LinuxPPS new functionalities

LinuxPPS new functionalities

[Rodolfo Giometti]

This patchset adds several PPS functionalities.

All patches have a specific comment but I prefer say something more:

* Patches 01-02 are just documentation about PPS support.

* Patch 03 adds PPS client support skeleton.

* Patches 04-06 add generic PPS serial support (with specific
  modifications requested by Alan Cox).

* Patches 07-08 add PPS serial support for specific hardware.

Ciao,

Rodolfo

[PATCH 1/8] pps: userland header file for PPS API.

[Rodolfo Giometti]

This patch adds into the PPS's documentation directory a possible
implementation of the PPS API (RFC 2783) by using the LinuxPPS's char
devices.

This file is not just an example but it can be used into real
systems. :)

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
---
 Documentation/pps/timepps.h |  198 +++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 198 insertions(+), 0 deletions(-)
 create mode 100644 Documentation/pps/timepps.h

diff --git a/Documentation/pps/timepps.h b/Documentation/pps/timepps.h
new file mode 100644
index 0000000..d2628d2
--- /dev/null
+++ b/Documentation/pps/timepps.h
@@ -0,0 +1,198 @@
+/*
+ * timepps.h -- PPS API main header
+ *
+ * Copyright (C) 2005-2007   Rodolfo Giometti <giometti@linux.it>
+ *
+ *   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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef _SYS_TIMEPPS_H_
+#define _SYS_TIMEPPS_H_
+
+#include <errno.h>
+#include <sys/time.h>
+#include <sys/ioctl.h>
+#include <linux/types.h>
+#include <linux/pps.h>
+
+#define LINUXPPS	1		/* signal we are using LinuxPPS */
+
+/*
+ * New data structures
+ */
+
+struct ntp_fp {
+	unsigned int integral;
+	unsigned int fractional;
+};
+
+union pps_timeu {
+	struct timespec tspec;
+	struct ntp_fp ntpfp;
+	unsigned long longpad[3];
+};
+
+struct pps_info {
+	unsigned long assert_sequence;	/* seq. num. of assert event */
+	unsigned long clear_sequence;	/* seq. num. of clear event */
+	union pps_timeu assert_tu;	/* time of assert event */
+	union pps_timeu clear_tu;	/* time of clear event */
+	int current_mode;		/* current mode bits */
+};
+
+struct pps_params {
+	int api_version;		/* API version # */
+	int mode;			/* mode bits */
+	union pps_timeu assert_off_tu;	/* offset compensation for assert */
+	union pps_timeu clear_off_tu;	/* offset compensation for clear */
+};
+
+typedef int pps_handle_t;		/* represents a PPS source */
+typedef unsigned long pps_seq_t;	/* sequence number */
+typedef struct ntp_fp ntp_fp_t;		/* NTP-compatible time stamp */
+typedef union pps_timeu pps_timeu_t;	/* generic data type for time stamps */
+typedef struct pps_info pps_info_t;
+typedef struct pps_params pps_params_t;
+
+#define assert_timestamp        assert_tu.tspec
+#define clear_timestamp         clear_tu.tspec
+
+#define assert_timestamp_ntpfp  assert_tu.ntpfp
+#define clear_timestamp_ntpfp   clear_tu.ntpfp
+
+#define assert_offset		assert_off_tu.tspec
+#define clear_offset		clear_off_tu.tspec
+
+#define assert_offset_ntpfp     assert_off_tu.ntpfp
+#define clear_offset_ntpfp      clear_off_tu.ntpfp
+
+/*
+ * The PPS API
+ */
+
+static __inline int time_pps_create(int source, pps_handle_t *handle)
+{
+	int ret;
+	struct pps_kparams dummy;
+
+	if (!handle) {
+		errno = EINVAL;
+		return -1;
+	}
+
+	/* First we check if current device is a valid PPS one by
+	 * doing a dummy PPS_GETPARAMS...
+	 */
+	ret = ioctl(source, PPS_GETPARAMS, &dummy);
+	if (ret) {
+		errno = EOPNOTSUPP;
+		return -1;
+	}
+
+	/* ... then since in LinuxPPS there are no differences between a
+	 * "PPS source" and a "PPS handle", we simply return the same value.
+	 */
+	*handle = source;
+
+	return 0;
+}
+
+static __inline int time_pps_destroy(pps_handle_t handle)
+{
+	return close(handle);
+}
+
+static __inline int time_pps_getparams(pps_handle_t handle,
+					pps_params_t *ppsparams)
+{
+	int ret;
+	struct pps_kparams __ppsparams;
+
+	ret = ioctl(handle, PPS_GETPARAMS, &__ppsparams);
+
+	ppsparams->api_version = __ppsparams.api_version;
+	ppsparams->mode = __ppsparams.mode;
+	ppsparams->assert_off_tu.tspec.tv_sec = __ppsparams.assert_off_tu.sec;
+	ppsparams->assert_off_tu.tspec.tv_nsec = __ppsparams.assert_off_tu.nsec;
+	ppsparams->clear_off_tu.tspec.tv_sec = __ppsparams.clear_off_tu.sec;
+	ppsparams->clear_off_tu.tspec.tv_nsec = __ppsparams.clear_off_tu.nsec;
+
+	return ret;
+}
+
+static __inline int time_pps_setparams(pps_handle_t handle,
+					const pps_params_t *ppsparams)
+{
+	struct pps_kparams __ppsparams;
+
+	__ppsparams.api_version = ppsparams->api_version;
+	__ppsparams.mode = ppsparams->mode;
+	__ppsparams.assert_off_tu.sec = ppsparams->assert_off_tu.tspec.tv_sec;
+	__ppsparams.assert_off_tu.nsec = ppsparams->assert_off_tu.tspec.tv_nsec;
+	__ppsparams.clear_off_tu.sec = ppsparams->clear_off_tu.tspec.tv_sec;
+	__ppsparams.clear_off_tu.nsec = ppsparams->clear_off_tu.tspec.tv_nsec;
+
+	return ioctl(handle, PPS_SETPARAMS, &__ppsparams);
+}
+
+/* Get capabilities for handle */
+static __inline int time_pps_getcap(pps_handle_t handle, int *mode)
+{
+	return ioctl(handle, PPS_GETCAP, mode);
+}
+
+static __inline int time_pps_fetch(pps_handle_t handle, const int tsformat,
+					pps_info_t *ppsinfobuf,
+					const struct timespec *timeout)
+{
+	struct pps_fdata __fdata;
+	int ret;
+
+	/* Sanity checks */
+	if (tsformat != PPS_TSFMT_TSPEC) {
+		errno = EINVAL;
+		return -1;
+	}
+
+	if (timeout) {
+		__fdata.timeout.sec = timeout->tv_sec;
+		__fdata.timeout.nsec = timeout->tv_nsec;
+		__fdata.timeout.flags = ~PPS_TIME_INVALID;
+	} else
+		__fdata.timeout.flags = PPS_TIME_INVALID;
+
+	ret = ioctl(handle, PPS_FETCH, &__fdata);
+
+	ppsinfobuf->assert_sequence = __fdata.info.assert_sequence;
+	ppsinfobuf->clear_sequence = __fdata.info.clear_sequence;
+	ppsinfobuf->assert_tu.tspec.tv_sec = __fdata.info.assert_tu.sec;
+	ppsinfobuf->assert_tu.tspec.tv_nsec = __fdata.info.assert_tu.nsec;
+	ppsinfobuf->clear_tu.tspec.tv_sec = __fdata.info.clear_tu.sec;
+	ppsinfobuf->clear_tu.tspec.tv_nsec = __fdata.info.clear_tu.nsec;
+	ppsinfobuf->current_mode = __fdata.info.current_mode;
+
+	return ret;
+}
+
+static __inline int time_pps_kcbind(pps_handle_t handle,
+					const int kernel_consumer,
+					const int edge, const int tsformat)
+{
+	/* LinuxPPS doesn't implement kernel consumer feature */
+	errno = EOPNOTSUPP;
+	return -1;
+}
+
+#endif				/* _SYS_TIMEPPS_H_ */
-- 
1.6.3.3

[PATCH 2/8] pps: documentation programs and examples.

[Rodolfo Giometti]

Here some utilities and examples about the PPS API and the LinuxPPS
support.

* ppstest.c implements an useful testing program, while

* ppsfind tries to help the user into finding a specific PPS source by
  using its name or path.

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
---
 Documentation/pps/Makefile  |   27 ++++++++
 Documentation/pps/ppsfind   |   17 +++++
 Documentation/pps/ppstest.c |  150 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 194 insertions(+), 0 deletions(-)
 create mode 100644 Documentation/pps/Makefile
 create mode 100644 Documentation/pps/ppsfind
 create mode 100644 Documentation/pps/ppstest.c

diff --git a/Documentation/pps/Makefile b/Documentation/pps/Makefile
new file mode 100644
index 0000000..3848aba
--- /dev/null
+++ b/Documentation/pps/Makefile
@@ -0,0 +1,27 @@
+TARGETS = ppstest
+
+CFLAGS += -Wall -O2 -D_GNU_SOURCE
+CFLAGS += -I . -I ../../include/
+CFLAGS += -ggdb
+
+# -- Actions section --
+
+.PHONY : all depend dep
+
+all : .depend $(TARGETS)
+
+.depend depend dep :
+	$(CC) $(CFLAGS) -M $(TARGETS:=.c) > .depend
+
+ifeq (.depend,$(wildcard .depend))
+include .depend
+endif
+
+
+# -- Clean section --
+
+.PHONY : clean
+
+clean :
+	rm -f *.o *~ core .depend
+	rm -f ${TARGETS}
diff --git a/Documentation/pps/ppsfind b/Documentation/pps/ppsfind
new file mode 100644
index 0000000..93c0e17
--- /dev/null
+++ b/Documentation/pps/ppsfind
@@ -0,0 +1,17 @@
+#!/bin/sh
+
+SYS="/sys/class/pps/"
+
+if [ $# -lt 1 ] ; then
+	echo "usage: ppsfind <name>" >&2
+	exit 1
+fi
+
+for d in $(ls $SYS) ; do
+	if grep $1 $SYS/$d/name >& /dev/null || \
+	   grep $1 $SYS/$d/path >& /dev/null ; then
+		echo "$d: name=$(cat $SYS/$d/name) path=$(cat $SYS/$d/path)"
+	fi
+done
+
+exit 0
diff --git a/Documentation/pps/ppstest.c b/Documentation/pps/ppstest.c
new file mode 100644
index 0000000..30c8f6d
--- /dev/null
+++ b/Documentation/pps/ppstest.c
@@ -0,0 +1,150 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+
+#include <timepps.h>
+
+int find_source(char *path, pps_handle_t *handle, int *avail_mode)
+{
+	pps_params_t params;
+	int ret;
+
+	printf("trying PPS source \"%s\"\n", path);
+
+	/* Try to find the source by using the supplied "path" name */
+	ret = open(path, O_RDWR);
+	if (ret < 0) {
+		fprintf(stderr, "unable to open device \"%s\" (%m)\n", path);
+		return ret;
+	}
+
+	/* Open the PPS source (and check the file descriptor) */
+	ret = time_pps_create(ret, handle);
+	if (ret < 0) {
+		fprintf(stderr, "cannot create a PPS source from device "
+				"\"%s\" (%m)\n", path);
+		return -1;
+	}
+	printf("found PPS source \"%s\"\n", path);
+
+	/* Find out what features are supported */
+	ret = time_pps_getcap(*handle, avail_mode);
+	if (ret < 0) {
+		fprintf(stderr, "cannot get capabilities (%m)\n");
+		return -1;
+	}
+	if ((*avail_mode & PPS_CAPTUREASSERT) == 0) {
+		fprintf(stderr, "cannot CAPTUREASSERT\n");
+		return -1;
+	}
+	if ((*avail_mode & PPS_OFFSETASSERT) == 0) {
+		fprintf(stderr, "cannot OFFSETASSERT\n");
+		return -1;
+	}
+
+	/* Capture assert timestamps, and compensate for a 675 nsec
+	 * propagation delay */
+	ret = time_pps_getparams(*handle, &params);
+	if (ret < 0) {
+		fprintf(stderr, "cannot get parameters (%m)\n");
+		return -1;
+	}
+	params.assert_offset.tv_sec = 0;
+	params.assert_offset.tv_nsec = 675;
+	params.mode |= PPS_CAPTUREASSERT | PPS_OFFSETASSERT;
+	params.mode &= ~(PPS_CAPTURECLEAR | PPS_OFFSETCLEAR);
+	ret = time_pps_setparams(*handle, &params);
+	if (ret < 0) {
+		fprintf(stderr, "cannot set parameters (%m)\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+int fetch_source(int i, pps_handle_t *handle, int *avail_mode)
+{
+	struct timespec timeout;
+	pps_info_t infobuf;
+	int ret;
+
+	/* create a zero-valued timeout */
+	timeout.tv_sec = 3;
+	timeout.tv_nsec = 0;
+
+retry:
+	if (*avail_mode & PPS_CANWAIT) /* waits for the next event */
+		ret = time_pps_fetch(*handle, PPS_TSFMT_TSPEC, &infobuf,
+				   &timeout);
+	else {
+		sleep(1);
+		ret = time_pps_fetch(*handle, PPS_TSFMT_TSPEC, &infobuf,
+				   &timeout);
+	}
+	if (ret < 0) {
+		if (ret == -EINTR) {
+			fprintf(stderr, "time_pps_fetch() got a signal!\n");
+			goto retry;
+		}
+
+		fprintf(stderr, "time_pps_fetch() error %d (%m)\n", ret);
+		return -1;
+	}
+
+	printf("source %d - "
+	       "assert %ld.%09ld, sequence: %ld - "
+	       "clear  %ld.%09ld, sequence: %ld\n",
+	       i,
+	       infobuf.assert_timestamp.tv_sec,
+	       infobuf.assert_timestamp.tv_nsec,
+	       infobuf.assert_sequence,
+	       infobuf.clear_timestamp.tv_sec,
+	       infobuf.clear_timestamp.tv_nsec, infobuf.clear_sequence);
+
+	return 0;
+}
+
+void usage(char *name)
+{
+	fprintf(stderr, "usage: %s <ppsdev> [<ppsdev> ...]\n", name);
+	exit(EXIT_FAILURE);
+}
+
+int main(int argc, char *argv[])
+{
+	int num;
+	pps_handle_t handle[4];
+	int avail_mode[4];
+	int i = 0;
+	int ret;
+
+	/* Check the command line */
+	if (argc < 2)
+		usage(argv[0]);
+
+	for (i = 1; i < argc && i <= 4; i++) {
+		ret = find_source(argv[i], &handle[i - 1], &avail_mode[i - 1]);
+		if (ret < 0)
+			exit(EXIT_FAILURE);
+	}
+
+	num = i - 1;
+	printf("ok, found %d source(s), now start fetching data...\n", num);
+
+	/* loop, printing the most recent timestamp every second or so */
+	while (1) {
+		for (i = 0; i < num; i++) {
+			ret = fetch_source(i, &handle[i], &avail_mode[i]);
+			if (ret < 0 && errno != ETIMEDOUT)
+				exit(EXIT_FAILURE);
+		}
+	}
+
+	for (; i >= 0; i--)
+		time_pps_destroy(handle[i]);
+
+	return 0;
+}
-- 
1.6.3.3

[PATCH 3/8] pps: LinuxPPS clients support.

[Rodolfo Giometti]

Each PPS source can be registered/deregistered into the system by
using special modules called "clients". They simply define the PPS
sources' attributes and implement the time signal registartion
mechanism.

This patch adds a special directory for such clients and adds a dummy
client that can be useful to test system integrity on real systems.

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
---
 drivers/pps/Kconfig          |    2 +
 drivers/pps/Makefile         |    1 +
 drivers/pps/clients/Kconfig  |   18 ++++++
 drivers/pps/clients/Makefile |    9 +++
 drivers/pps/clients/ktimer.c |  123 ++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 153 insertions(+), 0 deletions(-)
 create mode 100644 drivers/pps/clients/Kconfig
 create mode 100644 drivers/pps/clients/Makefile
 create mode 100644 drivers/pps/clients/ktimer.c

diff --git a/drivers/pps/Kconfig b/drivers/pps/Kconfig
index cc2eb8e..1afe4e0 100644
--- a/drivers/pps/Kconfig
+++ b/drivers/pps/Kconfig
@@ -30,4 +30,6 @@ config PPS_DEBUG
 	  messages to the system log.  Select this if you are having a
 	  problem with PPS support and want to see more of what is going on.
 
+source drivers/pps/clients/Kconfig
+
 endmenu
diff --git a/drivers/pps/Makefile b/drivers/pps/Makefile
index 19ea582..98960dd 100644
--- a/drivers/pps/Makefile
+++ b/drivers/pps/Makefile
@@ -4,5 +4,6 @@
 
 pps_core-y			:= pps.o kapi.o sysfs.o
 obj-$(CONFIG_PPS)		:= pps_core.o
+obj-y				+= clients/
 
 ccflags-$(CONFIG_PPS_DEBUG) := -DDEBUG
diff --git a/drivers/pps/clients/Kconfig b/drivers/pps/clients/Kconfig
new file mode 100644
index 0000000..60b83be
--- /dev/null
+++ b/drivers/pps/clients/Kconfig
@@ -0,0 +1,18 @@
+#
+# PPS clients configuration
+#
+
+if PPS
+
+comment "PPS clients support"
+
+config PPS_CLIENT_KTIMER
+	tristate "Kernel timer client (Testing client, use for debug)"
+	help
+	  If you say yes here you get support for a PPS debugging client
+	  which uses a kernel timer to generate the PPS signal.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called ktimer.ko.
+
+endif
diff --git a/drivers/pps/clients/Makefile b/drivers/pps/clients/Makefile
new file mode 100644
index 0000000..f3c1e39
--- /dev/null
+++ b/drivers/pps/clients/Makefile
@@ -0,0 +1,9 @@
+#
+# Makefile for PPS clients.
+#
+
+obj-$(CONFIG_PPS_CLIENT_KTIMER)	+= ktimer.o
+
+ifeq ($(CONFIG_PPS_DEBUG),y)
+EXTRA_CFLAGS += -DDEBUG
+endif
diff --git a/drivers/pps/clients/ktimer.c b/drivers/pps/clients/ktimer.c
new file mode 100644
index 0000000..6de5dfc
--- /dev/null
+++ b/drivers/pps/clients/ktimer.c
@@ -0,0 +1,123 @@
+/*
+ * ktimer.c -- kernel timer test client
+ *
+ *
+ * Copyright (C) 2005-2006   Rodolfo Giometti <giometti@linux.it>
+ *
+ *   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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/time.h>
+#include <linux/timer.h>
+#include <linux/pps_kernel.h>
+
+/*
+ * Global variables
+ */
+
+static int source;
+static struct timer_list ktimer;
+
+/*
+ * The kernel timer
+ */
+
+static void pps_ktimer_event(unsigned long ptr)
+{
+	struct timespec __ts;
+	struct pps_ktime ts;
+
+	/* First of all we get the time stamp... */
+	getnstimeofday(&__ts);
+
+	pr_info("PPS event at %lu\n", jiffies);
+
+	/* ... and translate it to PPS time data struct */
+	ts.sec = __ts.tv_sec;
+	ts.nsec = __ts.tv_nsec;
+
+	pps_event(source, &ts, PPS_CAPTUREASSERT, NULL);
+
+	mod_timer(&ktimer, jiffies + HZ);
+}
+
+/*
+ * The echo function
+ */
+
+static void pps_ktimer_echo(int source, int event, void *data)
+{
+	pr_info("echo %s %s for source %d\n",
+		event & PPS_CAPTUREASSERT ? "assert" : "",
+		event & PPS_CAPTURECLEAR ? "clear" : "",
+		source);
+}
+
+/*
+ * The PPS info struct
+ */
+
+static struct pps_source_info pps_ktimer_info = {
+	.name		= "ktimer",
+	.path		= "",
+	.mode		= PPS_CAPTUREASSERT | PPS_OFFSETASSERT | \
+			  PPS_ECHOASSERT | \
+			  PPS_CANWAIT | PPS_TSFMT_TSPEC,
+	.echo 		= pps_ktimer_echo,
+	.owner		= THIS_MODULE,
+};
+
+/*
+ * Module staff
+ */
+
+static void __exit pps_ktimer_exit(void)
+{
+	del_timer_sync(&ktimer);
+	pps_unregister_source(source);
+
+	pr_info("ktimer PPS source unregistered\n");
+}
+
+static int __init pps_ktimer_init(void)
+{
+	int ret;
+
+	ret = pps_register_source(&pps_ktimer_info,
+				PPS_CAPTUREASSERT | PPS_OFFSETASSERT);
+	if (ret < 0) {
+		printk(KERN_ERR "cannot register ktimer source\n");
+		return ret;
+	}
+	source = ret;
+
+	setup_timer(&ktimer, pps_ktimer_event, 0);
+	mod_timer(&ktimer, jiffies + HZ);
+
+	pr_info("ktimer PPS source registered at %d\n", source);
+
+	return  0;
+}
+
+module_init(pps_ktimer_init);
+module_exit(pps_ktimer_exit);
+
+MODULE_AUTHOR("Rodolfo Giometti <giometti@linux.it>");
+MODULE_DESCRIPTION("dummy PPS source by using a kernel timer (just for debug)");
+MODULE_LICENSE("GPL");
-- 
1.6.3.3

[PATCH 4/8] ldisc: new dcd_change() method for line disciplines.

[Rodolfo Giometti]

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
---
 Documentation/serial/tty.txt |    4 ++++
 include/linux/tty_ldisc.h    |    8 ++++++++
 2 files changed, 12 insertions(+), 0 deletions(-)

diff --git a/Documentation/serial/tty.txt b/Documentation/serial/tty.txt
index 8e65c44..3fc812a 100644
--- a/Documentation/serial/tty.txt
+++ b/Documentation/serial/tty.txt
@@ -100,6 +100,10 @@ write_wakeup()	-	May be called at any point between open and close.
 			is permitted to call the driver write method from
 			this function. In such a situation defer it.
 
+dcd_change()	-	Report to the tty line the current DCD pin status
+			changes and the relative timestamp. The timestamp
+			can be NULL.
+
 
 Driver Access
 
diff --git a/include/linux/tty_ldisc.h b/include/linux/tty_ldisc.h
index 0c4ee9b..526d66f 100644
--- a/include/linux/tty_ldisc.h
+++ b/include/linux/tty_ldisc.h
@@ -99,6 +99,12 @@
  *	cease I/O to the tty driver. Can sleep. The driver should
  *	seek to perform this action quickly but should wait until
  *	any pending driver I/O is completed.
+ *
+ * void (*dcd_change)(struct tty_struct *tty, unsigned int status,
+ * 			struct timespec *ts)
+ *
+ *	Tells the discipline that the DCD pin has changed its status and
+ *	the relative timestamp. Pointer ts can be NULL.
  */
 
 #include <linux/fs.h>
@@ -136,6 +142,8 @@ struct tty_ldisc_ops {
 	void	(*receive_buf)(struct tty_struct *, const unsigned char *cp,
 			       char *fp, int count);
 	void	(*write_wakeup)(struct tty_struct *);
+	void	(*dcd_change)(struct tty_struct *, unsigned int,
+				struct timespec *);
 
 	struct  module *owner;
 	
-- 
1.6.3.3

[PATCH 5/8] ldisc n_tty: add new method n_tty_inherit_ops()

[Rodolfo Giometti]

This new method can be used to init a new struct tty_ldisc_ops as the
default tty_ldisc_N_TTY struct.

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
---
 drivers/char/n_tty.c |    9 +++++++++
 include/linux/tty.h  |    1 +
 2 files changed, 10 insertions(+), 0 deletions(-)

diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c
index 2e50f4d..0f11375 100644
--- a/drivers/char/n_tty.c
+++ b/drivers/char/n_tty.c
@@ -48,6 +48,7 @@
 #include <linux/audit.h>
 #include <linux/file.h>
 #include <linux/uaccess.h>
+#include <linux/module.h>
 
 #include <asm/system.h>
 
@@ -2091,3 +2092,11 @@ struct tty_ldisc_ops tty_ldisc_N_TTY = {
 	.receive_buf     = n_tty_receive_buf,
 	.write_wakeup    = n_tty_write_wakeup
 };
+
+void n_tty_inherit_ops(struct tty_ldisc_ops *ops)
+{
+	*ops = tty_ldisc_N_TTY;
+	ops->owner = NULL;
+	ops->refcount = ops->flags = 0;
+}
+EXPORT_SYMBOL_GPL(n_tty_inherit_ops);
diff --git a/include/linux/tty.h b/include/linux/tty.h
index f0f43d0..5091438 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -485,6 +485,7 @@ extern void tty_ldisc_enable(struct tty_struct *tty);
 
 /* n_tty.c */
 extern struct tty_ldisc_ops tty_ldisc_N_TTY;
+extern void n_tty_inherit_ops(struct tty_ldisc_ops *ops);
 
 /* tty_audit.c */
 #ifdef CONFIG_AUDIT
-- 
1.6.3.3

[PATCH 6/8] pps: serial clients support.

[Rodolfo Giometti]

Adds support, by using the PPS line discipline, for the PPS sources
connected with the CD (Carrier Detect) pin of a serial port.

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
---
 drivers/pps/clients/Kconfig     |    7 ++
 drivers/pps/clients/Makefile    |    1 +
 drivers/pps/clients/pps-ldisc.c |  154 +++++++++++++++++++++++++++++++++++++++
 include/linux/serial_core.h     |   11 +++-
 4 files changed, 172 insertions(+), 1 deletions(-)
 create mode 100644 drivers/pps/clients/pps-ldisc.c

diff --git a/drivers/pps/clients/Kconfig b/drivers/pps/clients/Kconfig
index 60b83be..487c1c8 100644
--- a/drivers/pps/clients/Kconfig
+++ b/drivers/pps/clients/Kconfig
@@ -15,4 +15,11 @@ config PPS_CLIENT_KTIMER
 	  This driver can also be built as a module.  If so, the module
 	  will be called ktimer.ko.
 
+config PPS_CLIENT_LDISC
+	tristate "PPS line discipline"
+	depends on PPS
+	help
+	  If you say yes here you get support for a PPS source connected
+	  with the CD (Carrier Detect) pin of your serial port.
+
 endif
diff --git a/drivers/pps/clients/Makefile b/drivers/pps/clients/Makefile
index f3c1e39..9f5b988 100644
--- a/drivers/pps/clients/Makefile
+++ b/drivers/pps/clients/Makefile
@@ -3,6 +3,7 @@
 #
 
 obj-$(CONFIG_PPS_CLIENT_KTIMER)	+= ktimer.o
+obj-$(CONFIG_PPS_CLIENT_LDISC)	+= pps-ldisc.o
 
 ifeq ($(CONFIG_PPS_DEBUG),y)
 EXTRA_CFLAGS += -DDEBUG
diff --git a/drivers/pps/clients/pps-ldisc.c b/drivers/pps/clients/pps-ldisc.c
new file mode 100644
index 0000000..59e6301
--- /dev/null
+++ b/drivers/pps/clients/pps-ldisc.c
@@ -0,0 +1,154 @@
+/*
+ * pps-ldisc.c -- PPS line discipline
+ *
+ *
+ * Copyright (C) 2008	Rodolfo Giometti <giometti@linux.it>
+ *
+ *   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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/module.h>
+#include <linux/serial_core.h>
+#include <linux/tty.h>
+#include <linux/pps_kernel.h>
+
+#define PPS_TTY_MAGIC		0x0001
+
+static void pps_tty_dcd_change(struct tty_struct *tty, unsigned int status,
+				struct timespec *ts)
+{
+	int id = (int) tty->disc_data;
+	struct timespec __ts;
+	struct pps_ktime pps_ts;
+
+	/* First of all we get the time stamp... */
+	getnstimeofday(&__ts);
+
+	/* Does caller give us a timestamp? */
+	if (ts) {	/* Yes. Let's use it! */
+		pps_ts.sec = ts->tv_sec;
+		pps_ts.nsec = ts->tv_nsec;
+	} else {	/* No. Do it ourself! */
+		pps_ts.sec = __ts.tv_sec;
+		pps_ts.nsec = __ts.tv_nsec;
+	}
+
+	/* Now do the PPS event report */
+	pps_event(id, &pps_ts, status ? PPS_CAPTUREASSERT : PPS_CAPTURECLEAR,
+			NULL);
+
+	pr_debug("PPS %s at %lu on source #%d\n",
+			status ? "assert" : "clear", jiffies, id);
+}
+
+static int (*n_tty_open)(struct tty_struct *tty);
+
+static int pps_tty_open(struct tty_struct *tty)
+{
+	struct pps_source_info info;
+	struct tty_driver *drv = tty->driver;
+	int index = tty->index + drv->name_base;
+	int ret;
+
+	info.owner = THIS_MODULE;
+	info.dev = NULL;
+	snprintf(info.name, PPS_MAX_NAME_LEN, "%s%d", drv->driver_name, index);
+	snprintf(info.path, PPS_MAX_NAME_LEN, "/dev/%s%d", drv->name, index);
+	info.mode = PPS_CAPTUREBOTH | \
+			PPS_OFFSETASSERT | PPS_OFFSETCLEAR | \
+			PPS_CANWAIT | PPS_TSFMT_TSPEC;
+
+	ret = pps_register_source(&info, PPS_CAPTUREBOTH | \
+				PPS_OFFSETASSERT | PPS_OFFSETCLEAR);
+	if (ret < 0) {
+		pr_err("cannot register PPS source \"%s\"\n", info.path);
+		return ret;
+	}
+	tty->disc_data = (void *) ret;
+
+	/* Should open N_TTY ldisc too */
+	ret = n_tty_open(tty);
+	if (ret < 0)
+		pps_unregister_source((int) tty->disc_data);
+
+	pr_info("PPS source #%d \"%s\" added\n", ret, info.path);
+
+	return 0;
+}
+
+static void (*n_tty_close)(struct tty_struct *tty);
+
+static void pps_tty_close(struct tty_struct *tty)
+{
+	int id = (int) tty->disc_data;
+
+	pps_unregister_source(id);
+	n_tty_close(tty);
+
+	pr_info("PPS source #%d removed\n", id);
+}
+
+static struct tty_ldisc_ops pps_ldisc_ops;
+
+/*
+ * Module stuff
+ */
+
+static int __init pps_tty_init(void)
+{
+	int err;
+
+	/* Inherit the N_TTY's ops */
+	n_tty_inherit_ops(&pps_ldisc_ops);
+
+	/* Save N_TTY's open()/close() methods */
+	n_tty_open = pps_ldisc_ops.open;
+	n_tty_close = pps_ldisc_ops.close;
+
+	/* Init PPS_TTY data */
+	pps_ldisc_ops.owner = THIS_MODULE;
+	pps_ldisc_ops.magic = PPS_TTY_MAGIC;
+	pps_ldisc_ops.name = "pps_tty";
+	pps_ldisc_ops.dcd_change = pps_tty_dcd_change;
+	pps_ldisc_ops.open = pps_tty_open;
+	pps_ldisc_ops.close = pps_tty_close;
+
+	err = tty_register_ldisc(N_PPS, &pps_ldisc_ops);
+	if (err)
+		pr_err("can't register PPS line discipline\n");
+	else
+		pr_info("PPS line discipline registered\n");
+
+	return err;
+}
+
+static void __exit pps_tty_cleanup(void)
+{
+	int err;
+
+	err = tty_unregister_ldisc(N_PPS);
+	if (err)
+		pr_err("can't unregister PPS line discipline\n");
+	else
+		pr_info("PPS line discipline removed\n");
+}
+
+module_init(pps_tty_init);
+module_exit(pps_tty_cleanup);
+
+MODULE_ALIAS_LDISC(N_PPS);
+MODULE_AUTHOR("Rodolfo Giometti <giometti@linux.it>");
+MODULE_DESCRIPTION("PPS TTY device driver");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index db532ce..4bfb046 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -488,9 +488,13 @@ uart_handle_dcd_change(struct uart_port *uport, unsigned int status)
 {
 	struct uart_state *state = uport->state;
 	struct tty_port *port = &state->port;
+	struct tty_ldisc *ld = tty_ldisc_ref(port->tty);
+	struct timespec ts;
 
-	uport->icount.dcd++;
+	if (ld && ld->ops->dcd_change)
+		getnstimeofday(&ts);
 
+	uport->icount.dcd++;
 #ifdef CONFIG_HARD_PPS
 	if ((uport->flags & UPF_HARDPPS_CD) && status)
 		hardpps();
@@ -502,6 +506,11 @@ uart_handle_dcd_change(struct uart_port *uport, unsigned int status)
 		else if (port->tty)
 			tty_hangup(port->tty);
 	}
+
+	if (ld && ld->ops->dcd_change)
+		ld->ops->dcd_change(port->tty, status, &ts);
+	if (ld)
+		tty_ldisc_deref(ld);
 }
 
 /**
-- 
1.6.3.3

[PATCH 7/8] serial 8250: enable PPS support.

[Rodolfo Giometti]

Automagically function serial8250_enable_ms() is called when PPS ldisc
is selected.

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
---
 drivers/serial/8250.c |   16 ++++++++++++++++
 1 files changed, 16 insertions(+), 0 deletions(-)

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 737b4c9..e70187b 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -2415,6 +2415,21 @@ serial8250_set_termios(struct uart_port *port, struct ktermios *termios,
 }
 
 static void
+serial8250_set_ldisc(struct uart_port *port)
+{
+	int line = port->line;
+
+	if (line >= port->state->port.tty->driver->num)
+		return;
+
+	if (port->state->port.tty->ldisc->ops->num == N_PPS) {
+		port->flags |= UPF_HARDPPS_CD;
+		serial8250_enable_ms(port);
+	} else
+		port->flags &= ~UPF_HARDPPS_CD;
+}
+
+static void
 serial8250_pm(struct uart_port *port, unsigned int state,
 	      unsigned int oldstate)
 {
@@ -2628,6 +2643,7 @@ static struct uart_ops serial8250_pops = {
 	.startup	= serial8250_startup,
 	.shutdown	= serial8250_shutdown,
 	.set_termios	= serial8250_set_termios,
+	.set_ldisc	= serial8250_set_ldisc,
 	.pm		= serial8250_pm,
 	.type		= serial8250_type,
 	.release_port	= serial8250_release_port,
-- 
1.6.3.3

[PATCH 8/8] serial amba-pl010: enable PPS support.

[Rodolfo Giometti]

Function pl010_enable_ms() is automagically called when PPS ldisc
is selected.

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
---
 drivers/serial/amba-pl010.c |   15 +++++++++++++++
 1 files changed, 15 insertions(+), 0 deletions(-)

diff --git a/drivers/serial/amba-pl010.c b/drivers/serial/amba-pl010.c
index 429a8ae..e4b3c2c 100644
--- a/drivers/serial/amba-pl010.c
+++ b/drivers/serial/amba-pl010.c
@@ -471,6 +471,20 @@ pl010_set_termios(struct uart_port *port, struct ktermios *termios,
 	spin_unlock_irqrestore(&uap->port.lock, flags);
 }
 
+static void pl010_set_ldisc(struct uart_port *port)
+{
+	int line = port->line;
+
+	if (line >= port->state->port.tty->driver->num)
+		return;
+
+	if (port->state->port.tty->ldisc->ops->num == N_PPS) {
+		port->flags |= UPF_HARDPPS_CD;
+		pl010_enable_ms(port);
+	} else
+		port->flags &= ~UPF_HARDPPS_CD;
+}
+
 static const char *pl010_type(struct uart_port *port)
 {
 	return port->type == PORT_AMBA ? "AMBA" : NULL;
@@ -531,6 +545,7 @@ static struct uart_ops amba_pl010_pops = {
 	.startup	= pl010_startup,
 	.shutdown	= pl010_shutdown,
 	.set_termios	= pl010_set_termios,
+	.set_ldisc	= pl010_set_ldisc,
 	.type		= pl010_type,
 	.release_port	= pl010_release_port,
 	.request_port	= pl010_request_port,
-- 
1.6.3.3

Re: [PATCH 6/8] pps: serial clients support.

[Andrew Morton]

On Tue, 16 Feb 2010 10:51:23 +0100
Rodolfo Giometti <giometti@linux.it> wrote:

> Adds support, by using the PPS line discipline, for the PPS sources
> connected with the CD (Carrier Detect) pin of a serial port.
> 
> ...
>
> +static int (*n_tty_open)(struct tty_struct *tty);
> +
> +static int pps_tty_open(struct tty_struct *tty)
> +{
> +	struct pps_source_info info;
> +	struct tty_driver *drv = tty->driver;
> +	int index = tty->index + drv->name_base;
> +	int ret;
> +
> +	info.owner = THIS_MODULE;
> +	info.dev = NULL;
> +	snprintf(info.name, PPS_MAX_NAME_LEN, "%s%d", drv->driver_name, index);
> +	snprintf(info.path, PPS_MAX_NAME_LEN, "/dev/%s%d", drv->name, index);
> +	info.mode = PPS_CAPTUREBOTH | \
> +			PPS_OFFSETASSERT | PPS_OFFSETCLEAR | \
> +			PPS_CANWAIT | PPS_TSFMT_TSPEC;
> +
> +	ret = pps_register_source(&info, PPS_CAPTUREBOTH | \
> +				PPS_OFFSETASSERT | PPS_OFFSETCLEAR);
> +	if (ret < 0) {
> +		pr_err("cannot register PPS source \"%s\"\n", info.path);
> +		return ret;
> +	}
> +	tty->disc_data = (void *) ret;
> +
> +	/* Should open N_TTY ldisc too */
> +	ret = n_tty_open(tty);
> +	if (ret < 0)
> +		pps_unregister_source((int) tty->disc_data);
> +
> +	pr_info("PPS source #%d \"%s\" added\n", ret, info.path);
> +
> +	return 0;
> +}

hm.  The n_tty_open/close layering is a bit grubby but I guess we don't
have a need to create anything more comprehensive at this stage.

A small stylistic thing: the code's a bit confusing at present because
the names n_tty_open/n_tty_close are identical to the global functions
over in n_tty.c.  A user of ctags will be tricked.  Also, the _way_ in
which these function pointers are called:

	ret = n_tty_open(tty);

make it look like the code is calling
drivers/char/n_tty.c:n_tty_open().  Except it isn't.

So I think it's better to do it with

	ret = (*n_tty_open)(tty);

to avoid misleading the reader.

And perhaps to rename these variables to avoid the duplication with the
global functions.

Re: [PATCH 6/8] pps: serial clients support.

[Rodolfo Giometti]

On Fri, Feb 19, 2010 at 01:53:59PM -0800, Andrew Morton wrote:
> 
> hm.  The n_tty_open/close layering is a bit grubby but I guess we don't
> have a need to create anything more comprehensive at this stage.
> 
> A small stylistic thing: the code's a bit confusing at present because
> the names n_tty_open/n_tty_close are identical to the global functions
> over in n_tty.c.  A user of ctags will be tricked.  Also, the _way_ in
> which these function pointers are called:
> 
> 	ret = n_tty_open(tty);
> 
> make it look like the code is calling
> drivers/char/n_tty.c:n_tty_open().  Except it isn't.
> 
> So I think it's better to do it with
> 
> 	ret = (*n_tty_open)(tty);
> 
> to avoid misleading the reader.
> 
> And perhaps to rename these variables to avoid the duplication with the
> global functions.

It could be acceptable using functions
alias_n_tty_open()/alias_n_tty_close() modifying the code as follow?

diff --git a/drivers/pps/clients/pps-ldisc.c
b/drivers/pps/clients/pps-ldisc.c
index 59e6301..15c4622 100644
--- a/drivers/pps/clients/pps-ldisc.c
+++ b/drivers/pps/clients/pps-ldisc.c
@@ -53,7 +53,7 @@ static void pps_tty_dcd_change(struct tty_struct
*tty, unsigne
                        status ? "assert" : "clear", jiffies, id);
 }
 
-static int (*n_tty_open)(struct tty_struct *tty);
+static int (*alias_n_tty_open)(struct tty_struct *tty);
 
 static int pps_tty_open(struct tty_struct *tty)
 {
@@ -79,7 +79,7 @@ static int pps_tty_open(struct tty_struct *tty)
        tty->disc_data = (void *) ret;
 
        /* Should open N_TTY ldisc too */
-       ret = n_tty_open(tty);
+       ret = alias_n_tty_open(tty);
        if (ret < 0)
                pps_unregister_source((int) tty->disc_data);
 
@@ -88,14 +88,14 @@ static int pps_tty_open(struct tty_struct *tty)
        return 0;
 }
 
-static void (*n_tty_close)(struct tty_struct *tty);
+static void (*alias_n_tty_close)(struct tty_struct *tty);
 
 static void pps_tty_close(struct tty_struct *tty)
 {
        int id = (int) tty->disc_data;
 
        pps_unregister_source(id);
-       n_tty_close(tty);
+       alias_n_tty_close(tty);
 
        pr_info("PPS source #%d removed\n", id);
 }

Ciao,

Rodolfo

-- 

GNU/Linux Solutions                  e-mail: giometti@enneenne.com
Linux Device Driver                          giometti@linux.it
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti
Freelance ICT Italia - Consulente ICT Italia - www.consulenti-ict.it

Re: [PATCH 5/8] ldisc n_tty: add new method n_tty_inherit_ops()

[Andrew Morton]

On Tue, 16 Feb 2010 10:51:22 +0100
Rodolfo Giometti <giometti@linux.it> wrote:

>  };
> +
> +void n_tty_inherit_ops(struct tty_ldisc_ops *ops)
> +{
> +	*ops = tty_ldisc_N_TTY;
> +	ops->owner = NULL;
> +	ops->refcount = ops->flags = 0;
> +}
> +EXPORT_SYMBOL_GPL(n_tty_inherit_ops);

Most (nearly all) functions in n_tty.c have been documented in
kerneldoc form.  Let's continue that please.

n_tty_inherit_ops() is of course pretty simple, but OTOH it's a global,
export-to-modules interface.

Re: [PATCH 2/8] pps: documentation programs and examples.

[Alexander Gordeev]

[-- Attachment #1: Type: text/plain, Size: 6755 bytes --]

On Tue, 16 Feb 2010 10:51:19 +0100
Rodolfo Giometti <giometti@linux.it> wrote:

> Here some utilities and examples about the PPS API and the LinuxPPS
> support.
> 
> * ppstest.c implements an useful testing program, while
> 
> * ppsfind tries to help the user into finding a specific PPS source by
>   using its name or path.

Maybe these tools should be in a separate package (say pps-tools)? I've
written several tools too. They are based on your code. I think there
should be at least two of them:

 * ppsctl to get/set params, get capabilities and bind/unbind kernel
consumer

 * ppstest to fetch data and provide several representations for it

I can assist in this task too.

> Signed-off-by: Rodolfo Giometti <giometti@linux.it>
> ---
>  Documentation/pps/Makefile  |   27 ++++++++
>  Documentation/pps/ppsfind   |   17 +++++
>  Documentation/pps/ppstest.c |  150
> +++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 194
> insertions(+), 0 deletions(-) create mode 100644
> Documentation/pps/Makefile create mode 100644
> Documentation/pps/ppsfind create mode 100644
> Documentation/pps/ppstest.c
> 
> diff --git a/Documentation/pps/Makefile b/Documentation/pps/Makefile
> new file mode 100644
> index 0000000..3848aba
> --- /dev/null
> +++ b/Documentation/pps/Makefile
> @@ -0,0 +1,27 @@
> +TARGETS = ppstest
> +
> +CFLAGS += -Wall -O2 -D_GNU_SOURCE
> +CFLAGS += -I . -I ../../include/
> +CFLAGS += -ggdb
> +
> +# -- Actions section --
> +
> +.PHONY : all depend dep
> +
> +all : .depend $(TARGETS)
> +
> +.depend depend dep :
> +	$(CC) $(CFLAGS) -M $(TARGETS:=.c) > .depend
> +
> +ifeq (.depend,$(wildcard .depend))
> +include .depend
> +endif
> +
> +
> +# -- Clean section --
> +
> +.PHONY : clean
> +
> +clean :
> +	rm -f *.o *~ core .depend
> +	rm -f ${TARGETS}
> diff --git a/Documentation/pps/ppsfind b/Documentation/pps/ppsfind
> new file mode 100644
> index 0000000..93c0e17
> --- /dev/null
> +++ b/Documentation/pps/ppsfind
> @@ -0,0 +1,17 @@
> +#!/bin/sh
> +
> +SYS="/sys/class/pps/"
> +
> +if [ $# -lt 1 ] ; then
> +	echo "usage: ppsfind <name>" >&2
> +	exit 1
> +fi
> +
> +for d in $(ls $SYS) ; do
> +	if grep $1 $SYS/$d/name >& /dev/null || \
> +	   grep $1 $SYS/$d/path >& /dev/null ; then
> +		echo "$d: name=$(cat $SYS/$d/name) path=$(cat
> $SYS/$d/path)"
> +	fi
> +done
> +
> +exit 0
> diff --git a/Documentation/pps/ppstest.c b/Documentation/pps/ppstest.c
> new file mode 100644
> index 0000000..30c8f6d
> --- /dev/null
> +++ b/Documentation/pps/ppstest.c
> @@ -0,0 +1,150 @@
> +#include <stdio.h>
> +#include <stdlib.h>
> +#include <unistd.h>
> +#include <fcntl.h>
> +#include <sys/types.h>
> +#include <sys/stat.h>
> +
> +#include <timepps.h>
> +
> +int find_source(char *path, pps_handle_t *handle, int *avail_mode)
> +{
> +	pps_params_t params;
> +	int ret;
> +
> +	printf("trying PPS source \"%s\"\n", path);
> +
> +	/* Try to find the source by using the supplied "path" name
> */
> +	ret = open(path, O_RDWR);
> +	if (ret < 0) {
> +		fprintf(stderr, "unable to open device
> \"%s\" (%m)\n", path);
> +		return ret;
> +	}
> +
> +	/* Open the PPS source (and check the file descriptor) */
> +	ret = time_pps_create(ret, handle);
> +	if (ret < 0) {
> +		fprintf(stderr, "cannot create a PPS source from
> device "
> +				"\"%s\" (%m)\n", path);
> +		return -1;
> +	}
> +	printf("found PPS source \"%s\"\n", path);
> +
> +	/* Find out what features are supported */
> +	ret = time_pps_getcap(*handle, avail_mode);
> +	if (ret < 0) {
> +		fprintf(stderr, "cannot get capabilities (%m)\n");
> +		return -1;
> +	}
> +	if ((*avail_mode & PPS_CAPTUREASSERT) == 0) {
> +		fprintf(stderr, "cannot CAPTUREASSERT\n");
> +		return -1;
> +	}
> +	if ((*avail_mode & PPS_OFFSETASSERT) == 0) {
> +		fprintf(stderr, "cannot OFFSETASSERT\n");
> +		return -1;
> +	}
> +
> +	/* Capture assert timestamps, and compensate for a 675 nsec
> +	 * propagation delay */
> +	ret = time_pps_getparams(*handle, &params);
> +	if (ret < 0) {
> +		fprintf(stderr, "cannot get parameters (%m)\n");
> +		return -1;
> +	}
> +	params.assert_offset.tv_sec = 0;
> +	params.assert_offset.tv_nsec = 675;
> +	params.mode |= PPS_CAPTUREASSERT | PPS_OFFSETASSERT;
> +	params.mode &= ~(PPS_CAPTURECLEAR | PPS_OFFSETCLEAR);
> +	ret = time_pps_setparams(*handle, &params);
> +	if (ret < 0) {
> +		fprintf(stderr, "cannot set parameters (%m)\n");
> +		return -1;
> +	}
> +
> +	return 0;
> +}
> +
> +int fetch_source(int i, pps_handle_t *handle, int *avail_mode)
> +{
> +	struct timespec timeout;
> +	pps_info_t infobuf;
> +	int ret;
> +
> +	/* create a zero-valued timeout */
> +	timeout.tv_sec = 3;
> +	timeout.tv_nsec = 0;
> +
> +retry:
> +	if (*avail_mode & PPS_CANWAIT) /* waits for the next event */
> +		ret = time_pps_fetch(*handle, PPS_TSFMT_TSPEC,
> &infobuf,
> +				   &timeout);
> +	else {
> +		sleep(1);
> +		ret = time_pps_fetch(*handle, PPS_TSFMT_TSPEC,
> &infobuf,
> +				   &timeout);
> +	}
> +	if (ret < 0) {
> +		if (ret == -EINTR) {
> +			fprintf(stderr, "time_pps_fetch() got a
> signal!\n");
> +			goto retry;
> +		}
> +
> +		fprintf(stderr, "time_pps_fetch() error %d (%m)\n",
> ret);
> +		return -1;
> +	}
> +
> +	printf("source %d - "
> +	       "assert %ld.%09ld, sequence: %ld - "
> +	       "clear  %ld.%09ld, sequence: %ld\n",
> +	       i,
> +	       infobuf.assert_timestamp.tv_sec,
> +	       infobuf.assert_timestamp.tv_nsec,
> +	       infobuf.assert_sequence,
> +	       infobuf.clear_timestamp.tv_sec,
> +	       infobuf.clear_timestamp.tv_nsec,
> infobuf.clear_sequence); +
> +	return 0;
> +}
> +
> +void usage(char *name)
> +{
> +	fprintf(stderr, "usage: %s <ppsdev> [<ppsdev> ...]\n", name);
> +	exit(EXIT_FAILURE);
> +}
> +
> +int main(int argc, char *argv[])
> +{
> +	int num;
> +	pps_handle_t handle[4];
> +	int avail_mode[4];
> +	int i = 0;
> +	int ret;
> +
> +	/* Check the command line */
> +	if (argc < 2)
> +		usage(argv[0]);
> +
> +	for (i = 1; i < argc && i <= 4; i++) {
> +		ret = find_source(argv[i], &handle[i - 1],
> &avail_mode[i - 1]);
> +		if (ret < 0)
> +			exit(EXIT_FAILURE);
> +	}
> +
> +	num = i - 1;
> +	printf("ok, found %d source(s), now start fetching
> data...\n", num); +
> +	/* loop, printing the most recent timestamp every second or
> so */
> +	while (1) {
> +		for (i = 0; i < num; i++) {
> +			ret = fetch_source(i, &handle[i],
> &avail_mode[i]);
> +			if (ret < 0 && errno != ETIMEDOUT)
> +				exit(EXIT_FAILURE);
> +		}
> +	}
> +
> +	for (; i >= 0; i--)
> +		time_pps_destroy(handle[i]);
> +
> +	return 0;
> +}


-- 
  Alexander

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 489 bytes --]

Re: [PATCH 2/8] pps: documentation programs and examples.

[Rodolfo Giometti]

On Tue, Feb 16, 2010 at 03:44:56PM +0300, Alexander Gordeev wrote:
> On Tue, 16 Feb 2010 10:51:19 +0100
> Rodolfo Giometti <giometti@linux.it> wrote:
> 
> > Here some utilities and examples about the PPS API and the LinuxPPS
> > support.
> > 
> > * ppstest.c implements an useful testing program, while
> > 
> > * ppsfind tries to help the user into finding a specific PPS source by
> >   using its name or path.
> 
> Maybe these tools should be in a separate package (say pps-tools)? I've
> written several tools too. They are based on your code. I think there
> should be at least two of them:
> 
>  * ppsctl to get/set params, get capabilities and bind/unbind kernel
> consumer
> 
>  * ppstest to fetch data and provide several representations for it
> 
> I can assist in this task too.

Great! It's ok for me. Please, let me know how we can proceed.

Thanks a lot,

Rodolfo

-- 

GNU/Linux Solutions                  e-mail: giometti@enneenne.com
Linux Device Driver                          giometti@linux.it
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti
Freelance ICT Italia - Consulente ICT Italia - www.consulenti-ict.it

Re: [PATCH 1/8] pps: userland header file for PPS API.

[Alexander Gordeev]

[-- Attachment #1: Type: text/plain, Size: 7738 bytes --]

Hi Rodolfo,

On Tue, 16 Feb 2010 10:51:18 +0100
Rodolfo Giometti <giometti@linux.it> wrote:

> This patch adds into the PPS's documentation directory a possible
> implementation of the PPS API (RFC 2783) by using the LinuxPPS's char
> devices.
> 
> This file is not just an example but it can be used into real
> systems. :)

Isn't libc a proper place for this file? This header file provides a
generic standardized interface (PPSAPI) to the custom Linux
implementation (linux/pps.h). I think this is just what libc is for.

I wanted to file a bug in Debian BTS last week to add this header to
libc6-dev package and also push it upstream but something distracted
me. If you agree that the header should be in libc then I can do it now.

> Signed-off-by: Rodolfo Giometti <giometti@linux.it>
> ---
>  Documentation/pps/timepps.h |  198
> +++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 198
> insertions(+), 0 deletions(-) create mode 100644
> Documentation/pps/timepps.h
> 
> diff --git a/Documentation/pps/timepps.h b/Documentation/pps/timepps.h
> new file mode 100644
> index 0000000..d2628d2
> --- /dev/null
> +++ b/Documentation/pps/timepps.h
> @@ -0,0 +1,198 @@
> +/*
> + * timepps.h -- PPS API main header
> + *
> + * Copyright (C) 2005-2007   Rodolfo Giometti <giometti@linux.it>
> + *
> + *   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., 675 Mass Ave, Cambridge, MA 02139, USA.
> + */
> +
> +#ifndef _SYS_TIMEPPS_H_
> +#define _SYS_TIMEPPS_H_
> +
> +#include <errno.h>
> +#include <sys/time.h>
> +#include <sys/ioctl.h>
> +#include <linux/types.h>
> +#include <linux/pps.h>
> +
> +#define LINUXPPS	1		/* signal we are using
> LinuxPPS */ +
> +/*
> + * New data structures
> + */
> +
> +struct ntp_fp {
> +	unsigned int integral;
> +	unsigned int fractional;
> +};
> +
> +union pps_timeu {
> +	struct timespec tspec;
> +	struct ntp_fp ntpfp;
> +	unsigned long longpad[3];
> +};
> +
> +struct pps_info {
> +	unsigned long assert_sequence;	/* seq. num. of assert
> event */
> +	unsigned long clear_sequence;	/* seq. num. of clear
> event */
> +	union pps_timeu assert_tu;	/* time of assert event */
> +	union pps_timeu clear_tu;	/* time of clear event */
> +	int current_mode;		/* current mode bits */
> +};
> +
> +struct pps_params {
> +	int api_version;		/* API version # */
> +	int mode;			/* mode bits */
> +	union pps_timeu assert_off_tu;	/* offset compensation
> for assert */
> +	union pps_timeu clear_off_tu;	/* offset compensation
> for clear */ +};
> +
> +typedef int pps_handle_t;		/* represents a PPS source
> */ +typedef unsigned long pps_seq_t;	/* sequence number */
> +typedef struct ntp_fp ntp_fp_t;		/* NTP-compatible
> time stamp */ +typedef union pps_timeu pps_timeu_t;	/* generic
> data type for time stamps */ +typedef struct pps_info pps_info_t;
> +typedef struct pps_params pps_params_t;
> +
> +#define assert_timestamp        assert_tu.tspec
> +#define clear_timestamp         clear_tu.tspec
> +
> +#define assert_timestamp_ntpfp  assert_tu.ntpfp
> +#define clear_timestamp_ntpfp   clear_tu.ntpfp
> +
> +#define assert_offset		assert_off_tu.tspec
> +#define clear_offset		clear_off_tu.tspec
> +
> +#define assert_offset_ntpfp     assert_off_tu.ntpfp
> +#define clear_offset_ntpfp      clear_off_tu.ntpfp
> +
> +/*
> + * The PPS API
> + */
> +
> +static __inline int time_pps_create(int source, pps_handle_t *handle)
> +{
> +	int ret;
> +	struct pps_kparams dummy;
> +
> +	if (!handle) {
> +		errno = EINVAL;
> +		return -1;
> +	}
> +
> +	/* First we check if current device is a valid PPS one by
> +	 * doing a dummy PPS_GETPARAMS...
> +	 */
> +	ret = ioctl(source, PPS_GETPARAMS, &dummy);
> +	if (ret) {
> +		errno = EOPNOTSUPP;
> +		return -1;
> +	}
> +
> +	/* ... then since in LinuxPPS there are no differences
> between a
> +	 * "PPS source" and a "PPS handle", we simply return the
> same value.
> +	 */
> +	*handle = source;
> +
> +	return 0;
> +}
> +
> +static __inline int time_pps_destroy(pps_handle_t handle)
> +{
> +	return close(handle);
> +}
> +
> +static __inline int time_pps_getparams(pps_handle_t handle,
> +					pps_params_t *ppsparams)
> +{
> +	int ret;
> +	struct pps_kparams __ppsparams;
> +
> +	ret = ioctl(handle, PPS_GETPARAMS, &__ppsparams);
> +
> +	ppsparams->api_version = __ppsparams.api_version;
> +	ppsparams->mode = __ppsparams.mode;
> +	ppsparams->assert_off_tu.tspec.tv_sec =
> __ppsparams.assert_off_tu.sec;
> +	ppsparams->assert_off_tu.tspec.tv_nsec =
> __ppsparams.assert_off_tu.nsec;
> +	ppsparams->clear_off_tu.tspec.tv_sec =
> __ppsparams.clear_off_tu.sec;
> +	ppsparams->clear_off_tu.tspec.tv_nsec =
> __ppsparams.clear_off_tu.nsec; +
> +	return ret;
> +}
> +
> +static __inline int time_pps_setparams(pps_handle_t handle,
> +					const pps_params_t
> *ppsparams) +{
> +	struct pps_kparams __ppsparams;
> +
> +	__ppsparams.api_version = ppsparams->api_version;
> +	__ppsparams.mode = ppsparams->mode;
> +	__ppsparams.assert_off_tu.sec =
> ppsparams->assert_off_tu.tspec.tv_sec;
> +	__ppsparams.assert_off_tu.nsec =
> ppsparams->assert_off_tu.tspec.tv_nsec;
> +	__ppsparams.clear_off_tu.sec =
> ppsparams->clear_off_tu.tspec.tv_sec;
> +	__ppsparams.clear_off_tu.nsec =
> ppsparams->clear_off_tu.tspec.tv_nsec; +
> +	return ioctl(handle, PPS_SETPARAMS, &__ppsparams);
> +}
> +
> +/* Get capabilities for handle */
> +static __inline int time_pps_getcap(pps_handle_t handle, int *mode)
> +{
> +	return ioctl(handle, PPS_GETCAP, mode);
> +}
> +
> +static __inline int time_pps_fetch(pps_handle_t handle, const int
> tsformat,
> +					pps_info_t *ppsinfobuf,
> +					const struct timespec
> *timeout) +{
> +	struct pps_fdata __fdata;
> +	int ret;
> +
> +	/* Sanity checks */
> +	if (tsformat != PPS_TSFMT_TSPEC) {
> +		errno = EINVAL;
> +		return -1;
> +	}
> +
> +	if (timeout) {
> +		__fdata.timeout.sec = timeout->tv_sec;
> +		__fdata.timeout.nsec = timeout->tv_nsec;
> +		__fdata.timeout.flags = ~PPS_TIME_INVALID;
> +	} else
> +		__fdata.timeout.flags = PPS_TIME_INVALID;
> +
> +	ret = ioctl(handle, PPS_FETCH, &__fdata);
> +
> +	ppsinfobuf->assert_sequence = __fdata.info.assert_sequence;
> +	ppsinfobuf->clear_sequence = __fdata.info.clear_sequence;
> +	ppsinfobuf->assert_tu.tspec.tv_sec =
> __fdata.info.assert_tu.sec;
> +	ppsinfobuf->assert_tu.tspec.tv_nsec =
> __fdata.info.assert_tu.nsec;
> +	ppsinfobuf->clear_tu.tspec.tv_sec =
> __fdata.info.clear_tu.sec;
> +	ppsinfobuf->clear_tu.tspec.tv_nsec =
> __fdata.info.clear_tu.nsec;
> +	ppsinfobuf->current_mode = __fdata.info.current_mode;
> +
> +	return ret;
> +}
> +
> +static __inline int time_pps_kcbind(pps_handle_t handle,
> +					const int kernel_consumer,
> +					const int edge, const int
> tsformat) +{
> +	/* LinuxPPS doesn't implement kernel consumer feature */
> +	errno = EOPNOTSUPP;
> +	return -1;
> +}
> +
> +#endif				/* _SYS_TIMEPPS_H_ */


-- 
  Alexander

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 489 bytes --]

Re: [PATCH 1/8] pps: userland header file for PPS API.

[Rodolfo Giometti]

On Tue, Feb 16, 2010 at 03:33:01PM +0300, Alexander Gordeev wrote:
> Hi Rodolfo,
> 
> On Tue, 16 Feb 2010 10:51:18 +0100
> Rodolfo Giometti <giometti@linux.it> wrote:
> 
> > This patch adds into the PPS's documentation directory a possible
> > implementation of the PPS API (RFC 2783) by using the LinuxPPS's char
> > devices.
> > 
> > This file is not just an example but it can be used into real
> > systems. :)
> 
> Isn't libc a proper place for this file? This header file provides a
> generic standardized interface (PPSAPI) to the custom Linux
> implementation (linux/pps.h). I think this is just what libc is for.

You are right, but I added that file for documentation purpose also.
Just to illustrate how to implement an high level API with my simple
char device interface.

> I wanted to file a bug in Debian BTS last week to add this header to
> libc6-dev package and also push it upstream but something distracted
> me. If you agree that the header should be in libc then I can do it now.

I agree. :)

Can you please put me in Cc in your letters with the Debian people?

Thanks a lot,

Rodolfo

-- 

GNU/Linux Solutions                  e-mail: giometti@enneenne.com
Linux Device Driver                          giometti@linux.it
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti
Freelance ICT Italia - Consulente ICT Italia - www.consulenti-ict.it

Re: LinuxPPS new functionalities

[Alan Cox]

On Tue, 16 Feb 2010 10:51:17 +0100
Rodolfo Giometti <giometti@linux.it> wrote:

> This patchset adds several PPS functionalities.
> 
> All patches have a specific comment but I prefer say something more:
> 
> * Patches 01-02 are just documentation about PPS support.
> 
> * Patch 03 adds PPS client support skeleton.
> 
> * Patches 04-06 add generic PPS serial support (with specific
>   modifications requested by Alan Cox).

Thanks - the serial/tty stuff is much much cleaner in this patch set.

LinuxPPS new functionalities

[Rodolfo Giometti]

This patchset adds several PPS functionalities.

All patches have a specific comment but I prefer say something more:

* Patches 01-02 are just documentation about PPS support.

* Patch 03 adds PPS client support skeleton.

* Patches 04-06 add generic PPS serial support.

* Patches 07 and 10 add PPS serial support for specific hardware.

* Patch 08 adds PPS parallel support.

* Patch 09 is going to add a low level IRQ timestamps recording. Here
  some notes:

   1) This is just a proposal, so I'm quite sure it will be rejected,
      but I decided to propose it anyway in order to have some
      feedbacks.

   2) Due previous point I didn't split up the patch into several (and
      more logical) sub-parts in order to have a better review about
      what I wish to do.

* Patch 11 shows how to add a low level IRQ timestamps recording for a
  new platform once patch 09 has been added into main tree.

Even if patches 09 and 11 will be rejected I hope the others are
ok. :)

Ciao,

Rodolfo