Linux-OMAP Archive on lore.kernel.org
 help / color / Atom feed
* [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
@ 2020-04-30 17:46 Tony Lindgren
  2020-04-30 17:46 ` [PATCH 1/6] tty: n_gsm: Add support for serdev drivers Tony Lindgren
                   ` (9 more replies)
  0 siblings, 10 replies; 29+ messages in thread
From: Tony Lindgren @ 2020-04-30 17:46 UTC (permalink / raw)
  To: Greg Kroah-Hartman, Johan Hovold, Rob Herring
  Cc: Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap

Hi all,

Here's v6 set of these patches fixed up for Johan's earlier comments.

Getting rid of the custom chardev interface, and making things more
generic also simplified things quite a bit. Thanks a lot Johan :)

This series does the following:

1. Adds functions to n_gsm.c for serdev-ngsm.c driver to use

2. Adds a generic serdev-ngsm.c driver that brings up the TS 27.010
   TTY ports configured in devicetree with help of n_gsm.c

3. Allows the use of standard Linux device drivers for dedicated
   TS 27.010 channels for devices like GNSS and ALSA found on some
   modems for example

4. Adds a serdev-ngsm consumer driver for the GNSS device found on
   the Motorola Mapphone MDM6600 modem on devices like droid4

I've placed the serdev-ngsm.c under drivers/tty/serdev as it still
seems to make most sense with no better places available. It's no
longer an MFD driver as it really does not need to care what channel
specific consumer drivers might be configured. So it now just uses
of_platform_populate() to probe whatever child nodes it might find.

I'm not attached having the driver in drivers/tty/serdev. I just
don't have any better locations in mind. So jsing Johan's earlier
i2c example, the drivers/tty/serdev/serdev-ngsm.c driver is now a
generic protocol and bus driver, so it's getting closer to the
maybe the drivers/i2c/busses analogy :) Please do suggest better
locations other than MFD and misc if you have better ideas.

Now without the chardev support, the /dev/gsmtty* using apps need
to use "U1234AT+CFUN?" format for the packets. The advantage is
less kernel code, and we keep the existing /dev/gsmtty* interface.

If we still really need the custom chardev support, that can now
be added as needed with the channel specific consumer driver(s).

My guess is that at least with the pending ofono patches, we just
want to use the raw interface for /dev/gsmtty* interface and stop
pretending we have a modem that is AT compatible.

Regards,

Tony


Changes since v5:
- Based on comments from Johan, moved back to using the existing
  TS 27.010 TTYs created by n_gsm.c instaed of adding custom chardev
  support to deal with the Motorola custom protocol

- Based on comments from Johan, made the serdev-ngsm driver generic
  with just minimal quirk handling for the Motorola modem

- Dropped the Motorola custom protocol on top of TS 27.010 handling
  from serdev-ngsm.c as this can now be easily handled by the channel
  specific drivers as needed

- Added few more helpers to n_gsm.c for serdev-ngsm.c to use

- Added the channel specific GNSS driver for the Motorola modem

Changes since v4:
- Use drivers/tty/serdev/protocol directory for the driver instead of
  drivers/mfd as discussed on the lists for v3 set of patches
- Fix remove to call kfree only after removing device from the list

Changes since v3:
- Update list of folks in Cc, looks like I sent v3 only to Lee and lkml
- Init privdata before motmdm_register_dlci calls gsm_serdev_register_dlci
- Update binding based on Rob's comments for license and "allOf"

Changes since v2:
- Drop useless send_command indirection, use static motmdm_send_command

Changes since v1:

- Simplified usage and got rid of few pointless inline functions
- Added consumer MFD driver, devicetree binding, and dts changes


Tony Lindgren (6):
  tty: n_gsm: Add support for serdev drivers
  dt-bindings: serdev: ngsm: Add binding for serdev-ngsm
  serdev: ngsm: Add generic serdev-ngsm driver
  dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS
  gnss: motmdm: Add support for Motorola Mapphone MDM6600 modem
  ARM: dts: omap4-droid4: Configure modem for serdev-ngsm

 .../devicetree/bindings/gnss/motmdm.yaml      |  29 ++
 .../bindings/serdev/serdev-ngsm.yaml          |  64 +++
 .../devicetree/bindings/vendor-prefixes.yaml  |   2 +
 .../boot/dts/motorola-mapphone-common.dtsi    |  14 +
 drivers/gnss/Kconfig                          |   8 +
 drivers/gnss/Makefile                         |   3 +
 drivers/gnss/motmdm.c                         | 419 ++++++++++++++++
 drivers/tty/n_gsm.c                           | 428 +++++++++++++++++
 drivers/tty/serdev/Kconfig                    |  10 +
 drivers/tty/serdev/Makefile                   |   1 +
 drivers/tty/serdev/serdev-ngsm.c              | 449 ++++++++++++++++++
 include/linux/serdev-gsm.h                    | 163 +++++++
 12 files changed, 1590 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/gnss/motmdm.yaml
 create mode 100644 Documentation/devicetree/bindings/serdev/serdev-ngsm.yaml
 create mode 100644 drivers/gnss/motmdm.c
 create mode 100644 drivers/tty/serdev/serdev-ngsm.c
 create mode 100644 include/linux/serdev-gsm.h

-- 
2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

* [PATCH 1/6] tty: n_gsm: Add support for serdev drivers
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
@ 2020-04-30 17:46 ` Tony Lindgren
  2020-05-01 20:31   ` Pavel Machek
  2020-04-30 17:46 ` [PATCH 2/6] dt-bindings: serdev: ngsm: Add binding for serdev-ngsm Tony Lindgren
                   ` (8 subsequent siblings)
  9 siblings, 1 reply; 29+ messages in thread
From: Tony Lindgren @ 2020-04-30 17:46 UTC (permalink / raw)
  To: Greg Kroah-Hartman, Johan Hovold, Rob Herring
  Cc: Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap

We can make use of serdev drivers to do simple device drivers for
TS 27.010 chanels, and we can handle vendor specific protocols on top
of TS 27.010 with serdev drivers.

So far this has been tested with Motorola droid4 where there is a custom
packet numbering protocol on top of TS 27.010 for the MDM6600 modem.

I initially though about adding the serdev support into a separate file,
but that will take some refactoring of n_gsm.c. And I'd like to have
things working first. Then later on we might want to consider splitting
n_gsm.c into three pieces for core, tty and serdev parts. And then maybe
the serdev related parts can be just moved to live under something like
drivers/tty/serdev/protocol/ngsm.c.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 drivers/tty/n_gsm.c        | 428 +++++++++++++++++++++++++++++++++++++
 include/linux/serdev-gsm.h | 152 +++++++++++++
 2 files changed, 580 insertions(+)
 create mode 100644 include/linux/serdev-gsm.h

diff --git a/drivers/tty/n_gsm.c b/drivers/tty/n_gsm.c
--- a/drivers/tty/n_gsm.c
+++ b/drivers/tty/n_gsm.c
@@ -39,6 +39,7 @@
 #include <linux/file.h>
 #include <linux/uaccess.h>
 #include <linux/module.h>
+#include <linux/serdev.h>
 #include <linux/timer.h>
 #include <linux/tty_flip.h>
 #include <linux/tty_driver.h>
@@ -50,6 +51,7 @@
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/gsmmux.h>
+#include <linux/serdev-gsm.h>
 
 static int debug;
 module_param(debug, int, 0600);
@@ -150,6 +152,7 @@ struct gsm_dlci {
 	/* Data handling callback */
 	void (*data)(struct gsm_dlci *dlci, const u8 *data, int len);
 	void (*prev_data)(struct gsm_dlci *dlci, const u8 *data, int len);
+	struct gsm_serdev_dlci *ops; /* serdev dlci ops, if used */
 	struct net_device *net; /* network interface, if created */
 };
 
@@ -198,6 +201,7 @@ enum gsm_mux_state {
  */
 
 struct gsm_mux {
+	struct gsm_serdev *gsd;		/* Serial device bus data */
 	struct tty_struct *tty;		/* The tty our ldisc is bound to */
 	spinlock_t lock;
 	struct mutex mutex;
@@ -2346,6 +2350,430 @@ static int gsm_config(struct gsm_mux *gsm, struct gsm_config *c)
 	return 0;
 }
 
+#ifdef CONFIG_SERIAL_DEV_BUS
+
+/**
+ * gsm_serdev_get_config - read ts 27.010 config
+ * @gsd:	serdev-gsm instance
+ * @c:		ts 27.010 config data
+ *
+ * See gsm_copy_config_values() for more information.
+ */
+int gsm_serdev_get_config(struct gsm_serdev *gsd, struct gsm_config *c)
+{
+	struct gsm_mux *gsm;
+
+	if (!gsd || !gsd->gsm)
+		return -ENODEV;
+
+	gsm = gsd->gsm;
+
+	if (!c)
+		return -EINVAL;
+
+	gsm_copy_config_values(gsm, c);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_get_config);
+
+/**
+ * gsm_serdev_set_config - set ts 27.010 config
+ * @gsd:	serdev-gsm instance
+ * @c:		ts 27.010 config data
+ *
+ * See gsm_config() for more information.
+ */
+int gsm_serdev_set_config(struct gsm_serdev *gsd, struct gsm_config *c)
+{
+	struct gsm_mux *gsm;
+
+	if (!gsd || !gsd->serdev || !gsd->gsm)
+		return -ENODEV;
+
+	gsm = gsd->gsm;
+
+	if (!c)
+		return -EINVAL;
+
+	return gsm_config(gsm, c);
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_set_config);
+
+static struct gsm_dlci *gsd_dlci_get(struct gsm_serdev *gsd, int line,
+				     bool allocate)
+{
+	struct gsm_mux *gsm;
+	struct gsm_dlci *dlci;
+
+	if (!gsd || !gsd->gsm)
+		return ERR_PTR(-ENODEV);
+
+	gsm = gsd->gsm;
+
+	if (line < 1 || line >= 63)
+		return ERR_PTR(-EINVAL);
+
+	mutex_lock(&gsm->mutex);
+
+	if (gsm->dlci[line]) {
+		dlci = gsm->dlci[line];
+		goto unlock;
+	} else if (!allocate) {
+		dlci = ERR_PTR(-ENODEV);
+		goto unlock;
+	}
+
+	dlci = gsm_dlci_alloc(gsm, line);
+	if (!dlci) {
+		gsm = ERR_PTR(-ENOMEM);
+		goto unlock;
+	}
+
+	gsm->dlci[line] = dlci;
+
+unlock:
+	mutex_unlock(&gsm->mutex);
+
+	return dlci;
+}
+
+static int gsd_dlci_receive_buf(struct gsm_dlci *dlci,
+				const unsigned char *buf,
+				size_t len)
+{
+	struct tty_port *port;
+
+	port = &dlci->port;
+	tty_insert_flip_string(port, buf, len);
+	tty_flip_buffer_push(port);
+
+	return len;
+}
+
+static void gsd_dlci_data(struct gsm_dlci *dlci, const u8 *buf, int len)
+{
+	struct gsm_mux *gsm = dlci->gsm;
+	struct gsm_serdev *gsd = gsm->gsd;
+
+	if (!gsd || !dlci->ops)
+		return;
+
+	switch (dlci->adaption) {
+	case 0:
+	case 1:
+		if (dlci->ops->receive_buf)
+			dlci->ops->receive_buf(dlci->ops, buf, len);
+		else
+			gsd_dlci_receive_buf(dlci, buf, len);
+		break;
+	default:
+		pr_warn("dlci%i adaption %i not yet implemented\n",
+			dlci->addr, dlci->adaption);
+		break;
+	}
+}
+
+/**
+ * gsm_serdev_write - write data to a ts 27.010 channel
+ * @gsd:	serdev-gsm instance
+ * @ops:	channel ops
+ * @buf:	write buffer
+ * @len:	buffer length
+ */
+int gsm_serdev_write(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops,
+		     const u8 *buf, int len)
+{
+	struct gsm_mux *gsm;
+	struct gsm_dlci *dlci;
+	struct gsm_msg *msg;
+	int h, size, total_size = 0;
+	u8 *dp;
+
+	if (!gsd || !gsd->gsm)
+		return -ENODEV;
+
+	gsm = gsd->gsm;
+
+	dlci = gsd_dlci_get(gsd, ops->line, false);
+	if (IS_ERR(dlci))
+		return PTR_ERR(dlci);
+
+	h = dlci->adaption - 1;
+
+	if (len > gsm->mtu)
+		len = gsm->mtu;
+
+	size = len + h;
+
+	msg = gsm_data_alloc(gsm, dlci->addr, size, gsm->ftype);
+	if (!msg)
+		return -ENOMEM;
+
+	dp = msg->data;
+	switch (dlci->adaption) {
+	case 1:
+		break;
+	case 2:
+		*dp++ = gsm_encode_modem(dlci);
+		break;
+	}
+	memcpy(dp, buf, len);
+	gsm_data_queue(dlci, msg);
+	total_size += size;
+
+	return total_size;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_write);
+
+/**
+ * gsm_serdev_data_kick - indicate more data can be transmitted
+ * @gsd:	serdev-gsm instance
+ *
+ * See gsm_data_kick() for more information.
+ */
+void gsm_serdev_data_kick(struct gsm_serdev *gsd)
+{
+	struct gsm_mux *gsm;
+	unsigned long flags;
+
+	if (!gsd || !gsd->gsm)
+		return;
+
+	gsm = gsd->gsm;
+
+	spin_lock_irqsave(&gsm->tx_lock, flags);
+	gsm_data_kick(gsm);
+	spin_unlock_irqrestore(&gsm->tx_lock, flags);
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_data_kick);
+
+/**
+ * gsm_serdev_register_dlci - register a ts 27.010 channel
+ * @gsd:	serdev-gsm instance
+ * @ops:	channel ops
+ */
+int gsm_serdev_register_dlci(struct gsm_serdev *gsd,
+			     struct gsm_serdev_dlci *ops)
+{
+	struct gsm_dlci *dlci;
+	struct gsm_mux *gsm;
+	int retries;
+
+	if (!gsd || !gsd->gsm || !gsd->serdev)
+		return -ENODEV;
+
+	gsm = gsd->gsm;
+
+	if (!ops || !ops->line)
+		return -EINVAL;
+
+	dlci = gsd_dlci_get(gsd, ops->line, true);
+	if (IS_ERR(dlci))
+		return PTR_ERR(dlci);
+
+	if (dlci->state == DLCI_OPENING || dlci->state == DLCI_OPEN ||
+	    dlci->state == DLCI_CLOSING)
+		return -EBUSY;
+
+	mutex_lock(&dlci->mutex);
+	dlci->ops = ops;
+	dlci->modem_rx = 0;
+	dlci->prev_data = dlci->data;
+	dlci->data = gsd_dlci_data;
+	mutex_unlock(&dlci->mutex);
+
+	gsm_dlci_begin_open(dlci);
+
+	/*
+	 * Allow some time for dlci to move to DLCI_OPEN state. Otherwise
+	 * the serdev consumer driver can start sending data too early during
+	 * the setup, and the response will be missed by gms_queue() if we
+	 * still have DLCI_CLOSED state.
+	 */
+	for (retries = 10; retries > 0; retries--) {
+		if (dlci->state == DLCI_OPEN)
+			break;
+		msleep(100);
+	}
+
+	if (!retries)
+		dev_dbg(&gsd->serdev->dev, "dlci%i not currently active\n",
+			dlci->addr);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_register_dlci);
+
+/**
+ * gsm_serdev_unregister_dlci - unregister a ts 27.010 channel
+ * @gsd:	serdev-gsm instance
+ * @ops:	channel ops
+ */
+void gsm_serdev_unregister_dlci(struct gsm_serdev *gsd,
+				struct gsm_serdev_dlci *ops)
+{
+	struct gsm_mux *gsm;
+	struct gsm_dlci *dlci;
+
+	if (!gsd || !gsd->gsm || !gsd->serdev)
+		return;
+
+	gsm = gsd->gsm;
+
+	if (!ops || !ops->line)
+		return;
+
+	dlci = gsd_dlci_get(gsd, ops->line, false);
+	if (IS_ERR(dlci))
+		return;
+
+	mutex_lock(&dlci->mutex);
+	gsm_destroy_network(dlci);
+	dlci->data = dlci->prev_data;
+	dlci->ops = NULL;
+	mutex_unlock(&dlci->mutex);
+
+	gsm_dlci_begin_close(dlci);
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_unregister_dlci);
+
+static int gsm_serdev_output(struct gsm_mux *gsm, u8 *data, int len)
+{
+	struct serdev_device *serdev = gsm->gsd->serdev;
+
+	if (gsm->gsd->output)
+		return gsm->gsd->output(gsm->gsd, data, len);
+	else
+		return serdev_device_write_buf(serdev, data, len);
+}
+
+static int gsd_receive_buf(struct serdev_device *serdev, const u8 *data,
+			   size_t count)
+{
+	struct gsm_serdev *gsd = serdev_device_get_drvdata(serdev);
+	struct gsm_mux *gsm;
+	const unsigned char *dp;
+	int i;
+
+	if (WARN_ON(!gsd))
+		return 0;
+
+	gsm = gsd->gsm;
+
+	if (debug & 4)
+		print_hex_dump_bytes("gsd_receive_buf: ",
+				     DUMP_PREFIX_OFFSET,
+				     data, count);
+
+	for (i = count, dp = data; i; i--, dp++)
+		gsm->receive(gsm, *dp);
+
+	return count;
+}
+
+static void gsd_write_wakeup(struct serdev_device *serdev)
+{
+	serdev_device_write_wakeup(serdev);
+}
+
+static struct serdev_device_ops gsd_client_ops = {
+	.receive_buf = gsd_receive_buf,
+	.write_wakeup = gsd_write_wakeup,
+};
+
+int gsm_serdev_register_tty_port(struct gsm_serdev *gsd, int line)
+{
+	struct gsm_serdev_dlci *ops;
+	unsigned int base;
+	int error;
+
+	if (line < 1)
+		return -EINVAL;
+
+	ops = kzalloc(sizeof(*ops), GFP_KERNEL);
+	if (!ops)
+		return -ENOMEM;
+
+	ops->line = line;
+
+	error = gsm_serdev_register_dlci(gsd, ops);
+	if (error) {
+		kfree(ops);
+
+		return error;
+	}
+
+	base = mux_num_to_base(gsd->gsm);
+	tty_register_device(gsm_tty_driver, base + ops->line, NULL);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_register_tty_port);
+
+void gsm_serdev_unregister_tty_port(struct gsm_serdev *gsd, int line)
+{
+	struct gsm_dlci *dlci;
+	unsigned int base;
+
+	if (line < 1)
+		return;
+
+	dlci = gsd_dlci_get(gsd, line, false);
+	if (IS_ERR(dlci))
+		return;
+
+	base = mux_num_to_base(gsd->gsm);
+	tty_unregister_device(gsm_tty_driver, base + line);
+	gsm_serdev_unregister_dlci(gsd, dlci->ops);
+	kfree(dlci->ops);
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_unregister_tty_port);
+
+int gsm_serdev_register_device(struct gsm_serdev *gsd)
+{
+	struct gsm_mux *gsm;
+	int error;
+
+	if (WARN(!gsd || !gsd->serdev || !gsd->output,
+		 "serdev and output must be initialized\n"))
+		return -EINVAL;
+
+	serdev_device_set_client_ops(gsd->serdev, &gsd_client_ops);
+
+	gsm = gsm_alloc_mux();
+	if (!gsm)
+		return -ENOMEM;
+
+	gsm->encoding = 1;
+	gsm->tty = NULL;
+	gsm->gsd = gsd;
+	gsm->output = gsm_serdev_output;
+	gsd->gsm = gsm;
+	mux_get(gsd->gsm);
+
+	error = gsm_activate_mux(gsd->gsm);
+	if (error) {
+		gsm_cleanup_mux(gsd->gsm);
+		mux_put(gsd->gsm);
+
+		return error;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_register_device);
+
+void gsm_serdev_unregister_device(struct gsm_serdev *gsd)
+{
+	gsm_cleanup_mux(gsd->gsm);
+	mux_put(gsd->gsm);
+	gsd->gsm = NULL;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_unregister_device);
+
+#endif	/* CONFIG_SERIAL_DEV_BUS */
+
 /**
  *	gsmld_output		-	write to link
  *	@gsm: our mux
diff --git a/include/linux/serdev-gsm.h b/include/linux/serdev-gsm.h
new file mode 100644
--- /dev/null
+++ b/include/linux/serdev-gsm.h
@@ -0,0 +1,152 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _LINUX_SERDEV_GSM_H
+#define _LINUX_SERDEV_GSM_H
+
+#include <linux/device.h>
+#include <linux/serdev.h>
+#include <linux/types.h>
+
+struct gsm_serdev_dlci;
+struct gsm_config;
+
+/**
+ * struct gsm_serdev - serdev-gsm instance
+ * @serdev:		serdev instance
+ * @gsm:		ts 27.010 n_gsm instance
+ * @drvdata:		serdev-gsm consumer driver data
+ * @output:		read data from ts 27.010 channel
+ *
+ * Currently only serdev and output must be initialized, the rest are
+ * are initialized by gsm_serdev_register_dlci().
+ */
+struct gsm_serdev {
+	struct serdev_device *serdev;
+	struct gsm_mux *gsm;
+	void *drvdata;
+	int (*output)(struct gsm_serdev *gsd, u8 *data, int len);
+};
+
+/**
+ * struct gsm_serdev_dlci - serdev-gsm ts 27.010 channel data
+ * @line:		ts 27.010 channel, control channel 0 is not available
+ * @receive_buf:	function to handle data received for the channel
+ * @drvdata:		dlci specific consumer driver data
+ */
+struct gsm_serdev_dlci {
+	int line;
+	int (*receive_buf)(struct gsm_serdev_dlci *ops,
+			   const unsigned char *buf,
+			   size_t len);
+	void *drvdata;
+};
+
+#if IS_ENABLED(CONFIG_N_GSM) && IS_ENABLED(CONFIG_SERIAL_DEV_BUS)
+
+extern int gsm_serdev_register_device(struct gsm_serdev *gsd);
+extern void gsm_serdev_unregister_device(struct gsm_serdev *gsd);
+extern int gsm_serdev_register_tty_port(struct gsm_serdev *gsd, int line);
+extern void gsm_serdev_unregister_tty_port(struct gsm_serdev *gsd, int line);
+
+static inline void *gsm_serdev_get_drvdata(struct device *dev)
+{
+	struct serdev_device *serdev = to_serdev_device(dev);
+	struct gsm_serdev *gsd = serdev_device_get_drvdata(serdev);
+
+	if (gsd)
+		return gsd->drvdata;
+
+	return NULL;
+}
+
+static inline void gsm_serdev_set_drvdata(struct device *dev, void *drvdata)
+{
+	struct serdev_device *serdev = to_serdev_device(dev);
+	struct gsm_serdev *gsd = serdev_device_get_drvdata(serdev);
+
+	if (gsd)
+		gsd->drvdata = drvdata;
+}
+
+extern int gsm_serdev_get_config(struct gsm_serdev *gsd, struct gsm_config *c);
+extern int gsm_serdev_set_config(struct gsm_serdev *gsd, struct gsm_config *c);
+extern int
+gsm_serdev_register_dlci(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops);
+extern void
+gsm_serdev_unregister_dlci(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops);
+extern int gsm_serdev_write(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops,
+			    const u8 *buf, int len);
+extern void gsm_serdev_data_kick(struct gsm_serdev *gsd);
+
+#else	/* CONFIG_SERIAL_DEV_BUS */
+
+static inline
+int gsm_serdev_register_device(struct gsm_serdev *gsd)
+{
+	return -ENODEV;
+}
+
+static inline void gsm_serdev_unregister_device(struct gsm_serdev *gsd)
+{
+}
+
+static inline int
+gsm_serdev_register_tty_port(struct gsm_serdev *gsd, int line)
+{
+	return -ENODEV;
+}
+
+static inline
+void gsm_serdev_unregister_tty_port(struct gsm_serdev *gsd, int line)
+{
+}
+
+static inline void *gsm_serdev_get_drvdata(struct device *dev)
+{
+	return NULL;
+}
+
+static inline
+void gsm_serdev_set_drvdata(struct device *dev, void *drvdata)
+{
+}
+
+static inline
+int gsm_serdev_get_config(struct gsm_serdev *gsd, struct gsm_config *c)
+{
+	return -ENODEV;
+}
+
+static inline
+int gsm_serdev_set_config(struct gsm_serdev *gsd, struct gsm_config *c)
+{
+	return -ENODEV;
+}
+
+static inline
+int gsm_serdev_register_dlci(struct gsm_serdev *gsd,
+			     struct gsm_serdev_dlci *ops)
+{
+	return -ENODEV;
+}
+
+static inline
+void gsm_serdev_unregister_dlci(struct gsm_serdev *gsd,
+				struct gsm_serdev_dlci *ops)
+{
+}
+
+static inline
+int gsm_serdev_write(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops,
+		     const u8 *buf, int len)
+{
+	return -ENODEV;
+}
+
+static inline
+void gsm_serdev_data_kick(struct gsm_serdev *gsd)
+{
+}
+
+#endif	/* CONFIG_N_GSM && CONFIG_SERIAL_DEV_BUS */
+#endif	/* _LINUX_SERDEV_GSM_H */
-- 
2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

* [PATCH 2/6] dt-bindings: serdev: ngsm: Add binding for serdev-ngsm
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
  2020-04-30 17:46 ` [PATCH 1/6] tty: n_gsm: Add support for serdev drivers Tony Lindgren
@ 2020-04-30 17:46 ` Tony Lindgren
  2020-05-01 20:33   ` Pavel Machek
  2020-05-12 15:37   ` Rob Herring
  2020-04-30 17:46 ` [PATCH 3/6] serdev: ngsm: Add generic serdev-ngsm driver Tony Lindgren
                   ` (7 subsequent siblings)
  9 siblings, 2 replies; 29+ messages in thread
From: Tony Lindgren @ 2020-04-30 17:46 UTC (permalink / raw)
  To: Greg Kroah-Hartman, Johan Hovold, Rob Herring
  Cc: Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap

Add a binding document for a generic serdev-ngsm driver that can be
used to bring up TS 27.010 line discipline with Linux n_gsm support
on a serial port.

As the Motorola Mapphone modems require some custom handling, they
are handled with a separate compatible.

Let's also add vendor string for ETSI as we're using a ETSI 3GPP
TS 27.010 standard.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 .../bindings/serdev/serdev-ngsm.yaml          | 64 +++++++++++++++++++
 .../devicetree/bindings/vendor-prefixes.yaml  |  2 +
 2 files changed, 66 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/serdev/serdev-ngsm.yaml

diff --git a/Documentation/devicetree/bindings/serdev/serdev-ngsm.yaml b/Documentation/devicetree/bindings/serdev/serdev-ngsm.yaml
new file mode 100644
--- /dev/null
+++ b/Documentation/devicetree/bindings/serdev/serdev-ngsm.yaml
@@ -0,0 +1,64 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/serdev/serdev-ngsm.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Generic serdev-ngsm TS 27.010 driver
+
+maintainers:
+  - Tony Lindgren <tony@atomide.com>
+
+properties:
+  compatible:
+    enum:
+      - etsi,3gpp-ts27010-adaption1
+      - motorola,mapphone-mdm6600-serial
+
+  ttymask:
+    $ref: /schemas/types.yaml#/definitions/uint64
+    description: Mask of the TS 27.010 channel TTY interfaces to start (64 bit)
+
+  "#address-cells":
+    const: 1
+
+  "#size-cells":
+    const: 0
+
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: motorola,mapphone-mdm6600-serial
+    then:
+      properties:
+        phys:
+          $ref: /schemas/types.yaml#/definitions/phandle-array
+          description: USB PHY needed for shared GPIO PM wake-up pins
+          maxItems: 1
+
+        phy-names:
+          description: Name of the USB PHY
+          const: usb
+
+      required:
+        - phys
+        - phy-names
+
+required:
+  - compatible
+  - ttymask
+  - "#address-cells"
+  - "#size-cells"
+
+examples:
+  - |
+    modem {
+      compatible = "motorola,mapphone-mdm6600-serial";
+      ttymask = <0 0x00001fee>;
+      phys = <&fsusb1_phy>;
+      phy-names = "usb";
+      #address-cells = <1>;
+      #size-cells = <0>;
+    };
diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml
--- a/Documentation/devicetree/bindings/vendor-prefixes.yaml
+++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml
@@ -323,6 +323,8 @@ patternProperties:
     description: Espressif Systems Co. Ltd.
   "^est,.*":
     description: ESTeem Wireless Modems
+  "^etsi,.*":
+    description: ETSI
   "^ettus,.*":
     description: NI Ettus Research
   "^eukrea,.*":
-- 
2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

* [PATCH 3/6] serdev: ngsm: Add generic serdev-ngsm driver
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
  2020-04-30 17:46 ` [PATCH 1/6] tty: n_gsm: Add support for serdev drivers Tony Lindgren
  2020-04-30 17:46 ` [PATCH 2/6] dt-bindings: serdev: ngsm: Add binding for serdev-ngsm Tony Lindgren
@ 2020-04-30 17:46 ` Tony Lindgren
  2020-05-01 20:44   ` Pavel Machek
  2020-04-30 17:46 ` [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS Tony Lindgren
                   ` (6 subsequent siblings)
  9 siblings, 1 reply; 29+ messages in thread
From: Tony Lindgren @ 2020-04-30 17:46 UTC (permalink / raw)
  To: Greg Kroah-Hartman, Johan Hovold, Rob Herring
  Cc: Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap

We can have a generic serdev-ngsm driver bring up the TS 27.010 line
discipline on the selected serial ports based on device tree data.

And we can now do standard Linux device driver for the dedicated
TS 27.010 channels for devices like GNSS and ALSA found on modems.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 drivers/tty/serdev/Kconfig       |  10 +
 drivers/tty/serdev/Makefile      |   1 +
 drivers/tty/serdev/serdev-ngsm.c | 449 +++++++++++++++++++++++++++++++
 include/linux/serdev-gsm.h       |  11 +
 4 files changed, 471 insertions(+)
 create mode 100644 drivers/tty/serdev/serdev-ngsm.c

diff --git a/drivers/tty/serdev/Kconfig b/drivers/tty/serdev/Kconfig
--- a/drivers/tty/serdev/Kconfig
+++ b/drivers/tty/serdev/Kconfig
@@ -22,4 +22,14 @@ config SERIAL_DEV_CTRL_TTYPORT
 	depends on SERIAL_DEV_BUS != m
 	default y
 
+config SERIAL_DEV_N_GSM
+	tristate "Serial device TS 27.010 support"
+	help
+	  Select this if you want to use the TS 27.010 with a serial port with
+	  devices such as modems and GNSS devices.
+
+	  If unsure, say N.
+	depends on N_GSM
+	depends on SERIAL_DEV_CTRL_TTYPORT
+
 endif
diff --git a/drivers/tty/serdev/Makefile b/drivers/tty/serdev/Makefile
--- a/drivers/tty/serdev/Makefile
+++ b/drivers/tty/serdev/Makefile
@@ -4,3 +4,4 @@ serdev-objs := core.o
 obj-$(CONFIG_SERIAL_DEV_BUS) += serdev.o
 
 obj-$(CONFIG_SERIAL_DEV_CTRL_TTYPORT) += serdev-ttyport.o
+obj-$(CONFIG_SERIAL_DEV_N_GSM) += serdev-ngsm.o
diff --git a/drivers/tty/serdev/serdev-ngsm.c b/drivers/tty/serdev/serdev-ngsm.c
new file mode 100644
--- /dev/null
+++ b/drivers/tty/serdev/serdev-ngsm.c
@@ -0,0 +1,449 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Generic TS 27.010 serial line discipline serdev driver
+ * Copyright (C) 2020 Tony Lindgren <tony@atomide.com>
+ */
+
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/serdev.h>
+#include <linux/serdev-gsm.h>
+
+#include <linux/phy/phy.h>
+
+#include <uapi/linux/gsmmux.h>
+
+#define TS27010_C_N2		3	/* TS 27.010 default value */
+#define TS27010_RESERVED_DLCI	(BIT_ULL(63) | BIT_ULL(62) | BIT_ULL(0))
+
+struct serdev_ngsm_cfg {
+	const struct gsm_config *gsm;
+	unsigned int init_retry_quirk:1;
+	unsigned int needs_usb_phy:1;
+	unsigned int aggressive_pm:1;
+	int (*init)(struct serdev_device *serdev); /* for device quirks */
+};
+
+struct serdev_ngsm {
+	struct device *dev;
+	struct gsm_serdev gsd;
+	struct phy *phy;
+	u32 baudrate;
+	DECLARE_BITMAP(ttymask, 64);
+	const struct serdev_ngsm_cfg *cfg;
+};
+
+static int serdev_ngsm_tty_init(struct serdev_ngsm *ddata)
+{
+	struct gsm_serdev *gsd = &ddata->gsd;
+	struct device *dev = ddata->dev;
+	int bit, err;
+
+	for_each_set_bit(bit, ddata->ttymask, 64) {
+		if (BIT_ULL(bit) & TS27010_RESERVED_DLCI)
+			continue;
+
+		err = gsm_serdev_register_tty_port(gsd, bit);
+		if (err) {
+			dev_err(dev, "ngsm tty init failed for dlci%i: %i\n",
+				bit, err);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static void serdev_ngsm_tty_exit(struct serdev_ngsm *ddata)
+{
+	struct gsm_serdev *gsd = &ddata->gsd;
+	int bit;
+
+	for_each_set_bit(bit, ddata->ttymask, 64) {
+		if (BIT_ULL(bit) & TS27010_RESERVED_DLCI)
+			continue;
+
+		gsm_serdev_unregister_tty_port(gsd, bit);
+	}
+}
+
+/*
+ * Note that we rely on gsm_serdev_register_dlci() locking for
+ * reserved channels that serdev_ngsm_tty_init() and consumer
+ * drivers may have already reserved.
+ */
+int serdev_ngsm_register_dlci(struct device *dev,
+			      struct gsm_serdev_dlci *dlci)
+{
+	struct serdev_ngsm *ddata = gsm_serdev_get_drvdata(dev);
+	struct gsm_serdev *gsd = &ddata->gsd;
+	int err;
+
+	err = gsm_serdev_register_dlci(gsd, dlci);
+	if (err)
+		return err;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(serdev_ngsm_register_dlci);
+
+void serdev_ngsm_unregister_dlci(struct device *dev,
+				 struct gsm_serdev_dlci *dlci)
+{
+	struct serdev_ngsm *ddata = gsm_serdev_get_drvdata(dev);
+	struct gsm_serdev *gsd = &ddata->gsd;
+
+	gsm_serdev_unregister_dlci(gsd, dlci);
+}
+EXPORT_SYMBOL_GPL(serdev_ngsm_unregister_dlci);
+
+int serdev_ngsm_write(struct device *dev, struct gsm_serdev_dlci *ops,
+		      const u8 *buf, int len)
+{
+	struct serdev_ngsm *ddata = gsm_serdev_get_drvdata(dev);
+	struct gsm_serdev *gsd = &ddata->gsd;
+	int ret;
+
+	ret = pm_runtime_get_sync(dev);
+	if ((ret != -EINPROGRESS) && ret < 0) {
+		pm_runtime_put_noidle(dev);
+
+		return ret;
+	}
+
+	ret = gsm_serdev_write(gsd, ops, buf, len);
+
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_put_autosuspend(dev);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(serdev_ngsm_write);
+
+static int serdev_ngsm_set_config(struct device *dev)
+{
+	struct serdev_ngsm *ddata = gsm_serdev_get_drvdata(dev);
+	struct gsm_serdev *gsd = &ddata->gsd;
+	struct gsm_config c;
+	int err, n2;
+
+	memcpy(&c, ddata->cfg->gsm, sizeof(c));
+
+	if (ddata->cfg->init_retry_quirk) {
+		n2 = c.n2;
+		c.n2 *= 10;
+		err = gsm_serdev_set_config(gsd, &c);
+		if (err)
+			return err;
+
+		msleep(5000);
+		c.n2 = n2;
+	}
+
+	err = gsm_serdev_set_config(gsd, &c);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int serdev_ngsm_output(struct gsm_serdev *gsd, u8 *data, int len)
+{
+	struct serdev_device *serdev = gsd->serdev;
+	struct device *dev = &serdev->dev;
+	int err;
+
+	err = pm_runtime_get(dev);
+	if ((err != -EINPROGRESS) && err < 0) {
+		pm_runtime_put_noidle(dev);
+
+		return err;
+	}
+
+	serdev_device_write_buf(serdev, data, len);
+
+	pm_runtime_put(dev);
+
+	return len;
+}
+
+static int serdev_ngsm_runtime_suspend(struct device *dev)
+{
+	struct serdev_ngsm *ddata = gsm_serdev_get_drvdata(dev);
+	int err;
+
+	if (ddata->cfg->needs_usb_phy) {
+		err = phy_pm_runtime_put(ddata->phy);
+		if (err < 0) {
+			dev_warn(dev, "%s: phy_pm_runtime_put: %i\n",
+				 __func__, err);
+
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int serdev_ngsm_runtime_resume(struct device *dev)
+{
+	struct serdev_ngsm *ddata = gsm_serdev_get_drvdata(dev);
+	int err;
+
+	if (ddata->cfg->needs_usb_phy) {
+		err = phy_pm_runtime_get_sync(ddata->phy);
+		if (err < 0) {
+			dev_warn(dev, "%s: phy_pm_runtime_get: %i\n",
+				 __func__, err);
+
+			return err;
+		}
+	}
+
+	gsm_serdev_data_kick(&ddata->gsd);
+
+	return 0;
+}
+
+static const struct dev_pm_ops serdev_ngsm_pm_ops = {
+	SET_RUNTIME_PM_OPS(serdev_ngsm_runtime_suspend,
+			   serdev_ngsm_runtime_resume,
+			   NULL)
+};
+
+/*
+ * At least Motorola MDM6600 devices have GPIO wake pins shared between the
+ * USB PHY and the TS 27.010 interface. So for PM, we need to use the calls
+ * for phy_pm_runtime. Otherwise the modem won't respond to anything on the
+ * UART and will never idle either.
+ */
+static int serdev_ngsm_phy_init(struct device *dev)
+{
+	struct serdev_ngsm *ddata = gsm_serdev_get_drvdata(dev);
+	int err;
+
+	if (!ddata->cfg->needs_usb_phy)
+		return 0;
+
+	ddata->phy = devm_of_phy_get(dev, dev->of_node, NULL);
+	if (IS_ERR(ddata->phy)) {
+		err = PTR_ERR(ddata->phy);
+		if (err != -EPROBE_DEFER)
+			dev_err(dev, "%s: phy error: %i\n", __func__, err);
+
+		return err;
+	}
+
+	return 0;
+}
+
+/*
+ * Configure SoC 8250 device for 700 ms autosuspend delay, Values around 600 ms
+ * and shorter cause spurious wake-up events at least on droid 4. Also keep the
+ * SoC 8250 device active during use because of the OOB GPIO wake-up signaling
+ * shared with USB PHY.
+ */
+static int motmdm_init(struct serdev_device *serdev)
+{
+	pm_runtime_set_autosuspend_delay(serdev->ctrl->dev.parent, 700);
+	pm_suspend_ignore_children(&serdev->ctrl->dev, false);
+
+	return 0;
+}
+
+static const struct gsm_config adaption1 = {
+	.i = 1,			/* 1 = UIH, 2 = UI */
+	.initiator = 1,
+	.encapsulation = 0,	/* basic mode */
+	.adaption = 1,
+	.mru = 1024,		/* from android TS 27010 driver */
+	.mtu = 1024,		/* from android TS 27010 driver */
+	.t1 = 10,		/* ack timer, default 10ms */
+	.t2 = 34,		/* response timer, default 34 */
+	.n2 = 3,		/* retransmissions, default 3 */
+};
+
+static const struct serdev_ngsm_cfg adaption1_cfg = {
+	.gsm = &adaption1,
+};
+
+static const struct serdev_ngsm_cfg motmdm_cfg = {
+	.gsm = &adaption1,
+	.init_retry_quirk = true,
+	.needs_usb_phy = true,
+	.aggressive_pm = true,
+	.init = motmdm_init,
+};
+
+static const struct of_device_id serdev_ngsm_id_table[] = {
+	{
+		.compatible = "etsi,3gpp-ts27010-adaption1",
+		.data = &adaption1_cfg,
+	},
+	{
+		.compatible = "motorola,mapphone-mdm6600-serial",
+		.data = &motmdm_cfg,
+	},
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, serdev_ngsm_id_table);
+
+static int serdev_ngsm_probe(struct serdev_device *serdev)
+{
+	struct device *dev = &serdev->dev;
+	const struct of_device_id *match;
+	struct gsm_serdev *gsd;
+	struct serdev_ngsm *ddata;
+	u64 ttymask;
+	int err;
+
+	match = of_match_device(of_match_ptr(serdev_ngsm_id_table), dev);
+	if (!match)
+		return -ENODEV;
+
+	ddata = devm_kzalloc(dev, sizeof(*ddata), GFP_KERNEL);
+	if (!ddata)
+		return -ENOMEM;
+
+	ddata->dev = dev;
+	ddata->cfg = match->data;
+
+	gsd = &ddata->gsd;
+	gsd->serdev = serdev;
+	gsd->output = serdev_ngsm_output;
+	serdev_device_set_drvdata(serdev, gsd);
+	gsm_serdev_set_drvdata(dev, ddata);
+
+	err = serdev_ngsm_phy_init(dev);
+	if (err)
+		return err;
+
+	err = of_property_read_u64(dev->of_node, "ttymask", &ttymask);
+	if (err) {
+		dev_err(dev, "invalid or missing ttymask: %i\n", err);
+
+		return err;
+	}
+
+	bitmap_from_u64(ddata->ttymask, ttymask);
+
+	pm_runtime_set_autosuspend_delay(dev, 200);
+	pm_runtime_use_autosuspend(dev);
+	pm_runtime_enable(dev);
+	err = pm_runtime_get_sync(dev);
+	if (err < 0) {
+		pm_runtime_put_noidle(dev);
+
+		return err;
+	}
+
+	err = gsm_serdev_register_device(gsd);
+	if (err)
+		goto err_disable;
+
+	err = serdev_device_open(gsd->serdev);
+	if (err)
+		goto err_disable;
+
+	/* Optional serial port configuration */
+	of_property_read_u32(dev->of_node->parent, "current-speed",
+			     &ddata->baudrate);
+	if (ddata->baudrate)
+		serdev_device_set_baudrate(gsd->serdev, ddata->baudrate);
+
+	if (of_get_property(dev->of_node->parent, "uart-has-rtscts", NULL)) {
+		serdev_device_set_rts(gsd->serdev, true);
+		serdev_device_set_flow_control(gsd->serdev, true);
+	}
+
+	err = serdev_ngsm_set_config(dev);
+	if (err)
+		goto err_close;
+
+	err = serdev_ngsm_tty_init(ddata);
+	if (err)
+		goto err_tty;
+
+	if (ddata->cfg->init) {
+		err = ddata->cfg->init(serdev);
+		if (err)
+			goto err_tty;
+	}
+
+	err = of_platform_populate(dev->of_node, NULL, NULL, dev);
+	if (err)
+		goto err_tty;
+
+	/* Allow parent serdev device to idle when open, balanced in remove */
+	if (ddata->cfg->aggressive_pm)
+		pm_runtime_put(&serdev->ctrl->dev);
+
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_put_autosuspend(dev);
+
+	return 0;
+
+err_tty:
+	serdev_ngsm_tty_exit(ddata);
+
+err_close:
+	serdev_device_close(serdev);
+
+err_disable:
+	pm_runtime_dont_use_autosuspend(dev);
+	pm_runtime_put_sync(dev);
+	pm_runtime_disable(dev);
+	gsm_serdev_unregister_device(gsd);
+
+	return err;
+}
+
+static void serdev_ngsm_remove(struct serdev_device *serdev)
+{
+	struct gsm_serdev *gsd = serdev_device_get_drvdata(serdev);
+	struct device *dev = &serdev->dev;
+	struct serdev_ngsm *ddata;
+	int err;
+
+	ddata = gsm_serdev_get_drvdata(dev);
+
+	/* Balance the put done in probe for UART */
+	if (ddata->cfg->aggressive_pm)
+		pm_runtime_get(&serdev->ctrl->dev);
+
+	err = pm_runtime_get_sync(dev);
+	if (err < 0)
+		dev_warn(dev, "%s: PM runtime: %i\n", __func__, err);
+
+	of_platform_depopulate(dev);
+	serdev_ngsm_tty_exit(ddata);
+	serdev_device_close(serdev);
+	gsm_serdev_unregister_device(gsd);
+
+	pm_runtime_dont_use_autosuspend(dev);
+	pm_runtime_put_sync(dev);
+	pm_runtime_disable(dev);
+}
+
+static struct serdev_device_driver serdev_ngsm_driver = {
+	.driver = {
+		.name = "serdev_ngsm",
+		.of_match_table = of_match_ptr(serdev_ngsm_id_table),
+		.pm = &serdev_ngsm_pm_ops,
+	},
+	.probe = serdev_ngsm_probe,
+	.remove = serdev_ngsm_remove,
+};
+
+module_serdev_device_driver(serdev_ngsm_driver);
+
+MODULE_DESCRIPTION("serdev n_gsm driver");
+MODULE_AUTHOR("Tony Lindgren <tony@atomide.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/include/linux/serdev-gsm.h b/include/linux/serdev-gsm.h
--- a/include/linux/serdev-gsm.h
+++ b/include/linux/serdev-gsm.h
@@ -43,6 +43,17 @@ struct gsm_serdev_dlci {
 
 #if IS_ENABLED(CONFIG_N_GSM) && IS_ENABLED(CONFIG_SERIAL_DEV_BUS)
 
+/* TS 27.010 channel specific functions for consumer drivers */
+#if IS_ENABLED(CONFIG_SERIAL_DEV_N_GSM)
+extern int
+serdev_ngsm_register_dlci(struct device *dev, struct gsm_serdev_dlci *dlci);
+extern void serdev_ngsm_unregister_dlci(struct device *dev,
+					struct gsm_serdev_dlci *dlci);
+extern int serdev_ngsm_write(struct device *dev, struct gsm_serdev_dlci *ops,
+			     const u8 *buf, int len);
+#endif
+
+/* Interface for_gsm serdev support */
 extern int gsm_serdev_register_device(struct gsm_serdev *gsd);
 extern void gsm_serdev_unregister_device(struct gsm_serdev *gsd);
 extern int gsm_serdev_register_tty_port(struct gsm_serdev *gsd, int line);
-- 
2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

* [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
                   ` (2 preceding siblings ...)
  2020-04-30 17:46 ` [PATCH 3/6] serdev: ngsm: Add generic serdev-ngsm driver Tony Lindgren
@ 2020-04-30 17:46 ` Tony Lindgren
  2020-05-01 20:45   ` Pavel Machek
                     ` (2 more replies)
  2020-04-30 17:46 ` [PATCH 5/6] gnss: motmdm: Add support for Motorola Mapphone MDM6600 modem Tony Lindgren
                   ` (5 subsequent siblings)
  9 siblings, 3 replies; 29+ messages in thread
From: Tony Lindgren @ 2020-04-30 17:46 UTC (permalink / raw)
  To: Greg Kroah-Hartman, Johan Hovold, Rob Herring
  Cc: Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap

Add a binding document for Motorola Mapphone MDM6600 GNSS driver that
can be used with phones such as droid4.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 .../devicetree/bindings/gnss/motmdm.yaml      | 29 +++++++++++++++++++
 1 file changed, 29 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/gnss/motmdm.yaml

diff --git a/Documentation/devicetree/bindings/gnss/motmdm.yaml b/Documentation/devicetree/bindings/gnss/motmdm.yaml
new file mode 100644
--- /dev/null
+++ b/Documentation/devicetree/bindings/gnss/motmdm.yaml
@@ -0,0 +1,29 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/gnss/motmdm.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Motorola Mapphone MDM6600 GNSS binding
+
+maintainers:
+  - Tony Lindgren <tony@atomide.com>
+
+properties:
+  compatible:
+    const: motorola,mapphone-mdm6600-gnss
+
+  reg:
+    items:
+      description: TS 27.010 channel the GNSS device is using
+
+required:
+  - compatible
+  - reg
+
+examples:
+  - |
+    gnss@4 {
+      compatible = "motorola,mapphone-mdm6600-gnss";
+      reg = <4>;
+    };
-- 
2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

* [PATCH 5/6] gnss: motmdm: Add support for Motorola Mapphone MDM6600 modem
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
                   ` (3 preceding siblings ...)
  2020-04-30 17:46 ` [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS Tony Lindgren
@ 2020-04-30 17:46 ` Tony Lindgren
  2020-05-01 20:51   ` Pavel Machek
  2020-04-30 17:46 ` [PATCH 6/6] ARM: dts: omap4-droid4: Configure modem for serdev-ngsm Tony Lindgren
                   ` (4 subsequent siblings)
  9 siblings, 1 reply; 29+ messages in thread
From: Tony Lindgren @ 2020-04-30 17:46 UTC (permalink / raw)
  To: Greg Kroah-Hartman, Johan Hovold, Rob Herring
  Cc: Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap

Motorola is using a custom TS 27.010 based serial port line discipline
for various devices on the modem. These devices can be accessed on
dedicated channels using Linux kernel serdev-ngsm driver.

For the GNSS on these devices, we need to kick the GNSS device at a
desired rate. Otherwise the GNSS device stops sending data after a
few minutes. The rate we poll data defaults to 1000 ms, and can be
specified with a module option rate_ms between 1 to 16 seconds.

Note that AGPS with xtra2.bin is not yet supported, so getting a fix
can take quite a while. And a recent gpsd is needed to parse the
$GNGNS output, and to properly handle the /dev/gnss0 character device.
I've confirmed it works properly with gpsd-3.20.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 drivers/gnss/Kconfig  |   8 +
 drivers/gnss/Makefile |   3 +
 drivers/gnss/motmdm.c | 419 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 430 insertions(+)
 create mode 100644 drivers/gnss/motmdm.c

diff --git a/drivers/gnss/Kconfig b/drivers/gnss/Kconfig
--- a/drivers/gnss/Kconfig
+++ b/drivers/gnss/Kconfig
@@ -13,6 +13,14 @@ menuconfig GNSS
 
 if GNSS
 
+config GNSS_MOTMDM
+	tristate "Motorola Modem TS 27.010 serdev GNSS receiver support"
+	depends on SERIAL_DEV_N_GSM
+	---help---
+	  Say Y here if you have a Motorola modem using TS 27.010 line
+	  discipline for GNSS such as a Motorola Mapphone series device
+	  like Droid 4.
+
 config GNSS_SERIAL
 	tristate
 
diff --git a/drivers/gnss/Makefile b/drivers/gnss/Makefile
--- a/drivers/gnss/Makefile
+++ b/drivers/gnss/Makefile
@@ -6,6 +6,9 @@
 obj-$(CONFIG_GNSS)			+= gnss.o
 gnss-y := core.o
 
+obj-$(CONFIG_GNSS_MOTMDM)		+= gnss-motmdm.o
+gnss-motmdm-y := motmdm.o
+
 obj-$(CONFIG_GNSS_SERIAL)		+= gnss-serial.o
 gnss-serial-y := serial.o
 
diff --git a/drivers/gnss/motmdm.c b/drivers/gnss/motmdm.c
new file mode 100644
--- /dev/null
+++ b/drivers/gnss/motmdm.c
@@ -0,0 +1,419 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Motorola Modem TS 27.010 serdev GNSS driver
+ *
+ * Copyright (C) 2018 - 2020 Tony Lindgren <tony@atomide.com>
+ *
+ * Based on drivers/gnss/sirf.c driver example:
+ * Copyright (C) 2018 Johan Hovold <johan@kernel.org>
+ */
+
+#include <linux/errno.h>
+#include <linux/gnss.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/serdev-gsm.h>
+#include <linux/slab.h>
+
+#define MOTMDM_GNSS_TIMEOUT	1000
+#define MOTMDM_GNSS_RATE	1000
+
+/*
+ * Motorola MDM GNSS device communicates over a dedicated TS 27.010 channel
+ * using custom data packets. The packets look like AT commands embedded into
+ * a Motorola invented packet using format like "U1234AT+MPDSTART=0,1,100,0".
+ * But it's not an AT compatible serial interface, it's a packet interface
+ * using AT like commands.
+ */
+#define MOTMDM_GNSS_HEADER_LEN	5				/* U1234 */
+#define MOTMDM_GNSS_RESP_LEN	(MOTMDM_GNSS_HEADER_LEN + 4)	/* U1234+MPD */
+#define MOTMDM_GNSS_DATA_LEN	(MOTMDM_GNSS_RESP_LEN + 1)	/* U1234~+MPD */
+#define MOTMDM_GNSS_STATUS_LEN	(MOTMDM_GNSS_DATA_LEN + 7)	/* STATUS= */
+#define MOTMDM_GNSS_NMEA_LEN	(MOTMDM_GNSS_DATA_LEN + 8)	/* NMEA=NN, */
+
+enum motmdm_gnss_status {
+	MOTMDM_GNSS_UNKNOWN,
+	MOTMDM_GNSS_INITIALIZED,
+	MOTMDM_GNSS_DATA_OR_TIMEOUT,
+	MOTMDM_GNSS_STARTED,
+	MOTMDM_GNSS_STOPPED,
+};
+
+struct motmdm_gnss_data {
+	struct gnss_device *gdev;
+	struct device *modem;
+	struct gsm_serdev_dlci dlci;
+	struct delayed_work restart_work;
+	struct mutex mutex;	/* For modem commands */
+	ktime_t last_update;
+	int status;
+	unsigned char *buf;
+	size_t len;
+	wait_queue_head_t read_queue;
+	unsigned int parsed:1;
+};
+
+static unsigned int rate_ms = MOTMDM_GNSS_RATE;
+module_param(rate_ms, uint, 0644);
+MODULE_PARM_DESC(rate_ms, "GNSS refresh rate between 1000 and 16000 ms (default 1000 ms)");
+
+/*
+ * Note that multiple commands can be sent in series with responses coming
+ * out-of-order. For GNSS, we don't need to care about the out-of-order
+ * responses, and can assume we have at most one command active at a time.
+ * For the commands, can use just a jiffies base packet ID and let the modem
+ * sort out the ID conflicts with the modem's unsolicited message ID
+ * numbering.
+ */
+int motmdm_gnss_send_command(struct motmdm_gnss_data *ddata,
+			     const u8 *buf, int len)
+{
+	struct gnss_device *gdev = ddata->gdev;
+	const int timeout_ms = 1000;
+	unsigned char cmd[128];
+	int ret, cmdlen;
+
+	cmdlen = len + 5 + 1;
+	if (cmdlen > 128)
+		return -EINVAL;
+
+	mutex_lock(&ddata->mutex);
+	memset(ddata->buf, 0, ddata->len);
+	ddata->parsed = false;
+	snprintf(cmd, cmdlen, "U%04li%s", jiffies % 10000, buf);
+	ret = serdev_ngsm_write(ddata->modem, &ddata->dlci, cmd, cmdlen);
+	if (ret < 0)
+		goto out_unlock;
+
+	ret = wait_event_timeout(ddata->read_queue, ddata->parsed,
+				 msecs_to_jiffies(timeout_ms));
+	if (ret == 0) {
+		ret = -ETIMEDOUT;
+		goto out_unlock;
+	} else if (ret < 0) {
+		goto out_unlock;
+	}
+
+	if (!strstr(ddata->buf, ":OK")) {
+		dev_err(&gdev->dev, "command %s error %s\n",
+			cmd, ddata->buf);
+		ret = -EPIPE;
+	}
+
+	ret = len;
+
+out_unlock:
+	mutex_unlock(&ddata->mutex);
+
+	return ret;
+}
+
+/*
+ * Android uses AT+MPDSTART=0,1,100,0 which starts GNSS for a while,
+ * and then GNSS needs to be kicked with an AT command based on a
+ * status message.
+ */
+static void motmdm_gnss_restart(struct work_struct *work)
+{
+	struct motmdm_gnss_data *ddata =
+		container_of(work, struct motmdm_gnss_data,
+			     restart_work.work);
+	struct gnss_device *gdev = ddata->gdev;
+	const unsigned char *cmd = "AT+MPDSTART=0,1,100,0";
+	int error;
+
+	ddata->last_update = ktime_get();
+
+	error = motmdm_gnss_send_command(ddata, cmd, strlen(cmd));
+	if (error < 0) {
+		/* Timeouts can happen, don't warn and try again */
+		if (error != -ETIMEDOUT)
+			dev_warn(&gdev->dev, "%s: could not start: %i\n",
+				 __func__, error);
+
+		schedule_delayed_work(&ddata->restart_work,
+				      msecs_to_jiffies(MOTMDM_GNSS_RATE));
+
+		return;
+	}
+}
+
+static void motmdm_gnss_start(struct gnss_device *gdev, int delay_ms)
+{
+	struct motmdm_gnss_data *ddata = gnss_get_drvdata(gdev);
+	ktime_t now, next, delta;
+	int next_ms;
+
+	now = ktime_get();
+	next = ktime_add_ms(ddata->last_update, delay_ms);
+	delta = ktime_sub(next, now);
+	next_ms = ktime_to_ms(delta);
+
+	if (next_ms < 0)
+		next_ms = 0;
+	if (next_ms > delay_ms)
+		next_ms = delay_ms;
+
+	schedule_delayed_work(&ddata->restart_work, msecs_to_jiffies(next_ms));
+}
+
+static int motmdm_gnss_stop(struct gnss_device *gdev)
+{
+	struct motmdm_gnss_data *ddata = gnss_get_drvdata(gdev);
+	const unsigned char *cmd = "AT+MPDSTOP";
+
+	cancel_delayed_work_sync(&ddata->restart_work);
+
+	return motmdm_gnss_send_command(ddata, cmd, strlen(cmd));
+}
+
+static int motmdm_gnss_init(struct gnss_device *gdev)
+{
+	struct motmdm_gnss_data *ddata = gnss_get_drvdata(gdev);
+	const unsigned char *cmd = "AT+MPDINIT=1";
+	int error;
+
+	error = motmdm_gnss_send_command(ddata, cmd, strlen(cmd));
+	if (error < 0)
+		return error;
+
+	motmdm_gnss_start(gdev, 0);
+
+	return 0;
+}
+
+static int motmdm_gnss_finish(struct gnss_device *gdev)
+{
+	struct motmdm_gnss_data *ddata = gnss_get_drvdata(gdev);
+	const unsigned char *cmd = "AT+MPDINIT=0";
+	int error;
+
+	error = motmdm_gnss_stop(gdev);
+	if (error < 0)
+		return error;
+
+	return motmdm_gnss_send_command(ddata, cmd, strlen(cmd));
+}
+
+static int motmdm_gnss_receive_data(struct gsm_serdev_dlci *dlci,
+				    const unsigned char *buf,
+				    size_t len)
+{
+	struct gnss_device *gdev = dlci->drvdata;
+	struct motmdm_gnss_data *ddata = gnss_get_drvdata(gdev);
+	const unsigned char *msg;
+	size_t msglen;
+	int error = 0;
+
+	if (len <= MOTMDM_GNSS_RESP_LEN)
+		return 0;
+
+	/* Handle U1234+MPD style command response */
+	if (buf[MOTMDM_GNSS_HEADER_LEN] != '~') {
+		msg = buf + MOTMDM_GNSS_RESP_LEN;
+		strncpy(ddata->buf, msg, len - MOTMDM_GNSS_RESP_LEN);
+		ddata->parsed = true;
+		wake_up(&ddata->read_queue);
+
+		return len;
+	}
+
+	if (len <= MOTMDM_GNSS_DATA_LEN)
+		return 0;
+
+	/* Handle U1234~+MPD style unsolicted message */
+	switch (buf[MOTMDM_GNSS_DATA_LEN]) {
+	case 'N':	/* UNNNN~+MPDNMEA=NN, */
+		msg = buf + MOTMDM_GNSS_NMEA_LEN;
+		msglen = len - MOTMDM_GNSS_NMEA_LEN;
+
+		/*
+		 * Firmware bug: Strip out extra duplicate line break always
+		 * in the data
+		 */
+		msglen--;
+
+		/*
+		 * Firmware bug: Strip out extra data based on an
+		 * earlier line break in the data
+		 */
+		if (msg[msglen - 5 - 1] == 0x0a)
+			msglen -= 5;
+
+		error = gnss_insert_raw(gdev, msg, msglen);
+		break;
+	case 'S':	/* UNNNN~+MPDSTATUS=N,NN */
+		msg = buf + MOTMDM_GNSS_STATUS_LEN;
+		msglen = len - MOTMDM_GNSS_STATUS_LEN;
+
+		switch (msg[0]) {
+		case '1':
+			ddata->status = MOTMDM_GNSS_INITIALIZED;
+			break;
+		case '2':
+			ddata->status = MOTMDM_GNSS_DATA_OR_TIMEOUT;
+			if (rate_ms < MOTMDM_GNSS_RATE)
+				rate_ms = MOTMDM_GNSS_RATE;
+			if (rate_ms > 16 * MOTMDM_GNSS_RATE)
+				rate_ms = 16 * MOTMDM_GNSS_RATE;
+			motmdm_gnss_start(gdev, rate_ms);
+			break;
+		case '3':
+			ddata->status = MOTMDM_GNSS_STARTED;
+			break;
+		case '4':
+			ddata->status = MOTMDM_GNSS_STOPPED;
+			break;
+		default:
+			ddata->status = MOTMDM_GNSS_UNKNOWN;
+			break;
+		}
+		break;
+	case 'X':	/* UNNNN~+MPDXREQ=N for updated xtra2.bin needed */
+	default:
+		break;
+	}
+
+	return len;
+}
+
+static int motmdm_gnss_open(struct gnss_device *gdev)
+{
+	struct motmdm_gnss_data *ddata = gnss_get_drvdata(gdev);
+	struct gsm_serdev_dlci *dlci = &ddata->dlci;
+	int error;
+
+	dlci->drvdata = gdev;
+	dlci->receive_buf = motmdm_gnss_receive_data;
+
+	error = serdev_ngsm_register_dlci(ddata->modem, dlci);
+	if (error)
+		return error;
+
+	error = motmdm_gnss_init(gdev);
+	if (error) {
+		serdev_ngsm_unregister_dlci(ddata->modem, dlci);
+
+		return error;
+	}
+
+	return 0;
+}
+
+static void motmdm_gnss_close(struct gnss_device *gdev)
+{
+	struct motmdm_gnss_data *ddata = gnss_get_drvdata(gdev);
+	struct gsm_serdev_dlci *dlci = &ddata->dlci;
+	int error;
+
+	dlci->receive_buf = NULL;
+	error = motmdm_gnss_finish(gdev);
+	if (error < 0)
+		dev_warn(&gdev->dev, "%s: close failed: %i\n",
+			 __func__, error);
+
+	serdev_ngsm_unregister_dlci(ddata->modem, dlci);
+}
+
+static int motmdm_gnss_write_raw(struct gnss_device *gdev,
+				 const unsigned char *buf,
+				 size_t count)
+{
+	struct motmdm_gnss_data *ddata = gnss_get_drvdata(gdev);
+
+	return serdev_ngsm_write(ddata->modem, &ddata->dlci, buf, count);
+}
+
+static const struct gnss_operations motmdm_gnss_ops = {
+	.open		= motmdm_gnss_open,
+	.close		= motmdm_gnss_close,
+	.write_raw	= motmdm_gnss_write_raw,
+};
+
+static int motmdm_gnss_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct motmdm_gnss_data *ddata;
+	struct gnss_device *gdev;
+	u32 line;
+	int ret;
+
+	ddata = devm_kzalloc(dev, sizeof(*ddata), GFP_KERNEL);
+	if (!ddata)
+		return -ENOMEM;
+
+	ret = of_property_read_u32(dev->of_node, "reg", &line);
+	if (ret)
+		return ret;
+
+	if (!line)
+		return -EINVAL;
+
+	ddata->dlci.line = line;
+	ddata->modem = dev->parent;
+	ddata->len = PAGE_SIZE;
+	mutex_init(&ddata->mutex);
+	INIT_DELAYED_WORK(&ddata->restart_work, motmdm_gnss_restart);
+	init_waitqueue_head(&ddata->read_queue);
+
+	ddata->buf = devm_kzalloc(dev, ddata->len, GFP_KERNEL);
+	if (!ddata->buf)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, ddata);
+
+	gdev = gnss_allocate_device(dev);
+	if (!gdev)
+		return -ENOMEM;
+
+	gdev->type = GNSS_TYPE_NMEA;
+	gdev->ops = &motmdm_gnss_ops;
+	gnss_set_drvdata(gdev, ddata);
+	ddata->gdev = gdev;
+
+	ret = gnss_register_device(gdev);
+	if (ret)
+		goto err_put_device;
+
+	return 0;
+
+err_put_device:
+	gnss_put_device(ddata->gdev);
+
+	return ret;
+}
+
+static int motmdm_gnss_remove(struct platform_device *pdev)
+{
+	struct motmdm_gnss_data *data = platform_get_drvdata(pdev);
+
+	gnss_deregister_device(data->gdev);
+	gnss_put_device(data->gdev);
+
+	return 0;
+};
+
+#ifdef CONFIG_OF
+static const struct of_device_id motmdm_gnss_of_match[] = {
+	{ .compatible = "motorola,mapphone-mdm6600-gnss" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, motmdm_gnss_of_match);
+#endif
+
+static struct platform_driver motmdm_gnss_driver = {
+	.driver	= {
+		.name		= "gnss-mot-mdm6600",
+		.of_match_table	= of_match_ptr(motmdm_gnss_of_match),
+	},
+	.probe	= motmdm_gnss_probe,
+	.remove	= motmdm_gnss_remove,
+};
+module_platform_driver(motmdm_gnss_driver);
+
+MODULE_AUTHOR("Tony Lindgren <tony@atomide.com>");
+MODULE_DESCRIPTION("Motorola Mapphone MDM6600 GNSS receiver driver");
+MODULE_LICENSE("GPL v2");
-- 
2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

* [PATCH 6/6] ARM: dts: omap4-droid4: Configure modem for serdev-ngsm
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
                   ` (4 preceding siblings ...)
  2020-04-30 17:46 ` [PATCH 5/6] gnss: motmdm: Add support for Motorola Mapphone MDM6600 modem Tony Lindgren
@ 2020-04-30 17:46 ` Tony Lindgren
  2020-05-01 20:53   ` Pavel Machek
  2020-04-30 22:26 ` [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Pavel Machek
                   ` (3 subsequent siblings)
  9 siblings, 1 reply; 29+ messages in thread
From: Tony Lindgren @ 2020-04-30 17:46 UTC (permalink / raw)
  To: Greg Kroah-Hartman, Johan Hovold, Rob Herring
  Cc: Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap

Let's enable the TS 27.010 /dev/gsmmux* interfaces via Linux n_gsm that
can be used for voice calls and SMS with commands using a custom Motorola
format.

And let's also enable the kernel GNSS driver via serdev-ngsm that uses a
dedicated TS 27.010 channel.

Note that voice call audio mixer is not supported yet.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/boot/dts/motorola-mapphone-common.dtsi | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/arch/arm/boot/dts/motorola-mapphone-common.dtsi b/arch/arm/boot/dts/motorola-mapphone-common.dtsi
--- a/arch/arm/boot/dts/motorola-mapphone-common.dtsi
+++ b/arch/arm/boot/dts/motorola-mapphone-common.dtsi
@@ -702,6 +702,20 @@ &uart1 {
 			       &omap4_pmx_core 0x110>;
 	uart-has-rtscts;
 	current-speed = <115200>;
+
+	modem {
+		compatible = "motorola,mapphone-mdm6600-serial";
+		ttymask = <0 0x00001fee>;
+		phys = <&fsusb1_phy>;
+		phy-names = "usb";
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		gnss@4 {
+			compatible = "motorola,mapphone-mdm6600-gnss";
+			reg = <4>;
+		};
+	};
 };
 
 &uart3 {
-- 
2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
                   ` (5 preceding siblings ...)
  2020-04-30 17:46 ` [PATCH 6/6] ARM: dts: omap4-droid4: Configure modem for serdev-ngsm Tony Lindgren
@ 2020-04-30 22:26 ` Pavel Machek
  2020-05-01 14:52   ` Tony Lindgren
  2020-05-01  8:21 ` Pavel Machek
                   ` (2 subsequent siblings)
  9 siblings, 1 reply; 29+ messages in thread
From: Pavel Machek @ 2020-04-30 22:26 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 2841 bytes --]


> My guess is that at least with the pending ofono patches, we just
> want to use the raw interface for /dev/gsmtty* interface and stop
> pretending we have a modem that is AT compatible.

I tried to get it to work... it was not fun and I did not get far.

I pushed my results...

user@devuan:/my/ofono$ git push
Counting objects: 10, done.
Delta compression using up to 2 threads.
Compressing objects: 100% (10/10), done.
Writing objects: 100% (10/10), 1.17 KiB | 0 bytes/s, done.
Total 10 (delta 8), reused 0 (delta 0)
remote: Resolving deltas: 100% (8/8), completed with 8 local objects.
To github.com:pavelmachek/ofono.git
   fd34ca20..9042014b  mux-v1.29-1 -> mux-v1.29-1

Best regards,
								Pavel

ofonod[12922]: drivers/atmodem/network-registration.c:at_netreg_probe() Probing creg
Have command of length 9 (AT+CREG=?)
 or ^z 
ofonod[12922]: Voice: > U0000AT+FOO\r
ofonod[12922]: InSMS: > U0000AT+FOO\r
ofonod[12922]: OutSMS: > U0000AT+FOO\r
ofonod[12922]: Voice: < U0000+FOO:ERROR=9\n
new bytes 18
Have line: U0000+FOO:ERROR=9
Last character is 13
command response: U0000+FOO:ERROR=9
ofonod[12922]: plugins/motmdm.c:foo_cb() 
handle command response
ofonod[12922]: Voice: > U0000AT+SCRN=0\r
ofonod[12922]: OutSMS: < U0000+FOO:ERROR=9\n
new bytes 18
Have line: U0000+FOO:ERROR=9
Last character is 13
command response: U0000+FOO:ERROR=9
ofonod[12922]: plugins/motmdm.c:foo_cb() 
handle command response
ofonod[12922]: InSMS: < U0000+FOO:ERROR=9\n
new bytes 18
Have line: U0000+FOO:ERROR=9
Last character is 13
command response: U0000+FOO:ERROR=9
ofonod[12922]: plugins/motmdm.c:foo_cb() 
ofonod[12922]: plugins/motmdm.c:foo_cb() All channels working
handle command response
ofonod[12922]: Voice: < U0000+SCRN:OK\n
new bytes 14
Have line: U0000+SCRN:OK
Last character is 13
command response: U0000+SCRN:OK
ofonod[12922]: plugins/motmdm.c:scrn_cb() 
handle command response
ofonod[12922]: Voice: > U0000AT+CFUN=1\r
ofonod[12922]: Voice: < U0000+CFUN:OK\n
new bytes 14
Have line: U0000+CFUN:OK
Last character is 13
command response: U0000+CFUN:OK
ofonod[12922]: plugins/motmdm.c:cfun_cb() 
handle command response
ofonod[12922]: Voice: > U0000AT+CLIP=1\r
ofonod[12922]: Voice: < U0000+CLIP:OK\n
new bytes 14
Have line: U0000+CLIP:OK
Last character is 13
command response: U0000+CLIP:OK
handle command response
ofonod[12922]: Voice: > U0000AT+CCWA=1\r
ofonod[12922]: Voice: < U0000+CCWA:OK\n
new bytes 14
Have line: U0000+CCWA:OK
Last character is 13
command response: U0000+CCWA:OK
ofonod[12922]: drivers/motorolamodem/voicecall.c:motorola_voicecall_initialized() voicecall_init: registering to notifications



-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
                   ` (6 preceding siblings ...)
  2020-04-30 22:26 ` [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Pavel Machek
@ 2020-05-01  8:21 ` Pavel Machek
  2020-05-01 14:20   ` Tony Lindgren
  2020-05-01 20:19 ` Pavel Machek
  2020-05-01 20:23 ` Pavel Machek
  9 siblings, 1 reply; 29+ messages in thread
From: Pavel Machek @ 2020-05-01  8:21 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 1130 bytes --]

Hi!

> Now without the chardev support, the /dev/gsmtty* using apps need
> to use "U1234AT+CFUN?" format for the packets. The advantage is
> less kernel code, and we keep the existing /dev/gsmtty* interface.

Actually... yes, this works. But no, this is not "existing" tty
interface.

ttys work per character, and this interface definitely does not... it
is "packet" based, write() syscalls need exactly right lengths. You
can't just open minicom, and type "U1234...". You can't paste it,
either (I tried). tty controls like start/stop bits and baud rate are
useless here. CR/LF conversions are unwanted/dangerous because it is
confusing hard to debug if you get them wrong.

Now, I don't see reason why this could not be made to work, and it may
be more important to have something in mainline and work with that. So
if you can make this into -next, I'll not complain too loudly. But it
is... still wrong and I liked motmdm* more :-).

Best regards,

									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
  2020-05-01  8:21 ` Pavel Machek
@ 2020-05-01 14:20   ` Tony Lindgren
  0 siblings, 0 replies; 29+ messages in thread
From: Tony Lindgren @ 2020-05-01 14:20 UTC (permalink / raw)
  To: Pavel Machek
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap

* Pavel Machek <pavel@denx.de> [200501 08:22]:
> Hi!
> 
> > Now without the chardev support, the /dev/gsmtty* using apps need
> > to use "U1234AT+CFUN?" format for the packets. The advantage is
> > less kernel code, and we keep the existing /dev/gsmtty* interface.
> 
> Actually... yes, this works. But no, this is not "existing" tty
> interface.
> 
> ttys work per character, and this interface definitely does not... it
> is "packet" based, write() syscalls need exactly right lengths. You
> can't just open minicom, and type "U1234...". You can't paste it,
> either (I tried). tty controls like start/stop bits and baud rate are
> useless here. CR/LF conversions are unwanted/dangerous because it is
> confusing hard to debug if you get them wrong.

Yes.. That's what n_gsm spins up.

> Now, I don't see reason why this could not be made to work, and it may
> be more important to have something in mainline and work with that. So
> if you can make this into -next, I'll not complain too loudly. But it
> is... still wrong and I liked motmdm* more :-).

Yes.. There are issues too with the motmdm* char dev interface too.

I don't think it would work as is for devices with network interfaces
on some channel, and continuation packets could need more handling
possibly.

But let's try to get the basics sorted out first and use the "raw"
gsmtty* interface. More stuff can always be added later as needed.

Regards,

Tony




^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
  2020-04-30 22:26 ` [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Pavel Machek
@ 2020-05-01 14:52   ` Tony Lindgren
  2020-05-01 22:06     ` Pavel Machek
  0 siblings, 1 reply; 29+ messages in thread
From: Tony Lindgren @ 2020-05-01 14:52 UTC (permalink / raw)
  To: Pavel Machek, Stephen Boyd
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap

* Pavel Machek <pavel@denx.de> [200430 22:27]:
> 
> > My guess is that at least with the pending ofono patches, we just
> > want to use the raw interface for /dev/gsmtty* interface and stop
> > pretending we have a modem that is AT compatible.
> 
> I tried to get it to work... it was not fun and I did not get far.

OK. Yeah it's now 2020 and still dealing with serial port stuff :)

> I pushed my results...
> 
> user@devuan:/my/ofono$ git push
> Counting objects: 10, done.
> Delta compression using up to 2 threads.
> Compressing objects: 100% (10/10), done.
> Writing objects: 100% (10/10), 1.17 KiB | 0 bytes/s, done.
> Total 10 (delta 8), reused 0 (delta 0)
> remote: Resolving deltas: 100% (8/8), completed with 8 local objects.
> To github.com:pavelmachek/ofono.git
>    fd34ca20..9042014b  mux-v1.29-1 -> mux-v1.29-1

OK :) I still need to update the ALSA related patches on top
of this $subject series.

Also what I've noticed is that modprobe n_gsm debug=0xff hex output is
currently broken since commit 091cb0994edd ("lib/hexdump: make
print_hex_dump_bytes() a nop on !DEBUG builds"). Reverting the commit
fixes it.

Stephen, any ideas what should be changed to fix it?

Regards,

Tony

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
                   ` (7 preceding siblings ...)
  2020-05-01  8:21 ` Pavel Machek
@ 2020-05-01 20:19 ` Pavel Machek
  2020-05-01 20:32   ` Tony Lindgren
  2020-05-01 20:23 ` Pavel Machek
  9 siblings, 1 reply; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 20:19 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 635 bytes --]

Hi!

> My guess is that at least with the pending ofono patches, we just
> want to use the raw interface for /dev/gsmtty* interface and stop
> pretending we have a modem that is AT compatible.

Ok, so I got ofono back to work. ... I believe. It was not that
bad. SMS send/receive and outgoing call start/hangup worked at some
point (I did not play with mixers).

To github.com:pavelmachek/ofono.git
   61d3d727..195760e9  mux-v1.29-1 -> mux-v1.29-1

Best regards,
									Pavel

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
  2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
                   ` (8 preceding siblings ...)
  2020-05-01 20:19 ` Pavel Machek
@ 2020-05-01 20:23 ` Pavel Machek
  9 siblings, 0 replies; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 20:23 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 746 bytes --]

Hi!

> Now without the chardev support, the /dev/gsmtty* using apps need
> to use "U1234AT+CFUN?" format for the packets. The advantage is
> less kernel code, and we keep the existing /dev/gsmtty* interface.
> 
> If we still really need the custom chardev support, that can now
> be added as needed with the channel specific consumer driver(s).
> 
> My guess is that at least with the pending ofono patches, we just
> want to use the raw interface for /dev/gsmtty* interface and stop
> pretending we have a modem that is AT compatible.

The series:

Tested-by: Pavel Machek <pavel@ucw.cz>

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 1/6] tty: n_gsm: Add support for serdev drivers
  2020-04-30 17:46 ` [PATCH 1/6] tty: n_gsm: Add support for serdev drivers Tony Lindgren
@ 2020-05-01 20:31   ` Pavel Machek
  2020-05-01 21:31     ` Tony Lindgren
  0 siblings, 1 reply; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 20:31 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 1577 bytes --]

Hi!

> We can make use of serdev drivers to do simple device drivers for
> TS 27.010 chanels, and we can handle vendor specific protocols on top
> of TS 27.010 with serdev drivers.
> 
> So far this has been tested with Motorola droid4 where there is a custom
> packet numbering protocol on top of TS 27.010 for the MDM6600 modem.
> 
> I initially though about adding the serdev support into a separate file,
> but that will take some refactoring of n_gsm.c. And I'd like to have
> things working first. Then later on we might want to consider splitting
> n_gsm.c into three pieces for core, tty and serdev parts. And then maybe
> the serdev related parts can be just moved to live under something like
> drivers/tty/serdev/protocol/ngsm.c.
> 
> Signed-off-by: Tony Lindgren <tony@atomide.com>

If you fix error path below, you can add:

Reviewed-by: Pavel Machek <pavel@ucw.cz> 

> +static struct gsm_dlci *gsd_dlci_get(struct gsm_serdev *gsd, int line,
> +				     bool allocate)
> +{
> +	struct gsm_mux *gsm;
> +	struct gsm_dlci *dlci;
> +
> +	if (!gsd || !gsd->gsm)
> +		return ERR_PTR(-ENODEV);
> +
> +	gsm = gsd->gsm;
> +
> +	if (line < 1 || line >= 63)
> +		return ERR_PTR(-EINVAL);
> +
> +	mutex_lock(&gsm->mutex);
...
> +	dlci = gsm_dlci_alloc(gsm, line);
> +	if (!dlci) {
> +		gsm = ERR_PTR(-ENOMEM);
> +		goto unlock;

dlci = , or you get nice crash.

Best regards,
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
  2020-05-01 20:19 ` Pavel Machek
@ 2020-05-01 20:32   ` Tony Lindgren
  0 siblings, 0 replies; 29+ messages in thread
From: Tony Lindgren @ 2020-05-01 20:32 UTC (permalink / raw)
  To: Pavel Machek
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap

* Pavel Machek <pavel@denx.de> [200501 20:20]:
> Hi!
> 
> > My guess is that at least with the pending ofono patches, we just
> > want to use the raw interface for /dev/gsmtty* interface and stop
> > pretending we have a modem that is AT compatible.
> 
> Ok, so I got ofono back to work. ... I believe. It was not that
> bad. SMS send/receive and outgoing call start/hangup worked at some
> point (I did not play with mixers).
> 
> To github.com:pavelmachek/ofono.git
>    61d3d727..195760e9  mux-v1.29-1 -> mux-v1.29-1

OK good to hear and thanks for doing it.

Regards,

Tony

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 2/6] dt-bindings: serdev: ngsm: Add binding for serdev-ngsm
  2020-04-30 17:46 ` [PATCH 2/6] dt-bindings: serdev: ngsm: Add binding for serdev-ngsm Tony Lindgren
@ 2020-05-01 20:33   ` Pavel Machek
  2020-05-12 15:37   ` Rob Herring
  1 sibling, 0 replies; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 20:33 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 665 bytes --]

On Thu 2020-04-30 10:46:11, Tony Lindgren wrote:
> Add a binding document for a generic serdev-ngsm driver that can be
> used to bring up TS 27.010 line discipline with Linux n_gsm support
> on a serial port.
> 
> As the Motorola Mapphone modems require some custom handling, they
> are handled with a separate compatible.
> 
> Let's also add vendor string for ETSI as we're using a ETSI 3GPP
> TS 27.010 standard.
> 
> Signed-off-by: Tony Lindgren <tony@atomide.com>

Reviewed-by: Pavel Machek <pavel@ucw.cz>

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 3/6] serdev: ngsm: Add generic serdev-ngsm driver
  2020-04-30 17:46 ` [PATCH 3/6] serdev: ngsm: Add generic serdev-ngsm driver Tony Lindgren
@ 2020-05-01 20:44   ` Pavel Machek
  0 siblings, 0 replies; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 20:44 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 1501 bytes --]

Hi!

> We can have a generic serdev-ngsm driver bring up the TS 27.010 line
> discipline on the selected serial ports based on device tree data.
> 
> And we can now do standard Linux device driver for the dedicated
> TS 27.010 channels for devices like GNSS and ALSA found on modems.


> +++ b/drivers/tty/serdev/Kconfig
> @@ -22,4 +22,14 @@ config SERIAL_DEV_CTRL_TTYPORT
>  	depends on SERIAL_DEV_BUS != m
>  	default y
>  
> +config SERIAL_DEV_N_GSM
> +	tristate "Serial device TS 27.010 support"
> +	help
> +	  Select this if you want to use the TS 27.010 with a serial port with
> +	  devices such as modems and GNSS devices.
> +
> +	  If unsure, say N.
> +	depends on N_GSM
> +	depends on SERIAL_DEV_CTRL_TTYPORT

Normally, depends go after "tristate" and before "help" text. I did
not realize this is valid code.

Reviewed-by: Pavel Machek <pavel@ucw.cz>

> +/*
> + * Configure SoC 8250 device for 700 ms autosuspend delay, Values around 600 ms
> + * and shorter cause spurious wake-up events at least on droid 4. Also keep the

droid->Droid?

> +static const struct serdev_ngsm_cfg motmdm_cfg = {
> +	.gsm = &adaption1,
> +	.init_retry_quirk = true,
> +	.needs_usb_phy = true,
> +	.aggressive_pm = true,

Umm. These are unsigned int:1, not bools, so = 1 would be expected
here.

Best regards,
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS
  2020-04-30 17:46 ` [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS Tony Lindgren
@ 2020-05-01 20:45   ` Pavel Machek
  2020-05-01 21:14   ` Rob Herring
  2020-05-01 21:19   ` Rob Herring
  2 siblings, 0 replies; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 20:45 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 433 bytes --]

On Thu 2020-04-30 10:46:13, Tony Lindgren wrote:
> Add a binding document for Motorola Mapphone MDM6600 GNSS driver that
> can be used with phones such as droid4.

-> Droid 4.

> Signed-off-by: Tony Lindgren <tony@atomide.com>

Reviewed-by: Pavel Machek <pavel@ucw.cz>
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 5/6] gnss: motmdm: Add support for Motorola Mapphone MDM6600 modem
  2020-04-30 17:46 ` [PATCH 5/6] gnss: motmdm: Add support for Motorola Mapphone MDM6600 modem Tony Lindgren
@ 2020-05-01 20:51   ` Pavel Machek
  0 siblings, 0 replies; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 20:51 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 1047 bytes --]

On Thu 2020-04-30 10:46:14, Tony Lindgren wrote:
> Motorola is using a custom TS 27.010 based serial port line discipline
> for various devices on the modem. These devices can be accessed on
> dedicated channels using Linux kernel serdev-ngsm driver.
> 
> For the GNSS on these devices, we need to kick the GNSS device at a
> desired rate. Otherwise the GNSS device stops sending data after a
> few minutes. The rate we poll data defaults to 1000 ms, and can be
> specified with a module option rate_ms between 1 to 16 seconds.
> 
> Note that AGPS with xtra2.bin is not yet supported, so getting a fix
> can take quite a while. And a recent gpsd is needed to parse the
> $GNGNS output, and to properly handle the /dev/gnss0 character device.
> I've confirmed it works properly with gpsd-3.20.
> 
> Signed-off-by: Tony Lindgren <tony@atomide.com>

Reviewed-by: Pavel Machek <pavel@ucw.cz>

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 6/6] ARM: dts: omap4-droid4: Configure modem for serdev-ngsm
  2020-04-30 17:46 ` [PATCH 6/6] ARM: dts: omap4-droid4: Configure modem for serdev-ngsm Tony Lindgren
@ 2020-05-01 20:53   ` Pavel Machek
  0 siblings, 0 replies; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 20:53 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 624 bytes --]

On Thu 2020-04-30 10:46:15, Tony Lindgren wrote:
> Let's enable the TS 27.010 /dev/gsmmux* interfaces via Linux n_gsm that
> can be used for voice calls and SMS with commands using a custom Motorola
> format.
> 
> And let's also enable the kernel GNSS driver via serdev-ngsm that uses a
> dedicated TS 27.010 channel.
> 
> Note that voice call audio mixer is not supported yet.
> 
> Signed-off-by: Tony Lindgren <tony@atomide.com>

Acked-by: Pavel Machek <pavel@ucw.cz>

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS
  2020-04-30 17:46 ` [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS Tony Lindgren
  2020-05-01 20:45   ` Pavel Machek
@ 2020-05-01 21:14   ` Rob Herring
  2020-05-01 21:19   ` Rob Herring
  2 siblings, 0 replies; 29+ messages in thread
From: Rob Herring @ 2020-05-01 21:14 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Alan Cox, Lee Jones,
	Jiri Slaby, Merlijn Wajer, Pavel Machek, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap

On Thu, 30 Apr 2020 10:46:13 -0700, Tony Lindgren wrote:
> Add a binding document for Motorola Mapphone MDM6600 GNSS driver that
> can be used with phones such as droid4.
> 
> Signed-off-by: Tony Lindgren <tony@atomide.com>
> ---
>  .../devicetree/bindings/gnss/motmdm.yaml      | 29 +++++++++++++++++++
>  1 file changed, 29 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/gnss/motmdm.yaml
> 

My bot found errors running 'make dt_binding_check' on your patch:

Documentation/devicetree/bindings/gnss/motmdm.example.dts:19.11-21: Warning (reg_format): /example-0/gnss@4:reg: property has invalid length (4 bytes) (#address-cells == 1, #size-cells == 1)
Documentation/devicetree/bindings/gnss/motmdm.example.dt.yaml: Warning (pci_device_reg): Failed prerequisite 'reg_format'
Documentation/devicetree/bindings/gnss/motmdm.example.dt.yaml: Warning (pci_device_bus_num): Failed prerequisite 'reg_format'
Documentation/devicetree/bindings/gnss/motmdm.example.dt.yaml: Warning (simple_bus_reg): Failed prerequisite 'reg_format'
Documentation/devicetree/bindings/gnss/motmdm.example.dt.yaml: Warning (i2c_bus_reg): Failed prerequisite 'reg_format'
Documentation/devicetree/bindings/gnss/motmdm.example.dt.yaml: Warning (spi_bus_reg): Failed prerequisite 'reg_format'

See https://patchwork.ozlabs.org/patch/1280739

If you already ran 'make dt_binding_check' and didn't see the above
error(s), then make sure dt-schema is up to date:

pip3 install git+https://github.com/devicetree-org/dt-schema.git@master --upgrade

Please check and re-submit.

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS
  2020-04-30 17:46 ` [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS Tony Lindgren
  2020-05-01 20:45   ` Pavel Machek
  2020-05-01 21:14   ` Rob Herring
@ 2020-05-01 21:19   ` Rob Herring
  2020-05-01 21:25     ` Tony Lindgren
  2 siblings, 1 reply; 29+ messages in thread
From: Rob Herring @ 2020-05-01 21:19 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Alan Cox, Lee Jones,
	Jiri Slaby, Merlijn Wajer, Pavel Machek, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap

On Thu, Apr 30, 2020 at 10:46:13AM -0700, Tony Lindgren wrote:
> Add a binding document for Motorola Mapphone MDM6600 GNSS driver that
> can be used with phones such as droid4.
> 
> Signed-off-by: Tony Lindgren <tony@atomide.com>
> ---
>  .../devicetree/bindings/gnss/motmdm.yaml      | 29 +++++++++++++++++++
>  1 file changed, 29 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/gnss/motmdm.yaml
> 
> diff --git a/Documentation/devicetree/bindings/gnss/motmdm.yaml b/Documentation/devicetree/bindings/gnss/motmdm.yaml
> new file mode 100644
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/gnss/motmdm.yaml
> @@ -0,0 +1,29 @@
> +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/gnss/motmdm.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Motorola Mapphone MDM6600 GNSS binding
> +
> +maintainers:
> +  - Tony Lindgren <tony@atomide.com>
> +
> +properties:
> +  compatible:
> +    const: motorola,mapphone-mdm6600-gnss
> +
> +  reg:
> +    items:
> +      description: TS 27.010 channel the GNSS device is using
> +
> +required:
> +  - compatible
> +  - reg
> +
> +examples:
> +  - |
> +    gnss@4 {

This is a child of what's in patch 2? I think I'd just combine this 
schema into it.

Or you can leave it, but better to have complete examples (which will 
help fix what my bot reports). Examples have a default 
#address/size-cells of 1.

> +      compatible = "motorola,mapphone-mdm6600-gnss";
> +      reg = <4>;
> +    };
> -- 
> 2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS
  2020-05-01 21:19   ` Rob Herring
@ 2020-05-01 21:25     ` Tony Lindgren
  0 siblings, 0 replies; 29+ messages in thread
From: Tony Lindgren @ 2020-05-01 21:25 UTC (permalink / raw)
  To: Rob Herring
  Cc: Greg Kroah-Hartman, Johan Hovold, Alan Cox, Lee Jones,
	Jiri Slaby, Merlijn Wajer, Pavel Machek, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap

* Rob Herring <robh@kernel.org> [200501 21:20]:
> On Thu, Apr 30, 2020 at 10:46:13AM -0700, Tony Lindgren wrote:
> > Add a binding document for Motorola Mapphone MDM6600 GNSS driver that
> > can be used with phones such as droid4.
> > 
> > Signed-off-by: Tony Lindgren <tony@atomide.com>
> > ---
> >  .../devicetree/bindings/gnss/motmdm.yaml      | 29 +++++++++++++++++++
> >  1 file changed, 29 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/gnss/motmdm.yaml
> > 
> > diff --git a/Documentation/devicetree/bindings/gnss/motmdm.yaml b/Documentation/devicetree/bindings/gnss/motmdm.yaml
> > new file mode 100644
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/gnss/motmdm.yaml
> > @@ -0,0 +1,29 @@
> > +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
> > +%YAML 1.2
> > +---
> > +$id: http://devicetree.org/schemas/gnss/motmdm.yaml#
> > +$schema: http://devicetree.org/meta-schemas/core.yaml#
> > +
> > +title: Motorola Mapphone MDM6600 GNSS binding
> > +
> > +maintainers:
> > +  - Tony Lindgren <tony@atomide.com>
> > +
> > +properties:
> > +  compatible:
> > +    const: motorola,mapphone-mdm6600-gnss
> > +
> > +  reg:
> > +    items:
> > +      description: TS 27.010 channel the GNSS device is using
> > +
> > +required:
> > +  - compatible
> > +  - reg
> > +
> > +examples:
> > +  - |
> > +    gnss@4 {
> 
> This is a child of what's in patch 2? I think I'd just combine this 
> schema into it.

Yes it's a child node for patch 2 in this series. I'll combine them
into a single binding.

> Or you can leave it, but better to have complete examples (which will 
> help fix what my bot reports). Examples have a default 
> #address/size-cells of 1.

OK. In this case we have just #size-cells = <0>.

Regards,

Tony

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 1/6] tty: n_gsm: Add support for serdev drivers
  2020-05-01 20:31   ` Pavel Machek
@ 2020-05-01 21:31     ` Tony Lindgren
  0 siblings, 0 replies; 29+ messages in thread
From: Tony Lindgren @ 2020-05-01 21:31 UTC (permalink / raw)
  To: Pavel Machek
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap

* Pavel Machek <pavel@denx.de> [200501 20:32]:
> > +static struct gsm_dlci *gsd_dlci_get(struct gsm_serdev *gsd, int line,
> > +				     bool allocate)
> > +{
> > +	struct gsm_mux *gsm;
> > +	struct gsm_dlci *dlci;
> > +
> > +	if (!gsd || !gsd->gsm)
> > +		return ERR_PTR(-ENODEV);
> > +
> > +	gsm = gsd->gsm;
> > +
> > +	if (line < 1 || line >= 63)
> > +		return ERR_PTR(-EINVAL);
> > +
> > +	mutex_lock(&gsm->mutex);
> ...
> > +	dlci = gsm_dlci_alloc(gsm, line);
> > +	if (!dlci) {
> > +		gsm = ERR_PTR(-ENOMEM);
> > +		goto unlock;
> 
> dlci = , or you get nice crash.

Ah thanks yeah we return dlci and need to set dlci instead:

	if (!dlci) {
		dlci = ERR_PTR(-ENOMEM);
		goto unlock;
	}

Regards,

Tony

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4
  2020-05-01 14:52   ` Tony Lindgren
@ 2020-05-01 22:06     ` Pavel Machek
  0 siblings, 0 replies; 29+ messages in thread
From: Pavel Machek @ 2020-05-01 22:06 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Pavel Machek, Stephen Boyd, Greg Kroah-Hartman, Johan Hovold,
	Rob Herring, Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap


[-- Attachment #1: Type: text/plain, Size: 871 bytes --]

On Fri 2020-05-01 07:52:52, Tony Lindgren wrote:
> * Pavel Machek <pavel@denx.de> [200430 22:27]:
> > 
> > > My guess is that at least with the pending ofono patches, we just
> > > want to use the raw interface for /dev/gsmtty* interface and stop
> > > pretending we have a modem that is AT compatible.
> > 
> > I tried to get it to work... it was not fun and I did not get far.
> 
> OK. Yeah it's now 2020 and still dealing with serial port stuff :)

Yeah, and scary thing is... it is 2020 and serial port is _still_
complex and hard to understand and debug :-).

> OK :) I still need to update the ALSA related patches on top
> of this $subject series.

Let me know when you have these.

								Pavel
-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 2/6] dt-bindings: serdev: ngsm: Add binding for serdev-ngsm
  2020-04-30 17:46 ` [PATCH 2/6] dt-bindings: serdev: ngsm: Add binding for serdev-ngsm Tony Lindgren
  2020-05-01 20:33   ` Pavel Machek
@ 2020-05-12 15:37   ` Rob Herring
  1 sibling, 0 replies; 29+ messages in thread
From: Rob Herring @ 2020-05-12 15:37 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Pavel Machek, Peter Hurley, Alan Cox, Jiri Slaby, Merlijn Wajer,
	linux-serial, Greg Kroah-Hartman, devicetree, linux-omap,
	Lee Jones, Johan Hovold, Sebastian Reichel, linux-kernel

On Thu, 30 Apr 2020 10:46:11 -0700, Tony Lindgren wrote:
> Add a binding document for a generic serdev-ngsm driver that can be
> used to bring up TS 27.010 line discipline with Linux n_gsm support
> on a serial port.
> 
> As the Motorola Mapphone modems require some custom handling, they
> are handled with a separate compatible.
> 
> Let's also add vendor string for ETSI as we're using a ETSI 3GPP
> TS 27.010 standard.
> 
> Signed-off-by: Tony Lindgren <tony@atomide.com>
> ---
>  .../bindings/serdev/serdev-ngsm.yaml          | 64 +++++++++++++++++++
>  .../devicetree/bindings/vendor-prefixes.yaml  |  2 +
>  2 files changed, 66 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/serdev/serdev-ngsm.yaml
> 

Reviewed-by: Rob Herring <robh@kernel.org>

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 1/6] tty: n_gsm: Add support for serdev drivers
  2020-05-12 21:47 ` [PATCH 1/6] tty: n_gsm: Add support for serdev drivers Tony Lindgren
  2020-05-13 19:24   ` Pavel Machek
@ 2020-05-28  9:31   ` Johan Hovold
  1 sibling, 0 replies; 29+ messages in thread
From: Johan Hovold @ 2020-05-28  9:31 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap

On Tue, May 12, 2020 at 02:47:08PM -0700, Tony Lindgren wrote:
> We can make use of serdev drivers to do simple device drivers for
> TS 27.010 chanels, and we can handle vendor specific protocols on top
> of TS 27.010 with serdev drivers.
> 
> So far this has been tested with Motorola droid4 where there is a custom
> packet numbering protocol on top of TS 27.010 for the MDM6600 modem.
> 
> I initially though about adding the serdev support into a separate file,
> but that will take some refactoring of n_gsm.c. And I'd like to have
> things working first. Then later on we might want to consider splitting
> n_gsm.c into three pieces for core, tty and serdev parts. And then maybe
> the serdev related parts can be just moved to live under something like
> drivers/tty/serdev/protocol/ngsm.c.

Yeah, perhaps see where this lands first, but it should probably be done
before merging anything.

And it doesn't really make sense exporting these interfaces without the
actual serdev driver as they are closely tied and cannot be reviewed
separately anyway.

> Signed-off-by: Tony Lindgren <tony@atomide.com>
> ---
>  drivers/tty/n_gsm.c        | 435 +++++++++++++++++++++++++++++++++++++
>  include/linux/serdev-gsm.h | 154 +++++++++++++
>  2 files changed, 589 insertions(+)
>  create mode 100644 include/linux/serdev-gsm.h
> 
> diff --git a/drivers/tty/n_gsm.c b/drivers/tty/n_gsm.c
> --- a/drivers/tty/n_gsm.c
> +++ b/drivers/tty/n_gsm.c
> @@ -39,6 +39,7 @@
>  #include <linux/file.h>
>  #include <linux/uaccess.h>
>  #include <linux/module.h>
> +#include <linux/serdev.h>
>  #include <linux/timer.h>
>  #include <linux/tty_flip.h>
>  #include <linux/tty_driver.h>
> @@ -50,6 +51,7 @@
>  #include <linux/netdevice.h>
>  #include <linux/etherdevice.h>
>  #include <linux/gsmmux.h>
> +#include <linux/serdev-gsm.h>
>  
>  static int debug;
>  module_param(debug, int, 0600);
> @@ -150,6 +152,7 @@ struct gsm_dlci {
>  	/* Data handling callback */
>  	void (*data)(struct gsm_dlci *dlci, const u8 *data, int len);
>  	void (*prev_data)(struct gsm_dlci *dlci, const u8 *data, int len);
> +	struct gsm_serdev_dlci *ops; /* serdev dlci ops, if used */

Please rename the struct with a "_operations" suffix as you refer to
this as "ops" throughout.

>  	struct net_device *net; /* network interface, if created */
>  };
>  
> @@ -198,6 +201,7 @@ enum gsm_mux_state {
>   */
>  
>  struct gsm_mux {
> +	struct gsm_serdev *gsd;		/* Serial device bus data */
>  	struct tty_struct *tty;		/* The tty our ldisc is bound to */
>  	spinlock_t lock;
>  	struct mutex mutex;
> @@ -2346,6 +2350,437 @@ static int gsm_config(struct gsm_mux *gsm, struct gsm_config *c)
>  	return 0;
>  }
>  
> +#ifdef CONFIG_SERIAL_DEV_BUS
> +
> +/**
> + * gsm_serdev_get_config - read ts 27.010 config
> + * @gsd:	serdev-gsm instance
> + * @c:		ts 27.010 config data
> + *
> + * See gsm_copy_config_values() for more information.

Please document this properly since you're exporting these interfaces.

> + */
> +int gsm_serdev_get_config(struct gsm_serdev *gsd, struct gsm_config *c)
> +{
> +	struct gsm_mux *gsm;
> +
> +	if (!gsd || !gsd->gsm)
> +		return -ENODEV;

Is all this paranoia needed?

> +
> +	gsm = gsd->gsm;
> +
> +	if (!c)
> +		return -EINVAL;
> +
> +	gsm_copy_config_values(gsm, c);
> +
> +	return 0;
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_get_config);
> +
> +/**
> + * gsm_serdev_set_config - set ts 27.010 config
> + * @gsd:	serdev-gsm instance
> + * @c:		ts 27.010 config data
> + *
> + * See gsm_config() for more information.
> + */
> +int gsm_serdev_set_config(struct gsm_serdev *gsd, struct gsm_config *c)
> +{
> +	struct gsm_mux *gsm;
> +
> +	if (!gsd || !gsd->serdev || !gsd->gsm)
> +		return -ENODEV;

And why check for serdev here?

> +
> +	gsm = gsd->gsm;
> +
> +	if (!c)
> +		return -EINVAL;
> +
> +	return gsm_config(gsm, c);
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_set_config);
> +
> +static struct gsm_dlci *gsd_dlci_get(struct gsm_serdev *gsd, int line,
> +				     bool allocate)
> +{
> +	struct gsm_mux *gsm;
> +	struct gsm_dlci *dlci;
> +
> +	if (!gsd || !gsd->gsm)
> +		return ERR_PTR(-ENODEV);
> +
> +	gsm = gsd->gsm;
> +
> +	if (line < 1 || line >= 63)

Line 62 is reserved as well.

> +		return ERR_PTR(-EINVAL);
> +
> +	mutex_lock(&gsm->mutex);
> +
> +	if (gsm->dlci[line]) {
> +		dlci = gsm->dlci[line];
> +		goto unlock;
> +	} else if (!allocate) {
> +		dlci = ERR_PTR(-ENODEV);
> +		goto unlock;
> +	}
> +
> +	dlci = gsm_dlci_alloc(gsm, line);
> +	if (!dlci) {
> +		dlci = ERR_PTR(-ENOMEM);
> +		goto unlock;
> +	}
> +
> +	gsm->dlci[line] = dlci;
> +
> +unlock:
> +	mutex_unlock(&gsm->mutex);
> +
> +	return dlci;
> +}
> +
> +static int gsd_dlci_receive_buf(struct gsm_serdev_dlci *ops,
> +				const unsigned char *buf,
> +				size_t len)
> +{
> +	struct gsm_serdev *gsd = ops->gsd;

This looks backwards, why not pass in gsd instead?

> +	struct gsm_dlci *dlci;
> +	struct tty_port *port;
> +
> +	dlci = gsd_dlci_get(gsd, ops->line, false);
> +	if (IS_ERR(dlci))
> +		return PTR_ERR(dlci);
> +
> +	port = &dlci->port;
> +	tty_insert_flip_string(port, buf, len);
> +	tty_flip_buffer_push(port);
> +
> +	return len;
> +}
> +
> +static void gsd_dlci_data(struct gsm_dlci *dlci, const u8 *buf, int len)
> +{
> +	struct gsm_mux *gsm = dlci->gsm;
> +	struct gsm_serdev *gsd = gsm->gsd;
> +
> +	if (!gsd || !dlci->ops)
> +		return;
> +
> +	switch (dlci->adaption) {
> +	case 0:

0 isn't valid, right?

> +	case 1:
> +		if (dlci->ops->receive_buf)
> +			dlci->ops->receive_buf(dlci->ops, buf, len);
> +		break;

What about adaption 2 with modem status? Why are you not reusing
gsm_dlci_data()?

> +	default:
> +		pr_warn("dlci%i adaption %i not yet implemented\n",
> +			dlci->addr, dlci->adaption);

This needs to be rate limited. Use the dev_ versions when you can.

> +		break;
> +	}
> +}
> +
> +/**
> + * gsm_serdev_write - write data to a ts 27.010 channel
> + * @gsd:	serdev-gsm instance
> + * @ops:	channel ops
> + * @buf:	write buffer
> + * @len:	buffer length
> + */
> +int gsm_serdev_write(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops,
> +		     const u8 *buf, int len)
> +{
> +	struct gsm_mux *gsm;
> +	struct gsm_dlci *dlci;
> +	struct gsm_msg *msg;
> +	int h, size, total_size = 0;
> +	u8 *dp;
> +
> +	if (!gsd || !gsd->gsm)
> +		return -ENODEV;
> +
> +	gsm = gsd->gsm;
> +
> +	dlci = gsd_dlci_get(gsd, ops->line, false);
> +	if (IS_ERR(dlci))
> +		return PTR_ERR(dlci);
> +
> +	h = dlci->adaption - 1;
> +
> +	if (len > gsm->mtu)
> +		len = gsm->mtu;
> +
> +	size = len + h;
> +
> +	msg = gsm_data_alloc(gsm, dlci->addr, size, gsm->ftype);
> +	if (!msg)
> +		return -ENOMEM;
> +
> +	dp = msg->data;
> +	switch (dlci->adaption) {
> +	case 1:
> +		break;
> +	case 2:
> +		*dp++ = gsm_encode_modem(dlci);
> +		break;
> +	}
> +	memcpy(dp, buf, len);
> +	gsm_data_queue(dlci, msg);
> +	total_size += size;
> +
> +	return total_size;
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_write);
> +
> +/**
> + * gsm_serdev_data_kick - indicate more data can be transmitted
> + * @gsd:	serdev-gsm instance
> + *
> + * See gsm_data_kick() for more information.
> + */
> +void gsm_serdev_data_kick(struct gsm_serdev *gsd)
> +{
> +	struct gsm_mux *gsm;
> +	unsigned long flags;
> +
> +	if (!gsd || !gsd->gsm)
> +		return;
> +
> +	gsm = gsd->gsm;
> +
> +	spin_lock_irqsave(&gsm->tx_lock, flags);
> +	gsm_data_kick(gsm);
> +	spin_unlock_irqrestore(&gsm->tx_lock, flags);
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_data_kick);
> +
> +/**
> + * gsm_serdev_register_dlci - register a ts 27.010 channel
> + * @gsd:	serdev-gsm instance
> + * @ops:	channel ops
> + */
> +int gsm_serdev_register_dlci(struct gsm_serdev *gsd,
> +			     struct gsm_serdev_dlci *ops)
> +{
> +	struct gsm_dlci *dlci;
> +	struct gsm_mux *gsm;
> +	int retries;
> +
> +	if (!gsd || !gsd->gsm || !gsd->serdev)
> +		return -ENODEV;
> +
> +	gsm = gsd->gsm;
> +
> +	if (!ops || !ops->line)
> +		return -EINVAL;
> +
> +	dlci = gsd_dlci_get(gsd, ops->line, true);
> +	if (IS_ERR(dlci))
> +		return PTR_ERR(dlci);
> +
> +	if (dlci->state == DLCI_OPENING || dlci->state == DLCI_OPEN ||
> +	    dlci->state == DLCI_CLOSING)
> +		return -EBUSY;
> +
> +	mutex_lock(&dlci->mutex);
> +	ops->gsd = gsd;
> +	dlci->ops = ops;
> +	dlci->modem_rx = 0;
> +	dlci->prev_data = dlci->data;

I think this one is only used when bringing up a network interface.

> +	dlci->data = gsd_dlci_data;
> +	mutex_unlock(&dlci->mutex);
> +
> +	gsm_dlci_begin_open(dlci);

Why is this here? This should be handled when opening the serial device
(i.e. by gsmtty_open()).

> +
> +	/*
> +	 * Allow some time for dlci to move to DLCI_OPEN state. Otherwise
> +	 * the serdev consumer driver can start sending data too early during
> +	 * the setup, and the response will be missed by gms_queue() if we
> +	 * still have DLCI_CLOSED state.
> +	 */
> +	for (retries = 10; retries > 0; retries--) {
> +		if (dlci->state == DLCI_OPEN)
> +			break;
> +		msleep(100);
> +	}

What if you time out? This should be handled properly.

> +
> +	if (!retries)
> +		dev_dbg(&gsd->serdev->dev, "dlci%i not currently active\n",
> +			dlci->addr);
> +
> +	return 0;
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_register_dlci);
> +
> +/**
> + * gsm_serdev_unregister_dlci - unregister a ts 27.010 channel
> + * @gsd:	serdev-gsm instance
> + * @ops:	channel ops
> + */
> +void gsm_serdev_unregister_dlci(struct gsm_serdev *gsd,
> +				struct gsm_serdev_dlci *ops)
> +{
> +	struct gsm_mux *gsm;
> +	struct gsm_dlci *dlci;
> +
> +	if (!gsd || !gsd->gsm || !gsd->serdev)
> +		return;
> +
> +	gsm = gsd->gsm;
> +
> +	if (!ops || !ops->line)
> +		return;
> +
> +	dlci = gsd_dlci_get(gsd, ops->line, false);
> +	if (IS_ERR(dlci))
> +		return;
> +
> +	mutex_lock(&dlci->mutex);
> +	gsm_destroy_network(dlci);
> +	dlci->data = dlci->prev_data;
> +	dlci->ops->gsd = NULL;
> +	dlci->ops = NULL;
> +	mutex_unlock(&dlci->mutex);
> +
> +	gsm_dlci_begin_close(dlci);

This should all be handled when closing the serial device, that is, by
gsmtty_close().

> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_unregister_dlci);
> +
> +static int gsm_serdev_output(struct gsm_mux *gsm, u8 *data, int len)
> +{
> +	struct serdev_device *serdev = gsm->gsd->serdev;
> +
> +	if (gsm->gsd->output)
> +		return gsm->gsd->output(gsm->gsd, data, len);
> +	else
> +		return serdev_device_write_buf(serdev, data, len);
> +}

I guess this is needed to handle the motorola framing, but not easy to
tell currently due to the split.

> +
> +static int gsd_receive_buf(struct serdev_device *serdev, const u8 *data,
> +			   size_t count)
> +{
> +	struct gsm_serdev *gsd = serdev_device_get_drvdata(serdev);
> +	struct gsm_mux *gsm;
> +	const unsigned char *dp;
> +	int i;
> +
> +	if (WARN_ON(!gsd))
> +		return 0;

Probably better to take the NULL-deref. Can this ever happen?

> +
> +	gsm = gsd->gsm;
> +
> +	if (debug & 4)
> +		print_hex_dump_bytes("gsd_receive_buf: ",
> +				     DUMP_PREFIX_OFFSET,
> +				     data, count);
> +
> +	for (i = count, dp = data; i; i--, dp++)
> +		gsm->receive(gsm, *dp);
> +
> +	return count;
> +}
> +
> +static void gsd_write_wakeup(struct serdev_device *serdev)
> +{
> +	serdev_device_write_wakeup(serdev);
> +}
> +
> +static struct serdev_device_ops gsd_client_ops = {
> +	.receive_buf = gsd_receive_buf,
> +	.write_wakeup = gsd_write_wakeup,
> +};
> +
> +int gsm_serdev_register_tty_port(struct gsm_serdev *gsd, int line)
> +{
> +	struct gsm_serdev_dlci *ops;
> +	unsigned int base;
> +	int error;
> +
> +	if (line < 1)
> +		return -EINVAL;

Upper limit?

> +
> +	ops = kzalloc(sizeof(*ops), GFP_KERNEL);
> +	if (!ops)
> +		return -ENOMEM;
> +
> +	ops->line = line;
> +	ops->receive_buf = gsd_dlci_receive_buf;
> +
> +	error = gsm_serdev_register_dlci(gsd, ops);
> +	if (error) {
> +		kfree(ops);
> +
> +		return error;
> +	}
> +
> +	base = mux_num_to_base(gsd->gsm);
> +	tty_register_device(gsm_tty_driver, base + ops->line, NULL);

I would expect this to be tty_port_register_device_serdev() so that
serdev gets initialised properly for any client devices (e.g. gnss).

> +
> +	return 0;
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_register_tty_port);
> +
> +void gsm_serdev_unregister_tty_port(struct gsm_serdev *gsd, int line)
> +{
> +	struct gsm_dlci *dlci;
> +	unsigned int base;
> +
> +	if (line < 1)
> +		return;

As above.

> +
> +	dlci = gsd_dlci_get(gsd, line, false);
> +	if (IS_ERR(dlci))
> +		return;
> +
> +	base = mux_num_to_base(gsd->gsm);
> +	tty_unregister_device(gsm_tty_driver, base + line);
> +	gsm_serdev_unregister_dlci(gsd, dlci->ops);
> +	kfree(dlci->ops);
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_unregister_tty_port);
> +
> +int gsm_serdev_register_device(struct gsm_serdev *gsd)
> +{
> +	struct gsm_mux *gsm;
> +	int error;
> +
> +	if (WARN(!gsd || !gsd->serdev || !gsd->output,
> +		 "serdev and output must be initialized\n"))
> +		return -EINVAL;

Just oops if the driver is buggy and fails to set essential fields.

> +
> +	serdev_device_set_client_ops(gsd->serdev, &gsd_client_ops);
> +
> +	gsm = gsm_alloc_mux();
> +	if (!gsm)
> +		return -ENOMEM;
> +
> +	gsm->encoding = 1;
> +	gsm->tty = NULL;
> +	gsm->gsd = gsd;
> +	gsm->output = gsm_serdev_output;
> +	gsd->gsm = gsm;
> +	mux_get(gsd->gsm);
> +
> +	error = gsm_activate_mux(gsd->gsm);
> +	if (error) {
> +		gsm_cleanup_mux(gsd->gsm);
> +		mux_put(gsd->gsm);
> +
> +		return error;
> +	}
> +
> +	return 0;
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_register_device);
> +
> +void gsm_serdev_unregister_device(struct gsm_serdev *gsd)
> +{
> +	gsm_cleanup_mux(gsd->gsm);
> +	mux_put(gsd->gsm);
> +	gsd->gsm = NULL;
> +}
> +EXPORT_SYMBOL_GPL(gsm_serdev_unregister_device);
> +
> +#endif	/* CONFIG_SERIAL_DEV_BUS */

It looks like you may also have a problem with tty hangups, which serdev
does not support currently. There are multiple paths in n_gsm which can
trigger a hangup (e.g. based on remote input) and would likely lead to a
crash.

Johan

^ permalink raw reply	[flat|nested] 29+ messages in thread

* Re: [PATCH 1/6] tty: n_gsm: Add support for serdev drivers
  2020-05-12 21:47 ` [PATCH 1/6] tty: n_gsm: Add support for serdev drivers Tony Lindgren
@ 2020-05-13 19:24   ` Pavel Machek
  2020-05-28  9:31   ` Johan Hovold
  1 sibling, 0 replies; 29+ messages in thread
From: Pavel Machek @ 2020-05-13 19:24 UTC (permalink / raw)
  To: Tony Lindgren
  Cc: Greg Kroah-Hartman, Johan Hovold, Rob Herring, Alan Cox,
	Lee Jones, Jiri Slaby, Merlijn Wajer, Peter Hurley,
	Sebastian Reichel, linux-serial, devicetree, linux-kernel,
	linux-omap


[-- Attachment #1: Type: text/plain, Size: 1038 bytes --]

On Tue 2020-05-12 14:47:08, Tony Lindgren wrote:
> We can make use of serdev drivers to do simple device drivers for
> TS 27.010 chanels, and we can handle vendor specific protocols on top
> of TS 27.010 with serdev drivers.
> 
> So far this has been tested with Motorola droid4 where there is a custom
> packet numbering protocol on top of TS 27.010 for the MDM6600 modem.
> 
> I initially though about adding the serdev support into a separate file,
> but that will take some refactoring of n_gsm.c. And I'd like to have
> things working first. Then later on we might want to consider splitting
> n_gsm.c into three pieces for core, tty and serdev parts. And then maybe
> the serdev related parts can be just moved to live under something like
> drivers/tty/serdev/protocol/ngsm.c.
> 
> Signed-off-by: Tony Lindgren <tony@atomide.com>

Reviewed-by: Pavel Machek <pavel@ucw.cz>

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 29+ messages in thread

* [PATCH 1/6] tty: n_gsm: Add support for serdev drivers
  2020-05-12 21:47 [PATCHv8 " Tony Lindgren
@ 2020-05-12 21:47 ` Tony Lindgren
  2020-05-13 19:24   ` Pavel Machek
  2020-05-28  9:31   ` Johan Hovold
  0 siblings, 2 replies; 29+ messages in thread
From: Tony Lindgren @ 2020-05-12 21:47 UTC (permalink / raw)
  To: Greg Kroah-Hartman, Johan Hovold, Rob Herring
  Cc: Alan Cox, Lee Jones, Jiri Slaby, Merlijn Wajer, Pavel Machek,
	Peter Hurley, Sebastian Reichel, linux-serial, devicetree,
	linux-kernel, linux-omap

We can make use of serdev drivers to do simple device drivers for
TS 27.010 chanels, and we can handle vendor specific protocols on top
of TS 27.010 with serdev drivers.

So far this has been tested with Motorola droid4 where there is a custom
packet numbering protocol on top of TS 27.010 for the MDM6600 modem.

I initially though about adding the serdev support into a separate file,
but that will take some refactoring of n_gsm.c. And I'd like to have
things working first. Then later on we might want to consider splitting
n_gsm.c into three pieces for core, tty and serdev parts. And then maybe
the serdev related parts can be just moved to live under something like
drivers/tty/serdev/protocol/ngsm.c.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 drivers/tty/n_gsm.c        | 435 +++++++++++++++++++++++++++++++++++++
 include/linux/serdev-gsm.h | 154 +++++++++++++
 2 files changed, 589 insertions(+)
 create mode 100644 include/linux/serdev-gsm.h

diff --git a/drivers/tty/n_gsm.c b/drivers/tty/n_gsm.c
--- a/drivers/tty/n_gsm.c
+++ b/drivers/tty/n_gsm.c
@@ -39,6 +39,7 @@
 #include <linux/file.h>
 #include <linux/uaccess.h>
 #include <linux/module.h>
+#include <linux/serdev.h>
 #include <linux/timer.h>
 #include <linux/tty_flip.h>
 #include <linux/tty_driver.h>
@@ -50,6 +51,7 @@
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/gsmmux.h>
+#include <linux/serdev-gsm.h>
 
 static int debug;
 module_param(debug, int, 0600);
@@ -150,6 +152,7 @@ struct gsm_dlci {
 	/* Data handling callback */
 	void (*data)(struct gsm_dlci *dlci, const u8 *data, int len);
 	void (*prev_data)(struct gsm_dlci *dlci, const u8 *data, int len);
+	struct gsm_serdev_dlci *ops; /* serdev dlci ops, if used */
 	struct net_device *net; /* network interface, if created */
 };
 
@@ -198,6 +201,7 @@ enum gsm_mux_state {
  */
 
 struct gsm_mux {
+	struct gsm_serdev *gsd;		/* Serial device bus data */
 	struct tty_struct *tty;		/* The tty our ldisc is bound to */
 	spinlock_t lock;
 	struct mutex mutex;
@@ -2346,6 +2350,437 @@ static int gsm_config(struct gsm_mux *gsm, struct gsm_config *c)
 	return 0;
 }
 
+#ifdef CONFIG_SERIAL_DEV_BUS
+
+/**
+ * gsm_serdev_get_config - read ts 27.010 config
+ * @gsd:	serdev-gsm instance
+ * @c:		ts 27.010 config data
+ *
+ * See gsm_copy_config_values() for more information.
+ */
+int gsm_serdev_get_config(struct gsm_serdev *gsd, struct gsm_config *c)
+{
+	struct gsm_mux *gsm;
+
+	if (!gsd || !gsd->gsm)
+		return -ENODEV;
+
+	gsm = gsd->gsm;
+
+	if (!c)
+		return -EINVAL;
+
+	gsm_copy_config_values(gsm, c);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_get_config);
+
+/**
+ * gsm_serdev_set_config - set ts 27.010 config
+ * @gsd:	serdev-gsm instance
+ * @c:		ts 27.010 config data
+ *
+ * See gsm_config() for more information.
+ */
+int gsm_serdev_set_config(struct gsm_serdev *gsd, struct gsm_config *c)
+{
+	struct gsm_mux *gsm;
+
+	if (!gsd || !gsd->serdev || !gsd->gsm)
+		return -ENODEV;
+
+	gsm = gsd->gsm;
+
+	if (!c)
+		return -EINVAL;
+
+	return gsm_config(gsm, c);
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_set_config);
+
+static struct gsm_dlci *gsd_dlci_get(struct gsm_serdev *gsd, int line,
+				     bool allocate)
+{
+	struct gsm_mux *gsm;
+	struct gsm_dlci *dlci;
+
+	if (!gsd || !gsd->gsm)
+		return ERR_PTR(-ENODEV);
+
+	gsm = gsd->gsm;
+
+	if (line < 1 || line >= 63)
+		return ERR_PTR(-EINVAL);
+
+	mutex_lock(&gsm->mutex);
+
+	if (gsm->dlci[line]) {
+		dlci = gsm->dlci[line];
+		goto unlock;
+	} else if (!allocate) {
+		dlci = ERR_PTR(-ENODEV);
+		goto unlock;
+	}
+
+	dlci = gsm_dlci_alloc(gsm, line);
+	if (!dlci) {
+		dlci = ERR_PTR(-ENOMEM);
+		goto unlock;
+	}
+
+	gsm->dlci[line] = dlci;
+
+unlock:
+	mutex_unlock(&gsm->mutex);
+
+	return dlci;
+}
+
+static int gsd_dlci_receive_buf(struct gsm_serdev_dlci *ops,
+				const unsigned char *buf,
+				size_t len)
+{
+	struct gsm_serdev *gsd = ops->gsd;
+	struct gsm_dlci *dlci;
+	struct tty_port *port;
+
+	dlci = gsd_dlci_get(gsd, ops->line, false);
+	if (IS_ERR(dlci))
+		return PTR_ERR(dlci);
+
+	port = &dlci->port;
+	tty_insert_flip_string(port, buf, len);
+	tty_flip_buffer_push(port);
+
+	return len;
+}
+
+static void gsd_dlci_data(struct gsm_dlci *dlci, const u8 *buf, int len)
+{
+	struct gsm_mux *gsm = dlci->gsm;
+	struct gsm_serdev *gsd = gsm->gsd;
+
+	if (!gsd || !dlci->ops)
+		return;
+
+	switch (dlci->adaption) {
+	case 0:
+	case 1:
+		if (dlci->ops->receive_buf)
+			dlci->ops->receive_buf(dlci->ops, buf, len);
+		break;
+	default:
+		pr_warn("dlci%i adaption %i not yet implemented\n",
+			dlci->addr, dlci->adaption);
+		break;
+	}
+}
+
+/**
+ * gsm_serdev_write - write data to a ts 27.010 channel
+ * @gsd:	serdev-gsm instance
+ * @ops:	channel ops
+ * @buf:	write buffer
+ * @len:	buffer length
+ */
+int gsm_serdev_write(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops,
+		     const u8 *buf, int len)
+{
+	struct gsm_mux *gsm;
+	struct gsm_dlci *dlci;
+	struct gsm_msg *msg;
+	int h, size, total_size = 0;
+	u8 *dp;
+
+	if (!gsd || !gsd->gsm)
+		return -ENODEV;
+
+	gsm = gsd->gsm;
+
+	dlci = gsd_dlci_get(gsd, ops->line, false);
+	if (IS_ERR(dlci))
+		return PTR_ERR(dlci);
+
+	h = dlci->adaption - 1;
+
+	if (len > gsm->mtu)
+		len = gsm->mtu;
+
+	size = len + h;
+
+	msg = gsm_data_alloc(gsm, dlci->addr, size, gsm->ftype);
+	if (!msg)
+		return -ENOMEM;
+
+	dp = msg->data;
+	switch (dlci->adaption) {
+	case 1:
+		break;
+	case 2:
+		*dp++ = gsm_encode_modem(dlci);
+		break;
+	}
+	memcpy(dp, buf, len);
+	gsm_data_queue(dlci, msg);
+	total_size += size;
+
+	return total_size;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_write);
+
+/**
+ * gsm_serdev_data_kick - indicate more data can be transmitted
+ * @gsd:	serdev-gsm instance
+ *
+ * See gsm_data_kick() for more information.
+ */
+void gsm_serdev_data_kick(struct gsm_serdev *gsd)
+{
+	struct gsm_mux *gsm;
+	unsigned long flags;
+
+	if (!gsd || !gsd->gsm)
+		return;
+
+	gsm = gsd->gsm;
+
+	spin_lock_irqsave(&gsm->tx_lock, flags);
+	gsm_data_kick(gsm);
+	spin_unlock_irqrestore(&gsm->tx_lock, flags);
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_data_kick);
+
+/**
+ * gsm_serdev_register_dlci - register a ts 27.010 channel
+ * @gsd:	serdev-gsm instance
+ * @ops:	channel ops
+ */
+int gsm_serdev_register_dlci(struct gsm_serdev *gsd,
+			     struct gsm_serdev_dlci *ops)
+{
+	struct gsm_dlci *dlci;
+	struct gsm_mux *gsm;
+	int retries;
+
+	if (!gsd || !gsd->gsm || !gsd->serdev)
+		return -ENODEV;
+
+	gsm = gsd->gsm;
+
+	if (!ops || !ops->line)
+		return -EINVAL;
+
+	dlci = gsd_dlci_get(gsd, ops->line, true);
+	if (IS_ERR(dlci))
+		return PTR_ERR(dlci);
+
+	if (dlci->state == DLCI_OPENING || dlci->state == DLCI_OPEN ||
+	    dlci->state == DLCI_CLOSING)
+		return -EBUSY;
+
+	mutex_lock(&dlci->mutex);
+	ops->gsd = gsd;
+	dlci->ops = ops;
+	dlci->modem_rx = 0;
+	dlci->prev_data = dlci->data;
+	dlci->data = gsd_dlci_data;
+	mutex_unlock(&dlci->mutex);
+
+	gsm_dlci_begin_open(dlci);
+
+	/*
+	 * Allow some time for dlci to move to DLCI_OPEN state. Otherwise
+	 * the serdev consumer driver can start sending data too early during
+	 * the setup, and the response will be missed by gms_queue() if we
+	 * still have DLCI_CLOSED state.
+	 */
+	for (retries = 10; retries > 0; retries--) {
+		if (dlci->state == DLCI_OPEN)
+			break;
+		msleep(100);
+	}
+
+	if (!retries)
+		dev_dbg(&gsd->serdev->dev, "dlci%i not currently active\n",
+			dlci->addr);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_register_dlci);
+
+/**
+ * gsm_serdev_unregister_dlci - unregister a ts 27.010 channel
+ * @gsd:	serdev-gsm instance
+ * @ops:	channel ops
+ */
+void gsm_serdev_unregister_dlci(struct gsm_serdev *gsd,
+				struct gsm_serdev_dlci *ops)
+{
+	struct gsm_mux *gsm;
+	struct gsm_dlci *dlci;
+
+	if (!gsd || !gsd->gsm || !gsd->serdev)
+		return;
+
+	gsm = gsd->gsm;
+
+	if (!ops || !ops->line)
+		return;
+
+	dlci = gsd_dlci_get(gsd, ops->line, false);
+	if (IS_ERR(dlci))
+		return;
+
+	mutex_lock(&dlci->mutex);
+	gsm_destroy_network(dlci);
+	dlci->data = dlci->prev_data;
+	dlci->ops->gsd = NULL;
+	dlci->ops = NULL;
+	mutex_unlock(&dlci->mutex);
+
+	gsm_dlci_begin_close(dlci);
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_unregister_dlci);
+
+static int gsm_serdev_output(struct gsm_mux *gsm, u8 *data, int len)
+{
+	struct serdev_device *serdev = gsm->gsd->serdev;
+
+	if (gsm->gsd->output)
+		return gsm->gsd->output(gsm->gsd, data, len);
+	else
+		return serdev_device_write_buf(serdev, data, len);
+}
+
+static int gsd_receive_buf(struct serdev_device *serdev, const u8 *data,
+			   size_t count)
+{
+	struct gsm_serdev *gsd = serdev_device_get_drvdata(serdev);
+	struct gsm_mux *gsm;
+	const unsigned char *dp;
+	int i;
+
+	if (WARN_ON(!gsd))
+		return 0;
+
+	gsm = gsd->gsm;
+
+	if (debug & 4)
+		print_hex_dump_bytes("gsd_receive_buf: ",
+				     DUMP_PREFIX_OFFSET,
+				     data, count);
+
+	for (i = count, dp = data; i; i--, dp++)
+		gsm->receive(gsm, *dp);
+
+	return count;
+}
+
+static void gsd_write_wakeup(struct serdev_device *serdev)
+{
+	serdev_device_write_wakeup(serdev);
+}
+
+static struct serdev_device_ops gsd_client_ops = {
+	.receive_buf = gsd_receive_buf,
+	.write_wakeup = gsd_write_wakeup,
+};
+
+int gsm_serdev_register_tty_port(struct gsm_serdev *gsd, int line)
+{
+	struct gsm_serdev_dlci *ops;
+	unsigned int base;
+	int error;
+
+	if (line < 1)
+		return -EINVAL;
+
+	ops = kzalloc(sizeof(*ops), GFP_KERNEL);
+	if (!ops)
+		return -ENOMEM;
+
+	ops->line = line;
+	ops->receive_buf = gsd_dlci_receive_buf;
+
+	error = gsm_serdev_register_dlci(gsd, ops);
+	if (error) {
+		kfree(ops);
+
+		return error;
+	}
+
+	base = mux_num_to_base(gsd->gsm);
+	tty_register_device(gsm_tty_driver, base + ops->line, NULL);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_register_tty_port);
+
+void gsm_serdev_unregister_tty_port(struct gsm_serdev *gsd, int line)
+{
+	struct gsm_dlci *dlci;
+	unsigned int base;
+
+	if (line < 1)
+		return;
+
+	dlci = gsd_dlci_get(gsd, line, false);
+	if (IS_ERR(dlci))
+		return;
+
+	base = mux_num_to_base(gsd->gsm);
+	tty_unregister_device(gsm_tty_driver, base + line);
+	gsm_serdev_unregister_dlci(gsd, dlci->ops);
+	kfree(dlci->ops);
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_unregister_tty_port);
+
+int gsm_serdev_register_device(struct gsm_serdev *gsd)
+{
+	struct gsm_mux *gsm;
+	int error;
+
+	if (WARN(!gsd || !gsd->serdev || !gsd->output,
+		 "serdev and output must be initialized\n"))
+		return -EINVAL;
+
+	serdev_device_set_client_ops(gsd->serdev, &gsd_client_ops);
+
+	gsm = gsm_alloc_mux();
+	if (!gsm)
+		return -ENOMEM;
+
+	gsm->encoding = 1;
+	gsm->tty = NULL;
+	gsm->gsd = gsd;
+	gsm->output = gsm_serdev_output;
+	gsd->gsm = gsm;
+	mux_get(gsd->gsm);
+
+	error = gsm_activate_mux(gsd->gsm);
+	if (error) {
+		gsm_cleanup_mux(gsd->gsm);
+		mux_put(gsd->gsm);
+
+		return error;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_register_device);
+
+void gsm_serdev_unregister_device(struct gsm_serdev *gsd)
+{
+	gsm_cleanup_mux(gsd->gsm);
+	mux_put(gsd->gsm);
+	gsd->gsm = NULL;
+}
+EXPORT_SYMBOL_GPL(gsm_serdev_unregister_device);
+
+#endif	/* CONFIG_SERIAL_DEV_BUS */
+
 /**
  *	gsmld_output		-	write to link
  *	@gsm: our mux
diff --git a/include/linux/serdev-gsm.h b/include/linux/serdev-gsm.h
new file mode 100644
--- /dev/null
+++ b/include/linux/serdev-gsm.h
@@ -0,0 +1,154 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _LINUX_SERDEV_GSM_H
+#define _LINUX_SERDEV_GSM_H
+
+#include <linux/device.h>
+#include <linux/serdev.h>
+#include <linux/types.h>
+
+struct gsm_serdev_dlci;
+struct gsm_config;
+
+/**
+ * struct gsm_serdev - serdev-gsm instance
+ * @serdev:		serdev instance
+ * @gsm:		ts 27.010 n_gsm instance
+ * @drvdata:		serdev-gsm consumer driver data
+ * @output:		read data from ts 27.010 channel
+ *
+ * Currently only serdev and output must be initialized, the rest are
+ * are initialized by gsm_serdev_register_dlci().
+ */
+struct gsm_serdev {
+	struct serdev_device *serdev;
+	struct gsm_mux *gsm;
+	void *drvdata;
+	int (*output)(struct gsm_serdev *gsd, u8 *data, int len);
+};
+
+/**
+ * struct gsm_serdev_dlci - serdev-gsm ts 27.010 channel data
+ * @gsd:		serdev-gsm instance
+ * @line:		ts 27.010 channel, control channel 0 is not available
+ * @receive_buf:	function to handle data received for the channel
+ * @drvdata:		dlci specific consumer driver data
+ */
+struct gsm_serdev_dlci {
+	struct gsm_serdev *gsd;
+	int line;
+	int (*receive_buf)(struct gsm_serdev_dlci *ops,
+			   const unsigned char *buf,
+			   size_t len);
+	void *drvdata;
+};
+
+#if IS_ENABLED(CONFIG_N_GSM) && IS_ENABLED(CONFIG_SERIAL_DEV_BUS)
+
+extern int gsm_serdev_register_device(struct gsm_serdev *gsd);
+extern void gsm_serdev_unregister_device(struct gsm_serdev *gsd);
+extern int gsm_serdev_register_tty_port(struct gsm_serdev *gsd, int line);
+extern void gsm_serdev_unregister_tty_port(struct gsm_serdev *gsd, int line);
+
+static inline void *gsm_serdev_get_drvdata(struct device *dev)
+{
+	struct serdev_device *serdev = to_serdev_device(dev);
+	struct gsm_serdev *gsd = serdev_device_get_drvdata(serdev);
+
+	if (gsd)
+		return gsd->drvdata;
+
+	return NULL;
+}
+
+static inline void gsm_serdev_set_drvdata(struct device *dev, void *drvdata)
+{
+	struct serdev_device *serdev = to_serdev_device(dev);
+	struct gsm_serdev *gsd = serdev_device_get_drvdata(serdev);
+
+	if (gsd)
+		gsd->drvdata = drvdata;
+}
+
+extern int gsm_serdev_get_config(struct gsm_serdev *gsd, struct gsm_config *c);
+extern int gsm_serdev_set_config(struct gsm_serdev *gsd, struct gsm_config *c);
+extern int
+gsm_serdev_register_dlci(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops);
+extern void
+gsm_serdev_unregister_dlci(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops);
+extern int gsm_serdev_write(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops,
+			    const u8 *buf, int len);
+extern void gsm_serdev_data_kick(struct gsm_serdev *gsd);
+
+#else	/* CONFIG_SERIAL_DEV_BUS */
+
+static inline
+int gsm_serdev_register_device(struct gsm_serdev *gsd)
+{
+	return -ENODEV;
+}
+
+static inline void gsm_serdev_unregister_device(struct gsm_serdev *gsd)
+{
+}
+
+static inline int
+gsm_serdev_register_tty_port(struct gsm_serdev *gsd, int line)
+{
+	return -ENODEV;
+}
+
+static inline
+void gsm_serdev_unregister_tty_port(struct gsm_serdev *gsd, int line)
+{
+}
+
+static inline void *gsm_serdev_get_drvdata(struct device *dev)
+{
+	return NULL;
+}
+
+static inline
+void gsm_serdev_set_drvdata(struct device *dev, void *drvdata)
+{
+}
+
+static inline
+int gsm_serdev_get_config(struct gsm_serdev *gsd, struct gsm_config *c)
+{
+	return -ENODEV;
+}
+
+static inline
+int gsm_serdev_set_config(struct gsm_serdev *gsd, struct gsm_config *c)
+{
+	return -ENODEV;
+}
+
+static inline
+int gsm_serdev_register_dlci(struct gsm_serdev *gsd,
+			     struct gsm_serdev_dlci *ops)
+{
+	return -ENODEV;
+}
+
+static inline
+void gsm_serdev_unregister_dlci(struct gsm_serdev *gsd,
+				struct gsm_serdev_dlci *ops)
+{
+}
+
+static inline
+int gsm_serdev_write(struct gsm_serdev *gsd, struct gsm_serdev_dlci *ops,
+		     const u8 *buf, int len)
+{
+	return -ENODEV;
+}
+
+static inline
+void gsm_serdev_data_kick(struct gsm_serdev *gsd)
+{
+}
+
+#endif	/* CONFIG_N_GSM && CONFIG_SERIAL_DEV_BUS */
+#endif	/* _LINUX_SERDEV_GSM_H */
-- 
2.26.2

^ permalink raw reply	[flat|nested] 29+ messages in thread

end of thread, back to index

Thread overview: 29+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-04-30 17:46 [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Tony Lindgren
2020-04-30 17:46 ` [PATCH 1/6] tty: n_gsm: Add support for serdev drivers Tony Lindgren
2020-05-01 20:31   ` Pavel Machek
2020-05-01 21:31     ` Tony Lindgren
2020-04-30 17:46 ` [PATCH 2/6] dt-bindings: serdev: ngsm: Add binding for serdev-ngsm Tony Lindgren
2020-05-01 20:33   ` Pavel Machek
2020-05-12 15:37   ` Rob Herring
2020-04-30 17:46 ` [PATCH 3/6] serdev: ngsm: Add generic serdev-ngsm driver Tony Lindgren
2020-05-01 20:44   ` Pavel Machek
2020-04-30 17:46 ` [PATCH 4/6] dt-bindings: gnss: Add binding for Motorola Mapphone MDM6600 GNSS Tony Lindgren
2020-05-01 20:45   ` Pavel Machek
2020-05-01 21:14   ` Rob Herring
2020-05-01 21:19   ` Rob Herring
2020-05-01 21:25     ` Tony Lindgren
2020-04-30 17:46 ` [PATCH 5/6] gnss: motmdm: Add support for Motorola Mapphone MDM6600 modem Tony Lindgren
2020-05-01 20:51   ` Pavel Machek
2020-04-30 17:46 ` [PATCH 6/6] ARM: dts: omap4-droid4: Configure modem for serdev-ngsm Tony Lindgren
2020-05-01 20:53   ` Pavel Machek
2020-04-30 22:26 ` [PATCHv6 0/6] n_gsm serdev support and GNSS driver for droid4 Pavel Machek
2020-05-01 14:52   ` Tony Lindgren
2020-05-01 22:06     ` Pavel Machek
2020-05-01  8:21 ` Pavel Machek
2020-05-01 14:20   ` Tony Lindgren
2020-05-01 20:19 ` Pavel Machek
2020-05-01 20:32   ` Tony Lindgren
2020-05-01 20:23 ` Pavel Machek
2020-05-12 21:47 [PATCHv8 " Tony Lindgren
2020-05-12 21:47 ` [PATCH 1/6] tty: n_gsm: Add support for serdev drivers Tony Lindgren
2020-05-13 19:24   ` Pavel Machek
2020-05-28  9:31   ` Johan Hovold

Linux-OMAP Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-omap/0 linux-omap/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-omap linux-omap/ https://lore.kernel.org/linux-omap \
		linux-omap@vger.kernel.org
	public-inbox-index linux-omap

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-omap


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git