[PATCH v2 1/3] serial: 8250_fintek: Finish support for the F81865

[PATCH v2 1/3] serial: 8250_fintek: Finish support for the F81865

[Trevor Woerner]

This driver only partially supports the F81865 device. The UART portions of
this SuperIO chip behave very similarly to the UART of the F81866, except
that the F81866 has 128-byte FIFOs whereas the F81865 has 16-byte FIFOs,
and the IRQ configuration is different. Therefore fill out the support for
the F81865 in the places where it is missing.

Tested at 1500000 baud on the iEi NANO-PV-D5251-R10 board.

Signed-off-by: Trevor Woerner <twoerner@gmail.com>
---
 drivers/tty/serial/8250/8250_fintek.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/tty/serial/8250/8250_fintek.c b/drivers/tty/serial/8250/8250_fintek.c
index 251f0018ae8c..47b15d2d9901 100644
--- a/drivers/tty/serial/8250/8250_fintek.c
+++ b/drivers/tty/serial/8250/8250_fintek.c
@@ -63,7 +63,12 @@
 #define F81216_LDN_HIGH	0x4
 
 /*
- * F81866/966 registers
+ * F81866/865/966 registers
+ *
+ * The UART portion of the F81865 functions very similarly to the UART
+ * portion of the F81866, so there's no need to duplicate all the #defines
+ * etc. The only differences are: the F81866 has 128-byte FIFOs whereas the
+ * F81865 has 16-byte FIFOs, and the IRQ configuration is different.
  *
  * The IRQ setting mode of F81866/966 is not the same with F81216 series.
  *	Level/Low: IRQ_MODE0:0, IRQ_MODE1:0
@@ -316,6 +321,7 @@ static void fintek_8250_set_termios(struct uart_port *port,
 		break;
 	case CHIP_ID_F81966:
 	case CHIP_ID_F81866:
+	case CHIP_ID_F81865:
 		reg = F81866_UART_CLK;
 		break;
 	default:
@@ -363,6 +369,7 @@ static void fintek_8250_set_termios_handler(struct uart_8250_port *uart)
 	case CHIP_ID_F81216H:
 	case CHIP_ID_F81966:
 	case CHIP_ID_F81866:
+	case CHIP_ID_F81865:
 		uart->port.set_termios = fintek_8250_set_termios;
 		break;
 
-- 
2.35.1.455.g1a4874565f

Re: [PATCH v2 1/3] serial: 8250_fintek: Finish support for the F81865

[Greg Kroah-Hartman]

On Mon, Mar 14, 2022 at 08:18:56AM -0400, Trevor Woerner wrote:
> This driver only partially supports the F81865 device. The UART portions of
> this SuperIO chip behave very similarly to the UART of the F81866, except
> that the F81866 has 128-byte FIFOs whereas the F81865 has 16-byte FIFOs,
> and the IRQ configuration is different. Therefore fill out the support for
> the F81865 in the places where it is missing.
> 
> Tested at 1500000 baud on the iEi NANO-PV-D5251-R10 board.
> 
> Signed-off-by: Trevor Woerner <twoerner@gmail.com>
> ---
>  drivers/tty/serial/8250/8250_fintek.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/tty/serial/8250/8250_fintek.c b/drivers/tty/serial/8250/8250_fintek.c
> index 251f0018ae8c..47b15d2d9901 100644
> --- a/drivers/tty/serial/8250/8250_fintek.c
> +++ b/drivers/tty/serial/8250/8250_fintek.c
> @@ -63,7 +63,12 @@
>  #define F81216_LDN_HIGH	0x4
>  
>  /*
> - * F81866/966 registers
> + * F81866/865/966 registers
> + *
> + * The UART portion of the F81865 functions very similarly to the UART
> + * portion of the F81866, so there's no need to duplicate all the #defines
> + * etc. The only differences are: the F81866 has 128-byte FIFOs whereas the
> + * F81865 has 16-byte FIFOs, and the IRQ configuration is different.
>   *
>   * The IRQ setting mode of F81866/966 is not the same with F81216 series.
>   *	Level/Low: IRQ_MODE0:0, IRQ_MODE1:0
> @@ -316,6 +321,7 @@ static void fintek_8250_set_termios(struct uart_port *port,
>  		break;
>  	case CHIP_ID_F81966:
>  	case CHIP_ID_F81866:
> +	case CHIP_ID_F81865:
>  		reg = F81866_UART_CLK;
>  		break;
>  	default:
> @@ -363,6 +369,7 @@ static void fintek_8250_set_termios_handler(struct uart_8250_port *uart)
>  	case CHIP_ID_F81216H:
>  	case CHIP_ID_F81966:
>  	case CHIP_ID_F81866:
> +	case CHIP_ID_F81865:
>  		uart->port.set_termios = fintek_8250_set_termios;
>  		break;
>  
> -- 
> 2.35.1.455.g1a4874565f
> 

Please resend the whole series, not just one patch out of the lot.  Also
can you correctly "thread" them?  Using a tool like git send-email it
happens automatically.  That way our tools can pick them up properly.

thanks,

greg k-h