cpts: introduce time stamping code and a PTP hardware clock.

This patch adds a driver for the CPTS that offers time
stamping and a PTP hardware clock. Because some of the
CPTS hardware variants (like the am335x) do not support
frequency adjustment, we have implemented this in software
by changing the multiplication factor of the timecounter.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

4 files changed, 582 insertions, 1 deletions

diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig
index b26cbda5efa..cbc3905a0a1 100644
--- a/drivers/net/ethernet/ti/Kconfig
+++ b/drivers/net/ethernet/ti/Kconfig
@@ -60,6 +60,14 @@ config TI_CPSW
 	  To compile this driver as a module, choose M here: the module
 	  will be called cpsw.
 
+config TI_CPTS
+	boolean "TI Common Platform Time Sync (CPTS) Support"
+	depends on TI_CPSW && PTP_1588_CLOCK && !(TI_CPSW=y && PTP_1588_CLOCK=m)
+	---help---
+	  This driver supports the Common Platform Time Sync unit of
+	  the CPSW Ethernet Switch. The unit can time stamp PTP UDP/IPv4
+	  and Layer 2 packets, and the driver offers a PTP Hardware Clock.
+
 config TLAN
 	tristate "TI ThunderLAN support"
 	depends on (PCI || EISA)
diff --git a/drivers/net/ethernet/ti/Makefile b/drivers/net/ethernet/ti/Makefile
index 91bd8bba78f..c65148e8aa1 100644
--- a/drivers/net/ethernet/ti/Makefile
+++ b/drivers/net/ethernet/ti/Makefile
@@ -8,4 +8,4 @@ obj-$(CONFIG_TI_DAVINCI_EMAC) += davinci_emac.o
 obj-$(CONFIG_TI_DAVINCI_MDIO) += davinci_mdio.o
 obj-$(CONFIG_TI_DAVINCI_CPDMA) += davinci_cpdma.o
 obj-$(CONFIG_TI_CPSW) += ti_cpsw.o
-ti_cpsw-y := cpsw_ale.o cpsw.o
+ti_cpsw-y := cpsw_ale.o cpsw.o cpts.o
diff --git a/drivers/net/ethernet/ti/cpts.c b/drivers/net/ethernet/ti/cpts.c
new file mode 100644
index 00000000000..337766738ec
--- /dev/null
+++ b/drivers/net/ethernet/ti/cpts.c
@@ -0,0 +1,427 @@
+/*
+ * TI Common Platform Time Sync
+ *
+ * Copyright (C) 2012 Richard Cochran <richardcochran@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include <linux/err.h>
+#include <linux/if.h>
+#include <linux/hrtimer.h>
+#include <linux/module.h>
+#include <linux/net_tstamp.h>
+#include <linux/ptp_classify.h>
+#include <linux/time.h>
+#include <linux/uaccess.h>
+#include <linux/workqueue.h>
+
+#include <plat/clock.h>
+
+#include "cpts.h"
+
+#ifdef CONFIG_TI_CPTS
+
+static struct sock_filter ptp_filter[] = {
+	PTP_FILTER
+};
+
+#define cpts_read32(c, r)	__raw_readl(&c->reg->r)
+#define cpts_write32(c, v, r)	__raw_writel(v, &c->reg->r)
+
+static int event_expired(struct cpts_event *event)
+{
+	return time_after(jiffies, event->tmo);
+}
+
+static int event_type(struct cpts_event *event)
+{
+	return (event->high >> EVENT_TYPE_SHIFT) & EVENT_TYPE_MASK;
+}
+
+static int cpts_fifo_pop(struct cpts *cpts, u32 *high, u32 *low)
+{
+	u32 r = cpts_read32(cpts, intstat_raw);
+
+	if (r & TS_PEND_RAW) {
+		*high = cpts_read32(cpts, event_high);
+		*low  = cpts_read32(cpts, event_low);
+		cpts_write32(cpts, EVENT_POP, event_pop);
+		return 0;
+	}
+	return -1;
+}
+
+/*
+ * Returns zero if matching event type was found.
+ */
+static int cpts_fifo_read(struct cpts *cpts, int match)
+{
+	int i, type = -1;
+	u32 hi, lo;
+	struct cpts_event *event;
+
+	for (i = 0; i < CPTS_FIFO_DEPTH; i++) {
+		if (cpts_fifo_pop(cpts, &hi, &lo))
+			break;
+		if (list_empty(&cpts->pool)) {
+			pr_err("cpts: event pool is empty\n");
+			return -1;
+		}
+		event = list_first_entry(&cpts->pool, struct cpts_event, list);
+		event->tmo = jiffies + 2;
+		event->high = hi;
+		event->low = lo;
+		type = event_type(event);
+		switch (type) {
+		case CPTS_EV_PUSH:
+		case CPTS_EV_RX:
+		case CPTS_EV_TX:
+			list_del_init(&event->list);
+			list_add_tail(&event->list, &cpts->events);
+			break;
+		case CPTS_EV_ROLL:
+		case CPTS_EV_HALF:
+		case CPTS_EV_HW:
+			break;
+		default:
+			pr_err("cpts: unkown event type\n");
+			break;
+		}
+		if (type == match)
+			break;
+	}
+	return type == match ? 0 : -1;
+}
+
+static cycle_t cpts_systim_read(const struct cyclecounter *cc)
+{
+	u64 val = 0;
+	struct cpts_event *event;
+	struct list_head *this, *next;
+	struct cpts *cpts = container_of(cc, struct cpts, cc);
+
+	cpts_write32(cpts, TS_PUSH, ts_push);
+	if (cpts_fifo_read(cpts, CPTS_EV_PUSH))
+		pr_err("cpts: unable to obtain a time stamp\n");
+
+	list_for_each_safe(this, next, &cpts->events) {
+		event = list_entry(this, struct cpts_event, list);
+		if (event_type(event) == CPTS_EV_PUSH) {
+			list_del_init(&event->list);
+			list_add(&event->list, &cpts->pool);
+			val = event->low;
+			break;
+		}
+	}
+
+	return val;
+}
+
+/* PTP clock operations */
+
+static int cpts_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
+{
+	u64 adj;
+	u32 diff, mult;
+	int neg_adj = 0;
+	unsigned long flags;
+	struct cpts *cpts = container_of(ptp, struct cpts, info);
+
+	if (ppb < 0) {
+		neg_adj = 1;
+		ppb = -ppb;
+	}
+	mult = cpts->cc_mult;
+	adj = mult;
+	adj *= ppb;
+	diff = div_u64(adj, 1000000000ULL);
+
+	spin_lock_irqsave(&cpts->lock, flags);
+
+	timecounter_read(&cpts->tc);
+
+	cpts->cc.mult = neg_adj ? mult - diff : mult + diff;
+
+	spin_unlock_irqrestore(&cpts->lock, flags);
+
+	return 0;
+}
+
+static int cpts_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	s64 now;
+	unsigned long flags;
+	struct cpts *cpts = container_of(ptp, struct cpts, info);
+
+	spin_lock_irqsave(&cpts->lock, flags);
+	now = timecounter_read(&cpts->tc);
+	now += delta;
+	timecounter_init(&cpts->tc, &cpts->cc, now);
+	spin_unlock_irqrestore(&cpts->lock, flags);
+
+	return 0;
+}
+
+static int cpts_ptp_gettime(struct ptp_clock_info *ptp, struct timespec *ts)
+{
+	u64 ns;
+	u32 remainder;
+	unsigned long flags;
+	struct cpts *cpts = container_of(ptp, struct cpts, info);
+
+	spin_lock_irqsave(&cpts->lock, flags);
+	ns = timecounter_read(&cpts->tc);
+	spin_unlock_irqrestore(&cpts->lock, flags);
+
+	ts->tv_sec = div_u64_rem(ns, 1000000000, &remainder);
+	ts->tv_nsec = remainder;
+
+	return 0;
+}
+
+static int cpts_ptp_settime(struct ptp_clock_info *ptp,
+			    const struct timespec *ts)
+{
+	u64 ns;
+	unsigned long flags;
+	struct cpts *cpts = container_of(ptp, struct cpts, info);
+
+	ns = ts->tv_sec * 1000000000ULL;
+	ns += ts->tv_nsec;
+
+	spin_lock_irqsave(&cpts->lock, flags);
+	timecounter_init(&cpts->tc, &cpts->cc, ns);
+	spin_unlock_irqrestore(&cpts->lock, flags);
+
+	return 0;
+}
+
+static int cpts_ptp_enable(struct ptp_clock_info *ptp,
+			   struct ptp_clock_request *rq, int on)
+{
+	return -EOPNOTSUPP;
+}
+
+static struct ptp_clock_info cpts_info = {
+	.owner		= THIS_MODULE,
+	.name		= "CTPS timer",
+	.max_adj	= 1000000,
+	.n_ext_ts	= 0,
+	.pps		= 0,
+	.adjfreq	= cpts_ptp_adjfreq,
+	.adjtime	= cpts_ptp_adjtime,
+	.gettime	= cpts_ptp_gettime,
+	.settime	= cpts_ptp_settime,
+	.enable		= cpts_ptp_enable,
+};
+
+static void cpts_overflow_check(struct work_struct *work)
+{
+	struct timespec ts;
+	struct cpts *cpts = container_of(work, struct cpts, overflow_work.work);
+
+	cpts_write32(cpts, CPTS_EN, control);
+	cpts_write32(cpts, TS_PEND_EN, int_enable);
+	cpts_ptp_gettime(&cpts->info, &ts);
+	pr_debug("cpts overflow check at %ld.%09lu\n", ts.tv_sec, ts.tv_nsec);
+	schedule_delayed_work(&cpts->overflow_work, CPTS_OVERFLOW_PERIOD);
+}
+
+#define CPTS_REF_CLOCK_NAME "cpsw_cpts_rft_clk"
+
+static void cpts_clk_init(struct cpts *cpts)
+{
+	cpts->refclk = clk_get(NULL, CPTS_REF_CLOCK_NAME);
+	if (IS_ERR(cpts->refclk)) {
+		pr_err("Failed to clk_get %s\n", CPTS_REF_CLOCK_NAME);
+		cpts->refclk = NULL;
+		return;
+	}
+	clk_enable(cpts->refclk);
+	cpts->freq = cpts->refclk->recalc(cpts->refclk);
+}
+
+static void cpts_clk_release(struct cpts *cpts)
+{
+	clk_disable(cpts->refclk);
+	clk_put(cpts->refclk);
+}
+
+static int cpts_match(struct sk_buff *skb, unsigned int ptp_class,
+		      u16 ts_seqid, u8 ts_msgtype)
+{
+	u16 *seqid;
+	unsigned int offset;
+	u8 *msgtype, *data = skb->data;
+
+	switch (ptp_class) {
+	case PTP_CLASS_V1_IPV4:
+	case PTP_CLASS_V2_IPV4:
+		offset = ETH_HLEN + IPV4_HLEN(data) + UDP_HLEN;
+		break;
+	case PTP_CLASS_V1_IPV6:
+	case PTP_CLASS_V2_IPV6:
+		offset = OFF_PTP6;
+		break;
+	case PTP_CLASS_V2_L2:
+		offset = ETH_HLEN;
+		break;
+	case PTP_CLASS_V2_VLAN:
+		offset = ETH_HLEN + VLAN_HLEN;
+		break;
+	default:
+		return 0;
+	}
+
+	if (skb->len + ETH_HLEN < offset + OFF_PTP_SEQUENCE_ID + sizeof(*seqid))
+		return 0;
+
+	if (unlikely(ptp_class & PTP_CLASS_V1))
+		msgtype = data + offset + OFF_PTP_CONTROL;
+	else
+		msgtype = data + offset;
+
+	seqid = (u16 *)(data + offset + OFF_PTP_SEQUENCE_ID);
+
+	return (ts_msgtype == (*msgtype & 0xf) && ts_seqid == ntohs(*seqid));
+}
+
+static u64 cpts_find_ts(struct cpts *cpts, struct sk_buff *skb, int ev_type)
+{
+	u64 ns = 0;
+	struct cpts_event *event;
+	struct list_head *this, *next;
+	unsigned int class = sk_run_filter(skb, ptp_filter);
+	unsigned long flags;
+	u16 seqid;
+	u8 mtype;
+
+	if (class == PTP_CLASS_NONE)
+		return 0;
+
+	spin_lock_irqsave(&cpts->lock, flags);
+	cpts_fifo_read(cpts, CPTS_EV_PUSH);
+	list_for_each_safe(this, next, &cpts->events) {
+		event = list_entry(this, struct cpts_event, list);
+		if (event_expired(event)) {
+			list_del_init(&event->list);
+			list_add(&event->list, &cpts->pool);
+			continue;
+		}
+		mtype = (event->high >> MESSAGE_TYPE_SHIFT) & MESSAGE_TYPE_MASK;
+		seqid = (event->high >> SEQUENCE_ID_SHIFT) & SEQUENCE_ID_MASK;
+		if (ev_type == event_type(event) &&
+		    cpts_match(skb, class, seqid, mtype)) {
+			ns = timecounter_cyc2time(&cpts->tc, event->low);
+			list_del_init(&event->list);
+			list_add(&event->list, &cpts->pool);
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&cpts->lock, flags);
+
+	return ns;
+}
+
+void cpts_rx_timestamp(struct cpts *cpts, struct sk_buff *skb)
+{
+	u64 ns;
+	struct skb_shared_hwtstamps *ssh;
+
+	if (!cpts->rx_enable)
+		return;
+	ns = cpts_find_ts(cpts, skb, CPTS_EV_RX);
+	if (!ns)
+		return;
+	ssh = skb_hwtstamps(skb);
+	memset(ssh, 0, sizeof(*ssh));
+	ssh->hwtstamp = ns_to_ktime(ns);
+}
+
+void cpts_tx_timestamp(struct cpts *cpts, struct sk_buff *skb)
+{
+	u64 ns;
+	struct skb_shared_hwtstamps ssh;
+
+	if (!(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
+		return;
+	ns = cpts_find_ts(cpts, skb, CPTS_EV_TX);
+	if (!ns)
+		return;
+	memset(&ssh, 0, sizeof(ssh));
+	ssh.hwtstamp = ns_to_ktime(ns);
+	skb_tstamp_tx(skb, &ssh);
+}
+
+#endif /*CONFIG_TI_CPTS*/
+
+int cpts_register(struct device *dev, struct cpts *cpts,
+		  u32 mult, u32 shift)
+{
+#ifdef CONFIG_TI_CPTS
+	int err, i;
+	unsigned long flags;
+
+	if (ptp_filter_init(ptp_filter, ARRAY_SIZE(ptp_filter))) {
+		pr_err("cpts: bad ptp filter\n");
+		return -EINVAL;
+	}
+	cpts->info = cpts_info;
+	cpts->clock = ptp_clock_register(&cpts->info, dev);
+	if (IS_ERR(cpts->clock)) {
+		err = PTR_ERR(cpts->clock);
+		cpts->clock = NULL;
+		return err;
+	}
+	spin_lock_init(&cpts->lock);
+
+	cpts->cc.read = cpts_systim_read;
+	cpts->cc.mask = CLOCKSOURCE_MASK(32);
+	cpts->cc_mult = mult;
+	cpts->cc.mult = mult;
+	cpts->cc.shift = shift;
+
+	INIT_LIST_HEAD(&cpts->events);
+	INIT_LIST_HEAD(&cpts->pool);
+	for (i = 0; i < CPTS_MAX_EVENTS; i++)
+		list_add(&cpts->pool_data[i].list, &cpts->pool);
+
+	cpts_clk_init(cpts);
+	cpts_write32(cpts, CPTS_EN, control);
+	cpts_write32(cpts, TS_PEND_EN, int_enable);
+
+	spin_lock_irqsave(&cpts->lock, flags);
+	timecounter_init(&cpts->tc, &cpts->cc, ktime_to_ns(ktime_get_real()));
+	spin_unlock_irqrestore(&cpts->lock, flags);
+
+	INIT_DELAYED_WORK(&cpts->overflow_work, cpts_overflow_check);
+	schedule_delayed_work(&cpts->overflow_work, CPTS_OVERFLOW_PERIOD);
+
+	cpts->phc_index = ptp_clock_index(cpts->clock);
+#endif
+	return 0;
+}
+
+void cpts_unregister(struct cpts *cpts)
+{
+#ifdef CONFIG_TI_CPTS
+	if (cpts->clock) {
+		ptp_clock_unregister(cpts->clock);
+		cancel_delayed_work_sync(&cpts->overflow_work);
+	}
+	if (cpts->refclk)
+		cpts_clk_release(cpts);
+#endif
+}
diff --git a/drivers/net/ethernet/ti/cpts.h b/drivers/net/ethernet/ti/cpts.h
new file mode 100644
index 00000000000..e1bba3a496b
--- /dev/null
+++ b/drivers/net/ethernet/ti/cpts.h
@@ -0,0 +1,146 @@
+/*
+ * TI Common Platform Time Sync
+ *
+ * Copyright (C) 2012 Richard Cochran <richardcochran@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#ifndef _TI_CPTS_H_
+#define _TI_CPTS_H_
+
+#include <linux/clk.h>
+#include <linux/clkdev.h>
+#include <linux/clocksource.h>
+#include <linux/device.h>
+#include <linux/list.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/skbuff.h>
+
+struct cpsw_cpts {
+	u32 idver;                /* Identification and version */
+	u32 control;              /* Time sync control */
+	u32 res1;
+	u32 ts_push;              /* Time stamp event push */
+	u32 ts_load_val;          /* Time stamp load value */
+	u32 ts_load_en;           /* Time stamp load enable */
+	u32 res2[2];
+	u32 intstat_raw;          /* Time sync interrupt status raw */
+	u32 intstat_masked;       /* Time sync interrupt status masked */
+	u32 int_enable;           /* Time sync interrupt enable */
+	u32 res3;
+	u32 event_pop;            /* Event interrupt pop */
+	u32 event_low;            /* 32 Bit Event Time Stamp */
+	u32 event_high;           /* Event Type Fields */
+};
+
+/* Bit definitions for the IDVER register */
+#define TX_IDENT_SHIFT       (16)    /* TX Identification Value */
+#define TX_IDENT_MASK        (0xffff)
+#define RTL_VER_SHIFT        (11)    /* RTL Version Value */
+#define RTL_VER_MASK         (0x1f)
+#define MAJOR_VER_SHIFT      (8)     /* Major Version Value */
+#define MAJOR_VER_MASK       (0x7)
+#define MINOR_VER_SHIFT      (0)     /* Minor Version Value */
+#define MINOR_VER_MASK       (0xff)
+
+/* Bit definitions for the CONTROL register */
+#define HW4_TS_PUSH_EN       (1<<11) /* Hardware push 4 enable */
+#define HW3_TS_PUSH_EN       (1<<10) /* Hardware push 3 enable */
+#define HW2_TS_PUSH_EN       (1<<9)  /* Hardware push 2 enable */
+#define HW1_TS_PUSH_EN       (1<<8)  /* Hardware push 1 enable */
+#define INT_TEST             (1<<1)  /* Interrupt Test */
+#define CPTS_EN              (1<<0)  /* Time Sync Enable */
+
+/*
+ * Definitions for the single bit resisters:
+ * TS_PUSH TS_LOAD_EN  INTSTAT_RAW INTSTAT_MASKED INT_ENABLE EVENT_POP
+ */
+#define TS_PUSH             (1<<0)  /* Time stamp event push */
+#define TS_LOAD_EN          (1<<0)  /* Time Stamp Load */
+#define TS_PEND_RAW         (1<<0)  /* int read (before enable) */
+#define TS_PEND             (1<<0)  /* masked interrupt read (after enable) */
+#define TS_PEND_EN          (1<<0)  /* masked interrupt enable */
+#define EVENT_POP           (1<<0)  /* writing discards one event */
+
+/* Bit definitions for the EVENT_HIGH register */
+#define PORT_NUMBER_SHIFT    (24)    /* Indicates Ethernet port or HW pin */
+#define PORT_NUMBER_MASK     (0x1f)
+#define EVENT_TYPE_SHIFT     (20)    /* Time sync event type */
+#define EVENT_TYPE_MASK      (0xf)
+#define MESSAGE_TYPE_SHIFT   (16)    /* PTP message type */
+#define MESSAGE_TYPE_MASK    (0xf)
+#define SEQUENCE_ID_SHIFT    (0)     /* PTP message sequence ID */
+#define SEQUENCE_ID_MASK     (0xffff)
+
+enum {
+	CPTS_EV_PUSH, /* Time Stamp Push Event */
+	CPTS_EV_ROLL, /* Time Stamp Rollover Event */
+	CPTS_EV_HALF, /* Time Stamp Half Rollover Event */
+	CPTS_EV_HW,   /* Hardware Time Stamp Push Event */
+	CPTS_EV_RX,   /* Ethernet Receive Event */
+	CPTS_EV_TX,   /* Ethernet Transmit Event */
+};
+
+/* This covers any input clock up to about 500 MHz. */
+#define CPTS_OVERFLOW_PERIOD (HZ * 8)
+
+#define CPTS_FIFO_DEPTH 16
+#define CPTS_MAX_EVENTS 32
+
+struct cpts_event {
+	struct list_head list;
+	unsigned long tmo;
+	u32 high;
+	u32 low;
+};
+
+struct cpts {
+	struct cpsw_cpts __iomem *reg;
+	int tx_enable;
+	int rx_enable;
+#ifdef CONFIG_TI_CPTS
+	struct ptp_clock_info info;
+	struct ptp_clock *clock;
+	spinlock_t lock; /* protects time registers */
+	u32 cc_mult; /* for the nominal frequency */
+	struct cyclecounter cc;
+	struct timecounter tc;
+	struct delayed_work overflow_work;
+	int phc_index;
+	struct clk *refclk;
+	unsigned long freq;
+	struct list_head events;
+	struct list_head pool;
+	struct cpts_event pool_data[CPTS_MAX_EVENTS];
+#endif
+};
+
+#ifdef CONFIG_TI_CPTS
+extern void cpts_rx_timestamp(struct cpts *cpts, struct sk_buff *skb);
+extern void cpts_tx_timestamp(struct cpts *cpts, struct sk_buff *skb);
+#else
+static inline void cpts_rx_timestamp(struct cpts *cpts, struct sk_buff *skb)
+{
+}
+static inline void cpts_tx_timestamp(struct cpts *cpts, struct sk_buff *skb)
+{
+}
+#endif
+
+extern int cpts_register(struct device *dev, struct cpts *cpts,
+			 u32 mult, u32 shift);
+extern void cpts_unregister(struct cpts *cpts);
+
+#endif