Re: [PATCH v2] tty: serial: fsl_lpuart: select SERIAL_FSL_LPUART for SERIAL_FSL_LPUART_CONSOLE

From: Randy Dunlap
Date: Sun Feb 26 2023 - 10:45:11 EST


Hi Tom,

[adding the recent commit signers]

On 2/26/23 06:54, Tom Rix wrote:
> A rand config causes this link error
> ld: drivers/tty/serial/earlycon.o: in function `parse_options':
> drivers/tty/serial/earlycon.c:99: undefined reference to `uart_parse_earlycon'
>
> The rand config has
> CONFIG_SERIAL_CORE=m
> CONFIG_SERIAL_EARLYCON=y
> CONFIG_SERIAL_FSL_LPUART=m
> CONFIG_SERIAL_FSL_LPUART_CONSOLE=y
>
> SERIAL_FSL_LPUART should have been selected instead of depends on-ed.
>
> After applying the configs are
> CONFIG_SERIAL_CORE=y
> CONFIG_SERIAL_EARLYCON=y
> CONFIG_SERIAL_FSL_LPUART=y
> CONFIG_SERIAL_FSL_LPUART_CONSOLE=y

This works, but I thought that the point of the recent patch was to make the driver modular.

You said:
"commit 5779a072c248db7a40cfd0f5ea958097fd1d9a30 removed the =y. so it could be built as a module."

That's not the same as what is happening here.

I think it would be better to revert the commit mentioned above.

> Signed-off-by: Tom Rix <trix@xxxxxxxxxx>
> ---
> v2: Add how the configs changed to commit log
> ---
> drivers/tty/serial/Kconfig | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/tty/serial/Kconfig b/drivers/tty/serial/Kconfig
> index 625358f44419..b24d74d389fc 100644
> --- a/drivers/tty/serial/Kconfig
> +++ b/drivers/tty/serial/Kconfig
> @@ -1313,7 +1313,7 @@ config SERIAL_FSL_LPUART
>
> config SERIAL_FSL_LPUART_CONSOLE
> bool "Console on Freescale lpuart serial port"
> - depends on SERIAL_FSL_LPUART
> + select SERIAL_FSL_LPUART
> select SERIAL_CORE_CONSOLE
> select SERIAL_EARLYCON
> help

--
~Randy