Skip to contentSkip to conversation

Add and make use of a common rs485 device tree parsing function

PATCH New Uwe Kleine-König 2 participants 14 messages Jul 18, 2017 – Jul 31, 2017

Subsystems (5)
Device tree bindings Arm i.MX and Layerscape TTY layer and serial drivers Arm/Microchip (AT91) SoC support +1 more subsystems
lore
Share & tools
lore

Uwe Kleine-König

author
#0 lore raw

v4 cover Add and make use of a common rs485 device tree parsing function

Hello,

this is v4 of the series. Compared to v3 It fixes a compile error and is rebased to a newer kernel to apply fine on Greg's tree.

Sascha Hauer (6):
serial: Add common rs485 device tree parsing function
serial: atmel: Use common rs485 device tree parsing function
serial: fsl_lpuart: Use common rs485 device tree parsing function
serial: omap-serial: Use common rs485 device tree parsing function
serial: imx: default to half duplex rs485
serial: imx: Use common rs485 device tree parsing function

Uwe Kleine-K?nig (3):
serial: fsl_lpuart: clear unsupported options in .rs485_config()
dt-bindings: serial/rs485: make rs485-rts-delay optional
dt-bindings: serial: document rs485 bindings for various devices

 .../devicetree/bindings/serial/atmel-usart.txt     |  1 +
 .../devicetree/bindings/serial/fsl-imx-uart.txt    |  1 +
 .../devicetree/bindings/serial/fsl-lpuart.txt      |  1 +
 .../devicetree/bindings/serial/omap_serial.txt     |  1 +
 Documentation/devicetree/bindings/serial/rs485.txt |  5 +--
 drivers/tty/serial/Kconfig                         |  4 ++
 drivers/tty/serial/Makefile                        |  2 +
 drivers/tty/serial/atmel_serial.c                  | 25 +-----------
 drivers/tty/serial/fsl_lpuart.c                    | 29 ++++++++++---
 drivers/tty/serial/imx.c                           |  5 ++-
 drivers/tty/serial/of.c                            | 47 ++++++++++++++++++++++
 drivers/tty/serial/omap-serial.c                   | 13 +-----
 include/linux/serial_core.h                        | 13 ++++++
 13 files changed, 100 insertions(+), 47 deletions(-)
 create mode 100644 drivers/tty/serial/of.c
Signature · 1 line
2.11.0
author
#1 lore raw

PATCH v4 1/9 serial: fsl_lpuart: clear unsupported options in .rs485_config()

The struct serial_rs485 parameter is both input and output and is supposed to hold the actually used configuration on return. So clear unsupported settings.

Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/fsl_lpuart.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/tty/serial/fsl_lpuart.c +5 -0
index 343de8c384b0..4becfbf35c4c 100644
@@ -1019,6 +1019,11 @@ static int lpuart_config_rs485(struct uart_port *port,
 		~(UARTMODEM_TXRTSPOL | UARTMODEM_TXRTSE);
 	writeb(modem, sport->port.membase + UARTMODEM);
 
+	/* clear unsupported configurations */
+	rs485->delay_rts_before_send = 0;
+	rs485->delay_rts_after_send = 0;
+	rs485->flags &= ~SER_RS485_RX_DURING_TX;
+
 	if (rs485->flags & SER_RS485_ENABLED) {
 		/* Enable auto RS-485 RTS mode */
 		modem |= UARTMODEM_TXRTSE;
Signature · 1 line
2.11.0
author
#2 lore raw

PATCH v4 2/9 dt-bindings: serial/rs485: make rs485-rts-delay optional

In reply to Uwe Kleine-König

There are a few device trees that specify one of the already optional properties without also having the up to now required property rs485-rts-delay. Additionally there is no technical reason to require rs485-rts-delay and that's also what most drivers implement.

So give existing users and implementers a blessing and document rs485-rts-delay as optional.

Signed-off-by Uwe Kleine-K?nig <[email protected]>
Acked-by Rob Herring <[email protected]>
---
 Documentation/devicetree/bindings/serial/rs485.txt | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
Documentation/devicetree/bindings/serial/rs485.txt +2 -3
index 32b1fa1f2a5b..b8415936dfdb 100644
@@ -5,14 +5,13 @@ the built-in half-duplex mode.
 The properties described hereafter shall be given to a half-duplex capable
 UART node.
 
-Required properties:
+Optional properties:
 - rs485-rts-delay: prop-encoded-array <a b> where:
   * a is the delay between rts signal and beginning of data sent in milliseconds.
       it corresponds to the delay before sending data.
   * b is the delay between end of data sent and rts signal in milliseconds
       it corresponds to the delay after sending data and actual release of the line.
-
-Optional properties:
+  If this property is not specified, <0 0> is assumed.
 - linux,rs485-enabled-at-boot-time: empty property telling to enable the rs485
   feature at boot time. It can be disabled later with proper ioctl.
 - rs485-rx-during-tx: empty property that enables the receiving of data even
Signature · 1 line
2.11.0
author
#3 lore raw

PATCH v4 3/9 serial: Add common rs485 device tree parsing function

In reply to Uwe Kleine-König

From: Sascha Hauer <[email protected]>

Several drivers have the same device tree parsing code. Create a common helper function for it.

Signed-off-by Sascha Hauer <[email protected]>
[ukl: implement default <0 0> for rts-delay, unset unspecified flags]
Acked-by Nicolas Ferre <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/Kconfig  |  4 ++++
 drivers/tty/serial/Makefile |  2 ++
 drivers/tty/serial/of.c     | 47 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/serial_core.h | 13 +++++++++++++
 4 files changed, 66 insertions(+)
 create mode 100644 drivers/tty/serial/of.c
drivers/tty/serial/Kconfig +4 -0
index 1f096e2bb398..f7d7e4936940 100644
@@ -14,6 +14,10 @@ config SERIAL_EARLYCON
 	  the console before standard serial driver is probed. The console is
 	  enabled when early_param is processed.
 
+config OF_SERIAL
+	depends on SERIAL_CORE
+	def_bool y
+
 source "drivers/tty/serial/8250/Kconfig"
 
 comment "Non-8250 serial port support"
drivers/tty/serial/Makefile +2 -0
index fe88a75d9a59..a755faa25485 100644
@@ -7,6 +7,8 @@ obj-$(CONFIG_SERIAL_CORE) += serial_core.o
 obj-$(CONFIG_SERIAL_EARLYCON) += earlycon.o
 obj-$(CONFIG_SERIAL_EARLYCON_ARM_SEMIHOST) += earlycon-arm-semihost.o
 
+obj-$(CONFIG_OF_SERIAL) += of.o
+
 # These Sparc drivers have to appear before others such as 8250
 # which share ttySx minor node space.  Otherwise console device
 # names change and other unplesantries.
drivers/tty/serial/of.c new file +47 -0

47 lines not on this page · raw message

include/linux/serial_core.h +13 -0
index 1775500294bb..3a89e8925722 100644
@@ -501,4 +501,17 @@ static inline int uart_handle_break(struct uart_port *port)
 					 (cflag) & CRTSCTS || \
 					 !((cflag) & CLOCAL))
 
+/*
+ * Common device tree parsing helpers
+ */
+#ifdef CONFIG_OF_SERIAL
+int of_get_rs485_mode(struct device_node *np, struct serial_rs485 *rs485conf);
+#else
+static inline int of_get_rs485_mode(struct device_node *np,
+				    struct serial_rs485 *rs485conf)
+{
+	return 1;
+}
+#endif
+
 #endif /* LINUX_SERIAL_CORE_H */
Signature · 1 line
2.11.0
#4 lore raw

PATCH v4 3/9 serial: Add common rs485 device tree parsing function

Quoted 5 lines · Uwe Kleine-K?nig

From: Sascha Hauer <[email protected]>

Several drivers have the same device tree parsing code. Create a common helper function for it.

Quoted 8 lines
Signed-off-by Sascha Hauer <[email protected]>
[ukl: implement default <0 0> for rts-delay, unset unspecified flags]
Acked-by Nicolas Ferre <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/Kconfig  |  4 ++++
 drivers/tty/serial/Makefile |  2 ++
 drivers/tty/serial/of.c     | 47 +++++++++++++++++++++++++++++++++++++++++++++

Why is this a separate file, with a Kconfig option that is always enabled? Why not just add the single function to the serial core?

Quoted 9 lines

+/**
+ * of_get_rs485_mode() - Implement parsing rs485 properties
+ * @np: uart node
+ * @rs485conf: output parameter
+ *
+ * This function implements the device tree binding described in
+ * Documentation/devicetree/bindings/serial/rs485.txt.
+ *
+ * Return: 0 on success, 1 if the node doesn't contain rs485 stuff.

1 is not an error code :(

Return a proper error please.

Quoted 41 lines

+ */

drivers/tty/serial/of.c new file +32 -0
+int of_get_rs485_mode(struct device_node *np, struct serial_rs485 *rs485conf)
+{
+	u32 rs485_delay[2];
+	int ret;
+
+	if (!IS_ENABLED(CONFIG_OF) || !np)
+		return 1;
+
+	ret = of_property_read_u32_array(np, "rs485-rts-delay", rs485_delay, 2);
+	if (!ret) {
+		rs485conf->delay_rts_before_send = rs485_delay[0];
+		rs485conf->delay_rts_after_send = rs485_delay[1];
+	} else {
+		rs485conf->delay_rts_before_send = 0;
+		rs485conf->delay_rts_after_send = 0;
+	}
+
+	/*
+	 * clear full-duplex and enabled flags to get to a defined state with
+	 * the two following properties.
+	 */
+	rs485conf->flags &= ~(SER_RS485_RX_DURING_TX | SER_RS485_ENABLED);
+
+	if (of_property_read_bool(np, "rs485-rx-during-tx"))
+		rs485conf->flags |= SER_RS485_RX_DURING_TX;
+
+	if (of_property_read_bool(np, "linux,rs485-enabled-at-boot-time"))
+		rs485conf->flags |= SER_RS485_ENABLED;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(of_get_rs485_mode);
include/linux/serial_core.h +0 -0
index 1775500294bb..3a89e8925722 100644
@@ -501,4 +501,17 @@ static inline int uart_handle_break(struct uart_port *port)
 					 (cflag) & CRTSCTS || \
 					 !((cflag) & CLOCAL))

include/linux/serial_core.h static inline int uart_handle_break(struct uart_port *port) · line 513

Quoted 10 lines
include/linux/serial_core.h +10 -0
+/*
+ * Common device tree parsing helpers
+ */
+#ifdef CONFIG_OF_SERIAL
+int of_get_rs485_mode(struct device_node *np, struct serial_rs485 *rs485conf);
+#else
+static inline int of_get_rs485_mode(struct device_node *np,
+				    struct serial_rs485 *rs485conf)
+{
+	return 1;

Again, a real error.

And why not have a "generic" firmware function for this that defaults to the of_ for that platform type if present? What's the odds that acpi will need/want this soon anyway?

thanks,

greg k-h

author
#5 lore raw

PATCH v4 3/9 serial: Add common rs485 device tree parsing function

Quoted 15 lines · Greg KH
Quoted 5 lines · Uwe Kleine-K?nig

From: Sascha Hauer <[email protected]>

Several drivers have the same device tree parsing code. Create a common helper function for it.

Quoted 8 lines
Signed-off-by Sascha Hauer <[email protected]>
[ukl: implement default <0 0> for rts-delay, unset unspecified flags]
Acked-by Nicolas Ferre <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/Kconfig  |  4 ++++
 drivers/tty/serial/Makefile |  2 ++
 drivers/tty/serial/of.c     | 47 +++++++++++++++++++++++++++++++++++++++++++++
Quoted 2 lines

Why is this a separate file, with a Kconfig option that is always enabled? Why not just add the single function to the serial core?

Probably just a matter of taste. If you prefer I can put it into drivers/tty/serial/serial_core.c

Quoted 10 lines
Quoted 9 lines

+/**
+ * of_get_rs485_mode() - Implement parsing rs485 properties
+ * @np: uart node
+ * @rs485conf: output parameter
+ *
+ * This function implements the device tree binding described in
+ * Documentation/devicetree/bindings/serial/rs485.txt.
+ *
+ * Return: 0 on success, 1 if the node doesn't contain rs485 stuff.

Quoted 3 lines

1 is not an error code :(

Return a proper error please.

Well, it's not an error. These properties are optional and if I return (say) -ENOENT in this case, a caller looks as follows:

	ret = of_get_rs485_mode(...);
	if (ret == -ENOENT) {
		/*
		 * hmm, does this mean there are no rs485 properties, or
		 * is this a different error that I should propagate
		 * because there is a problem with the device tree?
		 * I'm optimistic and assume the former.
		 * As of_get_rs485_mode already did everything
		 * necessary, there is no need to do anything else.
		 */
	} else if (ret < 0) {
		return ret;
	}

With the semantic I chose it is just:

	ret = of_get_rs485_mode(...);
	if (ret < 0)
		return ret;

without any guessing.

Quoted 41 lines
Quoted 41 lines

+ */

drivers/tty/serial/of.c new file +32 -0
+int of_get_rs485_mode(struct device_node *np, struct serial_rs485 *rs485conf)
+{
+	u32 rs485_delay[2];
+	int ret;
+
+	if (!IS_ENABLED(CONFIG_OF) || !np)
+		return 1;
+
+	ret = of_property_read_u32_array(np, "rs485-rts-delay", rs485_delay, 2);
+	if (!ret) {
+		rs485conf->delay_rts_before_send = rs485_delay[0];
+		rs485conf->delay_rts_after_send = rs485_delay[1];
+	} else {
+		rs485conf->delay_rts_before_send = 0;
+		rs485conf->delay_rts_after_send = 0;
+	}
+
+	/*
+	 * clear full-duplex and enabled flags to get to a defined state with
+	 * the two following properties.
+	 */
+	rs485conf->flags &= ~(SER_RS485_RX_DURING_TX | SER_RS485_ENABLED);
+
+	if (of_property_read_bool(np, "rs485-rx-during-tx"))
+		rs485conf->flags |= SER_RS485_RX_DURING_TX;
+
+	if (of_property_read_bool(np, "linux,rs485-enabled-at-boot-time"))
+		rs485conf->flags |= SER_RS485_ENABLED;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(of_get_rs485_mode);
include/linux/serial_core.h +0 -0
index 1775500294bb..3a89e8925722 100644
@@ -501,4 +501,17 @@ static inline int uart_handle_break(struct uart_port *port)
 					 (cflag) & CRTSCTS || \
 					 !((cflag) & CLOCAL))
Quoted 12 lines

include/linux/serial_core.h static inline int uart_handle_break(struct uart_port *port) · line 513

Quoted 10 lines
include/linux/serial_core.h +10 -0
+/*
+ * Common device tree parsing helpers
+ */
+#ifdef CONFIG_OF_SERIAL
+int of_get_rs485_mode(struct device_node *np, struct serial_rs485 *rs485conf);
+#else
+static inline int of_get_rs485_mode(struct device_node *np,
+				    struct serial_rs485 *rs485conf)
+{
+	return 1;

Again, a real error.

OK, I could agree to return -ENOSYS here. I predict I get complaints about this though and people will do:

	ret = of_get_rs485_mode(...);
	if (ret == -ENOSYS) {
		/* ignore */
	} else if (ret < 0) {
		return ret;
	}

which one might consider troublesome or even wrong. (See the same discussions about 22c403676dbbb7c6f186099527af7f065498ef45 where I wanted to keep -ENOSYS.)

Quoted 3 lines

And why not have a "generic" firmware function for this that defaults to the of_ for that platform type if present? What's the odds that acpi will need/want this soon anyway?

I don't know about ACPI and all systems I care about are using dt. So I'd suggest we stick to dt and switch to "generic" if and when the need arises?

Best regards
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-K?nig            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
author
#6 lore raw

PATCH v4 4/9 serial: atmel: Use common rs485 device tree parsing function

In reply to Uwe Kleine-König

From: Sascha Hauer <[email protected]>

We just got a common helper for parsing the rs485 specific device tree properties. Use it and drop the open coded parser.

Note that there is a small difference between the removed and the now used implementation: The former cleared flags to 0 if rs485-rts-delay was given, the common helper clears SER_RS485_RX_DURING_TX and SER_RS485_ENABLED only but always which makes more sense.

Signed-off-by Sascha Hauer <[email protected]>
[ukleinek: point out semantic change in commit log]
Acked-by Nicolas Ferre <[email protected]>
Acked-by Richard Genoud <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/atmel_serial.c | 25 +------------------------
 1 file changed, 1 insertion(+), 24 deletions(-)
drivers/tty/serial/atmel_serial.c +1 -24
index 7551cab438ff..04e55f010894 100644
@@ -1667,29 +1667,6 @@ static void atmel_init_property(struct atmel_uart_port *atmel_port,
 	}
 }
 
-static void atmel_init_rs485(struct uart_port *port,
-				struct platform_device *pdev)
-{
-	struct device_node *np = pdev->dev.of_node;
-
-	struct serial_rs485 *rs485conf = &port->rs485;
-	u32 rs485_delay[2];
-
-	/* rs485 properties */
-	if (of_property_read_u32_array(np, "rs485-rts-delay",
-				       rs485_delay, 2) == 0) {
-		rs485conf->delay_rts_before_send = rs485_delay[0];
-		rs485conf->delay_rts_after_send = rs485_delay[1];
-		rs485conf->flags = 0;
-	}
-
-	if (of_get_property(np, "rs485-rx-during-tx", NULL))
-		rs485conf->flags |= SER_RS485_RX_DURING_TX;
-
-	if (of_get_property(np, "linux,rs485-enabled-at-boot-time", NULL))
-		rs485conf->flags |= SER_RS485_ENABLED;
-}
-
 static void atmel_set_ops(struct uart_port *port)
 {
 	struct atmel_uart_port *atmel_port = to_atmel_uart_port(port);

@@ -2373,7 +2350,7 @@ static int atmel_init_port(struct atmel_uart_port *atmel_port,
 	atmel_init_property(atmel_port, pdev);
 	atmel_set_ops(port);
 
-	atmel_init_rs485(port, pdev);
+	of_get_rs485_mode(pdev->dev.of_node, &port->rs485);
 
 	port->iotype		= UPIO_MEM;
 	port->flags		= UPF_BOOT_AUTOCONF | UPF_IOREMAP;
Signature · 1 line
2.11.0
#7 lore raw

PATCH v4 4/9 serial: atmel: Use common rs485 device tree parsing function

Quoted 54 lines · Uwe Kleine-K?nig

From: Sascha Hauer <[email protected]>

We just got a common helper for parsing the rs485 specific device tree properties. Use it and drop the open coded parser.

Note that there is a small difference between the removed and the now used implementation: The former cleared flags to 0 if rs485-rts-delay was given, the common helper clears SER_RS485_RX_DURING_TX and SER_RS485_ENABLED only but always which makes more sense.

Signed-off-by Sascha Hauer <[email protected]>
[ukleinek: point out semantic change in commit log]
Acked-by Nicolas Ferre <[email protected]>
Acked-by Richard Genoud <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/atmel_serial.c | 25 +------------------------
 1 file changed, 1 insertion(+), 24 deletions(-)
drivers/tty/serial/atmel_serial.c +0 -23
index 7551cab438ff..04e55f010894 100644
@@ -1667,29 +1667,6 @@ static void atmel_init_property(struct atmel_uart_port *atmel_port,
 	}
 }
 
-static void atmel_init_rs485(struct uart_port *port,
-				struct platform_device *pdev)
-{
-	struct device_node *np = pdev->dev.of_node;
-
-	struct serial_rs485 *rs485conf = &port->rs485;
-	u32 rs485_delay[2];
-
-	/* rs485 properties */
-	if (of_property_read_u32_array(np, "rs485-rts-delay",
-				       rs485_delay, 2) == 0) {
-		rs485conf->delay_rts_before_send = rs485_delay[0];
-		rs485conf->delay_rts_after_send = rs485_delay[1];
-		rs485conf->flags = 0;
-	}
-
-	if (of_get_property(np, "rs485-rx-during-tx", NULL))
-		rs485conf->flags |= SER_RS485_RX_DURING_TX;
-
-	if (of_get_property(np, "linux,rs485-enabled-at-boot-time", NULL))
-		rs485conf->flags |= SER_RS485_ENABLED;
-}
-
 static void atmel_set_ops(struct uart_port *port)
 {
 	struct atmel_uart_port *atmel_port = to_atmel_uart_port(port);

drivers/tty/serial/atmel_serial.c static int atmel_init_port(struct atmel_uart_port *atmel_port, · line 2353

Quoted 5 lines
drivers/tty/serial/atmel_serial.c +1 -1
@@ -2373,7 +2350,7 @@ static int atmel_init_port(struct atmel_uart_port *atmel_port,
 	atmel_init_property(atmel_port, pdev);
 	atmel_set_ops(port);
 
-	atmel_init_rs485(port, pdev);
+	of_get_rs485_mode(pdev->dev.of_node, &port->rs485);

No error checking?

author
#8 lore raw

PATCH v4 5/9 serial: fsl_lpuart: Use common rs485 device tree parsing function

In reply to Uwe Kleine-König

From: Sascha Hauer <[email protected]>

We just got a common helper for parsing the rs485 specific device tree properties. Use it and drop the open coded parser.

As a side effect this adds support for parsing rs485-rts-delay and rs485-rx-during-tx. As the driver doesn't support this though, probing fails if these are defined.

Signed-off-by Sascha Hauer <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/fsl_lpuart.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)
drivers/tty/serial/fsl_lpuart.c +18 -6
index 4becfbf35c4c..cac81b143abe 100644
@@ -2161,6 +2161,24 @@ static int lpuart_probe(struct platform_device *pdev)
 		return ret;
 	}
 
+	of_get_rs485_mode(np, &sport->port.rs485);
+
+	if (sport->port.rs485.flags & SER_RS485_RX_DURING_TX) {
+		dev_err(&pdev->dev, "driver doesn't support RX during TX\n");
+		return -ENOSYS;
+	}
+
+	if (sport->port.rs485.delay_rts_before_send ||
+	    sport->port.rs485.delay_rts_after_send) {
+		dev_err(&pdev->dev, "driver doesn't support RTS delays\n");
+		return -ENOSYS;
+	}
+
+	if (sport->port.rs485.flags & SER_RS485_ENABLED) {
+		sport->port.rs485.flags |= SER_RS485_RTS_ON_SEND;
+		writeb(UARTMODEM_TXRTSE, sport->port.membase + UARTMODEM);
+	}
+
 	sport->dma_tx_chan = dma_request_slave_channel(sport->port.dev, "tx");
 	if (!sport->dma_tx_chan)
 		dev_info(sport->port.dev, "DMA tx channel request failed, "

@@ -2171,12 +2189,6 @@ static int lpuart_probe(struct platform_device *pdev)
 		dev_info(sport->port.dev, "DMA rx channel request failed, "
 				"operating without rx DMA\n");
 
-	if (of_property_read_bool(np, "linux,rs485-enabled-at-boot-time")) {
-		sport->port.rs485.flags |= SER_RS485_ENABLED;
-		sport->port.rs485.flags |= SER_RS485_RTS_ON_SEND;
-		writeb(UARTMODEM_TXRTSE, sport->port.membase + UARTMODEM);
-	}
-
 	return 0;
 }
Signature · 1 line
2.11.0
#9 lore raw

PATCH v4 5/9 serial: fsl_lpuart: Use common rs485 device tree parsing function

Quoted 20 lines · Uwe Kleine-K?nig

From: Sascha Hauer <[email protected]>

We just got a common helper for parsing the rs485 specific device tree properties. Use it and drop the open coded parser.

As a side effect this adds support for parsing rs485-rts-delay and rs485-rx-during-tx. As the driver doesn't support this though, probing fails if these are defined.

Signed-off-by Sascha Hauer <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/fsl_lpuart.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)
drivers/tty/serial/fsl_lpuart.c +0 -0
index 4becfbf35c4c..cac81b143abe 100644

drivers/tty/serial/fsl_lpuart.c static int lpuart_probe(struct platform_device *pdev) · line 2164

Quoted 4 lines
drivers/tty/serial/fsl_lpuart.c +1 -0
@@ -2161,6 +2161,24 @@ static int lpuart_probe(struct platform_device *pdev)
 		return ret;
 	}
 
+	of_get_rs485_mode(np, &sport->port.rs485);

error checking?

Elsewhere in this series as well.

Also, why did you not cc: me on this series so I am sure to see it?

thanks,

greg k-h

author
#10 lore raw

PATCH v4 6/9 serial: omap-serial: Use common rs485 device tree parsing function

In reply to Uwe Kleine-König

From: Sascha Hauer <[email protected]>

We just got a common helper for parsing the rs485 specific device tree properties. Use it and drop the open coded parser.

Signed-off-by Sascha Hauer <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/omap-serial.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)
drivers/tty/serial/omap-serial.c +1 -12
index 1ea05ac57aa7..6d22097e6dd0 100644
@@ -1607,7 +1607,6 @@ static int serial_omap_probe_rs485(struct uart_omap_port *up,
 				   struct device_node *np)
 {
 	struct serial_rs485 *rs485conf = &up->port.rs485;
-	u32 rs485_delay[2];
 	enum of_gpio_flags flags;
 	int ret;
 

@@ -1638,17 +1637,7 @@ static int serial_omap_probe_rs485(struct uart_omap_port *up,
 		up->rts_gpio = -EINVAL;
 	}
 
-	if (of_property_read_u32_array(np, "rs485-rts-delay",
-				    rs485_delay, 2) == 0) {
-		rs485conf->delay_rts_before_send = rs485_delay[0];
-		rs485conf->delay_rts_after_send = rs485_delay[1];
-	}
-
-	if (of_property_read_bool(np, "rs485-rx-during-tx"))
-		rs485conf->flags |= SER_RS485_RX_DURING_TX;
-
-	if (of_property_read_bool(np, "linux,rs485-enabled-at-boot-time"))
-		rs485conf->flags |= SER_RS485_ENABLED;
+	of_get_rs485_mode(np, rs485conf);
 
 	return 0;
 }
Signature · 1 line
2.11.0
author
#11 lore raw

PATCH v4 7/9 serial: imx: default to half duplex rs485

In reply to Uwe Kleine-König

From: Sascha Hauer <[email protected]>

The i.MX driver defaulted to full duplex rs485 which is rather unusual and doesn't match the default implemented in other drivers.

So change the default to half duplex.

Signed-off-by Sascha Hauer <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/imx.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/tty/serial/imx.c +1 -2
index 80934e7bd67f..bb5cd8ff7d86 100644
@@ -2120,8 +2120,7 @@ static int serial_imx_probe(struct platform_device *pdev)
 	sport->port.fifosize = 32;
 	sport->port.ops = &imx_pops;
 	sport->port.rs485_config = imx_rs485_config;
-	sport->port.rs485.flags =
-		SER_RS485_RTS_ON_SEND | SER_RS485_RX_DURING_TX;
+	sport->port.rs485.flags = SER_RS485_RTS_ON_SEND;
 	sport->port.flags = UPF_BOOT_AUTOCONF;
 	init_timer(&sport->timer);
 	sport->timer.function = imx_timeout;
Signature · 1 line
2.11.0
author
#12 lore raw

PATCH v4 8/9 serial: imx: Use common rs485 device tree parsing function

In reply to Uwe Kleine-König

From: Sascha Hauer <[email protected]>

This adds support for the rs485 specific properties defined in Documentation/devicetree/bindings/serial/rs485.txt.

Signed-off-by Sascha Hauer <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 drivers/tty/serial/imx.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/tty/serial/imx.c +2 -0
index bb5cd8ff7d86..88957f31d8ed 100644
@@ -2059,6 +2059,8 @@ static int serial_imx_probe_dt(struct imx_port *sport,
 	if (of_get_property(np, "rts-gpios", NULL))
 		sport->have_rtsgpio = 1;
 
+	of_get_rs485_mode(np, &sport->port.rs485);
+
 	return 0;
 }
 #else
Signature · 1 line
2.11.0
author
#13 lore raw

PATCH v4 9/9 dt-bindings: serial: document rs485 bindings for various devices

In reply to Uwe Kleine-König

Atmel USART, Freescale UARTs and OMAP UART all support the rs485 binding described in rs485.txt, this commit just makes that explicit.

Acked-by Nicolas Ferre <[email protected]>
Acked-by Rob Herring <[email protected]>
Signed-off-by Uwe Kleine-K?nig <[email protected]>
---
 Documentation/devicetree/bindings/serial/atmel-usart.txt  | 1 +
 Documentation/devicetree/bindings/serial/fsl-imx-uart.txt | 1 +
 Documentation/devicetree/bindings/serial/fsl-lpuart.txt   | 1 +
 Documentation/devicetree/bindings/serial/omap_serial.txt  | 1 +
 4 files changed, 4 insertions(+)
Documentation/devicetree/bindings/serial/atmel-usart.txt +1 -0
index e6e6142e33ac..7c0d6b2f53e4 100644
@@ -24,6 +24,7 @@ Optional properties:
 	- dma-names: "rx" for RX channel, "tx" for TX channel.
 - atmel,fifo-size: maximum number of data the RX and TX FIFOs can store for FIFO
   capable USARTs.
+- rs485-rts-delay, rs485-rx-during-tx, linux,rs485-enabled-at-boot-time: see rs485.txt
 
 <chip> compatible description:
 - at91rm9200:  legacy USART support
Documentation/devicetree/bindings/serial/fsl-imx-uart.txt +1 -0
index 574c3a2c77d5..860a9559839a 100644
@@ -9,6 +9,7 @@ Optional properties:
 - fsl,irda-mode : Indicate the uart supports irda mode
 - fsl,dte-mode : Indicate the uart works in DTE mode. The uart works
                   in DCE mode by default.
+- rs485-rts-delay, rs485-rx-during-tx, linux,rs485-enabled-at-boot-time: see rs485.txt
 
 Please check Documentation/devicetree/bindings/serial/serial.txt
 for the complete list of generic properties.
Documentation/devicetree/bindings/serial/fsl-lpuart.txt +1 -0
index a1252a047f78..59567b51cf09 100644
@@ -16,6 +16,7 @@ Required properties:
 Optional properties:
 - dmas: A list of two dma specifiers, one for each entry in dma-names.
 - dma-names: should contain "tx" and "rx".
+- rs485-rts-delay, rs485-rx-during-tx, linux,rs485-enabled-at-boot-time: see rs485.txt
 
 Note: Optional properties for DMA support. Write them both or both not.
Documentation/devicetree/bindings/serial/omap_serial.txt +1 -0
index 7a71b5de77d6..43eac675f21f 100644
@@ -19,6 +19,7 @@ Optional properties:
 - dmas : DMA specifier, consisting of a phandle to the DMA controller
          node and a DMA channel number.
 - dma-names : "rx" for receive channel, "tx" for transmit channel.
+- rs485-rts-delay, rs485-rx-during-tx, linux,rs485-enabled-at-boot-time: see rs485.txt
 
 Example:
Signature · 1 line
2.11.0