Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1660162 > unrolled thread

[PATCH 06/15] serial: 8250_ingenic: Parse earlycon options

Started byPaul Cercueil <paul@crapouillou.net>
First post2017-06-07 22:10 +0200
Last post2017-06-08 23:20 +0200
Articles 3 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  [PATCH 06/15] serial: 8250_ingenic: Parse earlycon options Paul Cercueil <paul@crapouillou.net> - 2017-06-07 22:10 +0200
    Re: [PATCH 06/15] serial: 8250_ingenic: Parse earlycon options Marcin Nowakowski <marcin.nowakowski@imgtec.com> - 2017-06-08 09:40 +0200
      Re: [PATCH 06/15] serial: 8250_ingenic: Parse earlycon options Paul Cercueil <paul@crapouillou.net> - 2017-06-08 23:20 +0200

#1660162 — [PATCH 06/15] serial: 8250_ingenic: Parse earlycon options

FromPaul Cercueil <paul@crapouillou.net>
Date2017-06-07 22:10 +0200
Subject[PATCH 06/15] serial: 8250_ingenic: Parse earlycon options
Message-ID<tPPrc-2PB-17@gated-at.bofh.it>
In the devicetree, it is possible to specify the baudrate, parity,
bits, flow of the early console, by passing a configuration string like
this:

aliases {
	serial0 = &uart0;
};

chosen {
	stdout-path = "serial0:57600n8";
};

This, for instance, will configure the early console for a baudrate of
57600 bps, no parity, and 8 bits per baud.

This patches implements parsing of this configuration string in the
8250_ingenic driver, which previously just ignored it.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
---
 drivers/tty/serial/8250/8250_ingenic.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/tty/serial/8250/8250_ingenic.c b/drivers/tty/serial/8250/8250_ingenic.c
index b31b2ca552d1..59f3e632df49 100644
--- a/drivers/tty/serial/8250/8250_ingenic.c
+++ b/drivers/tty/serial/8250/8250_ingenic.c
@@ -99,14 +99,24 @@ static int __init ingenic_early_console_setup(struct earlycon_device *dev,
 					      const char *opt)
 {
 	struct uart_port *port = &dev->port;
-	unsigned int baud, divisor;
+	unsigned int divisor;
+	int baud = 115200;
 
 	if (!dev->port.membase)
 		return -ENODEV;
 
+	if (opt) {
+		char options[256];
+		unsigned int parity, bits, flow; /* unused for now */
+
+		strlcpy(options, opt, sizeof(options));
+		uart_parse_options(options, &baud, &parity, &bits, &flow);
+	}
+
 	ingenic_early_console_setup_clock(dev);
 
-	baud = dev->baud ?: 115200;
+	if (dev->baud)
+		baud = dev->baud;
 	divisor = DIV_ROUND_CLOSEST(port->uartclk, 16 * baud);
 
 	early_out(port, UART_IER, 0);
-- 
2.11.0

[toc] | [next] | [standalone]


#1660849

FromMarcin Nowakowski <marcin.nowakowski@imgtec.com>
Date2017-06-08 09:40 +0200
Message-ID<tQ0cW-1hd-33@gated-at.bofh.it>
In reply to#1660162
Hi Paul,

On 07.06.2017 22:04, Paul Cercueil wrote:
> In the devicetree, it is possible to specify the baudrate, parity,
> bits, flow of the early console, by passing a configuration string like
> this:
> 
> aliases {
> 	serial0 = &uart0;
> };
> 
> chosen {
> 	stdout-path = "serial0:57600n8";
> };
> 
> This, for instance, will configure the early console for a baudrate of
> 57600 bps, no parity, and 8 bits per baud.
> 
> This patches implements parsing of this configuration string in the
> 8250_ingenic driver, which previously just ignored it.
> 
> Signed-off-by: Paul Cercueil <paul@crapouillou.net>
> ---
>   drivers/tty/serial/8250/8250_ingenic.c | 14 ++++++++++++--
>   1 file changed, 12 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/tty/serial/8250/8250_ingenic.c b/drivers/tty/serial/8250/8250_ingenic.c
> index b31b2ca552d1..59f3e632df49 100644
> --- a/drivers/tty/serial/8250/8250_ingenic.c
> +++ b/drivers/tty/serial/8250/8250_ingenic.c
> @@ -99,14 +99,24 @@ static int __init ingenic_early_console_setup(struct earlycon_device *dev,
>   					      const char *opt)
>   {
>   	struct uart_port *port = &dev->port;
> -	unsigned int baud, divisor;
> +	unsigned int divisor;
> +	int baud = 115200;
>   
>   	if (!dev->port.membase)
>   		return -ENODEV;
>   
> +	if (opt) {
> +		char options[256];
> +		unsigned int parity, bits, flow; /* unused for now */
> +
> +		strlcpy(options, opt, sizeof(options));

Rather than adding this extra local copy maybe you could instead:

-void uart_parse_options(char *options, int *baud, int *parity, int *bits,
+void uart_parse_options(const char *options, int *baud, int *parity, 
int *bits,

I cannot see any reason why uart_parse_options shouldn't take 'const 
char *options' as an argument.

> +		uart_parse_options(options, &baud, &parity, &bits, &flow);
> +	}
> +
>   	ingenic_early_console_setup_clock(dev);
>   
> -	baud = dev->baud ?: 115200;
> +	if (dev->baud)
> +		baud = dev->baud;
>   	divisor = DIV_ROUND_CLOSEST(port->uartclk, 16 * baud);
>   
>   	early_out(port, UART_IER, 0);
> 


Marcin

[toc] | [prev] | [next] | [standalone]


#1661700

FromPaul Cercueil <paul@crapouillou.net>
Date2017-06-08 23:20 +0200
Message-ID<tQd0u-11f-25@gated-at.bofh.it>
In reply to#1660849
Hi,

[...]
>> diff --git a/drivers/tty/serial/8250/8250_ingenic.c 
>> b/drivers/tty/serial/8250/8250_ingenic.c
>> index b31b2ca552d1..59f3e632df49 100644
>> --- a/drivers/tty/serial/8250/8250_ingenic.c
>> +++ b/drivers/tty/serial/8250/8250_ingenic.c
>> @@ -99,14 +99,24 @@ static int __init 
>> ingenic_early_console_setup(struct earlycon_device *dev,
>>   					      const char *opt)
>>   {
>>   	struct uart_port *port = &dev->port;
>> -	unsigned int baud, divisor;
>> +	unsigned int divisor;
>> +	int baud = 115200;
>>     	if (!dev->port.membase)
>>   		return -ENODEV;
>>   +	if (opt) {
>> +		char options[256];
>> +		unsigned int parity, bits, flow; /* unused for now */
>> +
>> +		strlcpy(options, opt, sizeof(options));
> 
> Rather than adding this extra local copy maybe you could instead:
> 
> -void uart_parse_options(char *options, int *baud, int *parity, int 
> *bits,
> +void uart_parse_options(const char *options, int *baud, int *parity, 
> int *bits,
> 
> I cannot see any reason why uart_parse_options shouldn't take 'const
> char *options' as an argument.

Sure, good remark. I'll send a patch to change the prototype.

Thanks,
- Paul

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web