[PATCH V2] tty: serial: ralink: fix SERIAL_8250_RT288X dependency

[PATCH V2] tty: serial: ralink: fix SERIAL_8250_RT288X dependency

[John Crispin]

With every Ralink SoC that we add, we would need to extend the dependency. In
order to make life easier we make the symbol depend on MIPS & RALINK and then
select it from within arch/mips/ralink/.

Signed-off-by: John Crispin <blogic@openwrt.org>
---
Hi Greg,

this patch should go upstream via the mips tree to avoid merge conflicts.
The tty part however requires your Ack.

	John

 arch/mips/Kconfig               |    1 +
 drivers/tty/serial/8250/Kconfig |    4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index c1997db..2e8939f 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -441,6 +441,7 @@ config RALINK
 	select SYS_HAS_EARLY_PRINTK
 	select HAVE_MACH_CLKDEV
 	select CLKDEV_LOOKUP
+	select SERIAL_8250_RT288X
 
 config SGI_IP22
 	bool "SGI IP22 (Indy/Indigo2)"
diff --git a/drivers/tty/serial/8250/Kconfig b/drivers/tty/serial/8250/Kconfig
index 80fe91e..24ea3c8 100644
--- a/drivers/tty/serial/8250/Kconfig
+++ b/drivers/tty/serial/8250/Kconfig
@@ -295,8 +295,8 @@ config SERIAL_8250_EM
 	  If unsure, say N.
 
 config SERIAL_8250_RT288X
-	bool "Ralink RT288x/RT305x/RT3662/RT3883 serial port support"
-	depends on SERIAL_8250 && (SOC_RT288X || SOC_RT305X || SOC_RT3883)
+	bool
+	depends on SERIAL_8250 && MIPS && RALINK
 	help
 	  If you have a Ralink RT288x/RT305x SoC based board and want to use the
 	  serial port, say Y to this option. The driver can handle up to 2 serial
-- 
1.7.10.4

Re: [PATCH V2] tty: serial: ralink: fix SERIAL_8250_RT288X dependency

[Jonas Gorski]

On 16 April 2013 08:18, John Crispin <blogic@openwrt.org> wrote:
> With every Ralink SoC that we add, we would need to extend the dependency. In
> order to make life easier we make the symbol depend on MIPS & RALINK and then
> select it from within arch/mips/ralink/.
>
> Signed-off-by: John Crispin <blogic@openwrt.org>
> ---
> Hi Greg,
>
> this patch should go upstream via the mips tree to avoid merge conflicts.
> The tty part however requires your Ack.
>
>         John
>
>  arch/mips/Kconfig               |    1 +
>  drivers/tty/serial/8250/Kconfig |    4 ++--
>  2 files changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
> index c1997db..2e8939f 100644
> --- a/arch/mips/Kconfig
> +++ b/arch/mips/Kconfig
> @@ -441,6 +441,7 @@ config RALINK
>         select SYS_HAS_EARLY_PRINTK
>         select HAVE_MACH_CLKDEV
>         select CLKDEV_LOOKUP
> +       select SERIAL_8250_RT288X
>
>  config SGI_IP22
>         bool "SGI IP22 (Indy/Indigo2)"
> diff --git a/drivers/tty/serial/8250/Kconfig b/drivers/tty/serial/8250/Kconfig
> index 80fe91e..24ea3c8 100644
> --- a/drivers/tty/serial/8250/Kconfig
> +++ b/drivers/tty/serial/8250/Kconfig
> @@ -295,8 +295,8 @@ config SERIAL_8250_EM
>           If unsure, say N.
>
>  config SERIAL_8250_RT288X
> -       bool "Ralink RT288x/RT305x/RT3662/RT3883 serial port support"
> -       depends on SERIAL_8250 && (SOC_RT288X || SOC_RT305X || SOC_RT3883)
> +       bool
> +       depends on SERIAL_8250 && MIPS && RALINK

This won't work, Having RALINK=y, but SERIAL_8250=n will still result
in SERIAL_8250_RT288X=y, as select ignores dependencies. What could
work is removing the select from RALINK, and changing the depends from
this one to "default y if SERIAL_8250 && MIPS && RALINK".

>         help
>           If you have a Ralink RT288x/RT305x SoC based board and want to use the
>           serial port, say Y to this option. The driver can handle up to 2 serial
> --
> 1.7.10.4
>
>

Jonas

Re: [PATCH V2] tty: serial: ralink: fix SERIAL_8250_RT288X dependency

[John Crispin]

> This won't work, Having RALINK=y, but SERIAL_8250=n will still result
> in SERIAL_8250_RT288X=y, as select ignores dependencies. What could
> work is removing the select from RALINK, and changing the depends from
> this one to "default y if SERIAL_8250&&  MIPS&&  RALINK".
>
>

ok, these tty patches were rushed too much. i think trying to get them 
ready quickly for 3.10 was a bad bad idea, that already wasted too much 
of other peoples time. instead of making even more of a fool of myself 
and wasting yet more time we should just drop it and i will try again 
with more time for 3.11.

Sorry for the noise :-)
     John