		rcap = (lc->supported & ADVERT_MASK) | fw_fc | fw_fec;
+		lc->fc = lc->requested_fc & (PAUSE_RX | PAUSE_TX);
+		lc->fec = cc_fec;
+	} else if (lc->autoneg == AUTONEG_DISABLE) {
+		rcap = lc->requested_speed | fw_fc | fw_fec | fw_mdi;
+		lc->fc = lc->requested_fc & (PAUSE_RX | PAUSE_TX);
+		lc->fec = cc_fec;
+	} else {
+		rcap = lc->advertising | fw_fc | fw_fec | fw_mdi;
+	}
 
+	/* And send that on to the Firmware ...
+	 */
 	memset(&c, 0, sizeof(c));
 	c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) |
 				     FW_CMD_REQUEST_F | FW_CMD_EXEC_F |
@@ -3880,19 +3957,7 @@ int t4_link_l1cfg(struct adapter *adap, unsigned int mbox, unsigned int port,
 	c.action_to_len16 =
 		cpu_to_be32(FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_L1_CFG) |
 			    FW_LEN16(c));
-
-	if (!(lc->supported & FW_PORT_CAP_ANEG)) {
-		c.u.l1cfg.rcap = cpu_to_be32((lc->supported & ADVERT_MASK) |
-					     fc | fw_fec);
-		lc->fc = lc->requested_fc & (PAUSE_RX | PAUSE_TX);
-	} else if (lc->autoneg == AUTONEG_DISABLE) {
-		c.u.l1cfg.rcap = cpu_to_be32(lc->requested_speed | fc |
-					     fw_fec | mdi);
-		lc->fc = lc->requested_fc & (PAUSE_RX | PAUSE_TX);
-	} else
-		c.u.l1cfg.rcap = cpu_to_be32(lc->advertising | fc |
-					     fw_fec | mdi);
-
+	c.u.l1cfg.rcap = cpu_to_be32(rcap);
 	return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL);
 }
 
@@ -7630,19 +7695,28 @@ static const char *t4_link_down_rc_str(unsigned char link_down_rc)
 void t4_handle_get_port_info(struct port_info *pi, const __be64 *rpl)
 {
 	const struct fw_port_cmd *p = (const void *)rpl;
+	unsigned int acaps = be16_to_cpu(p->u.info.acap);
 	struct adapter *adap = pi->adapter;
 
 	/* link/module state change message */
-	int speed = 0, fc = 0;
+	int speed = 0, fc, fec;
 	struct link_config *lc;
 	u32 stat = be32_to_cpu(p->u.info.lstatus_to_modtype);
 	int link_ok = (stat & FW_PORT_CMD_LSTATUS_F) != 0;
 	u32 mod = FW_PORT_CMD_MODTYPE_G(stat);
 
+	/* Unfortunately the format of the Link Status returned by the
+	 * Firmware isn't the same as the Firmware Port Capabilities bitfield
+	 * used everywhere else ...
+	 */
+	fc = 0;
 	if (stat & FW_PORT_CMD_RXPAUSE_F)
 		fc |= PAUSE_RX;
 	if (stat & FW_PORT_CMD_TXPAUSE_F)
 		fc |= PAUSE_TX;
+
+	fec = fwcap_to_cc_fec(acaps);
+
 	if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_100M))
 		speed = 100;
 	else if (stat & FW_PORT_CMD_LSPEED_V(FW_PORT_CAP_SPEED_1G))
@@ -7659,11 +7733,20 @@ void t4_handle_get_port_info(struct port_info *pi, const __be64 *rpl)
 	lc = &pi->link_cfg;
 
 	if (mod != pi->mod_type) {
+		/* When a new Transceiver Module is inserted, the Firmware
+		 * will examine any Forward Error Correction parameters
+		 * present in the Transceiver Module i2c EPROM and determine
+		 * the supported and recommended FEC settings from those
+		 * based on IEEE 802.3 standards.  We always record the
+		 * IEEE 802.3 recommended "automatic" settings.
+		 */
+		lc->auto_fec = fec;
+
 		pi->mod_type = mod;
 		t4_os_portmod_changed(adap, pi->port_id);
 	}
 	if (link_ok != lc->link_ok || speed != lc->speed ||
-	    fc != lc->fc) {	/* something changed */
+	    fc != lc->fc || fec != lc->fec) {	/* something changed */
 		if (!link_ok && lc->link_ok) {
 			unsigned char rc = FW_PORT_CMD_LINKDNRC_G(stat);
 
@@ -7675,6 +7758,8 @@ void t4_handle_get_port_info(struct port_info *pi, const __be64 *rpl)
 		lc->link_ok = link_ok;
 		lc->speed = speed;
 		lc->fc = fc;
+		lc->fec = fec;
+
 		lc->supported = be16_to_cpu(p->u.info.pcap);
 		lc->lp_advertising = be16_to_cpu(p->u.info.lpacap);
 
@@ -7764,7 +7849,8 @@ static void get_pci_mode(struct adapter *adapter, struct pci_params *p)
 /**
  *	init_link_config - initialize a link's SW state
  *	@lc: structure holding the link state
- *	@caps: link capabilities
+ *	@pcaps: link Port Capabilities
+ *	@acaps: link current Advertised Port Capabilities
  *
  *	Initializes the SW state maintained for each link, including the link's
  *	capabilities and default speed/flow-control/autonegotiation settings.
@@ -7777,15 +7863,11 @@ static void init_link_config(struct link_config *lc, unsigned int pcaps,
 	lc->requested_speed = 0;
 	lc->speed = 0;
 	lc->requested_fc = lc->fc = PAUSE_RX | PAUSE_TX;
-	lc->auto_fec = 0;
 
 	/* For Forward Error Control, we default to whatever the Firmware
 	 * tells us the Link is currently advertising.
 	 */
-	if (acaps & FW_PORT_CAP_FEC_RS)
-		lc->auto_fec |= FEC_RS;
-	if (acaps & FW_PORT_CAP_FEC_BASER_RS)
-		lc->auto_fec |= FEC_BASER_RS;
+	lc->auto_fec = fwcap_to_cc_fec(acaps);
 	lc->requested_fec = FEC_AUTO;
 	lc->fec = lc->auto_fec;
 

include/linux/ethtool.h

@@ -374,5 +374,9 @@ struct ethtool_ops {
 				      struct ethtool_link_ksettings *);
 	int	(*set_link_ksettings)(struct net_device *,
 				      const struct ethtool_link_ksettings *);
+	int	(*get_fecparam)(struct net_device *,
+				      struct ethtool_fecparam *);
+	int	(*set_fecparam)(struct net_device *,
+				      struct ethtool_fecparam *);
 };
 #endif /* _LINUX_ETHTOOL_H */

include/uapi/linux/ethtool.h

@@ -1238,6 +1238,47 @@ struct ethtool_per_queue_op {
 	char	data[];
 };
 
+/**
+ * struct ethtool_fecparam - Ethernet forward error correction(fec) parameters
+ * @cmd: Command number = %ETHTOOL_GFECPARAM or %ETHTOOL_SFECPARAM
+ * @active_fec: FEC mode which is active on porte
+ * @fec: Bitmask of supported/configured FEC modes
+ * @rsvd: Reserved for future extensions. i.e FEC bypass feature.
+ *
+ * Drivers should reject a non-zero setting of @autoneg when
+ * autoneogotiation is disabled (or not supported) for the link.
+ *
+ */
+struct ethtool_fecparam {
+	__u32   cmd;
+	/* bitmask of FEC modes */
+	__u32   active_fec;
+	__u32   fec;
+	__u32   reserved;
+};
+
+/**
+ * enum ethtool_fec_config_bits - flags definition of ethtool_fec_configuration
+ * @ETHTOOL_FEC_NONE: FEC mode configuration is not supported
+ * @ETHTOOL_FEC_AUTO: Default/Best FEC mode provided by driver
+ * @ETHTOOL_FEC_OFF: No FEC Mode
+ * @ETHTOOL_FEC_RS: Reed-Solomon Forward Error Detection mode
+ * @ETHTOOL_FEC_BASER: Base-R/Reed-Solomon Forward Error Detection mode
+ */
+enum ethtool_fec_config_bits {
+	ETHTOOL_FEC_NONE_BIT,
+	ETHTOOL_FEC_AUTO_BIT,
+	ETHTOOL_FEC_OFF_BIT,
+	ETHTOOL_FEC_RS_BIT,
+	ETHTOOL_FEC_BASER_BIT,
+};
+
+#define ETHTOOL_FEC_NONE		(1 << ETHTOOL_FEC_NONE_BIT)
+#define ETHTOOL_FEC_AUTO		(1 << ETHTOOL_FEC_AUTO_BIT)
+#define ETHTOOL_FEC_OFF			(1 << ETHTOOL_FEC_OFF_BIT)
+#define ETHTOOL_FEC_RS			(1 << ETHTOOL_FEC_RS_BIT)
+#define ETHTOOL_FEC_BASER		(1 << ETHTOOL_FEC_BASER_BIT)
+
 /* CMDs currently supported */
 #define ETHTOOL_GSET		0x00000001 /* DEPRECATED, Get settings.
 					    * Please use ETHTOOL_GLINKSETTINGS
@@ -1330,6 +1371,8 @@ struct ethtool_per_queue_op {
 #define ETHTOOL_SLINKSETTINGS	0x0000004d /* Set ethtool_link_settings */
 #define ETHTOOL_PHY_GTUNABLE	0x0000004e /* Get PHY tunable configuration */
 #define ETHTOOL_PHY_STUNABLE	0x0000004f /* Set PHY tunable configuration */
+#define ETHTOOL_GFECPARAM	0x00000050 /* Get FEC settings */
+#define ETHTOOL_SFECPARAM	0x00000051 /* Set FEC settings */
 
 /* compatibility with older code */
 #define SPARC_ETH_GSET		ETHTOOL_GSET
@@ -1387,6 +1430,9 @@ enum ethtool_link_mode_bit_indices {
 	ETHTOOL_LINK_MODE_2500baseT_Full_BIT	= 47,
 	ETHTOOL_LINK_MODE_5000baseT_Full_BIT	= 48,
 
+	ETHTOOL_LINK_MODE_FEC_NONE_BIT	= 49,
+	ETHTOOL_LINK_MODE_FEC_RS_BIT	= 50,
+	ETHTOOL_LINK_MODE_FEC_BASER_BIT	= 51,
 
 	/* Last allowed bit for __ETHTOOL_LINK_MODE_LEGACY_MASK is bit
 	 * 31. Please do NOT define any SUPPORTED_* or ADVERTISED_*
@@ -1395,7 +1441,7 @@ enum ethtool_link_mode_bit_indices {
 	 */
 
 	__ETHTOOL_LINK_MODE_LAST
-	  = ETHTOOL_LINK_MODE_5000baseT_Full_BIT,
+	  = ETHTOOL_LINK_MODE_FEC_BASER_BIT,
 };
 
 #define __ETHTOOL_LINK_MODE_LEGACY_MASK(base_name)	\

net/core/ethtool.c

@@ -2512,6 +2512,33 @@ static int set_phy_tunable(struct net_device *dev, void __user *useraddr)
 	return ret;
 }
 
+static int ethtool_get_fecparam(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_fecparam fecparam = { ETHTOOL_GFECPARAM };
+
+	if (!dev->ethtool_ops->get_fecparam)
+		return -EOPNOTSUPP;
+
+	dev->ethtool_ops->get_fecparam(dev, &fecparam);
+
+	if (copy_to_user(useraddr, &fecparam, sizeof(fecparam)))
+		return -EFAULT;
+	return 0;
+}
+
+static int ethtool_set_fecparam(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_fecparam fecparam;
+
+	if (!dev->ethtool_ops->set_fecparam)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&fecparam, useraddr, sizeof(fecparam)))
+		return -EFAULT;
+
+	return dev->ethtool_ops->set_fecparam(dev, &fecparam);
+}
+
 /* The main entry point in this file.  Called from net/core/dev_ioctl.c */
 
 int dev_ethtool(struct net *net, struct ifreq *ifr)
@@ -2570,6 +2597,7 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	case ETHTOOL_GTUNABLE:
 	case ETHTOOL_PHY_GTUNABLE:
 	case ETHTOOL_GLINKSETTINGS:
+	case ETHTOOL_GFECPARAM:
 		break;
 	default:
 		if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
@@ -2779,6 +2807,12 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	case ETHTOOL_PHY_STUNABLE:
 		rc = set_phy_tunable(dev, useraddr);
 		break;
+	case ETHTOOL_GFECPARAM:
+		rc = ethtool_get_fecparam(dev, useraddr);
+		break;
+	case ETHTOOL_SFECPARAM:
+		rc = ethtool_set_fecparam(dev, useraddr);
+		break;
 	default:
 		rc = -EOPNOTSUPP;
 	}