Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:
 "Highlights:

   1) Maintain the TCP retransmit queue using an rbtree, with 1GB
      windows at 100Gb this really has become necessary. From Eric
      Dumazet.

   2) Multi-program support for cgroup+bpf, from Alexei Starovoitov.

   3) Perform broadcast flooding in hardware in mv88e6xxx, from Andrew
      Lunn.

   4) Add meter action support to openvswitch, from Andy Zhou.

   5) Add a data meta pointer for BPF accessible packets, from Daniel
      Borkmann.

   6) Namespace-ify almost all TCP sysctl knobs, from Eric Dumazet.

   7) Turn on Broadcom Tags in b53 driver, from Florian Fainelli.

   8) More work to move the RTNL mutex down, from Florian Westphal.

   9) Add 'bpftool' utility, to help with bpf program introspection.
      From Jakub Kicinski.

  10) Add new 'cpumap' type for XDP_REDIRECT action, from Jesper
      Dangaard Brouer.

  11) Support 'blocks' of transformations in the packet scheduler which
      can span multiple network devices, from Jiri Pirko.

  12) TC flower offload support in cxgb4, from Kumar Sanghvi.

  13) Priority based stream scheduler for SCTP, from Marcelo Ricardo
      Leitner.

  14) Thunderbolt networking driver, from Amir Levy and Mika Westerberg.

  15) Add RED qdisc offloadability, and use it in mlxsw driver. From
      Nogah Frankel.

  16) eBPF based device controller for cgroup v2, from Roman Gushchin.

  17) Add some fundamental tracepoints for TCP, from Song Liu.

  18) Remove garbage collection from ipv6 route layer, this is a
      significant accomplishment. From Wei Wang.

  19) Add multicast route offload support to mlxsw, from Yotam Gigi"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (2177 commits)
  tcp: highest_sack fix
  geneve: fix fill_info when link down
  bpf: fix lockdep splat
  net: cdc_ncm: GetNtbFormat endian fix
  openvswitch: meter: fix NULL pointer dereference in ovs_meter_cmd_reply_start
  netem: remove unnecessary 64 bit modulus
  netem: use 64 bit divide by rate
  tcp: Namespace-ify sysctl_tcp_default_congestion_control
  net: Protect iterations over net::fib_notifier_ops in fib_seq_sum()
  ipv6: set all.accept_dad to 0 by default
  uapi: fix linux/tls.h userspace compilation error
  usbnet: ipheth: prevent TX queue timeouts when device not ready
  vhost_net: conditionally enable tx polling
  uapi: fix linux/rxrpc.h userspace compilation errors
  net: stmmac: fix LPI transitioning for dwmac4
  atm: horizon: Fix irq release error
  net-sysfs: trigger netlink notification on ifalias change via sysfs
  openvswitch: Using kfree_rcu() to simplify the code
  openvswitch: Make local function ovs_nsh_key_attr_size() static
  openvswitch: Fix return value check in ovs_meter_cmd_features()
  ...

1 files changed, 234 insertions, 6 deletions

diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index a3c949ea7d1a..dcb8c39382e7 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -399,6 +399,12 @@ enum rtl_registers {
 	RxMaxSize	= 0xda,
 	CPlusCmd	= 0xe0,
 	IntrMitigate	= 0xe2,
+
+#define RTL_COALESCE_MASK	0x0f
+#define RTL_COALESCE_SHIFT	4
+#define RTL_COALESCE_T_MAX	(RTL_COALESCE_MASK)
+#define RTL_COALESCE_FRAME_MAX	(RTL_COALESCE_MASK << 2)
+
 	RxDescAddrLow	= 0xe4,
 	RxDescAddrHigh	= 0xe8,
 	EarlyTxThres	= 0xec,	/* 8169. Unit of 32 bytes. */
@@ -795,6 +801,7 @@ struct rtl8169_private {
 	u16 cp_cmd;
 
 	u16 event_slow;
+	const struct rtl_coalesce_info *coalesce_info;
 
 	struct mdio_ops {
 		void (*write)(struct rtl8169_private *, int, int);
@@ -1975,8 +1982,6 @@ static int rtl8169_set_speed_xmii(struct net_device *dev,
 		rtl_writephy(tp, MII_ADVERTISE, auto_nego);
 		rtl_writephy(tp, MII_CTRL1000, giga_ctrl);
 	} else {
-		giga_ctrl = 0;
-
 		if (speed == SPEED_10)
 			bmcr = 0;
 		else if (speed == SPEED_100)
@@ -2363,10 +2368,229 @@ static int rtl8169_nway_reset(struct net_device *dev)
 	return mii_nway_restart(&tp->mii);
 }
 
+/*
+ * Interrupt coalescing
+ *
+ * > 1 - the availability of the IntrMitigate (0xe2) register through the
+ * >     8169, 8168 and 810x line of chipsets
+ *
+ * 8169, 8168, and 8136(810x) serial chipsets support it.
+ *
+ * > 2 - the Tx timer unit at gigabit speed
+ *
+ * The unit of the timer depends on both the speed and the setting of CPlusCmd
+ * (0xe0) bit 1 and bit 0.
+ *
+ * For 8169
+ * bit[1:0] \ speed        1000M           100M            10M
+ * 0 0                     320ns           2.56us          40.96us
+ * 0 1                     2.56us          20.48us         327.7us
+ * 1 0                     5.12us          40.96us         655.4us
+ * 1 1                     10.24us         81.92us         1.31ms
+ *
+ * For the other
+ * bit[1:0] \ speed        1000M           100M            10M
+ * 0 0                     5us             2.56us          40.96us
+ * 0 1                     40us            20.48us         327.7us
+ * 1 0                     80us            40.96us         655.4us
+ * 1 1                     160us           81.92us         1.31ms
+ */
+
+/* rx/tx scale factors for one particular CPlusCmd[0:1] value */
+struct rtl_coalesce_scale {
+	/* Rx / Tx */
+	u32 nsecs[2];
+};
+
+/* rx/tx scale factors for all CPlusCmd[0:1] cases */
+struct rtl_coalesce_info {
+	u32 speed;
+	struct rtl_coalesce_scale scalev[4];	/* each CPlusCmd[0:1] case */
+};
+
+/* produce (r,t) pairs with each being in series of *1, *8, *8*2, *8*2*2 */
+#define rxtx_x1822(r, t) {		\
+	{{(r),		(t)}},		\
+	{{(r)*8,	(t)*8}},	\
+	{{(r)*8*2,	(t)*8*2}},	\
+	{{(r)*8*2*2,	(t)*8*2*2}},	\
+}
+static const struct rtl_coalesce_info rtl_coalesce_info_8169[] = {
+	/* speed	delays:     rx00   tx00	*/
+	{ SPEED_10,	rxtx_x1822(40960, 40960)	},
+	{ SPEED_100,	rxtx_x1822( 2560,  2560)	},
+	{ SPEED_1000,	rxtx_x1822(  320,   320)	},
+	{ 0 },
+};
+
+static const struct rtl_coalesce_info rtl_coalesce_info_8168_8136[] = {
+	/* speed	delays:     rx00   tx00	*/
+	{ SPEED_10,	rxtx_x1822(40960, 40960)	},
+	{ SPEED_100,	rxtx_x1822( 2560,  2560)	},
+	{ SPEED_1000,	rxtx_x1822( 5000,  5000)	},
+	{ 0 },
+};
+#undef rxtx_x1822
+
+/* get rx/tx scale vector corresponding to current speed */
+static const struct rtl_coalesce_info *rtl_coalesce_info(struct net_device *dev)
+{
+	struct rtl8169_private *tp = netdev_priv(dev);
+	struct ethtool_link_ksettings ecmd;
+	const struct rtl_coalesce_info *ci;
+	int rc;
+
+	rc = rtl8169_get_link_ksettings(dev, &ecmd);
+	if (rc < 0)
+		return ERR_PTR(rc);
+
+	for (ci = tp->coalesce_info; ci->speed != 0; ci++) {
+		if (ecmd.base.speed == ci->speed) {
+			return ci;
+		}
+	}
+
+	return ERR_PTR(-ELNRNG);
+}
+
+static int rtl_get_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
+{
+	struct rtl8169_private *tp = netdev_priv(dev);
+	void __iomem *ioaddr = tp->mmio_addr;
+	const struct rtl_coalesce_info *ci;
+	const struct rtl_coalesce_scale *scale;
+	struct {
+		u32 *max_frames;
+		u32 *usecs;
+	} coal_settings [] = {
+		{ &ec->rx_max_coalesced_frames, &ec->rx_coalesce_usecs },
+		{ &ec->tx_max_coalesced_frames, &ec->tx_coalesce_usecs }
+	}, *p = coal_settings;
+	int i;
+	u16 w;
+
+	memset(ec, 0, sizeof(*ec));
+
+	/* get rx/tx scale corresponding to current speed and CPlusCmd[0:1] */
+	ci = rtl_coalesce_info(dev);
+	if (IS_ERR(ci))
+		return PTR_ERR(ci);
+
+	scale = &ci->scalev[RTL_R16(CPlusCmd) & 3];
+
+	/* read IntrMitigate and adjust according to scale */
+	for (w = RTL_R16(IntrMitigate); w; w >>= RTL_COALESCE_SHIFT, p++) {
+		*p->max_frames = (w & RTL_COALESCE_MASK) << 2;
+		w >>= RTL_COALESCE_SHIFT;
+		*p->usecs = w & RTL_COALESCE_MASK;
+	}
+
+	for (i = 0; i < 2; i++) {
+		p = coal_settings + i;
+		*p->usecs = (*p->usecs * scale->nsecs[i]) / 1000;
+
+		/*
+		 * ethtool_coalesce says it is illegal to set both usecs and
+		 * max_frames to 0.
+		 */
+		if (!*p->usecs && !*p->max_frames)
+			*p->max_frames = 1;
+	}
+
+	return 0;
+}
+
+/* choose appropriate scale factor and CPlusCmd[0:1] for (speed, nsec) */
+static const struct rtl_coalesce_scale *rtl_coalesce_choose_scale(
+			struct net_device *dev, u32 nsec, u16 *cp01)
+{
+	const struct rtl_coalesce_info *ci;
+	u16 i;
+
+	ci = rtl_coalesce_info(dev);
+	if (IS_ERR(ci))
+		return ERR_CAST(ci);
+
+	for (i = 0; i < 4; i++) {
+		u32 rxtx_maxscale = max(ci->scalev[i].nsecs[0],
+					ci->scalev[i].nsecs[1]);
+		if (nsec <= rxtx_maxscale * RTL_COALESCE_T_MAX) {
+			*cp01 = i;
+			return &ci->scalev[i];
+		}
+	}
+
+	return ERR_PTR(-EINVAL);
+}
+
+static int rtl_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
+{
+	struct rtl8169_private *tp = netdev_priv(dev);
+	void __iomem *ioaddr = tp->mmio_addr;
+	const struct rtl_coalesce_scale *scale;
+	struct {
+		u32 frames;
+		u32 usecs;
+	} coal_settings [] = {
+		{ ec->rx_max_coalesced_frames, ec->rx_coalesce_usecs },
+		{ ec->tx_max_coalesced_frames, ec->tx_coalesce_usecs }
+	}, *p = coal_settings;
+	u16 w = 0, cp01;
+	int i;
+
+	scale = rtl_coalesce_choose_scale(dev,
+			max(p[0].usecs, p[1].usecs) * 1000, &cp01);
+	if (IS_ERR(scale))
+		return PTR_ERR(scale);
+
+	for (i = 0; i < 2; i++, p++) {
+		u32 units;
+
+		/*
+		 * accept max_frames=1 we returned in rtl_get_coalesce.
+		 * accept it not only when usecs=0 because of e.g. the following scenario:
+		 *
+		 * - both rx_usecs=0 & rx_frames=0 in hardware (no delay on RX)
+		 * - rtl_get_coalesce returns rx_usecs=0, rx_frames=1
+		 * - then user does `ethtool -C eth0 rx-usecs 100`
+		 *
+		 * since ethtool sends to kernel whole ethtool_coalesce
+		 * settings, if we do not handle rx_usecs=!0, rx_frames=1
+		 * we'll reject it below in `frames % 4 != 0`.
+		 */
+		if (p->frames == 1) {
+			p->frames = 0;
+		}
+
+		units = p->usecs * 1000 / scale->nsecs[i];
+		if (p->frames > RTL_COALESCE_FRAME_MAX || p->frames % 4)
+			return -EINVAL;
+
+		w <<= RTL_COALESCE_SHIFT;
+		w |= units;
+		w <<= RTL_COALESCE_SHIFT;
+		w |= p->frames >> 2;
+	}
+
+	rtl_lock_work(tp);
+
+	RTL_W16(IntrMitigate, swab16(w));
+
+	tp->cp_cmd = (tp->cp_cmd & ~3) | cp01;
+	RTL_W16(CPlusCmd, tp->cp_cmd);
+	RTL_R16(CPlusCmd);
+
+	rtl_unlock_work(tp);
+
+	return 0;
+}
+
 static const struct ethtool_ops rtl8169_ethtool_ops = {
 	.get_drvinfo		= rtl8169_get_drvinfo,
 	.get_regs_len		= rtl8169_get_regs_len,
 	.get_link		= ethtool_op_get_link,
+	.get_coalesce		= rtl_get_coalesce,
+	.set_coalesce		= rtl_set_coalesce,
 	.set_settings		= rtl8169_set_settings,
 	.get_msglevel		= rtl8169_get_msglevel,
 	.set_msglevel		= rtl8169_set_msglevel,
@@ -4401,10 +4625,9 @@ static void rtl_schedule_task(struct rtl8169_private *tp, enum rtl_flag flag)
 		schedule_work(&tp->wk.work);
 }
 
-static void rtl8169_phy_timer(unsigned long __opaque)
+static void rtl8169_phy_timer(struct timer_list *t)
 {
-	struct net_device *dev = (struct net_device *)__opaque;
-	struct rtl8169_private *tp = netdev_priv(dev);
+	struct rtl8169_private *tp = from_timer(tp, t, timer);
 
 	rtl_schedule_task(tp, RTL_FLAG_TASK_PHY_PENDING);
 }
@@ -8062,6 +8285,7 @@ static const struct rtl_cfg_info {
 	unsigned int align;
 	u16 event_slow;
 	unsigned features;
+	const struct rtl_coalesce_info *coalesce_info;
 	u8 default_ver;
 } rtl_cfg_infos [] = {
 	[RTL_CFG_0] = {
@@ -8070,6 +8294,7 @@ static const struct rtl_cfg_info {
 		.align		= 0,
 		.event_slow	= SYSErr | LinkChg | RxOverflow | RxFIFOOver,
 		.features	= RTL_FEATURE_GMII,
+		.coalesce_info	= rtl_coalesce_info_8169,
 		.default_ver	= RTL_GIGA_MAC_VER_01,
 	},
 	[RTL_CFG_1] = {
@@ -8078,6 +8303,7 @@ static const struct rtl_cfg_info {
 		.align		= 8,
 		.event_slow	= SYSErr | LinkChg | RxOverflow,
 		.features	= RTL_FEATURE_GMII | RTL_FEATURE_MSI,
+		.coalesce_info	= rtl_coalesce_info_8168_8136,
 		.default_ver	= RTL_GIGA_MAC_VER_11,
 	},
 	[RTL_CFG_2] = {
@@ -8087,6 +8313,7 @@ static const struct rtl_cfg_info {
 		.event_slow	= SYSErr | LinkChg | RxOverflow | RxFIFOOver |
 				  PCSTimeout,
 		.features	= RTL_FEATURE_MSI,
+		.coalesce_info	= rtl_coalesce_info_8168_8136,
 		.default_ver	= RTL_GIGA_MAC_VER_13,
 	}
 };
@@ -8450,11 +8677,12 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	tp->hw_start = cfg->hw_start;
 	tp->event_slow = cfg->event_slow;
+	tp->coalesce_info = cfg->coalesce_info;
 
 	tp->opts1_mask = (tp->mac_version != RTL_GIGA_MAC_VER_01) ?
 		~(RxBOVF | RxFOVF) : ~0;
 
-	setup_timer(&tp->timer, rtl8169_phy_timer, (unsigned long)dev);
+	timer_setup(&tp->timer, rtl8169_phy_timer, 0);
 
 	tp->rtl_fw = RTL_FIRMWARE_UNKNOWN;