Skip to contentSkip to conversation

LinuxPPS (RESUBMIT 6): the PPS Linux implementation.

PATCH In review Rodolfo Giometti 5 participants 26 messages Aug 19, 2008 – Aug 20, 2008

Subsystems (6)
lore
Share & tools
lore
#0 lore raw

This patch set adds the PPS support into Linux.

PPS means "pulse per second" and its API is specified by RFC 2783 (Pulse-Per-Second API for UNIX-like Operating Systems, Version 1.0).

The code has been tested with the NTPD program
(http://www.eecis.udel.edu/~mills/ntp/html/index.html) and several GPS antennae.

Please, note that since last submit a new line discipline has been added in order to leave untouched the serial ports code (as suggested by Alan).

Rodolfo

--

 b/Documentation/ABI/testing/sysfs-pps |   73 +++++++
 b/Documentation/ioctl-number.txt      |    2 
 b/Documentation/pps/Makefile          |   28 ++
 b/Documentation/pps/pps.txt           |  172 +++++++++++++++++
 b/Documentation/pps/ppsfind           |   17 +
 b/Documentation/pps/ppsldisc.c        |   46 ++++
 b/Documentation/pps/ppstest.c         |  151 +++++++++++++++
 b/Documentation/pps/timepps.h         |  195 +++++++++++++++++++
 b/Documentation/tty.txt               |    4 
 b/MAINTAINERS                         |    7 
 b/arch/x86/kernel/irq_32.c            |   18 +
 b/arch/x86/kernel/irq_64.c            |   22 ++
 b/drivers/Kconfig                     |    2 
 b/drivers/Makefile                    |    1 
 b/drivers/char/lp.c                   |   61 ++++++
 b/drivers/char/n_tty.c                |   54 +++--
 b/drivers/pps/Kconfig                 |   33 +++
 b/drivers/pps/Makefile                |    8 
 b/drivers/pps/clients/Kconfig         |   18 +
 b/drivers/pps/clients/Makefile        |    9 
 b/drivers/pps/clients/ktimer.c        |  124 ++++++++++++
 b/drivers/pps/clients/pps-ldisc.c     |  154 +++++++++++++++
 b/drivers/pps/kapi.c                  |  322 ++++++++++++++++++++++++++++++++
 b/drivers/pps/pps.c                   |  335 ++++++++++++++++++++++++++++++++++
 b/drivers/pps/sysfs.c                 |  104 ++++++++++
 b/drivers/serial/8250.c               |   13 +
 b/include/linux/Kbuild                |    1 
 b/include/linux/parport.h             |   22 ++
 b/include/linux/pps.h                 |  202 ++++++++++++++++++++
 b/include/linux/serial_core.h         |   10 -
 b/include/linux/tty.h                 |   16 +
 b/include/linux/tty_ldisc.h           |   10 +
 drivers/pps/Kconfig                   |   14 +
 drivers/pps/Makefile                  |    1 
 drivers/pps/clients/Kconfig           |   17 +
 drivers/pps/clients/Makefile          |    1 
 drivers/pps/clients/pps-ldisc.c       |   18 +
 include/linux/pps.h                   |    1 
 include/linux/tty.h                   |    3 
 39 files changed, 2263 insertions(+), 26 deletions(-)
#1 lore raw

PATCH 1/10 LinuxPPS core support.

This patch adds the kernel side of the PPS support currently named "LinuxPPS".

PPS means "pulse per second" and a PPS source is just a device which provides a high precision signal each second so that an application can use it to adjust system clock time.

Common use is the combination of the NTPD as userland program with a GPS receiver as PPS source to obtain a wallclock-time with sub-millisecond synchronisation to UTC.

To obtain this goal the userland programs shoud use the PPS API specification (RFC 2783 - Pulse-Per-Second API for UNIX-like Operating Systems, Version 1.0) which in part is implemented by this patch. It provides a set of chars devices, one per PPS source, which can be used to get the time signal. The RFC's functions can be implemented by accessing to these char devices.

---
 Documentation/ABI/testing/sysfs-pps |   73 ++++++++
 Documentation/ioctl-number.txt      |    2 +
 Documentation/pps/pps.txt           |  172 ++++++++++++++++++
 MAINTAINERS                         |    7 +
 drivers/Kconfig                     |    2 +
 drivers/Makefile                    |    1 +
 drivers/pps/Kconfig                 |   33 ++++
 drivers/pps/Makefile                |    8 +
 drivers/pps/kapi.c                  |  322 +++++++++++++++++++++++++++++++++
 drivers/pps/pps.c                   |  335 +++++++++++++++++++++++++++++++++++
 drivers/pps/sysfs.c                 |  104 +++++++++++
 include/linux/Kbuild                |    1 +
 include/linux/pps.h                 |  202 +++++++++++++++++++++
 13 files changed, 1262 insertions(+), 0 deletions(-)
 create mode 100644 Documentation/ABI/testing/sysfs-pps
 create mode 100644 Documentation/pps/pps.txt
 create mode 100644 drivers/pps/Kconfig
 create mode 100644 drivers/pps/Makefile
 create mode 100644 drivers/pps/kapi.c
 create mode 100644 drivers/pps/pps.c
 create mode 100644 drivers/pps/sysfs.c
 create mode 100644 include/linux/pps.h
Documentation/ABI/testing/sysfs-pps new file +73 -0
new file mode 100644
index 0000000..8cba1d9
@@ -0,0 +1,73 @@
+What:		/sys/class/pps/
+Date:		February 2008
+Contact:	Rodolfo Giometti <[email protected]>
+Description:
+		The /sys/class/pps/ directory will contain files and
+		directories that will provide a unified interface to
+		the PPS sources.
+
+What:		/sys/class/pps/ppsX/
+Date:		February 2008
+Contact:	Rodolfo Giometti <[email protected]>
+Description:
+		The /sys/class/pps/ppsX/ directory is related to X-th
+		PPS source into the system. Each directory will
+		contain files to manage and control its PPS source.
+
+What:		/sys/class/pps/ppsX/assert
+Date:		February 2008
+Contact:	Rodolfo Giometti <[email protected]>
+Description:
+		The /sys/class/pps/ppsX/assert file reports the assert events
+		and the assert sequence number of the X-th source in the form:
+
+			<secs>.<nsec>#<sequence>
+
+		If the source has no assert events the content of this file
+		is void.
+
+What:		/sys/class/pps/ppsX/clear
+Date:		February 2008
+Contact:	Rodolfo Giometti <[email protected]>
+Description:
+		The /sys/class/pps/ppsX/clear file reports the clear events
+		and the clear sequence number of the X-th source in the form:
+
+			<secs>.<nsec>#<sequence>
+
+		If the source has no clear events the content of this file
+		is void.
+
+What:		/sys/class/pps/ppsX/mode
+Date:		February 2008
+Contact:	Rodolfo Giometti <[email protected]>
+Description:
+		The /sys/class/pps/ppsX/mode file reports the functioning
+		mode of the X-th source in hexadecimal encoding.
+
+		Please, refer to linux/include/linux/pps.h for further
+		info.
+
+What:		/sys/class/pps/ppsX/echo
+Date:		February 2008
+Contact:	Rodolfo Giometti <[email protected]>
+Description:
+		The /sys/class/pps/ppsX/echo file reports if the X-th source
+		supports or not an "echo" function.
+
+What:		/sys/class/pps/ppsX/name
+Date:		February 2008
+Contact:	Rodolfo Giometti <[email protected]>
+Description:
+		The /sys/class/pps/ppsX/name file reports the name of the
+		X-th source.
+
+What:		/sys/class/pps/ppsX/path
+Date:		February 2008
+Contact:	Rodolfo Giometti <[email protected]>
+Description:
+		The /sys/class/pps/ppsX/path file reports the path name of
+		the device connected with the X-th source.
+
+		If the source is not connected with any device the content
+		of this file is void.
Documentation/ioctl-number.txt +2 -0
index 1c6b545..963516d 100644
@@ -97,6 +97,8 @@ Code	Seq#	Include File		Comments
 'M'	all	linux/soundcard.h
 'N'	00-1F	drivers/usb/scanner.h
 'P'	all	linux/soundcard.h
+'P'	00-04	linux/pps.h		LinuxPPS
+					<mailto:[email protected]>
 'Q'	all	linux/soundcard.h
 'R'	00-1F	linux/random.h
 'S'	all	linux/cdrom.h		conflict!
Documentation/pps/pps.txt new file +172 -0

172 lines not on this page · raw message

MAINTAINERS +7 -0
index 663485b..a6dfeaa 100644
@@ -3373,6 +3373,13 @@ P:	James Chapman
 M:	[email protected]
 S:	Maintained
 
+PPS SUPPORT
+P:	Rodolfo Giometti
+M:	[email protected]
+W:	http://wiki.enneenne.com/index.php/LinuxPPS_support
+L:	[email protected] (subscribers-only)
+S:	Maintained
+
 PREEMPTIBLE KERNEL
 P:	Robert Love
 M:	[email protected]
drivers/Kconfig +2 -0
index 59f33fa..473fa6f 100644
@@ -52,6 +52,8 @@ source "drivers/i2c/Kconfig"
 
 source "drivers/spi/Kconfig"
 
+source "drivers/pps/Kconfig"
+
 source "drivers/gpio/Kconfig"
 
 source "drivers/w1/Kconfig"
drivers/Makefile +1 -0
index 2735bde..2d46028 100644
@@ -66,6 +66,7 @@ obj-$(CONFIG_INPUT)		+= input/
 obj-$(CONFIG_I2O)		+= message/
 obj-$(CONFIG_RTC_LIB)		+= rtc/
 obj-y				+= i2c/
+obj-$(CONFIG_PPS)		+= pps/
 obj-$(CONFIG_W1)		+= w1/
 obj-$(CONFIG_POWER_SUPPLY)	+= power/
 obj-$(CONFIG_HWMON)		+= hwmon/
drivers/pps/Kconfig new file +33 -0
new file mode 100644
index 0000000..cc2eb8e
@@ -0,0 +1,33 @@
+#
+# PPS support configuration
+#
+
+menu "PPS support"
+
+config PPS
+	tristate "PPS support"
+	depends on EXPERIMENTAL
+	---help---
+	  PPS (Pulse Per Second) is a special pulse provided by some GPS
+	  antennae. Userland can use it to get a high-precision time
+	  reference.
+
+	  Some antennae's PPS signals are connected with the CD (Carrier
+	  Detect) pin of the serial line they use to communicate with the
+	  host. In this case use the SERIAL_LINE client support.
+
+	  Some antennae's PPS signals are connected with some special host
+	  inputs so you have to enable the corresponding client support.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called pps_core.ko.
+
+config PPS_DEBUG
+	bool "PPS debugging messages"
+	depends on PPS
+	help
+	  Say Y here if you want the PPS support to produce a bunch of debug
+	  messages to the system log.  Select this if you are having a
+	  problem with PPS support and want to see more of what is going on.
+
+endmenu
drivers/pps/Makefile new file +8 -0
new file mode 100644
index 0000000..19ea582
@@ -0,0 +1,8 @@
+#
+# Makefile for the PPS core.
+#
+
+pps_core-y			:= pps.o kapi.o sysfs.o
+obj-$(CONFIG_PPS)		:= pps_core.o
+
+ccflags-$(CONFIG_PPS_DEBUG) := -DDEBUG
drivers/pps/kapi.c new file +322 -0

322 lines not on this page · raw message

drivers/pps/pps.c new file +335 -0

335 lines not on this page · raw message

drivers/pps/sysfs.c new file +104 -0

104 lines not on this page · raw message

include/linux/Kbuild +1 -0
index 7d97067..f6bc168 100644
@@ -303,6 +303,7 @@ unifdef-y += pmu.h
 unifdef-y += poll.h
 unifdef-y += ppp_defs.h
 unifdef-y += ppp-comp.h
+unifdef-y += pps.h
 unifdef-y += ptrace.h
 unifdef-y += qnx4_fs.h
 unifdef-y += quota.h
include/linux/pps.h new file +202 -0

202 lines not on this page · raw message

Signature · 1 line
1.5.4.3
#2 lore raw

PATCH 2/10 PPS: userland header file for PPS API.

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. :)

---
 Documentation/pps/timepps.h |  195 +++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 195 insertions(+), 0 deletions(-)
 create mode 100644 Documentation/pps/timepps.h
Documentation/pps/timepps.h new file +195 -0
new file mode 100644
index 0000000..2372949
@@ -0,0 +1,195 @@
+/*
+ * timepps.h -- PPS API main header
+ *
+ * Copyright (C) 2005-2007   Rodolfo Giometti <[email protected]>
+ *
+ *   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 <sys/syscall.h>
+#include <unistd.h>
+#include <errno.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;
+
+	if (!handle) {
+		errno = EINVAL;
+		return -1;
+	}
+
+	/* First we check if current device is a PPS valid PPS one...
+	 */
+	ret = ioctl(source, PPS_CHECK);
+	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_ */
Signature · 1 line
1.5.4.3
#3 lore raw

PATCH 3/10 PPS: documentation programs and examples.

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

* ppsldisc.c shows how to manage PPS line discipline;

* 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.

---
 Documentation/pps/Makefile   |   28 ++++++++
 Documentation/pps/ppsfind    |   17 +++++
 Documentation/pps/ppsldisc.c |   46 +++++++++++++
 Documentation/pps/ppstest.c  |  151 ++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 242 insertions(+), 0 deletions(-)
 create mode 100644 Documentation/pps/Makefile
 create mode 100644 Documentation/pps/ppsfind
 create mode 100644 Documentation/pps/ppsldisc.c
 create mode 100644 Documentation/pps/ppstest.c
Documentation/pps/Makefile new file +28 -0
new file mode 100644
index 0000000..9841e6b
@@ -0,0 +1,28 @@
+TARGETS = ppstest ppsldisc
+
+CFLAGS += -Wall -O2 -D_GNU_SOURCE
+CFLAGS += -I .
+CFLAGS += -ggdb
+CFLAGS += -D__N_PPS=$(shell awk '/N_PPS/ {print $$3}' ../../include/linux/tty.h)
+
+# -- 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}
Documentation/pps/ppsfind new file +17 -0
new file mode 100644
index 0000000..93c0e17
@@ -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
Documentation/pps/ppsldisc.c new file +46 -0

46 lines not on this page · raw message

Documentation/pps/ppstest.c new file +151 -0

151 lines not on this page · raw message

Signature · 1 line
1.5.4.3
#4 lore raw

PATCH 4/10 PPS: LinuxPPS clients support.

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.

---
 drivers/pps/Kconfig          |    2 +
 drivers/pps/Makefile         |    1 +
 drivers/pps/clients/Kconfig  |   18 ++++++
 drivers/pps/clients/Makefile |    9 +++
 drivers/pps/clients/ktimer.c |  124 ++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 154 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
drivers/pps/Kconfig +2 -0
index cc2eb8e..1afe4e0 100644
@@ -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
drivers/pps/Makefile +1 -0
index 19ea582..98960dd 100644
@@ -4,5 +4,6 @@
 
 pps_core-y			:= pps.o kapi.o sysfs.o
 obj-$(CONFIG_PPS)		:= pps_core.o
+obj-y				+= clients/
 
 ccflags-$(CONFIG_PPS_DEBUG) := -DDEBUG
drivers/pps/clients/Kconfig new file +18 -0
new file mode 100644
index 0000000..60b83be
@@ -0,0 +1,18 @@
+#
+# PPS clients configuration
+#
+
+if PPS
+
+comment "PPS clients support"
+
+config PPS_CLIENT_KTIMER
+	tristate "Kernel timer client (Testing client, use for debug)"
+	help
+	  If you say yes here you get support for a PPS debugging client
+	  which uses a kernel timer to generate the PPS signal.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called ktimer.ko.
+
+endif
drivers/pps/clients/Makefile new file +9 -0
new file mode 100644
index 0000000..f3c1e39
@@ -0,0 +1,9 @@
+#
+# Makefile for PPS clients.
+#
+
+obj-$(CONFIG_PPS_CLIENT_KTIMER)	+= ktimer.o
+
+ifeq ($(CONFIG_PPS_DEBUG),y)
+EXTRA_CFLAGS += -DDEBUG
+endif
drivers/pps/clients/ktimer.c new file +124 -0

124 lines not on this page · raw message

Signature · 1 line
1.5.4.3
#5 lore raw

PATCH 5/10 ldisc: new dcd_change() method for line disciplines.

---
 Documentation/tty.txt     |    4 ++++
 include/linux/tty_ldisc.h |   10 ++++++++++
 2 files changed, 14 insertions(+), 0 deletions(-)
Documentation/tty.txt +4 -0
index 8e65c44..3fc812a 100644
@@ -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
 
include/linux/tty_ldisc.h +10 -0
index 40f38d8..e74a28b 100644
@@ -99,11 +99,19 @@
  *	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 uart_port *port, 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>
 #include <linux/wait.h>
 
+struct uart_port;
+
 struct tty_ldisc_ops {
 	int	magic;
 	char	*name;
@@ -136,6 +144,8 @@ 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 uart_port *, unsigned int,
+				struct timespec *);
 
 	struct  module *owner;
 	
Signature · 1 line
1.5.4.3
#6 lore raw

PATCH 6/10 ldisc n_tty: export all N_TTY ldisc methods.

---
 drivers/char/n_tty.c |   54 +++++++++++++++++++++++++++++--------------------
 include/linux/tty.h  |   16 ++++++++++++++
 2 files changed, 48 insertions(+), 22 deletions(-)
drivers/char/n_tty.c +32 -22
index 708c2b1..10d037e 100644
@@ -26,7 +26,7 @@
  *
  * 2002/03/18   Implemented n_tty_wakeup to send SIGIO POLL_OUTs to
  *		waiting writing processes-Sapan Bhatia <[email protected]>.
- *		Also fixed a bug in BLOCKING mode where write_chan returns
+ *		Also fixed a bug in BLOCKING mode where n_tty_write returns
  *		EAGAIN
  */
 
@@ -43,10 +43,10 @@
 #include <linux/mm.h>
 #include <linux/string.h>
 #include <linux/slab.h>
-#include <linux/poll.h>
 #include <linux/bitops.h>
 #include <linux/audit.h>
 #include <linux/file.h>
+#include <linux/module.h>
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
@@ -184,7 +184,7 @@ static void reset_buffer_flags(struct tty_struct *tty)
  *	Locking: ctrl_lock
  */
 
-static void n_tty_flush_buffer(struct tty_struct *tty)
+void n_tty_flush_buffer(struct tty_struct *tty)
 {
 	unsigned long flags;
 	/* clear everything and unthrottle the driver */
@@ -200,6 +200,7 @@ static void n_tty_flush_buffer(struct tty_struct *tty)
 	}
 	spin_unlock_irqrestore(&tty->ctrl_lock, flags);
 }
+EXPORT_SYMBOL(n_tty_flush_buffer);
 
 /**
  *	n_tty_chars_in_buffer	-	report available bytes
@@ -209,7 +210,7 @@ static void n_tty_flush_buffer(struct tty_struct *tty)
  *	at this instant in time.
  */
 
-static ssize_t n_tty_chars_in_buffer(struct tty_struct *tty)
+ssize_t n_tty_chars_in_buffer(struct tty_struct *tty)
 {
 	unsigned long flags;
 	ssize_t n = 0;
@@ -225,6 +226,7 @@ static ssize_t n_tty_chars_in_buffer(struct tty_struct *tty)
 	spin_unlock_irqrestore(&tty->read_lock, flags);
 	return n;
 }
+EXPORT_SYMBOL(n_tty_chars_in_buffer);
 
 /**
  *	is_utf8_continuation	-	utf8 multibyte check
@@ -346,7 +348,7 @@ static int opost(unsigned char c, struct tty_struct *tty)
  *	the simple cases normally found and helps to generate blocks of
  *	symbols for the console driver and thus improve performance.
  *
- *	Called from write_chan under the tty layer write lock. Relies
+ *	Called from n_tty_write under the tty layer write lock. Relies
  *	on lock_kernel for the tty->column state.
  */
 
@@ -895,13 +897,14 @@ handle_newline:
  *	IO must be woken up
  */
 
-static void n_tty_write_wakeup(struct tty_struct *tty)
+void n_tty_write_wakeup(struct tty_struct *tty)
 {
 	if (tty->fasync) {
 		set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
 		kill_fasync(&tty->fasync, SIGIO, POLL_OUT);
 	}
 }
+EXPORT_SYMBOL(n_tty_write_wakeup);
 
 /**
  *	n_tty_receive_buf	-	data receive
@@ -916,7 +919,7 @@ static void n_tty_write_wakeup(struct tty_struct *tty)
  *	calls one at a time and in order (or using flush_to_ldisc)
  */
 
-static void n_tty_receive_buf(struct tty_struct *tty, const unsigned char *cp,
+void n_tty_receive_buf(struct tty_struct *tty, const unsigned char *cp,
 			      char *fp, int count)
 {
 	const unsigned char *p;
@@ -990,6 +993,7 @@ static void n_tty_receive_buf(struct tty_struct *tty, const unsigned char *cp,
 	if (tty->receive_room < TTY_THRESHOLD_THROTTLE)
 		tty_throttle(tty);
 }
+EXPORT_SYMBOL(n_tty_receive_buf);
 
 int is_ignored(int sig)
 {
@@ -1009,7 +1013,7 @@ int is_ignored(int sig)
  *	when the ldisc is closed.
  */
 
-static void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old)
+void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old)
 {
 	if (!tty)
 		return;
@@ -1076,6 +1080,7 @@ static void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old)
 	wake_up_interruptible(&tty->write_wait);
 	wake_up_interruptible(&tty->read_wait);
 }
+EXPORT_SYMBOL(n_tty_set_termios);
 
 /**
  *	n_tty_close		-	close the ldisc for this tty
@@ -1087,7 +1092,7 @@ static void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old)
  *	ldisc methods are in progress.
  */
 
-static void n_tty_close(struct tty_struct *tty)
+void n_tty_close(struct tty_struct *tty)
 {
 	n_tty_flush_buffer(tty);
 	if (tty->read_buf) {
@@ -1095,6 +1100,7 @@ static void n_tty_close(struct tty_struct *tty)
 		tty->read_buf = NULL;
 	}
 }
+EXPORT_SYMBOL(n_tty_close);
 
 /**
  *	n_tty_open		-	open an ldisc
@@ -1106,7 +1112,7 @@ static void n_tty_close(struct tty_struct *tty)
  *	until a close.
  */
 
-static int n_tty_open(struct tty_struct *tty)
+int n_tty_open(struct tty_struct *tty)
 {
 	if (!tty)
 		return -EINVAL;
@@ -1125,6 +1131,7 @@ static int n_tty_open(struct tty_struct *tty)
 	tty->closing = 0;
 	return 0;
 }
+EXPORT_SYMBOL(n_tty_open);
 
 static inline int input_available_p(struct tty_struct *tty, int amt)
 {
@@ -1143,7 +1150,7 @@ static inline int input_available_p(struct tty_struct *tty, int amt)
  *	@b: user data
  *	@nr: size of data
  *
- *	Helper function to speed up read_chan.  It is only called when
+ *	Helper function to speed up n_tty_read.  It is only called when
  *	ICANON is off; it copies characters straight from the tty queue to
  *	user space directly.  It can be profitably called twice; once to
  *	drain the space from the tail pointer to the (physical) end of the
@@ -1210,7 +1217,7 @@ static int job_control(struct tty_struct *tty, struct file *file)
 	if (file->f_op->write != redirected_tty_write &&
 	    current->signal->tty == tty) {
 		if (!tty->pgrp)
-			printk(KERN_ERR "read_chan: no tty->pgrp!\n");
+			printk(KERN_ERR "n_tty_read: no tty->pgrp!\n");
 		else if (task_pgrp(current) != tty->pgrp) {
 			if (is_ignored(SIGTTIN) ||
 			    is_current_pgrp_orphaned())
@@ -1225,7 +1232,7 @@ static int job_control(struct tty_struct *tty, struct file *file)
 
 
 /**
- *	read_chan		-	read function for tty
+ *	n_tty_read		-	read function for tty
  *	@tty: tty device
  *	@file: file object
  *	@buf: userspace buffer pointer
@@ -1239,7 +1246,7 @@ static int job_control(struct tty_struct *tty, struct file *file)
  *	This code must be sure never to sleep through a hangup.
  */
 
-static ssize_t read_chan(struct tty_struct *tty, struct file *file,
+ssize_t n_tty_read(struct tty_struct *tty, struct file *file,
 			 unsigned char __user *buf, size_t nr)
 {
 	unsigned char __user *b = buf;
@@ -1255,7 +1262,7 @@ static ssize_t read_chan(struct tty_struct *tty, struct file *file,
 do_it_again:
 
 	if (!tty->read_buf) {
-		printk(KERN_ERR "n_tty_read_chan: read_buf == NULL?!?\n");
+		printk(KERN_ERR "n_tty_read: read_buf == NULL?!?\n");
 		return -EIO;
 	}
 
@@ -1442,9 +1449,10 @@ do_it_again:
 	n_tty_set_room(tty);
 	return retval;
 }
+EXPORT_SYMBOL(n_tty_read);
 
 /**
- *	write_chan		-	write function for tty
+ *	n_tty_write		-	write function for tty
  *	@tty: tty device
  *	@file: file object
  *	@buf: userspace buffer pointer
@@ -1458,7 +1466,7 @@ do_it_again:
  *	This code must be sure never to sleep through a hangup.
  */
 
-static ssize_t write_chan(struct tty_struct *tty, struct file *file,
+ssize_t n_tty_write(struct tty_struct *tty, struct file *file,
 			  const unsigned char *buf, size_t nr)
 {
 	const unsigned char *b = buf;
@@ -1530,9 +1538,10 @@ break_out:
 	remove_wait_queue(&tty->write_wait, &wait);
 	return (b - buf) ? b - buf : retval;
 }
+EXPORT_SYMBOL(n_tty_write);
 
 /**
- *	normal_poll		-	poll method for N_TTY
+ *	n_tty_poll		-	poll method for N_TTY
  *	@tty: terminal device
  *	@file: file accessing it
  *	@wait: poll table
@@ -1545,7 +1554,7 @@ break_out:
  *	Called without the kernel lock held - fine
  */
 
-static unsigned int normal_poll(struct tty_struct *tty, struct file *file,
+unsigned int n_tty_poll(struct tty_struct *tty, struct file *file,
 							poll_table *wait)
 {
 	unsigned int mask = 0;
@@ -1572,6 +1581,7 @@ static unsigned int normal_poll(struct tty_struct *tty, struct file *file,
 		mask |= POLLOUT | POLLWRNORM;
 	return mask;
 }
+EXPORT_SYMBOL(n_tty_poll);
 
 struct tty_ldisc_ops tty_ldisc_N_TTY = {
 	.magic           = TTY_LDISC_MAGIC,
@@ -1580,11 +1590,11 @@ struct tty_ldisc_ops tty_ldisc_N_TTY = {
 	.close           = n_tty_close,
 	.flush_buffer    = n_tty_flush_buffer,
 	.chars_in_buffer = n_tty_chars_in_buffer,
-	.read            = read_chan,
-	.write           = write_chan,
+	.read            = n_tty_read,
+	.write           = n_tty_write,
 	.ioctl           = n_tty_ioctl,
 	.set_termios     = n_tty_set_termios,
-	.poll            = normal_poll,
+	.poll            = n_tty_poll,
 	.receive_buf     = n_tty_receive_buf,
 	.write_wakeup    = n_tty_write_wakeup
 };
include/linux/tty.h +16 -0
index 0cbec74..826d695 100644
@@ -13,6 +13,7 @@
 #include <linux/tty_driver.h>
 #include <linux/tty_ldisc.h>
 #include <linux/mutex.h>
+#include <linux/poll.h>
 
 #include <asm/system.h>
 
@@ -393,6 +394,21 @@ extern void tty_ldisc_begin(void);
 /* This last one is just for the tty layer internals and shouldn't be used elsewhere */
 extern void tty_ldisc_enable(struct tty_struct *tty);
 
+extern void n_tty_flush_buffer(struct tty_struct *tty);
+extern ssize_t n_tty_chars_in_buffer(struct tty_struct *tty);
+extern void n_tty_write_wakeup(struct tty_struct *tty);
+extern void n_tty_receive_buf(struct tty_struct *tty, const unsigned char *cp,
+			char *fp, int count);
+extern void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old);
+extern void n_tty_close(struct tty_struct *tty);
+extern int n_tty_open(struct tty_struct *tty);
+extern ssize_t n_tty_read(struct tty_struct *tty, struct file *file,
+			unsigned char __user *buf, size_t nr);
+extern ssize_t n_tty_write(struct tty_struct *tty, struct file *file,
+			const unsigned char *buf, size_t nr);
+extern unsigned int n_tty_poll(struct tty_struct *tty, struct file *file,
+			poll_table *wait);
+
 
 /* n_tty.c */
 extern struct tty_ldisc_ops tty_ldisc_N_TTY;
Signature · 1 line
1.5.4.3
#7 lore raw

PATCH 7/10 tty: new PPS line discipline.

Add a new line discipline for "pulse per second" devices connected to a serial port.

---
 include/linux/tty.h |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
include/linux/tty.h +2 -1
index 826d695..1857db2 100644
@@ -24,7 +24,7 @@
  */
 #define NR_UNIX98_PTY_DEFAULT	4096      /* Default maximum for Unix98 ptys */
 #define NR_UNIX98_PTY_MAX	(1 << MINORBITS) /* Absolute limit */
-#define NR_LDISCS		18
+#define NR_LDISCS		19
 
 /* line disciplines */
 #define N_TTY		0
@@ -46,6 +46,7 @@
 #define N_HCI		15	/* Bluetooth HCI UART */
 #define N_GIGASET_M101	16	/* Siemens Gigaset M101 serial DECT adapter */
 #define N_SLCAN		17	/* Serial / USB serial CAN Adaptors */
+#define N_PPS		18	/* Pulse per Second */
 
 /*
  * This character is the same as _POSIX_VDISABLE: it cannot be used as
Signature · 1 line
1.5.4.3
#8 lore raw

PATCH 8/10 PPS: serial clients support.

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

---
 drivers/pps/clients/Kconfig     |    7 ++
 drivers/pps/clients/Makefile    |    1 +
 drivers/pps/clients/pps-ldisc.c |  154 +++++++++++++++++++++++++++++++++++++++
 drivers/serial/8250.c           |   13 +++
 include/linux/serial_core.h     |   10 ++-
 5 files changed, 184 insertions(+), 1 deletions(-)
 create mode 100644 drivers/pps/clients/pps-ldisc.c
drivers/pps/clients/Kconfig +7 -0
index 60b83be..487c1c8 100644
@@ -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
drivers/pps/clients/Makefile +1 -0
index f3c1e39..9f5b988 100644
@@ -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
drivers/pps/clients/pps-ldisc.c new file +154 -0

154 lines not on this page · raw message

drivers/serial/8250.c +13 -0
index 342e12f..1759d87 100644
@@ -2247,6 +2247,18 @@ 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->info->port.tty->driver->num)
+		return;
+
+	if (port->info->port.tty->ldisc.ops->num == N_PPS)
+		serial8250_enable_ms(port);
+}
+
+static void
 serial8250_pm(struct uart_port *port, unsigned int state,
 	      unsigned int oldstate)
 {
@@ -2450,6 +2462,7 @@ 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,
include/linux/serial_core.h +9 -1
index 3b2f6c0..1aa894c 100644
@@ -488,9 +488,13 @@ static inline void
 uart_handle_dcd_change(struct uart_port *port, unsigned int status)
 {
 	struct uart_info *info = port->info;
+	struct tty_ldisc *ld = tty_ldisc_ref(info->port.tty);
+	struct timespec ts;
 
-	port->icount.dcd++;
+	if (ld->ops->dcd_change)
+		getnstimeofday(&ts);
 
+	port->icount.dcd++;
 #ifdef CONFIG_HARD_PPS
 	if ((port->flags & UPF_HARDPPS_CD) && status)
 		hardpps();
@@ -502,6 +506,10 @@ uart_handle_dcd_change(struct uart_port *port, unsigned int status)
 		else if (info->port.tty)
 			tty_hangup(info->port.tty);
 	}
+
+	if (ld->ops->dcd_change)
+		ld->ops->dcd_change(port, status, &ts);
+	tty_ldisc_deref(ld);
 }
 
 /**
Signature · 1 line
1.5.4.3
#9 lore raw

PATCH 9/10 PPS: low level IRQ timestamps recording.

Add low level IRQ timestamps recording for x86 (32 and 64 bits) platforms and enable UART clients in order to use it.

This improves PPS precision. :)

---
 arch/x86/kernel/irq_32.c        |   18 ++++++++++++++++++
 arch/x86/kernel/irq_64.c        |   22 ++++++++++++++++++++--
 drivers/pps/Kconfig             |   12 ++++++++++++
 drivers/pps/clients/pps-ldisc.c |   18 ++++++++++++++++++
 include/linux/pps.h             |    1 +
 5 files changed, 69 insertions(+), 2 deletions(-)
arch/x86/kernel/irq_32.c +18 -0
index 1cf8c1f..3013d0d 100644
@@ -15,6 +15,7 @@
 #include <linux/notifier.h>
 #include <linux/cpu.h>
 #include <linux/delay.h>
+#include <linux/pps.h>
 
 #include <asm/apic.h>
 #include <asm/uaccess.h>
@@ -25,6 +26,11 @@ EXPORT_PER_CPU_SYMBOL(irq_stat);
 DEFINE_PER_CPU(struct pt_regs *, irq_regs);
 EXPORT_PER_CPU_SYMBOL(irq_regs);
 
+#ifdef CONFIG_PPS_IRQ_EVENTS
+struct pps_ktime pps_irq_ts[NR_IRQS];
+EXPORT_SYMBOL(pps_irq_ts);
+#endif
+
 /*
  * 'what should we do if we get a hw irq event on an illegal vector'.
  * each architecture has to answer this themselves.
@@ -225,6 +231,12 @@ unsigned int do_IRQ(struct pt_regs *regs)
 	/* high bit used in ret_from_ code */
 	int overflow, irq = ~regs->orig_ax;
 	struct irq_desc *desc = irq_desc + irq;
+#ifdef CONFIG_PPS_IRQ_EVENTS
+	struct timespec __ts;
+
+	/* Get IRQ timestamps as soon as possible for the PPS layer */
+	getnstimeofday(&__ts);
+#endif
 
 	if (unlikely((unsigned)irq >= NR_IRQS)) {
 		printk(KERN_EMERG "%s: cannot handle IRQ %d\n",
@@ -232,6 +244,12 @@ unsigned int do_IRQ(struct pt_regs *regs)
 		BUG();
 	}
 
+#ifdef CONFIG_PPS_IRQ_EVENTS
+	/* Then, after sanity check, store the IRQ timestamp */
+	pps_irq_ts[irq].sec = __ts.tv_sec;
+	pps_irq_ts[irq].nsec = __ts.tv_nsec;
+#endif
+
 	old_regs = set_irq_regs(regs);
 	irq_enter();
 
arch/x86/kernel/irq_64.c +20 -2
index 1f78b23..1ea77a7 100644
@@ -17,9 +17,15 @@
 #include <asm/io_apic.h>
 #include <asm/idle.h>
 #include <asm/smp.h>
+#include <linux/pps.h>
 
 atomic_t irq_err_count;
 
+#ifdef CONFIG_PPS_IRQ_EVENTS
+struct pps_ktime pps_irq_ts[NR_IRQS];
+EXPORT_SYMBOL(pps_irq_ts);
+#endif
+
 /*
  * 'what should we do if we get a hw irq event on an illegal vector'.
  * each architecture has to answer this themselves.
@@ -188,6 +194,12 @@ u64 arch_irq_stat(void)
 asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
 {
 	struct pt_regs *old_regs = set_irq_regs(regs);
+#ifdef CONFIG_PPS_IRQ_EVENTS
+	struct timespec __ts;
+
+	/* Get IRQ timestamps as soon as possible for the PPS layer */
+	getnstimeofday(&__ts);
+#endif
 
 	/* high bit used in ret_from_ code  */
 	unsigned vector = ~regs->orig_ax;
@@ -201,9 +213,15 @@ asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
 	stack_overflow_check(regs);
 #endif
 
-	if (likely(irq < NR_IRQS))
+	if (likely(irq < NR_IRQS)) {
+#ifdef CONFIG_PPS_IRQ_EVENTS
+		/* Then, after sanity check, store the IRQ timestamp */
+		pps_irq_ts[irq].sec = __ts.tv_sec;
+		pps_irq_ts[irq].nsec = __ts.tv_nsec;
+#endif
+
 		generic_handle_irq(irq);
-	else {
+	} else {
 		if (!disable_apic)
 			ack_APIC_irq();
 
drivers/pps/Kconfig +12 -0
index 1afe4e0..7c0b094 100644
@@ -22,6 +22,18 @@ config PPS
 	  To compile this driver as a module, choose M here: the module
 	  will be called pps_core.ko.
 
+config PPS_IRQ_EVENTS
+	bool "Use low level IRQ timestamps"
+	depends on PPS && (X86_32 || X86_64)
+	default yes
+	help
+	  Say Y here if you wish using low level IRQ timestamps to register
+	  PPS events.
+
+	  This should improve PPS resolution but it delays echo functions
+	  call. Note also that this function is not implemented on all
+	  platforms and PPS clients!
+
 config PPS_DEBUG
 	bool "PPS debugging messages"
 	depends on PPS
drivers/pps/clients/pps-ldisc.c +18 -0
index f1efa26..20d2c2e 100644
@@ -26,6 +26,22 @@
 
 #define PPS_TTY_MAGIC		0x0001
 
+#ifdef CONFIG_PPS_IRQ_EVENTS
+
+static void pps_tty_dcd_change(struct uart_port *port, unsigned int status,
+				struct timespec *ts)
+{
+	int id = (int) port->info->port.tty->disc_data;
+
+	pps_event(id, &pps_irq_ts[port->irq],
+			status ? PPS_CAPTUREASSERT : PPS_CAPTURECLEAR, port);
+
+	pr_debug("[IRQev] PPS %s at %lu on source #%d\n",
+			status ? "assert" : "clear", jiffies, id);
+}
+
+#else
+
 static void pps_tty_dcd_change(struct uart_port *port, unsigned int status,
 				struct timespec *ts)
 {
@@ -53,6 +69,8 @@ static void pps_tty_dcd_change(struct uart_port *port, unsigned int status,
 			status ? "assert" : "clear", jiffies, id);
 }
 
+#endif /* CONFIG_PPS_IRQ_EVENTS */
+
 static int pps_tty_open(struct tty_struct *tty)
 {
 	struct pps_source_info info;
include/linux/pps.h +1 -0
index 53e171c..bb9a2e6 100644
@@ -181,6 +181,7 @@ struct pps_device {
 
 extern spinlock_t pps_idr_lock;
 extern struct idr pps_idr;
+extern struct pps_ktime pps_irq_ts[];
 
 extern struct device_attribute pps_attrs[];
 
Signature · 1 line
1.5.4.3
#10 lore raw

PATCH 10/10 PPS: parallel port clients support.

Adds support for the PPS sources connected with the interrupt pin of a parallel port.

---
 drivers/char/lp.c           |   61 +++++++++++++++++++++++++++++++++++++++++++
 drivers/pps/clients/Kconfig |   10 +++++++
 include/linux/parport.h     |   22 +++++++++++++++
 3 files changed, 93 insertions(+), 0 deletions(-)
drivers/char/lp.c +61 -0
index 3f2719b..84e0bb0 100644
@@ -760,6 +760,27 @@ static struct console lpcons = {
 
 #endif /* console on line printer */
 
+/* Support for PPS signal on the line printer */
+
+#ifdef CONFIG_PPS_CLIENT_LP
+
+static void lp_pps_echo(int source, int event, void *data)
+{
+	struct parport *port = data;
+	unsigned char status = parport_read_status(port);
+
+	/* echo event via SEL bit */
+	parport_write_control(port,
+		parport_read_control(port) | PARPORT_CONTROL_SELECT);
+
+	/* signal no event */
+	if ((status & PARPORT_STATUS_ACK) != 0)
+		parport_write_control(port,
+			parport_read_control(port) & ~PARPORT_CONTROL_SELECT);
+}
+
+#endif
+
 /* --- initialisation code ------------------------------------- */
 
 static int parport_nr[LP_NO] = { [0 ... LP_NO-1] = LP_PARPORT_UNSPEC };
@@ -831,6 +852,38 @@ static int lp_register(int nr, struct parport *port)
 	}
 #endif
 
+#ifdef CONFIG_PPS_CLIENT_LP
+	port->pps_info.owner = THIS_MODULE;
+	port->pps_info.dev = port->dev;
+	snprintf(port->pps_info.path, PPS_MAX_NAME_LEN, "/dev/lp%d", nr);
+
+	/* No PPS support if lp port has no IRQ line */
+	if (port->irq != PARPORT_IRQ_NONE) {
+		strncpy(port->pps_info.name, port->name, PPS_MAX_NAME_LEN);
+
+		port->pps_info.mode = PPS_CAPTUREASSERT | PPS_OFFSETASSERT | \
+				PPS_ECHOASSERT | \
+				PPS_CANWAIT | PPS_TSFMT_TSPEC;
+
+		port->pps_info.echo = lp_pps_echo;
+
+		port->pps_source = pps_register_source(&(port->pps_info),
+				PPS_CAPTUREASSERT | PPS_OFFSETASSERT);
+		if (port->pps_source < 0)
+			dev_err(port->dev,
+					"cannot register PPS source \"%s\"\n",
+					port->pps_info.path);
+		else
+			dev_info(port->dev, "PPS source #%d \"%s\" added\n",
+					port->pps_source, port->pps_info.path);
+	} else {
+		port->pps_source = -1;
+		dev_err(port->dev, "PPS support disabled because port \"%s\" "
+					"is in polling mode\n",
+					port->pps_info.path);
+	}
+#endif
+
 	return 0;
 }
 
@@ -873,6 +926,14 @@ static void lp_detach (struct parport *port)
 		console_registered = NULL;
 	}
 #endif /* CONFIG_LP_CONSOLE */
+
+#ifdef CONFIG_PPS_CLIENT_LP
+	if (port->pps_source >= 0) {
+		pps_unregister_source(port->pps_source);
+		dev_dbg(port->dev, "PPS source #%d \"%s\" removed\n",
+			port->pps_source, port->pps_info.path);
+	}
+#endif
 }
 
 static struct parport_driver lp_driver = {
drivers/pps/clients/Kconfig +10 -0
index 487c1c8..b4054cd 100644
@@ -22,4 +22,14 @@ config PPS_CLIENT_LDISC
 	  If you say yes here you get support for a PPS source connected
 	  with the CD (Carrier Detect) pin of your serial port.
 
+comment "Parallel printer support (forced off)"
+	depends on !( PRINTER != n && !(PPS = m && PRINTER = y))
+
+config PPS_CLIENT_LP
+	bool "Parallel printer support"
+	depends on PRINTER != n && !(PPS = m && PRINTER = y)
+	help
+	  If you say yes here you get support for a PPS source connected
+	  with the interrupt pin of your parallel port.
+
 endif
include/linux/parport.h +22 -0
index 6a0d7cd..f2cfd6e 100644
@@ -100,6 +100,7 @@ typedef enum {
 #include <linux/proc_fs.h>
 #include <linux/spinlock.h>
 #include <linux/wait.h>
+#include <linux/pps.h>
 #include <linux/irqreturn.h>
 #include <linux/semaphore.h>
 #include <asm/system.h>
@@ -328,6 +329,11 @@ struct parport {
 
 	struct list_head full_list;
 	struct parport *slaves[3];
+
+#ifdef CONFIG_PPS_CLIENT_LP
+	struct pps_source_info pps_info;
+	int pps_source;		/* PPS source ID */
+#endif
 };
 
 #define DEFAULT_SPIN_TIME 500 /* us */
@@ -516,6 +522,22 @@ extern int parport_daisy_select (struct parport *port, int daisy, int mode);
 /* Lowlevel drivers _can_ call this support function to handle irqs.  */
 static inline void parport_generic_irq(struct parport *port)
 {
+#ifdef CONFIG_PPS_CLIENT_LP
+	struct timespec __ts;
+	struct pps_ktime ts;
+
+	/* First of all we get the time stamp... */
+	getnstimeofday(&__ts);
+
+	/* ... and translate it to PPS time data struct */
+	ts.sec = __ts.tv_sec;
+	ts.nsec = __ts.tv_nsec;
+
+	pps_event(port->pps_source, &ts, PPS_CAPTUREASSERT, port);
+	dev_dbg(port->dev, "PPS assert at %lu on source #%d\n",
+		jiffies, port->pps_source);
+#endif
+
 	parport_ieee1284_interrupt (port);
 	read_lock(&port->cad_lock);
 	if (port->cad && port->cad->irq_func)
Signature · 1 line
1.5.4.3
#11 lore raw

Re: [PATCH 09/10] PPS: low level IRQ timestamps recording.

In reply to Rodolfo Giometti
Quoted 5 lines · Rodolfo Giometti

Add low level IRQ timestamps recording for x86 (32 and 64 bits) platforms and enable UART clients in order to use it.

This improves PPS precision. :)

Quoted 8 lines
---
 arch/x86/kernel/irq_32.c        |   18 ++++++++++++++++++
 arch/x86/kernel/irq_64.c        |   22 ++++++++++++++++++++--
 drivers/pps/Kconfig             |   12 ++++++++++++
 drivers/pps/clients/pps-ldisc.c |   18 ++++++++++++++++++
 include/linux/pps.h             |    1 +
 5 files changed, 69 insertions(+), 2 deletions(-)

hm, NACK on this line of approach. If then this instrumentation should be done at the genirq level (kernel/irq/*.c). Also, the #ifdefs are ugly
- please hide them in include files.

Ingo

maintainer
#12 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

Quoted 4 lines · Alan Cox

Can we get the basics right first - drop the IRQ_EVENTS and HARDPPS hackery out of the patchset and the core code then looks quite clean. After that is sorted/merged we can come back to working out the best way to improve the IRQ and HARDPPS hacks.

Quoted 3 lines · Ingo Molnar

hm, NACK on this line of approach. If then this instrumentation should be done at the genirq level (kernel/irq/*.c). Also, the #ifdefs are ugly
- please hide them in include files.

Ok, I'll propose a new patch set without the IRQ_EVENTS part (or with this support as last patch to be easily discarted) to get kernel inclusion, but we had to solve this issue soon since without this trick the PPS precision is quite low.

Thanks,

Rodolfo

Signature · 4 lines
GNU/Linux Solutions                  e-mail: [email protected]
Linux Device Driver                          [email protected]
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti
#13 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

Quoted 9 lines · Rodolfo Giometti
Quoted 4 lines · Alan Cox

Can we get the basics right first - drop the IRQ_EVENTS and HARDPPS hackery out of the patchset and the core code then looks quite clean. After that is sorted/merged we can come back to working out the best way to improve the IRQ and HARDPPS hacks.

On Wed, Aug 20, 2008 at 12:31:15PM +0200, Ingo Molnar wrote:

Quoted 8 lines
Quoted 3 lines

hm, NACK on this line of approach. If then this instrumentation should be done at the genirq level (kernel/irq/*.c). Also, the #ifdefs are ugly
- please hide them in include files.

Ok, I'll propose a new patch set without the IRQ_EVENTS part (or with this support as last patch to be easily discarted) to get kernel inclusion, but we had to solve this issue soon since without this trick the PPS precision is quite low.

well, i didnt suggest you kill that aspect of the patchset - i suggested you change it to hook into kernel/irq/*.c instead of arch/x86/kernel/irq*.c.

That way all other architectures will benefit from it, not just x86. Doing an x86-only thing for such features is completely unacceptable.

Ingo

maintainer
#14 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

Quoted 1 line · Ingo Molnar

well, i didnt suggest you kill that aspect of the patchset - i suggested

I see, but I agree with Alan that is better to get a first working inclusion and then working on improments. 8)

Quoted 5 lines

you change it to hook into kernel/irq/*.c instead of arch/x86/kernel/irq*.c.

That way all other architectures will benefit from it, not just x86. Doing an x86-only thing for such features is completely unacceptable.

I see but doing like this I can record IRQs timestamps sooner than in kernel/irq/*.c, is that right? PPS precision is as better as sooner we record the timestamps.

Ciao,

Rodolfo

Signature · 4 lines
GNU/Linux Solutions                  e-mail: [email protected]
Linux Device Driver                          [email protected]
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti
#15 lore raw

Re: [PATCH 08/10] PPS: serial clients support.

In reply to Rodolfo Giometti

drivers/pps/clients/pps-ldisc.c line 32

Quoted 4 lines
drivers/pps/clients/pps-ldisc.c new file +4 -0
+static void pps_tty_dcd_change(struct uart_port *port, unsigned int status,
+				struct timespec *ts)
+{
+	int id = (int) port->info->port.tty->disc_data;

See you really do want to pass struct tty * upwards anyway 8)

Rest of this looks good. Maybe pps ldisc belongs in the n_tty file in the end but again that can be twiddled with after the merge.

Acked-by on v1 7/10
#16 lore raw

Re: [PATCH 07/10] tty: new PPS line discipline.

In reply to Rodolfo Giometti
Quoted 4 lines · On Tue, 19 Aug 2008 23:50:09 +0200 Rodolfo Giometti <[email protected]> wrote:

Add a new line discipline for "pulse per second" devices connected to a serial port.

Ack this and I'll push it upstream anyway so there is a stable number even if this doesn't all get upstream for a while.

#17 lore raw

Re: [PATCH 06/10] ldisc n_tty: export all N_TTY ldisc methods.

In reply to Rodolfo Giometti
Quoted 5 lines · On Tue, 19 Aug 2008 23:50:08 +0200 Rodolfo Giometti <[email protected]> wrote:
---
 drivers/char/n_tty.c |   54 +++++++++++++++++++++++++++++--------------------
 include/linux/tty.h  |   16 ++++++++++++++
 2 files changed, 48 insertions(+), 22 deletions(-)

Renames are really not part of the same change set but very sensible so no objection. EXPORT_SYMBOL_GPL please - pps is (with permission) sticking its nose into the innards of n_tty. Those are not good general interfaces however.

With the small changes noted I'm happy for the tty stuff to go upstream.

#18 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

In reply to Rodolfo Giometti
Quoted 6 lines

+ * void (*dcd_change)(struct uart_port *port, 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.
*/

uart_port is a specific property of a few random types of interface, pass a tty struct up. Otherwise looks fine. Might one day expand it to a more generic change type and pass info on what changed but thats a future mystery.

maintainer
#19 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

Quoted 11 lines · Alan Cox
Quoted 6 lines

+ * void (*dcd_change)(struct uart_port *port, 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.
*/

uart_port is a specific property of a few random types of interface, pass a tty struct up. Otherwise looks fine. Might one day expand it to a more generic change type and pass info on what changed but thats a future mystery.

How can I get the struct uart_port pointer starting from struct tty_struct one? I need it in order to know which is the IRQ line of a single port.

Thanks,

Rodolfo

Signature · 4 lines
GNU/Linux Solutions                  e-mail: [email protected]
Linux Device Driver                          [email protected]
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti
#20 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

Quoted 8 lines · On Wed, 20 Aug 2008 00:49:31 +0200 Rodolfo Giometti <[email protected]>…
Quoted 7 lines · Alan Cox
Quoted 6 lines

+ * void (*dcd_change)(struct uart_port *port, 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.
*/

Quoted 7 lines
Quoted 4 lines

uart_port is a specific property of a few random types of interface, pass a tty struct up. Otherwise looks fine. Might one day expand it to a more generic change type and pass info on what changed but thats a future mystery.

How can I get the struct uart_port pointer starting from struct tty_struct one?

A tty may not even have a uart structure. Why do you need to know the interrupt (if such a concept even exists for the given tty) ?

Alan

maintainer
#21 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

Quoted 16 lines · Alan Cox
Quoted 13 lines · On Wed, 20 Aug 2008 00:49:31 +0200 Rodolfo Giometti <[email protected]>…
Quoted 11 lines · Alan Cox
Quoted 6 lines

+ * void (*dcd_change)(struct uart_port *port, 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.
*/

uart_port is a specific property of a few random types of interface, pass a tty struct up. Otherwise looks fine. Might one day expand it to a more generic change type and pass info on what changed but thats a future mystery.

Quoted 5 lines
Quoted 2 lines

How can I get the struct uart_port pointer starting from struct tty_struct one?

A tty may not even have a uart structure. Why do you need to know the interrupt (if such a concept even exists for the given tty) ?

If you look at patch 9 (0009-PPS-low-level-IRQ-timestamps-recording.patch) you can see that we add a (dirty?) trick to improve PPS precision a lot: we record all IRQ timestamps into static array and then we return those values instead of ones passed by dcd_change()...

Maybe I can solve the problem defining dcd_change() as follow:

void (*dcd_change)(struct tty_struct *tty, unsigned int status,
void *data)

and then passing the timestamp or struct uart_port pointer?

Thanks,

Rodolfo

Signature · 4 lines
GNU/Linux Solutions                  e-mail: [email protected]
Linux Device Driver                          [email protected]
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti
maintainer
#22 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

Quoted 6 lines · Rodolfo Giometti

Maybe I can solve the problem defining dcd_change() as follow:

void (*dcd_change)(struct tty_struct *tty, unsigned int status,
void *data)

and then passing the timestamp or struct uart_port pointer?

After sleeping this night :) I think is better defing the uart_handle_dcd_change() into serial_core.h as follow:

uart_handle_dcd_change(struct uart_port *port, unsigned int status)

{
        struct uart_info *info = port->info;
        struct tty_ldisc *ld = tty_ldisc_ref(info->port.tty);
        struct timespec ts;

        if (ld->ops->dcd_change)
#ifdef CONFIG_PPS_IRQ_EVENTS
                ts = pps_irq_ts[port->irq];
#else
                getnstimeofday(&ts);

#endif

        port->icount.dcd++;
#ifdef CONFIG_HARD_PPS
        if ((port->flags & UPF_HARDPPS_CD) && status)
                hardpps();

#endif

        if (info->flags & UIF_CHECK_CD) {
                if (status)
                        wake_up_interruptible(&info->port.open_wait);
                else if (info->port.tty)
                        tty_hangup(info->port.tty);
        }

        if (ld->ops->dcd_change)
                ld->ops->dcd_change(info->port.tty, status, &ts);
        tty_ldisc_deref(ld);
}

This completely hides the (dirty) trick to the PPS clients and allow using struct tty_struct as first parameter.

Is that acceptable?

Rodolfo

Signature · 4 lines
GNU/Linux Solutions                  e-mail: [email protected]
Linux Device Driver                          [email protected]
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti
#23 lore raw

Re: [PATCH 05/10] ldisc: new dcd_change() method for line disciplines.

O> #ifdef CONFIG_PPS_IRQ_EVENTS

Quoted 2 lines
                ts = pps_irq_ts[port->irq];

#else

That bit won't work very shortly - IRQ numbers are going sparse not a single simple NR_IRQ array.

Quoted 2 lines

This completely hides the (dirty) trick to the PPS clients and allow using struct tty_struct as first parameter.

Can we get the basics right first - drop the IRQ_EVENTS and HARDPPS hackery out of the patchset and the core code then looks quite clean. After that is sorted/merged we can come back to working out the best way to improve the IRQ and HARDPPS hacks.

Alan

#24 lore raw

Re: [PATCH 01/10] LinuxPPS core support.

In reply to Rodolfo Giometti

Documentation/ioctl-number.txt Code Seq# Include File Comments · line 101

Quoted 3 lines · Rodolfo Giometti

'P' all linux/soundcard.h
+'P' 00-04 linux/pps.h LinuxPPS

Documentation/ioctl-number.txt +1 -0
+					<mailto:[email protected]>

This represents a reservation conflict even if it doesn't represent an actual conflict. If it is a reservation conflict only, please adjust the range reserved for soundcards (and verify with the sound maintainers that it's okay to do so); if it's an actual conflict please reassign.

-hpa

maintainer
#25 lore raw

Re: [PATCH 01/10] LinuxPPS core support.

Quoted 9 lines · H. Peter Anvin

Documentation/ioctl-number.txt Code Seq# Include File Comments · line 101

Quoted 3 lines · Rodolfo Giometti

'P' all linux/soundcard.h
+'P' 00-04 linux/pps.h LinuxPPS

Documentation/ioctl-number.txt +1 -0
+					<mailto:[email protected]>

This represents a reservation conflict even if it doesn't represent an actual conflict. If it is a reservation conflict only, please adjust the range reserved for soundcards (and verify with the sound maintainers that it's okay to do so); if it's an actual conflict please reassign.

I think is quicker that I reassing the codes. I'll use 'p' a0-a4 which are free.

Ciao,

Rodolfo

Signature · 4 lines
GNU/Linux Solutions                  e-mail: [email protected]
Linux Device Driver                          [email protected]
Embedded Systems                     phone:  +39 349 2432127
UNIX programming                     skype:  rodolfo.giometti