|PATCH] stmmac: dwmac-loongson: Add architecture dependency

|PATCH] stmmac: dwmac-loongson: Add architecture dependency

[Jean Delvare]

Only present the DWMAC_LOONGSON option on architectures where it can
actually be used.

This follows the same logic as the DWMAC_INTEL option.

Signed-off-by: Jean Delvare <jdelvare@suse.de>
Cc: Keguang Zhang <keguang.zhang@gmail.com>
---
I'm not familiar with the hardware, so please let me know if the
dependency needs to be adjusted somehow.

 drivers/net/ethernet/stmicro/stmmac/Kconfig |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- linux-6.6.orig/drivers/net/ethernet/stmicro/stmmac/Kconfig
+++ linux-6.6/drivers/net/ethernet/stmicro/stmmac/Kconfig
@@ -269,7 +269,7 @@ config DWMAC_INTEL
 config DWMAC_LOONGSON
 	tristate "Loongson PCI DWMAC support"
 	default MACH_LOONGSON64
-	depends on STMMAC_ETH && PCI
+	depends on MACH_LOONGSON64 && STMMAC_ETH && PCI
 	depends on COMMON_CLK
 	help
 	  This selects the LOONGSON PCI bus support for the stmmac driver,


-- 
Jean Delvare
SUSE L3 Support

Re: |PATCH] stmmac: dwmac-loongson: Add architecture dependency

[Simon Horman]

On Mon, Nov 13, 2023 at 03:45:22PM +0100, Jean Delvare wrote:
> Only present the DWMAC_LOONGSON option on architectures where it can
> actually be used.
> 
> This follows the same logic as the DWMAC_INTEL option.
> 
> Signed-off-by: Jean Delvare <jdelvare@suse.de>
> Cc: Keguang Zhang <keguang.zhang@gmail.com>
> ---
> I'm not familiar with the hardware, so please let me know if the
> dependency needs to be adjusted somehow.
> 
>  drivers/net/ethernet/stmicro/stmmac/Kconfig |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> --- linux-6.6.orig/drivers/net/ethernet/stmicro/stmmac/Kconfig
> +++ linux-6.6/drivers/net/ethernet/stmicro/stmmac/Kconfig
> @@ -269,7 +269,7 @@ config DWMAC_INTEL
>  config DWMAC_LOONGSON
>  	tristate "Loongson PCI DWMAC support"
>  	default MACH_LOONGSON64
> -	depends on STMMAC_ETH && PCI
> +	depends on MACH_LOONGSON64 && STMMAC_ETH && PCI

Hi Jean,

Could we consider also allowing the build to occur if COMPILE_TEST is set?
This would maintain the current level of build test coverage.

Something line this (completely untested!):

	depends on (MACH_LOONGSON64 || COMPILE_TEST) && STMMAC_ETH && PCI

>  	depends on COMMON_CLK
>  	help
>  	  This selects the LOONGSON PCI bus support for the stmmac driver,
> 
> 
> -- 
> Jean Delvare
> SUSE L3 Support
> 

Re: |PATCH] stmmac: dwmac-loongson: Add architecture dependency

[Jean Delvare]

Hi Simon,

On Mon, 13 Nov 2023 18:01:07 +0000, Simon Horman wrote:
> On Mon, Nov 13, 2023 at 03:45:22PM +0100, Jean Delvare wrote:
> > Only present the DWMAC_LOONGSON option on architectures where it can
> > actually be used.
> > 
> > This follows the same logic as the DWMAC_INTEL option.
> > 
> > Signed-off-by: Jean Delvare <jdelvare@suse.de>
> > Cc: Keguang Zhang <keguang.zhang@gmail.com>
> > ---
> > I'm not familiar with the hardware, so please let me know if the
> > dependency needs to be adjusted somehow.
> > 
> >  drivers/net/ethernet/stmicro/stmmac/Kconfig |    2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > --- linux-6.6.orig/drivers/net/ethernet/stmicro/stmmac/Kconfig
> > +++ linux-6.6/drivers/net/ethernet/stmicro/stmmac/Kconfig
> > @@ -269,7 +269,7 @@ config DWMAC_INTEL
> >  config DWMAC_LOONGSON
> >  	tristate "Loongson PCI DWMAC support"
> >  	default MACH_LOONGSON64
> > -	depends on STMMAC_ETH && PCI
> > +	depends on MACH_LOONGSON64 && STMMAC_ETH && PCI  
> 
> Could we consider also allowing the build to occur if COMPILE_TEST is set?
> This would maintain the current level of build test coverage.
> 
> Something line this (completely untested!):
> 
> 	depends on (MACH_LOONGSON64 || COMPILE_TEST) && STMMAC_ETH && PCI

Sure, that would be totally fine with me.

-- 
Jean Delvare
SUSE L3 Support

Re: |PATCH] stmmac: dwmac-loongson: Add architecture dependency

[Simon Horman]

On Mon, Nov 13, 2023 at 07:33:47PM +0100, Jean Delvare wrote:
> Hi Simon,
> 
> On Mon, 13 Nov 2023 18:01:07 +0000, Simon Horman wrote:
> > On Mon, Nov 13, 2023 at 03:45:22PM +0100, Jean Delvare wrote:
> > > Only present the DWMAC_LOONGSON option on architectures where it can
> > > actually be used.
> > > 
> > > This follows the same logic as the DWMAC_INTEL option.
> > > 
> > > Signed-off-by: Jean Delvare <jdelvare@suse.de>
> > > Cc: Keguang Zhang <keguang.zhang@gmail.com>
> > > ---
> > > I'm not familiar with the hardware, so please let me know if the
> > > dependency needs to be adjusted somehow.
> > > 
> > >  drivers/net/ethernet/stmicro/stmmac/Kconfig |    2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > --- linux-6.6.orig/drivers/net/ethernet/stmicro/stmmac/Kconfig
> > > +++ linux-6.6/drivers/net/ethernet/stmicro/stmmac/Kconfig
> > > @@ -269,7 +269,7 @@ config DWMAC_INTEL
> > >  config DWMAC_LOONGSON
> > >  	tristate "Loongson PCI DWMAC support"
> > >  	default MACH_LOONGSON64
> > > -	depends on STMMAC_ETH && PCI
> > > +	depends on MACH_LOONGSON64 && STMMAC_ETH && PCI  
> > 
> > Could we consider also allowing the build to occur if COMPILE_TEST is set?
> > This would maintain the current level of build test coverage.
> > 
> > Something line this (completely untested!):
> > 
> > 	depends on (MACH_LOONGSON64 || COMPILE_TEST) && STMMAC_ETH && PCI
> 
> Sure, that would be totally fine with me.

Thanks,

I think that if no one objects within the next day or so then posting a v2
would be a good next step.